IPR2017-01039 U.S. Patent 7,231,379
Filed on behalf of Unified Patents Inc. By: Jason R. Mudd, Reg. No. 57,700
Eric A. Buresh, Reg. No. 50,394 [email protected] [email protected] ERISE IP, P.A. 6201 College Blvd., Suite 300 Overland Park, Kansas 66211 Telephone: (913) 777-5600 Jonathan Stroud, Reg. No. 72,518 [email protected] Unified Patents Inc. 1875 Connecticut Ave. NW, Floor 10 Washington, D.C., 20009 Telephone: (202) 805-8931
Roshan Mansinghani, Reg. No. 62,429 [email protected] Unified Patents Inc. 13355 Noel Road, Suite 1100 Dallas, TX, 75240 Telephone: (214) 945-0200
UNITED STATES PATENT AND TRADEMARK OFFICE
____________
BEFORE THE PATENT TRIAL AND APPEAL BOARD ____________
UNIFIED PATENTS INC.
Petitioner
v.
GUADA TECHNOLOGIES LLC Patent Owner
____________
IPR2017-01039 Patent 7,231,379 ____________
PETITION FOR INTER PARTES REVIEW
OF U.S. PATENT NO. 7,231,379
IPR2017-01039 U.S. Patent No. 7,231,379
i
TABLE OF CONTENTS
I. INTRODUCTION ............................................................................................ 1
II. SUMMARY OF THE ’379 PATENT ............................................................. 2
A. Description of the alleged invention of the ’379 Patent ................................. 2
B. Summary of the prosecution history of the ’379 Patent ................................. 5
III. REQUIREMENTS FOR INTER PARTES REVIEW UNDER 37 C.F.R. §
42.104........................................................................................................................ 6
A. Grounds for standing under 37 C.F.R. § 42.104(a) ........................................ 6
B. Identification of challenge under 37 C.F.R. § 42.104(b) and relief requested 7
C. Level of skill of a person having ordinary skill in the art ............................... 8
D. Claim construction under 37 C.F.R. § 42.104(b)(3) ....................................... 8
IV. THERE IS A REASONABLE LIKELIHOOD THAT THE
CHALLENGED CLAIMS OF THE ’379 PATENT ARE UNPATENTABLE
13
A. Ground 1: Wesemann renders claims 1, 2, and 7 obvious ........................... 13
B. Ground 2: Wesemann in view of Rajaraman renders Claims 3-6 obvious. . 36
C. Ground 3: Fratkina renders claims 1, 2, and 7 obvious ................................ 50
D. Ground 4: Fratkina in view of Rajaraman renders Claims 3-6 obvious. ...... 66
V. CONCLUSION ............................................................................................... 68
VI. MANDATORY NOTICES UNDER 37 C.F.R. § 42.8(A)(1) ...................... 69
IPR2017-01039 U.S. Patent No. 7,231,379
ii
A. Real Party-In-Interest ................................................................................... 69
B. Related Matters ............................................................................................. 69
C. Lead and Back-Up Counsel .......................................................................... 70
IPR2017-01039 U.S. Patent No. 7,231,379
1
I. INTRODUCTION
Petitioner Unified Patents Inc. (“Petitioner”) respectfully requests an Inter
Partes Review (“IPR”) of claims 1-7 (collectively, the “Challenged Claims”) of
U.S. Patent 7,231,379 (“the ’379 Patent”). The ’379 Patent broadly claims the use
of keywords for searching a hierarchical network. The keywords are associated
with different points on the hierarchical network, referred to as “nodes” or
“vertices.” Put simply, there are only two concepts related to navigation of
hierarchical systems in the claims, and they are both obvious in view of the prior
art.
First, when a user inputs a given keyword, the claimed method causes the
system to “jump” to the node or vertex associated with that keyword, without
requiring the user to traverse through each intervening step in the hierarchical
network. The allowance of the ’379 Patent was largely based on this “jumping”
concept. However, as shown by the Wesemann and Fratkina prior art references,
such “jump[ing]” between different nodes was well-known in hierarchically
arranged systems before the filing of the ’379 Patent in 2002. See, e.g. Wesemann
(EX1004) at Abstract; see also, e.g., Smyth Decl. (EX1007) at ¶¶36-40, 49, 55-56,
84. Neither Wesemann nor Fratkina was cited during prosecution of the ’379
Patent.
IPR2017-01039 U.S. Patent No. 7,231,379
2
Second, the ’379 Patent includes four claims related to using a thesaurus to
search synonyms of user inputs and updating that thesaurus to understand new
synonyms. However, such thesaurus searching and updating had already been
significantly developed by the 1990s, and these thesaurus limitations fail to add
anything new over the prior art. Smyth Decl. (EX1007) at ¶¶41-45, 69-76.
Navigating hierarchical trees was not novel in 2002, and the claims of the
’379 Patent fail to include any limitations that would make it a non-obvious
improvement over what had already been practiced for years before the filing of
the patent. See infra Section IV; see also Smyth Decl. (EX1007) generally at ¶¶31-
93. Petitioner, therefore, respectfully requests institution of inter partes review of
the Challenged Claims.
II. SUMMARY OF THE ’379 PATENT
A. Description of the alleged invention of the ’379 Patent
The ’379 Patent relates to methods for searching a hierarchical “menu tree”
of nodes or vertices. ’379 Patent (EX1001) at 2:22-30, 3:5-28. The Applicant’s
allegedly novel take on this concept is a search system that “jumps” to different
nodes on a hierarchical tree without traversing through intervening nodes on the
hierarchical tree. See, e.g., ’379 Patent File History (EX1002) at 47 (Response to
Non-Compliant Appeal Brief (Nov. 6, 2006)) (“Appellant's claimed invention
solves the inadequacies of prior art systems, by allowing the system to cause the
IPR2017-01039 U.S. Patent No. 7,231,379
3
user to “jump” from one node in the hierarchy to another node that is not directly
connected to that node, without having to traverse through every intervening node
in the path . . . .”); see also Patent Owner’s Opposition to Motion to Dismiss
(EX1003) at 18-19 (“As discussed extensively during the prosecution history, the
‘jumping’ term, as defined above, was a point of novelty that distinguished the
claimed invention from the prior art.”). In prosecution and litigation, Patent Owner
(PO) has construed “jumping,” used in both independent claims 1 and 7, to mean
“a direct traversal from one node or vertex to another node or vertex that is not
directly connected to it (i.e., without traversal through any intervening nodes or
vertices or to a node or vertex whose only least common ancestor with that node or
vertex is the root node or vertex).” 1 Opposition to Motion to Dismiss (EX1003) at
18; ’379 Patent File History (EX1002) at 89 (Appeal Brief (May 31, 2005)). The
’379 Patent asserts that jumping may occur laterally (i.e., across branches of the
hierarchical tree) or vertically (i.e., up or down a corresponding hierarchical tree
branch). See, e.g., ’379 Patent (EX1001) at 12:49-56, 14:54-63. A simple example
of jumping explained by the ’379 Patent has been reproduced in reference to
Figure 2:
[W]hen a response to a verbal description is provided by a user, 1 For the purposes of this Petition, Petitioner does not contest this construction of
“jumping” as the broadest reasonable interpretation of this term. See Sec. III.D, infra.
IPR2017-01039 U.S. Patent No. 7,231,379
4
possible keywords are identified in the response and used to search
the index and identify any node to which the response may be
directed, irrespective of the hierarchy. Thus, a user response of “an
orange” to a verbal description located above the “fruit” node 202 in
the hierarchy, for example, “What would you like to buy today?”
would cause the system to identify “orange” as a key word from the
response, search the index, and directly identify node [](206) as the
node whose verbal description should be presented next, thereby
avoiding the need to traverse intervening nodes, for example, through
the “fruit” node (202)[], at all. This illustrates an example of a simple
jump according to the invention.
Id. at 6:7-21, Fig. 2.
As discussed herein, this concept of navigating between nodes or vertices in
a hierarchical system by “jumping” to and from different nodes or vertices on the
hierarchical tree without traversing through intervening nodes or vertices was well-
known prior to the filing of the ’379 Patent. For example, Wesemann, cited below,
IPR2017-01039 U.S. Patent No. 7,231,379
5
discloses both lateral and vertical jumping through a hierarchical network, and
Fratkina, also cited below, teaches that users may skip over parts of a hierarchical
menu.
Claims 3-6 of the ’379 Patent further recite a method for using a thesaurus
function to identify words input by a user as being synonymous with keywords.
Additional synonyms for keywords may be added to the thesaurus and associated
with nodes as users input new words into the system and navigate the system. See
id. at 9:65-10:2, 10:41-43. As discussed herein, these concepts were also known
prior to the filing of the ’379 Patent. See Smyth Decl. (EX1007) at ¶¶41-45, 69-76.
For example, Rajaraman, cited below, teaches each of these limitations in the
context of hierarchical searching. See infra Sec. IV.B.
B. Summary of the prosecution history of the ’379 Patent
The application that resulted in the ’379 Patent was filed on November 19,
2002. See ’379 Patent (EX1001). For purposes of this proceeding, Petitioner
assumes that the priority date for the Challenged Claims is November 19, 2002.
The original application included 26 claims, but was reduced to seven claims due
to a restriction requirement. ’379 Patent File History (EX1002) at 180-189. These
seven claims were not amended from their original application during prosecution,
even after three rejections. Id. at 74, 139, 164, 181.
During prosecution, the Applicants focused on the “jumping” concept of
IPR2017-01039 U.S. Patent No. 7,231,379
6
claims 1 and 7, insisting that “jumping” required a system (as opposed to a user) to
traverse from one node to a second, non-adjacent node based on an input from a
user, without traversing through intervening nodes in the path. See id. at 62-64; see
also, e.g., id. at 89-90, 127-30, 133, 156. The applicant took issue with cited prior
art that applicant contended graphically presented the user with the matching node
and required the user to select the node to cause the jump. See id. at 64
(distinguishing Pooser). The applicant made only general and conclusory remarks
with respect to claims 3-6. See, e.g., id. at 135, 160.
The claimed “jumping” feature, therefore, appears to have led to allowance
of independent claims 1 and 7. But, as discussed below, this concept was already
well-known in the prior art. For instance, Wesemann—not cited during
prosecution—emphasizes that a feature in its system is that users do not have to go
through “in-between” menu states. Unlike the cited prior art distinguished by
applicant, Wesemann (as well as Fratkina) uses interactive voice response systems
(like those disclosed in the ’379 Patent) that automatically perform the
“jumping”—no graphical display is present from which a user makes a selection to
cause the jump. See, e.g., Wesemann (EX1004) at 3:50-56.
III. REQUIREMENTS FOR INTER PARTES REVIEW UNDER 37 C.F.R. § 42.104
A. Grounds for standing under 37 C.F.R. § 42.104(a)
Petitioner certifies that the ‘379 Patent is available for IPR and that the
IPR2017-01039 U.S. Patent No. 7,231,379
7
Petitioner is not barred or estopped from requesting IPR challenging the claims of
the ‘379 Patent.
B. Identification of challenge under 37 C.F.R. § 42.104(b) and relief requested
In view of the prior art and evidence, claims 1-7 of the ’379 Patent are
unpatentable and should be cancelled. 37 C.F.R. § 42.104(b)(1). Based on the prior
art references identified below, IPR of the Challenged Claims should be granted.
37 C.F.R. § 42.104(b)(2).
Proposed Grounds of Unpatentability Exhibit Nos.
Ground 1: Claims 1, 2, and 7 are obvious under § 103(a) over U.S.
Pat. No. 6,731,724 to Wesemann et al. (“Wesemann”). EX1004
Ground 2: Claims 3-6 are obvious under § 103(a) over U.S. Pat.
No. 6,731,724 to Wesemann et al. (“Wesemann”) in view of U.S.
Pat. No. 6,366,910 to Rajaraman et al. (“Rajaraman”)
EX1004, EX1005
Ground 3: Claims 1, 2, and 7 are obvious under § 103(a) over U.S.
Pat. No. 7,539,656 to Fratkina et al. (“Fratkina”) EX1006
Ground 4: Claims 3-6 are obvious under § 103(a) over U.S. Pat.
No. 7,539,656 to Fratkina et al. (“Fratkina”) in view of U.S. Pat.
No. 6,366,910 to Rajaraman et al. (“Rajaraman”)
EX1006, EX1005
Section IV identifies where each element of the Challenged Claims is found in the
prior art. 37 C.F.R. § 42.104(b)(4). The exhibit numbers of the supporting evidence
relied upon to support the challenges are provided above and the relevance of the
IPR2017-01039 U.S. Patent No. 7,231,379
8
evidence to the challenges raised are provided in Section IV. 37 C.F.R.
§ 42.104(b)(5). Exhibits EX1001 – EX1025 are also attached.
C. Level of skill of a person having ordinary skill in the art
As explained by Petitioner’s expert, Dr. Padhraic Smyth, a Professor of
Computer Science at University of California, Irvine, a person having ordinary
skill in the art (“PHOSITA”) at the time of the ’379 Patent would have been a
person having the equivalent of a bachelor’s degree in computer science, electrical
engineering, or a similar discipline, and at least one year of experience working
with technology related to information retrieval and database searching, or an
equivalent amount of similar work experience or education, with additional
education substituting for experience and vice versa. Smyth Decl. (EX1007) at
¶¶28-30.
D. Claim construction under 37 C.F.R. § 42.104(b)(3)
The ’379 Patent has not expired. As such, the claim terms should be given
their “broadest reasonable construction in light of the specification[.]” 37 C.F.R.
§ 42.100(b). Petitioner proposes the following constructions. All claim terms not
specifically discussed below should be given their broadest reasonable
interpretation (BRI) in light of the specification.
i. “node”
All seven claims use the term “node.” Referring to Figure 1, the ‘379 Patent states,
IPR2017-01039 U.S. Patent No. 7,231,379
9
“[t]he individual boxes 102-120 are referred to as ‘nodes’ and each represents a
specific choice or option in the hierarchy.” ‘379 Patent (EX1001) at 4:22-26. Thus,
the BRI of “node” is “a specific choice or option in a hierarchy.”
ii. “vertex”
Only claim 7 uses the term “vertex” (and its plural form, “vertices”). “Node”
and “vertex” are used interchangeably throughout both the patent specification and
its prosecution history. See, e.g., ’379 Patent (EX1001) at 2:5-9 (“In general, there
will also be a combination of vertices or nodes in the graph that best represent or
are closest to the goal the user is trying to accomplish. We call these vertices the
‘goal vertices.’”) ’379 File History (EX1002) at 47-48 (describing jumping as
direct traversal “from one node or vertex to another node or vertex”). The only
apparent distinction in the patent between a “node” and a “vertex” is that a
“vertex” is a node in a hierarchy that can be represented as a graph:
In modern mathematics, graph theory is used to study networks of
hierarchical choices. The hierarchical networks can be represented as
a graph structure. . . . A graph structure is a collection of points,
called “vertices”, and a collection of lines, called “edges”. Each
edge joins a pair of vertices or a single point to itself.
’379 Patent (EX1001) at 1:27-35 (emphasis added). The patent continues:
Our invention is particularly applicable to transactional processing as
applied to instances where graph theory can be used to represent the
transactions as a set of options and when the options are structured
IPR2017-01039 U.S. Patent No. 7,231,379
10
according to a connected graph that contains no circuits. We call such a
graph a “tree”. . . . Informally, a “menu tree” can be regarded as a
series of vertices in a hierarchy or ordered pattern, arranged in
rows of increasing numbers of vertices. More precisely, a “menu
tree” can be represented as a “tree” in which (i) the vertices are all the
options provided anywhere in the “menu tree”, plus a first vertex,
(ii) every vertex except the first vertex, i.e., every “option vertex”, is
associated with the verbal description (or such other means) by which a
“menu” presents that option, (iii) an edge connects the first vertex to
each vertex that the first “menu” presents to the user as an option,
and (iv) each other vertex is similarly connected by edges to every
other vertex that the corresponding “menu” presents to the user as an
option.
’379 Patent (EX1001) at 3:5-27 (emphasis added); see also id. at 3:59-63 (“It
should be understood that the present invention is applicable to a wide range of
different networks, which can be mathematically represented by graph structures
consisting of vertices and edges”). Therefore, the BRI of “vertex” is a “a specific
choice or option in a hierarchy that can be represented in a graph.”
iii. “keyword”
The ’379 Patent teaches that each node in the hierarchy is associated with a
verbal description (whether audible or written) and that “[e]ach such description
contains ‘key’ words that are deemed to be of importance and other words that can
be disregarded.” Id. at 4:32-41, 1:49-52 (audible or written). “For example, one
IPR2017-01039 U.S. Patent No. 7,231,379
11
node may have the associated verbal description ‘Would you like to make a
reservation?’ In this description, there is only one ‘key’ word—‘reservation’
deemed important, so all of the other words in the description can be ignored.” Id.
at 4:37-41. Other nodes may have verbal descriptions with multiple keywords. For
example, the verbal description “Is the reservation for a domestic or international
flight?” is described as having two keywords, “domestic” and “international.” Id.
at 4:44-51. And the word “flight” could be a keyword if the system includes non-
air travel options, such as train, but it could also be an ignored term if, for example,
the system is only for airline reservations. Id.
The ’379 Patent describes an index that associates keywords with nodes. Id.
at 4:62-5:7. This index allows the menu tree of nodes to be searched by keyword
regardless of where in the hierarchy the user is currently located by allowing them
to jump to a node matching the keyword. Id. at 5:7-12. The patent teaches that the
specific format described for the index is only for illustration and “that other
techniques for interrelating data, such as hash tables, direct or indirect indexing,
etc. can be substituted in a straightforward manner.” Id. at 5:23-27.
The ’379 Patent describes embodiments based on an interactive voice
response (“IVR”) system in which a user responds vocally to prompts and
“keywords” are identified from the user’s speech. Id. at 6:63-7:9. Importantly, the
IPR2017-01039 U.S. Patent No. 7,231,379
12
’379 Patent explicitly teaches that a keyword can be more than just a single word
and can also include other forms of information, such as specific data patterns:
Note, there is no requirement for a [] ‘keyword’ to be a single
word, in some implementations, keywords could be single words,
phrases of two or more words, or even some other form of
information like a specific data pattern.
Id. at 7:5-9 (emphasis added).
Accordingly, the BRI of “keyword” is “one or more words or pieces of
information, such as a data pattern, that is associated with at least one node or
vertex.”
iv. “jumping”
As discussed above in the Summary of the ’379 Patent, the applicant during
prosecution and PO more recently have construed “jumping” to mean “a direct
traversal from one node or vertex to another node or vertex that is not
directly connected to it (i.e., without traversal through any intervening nodes
or vertices or to a node or vertex whose only least common ancestor with that
node or vertex is the root node or vertex).” See supra Sec. II.A (detailing Patent
Owner’s explicit definition during prosecution and recent motion to dismiss
(EX1003)). Petitioner notes that this jumping may occur laterally (i.e., across
branches of the hierarchical tree) or vertically (i.e., up or down a hierarchical tree
branch). See, e.g., ’379 Patent (EX1001) at 12:49-56, 14:54-63. Given Applicant’s
IPR2017-01039 U.S. Patent No. 7,231,379
13
repeated assertions regarding this term during prosecution and its subsequent
allowance, the BRI of “jumping” at least includes PO’s above quoted construction.
v. “verbal description”
The ’379 Patent teaches that each node in the hierarchy is associated with a
verbal description that describes the subject matter of the particular node. Id. at
3:37-43, 4:32-41. The ’379 Patent expressly defines a “verbal description” as “a set
of words relating to the subject matter whether presented audibly or in written
form.” Id. at 1:50-52. Accordingly, the BRI of “verbal description” is “a set of
words relating to the subject matter whether presented audibly or in written
form.”
IV. THERE IS A REASONABLE LIKELIHOOD THAT THE CHALLENGED CLAIMS OF THE ’379 PATENT ARE UNPATENTABLE
A. Ground 1: Wesemann renders claims 1, 2, and 7 obvious U.S. Patent No. 6,731,724 to Wesemann et al. (“Wesemann”) was filed on
June 22, 2001 and is prior art to the ’379 Patent under 35 U.S.C. § 102(e) (pre-
AIA). Wesemann (EX1004). Wesemann was not cited during prosecution of the
‘379 Patent. See ’379 Patent (EX1001).
Wesemann is in the same field of endeavor as, and reasonably pertinent to,
the ’379 Patent. As mentioned above, the ’379 Patent relates to methods for
navigating a hierarchical system of nodes, exemplified in the context of a “menu-
IPR2017-01039 U.S. Patent No. 7,231,379
14
type automated telephone voice response system.” See ’379 Patent (EX1001) at
3:49-58, see also id. at 3:5-14. The ’379 Patent purports to solve problems related
to inefficiencies in navigating through nodes in hierarchical networks by allowing
users to “jump” to a node in a hierarchy without traversing other nodes. Id. at
Abstract, 2:22-30. Like the ’379 Patent, Wesemann relates to improving the
efficiency of navigating through menu-type hierarchy systems. See Wesemann
(EX1004) at 2:45-65 (identifying the inefficiency of “expend[ing ] time . . . to
move systematically through a hierarchy of levels or menu states . . . even when a
user already knows what the final menu state will be” as a problem in the art). And
Wesemann teaches solving this problem in the same manner as the ’379 Patent
purports to do, by teaching a system which enables users to “jump from one menu
state to another menu state of the telephone service system without having to enter
input for each menu state between the first and the second menus states.”
Wesemann (EX1004) at Abstract (emphasis added); see also id. at 3:54-56 (“[T]he
invention enables a user to jump over ‘in between’ menu states, from a first menu
state to a second menu state with only a single user input.” (emphasis added)), 3:6-
14 (emphasis added); cf. ’379 Patent (EX1001) at Abstract, 2:22-30. Therefore,
Wesemann is analogous prior art to the claimed invention of the ’379 Patent. See
Smyth Decl. (EX1007) at ¶47.
IPR2017-01039 U.S. Patent No. 7,231,379
15
i. Claim 1
1. A method performed in a system having multiple navigable nodes interconnected in a hierarchical arrangement comprising:
Wesemann teaches navigation methods performed in a system having
multiple navigable nodes, referred to as “menu states” or “levels,” interconnected
in a hierarchical arrangement:
The voice-enabled user interface maps the hierarchy of menu states
and corresponding prompts of the telephone service system within a
template. . . . . A user can jump from one menu state to another menu
state by providing input that the voice-enabled user interface associates
with a corresponding menu state. The voice-enabled user interface
generates output that causes the telephone service system to transition
to the menu state that corresponds with the user input. Once the
telephone service system is in the appropriate menu state, the voice-
enabled user interface transmits a DTMF translation of the user input to
the telephone service system so that it can be processed.
Wesemann (EX1004) at 3:33-46 (emphasis added); see also id. at Abstract, 10:40-
64, Figs. 5-6. The user may navigate the menu hierarchy by systematically moving
between nodes as prompted by the system, or the user may jump to different nodes
without transitioning through in-between menu states. See id. at 9:66-10:20; see
also id. at 1:13-19, 11:65-12:12, Fig. 5. A PHOSITA would understand that the
menu states in Wesemann are choices or options in the hierarchy, or “nodes.”.
Smyth Decl. (EX1007) at ¶48-49.
IPR2017-01039 U.S. Patent No. 7,231,379
16
Wesemann describes its teachings in an exemplary embodiment of a
computer sales organization telephone system. See id. at 10:40-64, 11:32-46,
11:65-12:42, Fig. 6. A caller may navigate through multiple interconnected menus,
such as support, sales, and a personnel directory. Id. at 10:40-45. These menus may
have their own submenus. For example, the sales menu has different submenus for
home, business, and refurbished computers, and each of these menus has a
submenu for laptops and desktops. Id. at Fig. 6. The menu and submenus constitute
multiple nodes interconnected in a hierarchical arrangement. See id. at 10:62-64
(“Main menu 610 and each of the submenus 620-674 comprise discrete menu
states of menu hierarchy 600.”). Wesemann’s computer sales organization menu is
similar in structure to the flight system menu example described in the ’379 Patent:
Wesemann:
Id. at Fig. 6.
IPR2017-01039 U.S. Patent No. 7,231,379
17
’379 Patent:
’379 Patent (EX1001) at Fig. 6. [1(a)] at a first node, receiving an input from a user of the system, the input containing at least one word identifiable with at least one keyword from among multiple keywords,
Wesemann teaches a system that receives a user input, such as spoken words
from a user, at a first “level” or “menu state” (node) of a hierarchical arrangement
of levels or menu states (nodes). Wesemann (EX1004) at Abstract, 3:28-30, 6:56-
64, 11:47-12:6. For example, the system may receive at a main menu state 610
spoken words from a user, such as “refurbished laptop sales.” Id. at 11:65-12:6. A
template, such as template 232, maps “acceptable responses and inputs”
(keywords) with each of the menu states (nodes) in the hierarchy. Id. at 7:15-17
IPR2017-01039 U.S. Patent No. 7,231,379
18
(“A template, such as template 232, maps all of the menu states of a telephone
menu system and all of the corresponding prompts and acceptable responses and
inputs.”), 8:56-63, 12:13-16. The user’s spoken input (at least one word) is
analyzed by speech recognition software and compared to the “acceptable
responses and inputs” (keywords) stored in the template to determine the node with
the accepted response that is most similar to the input. Id. at 6:56-64, 7:6-14,
12:13-21. This mapping of inputs to nodes allows the ability for user input to cause
the system to “jump” to the most appropriate node. Id. at 12:25-36.
The user responds to prompts generated by the telephone service
system by speaking into the telephone device. . . . The voice-enabled
user interface maps the hierarchy of menu states and
corresponding prompts of the telephone service system within a
template. . . . A user can jump from one menu state to another
menu state by providing input that the voice-enabled user interface
associates with a corresponding menu state. The voice enabled user
interface generates output that causes the telephone service system to
transition to the menu state that corresponds with the user input.
Id. at 3:28-43 (emphasis added); see also id. at 10:65-11:16, 4:51-54 (“Users
perform functions with a voicemail system by entering input into a telephone
device that can be recognized and processed by the voicemail system.”), 12:13-16
(“The voice enabled user interface is able to determine what menu prompts are
most similar to the input entered by the user because all of the menu prompts and
IPR2017-01039 U.S. Patent No. 7,231,379
19
acceptable inputs and requests are mapped in a template”). For example, when
user input in the form of the spoken words “refurbished laptop sales” is received at
a main menu node 610, the system identifies this as an accepted response
(keyword) for node 672 (refurbished laptop sales) and “jump[s] directly” to that
node, as illustrated below. Id. at 11:65-12:6.
Id. at Fig. 6 (annotated).
The “template” described in Wesemann, which maps accepted responses and
inputs (keywords) associated with each of the menu states (nodes) in the hierarchy,
is just like the ’379 Patent’s “index,” which maps keyword associations with
different nodes. See ’379 Patent (EX1001) at 5:2-4 (“An index . . . associating
these keywords with the nodes containing them is then created.”); 6:7-11 (“[W]hen
a response to a verbal description is provided by a user, possible keywords are
IPR2017-01039 U.S. Patent No. 7,231,379
20
identified in the response and used to search the index and identify any node to
which the response may be directed … .”); see also Smyth Decl., (EX1007) at ¶51.
A PHOSITA, would understand that the “acceptable inputs and requests” mapped
in the template to associated menu states are keywords (e.g., one or more words or
patterns of data associated with nodes) to which the spoken words in the user’s
input are compared in Wesemann. 2 Smyth Decl., (EX1007) at ¶50. Further,
Wesemann compares the user’s spoken words to the “acceptable inputs and
requests” stored in the template for an IVR telephone system just like the ’379
Patent compares a user’s spoken words to keywords stored in an index for an IVR
telephone system. Id. at ¶¶49-52. The “acceptable inputs and requests” in
Wesemeann, even if not represented as “words” per se, would be understood by a
PHOSITA to be patterns of data (consistent with the ’379 Patent) against which the
user’s spoken words are compared so as to determine whether those words are
sufficiently similar to any node to be an accepted input and to identify the most
similar node. Id. Further, at a minimum, a PHOSITA would have found it
exceedingly obvious for the “acceptable inputs and requests” mapped in the
template to associated menu states to be implemented as “keywords.” Id. A 2 The BRI of “keyword” at least includes “one or more words or pieces of information,
such as a data pattern, that is associated with at least one node or vertex.” See Sec.
III.D.iii, supra.
IPR2017-01039 U.S. Patent No. 7,231,379
21
PHOSITA would have been motivated to use keywords, because using keywords
to search directory trees had long been well known, including in most conventional
computer operating systems’ file storage systems—so it would have been natural
for a PHOSITA to use keywords, and a PHOSITA could have obtained predictable
results doing so without undue experimentation. Id. Further, Wesemann’s express
teaching of “acceptable inputs and responses” would have, at a minimum, at least
been highly suggestive to a PHOSITA of keywords and would have also motivated
a PHOSITA to use keywords. Id.
Wesemann also teaches that its keywords are identifiable from among
multiple keywords in the system, because there are a plurality of nodes in
Wesemann’s system that are mapped to user inputs via the template and the system
identifies the nodes that are most similar to the user input. Wesemann (EX1004) at
7:15-17, 8:56-63, 12:13-16, Fig. 6; see also Smyth Decl., (EX1007) at ¶¶49-52. To
the extent Patent Owner improperly contends that the “first node” must itself be
associated with multiple keywords, Wesemann nonetheless satisfies such an
interpretation. 3 Wesemann teaches that the template may associate “common 3 Claim 1 merely recites “from among multiple keywords” and does not recite that the
“first node” must itself be associated with multiple keywords from among which the
keyword identifiable with the word in the user’s input is identified. Such an interpretation
would be contrary to the ’379 specification, which provides numerous examples of nodes
IPR2017-01039 U.S. Patent No. 7,231,379
22
correlating inputs” with the acceptable inputs and requests, thus allowing multiple
keywords to be assigned to (i.e., associated with) a menu state:
The voice-enabled user interface is able to determine what menu
prompts are most similar to the input entered by the user because all of
the menu prompts and acceptable inputs and requests are mapped in a
template, as described above. A simple process of comparing the user
input with the accepted inputs and requests on file can be used to
determine what the user is attempting to do or say. Additional data
can also be stored in the template to associate common correlating
inputs with the acceptable inputs and requests. If it cannot be
determined what a user is attempting to input, the voice-enabled user
interface reverts to step 534 and prompts the user for new input.
Wesemann (EX1004) at 12:13-24. A PHOSITA would understand from this
disclosure that Wesemann teaches that nodes can store multiple acceptable
responses (keywords), i.e., multiple pieces of information or data patterns, that can
be compared to the user’s spoken words. Smyth Decl. (EX1007) at ¶54. Thus,
that each only have only one associated keyword. See, e.g., ’379 Patent (EX1001) at
4:37-41 (“For example, one node may have the associated verbal description “Would you
like to make a reservation?” In this description, there is only one “key” word—
“reservation” deemed important, so all other words in the description can be ignored.”),
7:12-17 (with each node 1-7 having only one associated keyword, even though multiple
nodes may share a keyword), 9:17-21.
IPR2017-01039 U.S. Patent No. 7,231,379
23
Wesemann, nonetheless teaches this interpretation of “from among multiple
keywords” in the event Patent Owner were to improperly attempt to construe it as
requiring the “first node” to have multiple associated keywords. Therefore,
Wesemann teaches limitation [1(a)].
[1(b)] identifying at least one node, other than the first node, that is not directly connected to the first node but is associated with the at least one keyword, and jumping to the at least one node.
Patent Owner has specifically identified the “jumping” concept in this
limitation, found in both claims 1 and 7, as the alleged “point of novelty”
distinguishing the ‘379 Patent over the prior art. See Patent Owner’s Opposition to
Alice Motion (EX1003) at 12-13. As discussed, the BRI of “jumping” at least
includes “a direct traversal from one node or vertex to another node or vertex that
is not directly connected to it (i.e., without traversal through any intervening nodes
or vertices or to a node or vertex whose only least common ancestor with that node
or vertex is the root node or vertex).” See supra Sec. III.D.iv,., Wesemann teaches
this precise concept of allowing a user to “jump” over “in between” menu states:
The present invention also enables users to navigate through a
hierarchy of menu states of a telephone service system without
requiring the users to enter input to transition through “in
between” menu states. Instead, the invention enables a user to jump
over “in between” menu states, from a first menu state to a second
menu state with only a single user input.
IPR2017-01039 U.S. Patent No. 7,231,379
24
Wesemann (EX1004) at 3:47-56 (emphasis added); see also id. at Abstract, 3:10-
14, 3:38-40, 8:1-5, Fig. 5.
Specifically, Wesemann teaches identifying at least one node, other than the
first node, that is not directly connected to the first node but that is associated with
the acceptable response/input, i.e., keyword. This can be accomplished because all
menu states (nodes) in the hierarchy are mapped in template 232 to acceptable
responses/inputs (keywords). See id. at 7:15-17 (“A template, such as template
232, maps all of the menu states of a telephone menu system and all of the
corresponding prompts and acceptable responses and inputs.”). Wesemann teaches
a system that is capable of jumping to an identified node having a keyword
(acceptable response/input) matching the user’s input by jumping both laterally
and/or vertically across the hierarchical network of nodes:
It should be appreciated that by mapping the menu hierarchy 600,
including all menu prompts associated with menu states of the menu
hierarchy 600, the invention enables a user to jump from one menu
state to another menu state without having to enter input for every
“in between” menu state. For example, from main menu 610 a user
can state “123” and the user will automatically be transferred to
extension 123. Alternatively, a user can jump from home laptop
sales 652 to home computer support 646 by speaking “home
computer support.” It is not necessary for the user to first return to the
main menu 610 and then transition down to home computer support
646. It should be appreciated that this is one advantage of the invention
IPR2017-01039 U.S. Patent No. 7,231,379
25
over the prior art. This advantage should be appreciated even more
fully when considering that, according to the present embodiment, the
telephone service system of menu hierarchy 600 cannot recognize
audio or vocal commands, but instead is only responsive to DTMF
signals.
Id. at 12:25-42 (emphasis added), Fig. 6 (annotated). In the above example, a user
can jump vertically from main menu 610 to extension “123” without traversing
through “directory of personnel,” of which extension “123” is a submenu.
Alternatively, the user may jump laterally between nodes, such as jumping from
“home laptop sales” to “home computer support” without traversing up and down
the parent nodes “sales,” “main menu,” and “computer support.” Unlike the prior
art cited during prosecution of the ‘379 Patent, Wesemann teaches that this
jumping action is done automatically upon receiving the appropriate voice
IPR2017-01039 U.S. Patent No. 7,231,379
26
commands, or keywords, without requiring the user to select different menu items
from a display or navigate through the hierarchical menu. See, e.g., id. at 12:30-32
(“For example, from main menu 610 a user can state “123” and the user will
automatically be transferred to extension 123. (emphasis added)); see also id. at
11:65-12:6, 12:65-13:2 (“It should be appreciated that all of the communications
associated with jumping from one menu state to another menu state are conducted
without the knowledge and efforts of the user . . . .”). As discussed above for
limitation [1(a)], the “jumping” in Wesemann is provided by mapping acceptable
user inputs (keywords) to menu states (nodes) to allow the user to “jump” to the
menu state that is most similar to the user input. See Wesemann as applied to
limitation [1(a)], supra; see also Wesemann (EX1001) at 12:12-36; see also Smyth
Decl. (EX1007) at ¶¶47, 55-56.
Wesemann, therefore, discloses each of the limitations of claim 1. See Smyth
Decl. (EX1007) at ¶¶48-56, 63.
ii. Claim 2
2. The method of claim 1, providing a verbal description associated with the at least one node to the user.
As described above, Wesemann teaches all limitations of claim 1. In
addition, Wesemann teaches providing a verbal description associated with the at
least one node to the user. The ’379 Patent defines a “verbal description” as “a set
of words relating to the subject matter whether presented audibly or in written
IPR2017-01039 U.S. Patent No. 7,231,379
27
form,” and this is the BRI of this term. Id. at 1:50-52; see also supra Sec. III.D.v.
A verbal description may consist of message prompts, such as “Would you like to
make a reservation?” Id. at 4:32-45.
Wesemann teaches a system which provides verbal descriptions in the form
of menu prompts corresponding to each of the particular menu states (nodes), and
these menu prompts are verbally presented to a user over a telephone device when
the user is at that particular menu state (node):
Voicemail system 140 comprises voicemail data 150 that can include
any type of data that can be transmitted to a user over a telephone
device. In one preferred embodiment, voicemail data 150 comprises a
prerecorded message, such as, for example, a menu prompt, a
telephone phone message, a greeting, etc.
Id. at 5:62-67 (emphasis added); see also id. at 3:28-20 (“The user responds to
prompts generated by the telephone service system by speaking into the telephone
device.”); 7:15-17 (“A template, such as template 232, maps all of the menu states
of a telephone menu system and all of the corresponding prompts and
acceptable responses and inputs.”) (emphasis added), 8:56-59 (“In one presently
preferred embodiment, template 232 contains a map of all the menu states of a
telephone service system and all the prompts corresponding to each of the
menu states.”) (emphasis added). For example, these verbal menu prompts may
include “press 1 followed by # for sales,” “press 2 followed by # for computer
IPR2017-01039 U.S. Patent No. 7,231,379
28
support,” “or press 3 followed by # for a directory of personnel.” Id. at 11:33-38.
As other examples, these prompts may also include prompts for “home computer
sales,” “business computer sales,” or “refurbished computer sales,” for each of
those particular nodes. Id. at Fig. 6, 11:65-12:6. A PHOSITA, therefore, would
understand that after a user jumps to a node, the user will be presented with the
menu prompt (verbal description) for that node. See Smyth Decl. (EX1007) at ¶57.
Further, Wesemann also teaches the use of a “clarifying prompt” that
provides a verbal description of multiple potential nodes in the event that a user
input could apply to more than one node:
Another reason why user input may be deemed improper is because a
user may be entering input which the user believes is correct, but which
is incorrect because the user has made one or mistakes while entering
the input. If this is the case, the voice-enabled user interface can
provide the user with prompts for a correct input that is most similar to
what the user entered. The following example is given to illustrate how
this might transpire.
If a user enters the main menu 610 of the telephone service system of
the computer sales organization with the desire to purchase a
refurbished laptop computer, the user can systematically enter input in
response to menu prompts to reach refurbished laptop sales 672 by
navigating through sales 630 and refurbished computer sales 670 to
arrive at refurbished laptop sales 672. Alternatively, according to the
invention, a user can jump directly to refurbished laptop sales 672 by
IPR2017-01039 U.S. Patent No. 7,231,379
29
speaking the words “refurbished laptop sales.” If the user mistakenly
says, “refurbished laptop computers,” or “refurbished notebook
sales,” the voice-enabled user interface prompts the user with a
clarifying prompt, such as “Would you like refurbished laptop
sales or refurbished desktop sales?” The user can then enter the
appropriate input “refurbished laptop sales.”
Wesemann (EX1004) at 11:57-12:12 (emphasis added). Wesemann, therefore,
teaches providing a verbal description associated with the at least one node to the
user, and, thus, teaches the limitation of claim 2. See Smyth Decl. (EX1007) at ¶57,
63.
iii. Claim 7
7. A method performed in connection with an arrangement of nodes representable as a hierarchical graph containing vertices and edges connecting at least two of the vertices, the method comprising …
The limitations of claim 7 are very similar to the limitations presented in
claims 1 and 2. Claim 7 refers to “vertices,” but as discussed above, the ’379
patent uses “vertices” and “nodes” interchangeably, and the BRI of a “vertex” is
simply “a specific choice or option in a hierarchy that can be represented in a
graph.” See Sec. III.D.i, supra. Wesemann teaches a hierarchical menu tree that can
be represented as a graph containing vertices and edges, where the edges connect
at least two of the vertices. For example, in Figure 6 of Wesemann, the menu tree
IPR2017-01039 U.S. Patent No. 7,231,379
30
displays the vertices for “Sales” connected by edges to each of “Home Computer
Sales, “Business Computer Sales,” and “Refurbished Computer Sales:
Wesemann (EX1004) at Fig. 6 (annotated). As shown above, other vertices are
also connected by other edges in Wesemann’s hierarchy.
[7(a)] receiving an input from a user as a response to a verbal description associated with a first vertex
As discussed above with respect to claim 2, Wesemann teaches a system
which provides verbal descriptions (“a set of words relating to the subject matter
whether presented audibly or in written form”) through menu prompts, each
IPR2017-01039 U.S. Patent No. 7,231,379
31
corresponding to a particular menu state, or vertex, and users provide inputs to the
telephone service system in response to those prompts:
The user responds to prompts generated by the telephone service
system by speaking into the telephone device. . . . The voice-enabled
user interface maps the hierarchy of menu states and corresponding
prompts of the telephone service system within a template.
Wesemann (EX1004) at 3:28-35 (emphasis added); see also id. at 5:62-67, 7:15-17,
8:56-59, 11:65-12:12 (“[T]he user can systematically enter input in response to
menu prompts. . . .”); see Wesemann as applied to claim 2, supra. The menu
prompt verbal descriptions in Wesemann are each associated with a particular
menu state (vertex). Id. For example, the system provides the prompt of asking the
user whether they want “computer support,” “sales,” or “a directory of personnel”
at the main menu node 610. Id. at 11:33-55, Fig. 6. A PHOSITA, therefore, would
understand that Wesemann teaches receiving an input from a user as a response to
a verbal description associated with a first vertex. Smyth Decl. (EX1007) at ¶¶58-
60, 63.
[7(b)] analyzing the input to identify a meaningful term that can be associated with at least one keyword Wesemann teaches this limitation in three different ways. First, as explained
above for claim 1, Wesemann teaches using speech recognition software to analyze
the user’s spoken words/terms and comparing them to acceptable responses/inputs
IPR2017-01039 U.S. Patent No. 7,231,379
32
(keywords) stored in the template, which maps those keywords to each of the menu
states. See supra Sec. IV.A.i, Wesemann as applied to claim [1(a)] & [1(b)].
Wesemann teaches that the speech recognition software 230 interprets audio or
voice data received from a user and a comparison is made to determine if the input
closely resembles an acceptable response/input (keyword) stored in the template.
See id. at 6:56-7:23, 8:37-63, 10:35-39, 10:65-11:16, 12:13-24, Fig. 2. A
PHOSITA would understand that this process identifies one or more meaningful
terms from the user’s spoken input that can be associated/compared with a
particular acceptable response/input (keyword) stored in the template for
identifying an associated node to which to jump. Smyth Decl. (EX1007) at ¶¶61,
51-53. For example, the user can speak the words “refurbished laptop sales” to
jump to the node for refurbished laptop sales (i.e., node 672) or the user can speak
the words “home laptop sales” to jump to the node for home laptop sales (i.e., node
652) in Wesemann’s menu tree. Id.; see also Wesemann (EX1001) at 12:4-6,
12:58-65; see also supra Section IV.A.i, limitation [1(a)]. Figure 5 demonstrates
the flow chart for analyzing an input received from a user in the Wesemann
system:
IPR2017-01039 U.S. Patent No. 7,231,379
33
Id. at Fig. 5.
Second, Wesemann also teaches that additional data may be stored in the
template so as to allow the system to associate “common correlating inputs” with
the acceptable inputs and requests to allow the system to better determine what a
user is attempting to input. Id. at 12:13-24. A PHOSITA would understand that this
process allows additional meaningful terms from the user’s spoken input to be
associated with an acceptable response/input (keyword) used to identify a node to
jump to. Smyth Decl. (EX1007) at ¶61.
IPR2017-01039 U.S. Patent No. 7,231,379
34
Third, Wesemann teaches providing a user with a clarifying prompt when
voice inputs may be attributable to more than one menu state stored in the
template:
If the user mistakenly says, “refurbished laptop computers,” or
“refurbished notebook sales,” the voice-enabled user interface prompts
the user with a clarifying prompt, such as “Would you like refurbished
laptop sales or refurbished desktop sales?” The user can then enter the
appropriate input “refurbished laptop sales.”
Id. at 12:7-12. In this example, a PHOSITA would understand that the system has
identified “refurbished” as a meaningful term associated with acceptable
responses/inputs (keywords) for the nodes “refurbished desktop sales” and
“refurbished laptop sales,” and that the system has requested an additional
meaningful term associated with a keyword for one of these nodes. Smyth Decl.
(EX1007) ¶61.
Finally, for the same reasons as discussed for claim 1, to the extent PO
argues that Wesemann somehow does not teach identifying a word/term with a
“keyword” per se, it would have nonetheless been obvious to do so. See supra Sec.
IV.A.i, limitation [1(a)]; see also Smyth Decl. (EX1007) at ¶¶61, 53.
Wesemann, therefore, teaches limitation [7(b)].
[7(c)] selecting a vertex in the graph structure that is not connected by an edge to the first vertex, based upon an association between the meaningful term and the at least one keyword and a correlation between the at least one keyword and the vertex; and jumping to the vertex.
IPR2017-01039 U.S. Patent No. 7,231,379
35
As described with respect to Claim [1(b)], Wesemann teaches this “jumping”
limitation [7(c)]. See supra Wesemann as applied to limitation [1(b)], Sec. IV.A.i.
As summarized below, Wesemann teaches each step of limitation [7(c)]:
(a) a vertex, referred to as a menu-state or level in Wesemann (e.g.,
extension “123” or “home computer support,” see Fig. 6 of Wesemann, is selected
to jump to;
(b) the vertex selected to jump to (e.g., “123” or “home computer support”)
is not connected by an edge to the first vertex (e.g., the first vertex is “main menu”
node 610 for the vertical jump to extension “123”; alternatively, the first vertex is
“home computer sales” 650 for the lateral jump to “home computer support” 646);
and
(c) the vertex to jump to is selected based on analysis of a meaningful term
in the input (e.g., a voice command containing terms such as “123” or “home
computer support”), associated with a keyword (i.e., the acceptable user
inputs/responses which are mapped to these menu states in the template),
(d) the system jumps to the vertex without requiring the user to traverse
through “in-between” vertices (e.g., transferring directly from “main menu” to
“123” or from “home computer sales” to “home computer support”):
IPR2017-01039 U.S. Patent No. 7,231,379
36
See Wesemann (EX1004) at Fig. 6 (annotated); see also id. at 12:25-42; see also id.
at Abstract, 3:10-14, 3:47-56, 3:38-40, 8:1-5, 9:65-11:16, 11:33-56, Fig. 5.
Thus, Wesemann teaches each limitation of claim 7. See also Smyth Decl.
(EX1007) at ¶¶58-63.
B. Ground 2: Wesemann in view of Rajaraman renders Claims 3-6 obvious
U.S. Patent No. 6,366,910 to Rajaraman et al. (“Rajaraman”) was filed on
December 7, 1998, and, therefore, constitutes prior art to the ’379 Patent under 35
U.S.C. § 102(e) (pre-AIA). Rajaraman is entitled “Method and System for
Generation of Hierarchical Search Results” and teaches a “general purpose search”
(GPS) method and system “that generates search results for items that are
hierarchically classified.” See Rajaraman (EX1005) at 2:9-22; see also id. at 1:66-
67 (“Embodiments of the present invention provide a method and system for
IPR2017-01039 U.S. Patent No. 7,231,379
37
querying hierarchically classified data.”), Abstract. Like the ‘379 Patent,
Rajaraman teaches a system whereby users may navigate different
nodes/classifications, referred to in Rajaraman as categories, sub-categories, and
item-types, arranged hierarchically:
Rajaraman (EX1005) at Fig. 4; see also id. at 6:25-35. As discussed above, the
‘379 Patent also relates to methods for navigating a hierarchical system of nodes,
exemplified in the context of a “menu-type automated telephone response system.”
See ‘379 Patent (EX1001) at 3:49-58, see also id. at 3:5-14, Fig. 1. Therefore,
Rajaraman is within the field of endeavor of the ‘379 Patent and constitutes
analogous prior art. See also Smyth Decl. (EX1007) at ¶¶64-66.
Rajaraman is also analogous art because it is reasonably pertinent to
problems addressed by the ’379 Patent. First, the ’379 Patent seeks, in part, to
IPR2017-01039 U.S. Patent No. 7,231,379
38
address the problem of users becoming discouraged or frustrated in having to
navigate a hierarchy of nodes or choices to find a goal of interest, and the ’379
Patent uses keyword searching across nodes to address this issue. ’379 Patent
(EX1001) at 2:9-18. Similarly, Rajaraman notes the similar problem of users
becoming potentially frustrated in having to navigate a hierarchy of
nodes/classifications to find an item of interest, and Rajaraman similarly allows
keyword searching across nodes/classifications to address this issue. Rajaraman
(EX1005) at 1:42-2:22. Second, Rajaraman is cited in this Petition with respect to
claims 3-6 of the ’379 Patent, which relate to the use of a thesaurus. The ‘379
Patent purportedly sought to improve the efficiency of navigational systems by
allegedly enabling them to “learn to incorporate previously unknown words,
keyword or synonyms of keywords.” ’379 Patent (EX1001) at 3:44-48. As
described in more detail below, Rajaraman also teaches a “special terms file” that
allows the system to process queries using synonyms for different
classifications/categories or items; for example, the system could produce results
relating to the “women’s shirts” classification when a user queries for a “blouse”
and vice versa. See Rajaraman (EX1005) at 7:22-8:25. Rajaraman also teaches a
“log analyzer” that assists with learning user’s meanings for previously unknown
terms, such as misspelled words, synonyms, or homonyms. See id. at 7:63-8:25.
IPR2017-01039 U.S. Patent No. 7,231,379
39
Rajaraman, therefore, is also analogous art because it is reasonably pertinent to
problems addressed by the ‘379 Patent.
i. Claims 3-4
3. The method of claim 1 further comprising searching a thesaurus correlating keywords with synonyms. 4. The method of claim 3 wherein the searching further comprises identifying the at least one word as synonymous with the at least one keyword. Claim 3 depends from claim 1 and adds the additional limitation of
“searching a thesaurus correlating keywords with synonyms,” while claim 4
depends from claim 3 and requires the searching further comprise identifying the at
least one word from the user’s input as synonymous with the at least one keyword.
As the claim itself makes clear, a thesaurus correlates keywords with synonyms.
The ’379 specification describes the thesaurus as “equating” synonyms with
keywords so that a synonym can also cause the system to jump to the desired node.
See ’379 Patent (EX1001) at 8:3-32. The specification further describes that “the
equating of terms can be done in any of a myriad of ways” (e.g., by correlating
words, correlating synonyms to nodes, etc.), but that such details are “irrelevant to
the invention.” Id. at 8:28-41.
As described above, Wesemann teaches Claim 1. In addition, Rajaraman
teaches a general purpose searching system for searching data grouped into
hierarchically classifications that uses a “special terms file” (thesaurus) which
IPR2017-01039 U.S. Patent No. 7,231,379
40
“lists various words (i.e., ‘Good Terms’) that are synonymous with the
classification names.” See id. at 7:22-26 (providing “blouse” as an example
special term file word for the classification “women’s shirts”) (emphasis added);
see also id. at Fig. 7. During a search, Rajaraman’s “index builder” accesses the
special terms file (thesaurus) and assigns synonymous terms (so-called “Good
Terms”) a priority of 0 (although a different priority may be used) so as to include
classifications having synonymous terms in the search results. Id. at 8:26-30 and
Fig. 9; see also id. at 9:7-45 and Fig. 11. This addition of synonyms to the index is
depicted in step 902 of Fig. 9:
IPR2017-01039 U.S. Patent No. 7,231,379
41
Id. at Fig. 9 (annotated). Therefore, Rajaraman teaches the limitation of claim 3.
See Smyth Decl. at ¶¶64-69.
IPR2017-01039 U.S. Patent No. 7,231,379
42
Further, with regard to claim 4, Rajaraman’s searching comprises
identifying at least one word input by the user, such as “blouse,” as synonymous
with at least one node’s classification, or keyword, such as “women’s shirts”
through implementation of the “Good Terms” file of synonyms:
Rajaraman (EX1005) at Fig. 7 (annotated); see also id. at 7:22-26.
IPR2017-01039 U.S. Patent No. 7,231,379
43
A PHOSITA would have been motivated to combine the menu-state system
in Wesemann that jumps to a menu state having an acceptable response (keyword)
corresponding to the user’s spoken words with Rajaraman’s teaching of searching
an index of classifications that includes a special terms file that assigns “good
terms” to different classifications, as taught in Rajaraman. See Smyth Decl.
(EX1007) at ¶76. Use of such thesaurus functionality to search synonyms in
addition to keywords was well known by that time, and a PHOSITA would have
appreciated that such functionality would beneficially provide users unaware of the
predetermined keywords greater flexibility and more user-friendly operation in
navigating the hierarchical menu system in Wesemann. Id. Both Rajaraman and
Wesemann relate to hierarchical menu systems that allow keyword searching, and
modifying Wesemann to include this functionality would have required minor
modifications known to a PHOSITA. Id. Further, Wesemann already teaches that
additional data could be stored in its template to associate “common correlating
inputs” with acceptable inputs and requests (keywords), so Wesemann expressly
recognizes that additional data stored in the template to expand the scope/reach of
the keywords would be helpful to users. See Wesemann (EX1004) at 12:19-21. A
PHOSITA would have been motivated by this express teaching in Wesemann to
incorporate thesaurus functionality like that in Rajaraman. Smyth Decl. (EX1007)
IPR2017-01039 U.S. Patent No. 7,231,379
44
at ¶76. Further, a PHOSITA would have recognized that the application of known
methods related to searching keyword synonyms contained in a thesaurus would
yield predictable results when applied in the system of Wesemann and would not
render any features of Wesemann, such as the ability to jump between nodes,
inoperable. Id. Therefore, claims 3-4 are obvious over Wesemann in view of
Rajaraman.
ii. Claims 5-6
[5(a)] The method of claim 1 further comprising determining that the at least one word is neither a keyword nor a synonym of any keyword; and
As described above, Wesemann teaches claim 1. Rajaraman teaches that its
general purpose searching system determines that a word queried by a user is
neither a keyword nor a synonym of a keyword:
In one embodiment, the GPS system logs search requests along with
the search results and may also log which search results (i.e.,
classifications) are selected by the user. Periodically, these logs can be
analyzed to determine whether synonyms should be added for a search
term. For example, users may enter the search term “aparel,” rather
than “apparel.” Because the term “aparel” is not in the product
database and not in the classification hierarchy, the search result
will be empty. Therefore, it would be useful to add the term “aparel” as
a synonym of “apparel.” The GPS system provides a log analyzer to
help determine when to add synonyms. In one embodiment, the log
analyzer identifies the search requests that resulted in no search
IPR2017-01039 U.S. Patent No. 7,231,379
45
results or in very few classifications in the search results and
displays the identified search requests to an analyst responsible for
deciding on synonyms.
Rajaraman (EX1005) at 7:63-8:11 (emphasis added).
[5(b)] learning a meaning for the word so that the word will be treated as a learned synonym for at least one particular keyword of the multiple keywords. 6. The method of claim 5 further comprising adding the word to a thesaurus so that, when the word is input by a subsequent user, the word will be treated as synonymous with the at least one particular keyword. Rajaraman teaches that its search system includes a log analyzer to allow
addition to the thesaurus of common but previously unmapped query terms as
learned synonyms to search terms, based on analysis of users’ search behaviors:
In one embodiment, the GPS system logs search requests along with
the search results and may also log which search results (i.e.,
classifications) are selected by the user. Periodically, these logs can be
analyzed to determine whether synonyms should be added for a
search term. For example, users may enter the search term “aparel,”
rather than “apparel.” Because the term “aparel” is not in the product
database and not in the classification hierarchy, the search result will be
empty. Therefore, it would be useful to add the term “aparel” as a
synonym of “apparel.” The GPS system provides a log analyzer to
help determine when to add synonyms. In one embodiment, the log
analyzer identifies the search requests that resulted in no search results
or in very few classifications in the search results and displays the
identified search requests to an analyst responsible for deciding on
IPR2017-01039 U.S. Patent No. 7,231,379
46
synonyms. For example, the terms of the identified search requests
can be displayed along with a field so that the analyst can enter the
word(s) with which the displayed search term is synonymous. The
log analyzer may also display statistical information as to how
many times the displayed search term was entered by a user. Also,
the log analyzer may display additional information such as a
subsequent search request entered by the same user that does
return search results.
Id. at 7:63-8:19. Therefore, Rajaraman teaches the limitations of claims 5-6.4
It would have been obvious to a PHOSITA to enable the hierarchical
navigation system taught in Wesemann to learn new synonyms for keywords, as
taught by Rajaraman. Smyth Decl. (EX1007) at ¶76. As discussed for claims 3-4, it
would have been obvious to combine Rajaraman’s thesaurus functionality with
Wesemann’s system; and, it would have further been obvious to allow the
thesaurus to be updated by learning new synonyms for keywords because such
functionality was already well known in the art. Id. A PHOSITA would have
appreciated that allowing the thesaurus of synonyms for keywords to be updated
would have improved the system by making it even more user-friendly and by
further enhancing the efficiency with which a user can navigate the hierarchy of
nodes. Id. A PHOSITA would have further recognized that such an application in 4 See the discussion immediately below within this section for discussion regarding the
“subsequent user” limitation of claim 6.
IPR2017-01039 U.S. Patent No. 7,231,379
47
the system of Wesemann would have required minor modifications at most and
yielded predictable results. Id. A PHOSITA would have also been motivated to
make this combination based on Wesemann’s express teaching that “common
correlating inputs” could be added to its template. See id.; see also Wesemann, at
12:19-21. As such, Wesemann in view of Rajaraman renders claims 5-6 obvious.
Id.
Petitioner notes that claim 5 merely recites “learning a meaning for the word
so that the word will be treated as a learned synonym for at least one particular
keyword” and does not expressly recite “automatically learning” such a synonym
with no user involvement. Thus, a “learning” step that may involve some human
input to cause the “learning” to occur is still within the ambit of claim 5.
Patent Owner may conceivably try to argue a narrow interpretation of claim
5 that no human involvement can be involved in the claimed “learning” and that
the “analyst” mentioned in the paragraph cited above is a human, not part of the
system of Rajaraman. Such arguments are inapposite. First, “it is well settled that
it is not ‘invention’ to broadly provide a mechanical or automatic means to replace
manual activity which has accomplished the same result.” In Re Venner, 262 F.2d
91, 95 (C.C.P.A. 1958); see also, e.g., Ex Parte Brent Bursey, Appeal 2014-
003565, 2016 WL 4579139 at *3 (P.T.A.B. Aug. 26, 2016) (Decision on Appeal)
(“Rendering automatic by computer that which is done by hand per se would have
IPR2017-01039 U.S. Patent No. 7,231,379
48
been obvious to one of ordinary skill in the art.”); Western Union Co. v.
MoneyGram Payment Systems, Inc., 626 F.3d 1361, 1370-71 (Fed. Cir. 2010)
(finding that the use of electronic transaction device where prior art employed fax
machine was obvious, rejecting argument that the patentee's “effort … invested in
software implementation of its commercial system” tended to show that the patent
was inventive.). Merely replacing the manual activity of a human with software
logic that would allow the log analyzer to perform the broadly-recited learning step
of claim 5 would be obvious over Rajaraman. Second, a PHOSITA at the time of
the ’379 Patent would have found it obvious to automate the learning step in
Rajaraman using computer algorithms for automatic thesaurus updating, because
such algorithms were already well known. See Smyth Decl. (EX1007) at ¶74, see
also id. at ¶¶70-75, 41-45. A PHOSITA would have appreciated the benefit in
allowing automatic addition of synonyms for keywords to avoid the well-known
issues with labor-intensive manual updating of keyword thesauruses. Id. Therefore,
Wesemann in view of Rajaraman nonetheless renders claims 5-6 obvious even
under this alternative interpretation of claim 5. See id.
Petitioner notes that claim 6 merely recites “a subsequent user of the
system” and does not expressly recite that this subsequent user must be different
from the first user of the system. Further, the ’379 specification indicates that a
IPR2017-01039 U.S. Patent No. 7,231,379
49
subsequent user that benefits from the updated thesaurus may be either the same
user subsequently using the system or a different user:
[I]t tracks what the user did, what node of the tree the user went to, and
on this basis, it learns a new response to “coach class”. The next time a
caller (either the same person or a different person) uses the words
“coach class” the system does not offer the traditional path as it did the
first time, but instead it offers a new set of nodes based on what it
learned the first time.
’379 Patent (EX1001) at 14:32-38 (emphasis added). In the event PO argues that
“a subsequent user” must be a different person, Rajaraman’s thesaurus
functionality nonetheless satisfies such an interpretation, because Rajaraman does
not limit its thesaurus functionality to applying to only a specific user, but instead
bases it on the behavior of multiple users and even on homonyms and alternate
spellings:
For example, users may enter the search term “aparel,” rather than
“apparel.” Because the term “aparel” is not in the product database and
not in the classification hierarchy, the search result will be empty.
Therefore, it would be useful to add the term “aparel” as a synonym of
“apparel.”
Id. at 7:67-8:5 (emphasis added); see also id. at 8:22-25 (“For example, if users
enter the search request ‘sole’ and the search results relate only to shoes, the
analyst may want to indicate that ‘sole’ is a synonym for ‘soul,’ as in music.”
(emphasis added)). Further, Rajarman, teaches that its thesaurus (i.e., “the special
IPR2017-01039 U.S. Patent No. 7,231,379
50
terms file”) is a component of its overall system, which is used by all users of the
system, and a PHOSITA, therefore, would understand that Rajaraman is not
limiting its thesaurus to only a particular user, and, instead, subsequent users also
benefit from the thesaurus of synonyms, which is based on multiple users’ search
behvaiors. Id. at 4:55-5:9, Fig. 2, 7:22-42, Fig. 7; see also Smyth Decl. (EX1007) at
¶75. It also would have been obvious to a PHOSITA to have the thesaurus apply to
other subsequent users because it was well known in the art that thesauruses of
synonyms are improved when they are based on the learned search behaviors of
multiple users, which yields a more robust thesaurus functionality. Smyth Decl.
(EX1007) at ¶75; see also id. at ¶44. Thus, Wesemann in view of Rajaraman
renders claim 6 obvious even under such a “multiple user” interpretation.
Therefore, claims 5-6 are obvious over Wesemann in view of Rajaraman. Id.
at ¶¶70-76.
C. Ground 3: Fratkina renders claims 1, 2, and 7 obvious U.S. Patent No. 7,539,656 to Fratkina et al. (“Fratkina”) was filed on March
6, 2001, and therefore is prior art to the ’379 Patent under 35 U.S.C. § 102(e) (pre-
AIA). Fratkina (EX1006). Fratkina was not cited during prosecution of the ‘379
Patent. See ‘379 Patent (EX1001).
As mentioned above, the ’379 Patent relates to methods for navigating a
hierarchical system of nodes, exemplified in the context of a “menu-type
IPR2017-01039 U.S. Patent No. 7,231,379
51
automated telephone response system.” See ’379 Patent (EX1001) at 3:49-58, see
also id. at 3:5-14. The ’379 Patent purports to solve problems related to
inefficiencies in navigating through hierarchical decisional networks by allowing
users to “jump” to a node in a hierarchy without traversing other nodes. Id. at
Abstract, 2:22-30. Like the ’379 Patent, Fratkina teaches improving the efficiency
of navigating through menu-type hierarchy systems by enabling users to “jump” to
different “subgoals.” See Fratkina (EX1006) at 34:32-53; see also id. at 27:25-43;
see also id. at 37:54-63 and Fig. 15 (pointing dialog to a correct subgoal focus if a
user inputs a known menu item). Also like the ’379 Patent’s “IVR system”
embodiments, Fratkina teaches embodiments implemented using “an interactive
voice response (IVR) and/or speech recognition system” that recognizes spoken
words input by the user. See ’379 Patent (EX1001) at 6:66-67; Fratkina (EX1006)
at Abstract, 2:56-62, 13:15-29. Because Fratkina falls within the field of
navigating hierarchical systems and relates to solving problems related to the
efficiency of such navigational systems—including doing so by allowing users to
jump to different goals—Fratkina is both in the field of endeavor of and
reasonably pertinent to the ’379 Patent. See also Smyth Decl. (EX1007) at ¶¶77-79.
Therefore, Fratkina qualifies as analogous prior art to the claimed invention of the
’379 Patent.
IPR2017-01039 U.S. Patent No. 7,231,379
52
i. Claim 1
[1(P)] A method performed in a system having multiple navigable nodes interconnected in a hierarchical arrangement comprising:
Fratkina teaches the use of taxonomies, arranged hierarchically, in which
interconnected nodes may be navigated by a user:
Fratkina at Fig. 4. Fratkina describes its taxonomies as hierarchal arrangements of
nodes:
As shown, taxonomy 30 consists of a plurality of root nodes 300, a
plurality of concept nodes 310 coupled together by a plurality of edges
320. Each node 300, 310 in a taxonomy expresses a “concept,” or a
classification to which content and resources can be assigned. The set
of nodes 300, 310 for each taxonomy is created to model the
IPR2017-01039 U.S. Patent No. 7,231,379
53
taxonomy’s area of concern at an appropriate level for distinguishing
among the correct knowledge containers to return: neither too coarse a
representation which fails to differentiate among many knowledge
containers, nor too granular a representation which models more
distinctions than really exist among available knowledge containers.
See Fratkina, at 14:47-59 see also id. at 4:42-5:19, Fig. 5.
Fratkina teaches methods by which users may navigate through such
taxonomies, such as a meal menu:
Fratkina (EX1006) at Fig. 10; see also id. at 22:9-29 (describing navigating
through the menu through the use of “concept triggers”), 22:55-23:2 (describing
IPR2017-01039 U.S. Patent No. 7,231,379
54
navigating through “focus nodes”), Figs. 11-12, 20. Thus, to the extent the
preamble is limiting, Fratkina discloses a system having multiple navigable nodes
interconnected in a hierarchical arrangement. See Smyth Decl. (EX1007) at ¶80.
[1(a)] at a first node, receiving an input from a user of the system, the input containing at least one word identifiable with at least one keyword from among multiple keywords,
Fratkina teaches a system whereby the system receives input from a user
containing words at a first node (such as, for example, at a “Meal Type” node,
“Menu” node, or other node in a restaurant taxonomy for ordering food):
In some embodiments, a similar type of interaction can be achieved
vocally via an interactive voice response (IVR) type of system. In
these embodiments, the user speaks their requests and responses
into telephone 4 or other microphone and may also provide other
input by pressing buttons (e.g. buttons on the telephone's keypad).
The user's spoken responses are passed to a voice recognition system,
which turns the responses into data that dialog engine 232 can process.
The dialog engine 232 response is passed to a text-to-speech system
that turns it into a vocal response to the user.
Fratkina (EX1006) at 13:15-24 (emphasis added); see also id. at 13:25-39, 22:19-
29 (“[A] user states that they want breakfast . . .”), Fig. 10, 26:36-57 (“Breakfast”
node), Fig. 11, 34:9-53 (“Diet” node and “Menu” node); Fig. 12. Fratkina also
teaches that these inputs contain at least one word identifiable with at least one
keyword from among multiple keywords. For example, a user may use “keyword
IPR2017-01039 U.S. Patent No. 7,231,379
55
or natural language” queries to request actions or make choices, and a “dialog
engine” converts the user’s input into tags to be processed by the system using an
“autocontextualization” process. See id. at 5:13-28, 5:58-8:10 (describing various
query types), 14:27-31. The system uses the inputs to traverse the taxonomy:
The process of goal resolution is one of taxonomy traversal, going from
more general concepts at the root to more specific concepts closer to
the leaves of the taxonomy graph. Advancing a dialog from one node to
the next advances the focus of the goal, it causes the triggers associated
with each node to be evaluated and it causes constraints to be
generated. A goal is considered resolved when a target set of concept
nodes is reached during a traversal. When dialog engine 232 receives
an answer from a user, it identifies one or more goals for the
dialog, based on the dialog engine’s current understanding of the
user’s response. Dialog engine 232 improves its understanding of the
user’s initial question by conducting a multi-step dialog with the user.
Based on the user’s responses to follow-up questions, the dialog engine
is further able to focus its analysis of the user’s response (i.e., limit the
remaining sets of concepts by creating constraints). In other words,
dialog engine 232 seeks to describe the user’s information request in
more and more detail by mapping user responses to concept nodes in
the goal taxonomy.
See id. at 26:26-45 (emphasis added); see also id. at 14:27-31 (“Each taxonomy tag
60 includes the name or other unique identifier of a concept node within a
taxonomy 30 optionally followed by a number, typically between 0 and 1, which
IPR2017-01039 U.S. Patent No. 7,231,379
56
indicates the knowledge container's strength of association with that concept
node.”) (emphasis added), 26:46-27:27. A PHOSITA would understand that
Fratkina teaches a system where, at a first node, a user may input a
command/response containing one or more words identifiable with at least one or
more keywords for navigating the taxonomies. See Smyth Decl., (EX1007) at ¶¶80-
83.
The teachings of Fratkina can be further understood by way of its examples.
Figures 10-11 of Fratkina depict a menu hierarchy where “a user states that they
want breakfast”:
Fratkina (EX1006) at Fig. 10, 22:9-29.
US. Patent May 26, 2009 Sheet 10 6f 19 US 7,539,656 B2
1030 10540 1010 creoteGool
@@ node=breokf0st; @ ;
SCRAMBLED
@ POACHED BENEDICT
BREAKFAST
WITH SYRUP 1020 PANCAKES
CE} SYRUP @@ FIG. 10
IPR2017-01039 Unified EX1006 Page 13
IPR2017-01039 U.S. Patent No. 7,231,379
57
Id. at Fig. 11.
As shown in FIG. 11, the subsequent selection of a new concept node
by dialog engine 232 proceeds as the user answers questions posed by
the dialog engine. As shown at 1110, when the user answers “eggs” in
response to the question “Which of the following would you like to
get,” the goal of the dialog proceeds from the “Breakfast” node to the
“eggs” node. Similarly, in 1120 when the user answers “scrambled” in
response to the question “How would you like your eggs prepared,”
IPR2017-01039 U.S. Patent No. 7,231,379
58
the goal of the dialog proceeds from the “eggs” node to the
“scrambled” node 1130.
Id. at 26:50-57; see also id. at Fig. 10, 22:19-29.
Here, Fratkina provides an example in which a user moves systematically
through the menu hierarchy, providing input, where the input includes at least one
word (e.g., “eggs”) associated with a keyword (e.g., “eggs”) from among multiple
keywords (e.g., “eggs” and “pancakes”). Therefore, Fratkina teaches this
limitation. See Smyth Decl. (EX1007) at ¶¶81-83.
[1(b)] identifying at least one node, other than the first node, that is not directly connected to the first node but is associated with the at least one keyword, and jumping to the at least one node. Fratkina teaches jumping directly from one node to another that is not
directly connected without traversing intervening nodes through
“autocontextualization”:
As mentioned above, dialog engine 232 generates follow-up questions
to the user in the process of resolving active goals. The type of question
to be generated can be specified by the dialog designer. The answers to
these questions advance the state of the subgoal to a new location in the
taxonomy—e.g. change the focus of the subgoal. Changing the focus of
a subgoal may be by path traversal within the knowledge map (e.g., the
focus may change from parent node to child node).
Autocontextualization can be used to jump to a specific place in the
taxonomy and the dialog designer can explicitly specify a place to
jump to. In any event, the node-to-node path taken by dialog engine
IPR2017-01039 U.S. Patent No. 7,231,379
59
232 from the focus node(s) to the target node(s) heavily depends on
user responses to system questions.
Fratkina (EX1006) at 34:32-53 (emphasis added); see also id. at 27:25-43 (“In the
process of goal resolution, the focus is typically advanced along the edges of the
taxonomy graph, but may jump to a node (or set of nodes) more than one edge
away from the previous focus.”) (emphasis added), 37:54-63.
Therefore, using the menu example, a user may move through the menu
hierarchy directly by going from “breakfast” to “eggs” to “scrambled,” as
described above. However, Fratkina’s system may also jump to a specific place
depending on the user’s response to the system’s questions. Thus, if a user was at
the “breakfast node” (first node) and desired scrambled eggs, he may simply say he
wants “scrambled” eggs after being asked what he wanted for breakfast, which
would be associated with a node not directly connected to the “breakfast” node;
and a PHOSITA would have found it obvious that, using autocontextualization, the
dialog designer would be able to identify the “scrambled” node as the place the
user desired to go based on his response and “jump” directly to that node, without
requiring the user to first traverse through the “eggs” node. Id.; see also id. at 34:9-
53, Fig. 12; Smyth Decl. (EX1007) at ¶84. Fratkina’s express teachings regarding
use of autocontextualization to “jump” to non-connected nodes would have
motivated a PHOSITA to allow such jumping within the menu hierarchy to
IPR2017-01039 U.S. Patent No. 7,231,379
60
provide for a more natural/conversational and efficient flow in Fratkina’s system.
Id.
Therefore, a PHOSITA would understand Fratkina to teach, or at a
minimum, suggest, the use of the one or more keywords to jump directly to a node
within the taxonomy that is associated with the keyword without traversing
intervening nodes. See Smyth Decl. (EX1007) at ¶¶80-84.
ii. Claim 2
2. The method of claim 1, providing a verbal description associated with the at least one node to the user.
The ’379 Patent defines a “verbal description” as “a set of words relating to
the subject matter whether presented audibly or in written form,” and this is the
BRI of this term. ’379 Patent (EX1001) at 1:50-52; see also Sec. III.D.v, supra. A
verbal description may consist of message prompts, such as “Would you like to
make a reservation?” Id. at 4:32-45. Fratkina teaches that its system may provide a
verbal prompt to a user when she reaches a node:
FIG. 11 further illustrates the goal resolution process in accordance
with an embodiment of the present invention. As shown in FIG. 11, the
subsequent selection of a new concept node by dialog engine 232
proceeds as the user answers questions posed by the dialog engine. As
shown at 1110, when the user answers “eggs” in response to the
question “Which of the following would you like to get,” the goal of
the dialog proceeds from the “Breakfast” node to the “eggs” node.
IPR2017-01039 U.S. Patent No. 7,231,379
61
Similarly, in 1120 when the user answers “scrambled” in response
to the question “How would you like your eggs prepared,” the goal
of the dialog proceeds from the “eggs” node to the “scrambled”
node 1130. In this example, the nodes selected are confirmed nodes
since they represent nodes whose relevance to the user's information
need has been established.
Id. at 26:46-60. (emphasis added), Fig. 11, Fig. 21; see also id. at 13:15-24 (“The
dialog engine 232 response is passed to a text-to-speech system that turns it into a
vocal response to the user.”), 26:34-45, 37:12-30. Therefore, Fratkina teaches
claim 2. See also Smyth Decl. (EX1007), at ¶85.
iii. Claim 7
7. A method performed in connection with an arrangement of nodes representable as a hierarchical graph containing vertices and edges connecting at least two of the vertices, the method comprising …
The limitations of claim 7 are very similar to the limitations presented in
claims 1 and 2. Claim 7 refers to “vertices,” but as discussed above, the ‘379
patent uses “vertices” and “nodes” interchangeably, and the BRI of a “vertex” is
simply “a specific choice or option in a hierarchy that can be represented in a
graph.” See Sec. III.D.i, supra. Fratkina relates to a hierarchical network that can
be represented as a graphical menu tree containing vertices and edges, where the
edges connect at least two of the vertices. For example, in Figures 10-12 of
Fratkina, an embodiment featuring a meal-type taxonomy is displayed in a
IPR2017-01039 U.S. Patent No. 7,231,379
62
graphical arrangement. Referring to Figure 10, depicted below, the vertices
“breakfast” and “eggs” are connected by an edge:
Fratkina (EX1006) at Fig. 10 (annotated); see also id. at Figs 4, 11, 12; see also
Smyth Decl. (EX1007) at ¶¶86-87.
[7(a)] receiving an input from a user as a response to a verbal description associated with a first vertex
As discussed with respect to claim 2, Fratkina teaches a system that
provides verbal descriptions as part of a “multi-step dialog” including “follow-up
questions” (i.e., menu prompts), corresponding to a particular node, or vertex, and
users input responses to the questions accordingly:
IPR2017-01039 U.S. Patent No. 7,231,379
63
Dialog engine 232 improves its understanding of the user’s initial
question by conducting a multi-step dialog with the user. Based on the
user’s responses to follow-up questions, the dialog engine is further
able to focus its analysis of the user’s response (i.e., limit the remaining
sets of concepts by creating constraints) .
FIG. 11 further illustrates the goal resolution process in accordance
with an embodiment of the present invention. As shown in FIG. 11, the
subsequent selection of a new concept node by dialog engine 232
proceeds as the user answers questions posed by the dialog engine. As
shown at 1110, when the user answers “eggs” in response to the
question “Which of the following would you like to get,” the goal of
the dialog proceeds from the “Breakfast” node to the “eggs” node.
Similarly, in 1120 when the user answers “scrambled” in response
to the question “How would you like your eggs prepared,” the goal
of the dialog proceeds from the “eggs” node to the “scrambled”
node 1130.
Id. at 26:34-60. (emphasis added); see also id. at 13:15-24; 37:12-30; see also
Smyth Decl. (EX1007) at ¶88.
[7(b)] analyzing the input to identify a meaningful term that can be associated with at least one keyword; Fratkina teaches that its system analyzes user inputs to identify meaningful
terms that can be associated with one or more keywords associated with different
nodes in order to resolve a user’s “goals”:
Once dialog engine 232 creates an initial set of goals (target nodes and
focus nodes) based on the subject matter of the user’s inquiry, it begins
IPR2017-01039 U.S. Patent No. 7,231,379
64
to resolve the goals. The process of goal resolution is one of taxonomy
traversal, going from more general concepts at the root to more specific
concepts closer to the leaves of the taxonomy graph. . . . When dialog
engine 232 receives an answer from a user, it identifies one or more
goals for the dialog, based on the dialog engine's current
understanding of the user's response. Dialog engine 232 improves
its understanding of the user’s initial question by conducting a
multi-step dialog with the user. Based on the user’s responses to
follow-up questions, the dialog engine is further able to focus its
analysis of the user’s response (i.e., limit the remaining sets of concepts
by creating constraints). In other words, dialog engine 232 seeks to
describe the user's information request in more and more detail by
mapping user responses to concept nodes in the goal taxonomy.
Fratkina (EX1006) at 26:24-45 (emphasis added); see also id. at 5:13-28
(“keywords”), 13:15-24, 34:32-53; see also Fratkina applied to claim [1(a)] and
[1(b0] supra. For example, the user speaking/inputting a meaningful term (e.g.,
“eggs”) can be associated with a keyword (e.g., “eggs”). Id. at 26:50-57; see also
id. at Fig. 10, 22:19-29. A PHOSITA would understand that Fratkina teaches a
system in which a user may input a command containing one or more words that
may identifiable with at least one or more keywords for navigating the taxonomies.
See Smyth Decl., (EX1007) at ¶88.
[7(c)] selecting a vertex in the graph structure that is not connected by an edge to the first vertex, based upon an association between the meaningful term and the at least one keyword and a correlation between the at least one keyword and the vertex; and jumping to the vertex.
IPR2017-01039 U.S. Patent No. 7,231,379
65
As described with respect to claim [1(b)] above Fratkina teaches this
“jumping” limitation as it has been claimed in the ‘379 Patent, and, at a minimum,
it would have been obvious to implement such jumping in the context of the meal
menu example. See supra Sec. IV.C.i. As discussed, Fratkina teaches that
depending on the user’s responses, the system may “autocontextualize” the user’s
response and “jump” to a specific place in the taxonomy. Fratkina (EX1006) at
34:32-53 (“Autocontextualization can be used to jump to a specific place in the
taxonomy and the dialog designer can explicitly specify a place to jump to.”); see
also id. at 27:25-43 (“In the process of goal resolution, the focus is typically
advanced along the edges of the taxonomy graph, but may jump to a node (or set of
nodes) more than one edge away from the previous focus.”); see also id. at 37:54-
63. Using the meal hierarchy example, a PHOSITA would have understood
Fratkina’s system to be capable of navigating a hierarchical network consistent
with this limitation and it would have been obvious to do so (as discussed above
for limitation [1(b)], see supra Sec. IV.C.i):
(a) the user begins at the “breakfast” vertex;
(b) the vertex selected to jump to is not connected by an edge to the first
vertex (e.g., “scrambled” does not share an edge with “breakfast”);
IPR2017-01039 U.S. Patent No. 7,231,379
66
(c) the vertex to jump to is selected based on analysis of a meaningful term
(e.g., a user input of “scrambled eggs”) associated with a keyword associated with
the second vertex (e.g., “scrambled”)), and
(d) the system jumps directly to the second vertex using
autocontextualization (e.g., transferring directly from “breakfast” to “scrambled,”
without stopping at the intermediate level containing the “eggs” vertex).
See, e.g., id. at 26:34-60, Fig. 11; Smyth Decl., (EX1007) at ¶89.
Given the above, claims 1, 2 and 7 are obvious over Fratkina. Smyth Decl.,
(EX1007) at ¶¶77-90.
D. Ground 4: Fratkina in view of Rajaraman renders Claims 3-6 obvious
As discussed above, Rajaraman is cited in this Petition with respect to
claims 3, 4, 5, and 6, which each relate to the use of thesaurus functionality. See
Sec. IV.B, supra. As discussed previously, Rajaraman teaches a general purpose
searching system that uses a “special terms file” (thesaurus) for “good terms,”
which are designated as being synonymous to classification terms, and the system
includes a log analyzer that facilitates addition of new synonyms. See Rajaraman
(EX1005) at 7:22-8:53, 9:7-20; see also id. at Figs. 7, 9, and 11. Rajaraman
teaches and renders obvious each of the limitations of claims 3-6, and Petitioner
incorporates this discussion above by reference. See supra Sec. IV.B.
IPR2017-01039 U.S. Patent No. 7,231,379
67
Fratkina expressly teaches that user inputs can be processed using a
“thesaurus”:
Text disambiguation queries (DAQs): Identified by characteristics of
the user’s text, such as misspellings or ambiguity of words in the text
(as defined by an external machine-readable dictionary, thesaurus, or
lexicon, e.g., WordNet), or by a lack of information in the
autocontextualization engine about words in the text.
Fratkina (EX100X) at 7:47-52. And Fratkina expressly teaches that “[its] dialog
engine is designed to work with any prior art search/retrieval engines to produce a
search space.” Id. at 8:51-60. These express teachings would have motivated a
PHOSITA to combine Fratkina with the thesaurus functionality of Rajaraman’s
search engine provided by the special terms file that assigns “good terms” to
different classifications, as well as its log analyzer for adding new synonyms. See
Smyth Decl. (EX1007) at ¶¶91-92. A PHOSITA would have appreciated that such
functionality would beneficially provide users unaware of the predetermined
search terms greater flexibility and user-friendliness in navigating the taxonomies
in Fratkina. Id. Both Rajaraman and Fratkina relate to hierarchical menu systems
that allow keyword searching, and modifying Fratkina to include this functionality
would have required minor modifications known to a PHOSITA. Id. Further, a
PHOSITA would have recognized that the application of known methods related to
searching keyword synonyms contained in a thesaurus (which was a well-known
IPR2017-01039 U.S. Patent No. 7,231,379
68
functionality) would yield predictable results when applied in the system of
Fratkina and would not render any features of Fratkina, such as the ability to jump
between nodes, inoperable. Id.
Therefore, claims 3-6 are obvious over Fratkina in view of Rajaraman. Id.
V. CONCLUSION
For the forgoing reasons, Petitioner respectfully requests inter partes review
of claims 1-7 of the ’379 Patent.
Respectfully submitted, BY: /s/ Jason R. Mudd_______ Jason R. Mudd, Reg. No. 57,700
Eric A. Buresh, Reg. No. 50,394 Jonathan Stroud, Reg. No. 72,518 Roshan Mansinghani, Reg. No. 62,429
ATTORNEYS FOR PETITIONER
IPR2017-01039 U.S. Patent No. 7,231,379
69
VI. MANDATORY NOTICES UNDER 37 C.F.R. § 42.8(A)(1)
A. Real Party-In-Interest
The Petitioner is the real party-in-interest. 37 C.F.R. § 42.8(b)(1). No other
party exercised control or could exercise control over Petitioner’s participation in
this proceeding, the filing of this petition, or the conduct of any ensuing trial. In
this regard, Petitioner has submitted voluntary discovery. See Petitioner’s
Voluntary Interrogatory Responses (EX1008).
B. Related Matters
Pursuant to 37 C.F.R. § 42.8(b)(2), Petitioner states that the ‘379 Patent is
presently the subject of the following patent infringement lawsuits:
• Guada Technologies LLC v. Pandora Media, Inc., Case No. 2:16-cv-
01154 (E.D. Tex.).
• Guada Technologies LLC v. Netflix, Inc., Case No. 2:16-cv-01153
(E.D. Tex.).
• Guada Technologies LLC v. Spotify USA Inc., Case No. 2:16-cv-
01159 (E.D. Tex.).
• Guada Technologies LLC v. Batanga, Inc., Case No. 2:16-cv-01148
(E.D. Tex.).
• Guada Technologies LLC v. Defy Media, LLC., Case No. 2:16-cv-
IPR2017-01039 U.S. Patent No. 7,231,379
70
01149 (E.D. Tex.).
• Guada Technologies LLC v. Emusic.com Inc. Case No. 2:16-cv-
01150 (E.D. Tex.).
• Guada Technologies LLC v. IHeartmedia, Inc., Case No. 2:16-cv-
01151 (E.D. Tex.).
• Guada Technologies LLC v MLB Advanced Media, L.P.., Case No.
2:16-cv-01152 (E.D. Tex.).
• Guada Technologies LLC v. Reliance Majestic Holdings, LLC, Case
No. 2:16-cv-01155 (E.D. Tex.).
• Guada Technologies LLC v. Rhapsody International Inc., Case No.
2:16-cv-01156 (E.D. Tex.).
• Guada Technologies LLC v Slacker, Inc. Case No. 2:16-cv-01157
(E.D. Tex.).
• Guada Technologies LLC v. Smule, Inc., Case No. 2:16-cv-01158
(E.D. Tex.).
C. Lead and Back-Up Counsel
Petitioner provides the following designation and service information for
lead and back-up counsel. 37 C.F.R. § 42.8(b)(3) and (b)(4). Please direct all
correspondence regarding this proceeding to counsel at their respective email
addresses: [email protected], [email protected], [email protected],
IPR2017-01039 U.S. Patent No. 7,231,379
71
[email protected], and [email protected]. 37 C.F.R.
§ 42.8(b)(4).
Lead Counsel Back-Up Counsel Jason R. Mudd (Reg. No. 57,700) [email protected] [email protected] Postal and Hand-Delivery Address: ERISE IP, P.A. 6201 College Blvd., Suite 300 Overland Park, Kansas 66211 Telephone: (913) 777-5600
Eric A. Buresh (Reg. No. 50,394) [email protected] [email protected] Postal and Hand-Delivery Address: ERISE IP, P.A. 6201 College Blvd., Suite 300 Overland Park, Kansas 66211 Telephone: (913) 777-5600 Jonathan Stroud (Reg. No. 72,518) [email protected] Postal and Hand-Delivery Address: Unified Patents Inc. 1875 Connecticut Ave. NW, Floor 10 Washington, D.C., 20009 Telephone: (202) 805-8931 Roshan Mansinghani (Reg. No. 62,429) [email protected] Postal and Hand-Delivery Address: Unified Patents Inc. 13355 Noel Road, Suite 1100 Dallas, TX, 75240 Telephone: (214) 945-0200
IPR2017-01039 U.S. Patent No. 7,231,379
72
APPENDIX OF EXHIBITS
Exhibit 1001 U.S. Patent 7,231,379 to Parikh et al. (’379 Patent) Exhibit 1002 File History of U.S. Patent 7,231,379 to Parikh et al. (’379 Patent
File History) Exhibit 1003 Guada’s Combined Opposition to Defendants’ Motion to Dismiss Exhibit 1004 U.S. Pat. No. 6,731,724 to Wesemann et al. (“Wesemann”) Exhibit 1005 U.S. Pat. No. 6,366,910 to Rajaraman et al. (“Rajaraman”) Exhibit 1006 U.S. Pat. No. 7,539,656 to Fratkina et al. (“Fratkina”) Exhibit 1007 Declaration of Dr. Padhraic Smyth Exhibit 1008 Petitioner’s Voluntary Interrogatory Responses Exhibit 1009 Dr. Padraic Smyth Curriculum Vitae Exhibit 1010 Hopcroft, John E., and Jeffrey D. Ullman. Data Structures and
Algorithms. Boston, MA, USA, Addison-Wesley, pp. 75-106, 155-197, 306-346, 1983
Exhibit 1011 Donald, B. Crouch, Carolyn J. Crouch, and Glenn Andreas, The use of cluster hierarchies in hypertext information retrieval, Hypertext ’89 Proceedings, ACM Press, pp. 225-237, 1989
Exhibit 1012 Yvan Leclerc, Steven W. Zucker, Denis Leclerc, McGill University, A browsing approach to documentation, IEEE Computer, IEEE Press, pp 46-49, 1982
Exhibit 1013 Ricky E. Savage, James K. Habinek,Thomas W. Barnhart, The design, simulation, and evaluation of a menu driven user interface, Proceedings of the 1982 Conference on Human Factors in Computing Systems, ACM Press, pp 36-40, 1982
Exhibit 1014 Ricardo Baeza-Yates, Berthier Ribiero-Neto, Modern Information Retrieval, pp. 24-40, ACM Press, 1999
Exhibit 1015 Daniel Cunliffe, Carl Taylor, and Douglas Tudhope, Query-based navigation in semantically indexed hypermedia, Proceedings of the Eighth ACM Conference on Hypertext, pp. 87-95, ACM Press, 1997
Exhibit 1016 Hornstein, Telephone Voice Interfaces on the Cheap at § 2.3, Proceedings of the UBLAB '94 Conference, 1994
Exhibit 1017 De Bra, Paul, et al., Information Retrieval in Distributed Hypertexts, in RIAO, pp. 481–493, 1995
Exhibit 1018 U.S. Pat. No. 6,198,939 to Holström Exhibit 1019 Karen Sparck Jones, A look back and a look forward, Proceedings
of the 11th ACM SIGIR International Conference on Research and
IPR2017-01039 U.S. Patent No. 7,231,379
73
Development in Information Retrieval, pp. 13-29, ACM Press, 1988
Exhibit 1020 Gerard Salton, Anita Wong, and Chung-Shu Yang, A vector space model for automatic indexing, Communications of the ACM, 18(11): 613-620, 1975
Exhibit 1021 Jinxi Xu, W. Bruce Croft, Query expansion using local and global document analysis, Proceedings of the 19th ACM SIGIR International Conference on Research and Development in Information Retrieval, pp. 4-11, ACM, 1996
Exhibit 1022 Carolyn J. Crouch, A cluster-based approach to thesaurus construction, Proceedings of the 11th ACM SIGIR International Conference on Research and Development in Information Retrieval pp. 309-320. ACM, 1988
Exhibit 1023 Hinrich Schütze and Jan O. Pedersen, A cooccurrence-based thesaurus and two applications to information retrieval, 1 Intelligent Multimedia Information Retrieval Systems and Management, pp. 266-274, 1994
Exhibit 1024 Güntzer et al., Automatic Thesaurus Construction by Machine Learning from Retrieval Sessions, 25 Information Processing & Management No. 3 pp. 265-273, 1998
Exhibit 1025 Mostafa et al., A Multilevel Approach to Intelligent Information Filtering: Model, System, and Evaluation, 15 ACM Transactions on Information Systems No. 4, pp. 368-399, 1997
Exhibit 1026 U.S. Patent No. 6,006,225 to Bowman et al.
IPR2017-01039 U.S. Patent No. 7,231,379
2
CERTIFICATION OF WORD COUNT
The undersigned certifies pursuant to 37 C.F.R. §42.24 that the foregoing
Petition for Inter Partes Review, excluding any table of contents, mandatory notices under 37 C.F.R. §42.8, certificates of service or word count, or appendix of exhibits, contains 13,975 words according to the word-processing program used to prepare this document (Microsoft Word). Dated: March 21, 2017 BY: /s/ Jason R. Mudd Jason R. Mudd, Reg. No. 57,700 ATTORNEY FOR PETITIONER
IPR2017-01039 U.S. Patent No. 7,231,379
3
CERTIFICATE OF SERVICE ON PATENT OWNER
UNDER 37 C.F.R. § 42.105 Pursuant to 37 C.F.R. §§ 42.6(e) and 42.105, the undersigned certifies that
on March 21, 2017, a complete and entire copy of this Petition for Inter Partes Review including exhibits was provided via Federal Express to the Patent Owner by serving the correspondence address of record for the ‘379 Patent as listed on PAIR: Morgan & Finnegan Transition Team c/o Locke Lord LLP 3 World Financial Center New York NY 10281-2101 A courtesy copy of this Petition for Inter Partes Review was also provided via e-mail to the Patent Owner’s litigation counsel of record at the below e-mail address: David R. Bennett Direction IP Law P.O. Box 14184 Chicago, IL 60614 [email protected] BY: /s/ Jason R. Mudd Jason R. Mudd, Reg. No. 57,700 ATTORNEY FOR PETITIONER