Advanced Quantitative Research Methodology, LectureNotes: Text Analysis II: Unsupervised Learning via
Cluster Analysis1
Gary Kinghttp://GKing.Harvard.Edu
December 23, 2011
1©Copyright 2010 Gary King, All Rights Reserved.Gary King http://GKing.Harvard.Edu () Advanced Quantitative Research Methodology, Lecture Notes: Text Analysis II: Unsupervised Learning via Cluster AnalysisDecember 23, 2011 1 / 23
Reading
Justin Grimmer and Gary King. 2010. “Quantitative Discovery ofQualitative Information: A General Purpose Document ClusteringMethodology”http://gking.harvard.edu/files/abs/discov-abs.shtml.
Gary King (Harvard, IQSS) Quantitative Discovery from Text 2 / 23
The Problem: Discovery from Unstructured Text
Examples: scholarly literature, news stories, medical information, blogposts, comments, product reviews, emails, social media updates,audio-to-text summaries, speeches, press releases, legal decisions, etc.
10 minutes of worldwide email = 1 LOC equivalent
An essential part of discovery is classification: “one of the mostcentral and generic of all our conceptual exercises. . . . the foundationnot only for conceptualization, language, and speech, but also formathematics, statistics, and data analysis. . . . Without classification,there could be no advanced conceptualization, reasoning, language,data analysis or, for that matter, social science research.” (Bailey,1994).
We focus on cluster analysis: discovery through (1) classification and(2) simultaneously inventing a classification scheme
(We analyze text; our methods apply more generally)
Gary King (Harvard, IQSS) Quantitative Discovery from Text 3 / 23
Why Johnny Can’t Classify (Optimally)
Bell(n) = number of ways of partitioning n objects
Bell(2) = 2 (AB, A B)
Bell(3) = 5 (ABC, AB C, A BC, AC B, A B C)
Bell(5) = 52
Bell(100) ≈ 1028 × Number of elementary particles in the universe
Now imagine choosing the optimal classification scheme by hand!
That we think of all this as astonishing . . . is astonishing
Gary King (Harvard, IQSS) Quantitative Discovery from Text 4 / 23
Why HAL Can’t Classify Either
The Goal — an optimal application-independent cluster analysismethod — is mathematically impossible:
No free lunch theorem: every possible clustering method performsequally well on average over all possible substantive applications
Existing methods:
Many choices: model-based, subspace, spectral, grid-based, graph-based, fuzzy k-modes, affinity propogation, self-organizing maps,. . .Well-defined statistical, data analytic, or machine learning foundationsHow to add substantive knowledge: With few exceptions, who knows?!The literature: little guidance on when methods applyDeep problem in cluster analysis literature: no way to know whichmethod will work ex ante
Gary King (Harvard, IQSS) Quantitative Discovery from Text 5 / 23
If Ex Ante doesn’t work, try Ex Post
Methods and substance must be connected (no free lunch theorem)
The usual approach fails: hard to do it by understanding the model
We do it ex post (by qualitative choice). For example:
Create long list of clusterings; choose the bestToo hard for mere humans!An organized list will make the search possibleE.g.,: consider two clusterings that differ only because one document(of many) moves from category 5 to 6
Gary King (Harvard, IQSS) Quantitative Discovery from Text 6 / 23
Our Idea: Meaning Through Geography
We develop a (conceptual) geography of clusterings
Gary King (Harvard, IQSS) Quantitative Discovery from Text 7 / 23
A New StrategyMake it easy to choose best clustering from millions of choices
1 Code text as numbers (in one or more of several ways)
2 Apply all clustering methods we can find to the data — eachrepresenting different (unstated) substantive assumptions (<15 mins)
3 (Too much for a person to understand, but organization will help)
4 Develop an application-independent distance metric betweenclusterings, a metric space of clusterings, and a 2-D projection
5 “Local cluster ensemble” creates a new clustering at any point, basedon weighted average of nearby clusterings
6 A new animated visualization to explore the space of clusterings(smoothly morphing from one into others)
7 Millions of clusterings, easily comprehended (takes about 10-15minutes to choose a clustering with insight)
Gary King (Harvard, IQSS) Quantitative Discovery from Text 8 / 23
Many Thousands of Clusterings, Sorted & OrganizedYou choose one (or more), based on insight, discovery, useful information,. . .
Space of Cluster Solutions
biclust_spectral
clust_convex
mult_dirproc
dismea
rock
som
spec_cos spec_eucspec_man
spec_mink
spec_max
spec_canb
mspec_cos
mspec_euc
mspec_man
mspec_mink
mspec_max
mspec_canb
affprop cosine
affprop euclidean
affprop manhattan
affprop info.costs
affprop maximum
divisive stand.euc
divisive euclidean
divisive manhattan
kmedoids stand.euc
kmedoids euclidean
kmedoids manhattan
mixvmf
mixvmfVA
kmeans euclidean
kmeans maximum
kmeans manhattan
kmeans canberra
kmeans binary
kmeans pearson
kmeans correlation
kmeans spearman
kmeans kendall
hclust euclidean ward
hclust euclidean single
hclust euclidean complete
hclust euclidean average
hclust euclidean mcquitty
hclust euclidean median
hclust euclidean centroidhclust maximum ward
hclust maximum single
hclust maximum completehclust maximum averagehclust maximum mcquitty
hclust maximum medianhclust maximum centroid
hclust manhattan ward
hclust manhattan single
hclust manhattan complete
hclust manhattan averagehclust manhattan mcquittyhclust manhattan median
hclust manhattan centroid
hclust canberra ward
hclust canberra single
hclust canberra complete
hclust canberra average
hclust canberra mcquittyhclust canberra median
hclust canberra centroid
hclust binary ward
hclust binary single
hclust binary complete
hclust binary average
hclust binary mcquitty
hclust binary median
hclust binary centroid
hclust pearson ward
hclust pearson single
hclust pearson complete
hclust pearson averagehclust pearson mcquitty
hclust pearson medianhclust pearson centroid
hclust correlation ward
hclust correlation single
hclust correlation complete
hclust correlation averagehclust correlation mcquitty
hclust correlation medianhclust correlation centroid
hclust spearman ward
hclust spearman single
hclust spearman complete
hclust spearman average
hclust spearman mcquitty
hclust spearman median
hclust spearman centroid
hclust kendall ward
hclust kendall single
hclust kendall complete
hclust kendall average
hclust kendall mcquitty
hclust kendall median
hclust kendall centroid
●
●
Cluster Solution 1
Carter
Clinton
Eisenhower
Ford
Johnson
Kennedy
Nixon
Obama
Roosevelt
Truman
Bush
HWBush
Reagan
``Other Presidents ''
``Reagan Republicans''
Cluster Solution 2
Carter
Eisenhower
Ford
Johnson
Kennedy
Nixon
Roosevelt
Truman
Bush
ClintonHWBush
Obama
Reagan
``RooseveltTo Carter''
`` Reagan To Obama ''
Gary King (Harvard, IQSS) Quantitative Discovery from Text 9 / 23
Application-Independent Distance Metric: Axioms
Metric based on 3 assumptions1 Distance between clusterings: a function of the pairwise document
agreements (pairwise agreements ⇒ triples, quadruples, etc.)2 Invariance: Distance is invariant to the number of documents (for any
fixed number of clusters)3 Scale: the maximum distance is set to log(num clusters)
Only one measure satisfies all three (the “variation ofinformation”)
Meila (2007): derives same metric using different axioms (latticetheory)
Gary King (Harvard, IQSS) Quantitative Discovery from Text 10 / 23
Available March 2009: 304ppPb: 978-0-415-99701-0: $24.95www.routledge.com/politics
“The list of authors in The Future of Political Science is a 'who’s who' of political science. As I was reading it, I came to think of it as a platter of tasty hors d’oeuvres. It hooked me thoroughly.”
—Peter Kingstone, University of Connecticut
“In this one-of-a-kind collection, an eclectic set of contributors offer short but forceful forecasts about the future of the discipline. The resulting assortment is captivating, consistently thought-provoking, often intriguing, and sure to spur discussion and debate.”
—Wendy K. Tam Cho, University of Illinois at Urbana-Champaign
“King, Schlozman, and Nie have created a visionary and stimulating volume. The organization of the essays strikes me as nothing less than brilliant. . . It is truly a joy to read.”
—Lawrence C. Dodd, Manning J. Dauer Eminent Scholar in Political Science, University of Florida
The FuTure oF PoliTical Science100 Perspectivesedited by Gary King, harvard university, Kay Lehman Schlozman, Boston college and Norman H. Nie, Stanford university
Gary King (Harvard, IQSS) Quantitative Discovery from Text 11 / 23
Evaluators’ Rate Machine Choices Better Than Their Own
Scale: (1) unrelated, (2) loosely related, or (3) closely related
Table reports: mean(scale)
Pairs from Overall Mean Evaluator 1 Evaluator 2
Random Selection 1.38 1.16 1.60Hand-Coded Clusters 1.58 1.48 1.68Hand-Coding 2.06 1.88 2.24Machine 2.24 2.08 2.40
p.s. The hand-coders did the evaluation!
Gary King (Harvard, IQSS) Quantitative Discovery from Text 12 / 23
Evaluating Performance
Goals:
Validate Claim: computer-assisted conceptualization outperformshuman conceptualizationDemonstrate: new experimental designs for cluster evaluationInject human judgement: relying on insights from survey research
We now present three evaluations
Cluster Quality ⇒ RA codersInformative discoveries ⇒ Experienced scholars analyzing textsDiscovery ⇒ You’re the judge
Gary King (Harvard, IQSS) Quantitative Discovery from Text 13 / 23
Evaluation 1: Cluster Quality
What Are Humans Good For?
They can’t: keep many documents & clusters in their headThey can: compare two documents at a time=⇒ Cluster quality evaluation: human judgement of document pairs
Experimental Design to Assess Cluster Quality
automated visualization to choose one clusteringmany pairs of documentsfor coders: (1) unrelated, (2) loosely related, (3) closely relatedQuality = mean(within cluster) - mean(between clusters)Bias results against ourselves by not letting evaluators choose clustering
Gary King (Harvard, IQSS) Quantitative Discovery from Text 14 / 23
Evaluation 1: Cluster Quality
(Our Method) − (Human Coders)
−0.3 −0.2 −0.1 0.1 0.2 0.3
●
Lautenberg Press Releases
Lautenberg: 200 Senate Press Releases (appropriations, economy,education, tax, veterans, . . . )
Gary King (Harvard, IQSS) Quantitative Discovery from Text 15 / 23
Evaluation 1: Cluster Quality
(Our Method) − (Human Coders)
−0.3 −0.2 −0.1 0.1 0.2 0.3
●
Lautenberg Press Releases
●
Policy Agendas Project
Policy Agendas: 213 quasi-sentences from Bush’s State of the Union(agriculture, banking & commerce, civil rights/liberties, defense, . . . )
Gary King (Harvard, IQSS) Quantitative Discovery from Text 15 / 23
Evaluation 1: Cluster Quality
(Our Method) − (Human Coders)
−0.3 −0.2 −0.1 0.1 0.2 0.3
●
Lautenberg Press Releases
●
Policy Agendas Project
●
Reuter's Gold Standard
Reuter’s: financial news (trade, earnings, copper, gold, coffee, . . . ); “goldstandard” for supervised learning studies
Gary King (Harvard, IQSS) Quantitative Discovery from Text 15 / 23
Evaluation 2: More Informative Discoveries
Found 2 scholars analyzing lots of textual data for their work
Created 6 clusterings:
2 clusterings selected with our method (biased against us)2 clusterings from each of 2 other methods (varying tuning parameters)
Created info packet on each clustering (for each cluster: exemplardocument, automated content summary)
Asked for(62
)=15 pairwise comparisons
User chooses ⇒ only care about the one clustering that wins
Both cases a Condorcet winner:
“Immigration”:
Our Method 1 → vMF 1 → vMF 2 → Our Method 2 → K-Means 1 → K-Means 2
“Genetic testing”:
Our Method 1 → {Our Method 2, K-Means 1, K-means 2} → Dir Proc. 1 → Dir Proc. 2
Gary King (Harvard, IQSS) Quantitative Discovery from Text 16 / 23
Evaluation 3: What Do Members of Congress Do?
- David Mayhew’s (1974) famous typology
- Advertising- Credit Claiming- Position Taking
- Data: 200 press releases from Frank Lautenberg’s office (D-NJ)
- Apply our method
Gary King (Harvard, IQSS) Quantitative Discovery from Text 17 / 23
Example Discovery
: Partisan Taunting
biclust_spectral
clust_convex
mult_dirproc
dismeadist_cosdist_fbinarydist_ebinarydist_minkowskidist_maxdist_canbdist_binary
mec
rocksom
sot_euc
sot_cor
spec_cosspec_eucspec_man
spec_minkspec_maxspec_canbmspec_cosmspec_euc
mspec_manmspec_minkmspec_max
mspec_canb
affprop cosine
affprop euclidean
affprop manhattan
affprop info.costs
affprop maximum
divisive stand.euc
divisive euclidean
divisive manhattan
kmedoids stand.euckmedoids euclidean
kmedoids manhattan
mixvmf mixvmfVA
kmeans euclidean
kmeans maximum
kmeans manhattankmeans canberra
kmeans binary
kmeans pearson
kmeans correlation
kmeans spearmankmeans kendall
hclust euclidean ward
hclust euclidean single
hclust euclidean complete
hclust euclidean averagehclust euclidean mcquitty
hclust euclidean medianhclust euclidean centroid
hclust maximum ward
hclust maximum single
hclust maximum completehclust maximum average
hclust maximum mcquitty
hclust maximum medianhclust maximum centroid
hclust manhattan ward
hclust manhattan single
hclust manhattan complete
hclust manhattan average
hclust manhattan mcquitty
hclust manhattan medianhclust manhattan centroid
hclust canberra ward
hclust canberra single
hclust canberra complete
hclust canberra average
hclust canberra mcquitty
hclust canberra median
hclust canberra centroid
hclust binary ward
hclust binary single
hclust binary complete
hclust binary average
hclust binary mcquittyhclust binary median
hclust binary centroid
hclust pearson ward
hclust pearson single
hclust pearson completehclust pearson average
hclust pearson mcquittyhclust pearson median
hclust pearson centroid
hclust correlation ward
hclust correlation single hclust correlation completehclust correlation average
hclust correlation mcquitty
hclust correlation median
hclust correlation centroid
hclust spearman ward
hclust spearman single
hclust spearman complete
hclust spearman averagehclust spearman mcquitty
hclust spearman medianhclust spearman centroid
hclust kendall ward
hclust kendall singlehclust kendall complete
hclust kendall averagehclust kendall mcquittyhclust kendall medianhclust kendall centroid
affprop cosine
mixvmf
Red point: a clustering byAffinity Propagation-Cosine(Dueck and Frey 2007)Close to:Mixture of von Mises-Fisherdistributions (Banerjee et. al.2005)
Gary King (Harvard, IQSS) Quantitative Discovery from Text 18 / 23
Example Discovery
: Partisan Taunting
biclust_spectral
clust_convex
mult_dirproc
dismeadist_cosdist_fbinarydist_ebinarydist_minkowskidist_maxdist_canbdist_binary
mec
rocksom
sot_euc
sot_cor
spec_cosspec_eucspec_man
spec_minkspec_maxspec_canbmspec_cosmspec_euc
mspec_manmspec_minkmspec_max
mspec_canb
affprop cosine
affprop euclidean
affprop manhattan
affprop info.costs
affprop maximum
divisive stand.euc
divisive euclidean
divisive manhattan
kmedoids stand.euckmedoids euclidean
kmedoids manhattan
mixvmf mixvmfVA
kmeans euclidean
kmeans maximum
kmeans manhattankmeans canberra
kmeans binary
kmeans pearson
kmeans correlation
kmeans spearmankmeans kendall
hclust euclidean ward
hclust euclidean single
hclust euclidean complete
hclust euclidean averagehclust euclidean mcquitty
hclust euclidean medianhclust euclidean centroid
hclust maximum ward
hclust maximum single
hclust maximum completehclust maximum average
hclust maximum mcquitty
hclust maximum medianhclust maximum centroid
hclust manhattan ward
hclust manhattan single
hclust manhattan complete
hclust manhattan average
hclust manhattan mcquitty
hclust manhattan medianhclust manhattan centroid
hclust canberra ward
hclust canberra single
hclust canberra complete
hclust canberra average
hclust canberra mcquitty
hclust canberra median
hclust canberra centroid
hclust binary ward
hclust binary single
hclust binary complete
hclust binary average
hclust binary mcquittyhclust binary median
hclust binary centroid
hclust pearson ward
hclust pearson single
hclust pearson completehclust pearson average
hclust pearson mcquittyhclust pearson median
hclust pearson centroid
hclust correlation ward
hclust correlation single hclust correlation completehclust correlation average
hclust correlation mcquitty
hclust correlation median
hclust correlation centroid
hclust spearman ward
hclust spearman single
hclust spearman complete
hclust spearman averagehclust spearman mcquitty
hclust spearman medianhclust spearman centroid
hclust kendall ward
hclust kendall singlehclust kendall complete
hclust kendall averagehclust kendall mcquittyhclust kendall medianhclust kendall centroid
●
Space between methods:local cluster ensemble
Gary King (Harvard, IQSS) Quantitative Discovery from Text 18 / 23
Example Discovery
: Partisan Taunting
biclust_spectral
clust_convex
mult_dirproc
dismeadist_cosdist_fbinarydist_ebinarydist_minkowskidist_maxdist_canbdist_binary
mec
rocksom
sot_euc
sot_cor
spec_cosspec_eucspec_man
spec_minkspec_maxspec_canbmspec_cosmspec_euc
mspec_manmspec_minkmspec_max
mspec_canb
affprop cosine
affprop euclidean
affprop manhattan
affprop info.costs
affprop maximum
divisive stand.euc
divisive euclidean
divisive manhattan
kmedoids stand.euckmedoids euclidean
kmedoids manhattan
mixvmf mixvmfVA
kmeans euclidean
kmeans maximum
kmeans manhattankmeans canberra
kmeans binary
kmeans pearson
kmeans correlation
kmeans spearmankmeans kendall
hclust euclidean ward
hclust euclidean single
hclust euclidean complete
hclust euclidean averagehclust euclidean mcquitty
hclust euclidean medianhclust euclidean centroid
hclust maximum ward
hclust maximum single
hclust maximum completehclust maximum average
hclust maximum mcquitty
hclust maximum medianhclust maximum centroid
hclust manhattan ward
hclust manhattan single
hclust manhattan complete
hclust manhattan average
hclust manhattan mcquitty
hclust manhattan medianhclust manhattan centroid
hclust canberra ward
hclust canberra single
hclust canberra complete
hclust canberra average
hclust canberra mcquitty
hclust canberra median
hclust canberra centroid
hclust binary ward
hclust binary single
hclust binary complete
hclust binary average
hclust binary mcquittyhclust binary median
hclust binary centroid
hclust pearson ward
hclust pearson single
hclust pearson completehclust pearson average
hclust pearson mcquittyhclust pearson median
hclust pearson centroid
hclust correlation ward
hclust correlation single hclust correlation completehclust correlation average
hclust correlation mcquitty
hclust correlation median
hclust correlation centroid
hclust spearman ward
hclust spearman single
hclust spearman complete
hclust spearman averagehclust spearman mcquitty
hclust spearman medianhclust spearman centroid
hclust kendall ward
hclust kendall singlehclust kendall complete
hclust kendall averagehclust kendall mcquittyhclust kendall medianhclust kendall centroid
Found a region with particularlyinsightful clusterings
Gary King (Harvard, IQSS) Quantitative Discovery from Text 18 / 23
Example Discovery
: Partisan Taunting
biclust_spectral
clust_convex
mult_dirproc
dismeadist_cosdist_fbinarydist_ebinarydist_minkowskidist_maxdist_canbdist_binary
mec
rocksom
sot_euc
sot_cor
spec_cosspec_eucspec_man
spec_minkspec_maxspec_canbmspec_cosmspec_euc
mspec_manmspec_minkmspec_max
mspec_canb
affprop cosine
affprop euclidean
affprop manhattan
affprop info.costs
affprop maximum
divisive stand.euc
divisive euclidean
divisive manhattan
kmedoids stand.euckmedoids euclidean
kmedoids manhattan
mixvmf mixvmfVA
kmeans euclidean
kmeans maximum
kmeans manhattankmeans canberra
kmeans binary
kmeans pearson
kmeans correlation
kmeans spearmankmeans kendall
hclust euclidean ward
hclust euclidean single
hclust euclidean complete
hclust euclidean averagehclust euclidean mcquitty
hclust euclidean medianhclust euclidean centroid
hclust maximum ward
hclust maximum single
hclust maximum completehclust maximum average
hclust maximum mcquitty
hclust maximum medianhclust maximum centroid
hclust manhattan ward
hclust manhattan single
hclust manhattan complete
hclust manhattan average
hclust manhattan mcquitty
hclust manhattan medianhclust manhattan centroid
hclust canberra ward
hclust canberra single
hclust canberra complete
hclust canberra average
hclust canberra mcquitty
hclust canberra median
hclust canberra centroid
hclust binary ward
hclust binary single
hclust binary complete
hclust binary average
hclust binary mcquittyhclust binary median
hclust binary centroid
hclust pearson ward
hclust pearson single
hclust pearson completehclust pearson average
hclust pearson mcquittyhclust pearson median
hclust pearson centroid
hclust correlation ward
hclust correlation single hclust correlation completehclust correlation average
hclust correlation mcquitty
hclust correlation median
hclust correlation centroid
hclust spearman ward
hclust spearman single
hclust spearman complete
hclust spearman averagehclust spearman mcquitty
hclust spearman medianhclust spearman centroid
hclust kendall ward
hclust kendall singlehclust kendall complete
hclust kendall averagehclust kendall mcquittyhclust kendall medianhclust kendall centroid
●●
Mixture:
0.39 Hclust-Canberra-McQuitty
0.30 Spectral clusteringRandom Walk(Metrics 1-6)
0.13 Hclust-Correlation-Ward
0.09 Hclust-Pearson-Ward
0.05 Kmediods-Cosine
0.04 Spectral clusteringSymmetric(Metrics 1-6)
Gary King (Harvard, IQSS) Quantitative Discovery from Text 18 / 23
Example Discovery
: Partisan Taunting
biclust_spectral
clust_convex
mult_dirproc
dismeadist_cosdist_fbinarydist_ebinarydist_minkowskidist_maxdist_canbdist_binary
mec
rocksom
sot_euc
sot_cor
spec_cosspec_eucspec_man
spec_minkspec_maxspec_canbmspec_cosmspec_euc
mspec_manmspec_minkmspec_max
mspec_canb
affprop cosine
affprop euclidean
affprop manhattan
affprop info.costs
affprop maximum
divisive stand.euc
divisive euclidean
divisive manhattan
kmedoids stand.euckmedoids euclidean
kmedoids manhattan
mixvmf mixvmfVA
kmeans euclidean
kmeans maximum
kmeans manhattankmeans canberra
kmeans binary
kmeans pearson
kmeans correlation
kmeans spearmankmeans kendall
hclust euclidean ward
hclust euclidean single
hclust euclidean complete
hclust euclidean averagehclust euclidean mcquitty
hclust euclidean medianhclust euclidean centroid
hclust maximum ward
hclust maximum single
hclust maximum completehclust maximum average
hclust maximum mcquitty
hclust maximum medianhclust maximum centroid
hclust manhattan ward
hclust manhattan single
hclust manhattan complete
hclust manhattan average
hclust manhattan mcquitty
hclust manhattan medianhclust manhattan centroid
hclust canberra ward
hclust canberra single
hclust canberra complete
hclust canberra average
hclust canberra mcquitty
hclust canberra median
hclust canberra centroid
hclust binary ward
hclust binary single
hclust binary complete
hclust binary average
hclust binary mcquittyhclust binary median
hclust binary centroid
hclust pearson ward
hclust pearson single
hclust pearson completehclust pearson average
hclust pearson mcquittyhclust pearson median
hclust pearson centroid
hclust correlation ward
hclust correlation single hclust correlation completehclust correlation average
hclust correlation mcquitty
hclust correlation median
hclust correlation centroid
hclust spearman ward
hclust spearman single
hclust spearman complete
hclust spearman averagehclust spearman mcquitty
hclust spearman medianhclust spearman centroid
hclust kendall ward
hclust kendall singlehclust kendall complete
hclust kendall averagehclust kendall mcquittyhclust kendall medianhclust kendall centroid
●
Clusters in this Clustering
Mayhew
●
●
●
●
●
●● ●
●●
●
●
●
●●
●
●●
●
●
●
●
●
●●
Credit ClaimingPork
Credit Claiming, Pork:“Sens. Frank R. Lautenberg(D-NJ) and Robert Menendez(D-NJ) announced that the U.S.Department of Commerce hasawarded a $100,000 grant to theSouth Jersey EconomicDevelopment District”
Gary King (Harvard, IQSS) Quantitative Discovery from Text 18 / 23
Example Discovery
: Partisan Taunting
biclust_spectral
clust_convex
mult_dirproc
dismeadist_cosdist_fbinarydist_ebinarydist_minkowskidist_maxdist_canbdist_binary
mec
rocksom
sot_euc
sot_cor
spec_cosspec_eucspec_man
spec_minkspec_maxspec_canbmspec_cosmspec_euc
mspec_manmspec_minkmspec_max
mspec_canb
affprop cosine
affprop euclidean
affprop manhattan
affprop info.costs
affprop maximum
divisive stand.euc
divisive euclidean
divisive manhattan
kmedoids stand.euckmedoids euclidean
kmedoids manhattan
mixvmf mixvmfVA
kmeans euclidean
kmeans maximum
kmeans manhattankmeans canberra
kmeans binary
kmeans pearson
kmeans correlation
kmeans spearmankmeans kendall
hclust euclidean ward
hclust euclidean single
hclust euclidean complete
hclust euclidean averagehclust euclidean mcquitty
hclust euclidean medianhclust euclidean centroid
hclust maximum ward
hclust maximum single
hclust maximum completehclust maximum average
hclust maximum mcquitty
hclust maximum medianhclust maximum centroid
hclust manhattan ward
hclust manhattan single
hclust manhattan complete
hclust manhattan average
hclust manhattan mcquitty
hclust manhattan medianhclust manhattan centroid
hclust canberra ward
hclust canberra single
hclust canberra complete
hclust canberra average
hclust canberra mcquitty
hclust canberra median
hclust canberra centroid
hclust binary ward
hclust binary single
hclust binary complete
hclust binary average
hclust binary mcquittyhclust binary median
hclust binary centroid
hclust pearson ward
hclust pearson single
hclust pearson completehclust pearson average
hclust pearson mcquittyhclust pearson median
hclust pearson centroid
hclust correlation ward
hclust correlation single hclust correlation completehclust correlation average
hclust correlation mcquitty
hclust correlation median
hclust correlation centroid
hclust spearman ward
hclust spearman single
hclust spearman complete
hclust spearman averagehclust spearman mcquitty
hclust spearman medianhclust spearman centroid
hclust kendall ward
hclust kendall singlehclust kendall complete
hclust kendall averagehclust kendall mcquittyhclust kendall medianhclust kendall centroid
●
Clusters in this Clustering
Mayhew
●
●
●
●
●
●● ●
●●
●
●
●
●●
●
●●
●
●
●
●
●
●●
Credit ClaimingPork
● ●●
●
●
●●
●
●
●
●
●
●
●
●
●●
●●
●
●
●
●
●
●
Credit ClaimingLegislation
Credit Claiming, Legislation:“As the Senate begins its recess,Senator Frank Lautenberg todaypointed to a string of victories inCongress on his legislative agendaduring this work period”
Gary King (Harvard, IQSS) Quantitative Discovery from Text 18 / 23
Example Discovery
: Partisan Taunting
biclust_spectral
clust_convex
mult_dirproc
dismeadist_cosdist_fbinarydist_ebinarydist_minkowskidist_maxdist_canbdist_binary
mec
rocksom
sot_euc
sot_cor
spec_cosspec_eucspec_man
spec_minkspec_maxspec_canbmspec_cosmspec_euc
mspec_manmspec_minkmspec_max
mspec_canb
affprop cosine
affprop euclidean
affprop manhattan
affprop info.costs
affprop maximum
divisive stand.euc
divisive euclidean
divisive manhattan
kmedoids stand.euckmedoids euclidean
kmedoids manhattan
mixvmf mixvmfVA
kmeans euclidean
kmeans maximum
kmeans manhattankmeans canberra
kmeans binary
kmeans pearson
kmeans correlation
kmeans spearmankmeans kendall
hclust euclidean ward
hclust euclidean single
hclust euclidean complete
hclust euclidean averagehclust euclidean mcquitty
hclust euclidean medianhclust euclidean centroid
hclust maximum ward
hclust maximum single
hclust maximum completehclust maximum average
hclust maximum mcquitty
hclust maximum medianhclust maximum centroid
hclust manhattan ward
hclust manhattan single
hclust manhattan complete
hclust manhattan average
hclust manhattan mcquitty
hclust manhattan medianhclust manhattan centroid
hclust canberra ward
hclust canberra single
hclust canberra complete
hclust canberra average
hclust canberra mcquitty
hclust canberra median
hclust canberra centroid
hclust binary ward
hclust binary single
hclust binary complete
hclust binary average
hclust binary mcquittyhclust binary median
hclust binary centroid
hclust pearson ward
hclust pearson single
hclust pearson completehclust pearson average
hclust pearson mcquittyhclust pearson median
hclust pearson centroid
hclust correlation ward
hclust correlation single hclust correlation completehclust correlation average
hclust correlation mcquitty
hclust correlation median
hclust correlation centroid
hclust spearman ward
hclust spearman single
hclust spearman complete
hclust spearman averagehclust spearman mcquitty
hclust spearman medianhclust spearman centroid
hclust kendall ward
hclust kendall singlehclust kendall complete
hclust kendall averagehclust kendall mcquittyhclust kendall medianhclust kendall centroid
●
Clusters in this Clustering
Mayhew
●
●
●
●
●
●● ●
●●
●
●
●
●●
●
●●
●
●
●
●
●
●●
Credit ClaimingPork
● ●●
●
●
●●
●
●
●
●
●
●
●
●
●●
●●
●
●
●
●
●
●
Credit ClaimingLegislation
●●
●
●
●●
●
●●
●
AdvertisingAdvertising
Advertising:“Senate AdoptsLautenberg/Menendez ResolutionHonoring Spelling Bee Championfrom New Jersey”
Gary King (Harvard, IQSS) Quantitative Discovery from Text 18 / 23
Example Discovery: Partisan Taunting
biclust_spectral
clust_convex
mult_dirproc
dismeadist_cosdist_fbinarydist_ebinarydist_minkowskidist_maxdist_canbdist_binary
mec
rocksom
sot_euc
sot_cor
spec_cosspec_eucspec_man
spec_minkspec_maxspec_canbmspec_cosmspec_euc
mspec_manmspec_minkmspec_max
mspec_canb
affprop cosine
affprop euclidean
affprop manhattan
affprop info.costs
affprop maximum
divisive stand.euc
divisive euclidean
divisive manhattan
kmedoids stand.euckmedoids euclidean
kmedoids manhattan
mixvmf mixvmfVA
kmeans euclidean
kmeans maximum
kmeans manhattankmeans canberra
kmeans binary
kmeans pearson
kmeans correlation
kmeans spearmankmeans kendall
hclust euclidean ward
hclust euclidean single
hclust euclidean complete
hclust euclidean averagehclust euclidean mcquitty
hclust euclidean medianhclust euclidean centroid
hclust maximum ward
hclust maximum single
hclust maximum completehclust maximum average
hclust maximum mcquitty
hclust maximum medianhclust maximum centroid
hclust manhattan ward
hclust manhattan single
hclust manhattan complete
hclust manhattan average
hclust manhattan mcquitty
hclust manhattan medianhclust manhattan centroid
hclust canberra ward
hclust canberra single
hclust canberra complete
hclust canberra average
hclust canberra mcquitty
hclust canberra median
hclust canberra centroid
hclust binary ward
hclust binary single
hclust binary complete
hclust binary average
hclust binary mcquittyhclust binary median
hclust binary centroid
hclust pearson ward
hclust pearson single
hclust pearson completehclust pearson average
hclust pearson mcquittyhclust pearson median
hclust pearson centroid
hclust correlation ward
hclust correlation single hclust correlation completehclust correlation average
hclust correlation mcquitty
hclust correlation median
hclust correlation centroid
hclust spearman ward
hclust spearman single
hclust spearman complete
hclust spearman averagehclust spearman mcquitty
hclust spearman medianhclust spearman centroid
hclust kendall ward
hclust kendall singlehclust kendall complete
hclust kendall averagehclust kendall mcquittyhclust kendall medianhclust kendall centroid
●
Clusters in this Clustering
Mayhew
●
●
●
●
●
●● ●
●●
●
●
●
●●
●
●●
●
●
●
●
●
●●
Credit ClaimingPork
● ●●
●
●
●●
●
●
●
●
●
●
●
●
●●
●●
●
●
●
●
●
●
Credit ClaimingLegislation
●●
●
●
●●
●
●●
●
AdvertisingAdvertising
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
● ●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
● ●
●
Partisan Taunting
Partisan Taunting:“Republicans Selling Out Nationon Chemical Plant Security”
Gary King (Harvard, IQSS) Quantitative Discovery from Text 18 / 23
Example Discovery: Partisan Taunting
biclust_spectral
clust_convex
mult_dirproc
dismeadist_cosdist_fbinarydist_ebinarydist_minkowskidist_maxdist_canbdist_binary
mec
rocksom
sot_euc
sot_cor
spec_cosspec_eucspec_man
spec_minkspec_maxspec_canbmspec_cosmspec_euc
mspec_manmspec_minkmspec_max
mspec_canb
affprop cosine
affprop euclidean
affprop manhattan
affprop info.costs
affprop maximum
divisive stand.euc
divisive euclidean
divisive manhattan
kmedoids stand.euckmedoids euclidean
kmedoids manhattan
mixvmf mixvmfVA
kmeans euclidean
kmeans maximum
kmeans manhattankmeans canberra
kmeans binary
kmeans pearson
kmeans correlation
kmeans spearmankmeans kendall
hclust euclidean ward
hclust euclidean single
hclust euclidean complete
hclust euclidean averagehclust euclidean mcquitty
hclust euclidean medianhclust euclidean centroid
hclust maximum ward
hclust maximum single
hclust maximum completehclust maximum average
hclust maximum mcquitty
hclust maximum medianhclust maximum centroid
hclust manhattan ward
hclust manhattan single
hclust manhattan complete
hclust manhattan average
hclust manhattan mcquitty
hclust manhattan medianhclust manhattan centroid
hclust canberra ward
hclust canberra single
hclust canberra complete
hclust canberra average
hclust canberra mcquitty
hclust canberra median
hclust canberra centroid
hclust binary ward
hclust binary single
hclust binary complete
hclust binary average
hclust binary mcquittyhclust binary median
hclust binary centroid
hclust pearson ward
hclust pearson single
hclust pearson completehclust pearson average
hclust pearson mcquittyhclust pearson median
hclust pearson centroid
hclust correlation ward
hclust correlation single hclust correlation completehclust correlation average
hclust correlation mcquitty
hclust correlation median
hclust correlation centroid
hclust spearman ward
hclust spearman single
hclust spearman complete
hclust spearman averagehclust spearman mcquitty
hclust spearman medianhclust spearman centroid
hclust kendall ward
hclust kendall singlehclust kendall complete
hclust kendall averagehclust kendall mcquittyhclust kendall medianhclust kendall centroid
●
Clusters in this Clustering
Mayhew
●
●
●
●
●
●● ●
●●
●
●
●
●●
●
●●
●
●
●
●
●
●●
Credit ClaimingPork
● ●●
●
●
●●
●
●
●
●
●
●
●
●
●●
●●
●
●
●
●
●
●
Credit ClaimingLegislation
●●
●
●
●●
●
●●
●
AdvertisingAdvertising
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
● ●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
● ●
●
Partisan Taunting
Partisan Taunting:“Senator Lautenberg’samendment would change thename of ...the Republican bill...to‘More Tax Breaks for the Richand More Debt for OurGrandchildren Deficit ExpansionReconciliation Act of 2006’ ”
Gary King (Harvard, IQSS) Quantitative Discovery from Text 18 / 23
Example Discovery: Partisan Taunting
biclust_spectral
clust_convex
mult_dirproc
dismeadist_cosdist_fbinarydist_ebinarydist_minkowskidist_maxdist_canbdist_binary
mec
rocksom
sot_euc
sot_cor
spec_cosspec_eucspec_man
spec_minkspec_maxspec_canbmspec_cosmspec_euc
mspec_manmspec_minkmspec_max
mspec_canb
affprop cosine
affprop euclidean
affprop manhattan
affprop info.costs
affprop maximum
divisive stand.euc
divisive euclidean
divisive manhattan
kmedoids stand.euckmedoids euclidean
kmedoids manhattan
mixvmf mixvmfVA
kmeans euclidean
kmeans maximum
kmeans manhattankmeans canberra
kmeans binary
kmeans pearson
kmeans correlation
kmeans spearmankmeans kendall
hclust euclidean ward
hclust euclidean single
hclust euclidean complete
hclust euclidean averagehclust euclidean mcquitty
hclust euclidean medianhclust euclidean centroid
hclust maximum ward
hclust maximum single
hclust maximum completehclust maximum average
hclust maximum mcquitty
hclust maximum medianhclust maximum centroid
hclust manhattan ward
hclust manhattan single
hclust manhattan complete
hclust manhattan average
hclust manhattan mcquitty
hclust manhattan medianhclust manhattan centroid
hclust canberra ward
hclust canberra single
hclust canberra complete
hclust canberra average
hclust canberra mcquitty
hclust canberra median
hclust canberra centroid
hclust binary ward
hclust binary single
hclust binary complete
hclust binary average
hclust binary mcquittyhclust binary median
hclust binary centroid
hclust pearson ward
hclust pearson single
hclust pearson completehclust pearson average
hclust pearson mcquittyhclust pearson median
hclust pearson centroid
hclust correlation ward
hclust correlation single hclust correlation completehclust correlation average
hclust correlation mcquitty
hclust correlation median
hclust correlation centroid
hclust spearman ward
hclust spearman single
hclust spearman complete
hclust spearman averagehclust spearman mcquitty
hclust spearman medianhclust spearman centroid
hclust kendall ward
hclust kendall singlehclust kendall complete
hclust kendall averagehclust kendall mcquittyhclust kendall medianhclust kendall centroid
●
Clusters in this Clustering
Mayhew
●
●
●
●
●
●● ●
●●
●
●
●
●●
●
●●
●
●
●
●
●
●●
Credit ClaimingPork
● ●●
●
●
●●
●
●
●
●
●
●
●
●
●●
●●
●
●
●
●
●
●
Credit ClaimingLegislation
●●
●
●
●●
●
●●
●
AdvertisingAdvertising
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
● ●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
● ●
●
Partisan Taunting
Definition: Explicit, public, andnegative attacks on anotherpolitical party or its members
Taunting ruinsdeliberation
Gary King (Harvard, IQSS) Quantitative Discovery from Text 18 / 23
In Sample Illustration of Partisan Taunting
Taunting ruins deliberation
Sen. Lautenbergon Senate Floor4/29/04
- “Senator Lautenberg BlastsRepublicans as ‘Chicken Hawks’ ”[Government Oversight]
- “The scopes trial took place in1925. Sadly, President Bush’s vetotoday shows that we haven’tprogressed much since then”[Healthcare]
- “Every day the House Republicansdragged this out was a day thatmade our communities lesssafe.”[Homeland Security]
Gary King (Harvard, IQSS) Quantitative Discovery from Text 19 / 23
Out of Sample Confirmation of Partisan Taunting
- Discovered using 200 press releases; 1 senator.- Confirmed using 64,033 press releases; 301 senator-years.- Apply supervised learning method: measure proportion of press
releases a senator taunts other party
Prop. of Press Releases Taunting
Fre
quen
cy
0.1 0.2 0.3 0.4 0.5
1020
30
Gary King (Harvard, IQSS) Quantitative Discovery from Text 20 / 23
Out of Sample Confirmation of Partisan Taunting
- Discovered using 200 press releases; 1 senator.- Confirmed using 64,033 press releases; 301 senator-years.- Apply supervised learning method: measure proportion of press
releases a senator taunts other party
Prop. of Press Releases Taunting
Fre
quen
cy
0.1 0.2 0.3 0.4 0.5
1020
30
On Avg., Senators Taunt in 27 % of Press Releases
Gary King (Harvard, IQSS) Quantitative Discovery from Text 21 / 23
Advancing the Objective of Discovery
1) Conceptualization
2) Measurement
3) Validation
Quantitative Methods
Qualitative Methods (reading!)
Quantitative methods for conceptualization: aiding discovery
- Few formal methods designed explicitly for conceptualization
- Belittled: “Tom Swift and His Electric Factor Analysis Machine”(Armstrong 1967)
- Evaluation methods measure progress in discoveryGary King (Harvard, IQSS) Quantitative Discovery from Text 22 / 23
For more information:
http://GKing.Harvard.edu
Gary King (Harvard, IQSS) Quantitative Discovery from Text 23 / 23