Date post: | 01-Apr-2015 |
Category: |
Documents |
Upload: | angela-spragg |
View: | 222 times |
Download: | 3 times |
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Lu Huang, 1 Ying Guo, 1 TingTing Ma, 1,2 Alan L. Porter 2,3
1 School of Management and Economics, Beijing Institute of Technology, Beijing 100081, P. R. China 2 Technology Policy and Assessment Center, Georgia Tech, Atlanta, GA, USA
3 Search Technology, Inc., Norcross, GA, USA
The 4th International Seville Conference onFuture-Oriented Technology Analysis (FTA)
12 & 13 May 2011
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Introduction
We seek to provide usable intelligence, not only to get a handle on the discontinuous development of NEST’s, but also on the pertinent contextual forces and factors affecting possible technological innovation.
Technology Forecasting of Incrementally Advancing Technologies
Future-oriented Technology Aanalyses (FTA) Our endeavours: withine the context of
FTA
Toolsdevelop
New & Emerging Science & Technologies (NESTs) ……
NESTshigh uncertainty
high dynamics
FTA tools for NESTs pose notable challenges
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Data and Methods
We have devised a four-stage approach to Forecast Innovation Pathways
(“FIP”). This integrates a) heavily empirical “Tech Mining” with b) heavily
expert-based Multipath Mapping. The four FIP stages blend empirical and
expert knowledge.
Stage 1 – Understand the NEST and its critical environment
Stage 2 – Tech Mine
Stage 3 – Forecast likely innovation paths
Stage 4 – Synthesize and report
To operationalize these stages, we break them down into 10 steps. We label these A
through J, but should emphasize that forecasting innovation pathways is not a once-
through, linear process.
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Data and Methods
We exemplify our approach for a particular NEST case
Dye-Sensitized Solar Cells (“DSSCs”)
DSSC: one type of nano-enabled solar cells with special promise, are made of low-cost materials and are less equipment-intensive than other solar cell technologies.
This analysis treats DSSC abstract records through 2010 based on searches
in three databases:• 4104 documents (including 3134 articles) appearing in the Science Citation
Index (SCI) of the Web of Science (fundamental research emphasis)• 3730 documents from EI Compendex (journal and conference articles)• 3097 patent families from the Derwent World Patent Index (DWPI)
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Results----Profile R&D
Dye-sensitized solar cells publication & patent trends
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Results----Profile R&D
DSSC Science Overlay Map • DSSC research involves many fields• It concentrates in Materials Science and Chemistry• Could help locate expertise
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Results----Profile R&D
Geo-map of DSSC Research Organizations in China (based on SCI)
• Geo-map for China locating DSSC research activity
• Note several hotbeds
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Results----Profile innovation actors & activities
Cites Share thru 2008
Cites Share 2009 on
Pubs Share thru 2008
Pubs Share 2009 on
Chinese Acad Sciences (CAS) 6.0% 19.9% 19.5% 25.3%
Swiss Fed Inst Technol (EPFL) 49.3% 28.6% 20.5% 18.5%
AIST (Japan) 7.7% 4.4% 11.1% 7.2%
Uppsala University 8.1% 4.7% 5.7% 9.5%
Korea Inst Sci & Technol 1.9% 5.1% 6.3% 8.2%
Korea University 2.3% 10.3% 6.1% 8.2%
Natl Taiwan Univeristy 1.5% 5.2% 5.8% 7.2%
Imperial College, London 6.9% 6.9% 6.2% 6.4%
Royal Inst Technol 3.0% 8.0% 6.8% 4.5%
Kyoto University 3.3% 5.5% 5.8% 3.5%
NREL (U.S.) 10.0% 1.2% 6.1% 1.4%
Leading DSSC Research Institutions [Showing Percentages within these 11 organizations]
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Results----Profile innovation actors & activities
Cross-Data Analyses: Leading Industry “Actors”
• Organizations’ activity across these 4 databases varies a lot
• E.g., Samsung leads in publishing & patenting, but evidences little business activity in Factiva
• Dainippon Printing patents extensively, but does not publish
• Looking across different data types gains perspective
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Results----Determine potential applications
Focused DSSC Cross-Charting: Tracking Materials to Technology to Functions to Applications
• New technique – “cross-charting” to link technical attributes to functional advantages – to potential applications
• To help focus attention from “technology push” through “market pull”
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Results----Lay out alternative innovation pathways
This stage was completed in two rounds. The first round involved face-to-
face interviews with researchers at the Georgia Institute of Technology
(US), which provided input to allow a first evaluation of our analyses. The
second round entailed a campus workshop (~10 participants including ~5
with particular knowledge about nano-enhanced solar cells). This focused
on mapping likely innovation avenues.
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Results----Lay out alternative innovation pathways
Ingredients for the Multi-path Exploration
• Consolidate our empirical information
• Present in a chart showing
• To stimulate workshop discussion of Future Innovation Pathways
o Timeline (X axis)
o Innovation progression (Y axis)
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Results----Lay out alternative innovation pathways
Multi-Path Map for Dye Sensitized Solar Cells
• Depicts plausible innovation paths
• Identifies notable obstacles & opportunities along the paths
• Use to further discussion of this NEST and what to do to manage its innovation prospects
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Discussion
We have worked to various degrees at Forecasting Innovation Pathways (FIP) for several NESTs, including nano biosensors, deep brain stimulation, and nano-enhanced solar cells. This paper pursues FTA pertaining to the development of dye-sensitized solar cells (DSSCs).
- With Doug Robinson, we have tried “FIP” on several topics
o Nano biosensors
o Deep brain stimulation
o Nano enhanced solar cells (here focusing on DSSCs)
- Tingting Ma, in a related paper, investigates DSSCs through patent analyses of key technology components
- Here we share tools to identify major actors in the NEST development and to discern alternative development pathways for technology management and policy
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Discussion
- Nimble R&D profiling
- Challenge to identify key actors and innovation steps
- “Cross-charting” is our novel technique, still being refined to help do so
- 10-step process for FIP – Forecasting Innovation Pathways
- Integrates multiple empirical resources with expert contributions
- We invite your reactions?
Text Mining of Information Resources to Inform Forecasting of Innovation Pathways
Acknowledgements
This research was undertaken at Georgia Tech drawing on support from the
National Science Foundation (NSF) through the Center for Nanotechnology in
Society (Arizona State University; Award Numbers 0531194 and 0937591);
and the Science of Science Policy Program—“Measuring and Tracking
Research Knowledge Integration” (Georgia Tech; Award No. 0830207). The
findings and observations contained in this paper are those of the authors and
do not necessarily reflect the views of the National Science Foundation. We
thank Douglas Robinson and Chen Xu for their contributions.