Transcript Technology Intelligence From Chemical Information In

T H O M S O N
Converting Information Into
Technology Intelligence
Bob Stewart
Ron Kaminecki
Thomson Scientific
S C I E N T I F I C
T H O M S O N
S C I E N T I F I C
Agenda
• Information vs. Intelligence
• Converting Information to Intelligence
• Commonly available software tools
• Value-add tools provided by commercial search
engines
• Text-mining & Data-mining tools
• Conclusions
2
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Information vs. Intelligence
• Information
• Intelligence
• Who
• Inventor
• Patent Assignee
• What
• Abstract
• IPC
• When
• Application Date
• Publication Date
• Where
• Country
•
•
•
•
3
How does it work?
Why are they doing it?
What does it mean?
How does it fit into the “big
picture?”
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Information vs. Intelligence
D
a
t
a
Organize
RANK
Spreadsheets
I
n
f
o
r
m
a
t
I
o
n
4
Analyze
Charts
Graphs
Analysis Tools
I
n
t
e
l
l
i
g
e
n
c
e
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Organize – Readily Available Tools
• Simple Tools available from commercial search
services
• Dialog RANK
• Microsoft Excel
• Create charts and summaries
5
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Organize – Readily Available Tools
• Questions answered (Information Generated)
• Top companies with patents in a technology (IPC,
subject, chemical)
• Top inventors at an organization
• Technology trends
• Technology trends in other countries
• Identify subject experts from patents, scientific
literature
• Track trends from news, press releases, business
information
6
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Organize
Using
RANK
DWPI
SHAPE()MEMORY
OR
SUPERELASTIC
OR SUPER()ELASTIC
2005:2007
Result = 2526
RANK PANAME
7
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Organize
Using
RANK
DWPI
SHAPE()MEMORY
OR
SUPERELASTIC
OR SUPER()ELASTIC
2005:2007
Result = 2526
RANK PANAME
8
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Organize
Using
RANK
9
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Advantages of Online Tools
• No additional software
• Relatively easy to use
• Inexpensive
• Usually a few cents (or less) per document RANKed
• Results can be easily imported into more
sophisticated tools
• Microsoft Excel
• Microsoft Access
10
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
XML
• eXtensible Markup Language
• Collect information once and reuse it in a variety
of ways
11
Copyright 2007 The Thomson Corporation
T H O M S O N
Web Page
Excel Spreadsheet
S C I E N T I F I C
Portal
Word Document
Access Database
XML
Data
Text/Data Mining Application
12
RSS
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Download Data as XML
13
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Convert XML to Microsoft Excel
14
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Information Derived from XML Output
Organized Organized
by Inventorby Assignee
The “Big Picture”
Details in MS Word
15
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Analysis of Patents by Application Year
16
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Analysis of Patents by Application Year
for Top Assignees
17
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Analysis of Publication Country for Top
Assignees
18
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Alternate View of Publication Country for
Top Assignees
19
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Advantages of XML Download plus
Microsoft Excel & Word
• Both Excel & Word are readily available
• Organize & Analyze
• Relatively easy to use
• Pay once for data download in XML, then analyze
as necessary at no additional cost
• Sophisticated graphing of results in Excel
• Enhanced readability in Word
20
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Analyzing
• Data Mining
• Process of analyzing data from computers and other large
relational databases (i.e. structured data)
• Text Mining
• Process by computer of extracting information from unstructured
(i.e. natural language) text and making links to form new facts or
hypotheses
• Information Visualization
• A visual approach to data analysis to reveal insights or
unexpected relationships through lists, co-occurrence matrices,
maps etc
21
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Analyzing
• Data Mining
• Process of analyzing data from computers and other large
relational databases (i.e. structured data)
• Text Mining
Ah-ha!!!
• Process by computer of extracting information from unstructured
(i.e. natural language) text and making links to form new facts or
hypotheses
• Information Visualization
• A visual approach to data analysis to reveal insights or
unexpected relationships through lists, co-occurrence matrices,
maps etc
22
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Basic Analysis Tools Available in…
• Thomson Pharma
• Web of Knowledge
23
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Advanced Analysis Tools…
• Aureka
• Thomson Data Analyzer
24
Can import data
from Dialog
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Mining the Results of a Chemical
Structure Search
InfoChem
Search Engine
retrieves substances
where group occurs
twice
25
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Structure Search in DCR
26
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Sampling of Results
27
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
MAP DCR Results to DWPI
102 individual compounds in DCR
416 patent records in DWPI
28
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Import Results into Analysis Tool
29
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Top IPC’s
30
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Most Frequently Cited (Key) Patents
31
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Most Frequently Cited (Key) Assignees
32
Copyright 2007 The Thomson Corporation
T H O M S O N
33
S C I E N T I F I C
Copyright 2007 The Thomson Corporation
T H O M S O N
34
S C I E N T I F I C
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Mining…
• Important information can be mined from
•
•
•
•
•
•
Patents
SciSearch/Web of Knowledge
Inspec
Biosis
Embase
News and business publications
• Gale Group PROMT
• Dialog NEWSROOM
35
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Curcumin in SciSearch (TDA)
36
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Curcumin in SciSearch (TDA)
37
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Curcumin in SciSearch
38
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Curcumin in SciSearch (TDA)
39
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Curcumin in Web of Knowledge
40
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Curcumin in Web of Knowledge
41
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Curcumin in Web of Knowledge
42
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Curcumin in Web of Knowledge
43
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Curcumin in Web of Knowledge
44
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Curcumin in Thomson Pharma
45
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Thomson Pharma
46
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Thomson Pharma (Smart Charts)
47
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Curcumin in Gale Group PROMT (TDA)
48
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Curcumin in Gale Group PROMT (TDA)
49
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Conclusions
• Value add tools provided by commercial search
services (e.g. RANK) enable easy, relatively
inexpensive organization of data
• Readily available software (e.g. Microsoft Excel)
enables visualization and entry to analysis
• More sophisticated tools (e.g. Thomson Data
Analyzer) provide “ah-ha!!” moments
50
Copyright 2007 The Thomson Corporation
T H O M S O N
S C I E N T I F I C
Thank You!
Bob Stewart
[email protected]
Ron Kaminecki
[email protected]
51
Copyright 2007 The Thomson Corporation