SlideShare a Scribd company logo
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
Visualizing PLoS Data
DHUG 2014
February 11, 2014
Kevin Boyack, Bob Kasenchak, Margie Hlava
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
2
AGENDA
Background
Data
Science mapping
Mapping thesaurus directly
Mapping thesaurus indirectly
» SciTech S&T map
» Thesaurus map
Overlays
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
3
BACKGROUND
To answer questions posed by PLoS using PLoS index (thesaurus)
terms, a map of the thesaurus space is desirable
» Where is the coverage thin, where is it dense?
» What areas are core and which are emerging trends? Which fields are growing
quickly?
» Which fields are interrelated?
» Which fields have low activity but are related to very active fields?
Some questions require more than PLoS data
» How does the coverage in PLoS journals relate to coverage in other databases?
A map of the thesaurus space can function as a template upon which
various distributions can be overlaid
Some questions are more amenable to tables than maps
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
4
DATA
Thesaurus
» 10,551 unique terms in 15,164 locations
» 11 top level terms
» 296 second level terms in 305 locations
PLoS article data parsed / indexed by Access Innovations
» Article level data through mid-2013; 84,989 records
» 72,796 records with index (thesaurus) terms; 171,181 terms
Scopus data
» SciTech Strategies licenses use of Scopus data from Elsevier
» 1996-2011; 25M records
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
5
THESAURUS STRUCTURE
Biology CS Earth Env Eng Medicine People Physical Methods SS
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
6
SCIENCE MAPPING
A map is a visual representation of the relationships within a
» Classification system (e.g., journal categories, taxonomy)
» Document corpus
Science maps can/have been created using
» Documents, Journals, Authors, Terms, Taxonomies
30-40 year tradition of science mapping
» Well-established methodologies
» Current computing power and data availability enable large scale mapping and
analysis
Maps used for communication, strategy, planning, evaluation …
» Understanding of structure, relative location
» Basis for metrics
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
7
GENERAL PROCESS
Select data source and objects (e.g., docs) of interest Scopus / WoS
PubMed
Patents
Your own data
Calculate similarity between objects (doc-doc) Citations (links)
Words (title/abstr)
Create a visual layout (map) of the objects / clusters Pajek / Gephi
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
8
MAPPING PLOS TERMS DIRECTLY
Mapped second level terms
» All terms rolled up to the second level term at the top of the branch
Cosine similarity between terms based on co-occurrence in articles
Map created using Fruchterman-Reingold mapping routine in Pajek
» Based on only the top 3 links per term – strongest linkages
Each node (term) colored using top level terms
» Biology and life sciences (light green)
» Medicine and health sciences (salmon)
» Physical sciences (purple)
» Research and analysis methods (yellow)
» Social sciences (light orange)
» …
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
9
MAPPING PLOS TERMS
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
10
MAPPING PLOS TERMS DIRECTLY
Resulting map is difficult to use
» Using second level terms leaves out needed detail
» Term counts span several orders of magnitude – hard to map well
» Terms within the same branch of the thesaurus don’t cluster well
Resulting map doesn’t account well for context (areas where PLoS has
low coverage)
Need a map that does a better job of providing context, and that better
matches our current picture of science
So … we decided to index Scopus data using the PLoS rule-base and
map based on that more inclusive context
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
11
MAPPING PLOS TERMS INDIRECTLY
Use existing map of all of science
Triangulate term positions
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
12
SCITECH S&T MAP
20 million articles (1996-2011)
3 million patents (1996-2011)
220,000 document clusters
Cluster contents using citation
Cluster positions using text
The visual is the final step
(last 10%); the full process is
needed for robust analysis
Template on which other info
can be overlaid; documents (and
related info) tied to map positions
We are the only ones in the
world who do contextual analysis
at this scale
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
13
TRIANGULATION
Use full S&T map as basis
Based on Scopus 2007-2011
Locate each thesaurus term as the
average position of its documents
The resulting map contains 11,934
terms; full coverage; interpretable
Terms with multiple colors appear
multiple times
The term map becomes the basemap
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
14
THESAURUS BASEMAP
The term map becomes the basemap
Circle size - # documents
Circle color – top level term
Many small circles – detailed thesaurus
and rule base that differentiates
Few large circles – broad terms and
rule base that doesn’t differentiate
Location/size of terms facilitates questions
- Is term well located in the thesaurus?
- Is the rule base acting as desired?
- Should term be broken up?
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
15
NOT JUST PRETTY PICTURES
All visualizations are based on tabular data
Pictures help form hypotheses
Detailed analysis provides answers
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
16
ANALYSES
Biology and life sciences
Computer and information sciences
Earth sciences
Ecology and environmental sciences
Engineering and technology
Medical and health sciences
Physical sciences
Research and analysis methods
People and places
Social sciences
Science policy
Circuit models
Chromatin immunoprecipitation
Evolutionary modeling
Electron transport chain
Intelligence
Language
Cognition
Decision making
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
17
ANALYSES
Biology and life sciences
Computer and information sciences
Earth sciences
Ecology and environmental sciences
Engineering and technology
Medical and health sciences
Physical sciences
Research and analysis methods
People and places
Social sciences
Science policy
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
18
ANALYSES
Biology and life sciences
Computer and information sciences
Earth sciences
Ecology and environmental sciences
Engineering and technology
Medical and health sciences
Physical sciences
Research and analysis methods
People and places
Social sciences
Science policy
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
19
ANALYSES
Biology and life sciences
Computer and information sciences
Earth sciences
Ecology and environmental sciences
Engineering and technology
Medical and health sciences
Physical sciences
Research and analysis methods
People and places
Social sciences
Science policy
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
20
ANALYSES
Biology and life sciences
Computer and information sciences
Earth sciences
Ecology and environmental sciences
Engineering and technology
Medical and health sciences
Physical sciences
Research and analysis methods
People and places
Social sciences
Science policy
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
21
ANALYSES
Biology and life sciences
Computer and information sciences
Earth sciences
Ecology and environmental sciences
Engineering and technology
Medical and health sciences
Physical sciences
Research and analysis methods
People and places
Social sciences
Science policy
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
22
ANALYSES
Biology and life sciences
Computer and information sciences
Earth sciences
Ecology and environmental sciences
Engineering and technology
Medical and health sciences
Physical sciences
Research and analysis methods
People and places
Social sciences
Science policy
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
23
ANALYSES
Biology and life sciences
Computer and information sciences
Earth sciences
Ecology and environmental sciences
Engineering and technology
Medical and health sciences
Physical sciences
People and places
Social sciences
Science policy
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
24
ANALYSES
Biology and life sciences
Computer and information sciences
Earth sciences
Ecology and environmental sciences
Engineering and technology
Medical and health sciences
Physical sciences
Research and analysis methods
People and places
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
25
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
26
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
27
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
28
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
29
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
30
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
31
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
32
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
33
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
34
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
35
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
36
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
37
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
38
ANALYSES
PLoS – All journals and years
PLoS – All journals (2008)
PLoS – All journals (2009)
PLoS – All journals (2010)
PLoS – All journals (2011)
PLoS – All journals (2012)
PLoS – All journals (2013)
PLoS Biology
PLoS Computational Biology
PLoS Genetics
PLoS Medicine
PLoS NTD
PLoS Pathogens
PLoS One
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
39
ANALYSES
PLoS One – All years
PLoS One (2008)
PLoS One (2009)
PLoS One (2010)
PLoS One (2011)
PLoS One (2012)
PLoS One (2013)
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
40
ANALYSES
PLoS One – All years
PLoS One (2008)
PLoS One (2009)
PLoS One (2010)
PLoS One (2011)
PLoS One (2012)
PLoS One (2013)
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
41
ANALYSES
PLoS One – All years
PLoS One (2008)
PLoS One (2009)
PLoS One (2010)
PLoS One (2011)
PLoS One (2012)
PLoS One (2013)
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
42
ANALYSES
PLoS One – All years
PLoS One (2008)
PLoS One (2009)
PLoS One (2010)
PLoS One (2011)
PLoS One (2012)
PLoS One (2013)
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
43
ANALYSES
PLoS One – All years
PLoS One (2008)
PLoS One (2009)
PLoS One (2010)
PLoS One (2011)
PLoS One (2012)
PLoS One (2013)
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
44
ANALYSES
PLoS One – All years
PLoS One (2008)
PLoS One (2009)
PLoS One (2010)
PLoS One (2011)
PLoS One (2012)
PLoS One (2013)
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
45
ANALYSES
PLoS One (2007-2011)
PLoS One (2008)
PLoS One (2009)
PLoS One (2010)
PLoS One (2011)
PLoS One (2012)
PLoS One (2013)
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
46
ANALYSES
2007-2011
PLoS One
PNAS
JACS
J Mat Chem
Phys Rev Lett
J Nanosci Nanotechn
LNCS
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
47
ANALYSES
2007-2011
PLoS One
PNAS
JACS
J Mat Chem
Phys Rev Lett
J Nanosci Nanotechn
LNCS
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
48
ANALYSES
2007-2011
PLoS One
PNAS
JACS
J Mat Chem
Phys Rev Lett
J Nanosci Nanotechn
LNCS
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
49
ANALYSES
2007-2011
PLoS One
PNAS
JACS
J Mat Chem
Phys Rev Lett
J Nanosci Nanotechn
LNCS
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
50
ANALYSES
2007-2011
PLoS One
PNAS
JACS
J Mat Chem
Phys Rev Lett
J Nanosci Nanotechn
LNCS
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
51
ANALYSES
2007-2011
PLoS One
PNAS
JACS
J Mat Chem
Phys Rev Lett
J Nanosci Nanotechn
LNCS
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
52
ANALYSES
2007-2011
PLoS One
PNAS
JACS
J Mat Chem
Phys Rev Lett
J Nanosci Nanotechn
LNCS
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
53
SUMMARY
Mapping the PLoS thesaurus directly (using co-occurrence) gave maps
that were only partially useful.
Indirect mapping of the PLoS thesaurus using a comprehensive map of
the scientific literature (based on Scopus data) was very successful.
The resulting visualization of the PLoS thesaurus became a stand-alone
basemap, a template that can be used to:
» Examine the thesaurus structure, content, and level of detail.
» Show coverage and trends for various entities such as journals, institutions, authors,
etc.
This thesaurus mapping effort helped visualize which areas of the
PLOS thesaurus were covered well, or not well enough, as well as
which fields were emerging or well-established.
All overlays based on tabular data, so detailed analysis is possible.
Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents
54
QUESTIONS
Thank-you for your attention !

More Related Content

PPT
Biology spring 2012_cell_biology
PDF
Jonathan Tedds Distinguished Lecture at DLab, UC Berkeley, 12 Sep 2013: "The ...
PDF
How many citations are there in the Data Citation Index
PPT
Atypical combinations are confounded by disciplinary effects (Boyack & Klavans)
PPTX
Cancer Research Meets Data Science — What Can We Do Together?
PPTX
One View of Data Science
PPT
Moving From Small Science To Big Science
Biology spring 2012_cell_biology
Jonathan Tedds Distinguished Lecture at DLab, UC Berkeley, 12 Sep 2013: "The ...
How many citations are there in the Data Citation Index
Atypical combinations are confounded by disciplinary effects (Boyack & Klavans)
Cancer Research Meets Data Science — What Can We Do Together?
One View of Data Science
Moving From Small Science To Big Science

Similar to PLOS Visualization Project (20)

PPT
Taxonomy and Corpus Assessment: Using Visualization
PPTX
Optimising the use of existing knowledge
PPTX
Observation of The Hot Research on Disruptive Science and Technology via Scie...
PDF
Using Open Source Tools for Visualization and Semantic Mapping in a Large Sca...
PPTX
Visualizing Scientific Data - LATAM Faculty Summit 2011
PPTX
Text Mining, Term Mining, and Visualization - Improving the Impact of Scholar...
PPTX
Ontologies: Necessary, but not sufficient
PPTX
Visual exploration of scientific literature using VOSviewer and CitNetExplorer
PDF
Semantic Journal Mapping for Search Visualization in a Large Scale Article Di...
PDF
Dynamic Potential of Semantic Enrichment
PDF
Science Mapping and Research Positioning
PPTX
Found in Space: Creating and Visualizing IEEE Abstract Space for Publication ...
PDF
Data visualisations: drawing actionable insights from science and technology ...
PDF
II-SDV 2012 Text Mining, Term Mining and Visualization - Improving the Impac...
PPTX
Topic Maps - Human-oriented semantics?
PDF
The Student's and Researcher's Guide to Discovery: Exploring Scientific Field...
PDF
Multiple perspectives on bibliometric data
PPTX
Overview of Bibliometrics - IAP Course version 1.1
PDF
Scientific Knowledge Graphs: an Overview
PPTX
Found in Space: Creating and Visualizing IEEE Document Space
Taxonomy and Corpus Assessment: Using Visualization
Optimising the use of existing knowledge
Observation of The Hot Research on Disruptive Science and Technology via Scie...
Using Open Source Tools for Visualization and Semantic Mapping in a Large Sca...
Visualizing Scientific Data - LATAM Faculty Summit 2011
Text Mining, Term Mining, and Visualization - Improving the Impact of Scholar...
Ontologies: Necessary, but not sufficient
Visual exploration of scientific literature using VOSviewer and CitNetExplorer
Semantic Journal Mapping for Search Visualization in a Large Scale Article Di...
Dynamic Potential of Semantic Enrichment
Science Mapping and Research Positioning
Found in Space: Creating and Visualizing IEEE Abstract Space for Publication ...
Data visualisations: drawing actionable insights from science and technology ...
II-SDV 2012 Text Mining, Term Mining and Visualization - Improving the Impac...
Topic Maps - Human-oriented semantics?
The Student's and Researcher's Guide to Discovery: Exploring Scientific Field...
Multiple perspectives on bibliometric data
Overview of Bibliometrics - IAP Course version 1.1
Scientific Knowledge Graphs: an Overview
Found in Space: Creating and Visualizing IEEE Document Space
Ad

More from Access Innovations, Inc. (20)

PPTX
Understanding How Search Works November 7 2024.pptx
PDF
Supercharge your AI - SSP Industry Breakout Session 2024-v2_1.pdf
PPTX
Eureka, I found it! - Special Libraries Association 2021 Presentation
PDF
Making AI Behave: Using Knowledge Domains to Produce Useful, Trustworthy Results
PPTX
ISO 25964-1Working Group ISO/TC 46/SC 9/WG 8
PPTX
PDF
Plos taxonomy beyond search dhug 2021
PPTX
Hindawi taxonomy and personalization 27.10 (1)
PPTX
Data harmonycloudpowerpointclientfacing
PPTX
Data harmony update 2021
PPTX
Atypon dhug2021
PPTX
Asco using ai-taxos-for meta-titles-february-2021
PPTX
Asce more than just topic taxonomies
PPTX
Acs discoverability-dhug2021
PDF
Ai webinar 2 -what's in a name (consolidated pdf)
PDF
Tagging overview - Why Keywords Don't Cut It
PPTX
Health Affairs - Why Keywords Don't Cut It
PDF
Why Keywords Don't Cut It
PPTX
Data Harmony update 2020 final
PPTX
Data Harmony Update 2020 final
Understanding How Search Works November 7 2024.pptx
Supercharge your AI - SSP Industry Breakout Session 2024-v2_1.pdf
Eureka, I found it! - Special Libraries Association 2021 Presentation
Making AI Behave: Using Knowledge Domains to Produce Useful, Trustworthy Results
ISO 25964-1Working Group ISO/TC 46/SC 9/WG 8
Plos taxonomy beyond search dhug 2021
Hindawi taxonomy and personalization 27.10 (1)
Data harmonycloudpowerpointclientfacing
Data harmony update 2021
Atypon dhug2021
Asco using ai-taxos-for meta-titles-february-2021
Asce more than just topic taxonomies
Acs discoverability-dhug2021
Ai webinar 2 -what's in a name (consolidated pdf)
Tagging overview - Why Keywords Don't Cut It
Health Affairs - Why Keywords Don't Cut It
Why Keywords Don't Cut It
Data Harmony update 2020 final
Data Harmony Update 2020 final
Ad

PLOS Visualization Project

  • 1. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents Visualizing PLoS Data DHUG 2014 February 11, 2014 Kevin Boyack, Bob Kasenchak, Margie Hlava
  • 2. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 2 AGENDA Background Data Science mapping Mapping thesaurus directly Mapping thesaurus indirectly » SciTech S&T map » Thesaurus map Overlays
  • 3. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 3 BACKGROUND To answer questions posed by PLoS using PLoS index (thesaurus) terms, a map of the thesaurus space is desirable » Where is the coverage thin, where is it dense? » What areas are core and which are emerging trends? Which fields are growing quickly? » Which fields are interrelated? » Which fields have low activity but are related to very active fields? Some questions require more than PLoS data » How does the coverage in PLoS journals relate to coverage in other databases? A map of the thesaurus space can function as a template upon which various distributions can be overlaid Some questions are more amenable to tables than maps
  • 4. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 4 DATA Thesaurus » 10,551 unique terms in 15,164 locations » 11 top level terms » 296 second level terms in 305 locations PLoS article data parsed / indexed by Access Innovations » Article level data through mid-2013; 84,989 records » 72,796 records with index (thesaurus) terms; 171,181 terms Scopus data » SciTech Strategies licenses use of Scopus data from Elsevier » 1996-2011; 25M records
  • 5. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 5 THESAURUS STRUCTURE Biology CS Earth Env Eng Medicine People Physical Methods SS
  • 6. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 6 SCIENCE MAPPING A map is a visual representation of the relationships within a » Classification system (e.g., journal categories, taxonomy) » Document corpus Science maps can/have been created using » Documents, Journals, Authors, Terms, Taxonomies 30-40 year tradition of science mapping » Well-established methodologies » Current computing power and data availability enable large scale mapping and analysis Maps used for communication, strategy, planning, evaluation … » Understanding of structure, relative location » Basis for metrics
  • 7. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 7 GENERAL PROCESS Select data source and objects (e.g., docs) of interest Scopus / WoS PubMed Patents Your own data Calculate similarity between objects (doc-doc) Citations (links) Words (title/abstr) Create a visual layout (map) of the objects / clusters Pajek / Gephi
  • 8. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 8 MAPPING PLOS TERMS DIRECTLY Mapped second level terms » All terms rolled up to the second level term at the top of the branch Cosine similarity between terms based on co-occurrence in articles Map created using Fruchterman-Reingold mapping routine in Pajek » Based on only the top 3 links per term – strongest linkages Each node (term) colored using top level terms » Biology and life sciences (light green) » Medicine and health sciences (salmon) » Physical sciences (purple) » Research and analysis methods (yellow) » Social sciences (light orange) » …
  • 9. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 9 MAPPING PLOS TERMS
  • 10. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 10 MAPPING PLOS TERMS DIRECTLY Resulting map is difficult to use » Using second level terms leaves out needed detail » Term counts span several orders of magnitude – hard to map well » Terms within the same branch of the thesaurus don’t cluster well Resulting map doesn’t account well for context (areas where PLoS has low coverage) Need a map that does a better job of providing context, and that better matches our current picture of science So … we decided to index Scopus data using the PLoS rule-base and map based on that more inclusive context
  • 11. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 11 MAPPING PLOS TERMS INDIRECTLY Use existing map of all of science Triangulate term positions
  • 12. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 12 SCITECH S&T MAP 20 million articles (1996-2011) 3 million patents (1996-2011) 220,000 document clusters Cluster contents using citation Cluster positions using text The visual is the final step (last 10%); the full process is needed for robust analysis Template on which other info can be overlaid; documents (and related info) tied to map positions We are the only ones in the world who do contextual analysis at this scale
  • 13. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 13 TRIANGULATION Use full S&T map as basis Based on Scopus 2007-2011 Locate each thesaurus term as the average position of its documents The resulting map contains 11,934 terms; full coverage; interpretable Terms with multiple colors appear multiple times The term map becomes the basemap
  • 14. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 14 THESAURUS BASEMAP The term map becomes the basemap Circle size - # documents Circle color – top level term Many small circles – detailed thesaurus and rule base that differentiates Few large circles – broad terms and rule base that doesn’t differentiate Location/size of terms facilitates questions - Is term well located in the thesaurus? - Is the rule base acting as desired? - Should term be broken up?
  • 15. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 15 NOT JUST PRETTY PICTURES All visualizations are based on tabular data Pictures help form hypotheses Detailed analysis provides answers
  • 16. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 16 ANALYSES Biology and life sciences Computer and information sciences Earth sciences Ecology and environmental sciences Engineering and technology Medical and health sciences Physical sciences Research and analysis methods People and places Social sciences Science policy Circuit models Chromatin immunoprecipitation Evolutionary modeling Electron transport chain Intelligence Language Cognition Decision making
  • 17. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 17 ANALYSES Biology and life sciences Computer and information sciences Earth sciences Ecology and environmental sciences Engineering and technology Medical and health sciences Physical sciences Research and analysis methods People and places Social sciences Science policy
  • 18. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 18 ANALYSES Biology and life sciences Computer and information sciences Earth sciences Ecology and environmental sciences Engineering and technology Medical and health sciences Physical sciences Research and analysis methods People and places Social sciences Science policy
  • 19. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 19 ANALYSES Biology and life sciences Computer and information sciences Earth sciences Ecology and environmental sciences Engineering and technology Medical and health sciences Physical sciences Research and analysis methods People and places Social sciences Science policy
  • 20. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 20 ANALYSES Biology and life sciences Computer and information sciences Earth sciences Ecology and environmental sciences Engineering and technology Medical and health sciences Physical sciences Research and analysis methods People and places Social sciences Science policy
  • 21. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 21 ANALYSES Biology and life sciences Computer and information sciences Earth sciences Ecology and environmental sciences Engineering and technology Medical and health sciences Physical sciences Research and analysis methods People and places Social sciences Science policy
  • 22. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 22 ANALYSES Biology and life sciences Computer and information sciences Earth sciences Ecology and environmental sciences Engineering and technology Medical and health sciences Physical sciences Research and analysis methods People and places Social sciences Science policy
  • 23. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 23 ANALYSES Biology and life sciences Computer and information sciences Earth sciences Ecology and environmental sciences Engineering and technology Medical and health sciences Physical sciences People and places Social sciences Science policy
  • 24. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 24 ANALYSES Biology and life sciences Computer and information sciences Earth sciences Ecology and environmental sciences Engineering and technology Medical and health sciences Physical sciences Research and analysis methods People and places
  • 25. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 25 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 26. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 26 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 27. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 27 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 28. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 28 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 29. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 29 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 30. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 30 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 31. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 31 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 32. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 32 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 33. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 33 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 34. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 34 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 35. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 35 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 36. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 36 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 37. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 37 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 38. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 38 ANALYSES PLoS – All journals and years PLoS – All journals (2008) PLoS – All journals (2009) PLoS – All journals (2010) PLoS – All journals (2011) PLoS – All journals (2012) PLoS – All journals (2013) PLoS Biology PLoS Computational Biology PLoS Genetics PLoS Medicine PLoS NTD PLoS Pathogens PLoS One
  • 39. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 39 ANALYSES PLoS One – All years PLoS One (2008) PLoS One (2009) PLoS One (2010) PLoS One (2011) PLoS One (2012) PLoS One (2013)
  • 40. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 40 ANALYSES PLoS One – All years PLoS One (2008) PLoS One (2009) PLoS One (2010) PLoS One (2011) PLoS One (2012) PLoS One (2013)
  • 41. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 41 ANALYSES PLoS One – All years PLoS One (2008) PLoS One (2009) PLoS One (2010) PLoS One (2011) PLoS One (2012) PLoS One (2013)
  • 42. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 42 ANALYSES PLoS One – All years PLoS One (2008) PLoS One (2009) PLoS One (2010) PLoS One (2011) PLoS One (2012) PLoS One (2013)
  • 43. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 43 ANALYSES PLoS One – All years PLoS One (2008) PLoS One (2009) PLoS One (2010) PLoS One (2011) PLoS One (2012) PLoS One (2013)
  • 44. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 44 ANALYSES PLoS One – All years PLoS One (2008) PLoS One (2009) PLoS One (2010) PLoS One (2011) PLoS One (2012) PLoS One (2013)
  • 45. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 45 ANALYSES PLoS One (2007-2011) PLoS One (2008) PLoS One (2009) PLoS One (2010) PLoS One (2011) PLoS One (2012) PLoS One (2013)
  • 46. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 46 ANALYSES 2007-2011 PLoS One PNAS JACS J Mat Chem Phys Rev Lett J Nanosci Nanotechn LNCS
  • 47. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 47 ANALYSES 2007-2011 PLoS One PNAS JACS J Mat Chem Phys Rev Lett J Nanosci Nanotechn LNCS
  • 48. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 48 ANALYSES 2007-2011 PLoS One PNAS JACS J Mat Chem Phys Rev Lett J Nanosci Nanotechn LNCS
  • 49. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 49 ANALYSES 2007-2011 PLoS One PNAS JACS J Mat Chem Phys Rev Lett J Nanosci Nanotechn LNCS
  • 50. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 50 ANALYSES 2007-2011 PLoS One PNAS JACS J Mat Chem Phys Rev Lett J Nanosci Nanotechn LNCS
  • 51. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 51 ANALYSES 2007-2011 PLoS One PNAS JACS J Mat Chem Phys Rev Lett J Nanosci Nanotechn LNCS
  • 52. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 52 ANALYSES 2007-2011 PLoS One PNAS JACS J Mat Chem Phys Rev Lett J Nanosci Nanotechn LNCS
  • 53. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 53 SUMMARY Mapping the PLoS thesaurus directly (using co-occurrence) gave maps that were only partially useful. Indirect mapping of the PLoS thesaurus using a comprehensive map of the scientific literature (based on Scopus data) was very successful. The resulting visualization of the PLoS thesaurus became a stand-alone basemap, a template that can be used to: » Examine the thesaurus structure, content, and level of detail. » Show coverage and trends for various entities such as journals, institutions, authors, etc. This thesaurus mapping effort helped visualize which areas of the PLOS thesaurus were covered well, or not well enough, as well as which fields were emerging or well-established. All overlays based on tabular data, so detailed analysis is possible.
  • 54. Physics Chemistry Engineering Biology Disease Medicine Brain Health Social Computer Patents 54 QUESTIONS Thank-you for your attention !