DIVA VISUALIZATION SOFTWARE

Back to Home Page

UPDATE -- 16DEC07 --- In 2007 the DIVA software was updated and installed in a couple of academic locations for evaluation and application. Specifically, there are now routines for clustering and mapping 1) ISI's ID terms, 2) ISI's DE terms, 3) 1-word, 2-word, and 3-word title terms, 4) paper journals (journal bibliographic coupling analysis), and 5) reference journals (journal co-citation analysis). Additionally, institution and country analysis is also now available, but requires the user to extract the institutions and countries using some other program.

Presently a DIVA user's manual is under construction. You can access the user's manual here.

DIVA runs under MATLAB and requires the MATLAB Statistics and Database Toolboxes. An MSAccess database with associated Visual Basic programs is used to load and store the data from downloaded Web of Science files.

If you would like to install and try out DIVA, please contact Steven Morris by email. (email link here)

.
This software was written to provide analysis of and visulization of collections of journal articles and patents. At present it works mainly on abstracts downloaded through ISI's Web of Science product and stored in an MS Access database. It will also work with patent abstracts. DIVA will produce the following visualizations and analyses:

Maps

  • Research front timelines
  • 2D maps of papers clustered by bibliographic coupling
  • Research front to co-citation cluster crossmap
  • Research front to paper author cluster crossmap
  • Research front to author co-citation cluster crossmap
  • Reference usage plot by year
  • Reference  author usage plot by year
 Clustering

  • Clustering by co-occurence
    • Cluster by bibliographic coupling into research fronts
    • Cluster by co-citation into base reference groups
    • Cluster by author co-citation into base reference author groups
    • Cluster by co-authorship is author collaboration groupsl
  • Clustering reports
    • Research fronts
    • Co-citation clusters
    • Authors/inventors
    • Institutions
 Statistical plots

  • Reference frequency distribution
  • Cites per paper distribution
  • Paper author distribution
  • Reference author distribution
  • Paper journal distribution
  • Reference journal distribution
  • Co-citation distribution
  • Bibliographic coupling distribution
  • Bibliographic coupling clustering coefficient distribution.

MAIN MENU

The main menu is shown below:
 
 
 
MAIN MENU SUBMENUS
Project: 
  • New project
  • Open project
  • Save project
  • Save project as
  • Set database link
  • Convert patent set
 Similarity:
 
  • Read adjacency matrix from database
 Map: 
  • Make SOM 2D ordination
  • PFNET GUI
 Query
  • Highlight dots from query
Matrix:
  • Load paper to reference matrix
  • Bib to DIVA
  • Load paper to paper author matrix
  • Load paper to reference author matrix
  • Load DE terms matrix
  • Load assignee matrix
 Coocurrence: 
  • Use bibliographic coupling default
  • Use cocitation default
  • Use paper author co-occurrence default
  • Use author co-citation default
  • Make crossmaps
 Reports 
  • Research fronts
  • Co-citation clusters
  • Authors/inventors
  • Institutions
 Misc Plots
  • Cites per paper distribution
  • Co-citation distribution
  • Bibliographic coupling distribution
  • Bibliographic coupling clustering distribution
  • Reference frequency distribution

RESEARCH FRONT TIMELINES


  Timelines are used to  visualize  research fronts, which are defined as collections of papers that all deal with a common topic.

CROSSMAPS

research front to cocitation crossmap