Future of text analysis forrester briefing


Published on

A fall 2011 briefing for personnel at Forrester in Cambridge MA.

Published in: Technology, Education
  • Be the first to comment

  • Be the first to like this

Future of text analysis forrester briefing

  1. 1. The Future of Text AnalysisDr. Stuart ShulmanTexifter, LLC<br />Wednesday, June 15, 2011<br />
  2. 2. Briefing Agenda<br />R&D in Annotation and Public Comments<br />“The Future of Text Analysis” – The vision<br />“What is DiscoverText?” – The software<br /> The Features – The basics<br />Capturing social media importing other text<br />Creating archives, buckets and datasets<br />Coding a dataset or training a classifier<br />
  3. 3. Dr. Stuart W. Shulman<br />Founder & CEO, Texifter, LLCAssistant Professor, Department of Political ScienceUniversity of Massachusetts Amherst<br />Director, Qualitative Data Analysis Program (QDAP)Associate Director, National Center for Digital Government<br />Editor, Journal of Information Technology & Politics<br />413-545-5375 stu@polsci.umass.eduhttp://people.umass.edu/stu/<br />
  4. 4.
  5. 5.
  6. 6.
  7. 7. The Future of Projects<br />Projects leverage users’ credentials to control access to documents, tools, and resources<br />Major Project Components<br />Credentials<br />Documents<br />Peers<br />Advanced ‘Social’ Search<br />Metadata<br />Networks<br />Filtering<br />Tools for Tagging<br />Shared Analysis<br />Qualitative & Quantitative Findings<br />
  8. 8. The Future of Documents<br />Import & archive data from multiple sources into a single, searchable, unified repository<br />
  9. 9. The Future of Search<br />eDiscovery will search, merge, filter & classifyunlimited amounts of text and other data<br />
  10. 10. Well Worth Reading<br />
  11. 11. The Future of Tools<br />Text processing tools will enable quicker processing and more accurate results <br />
  12. 12. The Future of Peer Relations<br />Utilize trusted peers to scale your knowledge resources, increase productivity & lower total project costs<br />
  13. 13. Peers Groups<br />Securely segment your peers into project groups by agency, firm, department, location, or affiliation,<br />while controlling their access via credentials<br />
  14. 14. Security & Credentials<br />Data will be encrypted, secure and accessible by only peers who are granted specific permissions via their credentials<br />
  15. 15. Coding, Tagging or Labeling<br />Annotation enhances your analysis by applying human interpretation to machine results<br />
  16. 16. Coding in Flexible Teams<br />
  17. 17. Crowdsourcing<br />Crowdsourcing will bring widely distributed<br />wisdom to process of text analysis<br />“This is really the biggest paradigm shift in innovation since the Industrial Revolution”<br />- MIT professor Eric von Hippel, specialist in innovation management<br />
  18. 18. Active Machine Learning<br />By utilizing information and decisions previously captured, we can enhance future machine-based decisions<br />Active <br />Learning<br />Loop<br />
  19. 19. What is DiscoverText?<br />DiscoverText is a:<br />personal or organizational archive in the cloud<br />search engine for eDiscovery <br />social media comment aggregator<br />de-duplication and near duplicate clustering engine<br />FOIA redaction toolkit<br />coding, reporting and validation team workbench<br />repository of human annotation (text about text), and<br />customizable machine-learning classifier <br />(beta launched April 2011)<br />