Citations and Argumentation for Better Information Access

Add to your list(s) Download to your calendar using vCal

• Simone Teufel, Computer Laboratory
• Wednesday 14 May 2008, 14:15-15:15
• Lecture Theatre 1, Computer Laboratory.

If you have a question about this talk, please contact Timothy G. Griffin.

When publishing a scientific paper, authors pay great attention to how they talk about previous work: They cite it (in a easy-to-detect way), compare their own work to it (which is harder to detect), and make evaluative statements about its quality (which is the hardest to detect). All of these pieces of information can be invaluable when building information access systems such as summarisers or citation indexers. But without NLP , only citations are readily detectable (and Google Scholar and similar citation indexers do). My work looks at how to use discourse context and NLP to tease more information out of the text: sentiment towards citations, explicit statements of self-praise and comparison to others. Automatic annotation is based on supervised machine learning from lower-level sentential features. The data used comes from two different domains (computational linguistics and chemistry). Results are in the form of human annotation agreement, similarity of automatic and human annotations, and measures of usefulness for search.

This talk is part of the Wednesday Seminars - Department of Computer Science and Technology series.

This talk is included in these lists:

• All Talks (aka the CURE list)
• Cambridge talks
• Chris Davis' list
• Department of Computer Science and Technology talks and seminars
• Graduate-Seminars
• Guy Emerson's list
• Interested Talks
• Lecture Theatre 1, Computer Laboratory
• School of Technology
• Trust & Technology Initiative - interesting events
• Wednesday Seminars - Department of Computer Science and Technology
• bld31
• computer science
• se393's list

Note that ex-directory lists are not shown.