Entity-relation search: context pattern driven extraction and indexing
Zhang, Zequn
Loading…
Permalink
https://hdl.handle.net/2142/95610
Description
Title
Entity-relation search: context pattern driven extraction and indexing
Author(s)
Zhang, Zequn
Issue Date
2016-12-05
Director of Research (if dissertation) or Advisor (if thesis)
Chang, Kevin C.
Department of Study
Computer Science
Discipline
Computer Science
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
M.S.
Degree Level
Thesis
Keyword(s)
Relation search
Relation indexing
Abstract
Our research focuses on searching relations between entities with context constraints. In particular, we are interested in efficiently searching for the relations among medical entities (e.g. diseases, chemicals, species, genes, or mutations) in a professional medical corpus. Existing relation extraction systems, like OpenIE, are able to extract some relations between entities. However, its results are inseparable in terms of extraction contexts, which prevents it from being able to search for the relations of given contexts.
To address this issue, we propose to build an entity-relation search system with an awareness of extraction contexts. In order to achieve this goal, we propose to extract and index contexts for each extracted relation. We evaluate our search model over millions of professional medical abstracts and show that our context indexing is effective to support the task of searching relations into contexts.
Note that this rich and novel system is the product of a collaborative team effort: Tianxiao Zhang, Jiarui Xu and Varun Berry, and supervised by Professor Kevin Chang. While we separately document our individual contributions, we intentionally share some parts of our thesis to improve the readability of our overall system design. This thesis mainly focuses on the design of our context extraction and indexing method.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
