Paper: Understanding Tables in Context Using Standard NLP Toolkits

ACL ID P13-2116
Title Understanding Tables in Context Using Standard NLP Toolkits
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2013
Authors

Tabular information in text documents contains a wealth of information, and so tables are a natural candidate for in- formation extraction. There are many cues buried in both a table and its sur- rounding text that allow us to under- stand the meaning of the data in a ta- ble. We study how natural-language tools, such as part-of-speech tagging, dependency paths, and named-entity recognition, can be used to improve the quality of relation extraction from ta- bles. In three domains we show that (1) a model that performs joint probabilis- tic inference across tabular and natural language features achieves an F1 score that is twice as high as either a pure- table or pure-text system, and (2) us- ing only shallower features or non-joint inference results in lower quality.