The MSR Systems for Entity Linking and Temporal Slot Filling at TAC 2013

The paper describes the two systems from MSR that participated in the English Entity Linking and Temporal Slot Filling (TSF) tasks at TAC 2013. The entity linking system is built by using the same framework and architecture as the previous iterations that participated in the TAC 2011 and 2012 evaluations; therefore, its description focuses on the components that are novel with reference to those systems. The MSR system that addresses the newly introduced TSF task, employs a distant-supervision framework, in which language models for each targeted type of relation are built based on a corpus of sentences automatically extracted from Wikipedia text as guided by Wikipedia infobox data. For each of the TAC tracks addressed, we submitted two runs, which obtained the highest scores in their corresponding evaluations.

[1]  Christopher J. C. Burges,et al.  From RankNet to LambdaRank to LambdaMART: An Overview , 2010 .

[2]  Luis Gravano,et al.  Snowball: extracting relations from large plain-text collections , 2000, DL '00.

[3]  Dan Roth,et al.  Relational Inference for Wikification , 2013, EMNLP.

[4]  Avirup Sil,et al.  Extracting Action and Event Semantics from Web Text , 2010, AAAI Fall Symposium: Commonsense Knowledge.

[5]  David Yarowsky,et al.  A method for disambiguating word senses in a large corpus , 1992, Comput. Humanit..

[6]  Doug Downey,et al.  Web-scale information extraction in knowitall: (preliminary results) , 2004, WWW '04.

[7]  Andrew R. Golding,et al.  A Bayesian Hybrid Method for Context-sensitive Spelling Correction , 1996, VLC@ACL.

[8]  Avirup Sil,et al.  Extracting STRIPS Representations of Actions and Events , 2011, RANLP.

[9]  Heng Ji,et al.  Overview of the TAC 2010 Knowledge Base Population Track , 2010 .

[10]  Avirup Sil,et al.  Machine Reading Between the Lines: A Simple Evaluation Framework for Extracted Knowledge Bases , 2011 .

[11]  David Yarowsky,et al.  One Sense Per Discourse , 1992, HLT.

[12]  Silviu Cucerzan.  MSR System for Entity Linking at TAC 2012 , 2012, TAC.

[13]  Silviu Cucerzan,et al.  Large-Scale Named Entity Disambiguation Based on Wikipedia Data , 2007, EMNLP.

[14]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[15]  Angel X. Chang,et al.  SUTime: A library for recognizing and normalizing time expressions , 2012, LREC.

[16]  Lan Nie,et al.  Resolving Surface Forms to Wikipedia Topics , 2010, COLING.