Towards re-defining relation understanding in financial domain

  • Chenguang Wang
  • , Doug Burdick
  • , Laura Chiticariu
  • , Rajasekar Krishnamurthy
  • , Yunyao Li
  • , Huaiyu Zhu

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

We describe our experiences in participating in the scored task for the 2017 FEIII Data Challenge. Our approach is to model the problem as a binary classification problem and train an ensemble model leveraging domain features that capture financial terminology. We share challenge results for our submission, which performed well achieving the highest score in four out of six evaluation criteria. We describe semantic complexities encountered with regards to the task definition and ambiguities in the labeled dataset. We present an alternative task formulation Relationship Validation that addresses some of these semantic complexities and demonstrate how our approach naturally extends to this simplified task definition.

Original languageEnglish
Title of host publicationProceedings of the 3rd International Workshop on Data Science for Macro-Modeling with Financial and Economic Datasets, DSMM 2017 - In conjunction with the ACM SIGMOD/PODS Conference
PublisherAssociation for Computing Machinery, Inc
ISBN (Electronic)9781450350310
DOIs
StatePublished - May 14 2017
Event3rd International Workshop on Data Science for Macro-Modeling with Financial and Economic Datasets, DSMM 2017 - Chicago, United States
Duration: May 14 2017 → …

Publication series

NameProceedings of the 3rd International Workshop on Data Science for Macro-Modeling with Financial and Economic Datasets, DSMM 2017 - In conjunction with the ACM SIGMOD/PODS Conference

Conference

Conference3rd International Workshop on Data Science for Macro-Modeling with Financial and Economic Datasets, DSMM 2017
Country/TerritoryUnited States
CityChicago
Period05/14/17 → …

Keywords

  • FEIII
  • Financial Domain
  • Information Extraction
  • Relation Understanding
  • Text Classification

Fingerprint

Dive into the research topics of 'Towards re-defining relation understanding in financial domain'. Together they form a unique fingerprint.

Cite this