Elizabeth Roman
ABSTRACT
In this paper we describe our approach to the triple ranking task of the FEIII 2017 challenge. Our method leveraged different machine learning classifiers in an ensemble as well as Thomson Reuters knowledge bases and information services to bring in external world knowledge of mentioned entities and extract information from the contextual sentences. Internal evaluation of our method was done by computing the Normalized Discounted Cumulative Gain (NDCG) as tracked by the challenge and classification accuracy. The official FEIII Challenge evaluation showed our system performed highly in single ranking of all triples, placing in 2nd or 3rd place out of 17 participants for 4 of 6 scoring variants; the system also performed above average in per role ranking for 4 of 6 average role NDCG scoring variants.
- Louiqa Raschid, Doug Burdick, Mark Flood, John Grant, Joe Langsam, Ian Soboroff, and Elena Zotkina. Financial entity identification and information integration (FEIII) challenge 2017: The report of the organizing committee. In Proceedings of the Workshop on Data Science for Macro-Modeling (DSMM@SIGMOD), 2017. Google Scholar
Digital Library
- Open permid. http://developers.thomsonreuters.com/open-permid.Google Scholar
- Data fusion community edition. http://developers.thomsonreuters.com/data-fusion.Google Scholar
- Bring structure to unstructured content. http://www.opencalais.com.Google Scholar
- Fasttext. https://research.fb.com/projects/fasttext.Google Scholar
Recommendations
Thomson Reuters' Submission to the FEIII 2017 Challenge Non-scored Tasks
DSMM'17: Proceedings of the 3rd International Workshop on Data Science for Macro--Modeling with Financial and Economic DatasetsIn this paper we describe a machine learning approach to predict roles of extracted SEC triples for the non-scored task of the 2017 FEIII Challenge.
In addition, we describe a graph and data analysis derived from SEC triples.
Thomson Reuters and the FEIII Challenge
DSMM'16: Proceedings of the Second International Workshop on Data Science for Macro-ModelingIn this paper, we describe the approach to the FEIII Challenge taken by an inter-departmental team at Thomson Reuters along with interns from MIT.
The citation advantage for open access science journals with and without article processing charges
In this study of access models, we compared citation performance in journals that do and do not levy article processing charges (APCs) as part of their business model. We used a sample of journals from the Directory of Open Access Journals (DOAJ) science ...
Comments