论文信息 - An Ensemble-Based Named Entity Recognition Solution for Detecting Consumer Products

An Ensemble-Based Named Entity Recognition Solution for Detecting Consumer Products

This paper presents a technical description of a solution for International Conference on Data Mining 2012 Contest - Consumer Products number 1. The Contest provided a dataset including thousands of text items, a product catalog with over fifteen million products, and hundreds of manually annotated product mentions to support data-driven approaches. The task was to identify product mentions within a large user-generated web-based textual corpus and disambiguate the mentions against the large product catalog. The solution consists of an ensemble-based algorithm for processing a textual content. It uses Conditional Random Fields and a special approach which recognizes product mentions. This solution finished in the third place in the contest.

Lukasz Romaszko

[1] Thomas Hofmann,et al. Large Margin Methods for Structured and Interdependent Output Variables , 2005, J. Mach. Learn. Res..

[2] Burr Settles,et al. Biomedical Named Entity Recognition using Conditional Random Fields and Rich Feature Sets , 2004, NLPBA/BioNLP.

[3] Andrew McCallum,et al. Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data , 2001, ICML.

[4] Christopher D. Manning,et al. Incorporating Non-local Information into Information Extraction Systems by Gibbs Sampling , 2005, ACL.

[5] D. S. Kushwaha,et al. A Comparative Study of Named Entity Recognition for Hindi Using Sequential Learning Algorithms , 2009, 2009 IEEE International Advance Computing Conference.