An exploratory study of news article clustering for web-based bio-surveillance

Authors:
Manabu Torii;Burt-Ujin Bayarsaikhan;Hongfang Liu;Thang Nguyen;Kevin Jones;Noele P.> Nelson;David M. Hartley
Affiliations:
Georgetown University Medical Center, Washington, DC, USA;Georgetown University Medical Center, Washington, DC, USA;Georgetown University Medical Center, Washington, DC, USA;Georgetown University Medical Center, Washington, DC, USA;Georgetown University Medical Center, Washington, DC, USA;Georgetown University Medical Center, Washington, DC, USA;Georgetown University Medical Center, Washington, DC, USA
Venue:
Proceedings of the 1st ACM International Health Informatics Symposium
Year:
2010

Citing 8
Cited 0

A Comparative Study on Feature Selection in Text Categorization

ICML '97 Proceedings of the Fourteenth International Conference on Machine Learning
MiTAP, text and audio processing for bio-security: a case study

Eighteenth national conference on Artificial intelligence
Introduction to Information Retrieval

Introduction to Information Retrieval
Semantic Text Classification of Emergent Disease Reports

PKDD 2007 Proceedings of the 11th European conference on Principles and Practice of Knowledge Discovery in Databases
BNS feature scaling: an improved representation over tf-idf for svm text classification

Proceedings of the 17th ACM conference on Information and knowledge management
BioCaster

Bioinformatics
The role of roles in classifying annotated biomedical text

BioNLP '07 Proceedings of the Workshop on BioNLP 2007: Biological, Translational, and Clinical Language Processing
Automatic online news monitoring and classification for syndromic surveillance

Decision Support Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Online news articles provide rich and timely information for disease outbreak surveillance. Meanwhile, it is not trivial to search articles relevant to disease outbreaks among the large volume of online publications. In this study, we examined the use of text clustering techniques to organize online articles. To take into account surveillance analysts' expertise in clustering articles, we considered selection of informative word features in a supervised manner. Our experiments suggest that the supervised selection of features can significantly reduce the features size without affecting the utility of resulting clusters. In addition, we observed that the clustering algorithm could yield consistent results when a small number of selected features were used.