Learning to classify short and sparse text & web with hidden topics from large-scale data collections

Learning to classify short and sparse text & web with hidden topics from large-scale data collections (2008)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Xuan-hieu Phan , Susumu Horiguchi

Venue:	in WWW
Citations:	78 - 2 self

BibTeX

@INPROCEEDINGS{Phan08learningto,
    author = {Xuan-hieu Phan and Susumu Horiguchi},
    title = {Learning to classify short and sparse text & web with hidden topics from large-scale data collections},
    booktitle = {in WWW},
    year = {2008}
}

OpenURL

Abstract

This paper presents a general framework for building classifiers that deal with short and sparse text & Web segments by making the most of hidden topics discovered from largescale data collections. The main motivation of this work is that many classification tasks working with short segments of text & Web, such as search snippets, forum & chat messages, blog & news feeds, product reviews, and book & movie summaries, fail to achieve high accuracy due to the data sparseness. We, therefore, come up with an idea of gaining external knowledge to make the data more related as well as expand the coverage of classifiers to handle future data better. The underlying idea of the framework is that for each classification task, we collect a large-scale external data collection called “universal dataset”, and then build a classifier on both a (small) set of labeled training data and a rich set of hidden topics discovered from that data collection. The framework is general enough to be applied to different data domains and genres ranging from Web search results to medical text. We did a careful evaluation on several hundred megabytes of Wikipedia (30M words) and MEDLINE (18M words) with two tasks: “Web search domain disambiguation” and “disease categorization for medical text”, and achieved significant quality enhancement.

Keyphrases

hidden topic large-scale data collection sparse text web medical text book movie summary web search result blog news feed text web web search domain disambiguation chat message main motivation data collection product review short segment different data domain rich set sparse text web segment large-scale external data collection many classification task search snippet high accuracy largescale data collection universal dataset data sparseness general framework labeled training data external knowledge disease categorization future data classification task underlying idea careful evaluation significant quality enhancement