Automatically learning document taxonomies for hierarchical classification

Authors:
Kunal Punera;Suju Rajan;Joydeep Ghosh
Affiliations:
University of Texas at Austin;University of Texas at Austin;University of Texas at Austin
Venue:
WWW '05 Special interest tracks and posters of the 14th international conference on World Wide Web
Year:
2005

Citing 3
Cited 8

Making large-scale support vector machine learning practical

Advances in kernel methods
Hierarchical classification of Web content

SIGIR '00 Proceedings of the 23rd annual international ACM SIGIR conference on Research and development in information retrieval
Scalable feature selection, classification and signature generation for organizing large text databases into hierarchical topic taxonomies

The VLDB Journal — The International Journal on Very Large Data Bases

Acclimatizing Taxonomic Semantics for Hierarchical Content Classification

Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining
Topic taxonomy adaptation for group profiling

ACM Transactions on Knowledge Discovery from Data (TKDD)
Discovering relationships among categories using misclassification information

Proceedings of the 2008 ACM symposium on Applied computing
Detecting relationships among categories using text classification

Journal of the American Society for Information Science and Technology
Multi-label classification and extracting predicted class hierarchies

Pattern Recognition
A survey of hierarchical classification across different application domains

Data Mining and Knowledge Discovery
Editorial: Narrative-based taxonomy distillation for effective indexing of text collections

Data & Knowledge Engineering
Category hierarchy maintenance: a data-driven approach

SIGIR '12 Proceedings of the 35th international ACM SIGIR conference on Research and development in information retrieval

Quantified Score

Hi-index	0.00

Visualization

Abstract

While several hierarchical classification methods have been applied to web content, such techniques invariably rely on a pre-defined taxonomy of documents. We propose a new technique that extracts a suitable hierarchical structure automatically from a corpus of labeled documents. We show that our technique groups similar classes closer together in the tree and discovers relationships among documents that are not encoded in the class labels. The learned taxonomy is then used along with binary SVMs for multi-class classification. We demonstrate the efficacy of our approach by testing it on the 20-Newsgroup dataset.