Carnegie Mellon University
Browse
- No file added yet -

A Pairwise Ensemble Approach for Accurate Genre Classification

Download (291.27 kB)
journal contribution
posted on 2003-01-01, 00:00 authored by Yan Liu, Jaime G. Carbonell, Rong Jin

Text classification, whether by topic or genre, is an important task that contributes to text extraction, retrieval, summarization and question answering. In this paper we present a new pairwise ensemble approach, which uses pairwise Support Vector Machine (SVM) classifiers as base classifiers and “input-dependent latent variable” method for model combination. This new approach better captures the characteristics of genre classification, including its heterogeneous nature. Our experiments on two multi-genre collections and one topic-based classification datasets show that the pairwise ensemble method outperforms both boosting, which has been demonstrated as a powerful ensemble approach, and Error-Correcting Output Codes (ECOC), which applies pairwise-like classifiers for multiclass classification problems.

History

Date

2003-01-01

Usage metrics

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC