University of Illinois at Chicago
Browse

Incorporating conditional random fields and active learning to improve sentiment identification.

Download (442.46 kB)
journal contribution
posted on 2015-02-02, 00:00 authored by K. Zhang, Y. Xie, Y. Yang, A. Sun, H. Liu, . Choudhary
Many machine learning, statistical, and computational linguistic methods have been developed to identify sentiment of sentences in documents, yielding promising results. However, most of state-of-the-art methods focus on individual sentences and ignore the impact of context on the meaning of a sentence. In this paper, we propose a method based on conditional random fields to incorporate sentence structure and context information in addition to syntactic information for improving sentiment identification. We also investigate how human interaction affects the accuracy of sentiment labeling using limited training data. We propose and evaluate two different active learning strategies for labeling sentiment data. Our experiments with the proposed approach demonstrate a 5%-15% improvement in accuracy on Amazon customer reviews compared to existing supervised learning and rule-based methods.

History

Publisher Statement

NOTICE: This is the author’s version of a work that was accepted for publication in Neural Networks. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published Neural Networks, DOI: 10.1016/j.neunet.2014.04.005

Publisher

Neural Networks

issn

8805018

Issue date

2014-10-01

Usage metrics

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC