Subscribe Now Subscribe Today
Science Alert
Curve Top
Information Technology Journal
  Year: 2009 | Volume: 8 | Issue: 8 | Page No.: 1256-1262
DOI: 10.3923/itj.2009.1256.1262
Facebook Twitter Digg Reddit Linkedin StumbleUpon E-mail

K-Means Clustering to Improve the Accuracy of Decision Tree Response Classification

S.A. Ali, N. Sulaiman, A. Mustapha and N. Mustapha

The use of deep generation with statistical-based surface generation merits from response utterances readily available from corpus. Representation and quality of the instance data are the foremost factors that affect classification accuracy of the statistical-based method. Thus, in classification task, any irrelevant or unreliable tagging of response classes represented will result in low accuracy. This study focused on improving dialogue act classification of a user utterance into a response class by clustering the semantic and pragmatic features extracted from each user utterance. A Decision tree approach is used to classify 64 mixed-initiative, transaction dialogue corpus in theater domain. The experiment shows that by using clustering technique in pre-processing stage for re-tagging response classes, the Decision tree is able to achieve 97.5% recognition accuracy in classification, better than the 81.95% recognition accuracy when using Decision tree alone.
PDF Fulltext XML References Citation Report Citation
  •    Improving Intrusion Detection Using Genetic Algorithm
  •    The RFM-based Institutional Customers Clustering: Case Study of a Digital Content Provider
  •    Flexible Navigation Strategies by Predicting Human Motion Tendency
  •    Connection Stability Analyze Based on LEACH Algorithm for Mobile Ad Hoc Network
  •    Improving Accuracy of Intention-Based Response Classification using Decision Tree
How to cite this article:

S.A. Ali, N. Sulaiman, A. Mustapha and N. Mustapha, 2009. K-Means Clustering to Improve the Accuracy of Decision Tree Response Classification. Information Technology Journal, 8: 1256-1262.

DOI: 10.3923/itj.2009.1256.1262






Curve Bottom