Conscientious Ant Colony Optimization Based Support Vector Machine for Text Document Classification

Abstract

Document classification indicates the keyword extraction and it become a thrust research in text mining research. The main purpose of keyword extraction is to classify the documents in a more efficient manner. Misclassification of documents may lead the results to worst case. Hence, there exists a need for optimization to precede the document classification more efficiently. In this paper Conscientious Ant Colony Optimization based Support Vector Machine is proposed to classify the documents. Different keyword extraction methods are available for extracting the contents from documents. Proposed classifier is ensemble with selected keyword extraction methods to increase the classification accuracy. Results shows that the proposed classifier has got better accuracy when ensemble with different keyword extraction methods. The results show that the proposed classifier has better performance in terms of Classification Accuracy and F-Measure, than baseline classifiers.