The readability and accuracy of any excellent classifier are two essential characteristics. Associative classifiers have lately been utilised for many classification problems, for reasons such as acceptable accuracy, fast training, and good interpretability. While features may be extremely helpful for categorization of texts, owing to the great dimensionality of text documents, both training time and the number of rules generated will substantially rise. In this article we present an algorithm for classifying texts, which comprises a selection phase for features to pick essential characteristics and a classifying phase to address this shortcoming. The experimental findings from the application of the suggested algorithm show that our method surpasses others in efficiency and performance compared with the results of a chosen well-known classification algorithm.
K Aadinath
Mar Baselios College of Engineering and Technology, India
Classification, Text dataset, Naïve Bayes, Feature Selection. | Published By :
ICTACT
Published In :
ICTACT Journal on Data Science and Machine Learning
( Volume: 2 , Issue: 3 )
Date of Publication :
June 2021
|