• KOTHAGATTU RAMU
• KETHAVATH SRILATHA

Feature clustering is an efficient method for classification tasks that minimises the dimensionality of retrieved features. A Divisive Information-Theoretic Feature Clustering Algorithm for feature clustering is provided. The words in a document set's feature vector are sorted into clusters based on a similarity test. A cluster is formed by grouping words with similar meanings. Each cluster is distinguished by a membership function with a statistical mean and deviation. When some of the sentences are entered, the algorithm generates the necessary number of clusters. Each cluster has one extracted feature. A weighted combination of the cluster's phrases is used to extract the feature that corresponds to a cluster. This approach produces membership functions that closely match and properly represent the real distribution of the training data. Furthermore, the user is not required to specify the number of extracted features in advance, removing the need for trial-and-error to find the appropriate number of extracted features. Experiments show that it can extract attributes more quickly and correctly than earlier methods.

A Divisive Information-Theoretic Feature Clustering Algorithm for Text Classification

"A Divisive Information-Theoretic Feature Clustering Algorithm for Text Classification", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 2, page no.210-215, February-2019, Available :http://www.jetir.org/papers/JETIRFE06038.pdf

"A Divisive Information-Theoretic Feature Clustering Algorithm for Text Classification", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 2, page no. pp210-215, February-2019, Available at : http://www.jetir.org/papers/JETIRFE06038.pdf