• SARITHA BANOTH
• RAMAKRISHNA PASULA

The feature that corresponds to a cluster is extracted using a weighted combination of the cluster's words. This method yields membership functions that closely resemble and accurately represent the training data's true distribution. Furthermore, the user is not needed to indicate the number of extracted features in advance, which eliminates the requirement for trial-and-error to determine the correct number of extracted features. Experiments demonstrate that it extracts properties faster and more accurately than previous techniques. Feature clustering is a quicker classification approach that reduces the dimensionality of recovered information. For feature clustering, a KModes Clustering is proposed. A similarity test is used to arrange the words in a content set's training samples into clusters. Words having similar meanings are grouped together to form a cluster. A membership function with a statistical mean and deviation distinguishes each cluster. The algorithm produces the required number of clusters when part of the phrases are input. Each cluster has a single selected features.

KModes Clustering for Data Categorization

"KModes Clustering for Data Categorization", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 2, page no.55-60, February-2019, Available :http://www.jetir.org/papers/JETIRFE06011.pdf

"KModes Clustering for Data Categorization", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 2, page no. pp55-60, February-2019, Available at : http://www.jetir.org/papers/JETIRFE06011.pdf