<![CDATA[Clustering is a process of grouping documents that will form into several classes. The difference between clustering with classification is the classification will determine the class of the new document and the result is the new document will be joined into one class. In this research, clustering or grouping is used to group documents into classes based on threshold values. Several experiment is conducted to get the optimal threshold value. The optimal threshold will be used to train data clustering for naive bayes. The results of naive bayes training is used to determine the class of new document in testing phase. Results of clustering and classification depends on the words in the document, the narrower the discussion, the more accurate the results obtained from clustering and classification]]>
Copyrights © 2013
