• Dr.E Chandra Blessie
• Deepa A

Document classification using keyword extraction has become an important domain for researchers in text mining. The main intention of keyword extraction is to represent the document in a precise manner. The compatible details of documents acts as a multiple applications in different ways. Currently, the document classification based on keywords has become a main task. Most available classifiers are well suited only for the dataset which have minimum number of documents. In this paper, Fuzzy Logic Based Compatible Support Vector Machine (FL-CSVM) is proposed to classify the text documents in maximum accuracy. FL-CCSVM is designed to perform classification by dividing the documents dataset into multiple parts in a random manner, where the previous classification algorithm follows sequential manner classification leading to poor classification accuracy. The existing classifiers gives better results only when the dataset is small, but FL-CSVM is designed to accept the dataset with any size to give increased classification accuracy. To evaluate the proposed classifiers performance against previous classifiers this work chooses three text document collection dataset namely ACM Document collection dataset, Reuters-21578, and NBA Input document collection dataset holding 3506, 21578, and 1256 documents respectively. The results shows that FL-CSVM is having better performance in terms of Classification Accuracy and F-Measure, than baseline classifiers

Index Terms: Classification, Mining, Text, NBA,ACM, Reuters

"FUZZY LOGIC BASED COMPATIBLE SUPPORT VECTOR MACHINE (FL-CSVM) FOR TEXT DOCUMENT CLASSIFICATION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.436-446, June 2019, Available :http://www.jetir.org/papers/JETIR1907E63.pdf

"FUZZY LOGIC BASED COMPATIBLE SUPPORT VECTOR MACHINE (FL-CSVM) FOR TEXT DOCUMENT CLASSIFICATION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp436-446, June 2019, Available at : http://www.jetir.org/papers/JETIR1907E63.pdf