A new classification technique based on hybrid fuzzy soft set theory and supervised fuzzy c-means

Handaga, Bana (2013) A new classification technique based on hybrid fuzzy soft set theory and supervised fuzzy c-means. Doctoral thesis, Universiti Tun Hussein Onn Malaysia.


Download (857kB) | Preview
[img] Text (Copyright Declaration)
Restricted to Repository staff only

Download (275kB) | Request a copy
[img] Text (Full Text)
Restricted to Registered users only

Download (29MB) | Request a copy


Recent advances in information technology have led to significant changes in today‟s world. The generating and collecting data have been increasing rapidly. Popular use of the World Wide Web (www) as a global information system led to a tremendous amount of information, and this can be in the form of text document. This explosive growth has generated an urgent need for new techniques and automated tools that can assist us in transforming the data into more useful information and knowledge. Data mining was born for these requirements. One of the essential processes contained in the data mining is classification, which can be used to classify such text documents and utilize it in many daily useful applications. There are many classification methods, such as Bayesian, K-Nearest Neighbor, Rocchio, SVM classifier, and Soft Set Theory used to classify text document. Although those methods are quite successful, but accuracy and efficiency are still outstanding for text classification problem. This study is to propose a new approach on classification problem based on hybrid fuzzy soft set theory and supervised fuzzy c-means. It is called Hybrid Fuzzy Classifier (HFC). The HFC used the fuzzy soft set as data representation and then using the supervised fuzzy c-mean as classifier. To evaluate the performance of HFC, two well-known datasets are used i.e., 20 Newsgroups and Reuters-21578, and compared it with the performance of classic fuzzy soft set classifiers and classic text classifiers. The results show that the HFC outperforms up to 50.42% better as compared to classic fuzzy soft set classifier and up to 0.50% better as compare classic text classifier.

Item Type: Thesis (Doctoral)
Subjects: Q Science > QA Mathematics
Q Science > QA Mathematics > QA150-272.5 Algebra
Divisions: Faculty of Computer Science and Information Technology > Department of Software Engineering
Depositing User: Mrs. Sabarina Che Mat
Date Deposited: 31 Oct 2021 04:01
Last Modified: 31 Oct 2021 04:01
URI: http://eprints.uthm.edu.my/id/eprint/2198

Actions (login required)

View Item View Item