Clustering techniques for web mining

• Main Author: Qiu, Siyuan.
• Other Authors: Chen Lihui
• Format: Final Year Project
• Language: English
• Published: 2012
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• Online Access: http://hdl.handle.net/10356/50226
• Institution: Nanyang Technological University

With more and more high-dimensional data becoming prevalent, feature selection has been widely applied in data mining, machine learning and some other fields. The goal of feature selection is removing unneeded features because they might degrade the quality of discovered patterns. As a result, data mining process can be applied much quicker and more accurately. Various feature selection approaches in text categorization have been proposed in the literature. In this project, a Multitype Features Coselection for Web Document Clustering (MFCC) approach has been researched and implemented. MFCC is designed to improve identifying the most discriminative and remove the noisy features. In this project, other than the implementation of MFCC, we have also done the data processing which transforms the raw web documents to the format that can be used in MFCC JAVA program. Afterwards, several simulations have been conducted to test the accuracy and efficiency of MFCC.