Improved feature selection based on scatter degree
Arman Zare, Seyyed Hamed Fouladi · 2011
Feature reduction is important in machine learning, data mining and pattern recognition fields. Feature reduction consists of two methods: 1. Feature Extraction 2. Feature Selection. Feature selection methods try to select feature subset from feature set. Thus high dimension documents are projected on to lower dimension documents. The goal is selection of best subset that causes minimum error in classification. Scatter degree is one of the feature selection methods which attributes a degree of scattering for each feature. Features are selected that have higher scatter degree. In this paper, classification error has been reduced by considering other aspects in computing scatter degree (Improved Scatter Degree). Obtained results from this method have been compared with Scatter degree method.