Optimizing Performance Measures for Feature Selection

Software firms plan all development and management activities strategically to provide the best products and solutions to their user. IT professionals are involved in the process of studying the bugs reported and assign severity to make decisions regarding their resolution. To make the task fast and accurate, developers use automatic methods. Herein, the authors have used feature selection-based classification technique to decide about the severity of reported bugs. TF-IDF feature selection method is used to select the informative terms, determining the severity. Based on selected terms the support vector machine and artificial neural network classifiers are used for classification. A number of performance measures have been used to test the performance of classification. The bug reports of Eclipse project for JDT and platform products were collected from Bugzilla. The results show that classifying bugs on the basis of severity can be effectively improved by feature selection-based strategy.

Download Full-text

AN ADAPTIVE APPROACH TO FEATURE SELECTION IN PATTERN RECOGNITION

International Journal of Pattern Recognition and Artificial Intelligence ◽

10.1142/s0218001490000332 ◽

1990 ◽

Vol 04 (04) ◽

pp. 573-601

Author(s):

H.C. SHEN ◽

R. PILKEY

Keyword(s):

Pattern Recognition ◽

Feature Selection ◽

Texture Analysis ◽

Performance Measures ◽

Feature Space ◽

Selection Methods ◽

Performance Potential ◽

Analysis Problem ◽

Adaptive Scheme ◽

Ranking Scheme

Feature selection is an important phase in most pattern recognition problems, especially when the space of the extracted features is very large. Feature selection methods attempt to reduce the feature space to satisfy certain objectives. We propose the concept of defining a performance potential as a measure of the effectiveness of the set of selected features. This paper begins by outlining a ranking scheme for features based on a feature’s calculated “performance potential”. The performance potential is made up of a number of performance measures: extraction time, memory requirements, variance, covariance and classification success. An adaptive scheme is proposed to process a number of initial features and arrive at the “best” subset based on their performance potential. The approach is applied to a texture analysis problem. The results of the testing of the approach point to conclusions concerning its effectiveness.

Download Full-text

Bat algorithm and k-means techniques for classification performance improvement

Indonesian Journal of Electrical Engineering and Computer Science ◽

10.11591/ijeecs.v15.i3.pp1411-1418 ◽

2019 ◽

Vol 15 (3) ◽

pp. 1411

Author(s):

Rozlini Mohamed ◽

Munirah Mohd Yusof ◽

Noorhaniza Wahid ◽

Norhanifah Murli ◽

Muhaini Othman

Keyword(s):

Feature Selection ◽

Performance Improvement ◽

Performance Measures ◽

Selection Process ◽

Bat Algorithm ◽

Classification Performance ◽

Feature Selection Technique ◽

Selection Technique ◽

Discretization Technique

This paper presents Bat Algorithm and K-Means techniques for classification performance improvement. The objective of this study is to investigate efficiency of Bat Algorithm in discrete dataset and to find the optimum feature in discrete dataset. In this study, one technique that comprise the discretization technique and feature selection technique have been proposed. Our contribution is in two process of classification: pre-processing and feature selection process. First, to proposed discretization techniques called as BkMD, where we hybrid Bat Algorithm technique and K-Means classifier. Second, to proposed BkMDFS as feature selection technique where Bat Algorithm is embed into BkMD. In order to evaluate our proposed techniques, 14 continuous dataset from various applications are used in experiment. From the experiment, results show that BkMDFS outperforms in most performance measures. Hence it shows that, Bat Algorithm have potential to be one of the discretization technique and feature selection technique.

Download Full-text