Locality preserving projection on source code metrics for improved software maintainability

Authors:
Xin Jin;Yi Liu;Jie Ren;Anbang Xu;Rongfang Bie
Affiliations:
College of Information Science and Technology, Beijing Normal University, Beijing, P.R. China;College of Information Science and Technology, Beijing Normal University, Beijing, P.R. China;College of Information Science and Technology, Beijing Normal University, Beijing, P.R. China;Image Processing & Pattern Recognition Laboratory, Beijing Normal University, Beijing, P.R. China;College of Information Science and Technology, Beijing Normal University, Beijing, P.R. China
Venue:
AI'06 Proceedings of the 19th Australian joint conference on Artificial Intelligence: advances in Artificial Intelligence
Year:
2006

Citing 10
Cited 0

C4.5: programs for machine learning

C4.5: programs for machine learning
Support-Vector Networks

Machine Learning
Measuring the software process: a practical guide to functional measurements

Measuring the software process: a practical guide to functional measurements
Software metrics for reliability assessment

Handbook of software reliability engineering
Using self-organizing maps to analyze object-oriented software measures

Journal of Systems and Software
Empirical Analysis of CK Metrics for Object-Oriented Design Complexity: Implications for Software Defects

IEEE Transactions on Software Engineering
Genetic granular classifiers in modeling software quality

Journal of Systems and Software
Statistical and computational analysis of locality preserving projection

ICML '05 Proceedings of the 22nd international conference on Machine learning
A Novel Method for Early Software Quality Prediction Based on Support Vector Machine

ISSRE '05 Proceedings of the 16th IEEE International Symposium on Software Reliability Engineering
An Artificial Immune Recognition System-based Approach to Software Engineering Management: with Software Metrics Selection

ISDA '06 Proceedings of the Sixth International Conference on Intelligent Systems Design and Applications - Volume 01

Quantified Score

Hi-index	0.00

Visualization

Abstract

Software project managers commonly use various metrics to assist in the design, maintaining and implementation of large software systems. The ability to predict the quality of a software object can be viewed as a classification problem, where software metrics are the features and expert quality rankings the class labels. In this paper we propose a Gaussian Mixture Model (GMM) based method for software quality classification and use Locality Preserving Projection (LPP) to improve the classification performance. GMM is a generative model which defines the overall data set as a combination of several different Gaussian distributions. LPP is a dimensionality deduction algorithm which can preserve the distance between samples while projecting data to lower dimension. Empirical results on benchmark dataset show that the two methods are effective.