software-defect-prediction-server

Classification server to support Software Defect Prediction System

Team Dexter

Developers: Chirag Sejpal Rushil Shah Manthan Doshi

READMe

Software Defect Prediction System is a CK metric based defect prediction model that uses K-means clustering and Ant Colony Optimization algorithm to perform data mining on the logs of various defect tracking tools.

Set-up instructions

The system consists of two components.

Front-End

https://github.com/MANTHAN-D/software-defect-prediction-system
Back-End a. Set the ip address of machine to 10.0.0.239 b. Set-up Apache tomcat 7.0 server. c. Place the "software-defect-classification-service.war" from back-end folder of Source-code directory in webapps folder of Apache tomcat. d. Start the Apache tomcat server by executing start.sh from tomcat installation directory.

Note: If both the systems are to be executed on same machine, do following and compile the back-end to generate .war file: Back-end: -- Change line 36 of file project_dir/src/main/java/dao/ModellingDataDAO.java to following: private static final String RESOURCEBASEURL = "http://localhost:3000/clusters/getData?source=";

Front-end:
	-- Change line 76 of file project_dir/routes/upload.js to following:
		url: 'http://localhost:8080/software-defect-classification-service/rest/classificationService/classify'

Bibliography:

http://openscience.us/repo/defect/ck/ (Datasets)

MANTHAN-D/software-defect-prediction-server

software-defect-prediction-server