Observations on Fault Proneness Prediction Models of Object-Oriented System to Improve Software Quality

Dharmendra Lal Gupta, Anil Kumar Malviya


Software quality is the fundamental need of industry and also for a user. And the future business and reputation of the company
depends on the quality of the product. It is need of today to develop quality product. Software quality of system can be measured in terms of
fault-proneness of data. Effective prediction for the fault-proneness class early in software development plays a very important role in the
analysis of the software quality and balance of software cost.
For the effective prediction of fault-proneness the Pareto Principle will be helpful because it implies that 80% of all errors uncovered
during testing will be likely be traceable to 20% of all program components. The problem of course is to isolate these suspected components and
thoroughly test them. In this paper, we proposed fault-prone and not-fault-prone class using discriminant analysis , neural network and logistic
regression and prediction accuracy was measured and compared for each prediction system.



Keywords: Object-oriented, Reliability, Artificial Neural Networks, Fault, and Failure.

Full Text:


DOI: https://doi.org/10.26483/ijarcs.v2i2.367


  • There are currently no refbacks.

Copyright (c) 2016 International Journal of Advanced Research in Computer Science