Browse

The Virtual University, Pakistan’s first University based completely on modern Information and Communication Technologies, was established by the Government as a public sector, not-for-profit institution with a clear mission: to provide extremely affordable world class education to aspiring students all over the country.

Using free-to-air satellite television broadcasts and the Internet, the Virtual University allows students to follow its rigorous programs regardless of their physical locations. It thus aims at alleviating the lack of capacity in the existing universities while simultaneously tackling the acute shortage of qualified professors in the country. By identifying the top Professors of the country, regardless of their institutional affiliations, and requesting them to develop and deliver hand-crafted courses, the Virtual University aims at providing the very best courses to not only its own students but also to students of all other universities in the country.

  1. Home
  2. Thesis Detail

SOFTWAREDEFECT PREDICTION MODEL USING MULTI-LAYER FEED FORWARDNEURAL NETWORKS

Download

Author: FASEEHA MATLOOB

Citable URI : https://vspace.vu.edu.pk/detail.aspx?id=342

Publisher : Virtual University

Date Issued: 7/4/2020 12:00:00 AM

Abstract

Software defect prediction techniques are being focused by many researchers due to its effectiveness for cost reduction in testing process. Most of the software defect dataset contains uncleaned, noisy, high dimensional and imbalance data. These problems reduce the prediction accuracy of a classifier. In this paper, a framework is proposed which combines approaches to deal with all these problems. This framework comprises of four stages 1) data preprocessing 2) feature selection (FS) 3) class balancing, and 4) classification through ensemble learning. Normalization is performed on cleaned datasets. Multilayer perceptron (MLP) is used as subset evaluator in FS process with six search methods Best first (BF), Greedy Stepwise (GS), Genetic Search (GA), Particle swam Optimization Search (PSO), Rank Search (RS) and Linear forward selection (LFS). Resample and SMOTE algorithm are used for class balancing. In classification stacking ensemble model is applied to build model on 80% of the input data. Here meta classifier is set to MLP and base classifiers include decision tree (J48), Random forest (RF), Support vector machine (SVM), K nearest neighbor (kNN) and Bayes Net (BN). Parameter tuning of meta and base classifier is also performed. Performance is evaluated on NASA MDP using precision, recall, F-measure, MCC, ROC and accuracy. Results have shown that the proposed method showed significant improvement is defect prediction compared to base classifiers.


URI : https://vspace.vu.edu.pk/details.aspx?id=342

Citation: Matloob,F(2019).SOFTWAREDEFECT PREDICTION MODEL USING MULTI-LAYER FEED FORWARDNEURAL NETWORKS. Virtual University of Pakistan(Lahore, Pakistan)

Version : Final Version

Terms of Use :

Detailed Terms :

Journal :

Files in this item
Name Size Format
Fall 2019_CS720_MS160400300.pdf 402kb pdf

Copyright 2016 © Virtual University of Pakistan