Usage Patterns and Implementation of Random Forest Methods for Software Risk and Bugs Predictions
Alankrita Aggarwal1, Kanwalvir Singh Dhindsa2, P. K. Suri3
1Alankrita Aggarwal, Department of Computer Science and Engineering, I. K. Gujral Punjab Technical University, Jalandhar (Punjab), India.
2Kanwalvir Singh Dhindsa, Department of Computer Science and Engineering, I. K. Gujral Punjab Technical University, Jalandhar (Punjab), India.
3P. K. Suri, Department of Computer Science and Engineering, Kurukshetra University, Haryana, India.
Manuscript received on 08 August 2019 | Revised Manuscript received on 14 August 2019 | Manuscript Published on 26 August 2019 | PP: 927-932 | Volume-8 Issue-9S August 2019 | Retrieval Number: I11500789S19/19©BEIESP | DOI: 10.35940/ijitee.I1150.0789S19
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The software bugs predictions whereby the datasets of different types of bugs are evaluated for further predictions. In this research manuscript, the pragmatic evaluation of random forest approach is done and compared with results with traditional artificial neural networks (ANN) so that the results can be compared. From the outcome, the extracts from random forest are better on the accuracy level with the test datasets used in a specific format. The process of Random Forest (RF) Approach is adopted in this work that gives the effectual outcomes in most of the cases as compared to ANN and thereby the usage patterns of RF are performance aware. The paradigm of RF is used widely for the engineering optimization to solve the complex problems and generation of the dynamic trees. The outcomes and results obtained and presented in this work is giving the variations in favor random forest based optimization for the software risk management and predictive mining. The need of the proposed work and background of the study includes the effective and performance based software bugs detection. The current problem addressed includes the accuracy and multi-dimensional evaluations. The key methodology adopted here to solve the existing problem is the integration of Random Forest approach and the findings are quite effective and cavernous in assorted aspects.
Keywords: Artificial Neural Network, Random Forest Approach, Software Risk Management, Software Risk Prediction, Soft Computing for Software Bugs Prediction.
Scope of the Article: Systems and Software Engineering