TY  - JOUR
Y1  - 2021///
VL  - 1347
UR  - https://www.scopus.com/inward/record.uri?eid=2-s2.0-85101501260&doi=10.1007%2f978-981-33-6835-4_25&partnerID=40&md5=eaf980c10a87bc10d1121b41810af161
A1  - Balogun, A.O.
A1  - Basri, S.
A1  - Abdulkadir, S.J.
A1  - Mahamad, S.
A1  - Al-momamni, M.A.
A1  - Imam, A.A.
A1  - Kumar, G.M.
JF  - Communications in Computer and Information Science
KW  - Barium compounds; Decision trees; Forecasting; NASA; Predictive analytics; Security of data; Sodium compounds
KW  -  Feature selection methods; High dimensionality; Multiple-filter methods; Prediction performance; Rank aggregation; Selection problems; Software defect prediction; Stability problem
KW  -  Feature extraction
ID  - scholars15850
N2  - With the variety of different filter methods, selecting the most appropriate filter method which gives the best performance is a difficult task. Filter rank selection and stability problems make the selection of filter methods in SDP a hard choice. The best approach is to independently apply a mixture of filter methods and evaluate the results. This study presents a novel rank aggregation-based multi-filter feature selection method to address high dimensionality and filter rank selection problems in software defect prediction. The proposed method combines the rank list generated by individual filter methods from the software defect dataset using a rank aggregation mechanism into a single aggregated rank list. The proposed method aims to resolve the filter selection problem by using multiple filter methods of diverse computational characteristics to produce a more stable (non-disjoint) and complete feature rank list better than individual filter methods employed. The effectiveness of the proposed method was evaluated by applying with Decision Tree (DT) and Naïve Bayes (NB) models on defect datasets from NASA repository. From the experimental results, the proposed method had a superior effect (positive) on the prediction performance of selected prediction models than other experimented methods. This makes the combining of individual filter rank methods a viable solution to the filter rank selection problem and enhancement of prediction models in SDP. © 2021, Springer Nature Singapore Pte Ltd.
SN  - 18650929
PB  - Springer Science and Business Media Deutschland GmbH
EP  - 383
AV  - none
SP  - 371
TI  - Rank Aggregation Based Multi-filter Feature Selection Method for Software Defect Prediction
N1  - cited By 6; Conference of 2nd International Conference on Advances in Cyber Security, ACeS 2020 ; Conference Date: 8 December 2020 Through 9 December 2020; Conference Code:254989
ER  -