Title

A Review on Classification of Extracted Features from Software Requirements Specification Documents using Support Vector Machine Learning Technique

Author

1. Sadiq Waziri, Student, Abubakar Tafawa Balewa University Bauchi, Nigeria

Abstract

Manual classification of extracted features from large datasets can be tedious and time-consuming. This paper reviews the methods for classifying extracted features from SRS documents using Machine Learning (ML), with focus on linear Support Vector Machine (SVM) technique. We also explore other classification techniques, such as decision trees (DT), naïve Bayes (NB), and k-nearest neighbors (KNN)—for classifying the extracted features into mandatory and optional. Previous studies have compared different classification techniques for feature modeling. The primary goal of this review is to identify the best method for binary classification of features for software product lines engineering (SPLE). The proposed system will be tested on nine SRS documents that were chosen from the Public Requirements dataset with accuracy, precision, recall, and F1 scores used for evaluation.

Keywords

Conclusion

After implementing the proposed system, we found that SVM outperformed DT, NB, and KNN in terms of the average results shown in Table 1. This highlighted the potential of SVM as the most promising technique for feature classification in SPLE.

Table 1. Results of Performance Evaluation [22]

“Av.” means “Average”.

Future research could focus on:

1. Developing more efficient SVM algorithms for large-scale datasets.
2. Investigating the impact of different feature extraction techniques on classification performance.
3. Exploring alternative machine learning techniques that may be better suited for specific features or datasets.

Future research can contribute positively to the development of software product lines if these areas are addressed.

Author Contrubution

Sadiq Mohammed Waziri conceptualized and led the research; Fatima Umar Zambuk supervised the write-ups; Badamasi Imam Ya’u is the second supervisor and has contributed immensely to the literature review.

Funding

There was no funding received for the research work.

Conflict of Interest

The authors declare that there was no conflict of interest.

Data Sharing Statement

The data supporting the findings of this study is provided only upon reasonable request from the corresponding author.

Software And Tools Use

Acknowledgements

The authors acknowledge the Computer Science Department of Abubakar Tafawa Balewa University Bauchi for providing the necessary support required for the research.