Perbaikan Prediksi Kesalahan Perangkat Lunak Menggunakan Seleksi Fitur dan Cluster-Based Classification

Abstract

Abstract-High balance value of software fault prediction can help in conducting test effort, saving test costs, saving test resources, and improving software quality. Balance values in software fault prediction need to be considered, as in most cases, the class distribution of true and false in the software fault data set tends to be unbalanced. The balance value is obtained from trade-off between probability detection (pd) and probability false alarm (pf). Previous researchers had proposed Cluster-Based Classification (CBC) method which was integrated with EntropyBased Discretization (EBD). However, predictive models with irrelevant and redundant features in data sets can decrease balance value. This study proposes improvement of software fault prediction outcomes on CBC by integrating feature selection methods. Some feature selection methods are integrated with CBC, i.e. Information Gain (IG), Gain Ration (GR), One-R (OR), Relief-F (RFF), and Symmetric Uncertainty (SU). The result shows that combination of CBC with IG gives best average balance value, compared to other feature selection methods used in this research. Using five NASA public MDP data sets, the combination of IG and CBC generates 63.91% average of balance, while CBC method without feature selection produce 54.79% average of balance. It shows that IG can increase CBC balance average by 9.12%.Intisari-Tingginya nilai balance pada prediksi kesalahan perangkat lunak dapat membantu usaha pengujian, menghemat biaya pengujian, menghemat sumber daya pengujian, dan meningkatkan kualitas perangkat lunak. Nilai balance pada hasil prediksi kesalahan perangkat lunak perlu menjadi perhatian, karena pada umumnya, persebaran kelas true dan kelas false pada data set kesalahan perangkat lunak cenderung tidak seimbang. Nilai balance diperoleh dari hasil tarik-ulur (trade-off) antara nilai probability detection (pd) dan probability false alarm (pf). Peneliti sebelumnya mengusulkan metode Cluster-Based Classification (CBC) yang diintegrasikan dengan Entropy-Based Discretization (EBD). Namun, model prediksi dengan fitur yang redundan dan tidak relevan pada data set dapat menurunkan nilai balance. Dalam makalah ini diusulkan peningkatan hasil prediksi kesalahan perangkat lunak pada metode CBC dengan mengintegrasikan metode seleksi fitur. Adapun metode seleksi fitur akan diintegrasikan dengan CBC yaitu Perolehan Informasi (IG), Rasio Perolehan (GR), One-R (OR), Relief-F (RFF), dan Symmetric Uncertainty (SU). Hasil menunjukkan bahwa kombinasi CBC dengan IG menghasilkan nilai rata-rata balance terbaik, dibandingkan dengan keempat metode seleksi fitur lainnya. Kombinasi metode IG dan CBC menghasilkan 63,91% nilai rata-rata balance, sedangkan metode CBC tanpa seleksi fitur menghasilkan 54,79% nilai rata-rata balance. Jadi, IG dapat meningkatkan nilai rata-rata balance CBC sebesar 9,12%.Kata Kunci-Cluster-based Classification, Entropy-based Discretization, kesalahan perangkat lunak, seleksi fitur.I. PENDAHULUAN Prediksi kesalahan perangkat lunak sangat penting untuk dilakuka...