Prediksi Cacat Perangkat Lunak Untuk Kelas Tidak Seimbang Menggunakan Smote Bagging dan Ripper

• MUHAMAD NICKY
• 14002453

---

ABSTRAK

ABSTRAK

Nama              : Muhamad Nicky

NIM                 : 14002453

Program Studi : Ilmu Komputer

Jenjang           : Strata Dua (S2)

Konsentrasi     : Software Engineering

Judul Tesis   : “Prediksi Cacat Perangkat Lunak Untuk Kelas Tidak Seimbang Menggunakan Smote Bagging dan Ripper”

Memprediksi cacat perangkat lunak pada tahap awal proses pengembangan sangat penting untuk memastikan kualitas dan kehandalan produk perangkat lunak. Namun, tantangan distribusi kelas yang tidak seimbang pada dataset prediksi cacat menghambat efektivitas dari algoritma klasifikasi konvensional. Penelitian ini berfokus pada optimisasi kinerja algoritma Smote Bagging dan Ripper untuk prediksi cacat perangkat lunak dalam konteks dataset yang tidak seimbang. Kualitas suatu software dipengaruhi jumlah cacat yang ditemukan sebagai acuan untuk menemukan cacat software digunakan dataset software matrick. Dataset ini memiliki ketidak seimbangan kelas (Imbalanced Class), sehingga dataset ini hanya menghasilkan kelas yang dominan saja. Untuk menangani permasalahan ini, perlu adanya suatu pedekatan untuk menanganinya. Pada penelitian ini diusulkan metode Synthetic Minority Over-Sampling Technique (SMOTE). Sebagai algoritma pada proses klasifikasi digunakan algoritma BAGGING dan RIPPER. Hasil penelitian menunjukan bahwa algoritma SMOTE+BAGGING+RIPPER lebih baik dibandingkan dengan algoritma SMOTE+BAGGING dan SMOTE+RIPPER dengan nilai rata-rata akurasi 82.00% dan 83.00% serta nilai AUC sebesar 73.00% dan 74.00% sehingga dihasilkan perbedaan akurasi sebesar 1 % dan selisih perbedaan AUC sebesar 1%. Berdasarkan hasil analisis tersebut, maka algoritma SMOTE+BAGGING+RIPPER layak digunakan untuk memprediksi tingkat kecacatan software dengan rata-rata nilai akurasi sebesar 83% dan nilai AUC sebesar 74%. Keseluruhan model mampu menggeneralisasi dengan baik.

KATA KUNCI

Software Defect,Optimize Performance,Imbalanced Class,SMOTE,Bagging and Ripper

---

DAFTAR PUSTAKA

REFERENSI

[1] M. Sonhaji Akbar and S. Rochimah, “Prediksi Cacat Perangkat Lunak Dengan Optimasi Naive Bayes Menggunakan Pemilihan Fitur Gain Ratio,” J. Sist. Dan Inform., vol. 11, no. 1, pp. 147–155, 2016.

[2] T. M. Khoshgoftaar, K. Gao, and N. Seliya, “Attribute selection and imbalanced data: Problems in software defect prediction,” Proc. - Int. Conf. Tools with Artif. Intell. ICTAI, vol. 1, pp. 137–144, 2010, doi: 10.1109/ICTAI.2010.27.

[3] B. Ma, K. Dejaeger, J. Vanthienen, and B. Baesens, “Faculty of Business and Economics 6RIWZDUH GHIHFW SUHGLFWLRQ EDVHG RQ % DRMXQ 0D . DUHO ’ HMDHJHU -DQ 9DQWKLHQHQ DQG % DUW % DHVHQV DEPARTMENT OF DECISION SCIENCES AND INFORMATION MANAGEMENT (KBI) Software defect prediction based on asso- ciation rule c,” pp. 0–7.

[4] A. Niranjan, A. Prakash, N. Veena, M. Geetha, P. Deepa Shenoy, and K. R. Venugopal, “EBJRV: An Ensemble of Bagging, J48 and Random Committee by Voting for Efficient Classification of Intrusions,” WIECONECE 2017 - IEEE Int. WIE Conf. Electr. Comput. Eng. 2017, no. December, pp. 51–54, 2018, doi: 10.1109/WIECON-ECE.2017.8468876.

[5] J. Jia, X. Xiao, B. Liu, and L. Jiao, “Bagging-based spectral clustering ensemble selection,” Pattern Recognit. Lett., vol. 32, no. 10, pp. 1456– 1467, 2011, doi: 10.1016/j.patrec.2011.04.008.

[6] Y. Wu, Y. Ding, and J. Feng, “SMOTE-Boost-based sparse Bayesian model for flood prediction,” Eurasip J. Wirel. Commun. Netw., vol. 2020, no. 1, 2020, doi: 10.1186/s13638-020-01689-2.

[7] H. He et al., “Ensemble multiboost based on ripper classifier for prediction of imbalanced software defect data,” IEEE Access, vol. 7, pp. 110333– 110343, 2019, doi: 10.1109/ACCESS.2019.2934128.

[8] D. P. Rini and S. M. Shamsuddin, “Particle Swarm Optimization: Technique, System and Challenges,” Int. J. Appl. Inf. Syst., vol. 1, no. 1, 64 Program Studi Ilmu Komputer (S2) Universitas Nusa Mandiri pp. 33–45, 2011, doi: 10.5120/ijais-3651.

[9] R. S. Wahono, N. Suryana, and S. Ahmad, “Metaheuristic Optimization based Feature Selection for Software Defect Prediction,” J. Softw., vol. 9, no. 5, 2014, doi: 10.4304/jsw.9.5.1324-1333.

[10] G. Giray, K. E. Bennin, Ö. Köksal, Ö. Babur, and B. Tekinerdogan, “On the use of deep learning in software defect prediction,” J. Syst. Softw., vol. 195, p. 111537, 2023, doi: 10.1016/j.jss.2022.111537.

[11] R. S. Wahono and N. Suryana, “Combining particle swarm optimization based feature selection and bagging technique for software defect prediction,” Int. J. Softw. Eng. its Appl., vol. 7, no. 5, pp. 153–166, 2013, doi: 10.14257/ijseia.2013.7.5.16.

[12] D. Gray, D. Bowes, N. Davey, Y. Sun, and B. Christianson, “The misuse of the NASA Metrics Data Program data sets for automated software defect prediction,” IET Semin. Dig., vol. 2011, no. 1, pp. 96–103, 2011, doi: 10.1049/ic.2011.0012.

[13] Y. S. and B. C. David Gray, David Bowes, Neil Davey, “The Misuse of the NASA Metrics Data Program Data Sets for Automated Software Defect Prediction,” Sist. Otot, vol. 11, no. 2, pp. 1–26, 2013, [Online]. Available: http://eprints.uanl.mx/5481/1/1020149995.PDF

[14] G. A. Liebchen and M. Shepperd, “Software productivity analysis of a large data set and issues of confidentiality and data quality,” Proc. - Int. Softw. Metrics Symp., vol. 2005, no. Metrics, pp. 3–5, 2005, doi: 10.1109/METRICS.2005.43.

[15] M. A. H. Ian H. Witten, Eibe Frank, “Data Mining: Practical Machine Learning Tools and Techniques,”

[16] S. Wang and X. Yao, “Using class imbalance learning for software defect prediction,” IEEE Trans. Reliab., vol. 62, no. 2, pp. 434–443, 2013, doi: 10.1109/TR.2013.2259203.

[17] J. Han, M. Kamber, and J. Pei, Data Transformation by Normalization. 65 Program Studi Ilmu Komputer (S2) Universitas Nusa Mandiri 2011. doi: 10.1016/B978-0-12-381479-1.00001-0.

[18] Florin Gorunescu, Data Mining: Concepts, models and techniques. 2011.

[19] N. Lee, “Business Intelligence,” Faceb. Nation Total Inf. Aware. Third Ed., vol. 13, pp. 101–126, 2022, doi: 10.1007/978-1-0716-1867-7_5.

[20] D. Rodriguez, I. Herraiz, R. Harrison, J. Dolado, and J. C. Riquelme, “Preliminary Comparison of Techniques for Dealing with Imbalance in Software Defect Prediction Categories and Subject Descriptors,” Proc. 18th Int. Conf. Eval. Assess. Softw. Eng. - EASE ’14, 2014.

[21] V. López, A. Fernández, and F. Herrera, “On the importance of the validation technique for classification with imbalanced datasets: Addressing covariate shift when data is skewed,” Inf. Sci. (Ny)., vol. 257, pp. 1–13, 2014, doi: 10.1016/j.ins.2013.09.038.

[22] C. W. Dawson, “Projects in Computing and Information Systems. In Information Systems Journal”.