Software Effort Estimation Menggunakan Logarithmic Fuzzy Preference Programming dan Linear Regression

• I GUSTI BAGUS ARYA PRADNJA PARAMITHA
• 14210149

---

ABSTRAK

ABSTRAK

Nama              : I Gusti Bagus Arya Pradnja Paramitha

NIM                 : 14210149

Program Studi : Ilmu Komputer

Fakultas           : Teknologi Informasi

Jenjang            : Strata Dua (S2)

Konsentrasi      : Software Engineering

Judul             : “Software Effort Estimation Menggunakan Logarithmic Fuzzy Preference Programming dan Linear Regression”

Mengetahui perkiraan biaya dan waktu yang diperlukan saat membangun perangkat lunak sangat penting. Dalam penelitian sebelumnya, banyak orang menggunakan fuzzy atau machine learning saja untuk meningkatkan kinerja. fuzzy muncul sebagai pendekatan yang cocok untuk pengambilan keputusan multifaktor, sedangkan machine learning banyak digunakan untuk membuat model prediksi. Untuk mencapai hasil yang lebih baik dari upaya, penelitian ini menggabungkan pembelajaran mesin dan fuzzy. Untuk mencapai tujuan ini, penelitian ini menggunakan metode pemrograman preferensi fuzzy logarithmik (LFPP) dan Linear Regression serta dataset yang digunakan adalah COCOMO dana NASA. LFPP digunakan untuk mencari bobot kriteria dan langkah – langkah yang dilakukan adalah melakukan penentuan kritera Software Effort Estimation, lalu melakukan perbandingan kepentingan antar kriteria, melakukan perhitungan indeks konsistensi dan melakukan perhitungan bobot kriteria setelah itu dataset akan di masukan ke program machine learning berbasis Python. Dataset dibagi menjadi dua, yaitu data training dan data uji dengan perbandingan 80:20 dan pengembangan software effort menggunakan Linear Regression. Untuk pengujian kinerja menggunakan Magnitude Means of Relative Error (MMRE) dan Root Mean Square Error (RMSE). Dari metode yang di usulkan, hasil penelitian ini menunjukkan bahwa Prediksi usaha dalam pengembangan perangkat lunak dengan nilai MMRE sebesar 0,00107 dan nilai RMSE sebesar 0,0328 untuk dataset COCOMO, sedangkan pada dataset NASA didapatkan nilai MMRE sebesar 0.0487 dan RMSE sebesar 0,22068. Hasil dari penelitian ini lebih baik dibandingkan penelitian - penelitian sebelumnya.

KATA KUNCI

Software Effort Estimation,LOGARITHMIC FUZZY PREFERENCE PROGRAMMING (LFPP),Linear Regression,Magnitu

---

DAFTAR PUSTAKA

DAFTAR REFERENSI

[1] A. Trendowicz, J. Münch, and R. Jeffery, “State of the Practice in Software Effort Estimation: A Survey and Literature Review,” Int. Fed. Inf. Process., pp. 232–245, 2015.

[2] A. Purwanto and L. P. Manik, “Software Effort Estimation Using Logarithmic Fuzzy Preference Programming and Least Squares Support Vector Machines,” Sci. J. Informatics, vol. 10, no. 1, pp. 1–12, 2023, doi: 10.15294/sji.v10i1.39865.

[3] A. B. Nassif, M. Azzeh, A. Idri, and A. Abran, “Software Development Effort Estimation Using Regression Fuzzy Models,” Comput. Intell. Neurosci., vol. 2019, 2019, doi: https://doi.org/10.1155/2019/8367214.

[4] H. Azath, M. Mohanapriya, and S. Rajalakshmi, “Software Effort Estimation Using Modified Fuzzy C Means Clustering and Hybrid ABC-MCS Optimization in Neural Network,” J. Intell. Syst., vol. 29, no. 1, pp. 251–263, 2020, doi: 10.1515/jisys-2017- 0121.

[5] M. Kakkar, H. Rastogi, and S. Dhankar, “A Survey on Software Effort Estimation Techniques,” Int. Conf. Conflu. Next Gener. Inf. Technol. Summit, pp. 826–830, 2014.

[6] M. Baiquni, R. Sarno, Sarwosri, and Sholiq, “Improving the Accuracy of COCOMO II Using Fuzzy Logic and Local Calibration Method,” Int. Conf. Sci. Inf. Technol., pp. 284–289, 2017.

[7] R. K. Yadav and S. Niranjan, “OPTIMIZED MODEL FOR SOFTWARE EFFORT ESTIMATION USING COCOMO-2 METRICS WITH FUZZY LOGIC,” Int. J. Adv. Res. Comput. Sci., vol. 8, no. 7, pp. 121–125, 2017.

[8] R. V. S. Prathap and A. Suneetha, “SOFTWARE QUALITY PREDICTION BY USING MACHINE LEARNING ALGORITHMS,” Int. Res. J. Mod. Eng. Technol. Sci., no. 11, pp. 1217–1222, 2022, doi: DOI?: https://www.doi.org/10.56726/IRJMETS31403.

[9] A. G. Priya Varshini, K. Anitha Kumari, and V. Varadarajan, “Estimating software development efforts using a random forest-based stacked ensemble approach,” Electron., vol. 10, no. 10, pp. 1–19, 2021, doi: 10.3390/electronics10101195.

[10] A. ÇEB? and H. KARAL, “An application of fuzzy analytic hierarchy process ( FAHP ) for evaluating students ’ project,” Educ. Res. Rev., vol. 12(3), no. May, pp. 120– 132, 2017, doi: 10.5897/ERR2016.3065.

[11] S. K. Sehra, Y. S. Brar, and N. Kaur, “Applying Fuzzy-AHP for software effort estimation in data scarcity,” Int. J. Eng. Trends Technol., no. March, pp. 4–9, 2017.

[12] S. K. Sehra, Y. Singh, B. Navdeep, K. Sukhjit, and S. Sehra, “Software effort estimation using FAHP and weighted kernel LSSVM machine,” Soft Comput., vol. 23, no. 21, pp. 10881–10900, 2019, doi: 10.1007/s00500-018-3639-2. 60 Program Studi Ilmu Komputer (S2) Universitas Nusa Mandiri

[13] T. Wahyuningrum, Azhari, and Suprapto, “A Comparison of Extent Analysis and Fuzzy Preference Programming for Ebvaluating B2C Website Usability,” ICCSE, vol. 8th, pp. 6–11, 2018.

[14] T. Wahyuningrum and R. Pandiya, “Logaritmic Fuzzy Preference Programming Approach for Evaluating University Ranking Optimization,” Sci. J. Informatics, vol. 4, no. 1, pp. 1–7, 2017.

[15] M. Salaheldin, “Solution of fuzzy analytic hierarchy process using simulation,” Conf. Pap., no. January 2009, 2018, doi: 10.13140/RG.2.1.2892.3049.

[16] S. Do, T. Lutovac, and M. Kali, “Fuzzy AHP approach to passenger aircraft type selection zi,” J. Air Transp. Manag., 2017, doi: 10.1016/j.jairtraman.2017.08.003.

[17] T. Wahyuningrum, A. Azhari, and Suprapto Suprapto, “An Extended Consistent Fuzzy Preference Relation to Evaluating Website Usability,” Int. J. Adv. Comput. Sci. Appl., vol. 10, no. 9, pp. 109–116, 2019.

[18] T. Wahyuningrum, Azhari Azhari, and Suprapto Suprapto, “Modified LFPP to Improve the Accuracy of Matrix Pairwise Comparison Consistency Index in the Usability Evaluation,” Int. J. Intell. Eng. Syst., vol. 13, no. 3, 2020, doi: 10.22266/ijies2020.0630.36.

[19] X. Yang and W. Wen, “Ridge and Lasso Regression Models for Cross-Version Defect Prediction,” IEEE Trans. Reliab., vol. 67, no. 3, pp. 885–896, 2018, doi: 10.1109/TR.2018.2847353.

[20] L. ZHANG, X. WEI, J. LU, and J. PAN, “Lasso regression: From explanation to predection,” Adv. Psychol. Sci., vol. 28, no. 10, pp. 1777–1788, 2020, doi: 10.3724/sp.j.1042.2020.01777.

[21] T. O. Hodson, “Root-mean-square error (RMSE) or mean absolute error (MAE): when to use them or not,” Geosci. Model Dev., vol. 15, no. 14, pp. 5481–5487, 2022, doi: 10.5194/gmd-15-5481-2022.

[22] L. P. Manik, “Features Engineering With Generating Genetic Algorithm for Software Effort Estimation,” ICIC Express Lett. Part B Appl., vol. 13, no. 1, pp. 1–9, 2022, doi: 10.24507/icicelb.13.01.1.

[23] V. Fonti and E. Belitser, “Feature selection using lasso,” VU Amsterdam Res. Pap. Bus. Anal., vol. 30, pp. 1–25, 2017, [Online]. Available: https://www.researchgate.net/profile/David-Booth-7/post/Regression-ofpairwise-trait-similarity-on-similarity-in-personalattributes/attachment/5b18368d4cde260d15e3a4e3/AS%3A634606906785793% 401528313485788/download/werkstuk-fonti_tcm235-836234.pdf

[24] E. Papatheocharous, H. Papadopoulos, and A. S. Andreou, “Software Effort Estimation with Ridge Regression and Evolutionary Attribute Selection,” arXiv Prepr. arXiv\;1012.5754, 2010.

[25] Y. Wang and K. Chin, “Fuzzy analytic hierarchy process?: A logarithmic fuzzy preference programming methodology q,” Int. J. Approx. Reason., vol. 52, no. 4, pp. 541–553, 2011, doi: 10.1016/j.ijar.2010.12.004. 60 Program Studi Ilmu Komputer (S2) Universitas Nusa Mandiri

[26] T. Chai and R. R. Draxler, “Root mean square error (RMSE) or mean absolute error (MAE)? -Arguments against avoiding RMSE in the literature,” Geosci. Model Dev., vol. 7, no. 3, pp. 1247–1250, 2014, doi: 10.5194/gmd-7-1247-2014.

[27] J. Aczbl and T. L. Saaty, “Procedures for Synthesizing Ratio Judgements,” J. Math. Psychol., vol. 102, pp. 93–102, 1983.