Implementasi dan Evaluasi Explainable Recommender System pada Pemilihan Sepeda Motor Menggunakan Weighted Content-Based Filtering Berbasis Web

Yulia Darmi; Abiyyu Rahim; Rafly Nabil Lathif; Ramadhana Abdul Rasyid; Alifio Eren Atmaja; Yoga Andesmi; Tri Monica

doi:10.55606/sscj-amik.v4i4.6262

Authors

Yulia Darmi Universitas Muhammadiyah Bengkulu
Abiyyu Rahim Universitas Muhammadiyah Bengkulu
Rafly Nabil Lathif Universitas Muhammadiyah Bengkulu
Ramadhana Abdul Rasyid Universitas Muhammadiyah Bengkulu
Alifio Eren Atmaja Universitas Muhammadiyah Bengkulu
Yoga Andesmi Universitas Muhammadiyah Bengkulu
Tri Monica Universitas Muhammadiyah Bengkulu

DOI:

https://doi.org/10.55606/sscj-amik.v4i4.6262

Keywords:

Content-Based Filtering, Explainable Recommendation, Motorcycle Selection, Recommender System, Web Application

Abstract

Abstract. The increasing number and variety of motorcycles available in the market often make it difficult for prospective buyers to select a motorcycle that matches their needs and preferences. This study aims to implement and evaluate an Explainable Recommender System for motorcycle selection using the Weighted Content-Based Filtering method in a web-based environment. The proposed system utilizes motorcycle specification data and user preference profiles consisting of budget, motorcycle type, engine capacity, fuel efficiency priority, luggage capacity requirement, user height, daily usage, and touring needs. Criterion weights were obtained from questionnaire responses collected from 40 respondents and normalized before being applied in the recommendation process. Recommendation results were generated by calculating the similarity between user preferences and motorcycle specifications, followed by a weighted scoring mechanism to rank available alternatives. To improve transparency, the system provides explanations regarding the factors that influence each recommendation result. System evaluation was conducted through Black Box Testing, User Acceptance Test (UAT), and explainability evaluation involving 60 respondents. The Black Box Testing results showed that all system functions operated successfully with a success rate of 100%. The UAT evaluation achieved a score of 84.75%, while the explainability evaluation obtained a score of 81.50%, indicating that users positively perceived the usability, recommendation relevance, and explanation quality of the system. In addition, user satisfaction evaluation resulted in a score of 79.83%, indicating that the system is considered useful and feasible as a decision-support tool. The findings demonstrate that the integration of Weighted Content-Based Filtering and explainable recommendation features can generate personalized motorcycle recommendations while improving user understanding and trust in the recommendation process.

References

Adomavicius, G., & Tuzhilin, A. (2005). Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions. IEEE Transactions on Knowledge and Data Engineering, 17(6), 734–749.

Adam, S. I., & Mokodaser, W. G. (2024). Implementasi sistem rekomendasi produk e-commerce menggunakan Content-Based Filtering berbasis cosine similarity. SIMTEK: Jurnal Sistem Informasi dan Teknik Komputer, 10(2).https://doi.org/10.51876/simtek.v10i2.1665

Afzal Ziqri, & Ramadhan, N. G. (2023). Sistem rekomendasi pemilihan software berbasis Content-Based Filtering (studi kasus: PT. XYZ). Jurnal Informatika Polinema, 10(2). https://doi.org/10.33795/jip.v10i2.5008

Aggarwal, C. C. (2016). Recommender Systems: The Textbook. Cham: Springer.

Castro, J., Rodriguez, R. M., & Barranco, M. J. (2014). Weighting of features in content-based filtering with entropy and dependence measures. International Journal of Computational Intelligence Systems, 7(1), 80–89.

Chen, X., Zhang, Y., & Wen, J. R. (2022). Measuring why in recommender systems: A comprehensive survey on the evaluation of explainable recommendation. Foundations and Trends in Information Retrieval, 16(2), 123–221.

Kim, J., Kim, H., & Cho, Y. (2017). Content-based filtering for recommendation systems using multiattribute networks. Expert Systems with Applications, 89, 404–412.

Lops, P., De Gemmis, M., & Semeraro, G. (2011). Content-based recommender systems: State of the art and trends. In F. Ricci, L. Rokach, B. Shapira, & P. B. Kantor (Eds.), Recommender Systems Handbook (pp. 73–105). Springer.

Maharani, H., & Gunawan, F. A. (2014). Sistem rekomendasi mobil berdasarkan demographic dan Content-Based Filtering. Jurnal Telematika, 9(2).

Nugraha, A. A. A., & Sanjaya ER, N. A. (2023). Penyusunan sistem rekomendasi produk diecast mobil dengan metode Content-Based Filtering (CBF). Jurnal Nasional Teknologi Informasi dan Aplikasinya, 1(3), 973–976.

Pan, D., Li, X., Li, X., & Zhu, D. (2020). Explainable recommendation via interpretable feature mapping and evaluation of explainability. Proceedings of the International Conference on Artificial Intelligence and Data Analytics, 112–120.

Pressman, R. S., & Maxim, B. R. (2020). Software Engineering: A Practitioner’s Approach (9th ed.). New York: McGraw-Hill Education.

Ricci, F., Rokach, L., & Shapira, B. (2015). Recommender Systems Handbook (2nd ed.). Springer.

Suryono, A. B., Sumarlinda, S., & Oktaviani, I. (2024). Implementasi sistem rekomendasi produk mebel berbasis web menggunakan Content-Based Filtering. Jurnal Informatika Polinema, 11(4). https://doi.org/10.33795/jip.v11i4.7398

Tintarev, N., & Masthoff, J. (2015). Explaining recommendations: Design and evaluation. In F. Ricci, L. Rokach, & B. Shapira (Eds.), Recommender Systems Handbook (2nd ed., pp. 353–382). Springer.

Vultureanu-Albiși, A., & Bădică, C. (2026). Explainable and federated recommender systems: A survey and conceptual framework for trustworthy personalization. Electronics, 15(6), 1292.

Zhang, Y., & Chen, X. (2020). Explainable recommendation: A survey and new perspectives. Foundations and Trends in Information Retrieval, 14(1), 1–101.