SISTEM REKOMENDASI CROSS-SELLING BERBASIS ITEM SIMILARITY MENGGUNAKAN COSINE SIMILARITY PADA TRANSAKSI PENJUALAN

Erni Seniwati; Acihmah Sidauruk; Ninik Tri Hartanti; Rina Pramitasari

doi:10.30606/rjocs.v12i2.4938

Authors

Erni Seniwati Universitas Langlangbuana Bandung
Acihmah Sidauruk Univeritas Amikom Yogyakarta
Ninik Tri Hartanti Univeritas Amikom Yogyakarta
Rina Pramitasari Univeritas Amikom Yogyakarta

DOI:

https://doi.org/10.30606/rjocs.v12i2.4938

Keywords:

Cosine Similarity, cross-selling, Item Similarity, sistem rekomendasi, transaksi penjualan.

Abstract

Data transaksi ritel dapat dimanfaatkan untuk mengidentifikasi keterkaitan antarproduk dan mendukung strategi cross-selling, terutama ketika data rating, profil, atau identitas pelanggan tidak tersedia. Penelitian ini bertujuan membangun baseline sistem rekomendasi cross-selling berbasis Item Similarity menggunakan Cosine Similarity dari pola co-occurrence pada transaksi penjualan. Dataset terdiri atas 77 transaksi dan 296 produk. Setelah pembersihan data, transaksi ditransformasikan menjadi matriks transaksi-item biner dan ditransposisi menjadi vektor item-transaksi untuk menghitung kemiripan antarproduk dan menghasilkan rekomendasi Top-N. Evaluasi menggunakan protokol dua tingkat: leave-one-transaction-out, yaitu mengeluarkan seluruh transaksi uji dari data pembentuk matriks similarity untuk mencegah kebocoran informasi, kemudian leave-one-item-out di dalam transaksi uji, yaitu menahan satu item sebagai target dan menggunakan item lainnya sebagai masukan. Sebanyak 71 transaksi memenuhi syarat dan menghasilkan 368 kasus uji. Pasangan B148 dan B151 memperoleh Cosine Similarity 0,894. Pada Top-10, model menghasilkan Hit Rate/Recall@10 sebesar 8,15%, Precision@10 sebesar 0,82%, dan Mean Reciprocal Rank sebesar 0,040. Performa yang relatif rendah berkaitan dengan sparsity matriks sebesar 98,36% dan dominasi produk berfrekuensi rendah. Hasil ini menunjukkan bahwa Item Similarity berbasis transaksi dapat digunakan sebagai baseline yang sederhana dan interpretable untuk cross-selling pada ritel dengan data terbatas, sekaligus menegaskan perlunya data interaksi yang lebih kaya untuk meningkatkan kualitas rekomendasi.

Downloads

Download data is not yet available.

References

F. Pratama, S. Prastio, E. Seniwati, N. T. Hartanti, dan Sudarmanto, "Content-Based Filtering untuk Sistem Rekomendasi Produk E-Commerce," The Indonesian Journal of Computer Science Research (IJCSR), vol. 5, no. 1, pp. 86–98, 2026, doi: 10.59095/ijcsr.v5i1.254.

Y. S. Kim, H. Hwangbo, H. J. Lee, et al., "Sequence Aware Recommenders for Fashion E-Commerce," Electronic Commerce Research, vol. 24, pp. 2733–2753, 2024, doi: 10.1007/s10660-022-09627-8.

A. Bellogín, I. Cantador, et al., "Similarity Measures for Collaborative Filtering-Based Recommender Systems: Review and Experimental Comparison," Journal of King Saud University – Computer and Information Sciences, vol. 34, no. 9, pp. 7645–7669, 2022.

H. I. Abdalla, A. A. Amer, Y. A. Amer, L. Nguyen, et al., "Boosting the Item-Based Collaborative Filtering Model with Novel Similarity Measures," International Journal of Computational Intelligence Systems, vol. 16, Art. 123, 2023, doi: 10.1007/s44196-023-00299-2.

Hartatik, R. R. Isnanto, B. Warsito, N. Firdaus, and F. Y. A'La, "Towards the Application of Artificial Intelligence: Cosine Similarity in Recommendation Systems Based on Collaborative Filtering," in Proceedings of the 7th International Conference of Computer and Informatics Engineering (IC2IE 2024), IEEE, 2024, doi: 10.1109/IC2IE63342.2024.10747946.

Y. Wang, J. Wu, Z. Wu, dan G. Adomavicius, "Efficient and Flexible Long-Tail Recommendation Using Cosine Patterns," INFORMS Journal on Computing, vol. 37, no. 2, pp. 446–464, 2024, doi: 10.1287/ijoc.2022.0194.

A. P. Putra, D. P. S. Putri, dan A. A. K. A. Wiranatha, "Scientific Paper Recommendation System: Application of Sentence Transformers and Cosine Similarity Using arXiv Data," Journal of Applied Informatics and Computing, vol. 9, no. 4, 2025.

S. I. Adam dan W. G. Mokodaser, "Implementasi Sistem Rekomendasi Produk E-Commerce Menggunakan Content-Based Filtering Berbasis Cosine Similarity," SIMTEK: Jurnal Sistem Informasi dan Teknik Komputer, vol. 10, no. 2, pp. 418–424, 2025.

E. W. Pratomo dan E. Utami, "Analisis dan Implementasi Content-Based Filtering dengan Cosine Similarity untuk Sistem Rekomendasi Tugas Akhir Mahasiswa," AITI, vol. 23, no. 2, 2026.

A. Alsobhi dan N. Amare, "Ontology-Based Relational Product Recommendation System," Computational Intelligence and Neuroscience, 2022, Article ID 1591044.

M. Li, S. Jullien, M. Ariannezhad, and M. de Rijke, “A Next Basket Recommendation Reality Check,” ACM Transactions on Information Systems, vol. 41, no. 4, Art. 116, 2023, doi: 10.1145/3587153.

A. Romanov, O. Lashinin, M. Ananyeva, and S. Kolesnikov, “Time-Aware Item Weighting for the Next Basket Recommendations,” in Proceedings of the 17th ACM Conference on Recommender Systems (RecSys ’23), 2023, doi: 10.1145/3604915.3608859.

R. Esmeli, M. Bader-El-Den, H. Abdullahi, et al., “Implicit Feedback Awareness for Session Based Recommendation in E-Commerce,” SN Computer Science, vol. 4, Art. 320, 2023, doi: 10.1007/s42979-023-01752-x.

A. S. G. et al., “Session-Based Recommendations for e-Commerce with Graph-Based Data Modeling,” Applied Sciences, vol. 13, no. 1, Art. 394, 2023.

P. Kumar, M. K. Gupta, C. R. S. Rao, M. Bhavsingh, and M. Srilakshmi, “A Comparative Analysis of Collaborative Filtering Similarity Measurements for Recommendation Systems,” International Journal on Recent and Innovation Trends in Computing and Communication, vol. 11, no. 3s, pp. 184–192, 2023, doi: 10.17762/ijritcc.v11i3s.6180.

E. Zangerle and C. Bauer, “Evaluating Recommender Systems: Survey and Framework,” ACM Computing Surveys, vol. 55, no. 8, Art. 170, pp. 1–38, 2022, doi: 10.1145/3556536.

P. Castells and A. Moffat, “Offline Recommender System Evaluation: Challenges and New Directions,” AI Magazine, vol. 43, no. 2, pp. 225–238, 2022, doi: 10.1002/aaai.12051.

X. Wu, “Review of Collaborative Filtering Recommendation Systems,” Applied and Computational Engineering, vol. 43, pp. 76–82, 2024, doi: 10.54254/2755-2721/43/20230811.

Y. Ji, A. Sun, J. Zhang, and C. Li, “A Re-visit of the Popularity Baseline in Recommender Systems,” in Proc. 43rd Int. ACM SIGIR Conf. Res. Develop. Inf. Retrieval (SIGIR ’20), 2020, pp. 1749–1752, doi: 10.1145/3397271.3401233.