Analisis Komputasi Trust-based Pada Metode Collaborative Filtering Untuk Sistem Rekomendasi Hotel

Authors

  • Azka Apta Afrizal Telkomuniversity
  • Ibnu Asror Telkomuniversity
  • Yusza Reditya Murti Telkomuniversity

DOI:

https://doi.org/10.34818/eoe.v7i2.12788

Abstract

Abstrak Hotel adalah salah satu dari banyaknya jenis tempat tinggal yang ada hampir dalam setiap kota. Beberapa orang biasanya akan mencari hotel tergantung dari kebutuhan dan juga melihat ulasan dari orang-orang sebelumnya. Hotel dan juga ulasan yang terlalu banyak,terkadang akan membingungkan pencari hotel untuk menemukan hotel yang pas bagi mereka. Dari sinilah dibangun sebuah sistem yang akan mempermudah pencari hotel untuk menemukan hotel yang pas. Sistem ini adalah sebuah sistem rekomendasi. Dimana sistem akan merekomendasikan sebuah item yaitu hotel. Hotel yang akan direkomendasikan berdasarkan beberapa kriteria tertentu dan salah satunya yang terpenting yaitu rating dan ulasan user. Sistem rekomendasi yang dibangun di dalam tugas akhir ini menggunakan metode Collaborative Filtering(CF) dan dibantu dengan perhitungan komputasi Trust-Based. CF bekerja dengan cara melakukan perhitungan terhadap rating dari pengguna ataupun ulasan sebelumnya dan trust akan menghitung keaslian nilai yang diberikan suatu user terhadap sebuah item. Sistem akan merekomendasikan hotel berdasarkan karakteristik user dan nilai dari kesamaan(similarity) antar user lain yang sudah pernah melakukan review hotel dengan yang akan direkomendasi. Dari percobaan yang telah dilakukan,didapatkan nilai error sebesar 1.6924 dengan menggunakan metode CF, 0.8746 menggunakan metode CF ditambah Trust-Based dengan nilai threshold trust >3 dan 0.8716 dengan nilai threshold trust >5. Semakin kecil nilai error,maka semakin bagus rekomendasi bekerja. Dilihat dari pengujian yang sudah dijalankan,penggunaan trust dapat mengurangi besarnya error pada penggunaan metode collaborative filtering. Kata kunci : sistem rekomendasi,review,collaborative filtering,Trust-Based Abstract Hotels are the one of the many types of residences that exist in almost every city. Some people usually will search for hotels depending on their needs and also see reviews from the previous people. Hotels and review that too many will make hotel seekers confuse to find the right hotel for them. From here appointed a system that will facilitate hotel seekers to find the right hotel. This system is called recommendation system. Where the system will recommend an item which in this context is a hotel. Hotels to be recommended are based on certain criteria and one of the most important is user reviews. Recommendation system that was built using the Collaborative Filtering (CF) method. CF works by calculating the rating from users or previous reviews and trust will calculate the authenticity of the user value given by a user to an item . The system will recommend hotels based on the value of the similarity of other users who have rated one of the hotels and who are looking for hotel. From the experiment that has been done,there is an error value 1.6924 obtained using the CF method, 0.8746 with CF methods plus Trust-Based with trust threshold value >3 and 0.8716 with trust threshold value >5. The smaller error value,recommendation will have a better works. From the tests that has been done,the use of trust can reduce error in the use of collaborative filtering methods. Keywords: recommendation system,review,collaborative filtering,Trust-Based

Downloads

Download data is not yet available.

References

[1] J. Zhou and T. Luo, “Towards an introduction to collaborative filtering,” Proc. - 12th IEEE Int. Conf. Comput. Sci. Eng. CSE 2009, vol. 4, pp. 576–581, 2009.

[2] Q. Wu, A. Forsman, Z. Yu, and W. W. Song, “A Computational Model for Trust-Based Collaborative Filtering An Empirical Study on Hotel Recommendations,” pp. 266–279, 2014.

[3] A. N. Maus and A. K. Atwood, “Surveying Older Adults About a Recommender System for a Digital Library,” pp. 41–44, 2015.

[4] S. Parvatikar and B. Joshi, “Online book recommendation system by using collaborative filtering and association mining,” 2015 IEEE Int. Conf. Comput. Intell. Comput. Res. ICCIC 2015, pp. 6–9, 2016.

[5] N. Mittal, R. Nayak, M. C. Govil, and K. C. Jain, “Recommender system framework using clustering and collaborative filtering,” Proc. - 3rd Int. Conf. Emerg. Trends Eng. Technol. ICETET 2010, pp. 555– 558, 2010.

[6] D. Zeng and H. Chen, “for E-commerce.”

[7] H. Li, X. Diao, J. Cao, and Q. Zheng, “Collaborative Filtering Recommendation Based on All-Weighted Matrix Factorization and Fast Optimization,” IEEE Access, vol. 6, no. c, pp. 25248–25260, 2018.

[8] J. Li et al., “Category Preferred Canopy–K-means based Collaborative Filtering algorithm,” Futur. Gener. Comput. Syst., vol. 93, pp. 1046–1054, 2019.

[9] S. Publications, “Review Author ( s ): Peter Smith Ring Review by : Peter Smith Ring Source : Administrative Science Quarterly , Vol . 48 , No . 4 ( Dec ., 2003 ), pp . 720-722 Published by : Sage Publications , Inc . on behalf of the Johnson Graduate School of Management,” vol. 48, no. 4, pp. 720– 722, 2016.

[10] M. G. Ozsoy and F. Polat, “Trust based recommendation systems,” Proc. 2013 IEEE/ACM Int. Conf. Adv. Soc. Networks Anal. Mining, ASONAM 2013, pp. 1267–1274, 2013.

[11] N. Lathia, S. Hailes, and L. Capra, “Trust-based collaborative filtering Trust-Based Collaborative Filtering,” no. July, pp. 2650–2654, 2014.

[12] S. Cleger-Tamayo, J. M. Fernández-Luna, and J. F. Huete, “On the use of weighted mean absolute error in recommender systems,” CEUR Workshop Proc., vol. 910, pp. 24–26, 2012

Downloads

Published

2020-08-01

Issue

Pages

7729-7740

Section

Program Studi S1 Informatika

How to Cite

Afrizal, A. A., Asror, I., & Murti, Y. R. (2020). Analisis Komputasi Trust-based Pada Metode Collaborative Filtering Untuk Sistem Rekomendasi Hotel. EProceedings of Engineering, 7(2), 7729-7740. https://doi.org/10.34818/eoe.v7i2.12788