PEMODELAN SISTEM REKOMENDASI CERDAS MENGGUNAKAN HYBRID DEEP LEARNING

Kadek Cahya Dewi, Putu Indah Ciptayani

Abstract


Trend perkembangan teknologi saat ini adalah mengarah ke sistem cerdas. Namun saat ini belum ada yang menggabungkan dua metode deep learning pada algoritma rekomendasi, sehingga penting untuk melakukan pemodelan sistem rekomendasi cerdas menggunakan hybrid deep learning. Penelitian ini bertujuan untuk mendapatkan model hybrid deep learning yang optimal pada sistem rekomendasi cerdas. Penelitian ini menggunakan pendekatan penelitian eksperimen. Teknik pengumpulan data yang digunakan meliputi observasi, penelusuran online dan pencatatan dataset. Tahapan penelitian terdiri dari: (a) literature review, (b) observasi dan pencarian online, (c) modeling (d) prototyping dan (e) testing. Model yang dihasilkan merupakan model hybrid deep learning yang terdiri dari dua layer yaitu layer Self Organizing Map (SOM) dan layer Recurrent Neural Network (RNN). Penelitian ini menggunakan bahasa pemrograman Python pada tahap pembuatan prototipe. Beberapa modul library dalam python yang digunakan antara lain numpy, pandas, tensorflow, hard, torch, sklearn. Program diuji dengan dataset dari kaggle.com. Hasil pengujian berhasil meningkatkan kinerja dengan meningkatkan akurasi hingga 100%. Dapat disimpulkan bahwa model SOM-RNN dapat meningkatkan kinerja sistem rekomendasi cerdas.

Keywords


Deep Learning; Hybrid Deep Learning; Intelligent Recommendation Systems; Recommendation Systems

Full Text:

PDF

References


ADIYANSJAH, GUNAWAN, AAS., dan SUHARTONO, D., 2019. Music Recommender System Based on Genre using Convolutional Recurrent Neural Networks. Procedia Computer Science. vol. 157. pp. 99 – 109.

BENABDERRAHMANE, S., MELLOULI, N., dan LAMOLLE, M., 2018. On the predictive analysis of behavioral massive job data using embedded clustering and deep recurrent neural networks. Knowledge-Based Systems. vol. 151. pp. 95–113.

BRUSAFERRIA, A., MATTEUCCIB, M, SPINELLIA, S. dan VITALI, A., 2020, Learning behavioral models by recurrent neural networks with discrete latent representations with application to a flexible industrial conveyor. Computers in Industry. vol. 122. pp. 103263.

DEWI, K. C. dan HARJOKO, A., 2010. Kid's song classification based on mood parameters using K-Nearest Neighbor classification method and Self Organizing Map. 2010 International Conference on Distributed Frameworks for Multimedia Applications. Yogyakarta. pp. 1-5.

DJELLALI, C. dan ADDA, M. 2020. A New Hybrid Deep Learning Model based-Recommender System using Artificial Neural Network and Hidden Markov Model. Procedia Computer Science. vol. 175. pp. 214–220.

JHA, S., PRASHAR, D., LONG, H. V., dan TANIAR, D., 2020. Recurrent neural network for detecting malware. Computers & Security. Vol.99, pp.102037.

KOEHN, D, LESSMANN, S., dan SCHAAL, M., 2020. Predicting online shopping behaviour from clickstream data using deep learning. Expert Systems With Applications. vol. 150. pp. 113342.

NILASHI, M., IBRAHIM, O., dan ITHNIN, N., 2014. Hybrid recommendation approaches for multi-criteria collaborative filtering. Expert Systems with Applications. vol. 41. pp.3879–3900.

NILASHI, M., BAGHERIFARD, K., RAHMANI, M., dan RAFE, V., 2017. A recommender system for tourism industry using cluster ensemble and prediction machine learning techniques. Computers & Industrial Engineering. vol. 109. pp. 357–368.

TURKUT, Ü., TUNCER, A., SAVRAN, H., dan YILMAZ, S., 2020. An Online Recommendation System Using Deep Learning for Textile Products. 2020 International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA). Ankara. Turkey. pp. 1-4.

WANG, K., ZHANG, T., XUE, T., LU, Y., dan NA, SG., 2019. E-Commerce Personalized Recommendation Analysis by Deeply-learned Clustering, J. Vis. Commun. Image R.

WU, B. dan YE, Y., 2020, BSPR: Basket-Sensitive Personalized Ranking for Product Recommendation, Information Sciences.

XU, Y., YANG, Y., HAN, J., WANG, E., MING, J., dan XIONG, H., 2019. Slanderous user detection with modified recurrent neural networks in recommender system. Information Sciences. vol. 505. pp. 265–281.

ZHANG, M. dan BOCKSTEDT, J., 2020. Complements and Substitutes in Online Product Recommendations: The Differential Effects on Consumers’ Willingness to Pay. Information and Management.




DOI: https://doi.org/10.31326/sistek.v4i2.1157

Refbacks

  • There are currently no refbacks.


JOURNAL IDENTITY

Journal Name:  Journal Information System and Science Technology

Jurnal Sistem Informasi dan Sains Teknologi


e-ISSN: 2684-8260
Publisher: Program Studi Sistem Informasi, Universitas Trilogi, Jakarta Selatan, Indonesia
Publication Schedule: February and August
Language: Indonesian and English
APC: Free of charge (submission, publishing) 
Indexing:  Google Scholar, Garuda, Neliti, One Search, Base, DRJI, Road, Crossref, Index CopernicusWorldCat, ScilitDimensions (find by DOI article)
OAI addresshttp://trilogi.ac.id/journal/ks/index.php/SISTEK/oai?verb=ListRecords&metadataPrefix=oai_dc
Collaboration Partners: Indonesian Association of Higher Education in Informatics and Computing (APTIKOM)

Contactjurnalsistek@gmail.com (Whatsapp Number: +628192454119)

license :
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Journal Information System and Science Technology (Jurnal Sistem Informasi dan Sains Teknologi) is Published by Information System Department Trilogi University, South Jakarta, Indonesia. 

Under license CC-BY from Creative Commons Attribution.