BUILDING FOOD RECOMMENDATION SYSTEMS USING ALTERNATING LEAST SQUARES AND COSINE SIMILARITY TECHNIQUES

Year 2024, Volume: 9 Issue: Issue:1, 1 - 17, 06.06.2024

Merve Cengiz , Tuğba Yıldız

https://doi.org/10.53070/bbd.1389078

https://izlik.org/JA67KC26LA

Abstract

The aim of this study was to develop a food recommendation system based on the recipes on the Allrecipes.com website and the ratings given by the members. Total 1840 recipes (Diabetic - Gluten Free - Ketogenic - Low Sodium - Low Cholesterol - Vegetarian - Vegan) were scraped from Allrecipes.com and analysed in Python. Recommendation System was performed by using Alternating Least Square (ALS). Diet Food Recommendation System was performed by using cosine similarity technique. The application of ALS technique with big data was performed on the cloud. The root mean squared error of the model was found 0.495. The foods recommended by the model were examined on a user basis and it was determined that the results were consistent. When the most recommended foods were examined, vegetarian recipes were ranked first, and in total, a high number of ketogenic recipes were recommended. Consequently, we created a web-based food recommendation system that generates accurate recommendations to users who want to have ideas about foods via recipes and choose food according to their diet.

Keywords

Recommendation system , alternating least squares , cosine similarity , big data

DEĞİŞİMLİ EN KÜÇÜK KARELER VE KOSİNÜS BENZERLİK TEKNİKLERİ KULLANILARAK YEMEK TAVSİYE SİSTEMİ OLUŞTURMA

https://izlik.org/JA67KC26LA

Abstract

Bu çalışmada, Allrecipes.com web sitesindeki yemek tariflerine ve üyeler tarafından verilen oylara dayalı bir yemek tavsiye sistemi geliştirildi. Toplam 1840 yemek tarifi (Diyabetik - Glutensiz - Ketojenik - Düşük Sodyum - Düşük Kolesterol - Vejetaryen – Vegan) Allrecipes.com'dan web scraping yöntemi ile kazındı ve Python'da analiz edildi. Tavsiye Sistemi, Değişimli En Küçük Kareler (DEKK) yöntemi kullanılarak oluşturuldu. Diyet Yemek Tavsiye Sistemi, kosinüs benzerlik yöntemi kullanılarak gerçekleştirildi. DEKK yönteminin büyük veri ile uygulaması bulut üzerinde gerçekleştirildi. Modelin hata kareler ortalamasının karekökü 0.495 olarak bulundu. Modelin önerdiği yemekler kullanıcı bazlı incelendi ve sonuçların tutarlı olduğu belirlendi. En çok tavsiye edilen yemekler incelendiğinde, vejetaryen tariflerin ilk sırada yer aldığı; toplamda ise ketojenik tariflerin yüksek sayıda önerildiği görüldü. Sonuç olarak, yemek tarifleri aracılığıyla yiyecekler hakkında fikir sahibi olmak ve diyetlerine göre yiyecek seçmek isteyen kullanıcılara doğru öneriler üreten web tabanlı bir yemek öneri sistemi oluşturuldu.

Keywords

Tavsiye sistemi , değişimli en küçük kareler , kosinüs benzerlik , büyük veri

References

• Awan M J, Khan R A, Nobanee H, Yasin A, Anwar S , Naseem U, Singh V P (2021). A recommendation engine for predicting movie ratings using a big data approach. Electronics, 10(10): 1215. https://doi.org/10.3390/electronics10101215
• Barakat M O S (2020) Pubmed Article Recommendation System Based On Collaborative Filtering, Master's thesis, Dokuz Eylul University, Izmir.
• Bozkurt M, Acı Ç İ (2021) Öneri algoritmalarının film önerme problemi üzerinde karşılaştırılması: Movielens örneği. Bilgisayar Bilimleri ve Teknolojileri Dergisi, 2(2): 36-42.
• Chen J, Fang J, Liu W, Tang T, Chen X, Yang C. (2017) Efficient And Portable Als Matrix Factorization For Recommender Systems. 2017 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), pp.409-418.
• Gündoğan, E, Kaya, M (2021) Bilimsel dergi tavsiyesi için içerik tabanlı bir yaklaşım. Computer Science, (Special): 41-47.
• Han, J., Kamber, M., & Pei, J. (2012, January). Getting to know your data. In Data mining (Vol. 3, pp. 39-82). Boston, MA: Morgan Kaufmann. A chapter in the book, Data Mining (Third Edition) The Morgan Kaufmann Series in Data Management Systems https://doi.org/10.1016/B978-0-12-381479-1.00002-2.
• Jiang J, Li W, Dong A, Gou Q, Luo X (2020) A fast deep autoencoder for high-dimensional and sparse matrices in recommender systems. Neurocomputing, 412: 381-391.
• Kaya TS (2019) Veri Madenciliği Algoritmaları İle Kredi Kartı Kullanım Alışkanlıklarının Incelenmesi Ve Kişiye Özgü Kampanya Teklifi, Master's Thesis. İstanbul Üniversitesi.
• Koren Y, Bell R, Volinsky C (2009) Matrix factorization techniques for recommender systems. Computer, 42(8): 30-37.
• Lawson R. (2015) Web Scraping With Python, Packt Publishing Ltd.
• Li JB, Lin SY, Hsu YH, Huang YC. (2018) Implementation Of An Alternating Least Square Model Based Collaborative Filtering Movie Recommendation System On Hadoop And Spark Platforms. International Conference on Broadband and Wireless Computing, Communication and Applications, pp.237-249.
• Li S, McAuley J (2020) Recipes for Success: Data Science in the Home Kitchen. Harvard Data Science Review, 2. https://assets.pubpub.org/nzhfriaw/ca2af84f-38f9-48c4-8b38-5cfe915a7b7e.pdf. Accessed 02 Nov 2022
• Linden G, Smith B, York J (2003) Amazon.com recommendations: item-to-item collaborative filtering. IEEE Internet computing, 7(1): 76-80.
• Mitchell R. (2018) Web Scraping With Python: Collecting More Data From The Modern Web, O'Reilly Media, Inc.
• Nguyen W. (2021) A Literature Review of Collaborative Filtering Recommendation System using Matrix Factorization algorithms. In Proceedings of ACM Conference (Conference’17). ACM.
• Oh Y, Choi A, Woo W (2010) u-BabSang: a context-aware food recommendation system. The Journal of Supercomputing, 54: 61-81.
• Özcan İ, Çelik M. (2018) Developing Recommendation System Using Genetic Algorithm Based Alternative Least Squares. 2018 International Conference on Artificial Intelligence and Data Processing, pp.1-5.
• Philip S, Shola P, Ovye A (2014) Application of content-based approach in research paper recommendation system for a digital library. International Journal of Advanced Computer Science and Applications, 5(10): 37-40.
• Sarwar B, Karypis G, Konstan J, Riedl J (2001) Item-Based Collaborative Filtering Recommendation Algorithms. In Proceedings of the 10th international conference on World Wide Web: 285-295.
• Schafer J B, Frankowski D, Herlocker J, Sen S. (2007) Collaborative Filtering Recommender Systems. In The Adaptive Web, Springer, Berlin, Heidelberg.
• Fathollahi M S, Razzazi F (2021) Music similarity measurement and recommendation system using convolutional neural networks. International Journal of Multimedia Information Retrieval, 10(1): 43-53.
• Thorat P B, Goudar R M and Barve S (2015) Survey on collaborative filtering, content-based filtering and hybrid recommendation system. International Journal of Computer Applications, 110(4): 31-36.
• Uluyağmur M (2012) Hibrit Film Öneri Sistemi, Master's thesis, Istanbul Teknik Üniversitesi.
• Ünaldı, S S (2022) Veri Seyrekliği ve Ölçeklenebilirlik Problemlerini Gidermek İçin Derin Otomatik Kodlayıcı Tabanlı Yeni Bir Tavsiye Sistemi Modeli, Master's thesis, Necmettin Erbakan Üniversitesi.
• Vairale VS, Shukla S. (2021) Recommendation Of Food Items For Thyroid Patients Using Content-Based Knn Method. Data Science and Security: Proceedings of IDSCS 2020, pp.71-77.
• Xie L, Zhou W, Li Y (2016) Application of improved recommendation system based on spark platform in big data analysis. Cybernetics and Information Technologies, 16(6): 245-255.