Advertisement

Machine Learning Based Food Recipe Recommendation System

Conference paper
Part of the Lecture Notes in Networks and Systems book series (LNNS, volume 14)

Abstract

Recommender systems make use of user profiles and filtering technologies to help users to find appropriate information over large volume of data. Users profile is important for successful recommendations. In this paper, we present two approaches to recommend recipes based on preferences of the user given in the form of ratings and compare them to identify which approach suits the dataset better. We use two approaches namely, item based approach and user based approach to recommend recipes. For item based approach Tanimoto Coefficient Similarity and Log Likelihood Similarity would be used to compute similarities between different recipes. For user based approach Euclidean Distance and Pearson Correlation are used. We use similarity techniques of user based approach and introduce fixed size neighborhood and threshold-based neighborhood to the same. The performance of the user based approach is found to be better than item based approach. The performance for the Allrecipe data set is found to be better than the simulated dataset since there are more number of interactions between users and items.

Keywords

Collaborative filtering Item based User based Fixed size neighborhood Threshold-based neighborhood 

References

  1. 1.
    Adomavicius G, Tuzhilin A (2005) Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions. IEEE Trans Knowl Data Eng 17(6), 734–749Google Scholar
  2. 2.
    Introduction to recommender systems handbook, Recommender systems handbook, Online ISBN 978-0- 387-85820-3, Springer US (2011) Google Scholar
  3. 3.
    Facebook (2010) Pandora lead rise of recommendation engines—TIMEGoogle Scholar
  4. 4.
    Felfernig A, Isak K, Szabo K, Zachar P (2007) The VITA financial services sales support environment. In: AAAI/IAAI 2007, pp 1692–1699. Vancouver, CanadaGoogle Scholar
  5. 5.
    Gupta P, Goel A, Lin J, Sharma A, Wang D, Zadeh RB (2013) WTF: the who-to-follow system at Twitter. In: Proceedings of the 22nd International Conference on World Wide Web WWW 2013, Rio de Janeiro, Brazil. ACM 978-1-4503-2035Google Scholar
  6. 6.
    Almazro D, Shahatah G, Albdulkarin L (2010) A survey paper on recommender systems. arXiv:1006.5278v4[cs.IR]
  7. 7.
    Jafarkarimi H, Sim ATH, Saadatdoost R (2012) A Naïve recommendation model for large databases. Int J Inf Educ Technol (2012)Google Scholar
  8. 8.
    Mooney RJ, Roy L (1999) Content-based book recommendation using learning for text categorization. In: Workshop on recommender systems: algorithms and evaluationGoogle Scholar
  9. 9.
    Burke R (2002) Hybrid recommender systems: survey and experiments. User Model User-Adap Interact 12(4):331–370CrossRefMATHGoogle Scholar
  10. 10.
    Herlocker JL, Konstan JA, Terveen LG, Riedl JT (2004) Evaluating collaborative filtering recommender systems. ACM Trans Inf Syst 22(1):5–53. doi: 10.1145/963770.963772 CrossRefGoogle Scholar
  11. 11.
    Beel J, Langer S, Genzmehr M, Gipp B (2013) A comparative analysis of offline and online evaluations and discussion of research paper recommender system evaluation. In: Proceedings of the workshop on reproducibility and replication in recommender systems evaluation (RepSys) at the ACM recommender system conference (RecSys)Google Scholar
  12. 12.
    Srivastava J, Cooley R, Deshpande M, Tan P (2000) Web usage mining: discovery and applications of usage patterns from web data. SIGKDD Explor 1(2):12–23CrossRefGoogle Scholar
  13. 13.
    Talavera L, Gaudioso E (2004) Mining student data to characterize similar behavior groups in unstructured collaboration spaces. In: Workshop on AI in CSCL, pp 17–23 (2004)Google Scholar
  14. 14.
    Zaïane OR (2006) Recommender system for e-learning: towards non-instructive web mining. In: Data mining in e-learning, pp 79–96Google Scholar
  15. 15.
    Bellogin A, Cantador I, Castells P, Ortigosa A (2008) Discovering relevant preferences in a personalized recommender system using machine learning techniques. Spanish Ministry of Science and Education (TIN2005–6885 and TIN2007-64718)Google Scholar
  16. 16.
    Rajabi S, Harounabadi A, Aghazarian V (2014) A recommender system for the web: using user profiles and machine learning methods. Int J Comput Appl (0975–8887) 96(11) (2014)Google Scholar
  17. 17.
    Seifoddini H, Djassemi M (2007) Merits of the production volume based similarity coefficient in machine cell formationGoogle Scholar
  18. 18.
    Abreu R, Zoeteweij P, Van Gemund A (2010) An evaluation of similarity coefficients for software fault localizationGoogle Scholar
  19. 19.
    Evaluating and implementing recommender systems as web services using Apache Mahout. http://cslab1.bc.edu/~csacademics/pdf/14Casinelli.pdf
  20. 20.
    Madylova A, Oguducu SG (2009) A taxonomy based semantic similarity of documents using the cosine measure. In: Proceedings of international symposium on computer and information sciences, pp 129–134Google Scholar
  21. 21.
    Pearson correlation: definition and easy steps for use. http://www.statisticshowto.com/what-is-the-pearson-correlation-coefficient/
  22. 22.
  23. 23.
    Surprise and coincidence—musings from the longtail. http://tdunning.blogspot.in/2008/03/surprise-and-coincidence.html
  24. 24.
  25. 25.

Copyright information

© Springer Nature Singapore Pte Ltd. 2018

Authors and Affiliations

  1. 1.Information Science and EngineeringSri Jayachamarajendra College of EngineeringMysoreIndia
  2. 2.Computer Science and EngineeringSri Jayachamarajendra College of EngineeringMysoreIndia

Personalised recommendations