David López Caballero

Advisors

Rebecca Lennartz (M. Sc.),  Yannick Wiesner (BesserEsser), Pauline Nöldemann (BesserEsser), Prof. Dr. Björn Eskofier

Duration

09 / 2024 – 09/ 2025

Abstract

In Germany around 20 million people suffer from food intolerances [1] [2]. As most often food intolerances cannot be determined by an accurate medical test, the start-up BesserEsser implemented an application to support the identification of food intolerances [3]. One challenge is the selection of suitable recipes respecting the food intolerance of a user. The project team of Happy Plate has implemented an algorithm to select a recipe according to the configured identified food intolerances, for example lactose intolerance or gluten free recipes [4]. The result is an unsorted list of recipes categorized in four categories (e. g. main dish or dessert). One additional option evaluated but not implemented is the use of recommender systems. Recommender systems can propose recipes and possibly increase the acceptance rate of the listed recipes. Recommender systems are numerous and are applied in different fields [5]. In the domain of recipe recommendations, recommender systems consider demographic aspects [6], nutritional expectations [7] or a broad set of criteria, such as dietary preferences and ingredient restrictions [8]. The evaluation of such recommender systems spans from image-based studies [7] and limited questionnaires with a mixed user group [9], to periodical user studies comparing the suitability of varying recommender algorithms for recipe recommendations [10].

The aim of the master research is to implement a suitable collaborative based recommender system in the given context of dietary plans for people with food intolerances. By doing so, the benefit of recommender systems for the specific user group is investigated in detail. The research includes the evaluation of a collaborative based recommender system in comparison with a basic knowledge-based recommender system (the filtered list of recipes based on pre-defined parameters) with respect to the adherence to dietary plans and well-being of treated people.

The research question derived is the following:

• Can a collaborative-based recommender system be implemented for people with food intolerances?
• Does the collaborative-based recommender system increase the well-being of people with food intolerances by adapting to recommended recipes compared to an unsorted list of recipes presented to users?

References

[1] YouGov. (2021, August 16). statista. Retrieved 06 02, 2024, from https://de.statista.com/statistik/daten/studie/1267347/umfrage/umfrage-zu lebensmittelunvertraeglichkeiten-in-deutschland/

[2] Weber, N. (2014). DER SPIEGEL. Retrieved 06 08, 2024, from DER SPIEGEL: https://www.spiegel.de/gesundheit/ernaehrung/gluten-laktose-histamin-23-prozent klagen-ueber-unvertraeglichkeiten-a-975015.html

[3] PYRA MEDI. (2024). https://besseresser-app.de/

[4] Machine Learning and Data Analytics (MaD) Lab (2024). FAU. Retrieved 04.09.2024, from https://www.mad.tf.fau.de/research/projects/innovation-lab-for-wearable-and-ubiquitous-computing/happy-plate/

[5]  Charu C. Aggarwal. (28 March 2016). Recommender Systems. Springer Cham. https://doi.org/10.1007/978-3-319-29659-3

[6] Sobecki, J., Babiak, E., Słanina, M. (2006). Application of Hybrid Recommendation in Web-Based Cooking Assistant. In: Gabrys, B., Howlett, R.J., Jain, L.C. (eds) Knowledge-Based Intelligent Information and Engineering Systems. KES 2006. Lecture Notes in Computer Science(), vol 4253. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11893011_101

[7] Longqi Yang, Cheng-Kang Hsieh, Hongjian Yang, John P. Pollak, Nicola Dell, Serge Belongie, Curtis Cole, and Deborah Estrin. 2017. Yum-Me: A Personalized Nutrient-Based Meal Recommender System. ACM Trans. Inf. Syst. 36, 1, Article 7 (January 2018), 31 pages. https://doi.org/10.1145/3072614

[8] Xing, T., Gao, J. (2024). RecipeRadar: An AI-Powered Recipe Recommendation System. In: Arai, K. (eds) Intelligent Systems and Applications. IntelliSys 2024. Lecture Notes in Networks and Systems, vol 1067. Springer, Cham. https://doi.org/10.1007/978-3-031-66431-1_7

[9] Zioutos, K., Kondylakis, H., & Stefanidis, K. (2023). SHARE: A Framework for Personalized and Healthy Recipe Recommendations. EDBT/ICDT Workshops. https://researchportal.tuni.fi/files/95244801/HeDAI_2023_paper405.pdf

[10] Jill Freyne and Shlomo Berkovsky (2010). Intelligent food planning: personalized recipe recommendation. In Proceedings of the 15th international conference on Intelligent user interfaces (IUI ’10). Association for Computing Machinery, New York, NY, USA, 321–324. https://doi.org/10.1145/1719970.1720021