Yum-Me: A personalized nutrient-based meal recommender system
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
Yum-Me : A personalized nutrient-based meal recommender system. / Yang, Longqi; Hsieh, Cheng Kang; Yang, Hongjian; Pollak, John P.; Dell, Nicola; Belongie, Serge; Cole, Curtis; Estrin, Deborah.
In: ACM Transactions on Information Systems, Vol. 36, No. 1, 7, 04.2017.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Yum-Me
T2 - A personalized nutrient-based meal recommender system
AU - Yang, Longqi
AU - Hsieh, Cheng Kang
AU - Yang, Hongjian
AU - Pollak, John P.
AU - Dell, Nicola
AU - Belongie, Serge
AU - Cole, Curtis
AU - Estrin, Deborah
N1 - Funding Information: This work is funded through Awards from NSF (#1344587, #1343058) and NIH (#1U54EB020404); as well as gift funding from AOL, RWJF, UnitedHealth Group, Google, and Adobe. Publisher Copyright: Copyright 2017 is held by the owner/author(s). Publication rights licensed to ACM.
PY - 2017/4
Y1 - 2017/4
N2 - Nutrient-based meal recommendations have the potential to help individuals prevent or manage conditions such as diabetes and obesity. However, learning people's food preferences and making recommendations that simultaneously appeal to their palate and satisfy nutritional expectations are challenging. Existing approaches either only learn high-level preferences or require a prolonged learning period. We propose Yum-me, a personalized nutrient-based meal recommender system designed to meet individuals' nutritional expectations, dietary restrictions, and fine-grained food preferences. Yum-me enables a simple and accurate food preference profiling procedure via a visual quiz-based user interface and projects the learned profile into the domain of nutritionally appropriate food options to find ones that will appeal to the user.We present the design and implementation of Yum-me and further describe and evaluate two innovative contributions. The first contriution is an open source state-of-the-art food image analysis model, named FoodDist. We demonstrate FoodDist's superior performance through careful benchmarking and discuss its applicability across a wide array of dietary applications. The second contribution is a novel online learning framework that learns food preference from itemwise and pairwise image comparisons. We evaluate the framework in a field study of 227 anonymous users and demonstrate that it outperforms other baselines by a significant margin. We further conducted an end-to-end validation of the feasibility and effectiveness of Yum-me through a 60-person user study, in which Yum-me improves the recommendation acceptance rate by 42.63%.
AB - Nutrient-based meal recommendations have the potential to help individuals prevent or manage conditions such as diabetes and obesity. However, learning people's food preferences and making recommendations that simultaneously appeal to their palate and satisfy nutritional expectations are challenging. Existing approaches either only learn high-level preferences or require a prolonged learning period. We propose Yum-me, a personalized nutrient-based meal recommender system designed to meet individuals' nutritional expectations, dietary restrictions, and fine-grained food preferences. Yum-me enables a simple and accurate food preference profiling procedure via a visual quiz-based user interface and projects the learned profile into the domain of nutritionally appropriate food options to find ones that will appeal to the user.We present the design and implementation of Yum-me and further describe and evaluate two innovative contributions. The first contriution is an open source state-of-the-art food image analysis model, named FoodDist. We demonstrate FoodDist's superior performance through careful benchmarking and discuss its applicability across a wide array of dietary applications. The second contribution is a novel online learning framework that learns food preference from itemwise and pairwise image comparisons. We evaluate the framework in a field study of 227 anonymous users and demonstrate that it outperforms other baselines by a significant margin. We further conducted an end-to-end validation of the feasibility and effectiveness of Yum-me through a 60-person user study, in which Yum-me improves the recommendation acceptance rate by 42.63%.
KW - Food preferences
KW - Nutrient-based meal recommendation
KW - Online learning
KW - Personalization
KW - Visual interface
UR - http://www.scopus.com/inward/record.url?scp=85026445680&partnerID=8YFLogxK
U2 - 10.1145/3072614
DO - 10.1145/3072614
M3 - Journal article
AN - SCOPUS:85026445680
VL - 36
JO - ACM Transactions on Information Systems
JF - ACM Transactions on Information Systems
SN - 1046-8188
IS - 1
M1 - 7
ER -
ID: 301827380