The task of point of interest (POI) recommendation is to recommend POIs (e.g., restaurants, coffee shops, museums) that the end-user may be interested in, but has never visited in a given time window. While POI recommendation in general inherits the large body of work in the community of recommender systems, it also carries new constraints and challenges that might not be the case for a traditional recommender.
An ideal POI recommender should return personalized and contextual results. Personalization has been the focus of many approaches in the past where historical check-ins are exploited to predict which POI the user prefers to visit next, using techniques such as Matrix Factorization and Poisson Factor Model. Contextuality is to capture the current situation of the user (aka context) to bias the recommendation results. While the literature focuses on time and location to form context, user’s intent has received less attention.
In this internship, the objective is to predict user’s intent and employ it as a strong signal to improve POI recommendations. A deep ML model will be developed to predict user’s intent for any given situation. The predicted intent will then be used to increase the prediction probability of POIs which are relevant to the prediction. For instance, in case the intent is predicted as “being hungry”, restaurants and coffee shops are more probable to be recommended.
The ML model should then be evaluated on public POI datasets such as Foursquare and Gowalla, as well as NAVER Maps dataset. The evaluation may also consist of a crowdsourcing campaign to measure the usefulness of the intent predictor.
Supervisor: Behrooz Omidvar-Tehrani
 Behrooz Omidvar-Tehrani, Sruthi Viswanathan, Frederic Roulland, and Jean-Michel Renders. “SAGE: Interactive State-aware Point-of-Interest Recommendation”. WSDM SUM 2020.
 Jie Bao, Yu Zheng, David Wilkie, and Mohamed Mokbel. “Recommendations in location-based social networks: a survey”. GeoInformatica 2015.
 Hongzhi Yin, Bin Cui, Yizhou Sun, Zhiting Hu, and Ling Chen. “Lcars: A spatial item, recommender system.” TOIS 2014.
 Bin Liu, Yanjie Fu, Zijun Yao, and Hui Xiong. “Learning geographical preferences for point-ofinterest recommendation”. SIGKDD 2013.
 Justin J Levandoski, Mohamed Sarwat, Ahmed Eldawy, and Mohamed F Mokbel. “Lars: A location-aware recommender system”. ICDE 2012.
NAVER LABS is a world class team of self-motivated and highly engaged researchers, engineers and interface designers collaborating together to create next generation ambient intelligence technology and services that are rich with the organic understanding they have of users, their contexts and situations.
Since 2013 LABS has led NAVER’s innovation in technology through products such as the AI-based translation app ‘Papago’, the omni-tasking web browser ‘Whale’, the virtual AI assistant ‘WAVE’, in-vehicle information entertainment system ‘AWAY’ and M1, the 3D indoor mapping robot.
The team in Europe is multidisciplinary and extremely multicultural specializing in artificial intelligence, machine learning, computer vision, natural language processing, UX and ethnography. We collaborate with many partners in the European scientific community on R&D projects.
NAVER LABS Europe is located in the south east of France in Grenoble. The notoriety of Grenoble comes from its exceptional natural environment and scientific ecosystem with 21,000 jobs in public and private research. It is home to 1 of the 4 French national institutes in AI called MIAI (Multidisciplinary Innovation in Ai) It has a large student community (over 62,000 students) and is a lively and cosmopolitan place, offering a host of leisure opportunities. Grenoble is close to both the Swiss and Italian borders and is the ideal place for skiing, hiking, climbing, hang gliding and all types of mountain sports.