Point-of-Interest (PoI) Recommender System
Project Card
- Category: Recommender System
- Date: January 2021 to February 2022
- Affiliation: Independent Research
- Stack: Python
- Related Links: GitHub (CAPRI), Paper (CAPRI), GitHub (FairPOI), Paper (FairPOI), GitHub (ContextsFair), Paper (ContextsFair)
A set of Point-of-Interest (PoI) recommender system tools and framework for suggesting the most relevant PoIs to users based on their preferences and the context of the recommendation. The investigations include the fairness and bias in the recommendations, as well as the context-awareness of the suggestions. Finally, a framework for generating synthetic PoI recommendations is developed. The code is publicly available and the results were published in the papers referenced below:
A. Tourani, H.A. Rahmani, M. Naghiaei, and Y. Deldjoo, "CAPRI: Context-aware point-of-interest recommendation framework," Software Impacts, vol. 19, p. 100606, 2024, DOI: 10.1016/j.simpa.2023.100606.
H.A. Rahmani, M. Naghiaei, A. Tourani, and Y. Deldjoo, "Exploring the Impact of Temporal Bias in Point-of-Interest Recommendation," Proceedings of the 16th ACM Conference on Recommender Systems (RecSys'22), pp. 598-603, Seattle, WA, USA, 2022, DOI: 10.1145/3523227.3551481.
H.A. Rahmani, Y. Deldjoo, A. Tourani, and M. Naghiaei, "The Unfairness of Active Users and Popularity Bias in Point-of-Interest Recommendation," International Workshop on Algorithmic Bias in Search and Recommendation (BIAS'22), pp. 56-68, 2022, DOI: 10.1007/978-3-031-09316-6_6.