In the era of information explosion, users of online services are urgently waiting for timely and effective recommendations. In this paper, we present the first study on the responsiveness aspect of recommender system and present Responsive Online RecOmmender System (RØROS) based on Meta-Gradients Update (MGU) techniques, which helps improve the recommendation quality for both existing and new users when the system only observes a limited number of in-coming interactions. Technically, we propose a new sampling strategy to enable the recommender model to update itself in the unit of user-centered interaction graphs, which helps balance the responsiveness to different types of users. For model updating, we propose an MGU-based strategy to effectively capture users’ current short-term preference, which hence improve RØROS’s responsiveness to users. Mean-while, RØROS also maintains users’ long-term preference with a preference memory module. Extensive experiments and case studies on standard benchmarks validate the responsiveness of our novel recommender system design, which we hope would foster future studies in this new direction.

Lanqing Xue, Kaitao Song, Duocai Wu, Xu Tan, Nevin L. Zhang, Tao Qin, Wei-Qiang Zhang, Tie-Yan Liu. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Location-based embedding is a fundamental problem to solve in location-based social networks (LBSN). In this paper, we propose a geographical convolutional neural tensor network (GeoCNTN) as a generic embedding model. GeoCNTN first takes the raw location data and extracts from it a more well-conditioned representation by our proposed Geo-CMeans algorithm. We then use a convolutional neural network (CNN) and an embedding structure to extract individual latent structural patterns from the preprocessed data. Finally, we apply a neural tensor network (NTN) to craft the implicitly related features we have obtained into a unified geographical feature. The advantages of our GeoCNTN mainly come from its novel neural network structure, which intrinsically offers a mechanism to extract latent structural features from the geographical data, as well as its wide applicability in various LBSN-related tasks. From two case studies, i.e. link prediction and entity classification in user-group LBSN, we evaluate the embedding efficacy of our model. Results show that GeoCNTN significantly performs better on at least two tasks, with improvement by 9% w.r.t. NDCG and 11% w.r.t. F1 score respectively, using the Meetup-USA dataset.