Session-based recommendation is a task to predict users’ next actions given a sequence of previous actions in the same session. Existing methods either encode the previous actions in a strict order or completely ignore the order. It is not necessary to always capture the sequential information in sessions by following a strict order, because sometimes, the order of actions in a short sub-sequence, called the detailed order, may not be important, e.g., when a user is just comparing the same kind of products from different brands. We term the property that the order of actions in the sub-session level does not matter the local invariance. Nevertheless, the high-level ordering information is still useful because the data is sequential in nature. Therefore, a good session-based reco...[ Read more ]

Session-based recommendation is a task to predict users’ next actions given a sequence of previous actions in the same session. Existing methods either encode the previous actions in a strict order or completely ignore the order. It is not necessary to always capture the sequential information in sessions by following a strict order, because sometimes, the order of actions in a short sub-sequence, called the detailed order, may not be important, e.g., when a user is just comparing the same kind of products from different brands. We term the property that the order of actions in the sub-session level does not matter the local invariance. Nevertheless, the high-level ordering information is still useful because the data is sequential in nature. Therefore, a good session-based recommender should consider the local invariance property while capturing the sequential information by paying different attention to the ordering information in different levels of granularity. To this end, we propose a novel model called LINet to automatically ignore the insignificant detailed ordering information in some sub-sessions, while keeping the high-level sequential information of the whole sessions. In the model, we first use a full self-attention layer with Gaussian weighting to extract features of sub-sessions, and then we apply a recurrent neural network to capture the high-level sequential information. Extensive experiments on two real-world datasets show that our method outperforms or matches the state-of-the-art methods and the proposed mechanism to consider the local invariance property plays an important role.