The significance of estimating video watch time has been highlighted by the rising importance of (short) video recommendation, which has become a core product of mainstream social media platforms. Modeling video watch time, however, has been challenged by the complexity of user-video interaction, such as different user behavior modes in watching the recommended videos and varying watching probability over the video progress bar. Despite the importance and challenges, existing literature on modeling video watch time mostly focuses on relatively black-box mechanical enhancement of the classical regression/classification losses, without factoring in user behavior in a principled manner. In this paper, we for the first time take on a user-centric perspective to model video watch time, from which we propose a white-box statistical framework that directly translates various user behavior assumptions in watching (short) videos into statistical watch time models. These behavior assumptions are portrayed by our domain knowledge on users' behavior modes in video watching. We further employ bucketization to cope with user's non-stationary watching probability over the video progress bar, which additionally helps to respect the constraint of video length and facilitate the practical compatibility between the continuous regression event of watch time and other binary classification events. We test our models extensively on two public datasets, a large-scale offline industrial dataset, and an online A/B test on a short video platform with hundreds of millions of daily-active users. On all experiments, our models perform competitively against strong relevant baselines, demonstrating the efficacy of our user-centric perspective and proposed framework.

Abstract Session-based recommendation aims to generate recommendations for the next item of users’ interest based on a given session. In this manuscript, we develop intention enhanced mixed attentive model () to generate session-based recommendations using two important factors: temporal patterns and estimates of users’ intentions. Unlike existing methods which primarily leverage complicated gated recurrent units to model the temporal patterns, models the temporal patterns using a light-weight while effective position-sensitive attention mechanism. In , we also leverage the estimate of users’ prospective preferences to signify important items, and generate better recommendations. Our experimental results demonstrate that models significantly outperform the state-of-the-art methods in six benchmark datasets, with an improvement as much as 19.2%. In addition, our run-time performance comparison demonstrates that during testing, models are much more efficient than the best baseline method, with a significant average speedup of 47.7 folds.

Self-attention (SA) mechanisms have been widely used in developing sequential recommendation (SR) methods, and demonstrated state-of-the-art performance. However, in this article, we show that self-attentive SR methods substantially suffer from the over-smoothing issue that item embeddings within a sequence become increasingly similar across attention blocks. As widely demonstrated in the literature, this issue could lead to a loss of information in individual items, and significantly degrade models’ scalability and performance. To address the over-smoothing issue, in this article, we view items within a sequence constituting a star graph and develop a method, denoted as \(\mathop{\mathtt{MSSG}}\limits\) , for SR. Different from existing self-attentive methods, \(\mathop{\mathtt{MSSG}}\limits\) introduces an additional internal node to specifically capture the global information within the sequence, and does not require information propagation among items. This design fundamentally addresses the over-smoothing issue and enables \(\mathop{\mathtt{MSSG}}\limits\) a linear time complexity with respect to the sequence length. We compare \(\mathop{\mathtt{MSSG}}\limits\) with eleven state-of-the-art baseline methods on six public benchmark datasets. Our experimental results demonstrate that \(\mathop{\mathtt{MSSG}}\limits\) significantly outperforms the baseline methods, with an improvement of as much as 10.10%. Our analysis shows the superior scalability of \(\mathop{\mathtt{MSSG}}\limits\) over the state-of-the-art self-attentive methods. Our complexity analysis and runtime performance comparison together show that \(\mathop{\mathtt{MSSG}}\limits\) is both theoretically and practically more efficient than self-attentive methods. Our analysis of the attention weights learned in SA-based methods indicates that on sparse recommendation data, modeling dependencies in all item pairs using the SA mechanism yields limited information gain, and thus, might not benefit the recommendation performance. Our source code and data are publicly accessible through GitHub .

Sequential recommendation aims to identify and recommend the next few items of users’ interest. It becomes an effective tool to help users select their favorite items from a variety of options. A key challenge in sequential recommendation is to learn the patterns and dynamics, which are most pertinent to inform future interactions of users. With the prosperity of deep learning, many deep models, particularly based on recurrent neural networks [1] and with attention mechanisms [2] , [3] , have been developed for sequential recommendation purposes. However, our analysis demonstrates that, these deep models, particularly those with attention mechanisms, may not always learn meaningful attention weights from the extremely sparse recommendation data, and thus, could degrade the recommendation performance. Therefore, in this study, instead of deep models, we develop novel, effective and efficient hybrid associations models (HAM) to better learn from the sparse and limited recommendation data. This study has been published in IEEE Transactions on Knowledge and Data Engineering. Please refer to the full manuscript [4] for more details.

Next-basket recommendation considers the problem of recommending a set of items into the next basket that users will purchase as a whole. In this paper, we develop a novel mixed model with preferences, popularities and transitions ($\mathop {\mathtt {M^2}}\limits$) for the next-basket recommendation. This method models three important factors in next-basket generation process: 1) users’ general preferences, 2) items’ global popularities and 3) transition patterns among items. Unlike existing recurrent neural network-based approaches,$\mathop {\mathtt {M^2}}\limits$does not use the complicated networks to model the transitions among items, or generate embeddings for users. Instead, it has a simple encoder-decoder based approach ($\mathop {\mathtt {ed\text{-}Trans}}\limits$) to better model the transition patterns among items. We compared$\mathop {\mathtt {M^2}}\limits$with different combinations of the factors with 5 state-of-the-art next-basket recommendation methods on 4 public benchmark datasets in recommending the first, second and third next basket. Our experimental results demonstrate that$\mathop {\mathtt {M^2}}\limits$significantly outperforms the state-of-the-art methods on all the datasets in all the tasks, with an improvement of up to 22.1%. In addition, our ablation study demonstrates that the$\mathop {\mathtt {ed\text{-}Trans}}\limits$is more effective than recurrent neural networks in terms of the recommendation performance. We also have a thorough discussion on various experimental protocols and evaluation metrics for next-basket recommendation evaluation.

Sequential recommendation aims to identify and recommend the next few items for a user that the user is most likely to purchase/review, given the user's purchase/rating trajectories. It becomes an effective tool to help users select favorite items from a variety of options. In this manuscript, we developed hybrid associations models (HAM) to generate sequential recommendations. using three factors: 1) users' long-term preferences, 2) sequential, high-order and low-order association patterns in the users' most recent purchases/ratings, and 3) synergies among those items. HAM uses simplistic pooling to represent a set of items in the associations, and element-wise product to represent item synergies of arbitrary orders. We compared HAM models with the most recent, state-of-the-art methods on six public benchmark datasets in three different experimental settings. Our experimental results demonstrate that HAM models significantly outperform the state of the art in all the experimental settings. with an improvement as much as 46.6%. In addition, our run-time performance comparison in testing demonstrates that HAM models are much more efficient than the state-of-the-art methods. and are able to achieve significant speedup as much as 139.7 folds.