Table 2 - uploaded by Pengjie Ren
Content may be subject to copyright.
Source publication
Given e-commerce scenarios that user profiles are invisible, session-based recommendation is proposed to generate recommendation results from short sessions. Previous work only considers the user's sequential behavior in the current session, whereas the user's main purpose in the current session is not emphasized. In this paper, we propose a novel...
Similar publications
Facial micro-expressions, which usually last only for a fraction of a second, are challenging to detect by the human eye or machine. They are useful for understanding the genuine emotional state of a human face, and have various applications in education, medical, surveillance and legal sectors. Existing works on micro-expressions are focused on bi...
Citations
... Tang and Wang [29] combines convolutional sequence embedding model with Top-N sequential recommendation as a way for temporal recommendation. Li et al. [30] designed a new neural attention recommender network to consider the sequential behavior of user in current session. Zhang et al. [31] introduced attention mechanism into the sequence-sense recommendation model to represent user's temporal interest. ...
... Tang and Wang [29] combined convolutional sequence embedding model with Top-N sequential recommendation as a way for temporal recommendation. Li et al. [30] designed a new neural attention recommender network that considered the sequential behavior of user in current session. Zhang et al. [31] introduced attention mechanism into the sequence-sense recommendation model that represented user's temporal interest. ...
With the development of Internet technology, the problem of information overload has increasingly attracted attention. Nowadays, the recommendation system with excellent performance in information retrieval and filtering would be widely used in the business field. However, most existing recommendation systems are considered a static process, during which recommendations for internet users are often based on pre-trained models. A major disadvantage of these static models is that they are incapable of simulating the interaction process between users and their systems. Moreover, most of these models only consider users’ real-time interests while ignoring their long-term preferences. This paper addresses the abovementioned issues and proposes a new recommendation model, DRR-Max, based on deep reinforcement learning (DRL). In the proposed framework, this paper adopted a state generation module specially designed to obtain users’ long-term and short-term preferences from user profiles and user history score item information. Next, Actor-Critical algorithm is used to simulate the real-time recommendation process.Finally, this paper uses offline and online methods to train the model. In the online mode, the network parameters were dynamically updated to simulate the interaction between the system and users in a real recommendation environment. Experimental results on the two publicly available data sets were used to demonstrate the effectiveness of our proposed model.
... Recently, a sequential-recommendation approach has been proposed to model user behavior from user historical sequences. Many such methods employ recurrent neural networks (RNN) [2]- [4] and the most recent methods improve RNN-based methods by applying an attention mechanism [5] both unidirectionally [6] and bidirectionally [7]. However, all of the above methods model the user behavior for the whole sequence, whereas the user behavior may have changed over time. ...
... RRN [17] utilizes the long short-term memory concept [18] to learn the dynamic embeddings for users and items. NARM [4] combines GRU with an attention mechanism to model the user sequential behavior and capture the user's main purpose in session-based recommendations. ACA-GRU [19] applies GRU with an attention mechanism to model the user behavior from four classified contexts. ...
Previous sequential-recommendation methods have been able to capture patterns of item characteristics that interact with the user. However, they modeled user behavior using a whole interaction sequence, despite possible changes in a user’s behavior over time, which can make some behaviors no longer relevant by the end of the user–item interaction period. That is, each item representation was derived from the influences of all items in the whole sequence (i.e. global representation) without considering item-adjacency factors that can affect the item characteristics during the interaction period (i.e. local representation). Furthermore, these methods modeled only user behavior, ignoring item behavior, which can involve patterns of user characteristics for users who interact with the item. In this paper, we therefore propose a novel attentive local-interaction model for sequential recommendation called ARERec, which applies a region-embedding technique to both user and item historical sequences. The aim is to model user and item behavior over the user–item interaction sequence while considering local representations that contain specific characteristics of both user and item in the sequence. In this way, information is derived for corresponding periods that reflect more-specific reasons behind the interaction in the sequence. Moreover, to account for ratings and neighbor-related factors, we adopt the concept of neighbor-based collaborative filtering in our predictions. One issue is that neighbors have the same similarity levels for all users, resulting in similar predictions, even though there may be different user-specific information. To address this problem, we apply a multi-head attention mechanism to personalize each neighbor based on the user’s characteristics. Extensive experiments on three datasets demonstrate that ARERec consistently outperforms state-of-the-art sequential methods, including Recurrent Neural Networks and the attention-based methods (both unidirectional and bidirectional) HitRate and Normalized Discounted Cumulative Gain. Our experiments also show that ARERec provides superior results by considering interaction periods with local representation rather than the whole sequence using global representation.
... The purpose of both is to explore the user preference timing relationship in the sessions. In 2017, Tuan et al. combined RNNs with session recommendations to mine information from time changes in user behaviour and implement recommendations [12]. Hidasi et al. [10] studied the application of the RNN and session in the next item recommendation, which used Gated Recurrent Neural Networks to redefine the classic RNNs. ...
The biggest challenge to recommendation systems based on user preferences is how to improve the ability of the recommendation system to mine and analyse user preferences and behaviours. In this process, we must not only consider the continuation of the user’s long-term preference but also improve the system’s ability to accommodate short-term preferences and discrete preferences. To this end, we focus on the performance of time factors of user preferences. However, the issue we are concerned about has not received much attention in the existing research. We propose a new recommendation model based on the perspective of user sessions, namely GACOforRec. This model can handle long-term and stable preferences at the same time and preserve the hierarchy of potential preferences. We conducted a large number of comparative experiments on two real datasets, and the results show that GACOforRec is significantly better than other state-of-the-art methods in the study of user sessions.
... However, as a user may accidentally click on wrong or irrelevant items, traditional RNNs that do not differentiate the importance of all the visited items may fail to capture the user's main purpose of the current session. To address this issue, Li et al. [18] improved the RNN-based model by introducing the attention mechanism to help the recommender capture the user's current intention. However, there are two limitations w.r.t. ...
... Ren et al. [25] explored the repeated consumption phenomenon in session recommendation, and incorporated a repeat-explore mechanism into neural networks. Li et al. [18] adopted an attention mechanism to capture the user's main intention in the current session, since users' interactions are often accompanied by randomness. Liu et al. [20] used the embedding of the last-click to represent the user's current interests, and built the attention model on top of it to capture the user's short-term intention. ...
... Differences: Our method has significant differences with these existing methods. First, although the previous work [18] has introduced an attention mechanism to capture the user's main purposes of the current session, they did not consider the user's intention from their historical interactions, which might be useful to the session recommender. Our method is also different from the works in [24] and [19], which try to model both the user's long and shortterm interests, but the randomness of user interactions was not considered, and the user's current intention cannot be captured. ...
Session-based Recommendation (SR) is the task of recommending the next item based on previously recorded user interactions. In this work, we study SR in a practical streaming scenario, namely Streaming Session-based Recommendation (SSR), which is a more challenging task due to (1) the uncertainty of user behaviors, and (2) the continuous, large-volume, high-velocity nature of the session data. Recent studies address (1) by exploiting the attention mechanism in Recurrent Neural Network (RNN) to better model the user's current intent, which leads to promising improvements. However, the proposed attention models are based solely on the current session. Moreover, existing studies only perform SR under static offline settings and none of them explore (2). In this work, we target SSR and propose a Streaming Session-based Recommendation Machine (SSRM) to tackle these two challenges. Specifically, to better understand the uncertainty of user behaviors, we propose a Matrix Factorization (MF) based attention model, which improves the commonly used attention mechanism by leveraging the user's historical interactions. To deal with the large-volume and high-velocity challenge, we introduce a reservoir-based streaming model where an active sampling strategy is proposed to improve the efficiency of model updating. We conduct extensive experiments on two real-world datasets. The experimental results demonstrate the superiority of the SSRM method compared to several state-of-the-art methods in terms of MRR and Recall.
Learning resource recommendation systems can help learners find suitable resources (e.g., books, journals, …) for learning and research. In particular, in the context of online learning due to the impact of the COVID-19 pandemic, the learning resource recommendation is very necessary. In this study, we propose using session-based recommendation systems to suggest the learning resources to the learners. Experiments are performed on a learning resource dataset collected at a local university and a public dataset. After preprocessing the data to convert it to session form, the Neural Attentive Session-based Recommendation (NARM) and Recurrent Neural Networks (GRU4Rec) models were used for training, testing, and comparison. The results show that recommending learning resources according to the NARM model is more effective than that of the GRU4Rec model, and thus, using the session-based recommendation system would be a promising approach for learning resource recommendation.KeywordsLearning resource recommendationSession-based recommender systemNARMGRU4Rec
In a recent study, it was shown that, with batch training of a graph neural network, it is possible to recommend suitable items for users. Although the method has obtained item embedding and considered the complex transitions between items, there is no multi-dimensional focus on the users’ interests and preferences. In this paper, we propose a multi-head attention graph neural network (MAE-GNN) for session-based recommendation by combining a dual-gated graph neural network and multi-head attention mechanisms. MAE-GNN can select important node information and extract users’ interests and preferences from multiple dimensions. Experimental evaluation has been conducted to show that, compared with the state-of-the-art methods, the proposed model has significant improvement in term of P@K and MRR@K for session-based recommendation.
Recommender systems have achieved great success in many fields, benefiting from the accessibility of massive amounts of user behavior data. However, the increasing concern about privacy issues brings new research challenges to traditional data-centralized recommender systems. As a privacy-preserving machine learning technique, federated learning provides potential solutions to these research challenges, by jointly training perceptive global recommendation models based on the decentralized user behavior data. Although existing federated recommendation methods have achieved promising results, they still suffer the following limitations. Firstly, they mainly focus on the horizontal federated recommendation scenarios where the participants are individual users, without considering the scenarios where participants are organizations. Secondly, they tend to model static user preferences and cannot learn the preference evolution in users’ interaction sequences. To address these limitations, we propose a horizontal Federated recommendation framework for Sequential Recommendation (FedSeqRec), where the participants are organizations. As the data in different organizations are usually not independent and identically distributed (Non-IID), federated training tends to yield suboptimal results. To alleviate this problem, we propose a federated re-average algorithm, which takes the consistency of the client’s data distribution with the overall data distribution into consideration. Moreover, we also leverage a low-rank tensor projection to model user long-term preferences. Experimental results on two real-world datasets demonstrate that FedSeqRec outperforms state-of-the-art federated recommendation methods. The implementation code of FedSeqRec is avaiable at https://github.com/MuziLee-x/FedSeqRec.
