Sequential Recommendation Research Articles

Disentangled Dynamic Graph Attention Network for Out-of-distribution Sequential Recommendation

Sequential recommendation, leveraging user-item interaction histories to provide personalized and timely suggestions, has drawn significant research interest recently. With the power of exploiting spatio-temporal dynamics, dynamic graph neural networks (DyGNNs) show great potential in sequential recommendation by modeling the dynamic relationship between users and items. However, spatio-temporal distribution shifts naturally exist in out-of-distribution sequential recommendation, where both user-item relationships and temporal sequences demonstrate pattern shifts. The out-of-distribution scenarios may lead to the failure of existing DyGNNs in handling spatio-temporal distribution shifts in sequential recommendation, given that the patterns they exploit tend to be variant w.r.t labels under distribution shifts. In this paper, we propose D isentangled I ntervention-based D ynamic graph A ttention networks with I nvariance Promotion ( I-DIDA ) to handle spatio-temporal distribution shifts in sequential recommendation by discovering and utilizing invariant patterns, ie , structures and features whose predictive abilities are stable across distribution shifts. Specifically, we first propose a disentangled spatio-temporal attention network to capture the variant and invariant patterns. By utilizing the disentangled patterns, we design a spatio-temporal intervention mechanism to create multiple interventional distributions and an environment inference module to infer the latent spatio-temporal environments, and minimize the invariance loss to leverage the invariant patterns with stable predictive abilities under distribution shifts. Extensive experiments demonstrate the superiority of our method over state-of-the-art sequential recommendation baselines under distribution shifts.

Efficient Session-based Recommendation with Contrastive Graph-based Shortest Path Search

Session-based recommendation aims to predict the next item based on a set of anonymous sessions. Capturing user intent from a short interaction sequence imposes a variety of challenges since no user profiles are available and interaction data is naturally sparse. Recent approaches relying on graph neural networks (GNNs) for session-based recommendation use global item relations to explore collaborative information from different sessions. These methods capture the topological structure of the graph and rely on multi-hop information aggregation in GNNs to exchange information along edges. Consequently, graph-based models suffer from noisy item relations in the training data and introduce high complexity for large item catalogs. We propose to explicitly model the multi-hop information aggregation mechanism over multiple layers via shortest-path edges based on knowledge from the sequential recommendation domain. Our approach does not require multiple layers to exchange information and ignores unreliable item-item relations. Furthermore, to address inherent data sparsity, we are the first to apply supervised contrastive learning by mining data-driven positive and hard negative item samples from the training data. Extensive experiments on four different datasets show that the proposed approach outperforms almost all of the state-of-the-art methods.

A knowledge-guided process planning approach with reinforcement learning

With the widespread application of computer-aided technologies such as CAD and CAM in the manufacturing industry, a growing number of process documents and design documents generate multi-source process knowledge and expert experience. However, due to the diverse and complex representation of process knowledge, there is a need for more effective methods to mine a large amount of multi-source information and to exploit the explicit and implicit relationships between the knowledge contained in process knowledge. Effective knowledge reuse in process planning still needs to be improved. This paper proposes a reinforcement learning approach that combines knowledge graphs and process decision-making activities in process planning to exploit the learning potential of process knowledge graphs. Firstly, a reinforcement learning environment for process planning is introduced to model the process planning decision-making phase as a sequential recommendation of process knowledge. Then, this paper designs in detail the state representation method that combines process decision sequences and potential relationships between processes. This paper also creates the composite reward function that combines the process planning environment. In addition, a new algorithm is proposed for learning the proposed model more efficiently. Experimental results show that the network structure has more accurate recommendation results than other methods.

TriMLP : A Foundational MLP-Like Architecture for Sequential Recommendation

In this work, we present TriMLP as a foundational MLP-like architecture for the sequential recommendation, simultaneously achieving computational efficiency and promising performance. First, we empirically study the incompatibility between existing purely MLP-based models and sequential recommendation, that the inherent fully-connective structure endows historical user–item interactions (referred as tokens) with unrestricted communications and overlooks the essential chronological order in sequences. Then, we propose the MLP-based Triangular Mixer to establish ordered contact among tokens and excavate the primary sequential modeling capability under the standard auto-regressive training fashion. It contains (1) a global mixing layer that drops the lower-triangle neurons in MLP to block the anti-chronological connections from future tokens and (2) a local mixing layer that further disables specific upper-triangle neurons to split the sequence as multiple independent sessions. The mixer serially alternates these two layers to support fine-grained preferences modeling, where the global one focuses on the long-range dependency in the whole sequence, and the local one calls for the short-term patterns in sessions. Experimental results on 12 datasets of different scales from 4 benchmarks elucidate that TriMLP consistently attains favorable accuracy/efficiency tradeoff over all validated datasets, where the average performance boost against several state-of-the-art baselines achieves up to 14.88%, and the maximum reduction of inference time reaches 23.73%. The intriguing properties render TriMLP a strong contender to the well-established RNN-, CNN-, and Transformer-based sequential recommenders. Code is available at https://github.com/jiangyiheng1/TriMLP .

Learning Robust Sequential Recommenders through Confident Soft Labels

Sequential recommenders that are trained on implicit feedback are usually learned as a multi-class classification task through softmax-based loss functions on one-hot class labels. However, one-hot training labels are sparse and may lead to biased training and sub-optimal performance. Dense, soft labels have been shown to help improve recommendation performance. However, how to generate high-quality and confident soft labels from noisy sequential interactions between users and items is still an open question. We propose a new learning framework for sequential recommenders, CSRec, which introduces c onfident s oft labels to provide robust guidance when learning from user-item interactions. CSRec contains a teacher module that generates high-quality and confident soft labels and a student module that acts as the target recommender and is trained on the combination of dense, soft labels and sparse, one-hot labels. We propose and compare three approaches to constructing the teacher module: (i) model-level, (ii) data-level, and (iii) training-level. To evaluate the effectiveness and generalization ability of CSRec, we conduct experiments using various state-of-the-art sequential recommendation models as the target student module on four benchmark datasets. Our experimental results demonstrate that CSRec is effective in training better-performing sequential recommenders.

CBRec: A causal way balancing multidimensional attraction effect in POI recommendations

In the next Point-of-Interest recommendation, sparse and uneven location data generate biases, resulting in homogeneous recommendation outcomes that fail to reflect user preferences. Although there are many related unbiased studies, they still exhibit limitations. They lack a unified debiasing paradigm and typically employ different methods to address various biases, resulting in complex and incompatible debiasing models. Additionally, they often overlook the potential advantages of biases, thus harming the quality of location features. To address these challenges, we propose a unified debiasing paradigm by intervening in location attraction to balance the positive and negative effects of bias. By analyzing the structural causal graph, we identify attraction as a feature influenced by bias. By comparing observational results affected by attraction with counterfactual results unaffected by it, we derive a unified debiasing paradigm that eliminates the effects of bias. Additionally, through feature fusion, we embed multidimensional attraction into user features, leveraging the advantages of bias to preserve the quality of location features. Finally, experimental results on five real-world datasets demonstrate that our proposed model outperforms recent sequential recommendation models.

FELRec: efficient handling of item cold-start with dynamic representation learning in recommender systems

AbstractRecommender systems suffer from the cold-start problem whenever a new user joins the platform or a new item is added to the catalog. To address item cold-start, we propose to replace the embedding layer in sequential recommenders with a dynamic storage that has no learnable weights and can keep an arbitrary number of representations. In this paper, we present , a large embedding network that refines the existing representations of users and items in a recursive manner, as new information becomes available. In contrast to similar approaches, our model represents new users and items without side information and time-consuming finetuning, instead it runs a single forward pass over a sequence of existing representations. During item cold-start, our method outperforms similar method by 29.50–47.45%. Further, our proposed model generalizes well to previously unseen datasets in zero-shot settings. The source code is publicly available at https://github.com/kweimann/FELRec.

Sequential Recommendation Model Based on Deep Residual Recurrent Neural Network

In order to solve the problem that the sequence recommendation model based on RNN (Recurrent Neural Network) is prone to gradient vanishing or exploding when processing long sequences, which leads to the instability of the recommendation model training process, residual connection, layer normalization and feedforward neural network modules are introduced on the basis of the traditional gated recurrent unit (GRU), and a sequence recommendation model DeepGRU based on deep residual recurrent neural network is proposed. It is verified on 3 public datasets. The experimental results show that the DeepGRU has obvious advantages over the most advanced sequence recommendation methods (the average recommendation accuracy is improved 8.68%). The ablation experiment verifies the effectiveness of the introduced residual connection and other modules under the DeepGRU framework. In addition, the DeepGRU effectively alleviates the problem of unstable training process when processing long sequences.

Improving Effectiveness by Reducing Overconfidence in Large Catalogue Sequential Recommendation with gBCE loss

A large catalogue size is one of the central challenges in training recommendation models: a large number of items makes them memory and computationally inefficient to compute scores for all items during training, forcing these models to deploy negative sampling. However, negative sampling increases the proportion of positive interactions in the training data, and therefore, models trained with negative sampling tend to overestimate the probabilities of positive interactions – a phenomenon we call overconfidence . While the absolute values of the predicted scores/probabilities are not important for the ranking of retrieved recommendations, overconfident models may fail to estimate nuanced differences in the top-ranked items, resulting in degraded performance. In this paper, we show that overconfidence explains why the popular SASRec model underperforms when compared to BERT4Rec. This is contrary to the BERT4Rec authors’ explanation that the difference in performance is due to the bi-directional attention mechanism. To mitigate overconfidence, we propose a novel Generalised Binary Cross-Entropy Loss function (gBCE) and theoretically prove that it can mitigate overconfidence. We further propose the gSASRec model, an improvement over SASRec that deploys an increased number of negatives and the gBCE loss. Through detailed experiments on three datasets, we show that gSASRec does not exhibit the overconfidence problem. As a result, gSASRec can outperform BERT4Rec (e.g. +9.47% NDCG on the MovieLens-1M dataset), while requiring less training time (e.g. -73% training time on MovieLens-1M). Moreover, in contrast to BERT4Rec, gSASRec is suitable for large datasets that contain more than 1 million items. Finally, we show how addressing overconfidence can improve model calibration – the ability of a model to predict actual interaction probabilities accurately. By applying gBCE to the SASRec model on MovieLens-1M dataset, we reduce the models’ expected calibration error by 98.9% (from 0.966 to 0.01).

Exploiting explicit item–item correlations from knowledge graphs for enhanced sequential recommendation

In recent years, the research of employing knowledge graphs (KGs) in sequential recommendation (SR) has received a lot of attention, since the side information extracted from KGs, especially the information of the correlations between items, indeed helps the SR models achieve better performance. However, many previous KG-based SR models tend to introduce some noise information when learning item embeddings, or insufficiently fuse item–item correlations into their sequential modeling, thus limiting their performance improvements. In this paper, we propose a Distance-Aware Knowledge-based Sequential Recommendation model (DAKSR), which exploits the explicit item–item correlations from KGs to achieve enhanced SR. Specifically, as one critical component in our DAKSR, the distance score matrix (DSM) is first obtained to indicate the correlations between items, and then leveraged in the following three major modules of DAKSR. First, in the Item-Set Embedding layer (ISE) all item embeddings are learned based on DSM, in which the noise information is eliminated effectively. Meanwhile, the Knowledge-Infused Transformer (KIT) incorporates DSM into its attention mechanism to improve the feature extraction. Furthermore, the Knowledge Contrastive Learning module (KCL) also leverages the item–item correlations presented in DSM to generate two credible sequence views, which are used to refine sample representations through a contrastive learning strategy, and thus improve the model’s robustness. Our extensive experiments on three SR benchmarks obviously demonstrate our DAKSR’s superior performance over the state-of-the-art (SOTA) KG-based recommendation models. The implementation of our DAKSR is available at https://github.com/Easonsi/DAKSR for reproducing our experiment results conveniently.

IPSRM: An intent perceived sequential recommendation model

Objectives:Sequential recommendation aims to recommend items that are relevant to users’ interests based on their existing interaction sequences. Current models lack in capturing users’ latent intentions and do not sufficiently consider sequence information during the modeling of users and items. Additionally, noise in user interaction sequences can affect the model’s optimization process. Methods:This paper introduces an intent perceived sequential recommendation model (IPSRM). IPSRM employs the generalized expectation–maximization (EM) framework, alternating between learning sequence representations and optimizing the model to better capture the underlying intentions of user interactions. Specifically, IPSRM maps unlabeled behavioral sequences into frequency domain filtering and random Gaussian distribution space. These mappings reduce the impact of noise and improve the learning of user behavior representations. Through clustering process, IPSRM captures users’ potential interaction intentions and incorporates them as one of the supervisions into the contrastive self-supervised learning process to guide the optimization process. Results:Experimental results on four standard datasets demonstrate the superiority of IPSRM. Comparative experiments also verify that IPSRM exhibits strong robustness under cold start and noisy interaction conditions. Conclusions:Capturing latent user intentions, integrating intention-based supervision into model optimization, and mitigating noise in sequential modeling significantly enhance the performance of sequential recommendation systems.

Multi-interest sequential recommendation with contrastive learning and temporal analysis

Sequential recommendation systems aim to forecast the subsequent item of interest to users by analyzing their historical behaviors. While existing approaches, which employ attention mechanisms, have significantly advanced by capturing users’ multiple interests, they encounter two primary challenges. Firstly, they often fail to effectively capture the transient shifts in users’ interests across a sequence of items and neglect the interdependencies among these items, leading to a misalignment between the identified and actual interests. Secondly, conventional multi-interest models struggle to ensure that the identified interests are distinct, which results in overly similar interests that may not adequately satisfy user requirements. To address these issues, we propose a novel multi-interest recommendation method, which models the temporal features and user’s preference features from the user level. In order to capture short-term variations in interest, we introduce a time period module to encode the behavioral intervals between items and capture the periodicity of users clicking on similar items by extracting temporal information. In addition, we integrate similar types of items into the interest subgraph through preference feature extraction to capture users’ short-term changes in relevance term interests, and incorporate contrastive learning to enhance the differences between the captured interests. Extensive experiments conducted on two datasets Amazon Books and Taobao show that the model outperforms current state-of-the-art methods.

A Survey on Reinforcement Learning for Recommender Systems.

Recommender systems have been widely applied in different real-life scenarios to help us find useful information. In particular, reinforcement learning (RL)-based recommender systems have become an emerging research topic in recent years, owing to the interactive nature and autonomous learning ability. Empirical results show that RL-based recommendation methods often surpass supervised learning methods. Nevertheless, there are various challenges in applying RL in recommender systems. To understand the challenges and relevant solutions, there should be a reference for researchers and practitioners working on RL-based recommender systems. To this end, we first provide a thorough overview, comparisons, and summarization of RL approaches applied in four typical recommendation scenarios, including interactive recommendation, conversational recommendation, sequential recommendation, and explainable recommendation. Furthermore, we systematically analyze the challenges and relevant solutions on the basis of existing literature. Finally, under discussion for open issues of RL and its limitations of recommender systems, we highlight some potential research directions in this field.

Item attributes fusion based on contrastive learning for sequential recommendation

Item attributes fusion based on contrastive learning for sequential recommendation

HyperCLR: A Personalized Sequential Recommendation Algorithm Based on Hypergraph and Contrastive Learning

Sequential recommendations aim to predict users’ next interactions by modeling their interaction sequences. Most existing work concentrates on user preferences within these sequences, overlooking the complex item relationships across sequences. Additionally, these studies often fail to address the diversity of user interests, thus not capturing their varied latent preferences effectively. To tackle these problems, this paper develops a novel recommendation algorithm based on hypergraphs and contrastive learning named HyperCLR. It dynamically incorporates the time and location embeddings of items to model high-order relationships in user preferences. Moreover, we developed a graph construction approach named IFDG, which utilizes global item visit frequencies and spatial distances to discern item relevancy. By sampling subgraphs from IFDG, HyperCLR can align the representations of identical interaction sequences closely while distinguishing them from the broader global context on IFDG. This approach enhances the accuracy of sequential recommendations. Furthermore, a gating mechanism is designed to tailor the global context information to individual user preferences. Extensive experiments on Taobao, Books and Games datasets have shown that HyperCLR consistently surpasses baselines, demonstrating the effectiveness of the method. In particular, in comparison to the best baseline methods, HyperCLR demonstrated a 29.1% improvement in performance on the Taobao dataset.

A global contextual enhanced structural-aware transformer for sequential recommendation

Sequential recommendation (SR) has become a research hotspot recently. In our research, we observe that most existing SR models only leverage each user’s own interaction sequence to make recommendation. We argue that leveraging global contextual information across different interaction sequences could enrich item representations and thereby improve recommendation performance. To achieve this, we formulate a global graph from different sequences, providing global contextual information for each sequence. Specifically, we propose to conduct graph contrastive learning on a subgraph sampled from the global graph and a local sequence graph built from each sequence to augment item representations within each sequence. At the same time, we observe that structural dependencies, referring to relationships between items based on the graphic structure, can be extracted from the constructed global graph. Capturing structural dependencies between items may enrich the item representations. To leverage structural dependencies, we propose a new attention mechanism referred to as the Jaccard attention. While prevalent Transformer-based SR models capture semantic dependencies, referring to relationships between items based on item embeddings, in a sequence through self-attention. Therefore, it is beneficial to capture both semantic and structural dependencies between items in a sequence to further enrich item representations. Specifically, we employ two sequence encoders based on the self-attention and the proposed Jaccard attention to capture semantic and structural dependencies between items in a sequence, respectively. Overall, we propose a Global Contextual enhanced Structural-aware Transformer (GC-ST) for SR. Extensive experiments carried out on three widely used datasets demonstrate the effectiveness of GC-ST.

A Multi-view Graph Contrastive Learning Framework for Cross-Domain Sequential Recommendation

A Multi-view Graph Contrastive Learning Framework for Cross-Domain Sequential Recommendation

Enhanced side information fusion framework for sequential recommendation

Enhanced side information fusion framework for sequential recommendation

Graphical contrastive learning for multi-interest sequential recommendation

Multi-interest sequential recommendation aims to deliver accurate recommendations by modeling users’ sequential behaviors and multi-faceted interests. Contrastive learning (CL) possesses an inborn advantage in distinguishing users’ multiple interests. However, existing CL-based multi-interest sequential recommenders tend to overlook the issue of data sparsity which deteriorates model performance. Graph neural network (GNN), on the other hand, has been proven as an effective measure of tackling the data sparsity problem. However, incorporating GNN into CL requires a delicate design to simultaneously achieve distinctive interest representation learning and relieve data sparsity. To address the challenges, we propose a Graphical Contrastive Learning for Multi-Interest Sequential Recommendation (GCL4MI), which first extracts user interest and constructs interest-specific graph based on item clusters and then perform graphical contrastive learning to enhance the independence between multiple interests. Furthermore, we explore different extensible CL sampling strategies with the facilitation of GNN to learn distinct interests in a high-cohesion and low-coupling manner. Extensive experiments on three real-world datasets demonstrate that our proposed GCL4MI significantly outperforms state-of-the-arts, with average lifts of 7.0%, 7.1%, and 5.8% on recall, NDCG, and HR, respectively.

Modeling Sequences as Star Graphs to Address Over-Smoothing in Self-Attentive Sequential Recommendation

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 .