Session-based Recommendations with Recurrent Neural Networks
By Balázs Hidasi, Alexandros Karatzoglou, Linas Baltrunas, Domonkos Tikk. 2015
2,897 citations recorded by the source corpus.
We apply recurrent neural networks (RNN) on a new domain, namely recommender systems. Real-life recommender systems often face the problem of having to base recommendations only on short session-based data (e.g. a small sportsware website) instead of long user histories (as in the case of Netflix). In this situation the frequently praised matrix factorization approaches are not accurate. This problem is usually overcome in practice by resorting to item-to-item recommendations, i.e. recommending similar items. We argue that by modeling the whole session, more accurate recommendations can be provided. We therefore propose an RNN-based approach for session-based recommendations. Our approach also considers practical aspects of the task and introduces several modifications to classic RNNs such as a ranking loss function that make it more viable for this specific problem. Experimental results on two data-sets show marked improvements over widely used approaches.
Subfields: Artificial Intelligence, Information Systems, Management Science and Operations Research, Model Architectures, Recommender Systems
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Xiangnan He
- Yongfeng Zhang
- Wayne Xin Zhao
- Julian McAuley
- Xu Chen
- Xiang Wang
- Ji-rong Wen
- Tat-seng Chua
In Information Systems
- Xiangnan He
- Yongfeng Zhang
- Xu Chen
- Xiang Wang
- Dietmar Jannach
- Julian McAuley
- Wayne Xin Zhao
- Hongzhi Yin
In Management Science and Operations Research
- Xiangnan He
- Dietmar Jannach
- Malte Ludewig
- Julian McAuley
- Wang-Cheng Kang
- Yan Wang
- Longbing Cao
- Quan Z. Sheng
In Model Architectures
- Dietmar Jannach
- Malte Ludewig
- Alexandros Karatzoglou
- Julian McAuley
- Balázs Hidasi
- Hao Wang
- Wang-Cheng Kang
- Xiangnan He
In Recommender Systems
- Xiangnan He
- Yongfeng Zhang
- Julian McAuley
- Xu Chen
- Dietmar Jannach
- Malte Ludewig
- Alexandros Karatzoglou
- Xiang Wang
Citations in the corpus
- Variational Autoencoders for Collaborative Filtering (2018, methodology): “For example, it has been used in recurrent neural networks for session-based sequential recommendation [8, 15, 16, 42, 45] and in feedward neural networks applied to Youtube recommendation [9].”
- Contextual Sequence Modeling for Recommendation with Recurrent Neural Networks (2017, methodology): “[5] applied RNNs to the task of session-based recommendation.”
- Recurrent Latent Variable Networks for Session-Based Recommendation (2017, methodology): “In this context, the work recently presented in [13] constitutes the most characteristic development.”
- Recurrent Neural Networks with Top-k Gains for Session-based Recommendations (2017, positive): “Recurrent Neural Networks have been used with success in the area of session-based recommendations; [7] proposed a Recurrent Neural Network with a pairwise ranking loss for this task.”
- Improved Recurrent Neural Networks for Session-based Recommendations (2016, methodology): “The dataset is split following [10], where sessions in the last day are placed in the test set, and everything else is placed in the training set.”
- AutoEmb: Automated Embedding Dimensionality Search in Streaming Recommendations (2020, background): “Hidasi et al.[10] firstly proposed a session-based recommendation model named GRU4Rec to model the sequential influence of items’ transition.”
- Online Fast Adaptation and Knowledge Accumulation: a New Approach to Continual Learning (2020, background): “…on locally stationary data distributions, e.g., an autonomous vehicle navigating specific weather, a recommendation system dealing with users with time-varying interest (Hidasi et al., 2015), or an embedded agent doing a specific task for a while (Lomonaco et al., 2019a; Padakandla et al., 2019).”
- Dynamical Isometry and a Mean Field Theory of RNNs: Gating Enables Signal Propagation in Recurrent Neural Networks (2018, neutral): “, 2013) and recommendation systems (Hidasi et al., 2015; Wu et al., 2017).”
- Imitation-Regularized Offline Learning (2019, background): “Formally, we estimate Q(a, x), the expected reward from taking an action a in context x; decisions are then implied by the greedy policy that selects actions with the highest expected reward in each decision context [15, 23, 11, 21].”
- Handling Information Loss of Graph Neural Networks for Session-based Recommendation (2020, methodology): “We choose GRU instead of LSTM because it was shown that GRU outperforms LSTM for the session-based recommendation task [8, 11].”
- RecSim: A Configurable Simulation Platform for Recommender Systems (2019, methodology): “However in recent years, RS research has increasingly started to explore sequential patterns in user interaction using HMMs, RNNs and related methods [He and McAuley, 2016, Hidasi et al., 2016, Wu et al., 2017].”
- Self-Attentive Sequential Recommendation (2018, methodology): “GRU4Rec [2]) have a dependency on time steps (i.”
- A Survey on Session-based Recommender Systems (2019, methodology): “Session-based Recommendation with Graph Neural Networks (SR-GNN) [129] is the pioneering work that introduced GNN into SBRSs and is claimed to have achieved superior performance, compared with non-GNN approaches including Short-Term Attention/Memory Priority Model (STAMP) [58], Neural Attentive Reco”
- Large Language Models are Zero-Shot Rankers for Recommender Systems (2023, background): “…the real user preference, since existing models mainly capture user preference from historical interaction behaviors, e.g., clicked item sequences [14, 19, 39, 17], limiting the expressive power to model the complicated user interests ( e.g., user intentions expressed in natural language).”
- Neural Attentive Session-based Recommendation (2017, methodology): “• GRU-Rec : We denote the model proposed in [12] as GRU-Rec, which utilizes session-parallel mini-batch training process and also employs ranking-based loss functions for learning the model.”
- When large language models meet personalization: perspectives of challenges and opportunities (2023, neutral): “Self-Attentions [20] and Gated Recurrent Units (GRUs) [21] are popular choices for modeling sequential data in recommender systems.”
- Recommender Systems with Generative Retrieval (2023, methodology): “On the other hand, sequential recommenders [6, 11, 17, 23, 31, 46, 49], which explicitly take into account the order of user-item interactions, are also recently popular.”
- Artificial intelligence in recommender systems (2020, background): “RNNs have emerged as a new direction known as session-based recommender systems or sequential recommender systems where the real-time recommendation is refined according to the historical sequential data [96, 97].”