Jointly Modeling Intra- and Inter-transaction Dependencies with Hierarchical Attentive Transaction Embeddings for Next-item Recommendation

@article{Wang2020JointlyMI,
  title={Jointly Modeling Intra- and Inter-transaction Dependencies with Hierarchical Attentive Transaction Embeddings for Next-item Recommendation},
  author={Shoujin Wang and Longbing Cao and Liang Hu and Shlomo Berkovsky and Xiaoshui Huang and Lin Xiao and Wenpeng Lu},
  journal={ArXiv},
  year={2020},
  volume={abs/2006.04530}
}
A transaction-based recommender system (TBRS) attempts to predict the next item by modeling dependencies in transactional data. Generally, two kinds of dependency considered are intra-transaction dependency and inter-transaction dependency. Most existing TBRSs recommend next item by only modeling the intra-transaction dependency within the current transaction while ignoring inter-transaction dependency with recent transactions that may also affect the next item. However, not all recent… 

Figures and Tables from this paper

SASNet: Stage-aware Sequential Matching for Online Travel Recommendation

This paper proposes to capture the deep sequential context by modeling the evolving of user stages, and develops a novel stage-aware deep sequential matching network (SASNet) that incorporates inter-stage and intra-stage dependencies over stage-augmented interaction sequence for more accurate and interpretable recommendation.

Session-Based Recommendations for e-Commerce with Graph-Based Data Modeling

Two session-based recommendation methods for anonymous browsing in a generic e-commerce framework using a graph where items are connected to sessions and to each other based on the order of appearance or their co-occurrence, which compete favorably on a number of datasets.

A Survey on Session-based Recommender Systems

A systematic and comprehensive review on SBRS is provided and a hierarchical framework is created to categorize the related research issues and methods of SBRS and to reveal its intrinsic challenges and complexities.

Cross-Session Aware Temporal Convolutional Network for Session-based Recommendation

CA-TCN is proposed, which can perform convolution operation on multi-hops items and maintain sequence information in the process of convolution, and outperforms state-of-the-art methods consistently.

References

SHOWING 1-10 OF 23 REFERENCES

Attention-Based Transactional Context Embedding for Next-Item Recommendation

An effective attention based transaction embedding model (ATEM) for context embedding to weight each observed item in a transaction without assuming order is designed.

Perceiving the Next Choice with Comprehensive Transaction Embeddings for Online Recommendation

A neural network-based comprehensive transaction embedding model (NTEM) which can effectively perceive the next choice in a transaction context and outperforms three typical TBRS models FPMC, PRME and GRU4Rec in terms of recommendation accuracy and novelty.

Neural Cross-Session Filtering: Next-Item Prediction Under Intra- and Inter-Session Context

This work designs an efficient neural architecture to model both intra- and inter-context for next item prediction in session-based RSs, inspired by the successful experience in modern language modeling.

Interacting Attention-gated Recurrent Networks for Recommendation

The Interacting Attention-gated Recurrent Network (IARN) is presented which adopts the attention model to measure the relevance of each time step of user and item interaction history and can selectively memorize different time steps of a user's history when predicting her preferences over different items.

Diversifying Personalized Recommendation with User-session Context

It is argued that recommendation should be diversified by leveraging session contexts with personalized user profiles, and an efficient SBRS with shallow wide-in-wide-out networks is designed, inspired by the successful experience in modern language modelings.

Inferring Implicit Rules by Learning Explicit and Hidden Item Dependency

Experimental results show that IRRMiner not only infers implicit rules of various sizes consisting of both frequent and infrequent items effectively, it also runs at least four times faster than IARMiner, a typical indirect association rule mining algorithm which can only mine size-2 indirect association rules between frequent items.

Modeling Multi-Purpose Sessions for Next-Item Recommendations via Mixture-Channel Purpose Routing Networks

A mixture-channel model to accommodate the multi-purpose item subsets for more precisely representing a session and the superiority of MCPRN over the state-of-the-art methods in terms of both recommendation accuracy and diversity is shown.

Factorizing personalized Markov chains for next-basket recommendation

This paper introduces an adaption of the Bayesian Personalized Ranking (BPR) framework for sequential basket data and shows that the FPMC model outperforms both the common matrix factorization and the unpersonalized MC model both learned with and without factorization.

Context-aware music recommendation based on latenttopic sequential patterns

This paper presents a context-aware music recommender system which infers contextual information based on the most recent sequence of songs liked by the user, and uses topic modeling to determine a set of latent topics for each song, representing different contexts.

Personalized Ranking Metric Embedding for Next New POI Recommendation

This paper proposes a personalized ranking metric embedding method (PRME) to model personalized check-in sequences and develops a PRME-G model, which integrates sequential information, individual preference, and geographical influence, to improve the recommendation performance.