[Paper Review] Deep Learning for Sequential Recommendation: Algorithms, Influential Factors, and Evaluations

Hui Fang, Danning Zhang+2 more|arXiv (Cornell University)|Apr 30, 2019

Recommender Systems and TechniquesComputer Science146 references78 citations

TL;DR

A comprehensive survey of DL-based sequential recommender systems, introducing a three-type behavior-sequence taxonomy, influential factors, and empirical evaluations to guide design and practice.

ABSTRACT

In the field of sequential recommendation, deep learning (DL)-based methods have received a lot of attention in the past few years and surpassed traditional models such as Markov chain-based and factorization-based ones. However, there is little systematic study on DL-based methods, especially regarding to how to design an effective DL model for sequential recommendation. In this view, this survey focuses on DL-based sequential recommender systems by taking the aforementioned issues into consideration. Specifically,we illustrate the concept of sequential recommendation, propose a categorization of existing algorithms in terms of three types of behavioral sequence, summarize the key factors affecting the performance of DL-based models, and conduct corresponding evaluations to demonstrate the effects of these factors. We conclude this survey by systematically outlining future directions and challenges in this field.

Motivation & Objective

• Provide a comprehensive overview of DL techniques used in sequential recommendation.
• Propose a classification framework for sequential recommendation tasks (three behavioral sequence types).
• Identify and analyze influential factors affecting DL-based sequential recommenders and validate their effects with experiments.
• Compare DL-based approaches with traditional methods and outline practical guidelines and open challenges.

Proposed method

• Survey and taxonomy of DL-based sequential recommender systems across three task categories: experience-based, transaction-based, and interaction-based recommendations.
• Review of DL techniques (RNNs, CNNs, MLPs, attention mechanisms, GNNs) and how they are applied to sequential data.
• Empirical evaluation on real datasets to demonstrate the impact of influential factors on performance.
• Synthesis of open issues and future research directions based on literature and experimental insights.

Experimental results

Research questions

• RQ1 What DL techniques are used for sequential recommendation and how can tasks be categorized by behavior sequences?
• RQ2 What factors influence the performance of DL-based sequential recommenders and how do they affect accuracy?
• RQ3 How do DL-based approaches compare to traditional methods for sequential recommendation in practice?
• RQ4 What are the key open issues and promising directions for future research in DL-based sequential recommendation.

Key findings

• DL-based sequential recommender models achieve state-of-the-art performance compared to traditional approaches like Markov chains and MF-based methods.
• A long sequence can be leveraged to learn the overall theme of user behavior, improving robustness on sparse data and handling varied input lengths.
• Different DL techniques (RNNs, CNNs, MLPs, attention, GNNs) offer complementary strengths for capturing intra- and inter-session dynamics.
• Multi-behavior and multi-task learning can enhance modeling of user intents expressed via different behavior types.
• There are important challenges in explainability and training efficiency, motivating future work on interpretable and scalable models.