Time-Aware CF and Temporal Association Rule-Based Personalized Hybrid Recommender System

Time-Aware CF and Temporal Association Rule-Based Personalized Hybrid Recommender System

Dan Yang, Zheng Tie Nie, Fajun Yang
Copyright: © 2021 |Pages: 16
DOI: 10.4018/JOEUC.20210501.oa2
Article PDF Download
Open access articles are freely available for download


Most recommender systems usually combine several recommendation methods to enhance the recommendation accuracy. Collaborative filtering (CF) is a best-known personalized recommendation technique. While temporal association rule-based recommendation algorithm can discover users' latent interests with time-specific leveraging historical behavior data without domain knowledge. The concept-drifting and user interest-drifting are two key problems affecting the recommendation performance. Aiming at the above problems, a time-aware CF and temporal association rule-based personalized hybrid recommender system, TP-HR, is proposed. The proposed time-aware CF algorithm considers evolving features of users' historical feedback. And time-aware users' similar neighbors selecting measure and time-aware item rating prediction function are proposed to keep track of the dynamics of users' preferences. The proposed temporal association rule-based recommendation algorithm considers the time context of users' historical behaviors when mining effective temporal association rules. Experimental results on real datasets show the feasibility and performance improvement of the proposed hybrid recommender system compared to other baseline approaches.
Article Preview

1. Introduction

Recommender systems (Adomavicius, G., & Tuzhilin, A., 2005) have been developed and researched more than 20 years. Various recommendation methods and techniques have been proposed including demographic, content based, collaborative filtering (CF) (Breese et al., 1998), knowledge based (Wu et al., 2018), utility based, association rule based (Ghoshal, A., & Sarkar, S., 2014; Ghoshal et al., 2015), and deep neural network based recommendation (Bai et al., 2017) recently. Currently almost all search engines and e-commerce websites have widely applied recommendation algorithms to bring better user experience and huge economic benefits. And cloud service provider needs to recommend its services to users (Qi et al., 2017). CF algorithm is a successful widely used technique in recommender systems which primarily works by aggregating and ranking the preferences of users with similar past behaviors. However users’ preferences and interests are not static but changing over time. Traditional user-based CF algorithm does not consider the effect of temporal information of users’ historical behaviors on user similarity, not consider the users’ interests changing over time on item rating prediction, which seriously affect the recommendation performance. In order to solve these problems, this paper proposes a time-aware user-based CF algorithm introducing time-aware similar neighbors selecting measure to find neighborhood set which is most similar to the interests and preferences of the target user. In most cases, a single recommendation algorithm has its advantages and disadvantages in different specific application scenarios. Usually most commercial recommender systems use a combination of various recommendation algorithms to form hybrid recommender systems to improve the recommendation accuracy. CF algorithm has the problem of sparsity. Temporal association rule based recommendation algorithm can discover users’ latent interests with time specific leveraging users’ historical behavior data without domain knowledge. Therefore, this paper proposes a personalized hybrid recommender system TP-HR based on time-aware CF and temporal association rule based recommendation algorithm.

The major contribution of this paper can be summarized as:

  • A comparative study of various existing time-aware recommender systems and hybrid recommender systems;

  • We propose the time-aware personalized hybrid recommender system TP-HR by combining time-aware user-based CF algorithm and temporal association rule based recommendation algorithm which not only capture temporal dynamics of users’ interests over time and considers users’ time context to recommend;

  • We propose the time-aware CF algorithm to improve the quality of CF recommendation, introducing the time decay function to the user-based CF algorithm when calculating similar users for the target user. Moreover we propose the time-aware item rating prediction function;

  • Find a suitable temporal association rule based recommendation algorithm taking into account the time context of users’ behaviors to improve the recommendation accuracy;

  • Comparative performance analysis of proposed TP-HR on real datasets with other baselines.

The remainder of the paper is structured as follows. Section II discusses related works reported in the literature. Section III presents the TP-HR framework. Section IV introduces our proposed time-aware CF algorithm in detail. Section V gives recommendation algorithm based on temporal association rule in detail. Experimental results are presented in Section VI, and Section VII concludes the paper.


In this section, we first introduce related work on time-aware recommender systems, and then association rule mining and temporal association rule mining, and finally hybrid recommender systems, which related to our algorithms.

Complete Article List

Search this Journal:
Volume 36: 1 Issue (2024)
Volume 35: 3 Issues (2023)
Volume 34: 10 Issues (2022)
Volume 33: 6 Issues (2021)
Volume 32: 4 Issues (2020)
Volume 31: 4 Issues (2019)
Volume 30: 4 Issues (2018)
Volume 29: 4 Issues (2017)
Volume 28: 4 Issues (2016)
Volume 27: 4 Issues (2015)
Volume 26: 4 Issues (2014)
Volume 25: 4 Issues (2013)
Volume 24: 4 Issues (2012)
Volume 23: 4 Issues (2011)
Volume 22: 4 Issues (2010)
Volume 21: 4 Issues (2009)
Volume 20: 4 Issues (2008)
Volume 19: 4 Issues (2007)
Volume 18: 4 Issues (2006)
Volume 17: 4 Issues (2005)
Volume 16: 4 Issues (2004)
Volume 15: 4 Issues (2003)
Volume 14: 4 Issues (2002)
Volume 13: 4 Issues (2001)
Volume 12: 4 Issues (2000)
Volume 11: 4 Issues (1999)
Volume 10: 4 Issues (1998)
Volume 9: 4 Issues (1997)
Volume 8: 4 Issues (1996)
Volume 7: 4 Issues (1995)
Volume 6: 4 Issues (1994)
Volume 5: 4 Issues (1993)
Volume 4: 4 Issues (1992)
Volume 3: 4 Issues (1991)
Volume 2: 4 Issues (1990)
Volume 1: 3 Issues (1989)
View Complete Journal Contents Listing