Title: Toward business process recommendation-based collaborative filtering
Authors: Wei Luo; Zhihao Peng; Ansheng Deng; Xiaoming Bi
Addresses: Department of Complex System Modeling and Simulation, Hangzhou Dianzi University, No.1 Avenue, 2 Xiasha, Jianggan District, Hangzhou, China ' Department of Complex System Modeling and Simulation, Hangzhou Dianzi University, No.1 Avenue, 2 Xiasha, Jianggan District, Hangzhou, China ' Department of Complex System Modeling and Simulation, Hangzhou Dianzi University, No.1 Avenue, 2 Xiasha, Jianggan District, Hangzhou, China ' Department of Complex System Modeling and Simulation, Hangzhou Dianzi University, No.1 Avenue, 2 Xiasha, Jianggan District, Hangzhou, China
Abstract: Existing process recommendation methods cannot meet the various needs of personalised users. To address this problem, this paper proposed a personalised process recommendation method that is based on user behaviour preference. This method combines traditional process recommendation with user behaviour similarity and mines user behaviour preference according to the historical tracks of processes that were performed by users. In the execution of a process, the execution trace of a behaviour-similar user and executable candidate activities to be recommended that are provided by conventional process recommendation are analysed. Then, activities or recommended activities for the current user are selected to realise the automatic construction of the entire process to meet the personalised needs of users. The experimental results show that the proposed method outperforms other methods in terms of accuracy and efficiency.
Keywords: business process recommendation; flow similarity; user preferences; collaborative filtering; personalised process recommendation.
DOI: 10.1504/IJIMS.2019.103860
International Journal of Internet Manufacturing and Services, 2019 Vol.6 No.4, pp.389 - 402
Received: 05 Dec 2017
Accepted: 25 May 2018
Published online: 02 Dec 2019 *