Short-term free parking berths prediction based on multitask – DBN neural network

Issue title: Special Section: Fuzzy theoretical model analysis for signal processing

Guest editors: Valentina E. Balas, Jer Lang Hong, Jason Gu and Tsung-Chih Lin

Article type: Research Article

Authors: Zhao, Hongwei^{a; c} | Liu, Yuqi^{a; c} | Huang, Yongping^{a; c; *} | Lu, Xuwang^{a; b} | Tu, Xiaohang^{a; c}

Affiliations: [a] College of Computer Science and Technology, Jilin University, Changchun, Jilin, P. R. China | [b] College of Software, Jilin University, Changchun, Jilin, P. R. China | [c] Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun, Jilin, P. R. China

Correspondence: [*] Corresponding author. Yongping Huang. E-mail: hyp@jlu.edu.cn.

Abstract: Aiming at the problem of low prediction accuracy and slow training time for Neural network with single hidden layer forecast, this paper proposes a combination of Multitask and DBN Neural network used to predict the short-term free parking berths. Firstly, the DBN Neural Network is used to carry out auto correlation analysis of the original data, and the characteristics of the data inclusion are obtained. Combined with Multitask Learning, the paper studies several related tasks simultaneously, the Neural Network can have the knowledge to new things for forecasting, and compared with the existing Neural network with single hidden layer, the data preprocessing process is reduced and better prediction results are obtained. The results show that this method shortens the training time and improves the prediction results.

Keywords: Free parking berths, multitask learning, DBN neural network

DOI: 10.3233/JIFS-179282

Journal: Journal of Intelligent & Fuzzy Systems, vol. 37, no. 4, pp. 4493-4498, 2019