Multilayer Perceptron Based on Joint Training for Predicting Popularity

She，Wei1,2,3; Xu，Li4; Xu，Huibo1; Zhang，Xiaoqing5; Hu，Yue1,2,4; Tian，Zhao1

10.1007/978-3-030-57881-7_50

2020

0302-9743

1611-3349

LECT NOTES ARTIF INT

12240 LNCS

570-580

For predictive analysis, Independent features and feature combination are of equal importance, but most models only focus on either independent features or feature combinations. In this paper, we propose a novel deep network model for predictive analysis. It incorporates two components: wide simple feed-forward neural network and MLP (multilayer perceptron) neural network. The wide simple feed-forward neural network is used to generalize to unseen feature combinations, and MLP neural network’s aim to select and memorize vital independent features. The Feed-forward & MLP models are jointly trained for the Feed-forward & MLP model, in order to combine the benefits of selection, memorization and generalization. The results from the experiments show the jointly trained Neural Networks model can achieve ideal accuracy.

Generalization Memorization MLP Popularity prediction Selection

其他

英语

EI

20203909228421

Multilayer neural networks ; Predictive analytics

2-s2.0-85091268791

Scopus

1.School of Software,Zhengzhou University,Zhengzhou,450003,China
2.Yellow River Institute of Hydraulic Research,Zhengzhou,China
3.Research Center on Levee Safety Disaster Prevention,Zhengzhou,China
4.Cooperative Innovation Center of Internet Healthcare,Zhengzhou University,Zhengzhou,450052,China
5.Southern University of Science and Technology,Shenzhen,China

She，Wei,Xu，Li,Xu，Huibo,et al. Multilayer Perceptron Based on Joint Training for Predicting Popularity[C],2020:570-580.