A Hybrid Intelligent Directional Push Model for Service Platform Based on Deep Learning

2019 2nd International Conference on Advanced Materials, Intelligent Manufacturing and Automation

Shuai, Yong^1^2 ; Song, Tailiang^3 ; Wang, Yong^1^2 ; Xia, Qing^1^2 ; Su, Xinyi^4

Chongqing Ceprei Industrial Technology Research Institute, Chongqing

401332, China^1

Chongqing Engineering Research Center of Electronic Information Products Reliability, Chongqing

401332, China^2

Institute of Optoelectronics, Beijing Institute of Technology, Beijing

100081, China^3

Faculty of Information Technology, Monash University, Melbourne

3800, Australia^4

关键词: Collaborative filtering algorithms;    Data preparation;    Existing problems;    Intelligent search;    Normalization algorithms;    Service platforms;    User associations;    User satisfaction;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/569/5/052076/pdf
DOI  :  10.1088/1757-899X/569/5/052076

来源: IOP

【 摘 要 】

In order to improve the push precision and user satisfaction of the service platform, this paper proposes an intelligent directional push model based on the existing problems of the intelligent search model and the actual characteristics of the service platform. Firstly, use data preparation method to finish collecting, integrating, cleaning, conversion and protocol of the user association data in the service platform. Then deep learning model is used to build a local service platform semantic library. Thirdly, use jieba algorithm to match the user-associated data with the local service platform semantic library, set the weights of input data based on its importance, use the normalization algorithm to obtain the access matching matrix of the target users. Finally the collaborative filtering algorithm is used to calculate the user matching degree. The case analysis proves that the model in this paper has higher accuracy.