Research and Application of Project Investment Conversion Prediction Based on Improved BP Neural Network

Loading...
Thumbnail Image
Paper number
758
Working Group Number
Conference name
CIRED 2019
Conference date
3-6 June 2019
Conference location
Madrid, Spain
Peer-reviewed
Yes
Short title
Convener
Authors
Ru Sen, Fan, State Grid Shanghai Qingpu Electric Power Supply Company, China
Ya, Li, State Grid Shanghai Qingpu Electric Power Supply Company, China
Tao Tao, Ma, State Grid Shanghai Qingpu Electric Power Supply Company, China
Abstract
In order to achieve the goal of "precise investment" and promote the company's operating income and the level of investment management, an improved BP neural network prediction model for project investment conversion is proposed. With fusion of massive online and offline basic data, and generating the model sample data sets, an improved forward BP neural network prediction model is established. When handling the correction error, an additional momentum is added dynamically to adjust the momentum learning rate, so as to speed up the convergence. Based on the predicted project final amount, the rate of project investment conversion can be obtained, and also the probability of predicted value occurrence. The actual test results show that the predicted project final amounts basically fit very well with the actual values, the Mean Absolute Percentage Error is 3.96%.
Table of content
Keywords
Publisher
AIM
Date
2019-06-03
Permanent link to this record
https://cired-repository.org/handle/20.500.12455/141
http://dx.doi.org/10.34890/284
ISSN
2032-9644
ISBN
978-2-9602415-0-1