Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Overdue monitoring method for optimizing recurrent neural network based on ant colony algorithm

A technology of cyclic neural network and ant colony algorithm, which is applied in the field of overdue monitoring based on an ant colony algorithm to optimize the cyclic neural network, can solve the problems of model convergence speed and prediction accuracy that are not ideal

Inactive Publication Date: 2021-04-09
百维金科(上海)信息科技有限公司
View PDF2 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The convergence speed of the genetic algorithm is too fast in the early stage, and it is easy to fall into the local optimal solution; although the particle swarm optimization algorithm is simple to operate and can converge quickly, as the number of iterations continues to increase, while the population converges, each particle is getting more and more The more similar, it may not be possible to jump out of the local optimal solution, and the convergence speed and prediction accuracy of the obtained model are still not ideal

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Overdue monitoring method for optimizing recurrent neural network based on ant colony algorithm
  • Overdue monitoring method for optimizing recurrent neural network based on ant colony algorithm
  • Overdue monitoring method for optimizing recurrent neural network based on ant colony algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0091] see figure 1 , the present invention provides a technical solution:

[0092] An overdue monitoring method based on an ant colony algorithm to optimize a recurrent neural network, a kind of overdue monitoring method based on an ant colony algorithm to optimize a recurrent neural network, a kind of overdue monitoring method based on an ant colony algorithm to optimize a recurrent neural network, comprising the following six steps :

[0093] S1. Collect a certain proportion of customers with normal and overdue repayment performance as modeling samples, and collect basic personal information of customer account registration, user repayment, overdue, consumption and other behavior data as credit data, based on normal and overdue repayment performance. Overdue repayment performance labeling of customers;

[0094] S2. Preprocessing the credit data collected in step S1, including removing abnormal data, reducing noise and normalizing processing, and obtaining the preprocessed...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an overdue monitoring method for optimizing a recurrent neural network based on an ant colony algorithm, and the method comprises the following steps: firstly, selecting a modeling sample, and collecting credit data of the sample, and labelling the performance data of the sample; preprocessing the collected credit data, and randomly segmenting the modeling sample into a training set and a test set; determining a recurrent neural network topology structure and initializing network parameters according to the training set features, and establishing a recurrent neural network model; pre-training the weight and bias in the recurrent neural network by using an ant colony algorithm, and training the recurrent neural network by using a gradient descent algorithm; inputting a test set sample to the trained recurrent neural network for prediction comparison and evaluation. According to the method, the optimal weight and bias of the recurrent neural network are determined by using the ant colony algorithm, the convergence rate of the neural network is increased, the accuracy of the prediction model is improved, and the requirement of real-time detection of Internet financial fraud behaviors can be met.

Description

technical field [0001] The invention relates to the technical field of risk control in the Internet financial industry, in particular to an overdue monitoring method based on an ant colony algorithm to optimize a recurrent neural network. Background technique [0002] The deep learning system includes deep neural network, convolutional neural network, and recurrent neural network. The advantage of deep learning is that it can automatically use unsupervised or semi-supervised feature learning and hierarchical feature extraction efficient algorithms to replace manual feature acquisition. Its application The effect has significantly exceeded the traditional machine learning algorithm. [0003] Recurrent Neural Network (RNN) is a type of recurrent neural network that takes sequence data as input, recurses in the evolution direction of the sequence, and connects all cyclic units in a chain to form a closed loop. The cyclic neural network has the characteristics of memory and par...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06Q40/02G06N3/00G06N3/04G06N3/08G06K9/62
CPCG06N3/006G06N3/08G06N3/045G06Q40/03G06F18/214
Inventor 江远强
Owner 百维金科(上海)信息科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products