Fraud behavior detection method based on whale algorithm optimization LVQ neural network

Abstract

The invention relates to the technical field of risk control of the Internet financial industry, particularly to a fraud behavior detection method based on whale algorithm optimization LVQ neural network. The method comprises six steps. According to the invention, compared with BP, RBF, SOM and other neural networks, the LVQ neural network has the advantages of simple structure, short training time, stronger non-linear classification processing capability and the like; compared with optimization algorithms such as a genetic algorithm, a particle swarm algorithm and the like, the whale optimization algorithm has the characteristics of simple parameter setting, strong function optimization capability, strong global optimization capability, good convergence stability and the like; and the initial weight of the LVQ neural network is optimized by using the whale optimization algorithm to improve the global fitting capability, the learning rate and the prediction precision of the LVQ neuralnetwork, so that the requirement of real-time detection of networked 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 a fraud detection method based on a whale algorithm to optimize an LVQ neural network. Background technique [0002] In recent years, with the continuous progress and development of my country's economy, more and more people have begun to use credit consumption, and the demand for credit evaluation is also increasing. Machine learning algorithms such as logistic regression, decision tree, and support vector machine credit evaluation All have been successfully applied, but these algorithms are not effective in identifying financial fraud. With the development of artificial intelligence, neural networks have played an important role in the identification of Internet financial fraud. Neural networks such as error backpropagation (BP), radial basis function (RBF), and self-organizing map (SOM) have become important research fields for Internet f...

AI Technical Summary

Problems solved by technology

[0004] The LVQ neural network also has shortcomings. It only learns the winning node each time, resulting in a waste of information resources between input samples and competing nodes. It is very sensitive to the initial value in the initial stage of network learning, and the initial weight value with too much deviation will As a result, some nodes are not fully utilized and become "dead" points, and the nodes are not fully utilized, resulting in the input vectors not being well clustered in the competition layer, affecting the network convergence speed and classification accuracy

[0005] In the prior art, the swarm intelligence algorithm is used to optimize the initial weight of the LVQ neural network, such as the genetic algorithm and the particle swarm algorithm, but the genetic algorithm takes a long time to run in actual use, and the local search ability is poor; The reason for the update mechanism is that the global search ability is poor in the later stage, which reduces the prediction accuracy of the neural network

Images