Method for determining weight of weighted naive Bayesian algorithm

Abstract

The invention relates to a method for determining the weight of a weighted naive Bayesian algorithm. The method comprises the following steps: S1, setting an initial weight and selecting a weight increase amplitude; s2, in the training set, calculating the prediction accuracy of the program under the initial weight; s3, adjusting the weight so as to enable the adjusted weight to be equal to the sum of the initial weight and the weight increase amplitude, and obtaining the prediction accuracy of the program under the adjusted weight; s4, comparing the prediction accuracies before and after weight modification, if the prediction accuracy after weight modification is higher, keeping the weight after weight modification unchanged, and increasing the weight increase amplitude by 1; otherwise, recovering the weight to be in a state before modification, and reducing the weight increase amplitude by 0.1; s5, judging whether the weight increasing amplitude is 0 or not, and if so, turning to thestep S6; otherwise, turning to the step S3; s6, taking the obtained weight as the optimal weight, and outputting the optimal weight. According to the method, under the condition that attributive characters of the naive Bayesian algorithm do not meet'naive ', an optimization method for quickly determining the weight is carried out.

Description

technical field [0001] The invention belongs to the field of machine learning, and in particular relates to a method for determining the weight of a weighted naive Bayesian algorithm. Background technique [0002] Epilepsy is a common chronic brain disorder. Not only in China, but also in the whole world, many people are deeply affected by epilepsy. Epilepsy is usually detected by EEG, and the analysis work in the previous research process was done manually, which has many disadvantages. Therefore, people began to seek to replace labor by automated detection. The Naive Bayesian algorithm is a very classic algorithm in the field of machine learning, which is based on Bayesian theorem and the independent assumption of feature conditions. Prediction of epilepsy using Naive Bayesian method is very simple. The model assumes that the attributes are independent of each other, but this assumption is often not true in practical applications. When the number of attributes is large...

AI Technical Summary

Problems solved by technology

The model assumes that the attributes are independent of each other, but this assumption is often not true in practical applications. When the number of attributes is large or the correlation between attributes is large, the classification effect of the Naive Bayesian method is not good.

Images