A Novel Method to Determine Basic Probability Assignment Based on Adaboost and Its Application in Classification

Abstract

In the framework of evidence theory, one of the open and crucial issues is how to determine the basic probability assignment (BPA), which is directly related to whether the decision result is correct. This paper proposes a novel method for obtaining BPA based on Adaboost. The method uses training data to generate multiple strong classifiers for each attribute model, which is used to determine the BPA of the singleton proposition since the weights of classification provide necessary information for fundamental hypotheses. The BPA of the composite proposition is quantified by calculating the area ratio of the singleton proposition’s intersection region. The recursive formula of the area ratio of the intersection region is proposed, which is very useful for computer calculation. Finally, BPAs are combined by Dempster’s rule of combination. Using the proposed method to classify the Iris dataset, the experiment concludes that the total recognition rate is 96.53% and the classification accuracy is 90% when the training percentage is 10%. For the other datasets, the experiment results also show that the proposed method is reasonable and effective, and the proposed method performs well in the case of insufficient samples.

Evaluating Machine Learning Methods of Analyzing Multiclass Metabolomics

“…A detailed introduction to these methods of constructing multiclass classification models is as follows: AdaBo o st : AdaBoost is short for adaptive boosting and is based on the boosting algorithm . The structure of the model is simple, and the classification accuracy is high because it concentrates on the misclassified samples and finally constructs a well-performing model Bagging : Bagging is short for bootstrap aggregation, and it can improve accuracy by reducing variances .…”

“…AdaBo o st : AdaBoost is short for adaptive boosting and is based on the boosting algorithm . The structure of the model is simple, and the classification accuracy is high because it concentrates on the misclassified samples and finally constructs a well-performing model …”

“…70 The structure of the model is simple, and the classification accuracy is high because it concentrates on the misclassified samples and finally constructs a well-performing model. 71 • Bagging: Bagging is short for bootstrap aggregation, and it can improve accuracy by reducing variances. 72 The best prediction result is obtained based on multiple predictions of several weak models.…”

The Multisensor Data Fusion Method Based on Improved Fuzzy Evidence Theory in the Coal Mine Environment

“…Wang and Tang [13] constructed a Gaussian distribution based on the mean and variance of the training samples in the data set, and generated the BBA by calculating the function value of the test sample on the Gaussian distribution. Fu et al [14] constructed multiple strong classifiers through the Adaboost algorithm and recorded the corresponding weights for determining the BBA of singleton propositions and obtained the BBA of composite propositions by quantifying the cross-sectional area of singleton proposition intersections. In order to quantify the uncertainty of the BBA function, Pan et al [15] proposed a new belief entropy based on likelihood transformation and weighted Hartley entropy.…”

Conflicting data fusion method based on differences between fuzzy information of the single subset of focal elements

“…Therefore, it can be applied in many fields including pattern recognition [9,10], fault diagnosis [11][12][13][14][15], risk assessment [16,17]. In recent years, D-S evidence theory has received much attention as a promising theory for the handing of uncertain information in multi-sensor fusion system [18][19][20][21][22][23][24][25][26][27].…”