Improved naive Bayes classification algorithm for traffic risk management

Abstract

Naive Bayesian classification algorithm is widely used in big data analysis and other fields because of its simple and fast algorithm structure. Aiming at the shortcomings of the naive Bayes classification algorithm, this paper uses feature weighting and Laplace calibration to improve it, and obtains the improved naive Bayes classification algorithm. Through numerical simulation, it is found that when the sample size is large, the accuracy of the improved naive Bayes classification algorithm is more than 99%, and it is very stable; when the sample attribute is less than 400 and the number of categories is less than 24, the accuracy of the improved naive Bayes classification algorithm is more than 95%. Through empirical research, it is found that the improved naive Bayes classification algorithm can greatly improve the correct rate of discrimination analysis from 49.5 to 92%. Through robustness analysis, the improved naive Bayes classification algorithm has higher accuracy.

Expecting individuals’ body reaction to Covid-19 based on statistical Naïve Bayes technique

“…Two main scenarios for the implementation of the CPES will be pursued. According to the first scenario, the IBGA as a hybrid feature selection method will be tested against other modern selection methods using NB classifier as a standard classification method [ 29 , 30 , 31 ]. Then, the complete strategy called CPES will be tested against other modern diagnostic strategies.…”

“…The modern selection methods used in the comparison are Genetic Algorithm (GA) [ 2 , 27 , 28 ], Feature Selection via Directional Outliers Correcting (FSDOC) [36] , Orthogonal Least Squares (OLS) based feature selection method [37] , the Modified Grasshopper Optimization Algorithm (MGOA) [38] , and Stochastic Diffusion Search (SDS) algorithm [29] . To evaluate these features selection methods, NB classifier is used as a standard method [ 29 , 30 , 31 ]. The figures (8→13) show the accuracy, error, precision, recall, and run-time of the used feature selection methods.…”

“…According to table 2 , it is assumed that the used classifiers are Naïve Bayes (NB) [ 29 , 30 , 31 ], K-Nearest Neighbors (KNN) [5] , and Support Vector Machine (SVM) [11] . Based on their accuracy values for chromosomes, it is noted that NB and KNN proven that the first chromosome ( Ch 1 ) is better than the second one.…”

“…Generally, the success of the classification is built upon the accurate selection of discriminative features. Hence, a weight can be given to each feature to measure its ability for discrimination, which can be called the feature weight [31] . In the proposed SNB, the weight of each feature is calculated according to how close the value of that feature is in the elements of the same class and how far apart the values ​​of that feature are between the items of the different classes.…”

Multinomial Naïve Bayes Classifier: Bayesian versus Nonparametric Classifier Approach

“…However, KDE is an effective way of estimating probabilistic function of a given data when the distribution of the data is unknown. In other words, KDE makes it easy to solve the problem of non-Gaussian distribution when the dataset is of continuous or numeric attributes (Kaviani & Sunita, 2017;Chen et al, 2021).…”