Application Of LDWPSO Based Weighted Naive Bayes Algorithm In Logistics Demand Prediction

The State Council has made clear positioning of logistics industry as the foundation of support for the national economy and the strategic industry.In this context,as a super large city in the Yangtze River economic belt,Nanjing's economic development level,industrial structure,infrastructure,consumption level,consumption concept,logistics technology,urban logistics infrastructure layout and logistics service level are constantly improving,changing and improving.Therefore,it is necessary to establish a scientific prediction model for the logistics needs of Nanjing.Based on the actual situation of logistics development in Nanjing and related theories of logistics demand,this paper studies logistics demand prediction according to the related characteristics of logistics demand.The main research work has the following aspects:(1)On the basis of reading and studying the periodical literature of logistics demand at home and abroad,the gray system theory and exponential smoothing method are selected as the prediction models which are compared with the prediction models established in this paper.At the same time,on the basis of the factors that affect the logistics demand,the freight volume is selected as the forecast index of logistics demand.(2)Aiming at the problem of the independence of the attributes in the traditional naive Bayes model,this paper uses the method of adding weights for each attribute to weaken the independence of each attribute.Combining particle swarm optimization algorithm with strong fitting ability,a weighted naive Bayes prediction model based on particle swarm optimization is established for logistics demand prediction.At the same time,particle swarm optimization algorithm is easy to fall into the local optimal solution when dealing with discrete data.,this paper uses inertia weight linear decreasing method to solve this problem,so as to improve the robustness and accuracy of prediction model.(3)The empirical analysis of logistics demand prediction in Nanjing is carried out.The experimental results show that the prediction error rate of the weighted naive Bias prediction model based on LDWPSO is significantly lower than that of the other single prediction models,which verifies the accuracy of the prediction model.