Based On The Rf Update Of Phtp Demand Influencing Factors Analysis

With the emerging of hi-tech products, renewal and substitution of product appears frequently and renewal between products with new and old technique becomes more and more frequently. Probing into the influencing factors of demand of perishable hi-tech product in renewal period can not only provides pricing and decreasing price strategy for product in the renewal period but also provides principles of the direction of product renewal and innovation. This paper comes from program of National Nature Science Foundation of China. It continues the related researches of perishable hi-tech product of the previous scholars. And with the combination of the features of perishable hi-tech product, it uses Random Forests classifier combination and takes the notebook PC as an example to analyze the influencing factors of the demand of perishable hi-tech product.At first, combing the features of perishable hi-tech product, this paper makes qualitative analyses to the influencing factors of demand. There appears a lot of influencing factors of the product including factors of the product itself and allocation of the renewal product and competitive product and the potential influence of the appearance and price. In other words, the choice collection of the consumers stays not only at the various products in the renewal period but the property of these products. In addition, the environmental factors especially the competitive surrounding is also the factor needed to be considered carefully.Then, considering that there is abundant demand information in degree of attention, this paper makes the degree of attention the target variables influencing the demand. Degree of attention is the accumulative value of click rate and related statement to notebooks of various types of the consumers on the internet. It contains not only the real demand information that is who clicks and buys the product but also the potential demand information that is who clicks but doesn't buy the product. It puts forward the features the consumers on the internet notice the data that is complexity, heterogeneity, high dimension and great amount.At last, in view of the features of the consumer on the internet notice the data, this paper uses random forest classifier combination and takes note book as an example to analyze the influencing factors of demand and gets the contribution rate of every factor to degree of attention. Through comparison with Neural Network method and decision tree method and after analysis, it proves that Random Forests classifier can get the optimal result.This paper puts forward the features of the consumers on the internet notice the data of perishable hi-tech products firstly, and then with the combination of the feature of the data, it introduces the Random Forests classifier combination firstly. Through theories and comparative analysis, it proves that Random Forests classifier combination is the most effective way to analyze the features of consumers on the internet notice the data of perishable hi-tech products, which paves the ground for quantitative analysis of the demand of perishable hi-tech products...