Study On The Methods And Applications For Product Sales Forecasting Based On Historical Data And Online Reviews

With the rapid development of the market economy and the challenges of econmy globalization,the firms facing the cruel and fierce market competition.Sales forecasting has always been the focus of enterprise managers.Firms use product sales forecasting as a foundation to estimate sales revenue and make decisions regarding production,operation and marketing strategies.Through product sales forecasting,firms can improve their economic benefits and reduce losses caused by weaknesses in the production plan.Advancements in Internet technologies have provided new purchasing platforms for the consumers,and consumers are paying more attention on online reviews.A number of researchers have studied the product sales forecasting problem and developed effective forecasting models that take relevant factors into account.However,few studies have developed improved versions of the forecasting model to consider the sentiments expressed in the content of online reviews.Therefore,the framework and method of sales forecasting need being explore based on historical data and online reviews.On the one hand,the sales forecasting method can fill the gaps in the existing research,it has vital significance on enriching and improving the sales forecasting method.On the other hand,enhancing the accuracy of sales forecasting can improve the firm cognition to the market and economic benefits.A series of research works are conducted as follows.(1)Proposed the method of sales forecasting for one generation product based on online reviews analysis and Bass model.Analyzed the influence of each online reviews index on sales forecasting and considering the online reviews indexes are sufficient or not,this paper proposed the sales forecasting model based on sentiment index and the sales forecasting model based on several online reviews indexes.(2)Proposed the method of sales forecasting for generations product based on online reviews analysis and Norton model.Analyzed the influence of online reviews each index on sales forecasting for generations product and considering the online reviews indexes are sufficient or not,this paper proposed the generations product sales forecasting model based on sentiment index and the sales forecasting model based on several online reviews indexes.(3)Showed the application of the product sales forecasting model based on online reviews and historical data.To verify the the practicability and rationality of the method this paper prososed,this paper take Beijing Hyundai product sales forecasting as an example.The Experimental results show that the online reviews have an effect on product sales.The sales forecasting method using online reviews and historical data can be more accurate to forecast product sales.