| In the background of big data era,the development of the business models has undergone significant changes.Many emerging companies have emerged with new business models.But most of them have quickly gone bankrupt or have closed down in the long-term competition.Therefore,how to quickly and accurately predict the future trend of a new company or product is of great significance.In this paper,we try to prove that the Lotka-Volterra model outperforms most models in predicting sales or consumer groups.However,a common feature of these emerging companies is their over-reliance on the Internet.Therefore,we start with the number of Chinese Internet users who is the most easily accessible and analyzed and has a small amount of data to provide a certain research basis and appropriate research ideas for e-commerce in the era of big data.However,the growth trend of the number of Internet users can be described by the diffusion model.Thus,it is very important to find a diffusion model that can most accurately describe the growth trend of the number of Chinese Internet users.In this paper,we respectively describe four classical diffusion models and their parameter estimation methods,that is,the Gompertz model,the Logistic model,the Bass model,and the Lotka-Volterra model.And we use the four diffusion models and the data of Chinese Internet users during 2007 to 2014 to fit the models and estimate the parameters.Then,we compare the goodness-of-fit and prediction accuracy criteria(ie,mean absolute percent error(MAPE),mean absolute error(MAE)and root mean squared error(RMSE))to compare the fitting effect of the four models.Then we calculate the predicted value for 2015-2017,and use MAPE to compare the prediction accuracy of the four models.We find that the Lotka-Volterra model has the highest goodness of fit and prediction accuracy in fitting and predicting the number of Chinese Internet users.This shows that the diffusion trajectory of the Lotka-Volterra model is more in line with the growing trend of the number of Chinese Internet users.We also use another feature of the Lotka-Volterra model that the Lotka-Volterra model can be used to analyze the competitive relationship between two populations to investigate the relationship between the rural Internet users and the urban Internet users in China.And analyzing the reasons for this kind of result from the relation we have obtained,with the purpose of providing decision support for the government or the enterprise.The experimental result shows that the relationship between rural netizens and urban netizens in China is commensalism.That is to say,the penetration of rural Internet users is not influenced by the fluctuating penetration of urban Internet users,but the penetration of urban Internet users will increase as the penetration of rural Internet users increases.The reason for this kind of relation with the irreversible urbanization process of China during the past two decades.Therefore,we believe that promoting the growth of the number of rural Internet users will help promoting the growth of the overall number of Internet users.In terms of increasing the penetration rate of rural Internet,the government or enterprises can focus on mobile smart phones,and dig out mobile Internet service scenarios to promote the popularity of rural Internet in the future.In addition,in order to further verify the advantages of the Lotka-Volterra model in predicting consumer groups and analyzing the characteristics of competing relationships among different consumer groups,we conduct behavioral research of Internet users.Similar to the method of analyzing the number of Internet users in China,four models are used to fit and predict the scale of online game users,and the competition relationship between instant messaging and online news is also analyzed.The experimental results are consistent with our expected results.In the end,we have summarized the full text and made some prospects for future research directions. |
PDF Full Text Request