PROJECT TITLE :

Product Adoption Rate Prediction in a Competitive Market - 2018

ABSTRACT:

Because the worlds of commerce and therefore the Internet technology become additional inextricably linked, a giant variety of user consumption series become obtainable for on-line market intelligence analysis. A critical demand along this line is to predict the longer term product adoption state of each user, that enables a wide range of applications like targeted marketing. Nevertheless, previous works solely aimed at predicting if a user would adopt a specific product or not with a binary get-or-not representation. The problem of tracking and predicting users' adoption rates, i.e., the frequency and regularity of using each product over time, continues to be below-explored. To this finish, we have a tendency to present a comprehensive study of product adoption rate prediction in a competitive market. This task is nontrivial as there are 3 major challenges in modeling users' advanced adoption states: the heterogeneous data sources around users, the unique user preference and the competitive product selection. To cater to these challenges, we 1st introduce a versatile issue-based call operate to capture the modification of users' product adoption rate over time, where various factors that will influence users' choices from heterogeneous data sources will be leveraged. Using this factor-based mostly call perform, we have a tendency to then provide two corresponding models to find out the parameters of the choice operate with each generalized and personalized assumptions of users' preferences. We have a tendency to further study the way to leverage the competition among different product and simultaneously learn product competition and users' preferences with each generalized and personalised assumptions. Finally, in depth experiments on 2 real-world datasets show the superiority of our proposed models.