Multiple User Characteristic Models for Online Survey Based on FP-Tree Algorithm

Online survey recently has received many attractions and become an important way for enterprises to accumulate data and facilitate their business development. Under the analysis of user attributes, user behavior and survey attributes in the domain of online survey, this paper proposes five user characteristic models for online survey based on improved FP-Tree algorithm. First, by analyzing the attributes of user and online survey, the users are divided into different categories, using our proposed improved FP-Tree algorithm to mine frequent items of user characteristics. By doing so, we then discover association rules between the user and the survey. Based on generated association rules, multiple user characteristic models are built to support enterprise for online surveys. Experimental results show that the improved FP-Tree algorithm can significantly benefit the performance compared with the traditional algorithm. By the analysis of different user characteristics models, it is concluded that there are obvious characteristics of online survey user and strong association rules between user attributes and the survey type.