Effective User Invitation Models for Online Survey Using Clustering Algorithm

With the continuous and rapid development of online questionnaire survey, the low response rate has plagued operating companies. To solve this problem, this paper proposed an effective user invitation model by our improved clustering algorithm, which analyzed large-scale historical user behavior characteristic data, including users´ quality data, users´ preferential data and users´ similarity data. Extensive experiments with large-scale data from an online survey company have been conducted to validate the feasibility and effectiveness of our proposed approach. Experimental results demonstrate that the questionnaire response rate is increased and our approach can be easily deployed in real-world online survey application for effective personalized survey recommendation.