User Workflow Preference Analysis Based on Confidence and Lift Value of Association Rule

Recommender systems have been widely used to improved customer experienced and to support personalized service to the consumer. In this paper, we proposed a method to obtain preferred service workflow which has the highest probability of going to be used by the customer to facilitate market analysis. The approach is by mining the association rules that correspond to the workflow of a universal service then performs a workflow reduction to obtain the preferred workflow. We first proposed a problem for determining preferred workflow. Then, we proposed a workflow reduction method based on association rules generation and rules filtering. Finally, we illustrated the approach of the proposed method with an example of a user workflow prediction.