Association Rule Mining for Precision Marketing of Power Companies with User Features Extraction

To help power companies quickly and precisely find target consumers for different purposes from resident users, a method based on association rule mining is proposed. First, users’ four electricity features including daily electricity consumption (DEC), electricity consumption stability (ECS), potential of peak load shifting (PPLS) and typical electricity consumption patterns (TECP) and basic attributes (population, employment status, cooking type, etc.) are extracted. Then, the association rules between basic attributes and electricity features are obtained by FP-growth algorithm, which can guide power companies to select the target users. Finally, using a sample to evaluate obtained rules, the results demonstrate that the proposed method is practical and effective.