Complex Event Processing (CEP) is proposed in this paper to tackle the changeability, tense correlation, and association in massive data process. This paper takes RFID (Radio Frequency Identification) data for example to illustrate how CEP works on dealing with massive data processing. First, semantic expressions are introduced as the reason of modeling for CEP. Then, model solution based on semantic expressions is proposed. Finally, this methodology achieves good results in processing massive RFID data in terms of speed, efficiency and veracity. The experiment results demonstrate that it is better in getting rid of complexity data after comparing with traditional data management based on database with massive data.