A Novel Automatic Classification Method for Flash Based on BP Neural Network

With the rapid development of Internet, the network has penetrated into many areas of people´s lives. It has become an indispensable communication tool, information access tool and entertainment tool for people. Flash as a new multimedia has become an important part on the Internet. Because the advantages of strong artistic expression, simplified manufacture, flexible interaction, small storage and easy communication, it has been widely used on the Internet. Faced with the growing number of Flash, how to retrieve Flash accurately and efficiently is becoming a noticeable problem. Based on the study of file structures, forming principles and content features, a novel automatic classification method for Flash is proposed in this paper. The algorithm of BP neural network algorithm is adopted in this method. It is called BPFAC which can realize the automatic classification of Flash, and can make retrieval system for Flash more efficient and accurate. On the basis of the optimizing the efficiency of algorithm, we have developed the prototype of BPFAC by C++. Based on two experimental parameters, the Accuracy Rate and the Time Consumption, several comparative experiments have been carried out respectively. Results show that BPFAC has higher accuracy in classification of Flash, and has relatively stable time consumption at the same time.