Multi scale passenger frequent pattern mining method based on single mode data

摘要

In order to overcome the problems of low accuracy and efficiency in traditional passenger frequent pattern mining methods, a multi-scale passenger frequent pattern mining method based on single mode data is proposed. Set up a passenger travel data acquisition framework to obtain the travel displacement and time data of passengers, and denoise the relevant data. Based on the denoised passenger data, the frequent pattern mining database is constructed, and the hash index process is set up. Finally, the travel characteristics are extracted, and the PrefixSpan algorithm is used to complete the passenger frequent pattern mining. Experimental results show that, compared with traditional mining methods, the proposed mining method has higher mining accuracy and efficiency, and has higher practical application value.