The invention provides a two-stage passenger crowd travel purpose analysis method integrating bus data and POI data, aims to solve the problem of identifying bus passenger crowd travel purposes by combining the bus data and the POI data, and provides a novel, reasonable and effective passenger crowd travel purpose analysis method. The method mainly comprises two stages, in the first stage, firstly, the bus data and the POI data are extracted, cleaned and converted, station information is converted into POI vectors, and finally, a clustering method is adopted to analyze the travel purpose of bus passengers; in the second stage, firstly, card swiping data and intelligent card data are combined to extract the demographic information of passengers, then the riding distance of the passengers iscalculated, the travel purpose of the first stage is used as the characteristics of passenger group clustering, and finally, a clustering algorithm is used for clustering passengers, and the travel purpose in each cluster is analyzed, so that the effect of analyzing different travel purposes of different riding crowds is achieved. The method is used for analyzing the travel purpose of the publictransport passenger crowd.