The　mining　of　IC　card　data　for　commuters＇　classification　and　analysis　of　bus　route　choice　plays　an　important　role　for　public　transport　passengers＇　choice　behavior　analysis,　and　formulating　a　scientific　and　reasonable　traffic　planning　strategy.　The　authors　choose　some　indicators　for　passengers＇　classification　in　Beijing　and　use　classification　and　regression　tree　model　to　build　the　classifier.　Three　parameters:　departure　time,　travel　distance　and　travel　frequency,　as　the　classification　parameters,　are　input　into　the　model　and　the　results　show　45　classification　passengers;　Among　them,　strong　commuting　passengers　accounted　for　7.25%,　weak　commuting　passengers　accounted　for　30.27%,　and　accidental　commuting　passengers　accounted　for　62.48.　This　paper　also　analyzes　the　characteristics　of　the　public　transport　route　choice　for　all　classification　groups　from　four　aspects:　the　overall　characteristics,　transfer　characteristics,　travel　characteristics　and　subway　travel　characteristics.　The　conclusions　of　this　paper　can　provide　a　theoretical　basis　for　the　analysis　of　passenger　flow　composition　of　multimodel　public　transport　and　the　establishment　of　multi-mode　bus　route　selection　model.　©　Published　under　licence　by　IOP　Publishing　Ltd.