Research Article
Alighting Stop Determination of Unlinked Trips Based on a Two-Layer Stacking Framework
Table 6
Indicators of smart card data for clustering.
| Type | Index | Indicators | Quantification |
| Base indicators | 1 | ID | —— | 2 | Type | —— | 3 | Date | Making data into numbers, such as using 20181101 rather than November 1, 2018 | 4 | Transaction time | Changing transaction time to numbers in seconds with 00 : 00 : 00 a day as a reference, such as using 28800 rather than 8 : 00 : 00 | 5 | Boarding stop index | —— | Constructed indicators | 6 | Weekday | =0 : The boarding day is not a weekday | =1 : The boarding day is a weekday | 7 | Weather | =0 : It rains on the boarding day | =1 : It doesn’t rain on the boarding day | 8 | Peak hours | =0 : Transaction time is not in peak hours | =1 : Transaction time is in peak hours | 9 | Type of boarding stop | =1 : The boarding stop is the first-class passenger flow center | =2 : The boarding stop is the second-class passenger flow center | =3 : The boarding stop is the third-class passenger flow center | 10 | Administrative region of boarding stop | =1 : Boarding stop is in administrative region 1 | =2 : Boarding stop is in administrative region 2 | The rest can be done in the same manner | 11 | Loyalty of passenger | =1 : Bus is an occasional selection | =2 : Bus is an alternative selection | =3 : Bus is a loyal selection |
|
|