Page 135 - Proceeding of Atrans Young Researcher's Forum 2019_Neat

“Transportation for A Better Life:
Smart Mobility for Now and Then”

23 August 2019, Bangkok, Thailand

into two subsets: the training dataset (70% of data) variables (e.g. trips length, trip cost, on-vehicle
and the testing dataset (30% of data). duration, parking and walking duration, real parking
It is noteworthy noticed that among 15 inputs charge, proposed parking charge). To avoid
for travel users, there were 7 inputs (Table 1) being numerical errors, the currency of Vietnam (VND) in
considered as categorical variables (e.g. location, trip cost and real parking charge was changed to
gender, age, study, occupation, trip mode, trip x1000 unit (e.g. 3000 transformed to 3).
purpose), 2 inputs being transformed into categorical
variables (e.g. personal income, household income), Table 1. Socio-economic characteristics
the remaining 6 inputs were kept as continuous of interviewees

Variables Freq. (%) Preprocessing
Location X1
Office building 125 40.2 1
Shopping mall 110 35.4 2
Condominium 76 24.4 3
Gender X2
Male 171 55 1
Female 140 45 2
Age X3
Under 18 12 3.9 1
Age 18-24 86 27.7 2
Age 25-35 151 48.6 3
Age 36-50 44 14.1 4
Age 51-60 14 4.5 5
Above 60 4 1.3 6
Education X4
Uneducated 1 0.3 1
Primary-secondary school 4 1.3 2
High school 80 25.7 3
Vocational college 56 18.0 4
University 155 49.8 5
Post graduated or doctor 15 4.8 6
Occupation X5
Office worker/Government officer 148 47.6 1
Worker 12 3.9 2
Farmer 0 0 3
Small business/Self-employed 52 16.7 4
University student 36 11.6 5
Pupil (primary to high school) 12 3.9 6
Seasonal worker 22 7.1 7
Housewife/Retired/Jobless 16 5.1 8
Others 13 4.2 9
Trip Mode X8
Bicycle 27 8.7 1
Motorcycle 242 77.8 2
Car 24 7.7 3
Bus 15 4.8 4
Others 3 1.0 5
Trip Purpose X9
To work 183 58.8 1

110