Bike sharing
Many cities provide the bike sharing service. Some of them made the bike trip data open. For example:
| New York | Citibike | data |
| Chicago | Divvy | data |
| Boston | Bluebikes | data |
| Pittsburgh | HealthyRide | data |
| Philadelphia | Indego | data |
| Washington, DC | Capital | data |
| San Francisco | Bay Wheels | data |
Each trip is anonymized and includes:
These data can be aggregated into frequency distributions in different ways. A trip t adds 1 to a bin iff (if and only if) [t.start,t.end) ∩ bin ≠ ∅. For example
DURA: A trip duration (= t.end - t.start) distribution for a selected year (bin size = 5 min) for each usertype and total.
WEEK: A weekly activity for a selected year for each usertype and total. The units U are the weeks, bins are days of the week (Mo, Tu, We, Th, Fr, Sa, Su).
DAY: A daily activity for a selected year for each usertype and total. The units U are days in a year, bins are half hours of a day.
ISTA: A station daily arrival activity for a selected year for each usertype and total. The units U are stations, bins are half hours of a day. A trip t adds 1 to a bin iff t.end ∈ bin.
OSTA: A station daily departure activity for a selected year for each usertype and total. The units U are stations, bins are half hours of a day. A trip t adds 1 to a bin iff t.start ∈ bin(station,usertype).
For each student number s the TYPE and service are determined by the following table
| Citibike | Divvy | Bluebikes | HealthyRide | Indego | Capital | Bay Wheels | |
|---|---|---|---|---|---|---|---|
| WEEK | 1 | 4 | 7 | 10 | 13 | 16 | 19 |
| DAY | 2 | 5 | 8 | 11 | 14 | 17 | 20 |
| ISTA | 3 | 6 | 9 | 12 | 15 | 18 | 21 |
| OSTA | 22 | 23 | 24 | 25 | 26 | 27 | 28 |
For example, to the student s=5 correspond TYPE=DAY and service=Divvy.
For the year 2019 and “your” service construct the aggregated descriptions for types DURA and “your” TYPE.
Each distribution is a vector. To get a unit description join the corresponding vectors into a named list. To get a data set description join the unit descriptions into a named list. Save both data set descriptions in JSON format.
Visualize/analyze the created data sets. Report your observations.