metadata

license: odbl
task_categories:
  - tabular-regression
language:
  - en
  - fr
  - de
  - nl
tags:
  - mobility
  - hitchhiking
  - transport
pretty_name: Largest Dataset of Hitchhiking Rides
size_categories:
  - 10K<n<100K
dataset_info:
  features:
    - name: stops
      list:
        - name: arrival_time
          dtype: string
        - name: departure_time
          dtype: string
        - name: location
          struct:
            - name: is_exact
              dtype: bool
            - name: latitude
              dtype: float64
            - name: longitude
              dtype: float64
        - name: waiting_duration
          dtype: string
    - name: rating
      dtype: int64
    - name: hitchhikers
      list:
        - name: gender
          dtype: 'null'
        - name: hitchhiking_since
          dtype: 'null'
        - name: languages
          dtype: 'null'
        - name: nickname
          dtype: string
        - name: origin_country
          dtype: 'null'
        - name: origin_location
          dtype: 'null'
        - name: reasons_to_hitchhike
          dtype: 'null'
        - name: was_driver
          dtype: 'null'
        - name: year_of_birth
          dtype: 'null'
    - name: comment
      dtype: string
    - name: signals
      list:
        - name: asking_content
          dtype: 'null'
        - name: asking_languages
          dtype: 'null'
        - name: duration
          dtype: 'null'
        - name: methods
          sequence: string
        - name: sign_content
          dtype: 'null'
        - name: sign_languages
          dtype: 'null'
        - name: total_solicited
          dtype: 'null'
    - name: occupants
      dtype: 'null'
    - name: mode_of_transportation
      dtype: 'null'
    - name: ride
      dtype: 'null'
    - name: declined_rides
      dtype: 'null'
    - name: source
      dtype: string
    - name: license
      dtype: string
    - name: submission_time
      dtype: string
  splits:
    - name: train
      num_bytes: 14602124
      num_examples: 63291
  download_size: 6888947
  dataset_size: 14602124
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

The Hitchhiking Rides Dataset

Here the dataset described in Hitchhiking Rides Dataset: Two decades of crowd-sourced records on stochastic traveling is published.

This is the largest dataset of crowd-sourced real-world hitchhiking rides. Data has been collected on online platforms such das liftershalte.info, hitchwiki.org and hitchmap.com starting in 2005 until today.

Dataset Details

Dataset Description

Curated by: Till Wenke
License: Open Data Commons Open Database License (ODbL)

Credit

parts of the data stem from hitchwiki.org and were published under CC BY-SA 4.0, see their home page
parts of the data stem from hitchmap.com and were published under ODbL, see their copyright notice

Dataset Sources

Sources are given for each entry in the respective dataset column.

Data was fetched from the different sources and published here running this scrip.

Uses

To inform hitchhikers to make more data-based decisions on the road.

General research towards hitchhiking as a cultural phenomenon.

Data Science Uses

Waiting time predition - regression task e.g. in Heatchmap: A Gaussian process approach to predict hitchhiking waiting times

Dataset Structure

The structure that this dataset follows is described in the Hitchhiking Data Standard.

Dataset Creation & Bias, Risks, and Limitations

A dataset paper was published on arXiv to address the above questions.

Citation

BibTeX:

[More Information Needed]

Dataset Card Authors

Till Wenke