Abstract
Hitchhiking is a well-known form of transportation, which has been extensively studied from a sociological perspective. While this tourism approach has been popular in some countries for decades, the question of making hitchhiking fast has never been studied from a comprehensive point of view. In this research, we aim to find the best locations for hitchhiking, which may be used in travel recommender systems. For this purpose, we study the relationships between certain spatial properties and waiting time at pick-up locations, estimate statistical significance of several factors according to various countries. We extracted features for the data from hitchhiking logs, analyzed their importance and built corresponding classification and regression models for estimating hitchhiking experience with a sufficient accuracy. This paper is the first to analyze important spatial factors for a successful hitchhiking trip and construct models to predict the quality and waiting time at hitchhiking locations.
Highlights
Hitchhiking is a well-known form of transportation, which has been extensively studied from a sociological perspective
We investigate the waiting time and rating of a given location based on several spatial factors taken from hitchhikers’ descriptions
We have developed a feature set for classifying good and bad ratings of locations based on location features as well as a regression model to predict exact ratings
Summary
Motivation Hitchhiking is the first well-known form of ridesharing that became popular in 20th century in the US, and wide spread all around the world. Certain location properties make it easier for drivers to stop and have enough traffic to increase the probability of a successful ride for a hitchhiker. Gas stations are popular hitchhiking points on motorways, while bus stops may be significant pick-up points on regional roads. The results of this paper describe certain properties needed for a successful hitchhiking location, show how hitchhiking experience is different for various roads, and how users’ comments can improve the accuracy of a recommender system. These results are highly important to assist hitchhikers in helping them to plan an optimal trip and can assist in developing a trip planning application. “Conclusions” section provides conclusions and future directions of hitchhiking-related researches
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
