Who Will You Share a Ride With: Factors that Influence Trust of Potential Rideshare Partners

TL;DR

This study explores how social media influences trust in peer-to-peer ridesharing, identifying key factors like social proof, approval, and self-disclosure that significantly impact trust decisions and improve prediction models.

Contribution

It introduces a novel approach of using social media profiles to assess trust in ridesharing, highlighting three main factors that enhance trust prediction accuracy.

Findings

Social proof, social approval, and self-disclosure are key trust factors.

Trust prediction model achieved 87.4% F-measure, outperforming previous methods.

Social media data significantly improves trust assessment in ridesharing.

Abstract

Ridesharing can save energy, reduce pollution, and alleviate congestion. Ridesharing systems allow peer-to-peer ride matching through mobile-phone applications. In many systems people have little information about the driver and co-riders. Trust heavily influences the usage of a ridesharing system, however, work in this area is limited. In this paper, we investigate the role of social media in shaping trust and delineate factors that influence the decisions of trust placed on a driver or co-rider. We design a ridesharing system where a rider can see the Instagram profile and 10 recent pictures of a co-rider and ask them to make judgments on six trust related questions. We discovered three important factors that shape trust forming: \textit{social proof}, \textit{social approval}, and \textit{self-disclosure}. These three factors account for 40\%, 33\%, and 13\% of the variance, and subsequent factors extracted accounted for substantially less variance. Using these three factors, we successfully build a trust prediction model with F-measure of 87.4\%, an improvement from 78.6\% on using all factors collected.