Qualidade de experiência em serviços de mobilidade compartilhada: proposta de um sistema de recomendação para usuários

Abstract

In recent years, new models of urban transport system gained popularity for their advantages and lower costs for users. One of the new service models that has been highlighted is the Ridesharing service, which allows shared travel and can reduce the number of vehicles in the streets and contributes for urban mobility. The evolution and dissemination of smartphones allowed the emergence of several applications (APP), developing innovative solutions for mobility, and contributing to the rapid growth of the number of users. Quality systems have been present in the automotive industry for decades, ensuring customers’ satisfaction with their products. The user’s quality perception in mobility services, has become a key factor in transport solutions. In this way, the Quality of Experience (QoE) can be a new tangible evaluation parameter for the new models of sharing transport. Based on this scenario, the present work proposes to use the QoE concept in Ridesharing services. A recommendation model for shared mobility services is proposed, considering the profile information, extracted from the social network sites, and user preferences. The main objective of the proposed recommendation model is to improve the user’s QoE. For this, subjective tests are conducted, creating a database capable of providing the conditions for elaborating the classification model with application of machine learning algorithms. The results of these algorithms compose the recommendation model, this in turn can identify users with similar preferences through a similarity function, thus users with similar preferences and characteristic share a ride. An experimental results showed that Random Forest algorithm achieved the best performance, reaching the F-Meansure of 0,92. In addition, the results showed that 94.2% of participants agree with the recommendation model results.