A Recommendation System for Shared-Use Mobility Service through Data Extracted from Online Social Networks

Abstract

In recent years, the shared mobility service has increased in many countries across the world because its low cost and several shared-use mobility applications on mobile devices [1]. Commonly, if a ride is shared between people with similar preferences, users likely feel both more comfortable and safe. In this context, the main goal of this article is to classify users with similar preferences, in automatic manner, to improve user's quality of experience in ridesharing service. To obtain initial data, subjective tests are carried out using questionnaires and their results are used to determine ridesharing profiles. Then, some basic user profile information is extracted from Online Social Networks (OSN) to determine an user profile based on preferences in ridesharing service. The user profile classification is performed through different machine learning algorithms, which use as input the data extracted from OSN. Two case studies of shared-mobility are treated, (i) sharing a ride with a passenger with a similar hobby [2], and (ii) sharing a ride with people that support an opposite football teams. The novelty of this work consists in the application of the Hybrid Discriminative Restricted Boltzmann Machines (HDRBM) algorithm for classification whose performance overperforms widely used algorithms such as Random Forest, SVM and DRBM. The experimental results presented a correctly classified instance of 96.9% and 97.3% for the cases of sharing a ride with people with similar hobby and support different football team, respectively. Finally, a Recommendation System (RS) is proposed, which efficiency is compared with a basic RS, obtaining a Pearson correlation coefficient of 0.97 and 0.71, respectively. Index Terms-Recommendation system; ridesharing; mobile applications, online social networks, machine learning, HDRBM, social web analysis tool. I. INTRODUCTION T HE mobile devices and applications have facilitated the manner of people request services. The urban mobility is a kind of service that became very popular, in it is common that people share the rides among others. The advantages of this service are related to benefits for urban traffic, such as reduction of environment pollution, and low cost.

A Systematic Literature Review on Machine Learning in Shared Mobility

“…The existing literature advocating the use of ML techniques for SMS providers spans various sharing schemes, including dockless systems [82,83,85,86] and hybrid models encompassing both docked and dockless systems [44,84]. These studies encompass a range of SMS types, from ride-hailing and car-sharing to bike and e-scooter sharing.…”

“…From our review, methodologies like sentiment analysis, topic modeling, and classification algorithms have been em- Scooter Dockless Survey Generalized Linear Models [86] ployed to dissect data from diverse sources, including social media, surveys, and online news platforms. For instance, [83] introduces a framework to profile ride-sharing users, aiming to enhance service quality by matching users with similar interests. [82] delves into user sentiments about ridehailing services, pinpointing areas like customer satisfaction.…”

P2PCF: A collaborative filtering based recommender system for peer to peer social networks

“…The new definition covers systems recommending any objects in social media domains such as items, tags, links, people, and communities [17,57]. It has been shown that social considerations improve recommender systems since there is a significant overlap between users' interest and connections [48,56,61].…”

Novel m-PSO-SVM based Interface Controller Design for Haptic System

“…Velocity and position of each particle are updated with population law locally in pbest and globally gbest. Velocity and position update formulas are given in (19) and (20), respectively [28]. Acceleration factors wc Inertia weight factor rand(1) Random number in between 0 to 1 rand(2) Random number in between 0 to 1 pbesti,j i th particle best position at iteration k (local) gbestj swarm particle best position until iteration k (Global)…”

“…But it has been found that NN is prone to local convergence, slow learner, and overfitting problems in some applications [16]. Recently, support vector machine (SVM) has emerged as a tool to overcome the drawbacks of NN [11], [18], [19]. The SVM has three independent factors that affect its performance significantly.…”