Cookies
O website necessita de alguns cookies e outros recursos semelhantes para funcionar. Caso o permita, o INESC TEC irá utilizar cookies para recolher dados sobre as suas visitas, contribuindo, assim, para estatísticas agregadas que permitem melhorar o nosso serviço. Ver mais
Aceitar Rejeitar
  • Menu
Tópicos
de interesse
Detalhes

Detalhes

  • Nome

    Joel Ribeiro
  • Cargo

    Investigador Colaborador Externo
  • Desde

    01 dezembro 2018
003
Publicações

2023

Leveraging Social Media as a Source of Mobility Intelligence: An NLP-Based Approach

Autores
Fontes, T; Murcos, F; Carneiro, E; Ribeiro, J; Rossetti, RJF;

Publicação
IEEE OPEN JOURNAL OF INTELLIGENT TRANSPORTATION SYSTEMS

Abstract
This work presents a deep learning framework for analyzing urban mobility by extracting knowledge from messages collected from Twitter. The framework, which is designed to handle large-scale data and adapt automatically to new contexts, comprises three main modules: data collection and system configuration, data analytics, and aggregation and visualization. The text data is pre-processed using NLP techniques to remove informal words, slang, and misspellings. A pre-trained, unsupervised word embedding model, BERT, is used to classify travel-related tweets using a unigram approach with three dictionaries of travel-related target words: small, medium, and big. Public opinion is evaluated using VADER to classify travel-related tweets according to their sentiments. The mobility of three major cities was assessed: London, Melbourne, and New York. The framework demonstrates consistently high average performance, with a Precision of 0.80 for text classification and 0.77 for sentiment analysis. The framework can aggregate sparse information from social media and provide updated information in near real-time with high spatial resolution, enabling easy identification of traffic-related events. The framework is helpful for transportation decision-makers in operational control, tactical-strategic planning, and policy evaluation. For example, it can be used to improve the management of resources during traffic congestion or emergencies.

2022

Real-Time Detection of Vehicle-Based Logistics Operations

Autores
Ribeiro, J; Tavares, J; Fontes, T;

Publicação
INTELLIGENT TRANSPORT SYSTEMS (INTSYS 2021)

Abstract
Geolocation data is fundamental to businesses relying on vehicles such as logistics and transportation. With the advance of the technology, collecting geolocation data become increasingly accessible and affordable, which raised new opportunities for business intelligence. This paper addresses the application of geolocation data for monitoring logistics processes, namely for detecting vehicle-based operations in real time. A stream of geolocation entries is used for inferring stationary events. Data from an international logistics company is used as a case study, in which operations of loading/unloading of goods are not only identified but also quantified. The results of the case study demonstrate the effectiveness of the solution, showing that logistics operations can be inferred from geolocation data. Further meaningful information may be extracted from these inferred operations using process mining techniques.

2022

Detection of vehicle-based operations from geolocation data

Autores
Tavares, J; Ribeiro, J; Fontes, T;

Publicação
Transportation Research Procedia

Abstract
Geolocation data identifies the geographic location of people or objects, which may unveil the performance of some activity or operation. A good example is, if a vehicle is in a gas station then one may assume that the vehicle is being refuelled. This work aims to obtain vehicle-based operations from geolocation data by analysing the stationary states of vehicles, which may identify some motionless event (e.g. bus line stops and traffic incidents). Ultimately, these operations may be analysed with Process Mining techniques in order to discover the most significant ones and extract process related information. In this work, we studied the application of diverse approaches for detecting vehicle-based operations and identified different operations related to the bus services. The operations were also characterized according the distribution of their events, allowing to identify specific operations characteristics. The public transport network of Rio de Janeiro is used as a case study, which is supported by a real-time data stream of buses geolocations.

2020

Process discovery on geolocation data

Autores
Ribeiro, J; Fontes, T; Soares, C; Borges, JL;

Publicação
Transportation Research Procedia

Abstract
Fleet tracking technology collects real-time information about geolocation of vehicles as well as driving-related data. This information is typically used for location monitoring as well as for analysis of routes, vehicles and drivers. From an operational point of view, the geolocation simply identifies the state of a vehicle in terms of positioning and navigation. From a management point of view, the geolocation may be used to infer the state of a vehicle in terms of process (e.g., driving, fueling, maintenance, or lunch break). Meaningful information may be extracted from these inferred states using process mining. An innovative methodology for inferring process states from geolocation data is proposed in this paper. Also, it is presented the potential of applying process mining techniques on geolocation data for process discovery. © 2020 The Authors. Published by Elsevier B.V.

2020

A Deep Learning Approach for Predicting Bus Passenger Demand Based on Weather Conditions

Autores
Fontes, T; Correia, R; Ribeiro, J; Borges, JL;

Publicação
Transport and Telecommunication

Abstract
This work apply a deep learning artificial neural network model-the Multilayer Perceptron- A s a regression model to estimate the demand of bus passengers. Transit bus ridership and weather conditions were collected over a year from a medium-size European metropolitan area and linked under the assumption: Individuals choose the travel mode based on the weather conditions that are observed during (a) the departure hour, (b) the hour before or (c) two hours prior to the travel start. The transit ridership data were also labelled according to the hour of the day, day of the week, month, and whether there was a strike and/or holiday or not. The results show that the prediction error of the model decrease by ~9% when the weather conditions observed two hours before travel start is taken into account. The model sensitivity analyses reveals that the worst performance is obtained for a strike day of a weekday in spring (typically Wednesdays or Thursdays). © 2020 Tânia Fontes et al., published by Sciendo.