Marcelo Tílio Monteiro de Carvalho

A methodology for traffic-related Twitter messages interpretation

HighlightsThis paper focuses on the problem of interpreting tweets that describe traffic-related events.We introduce a traffic event domain ontology, called TEDO, that models traffic-related events.We describe a new tool to automatically interpret traffic-related tweets.This tool translates each tweet into a set of RDF triples structured according to TEDO. This paper addresses the problem of interpreting tweets that describe traffic-related events and that are distributed by government agencies in charge of road networks or by news agencies. Processing such tweets is of interest for two reasons. First, albeit phrased in natural language, such tweets use a much more regular and well-behaved prose than generic user-generated tweets. This characteristic facilitates automating their interpretation and achieving high precision and recall. Second, government agencies and news agencies use Twitter channels to distribute real-time traffic conditions and to alert drivers about planned changes on the road network and about future events that may affect traffic conditions. Hence, such tweets provide exactly the kind of information that proactive truck fleet monitoring and similar applications require. The main contribution of the paper is an automatic tweet interpretation tool, based on Machine Learning techniques, that achieves good performance for traffic-related tweets distributed by traffic authorities and news agencies. The paper also covers in detail experiments with real traffic-related tweets to access the precision and recall of the tool.

A Proactive Application to Monitor Truck Fleets

Positioning systems, combined with inexpensive communication technologies, open interesting possibilities to implement real-time applications that monitor vehicles and support decision making. This paper first discusses basic requirements for proactive real-time monitoring applications. Then, it describes how to structure and geo-reference unstructured text information available on the Internet, with a focus on road conditions change and using available geocoding services. Lastly, the paper outlines an application that monitors a fleet of trucks and incorporates proactive features.

Travel time prediction using machine learning

This paper investigates the application of a Machine Learning technique to predict the time that will be spent by a vehicle between any two points in an approximated area. The prediction is based on a learning process based on historical data about the movements performed by the vehicles taking into account a set of semantic variables to get estimated time accurately. The paper also describes an experiment with real-world data. Although this is preliminary work, the results were satisfactory.

The TRANSPETRO Emergency Response System

This paper describes the TRANSPETRO Emergency Response System, developed by PETROBRAS / TRANSPETRO, which is applicable to oils terminals, mostly located in sea or river harbor areas, and a large oil pipeline mesh that spans the Brazilian territory. The engine of the system consists of a distributed workflow management software, called InfoPAE. The software runs locally at each major installation, all connected to a central site. The deployment of the TRANSPETRO Emergency Response System is part of a larger effort to achieve a standard of excellence in the response to emergency situations at each operational unity, including full compliance with environmental regulations set by the federal government.

SAAAP: An Automated Pipeline Routing System

SAAAP — Routing Alternatives Environmental Evaluation System — was designed to select the best alternative route, within an area of interest, to implement a new pipeline project. The system takes into account economic, environmental and engineering factors, according to an optimality criterion that combines several variables, such as vegetation coverage, soil type and declivity. The system is operational and has been tested in several realistic projects.Copyright