No abstract

No abstract

Around the world, cities suffer from a large variety of problems. One of them is urban mobility, and the most common cause is related to traffic incidents, which unexpectedly provoke delays to people and produce losses to businesses. Through natural-language-processing methods, this chapter proposes a way to inform people about events that could have happened on the road. The proposed application takes information from news reports (published by a local newspaper) that have been previously classified as 'traffic incidents'; the app tries to extract the location where these events occurred. These data are then served on a web application, which shows a map that marks all the recent incidents. In this way, the authors offer an alternative to allow citizens to be informed about this kind of event so they can take preventive actions.

The primary objective of this chapter is to analyze the content retrieved from an RSS newspaper report written in the Spanish language and determine if it describes a traffic incident. The real-world case study occurs in a Mexican city with 1.5 million inhabitants, and its purpose is to offer an application to inform in real-time about traffic incidents that happen in the town, gathering its information from the reports published by a local newspaper. However, to do so, the authors first need to classify the articles and ignore those whose content does not imply any kind of road accident.