ASFINAG Drivers Dashboard

The Driver‘ Dashboard was developed for the Austrian Motorway Operator ASFINAG, assisting them in identifying the root causes of traffic jams and delays on their motorway network. Social media posts containing traffic relevant information are linked to the traffic telemetry data to identify the traffic events described by both in different ways. The traffic telemetry data is provided by ASFINAG and the social media text is gathered from the Facebook page of ASFINAG, RSS feed, and the RSS feed. To establish the link between the telemetry data and the social media texts, the texts are annotated by means of Natural Language Processing (NLP). The annotations describe traffic relevant concepts mentioned in the texts. The annotations are split into general annotations and traffic specific annotations. The general annotations are names of persons and locations; the traffic specific annotations are roads, junctions, parking spaces, and freely defined keywords.

  • The feely defined keywords can be added or removed via the web interface.
  • To retrieve annotated texts, the sources of the texts have to be select first.
  • Next, the texts to retrieve from the selected sources can be specified in further detail. There is the possibility to retrieve all texts, only the texts with annotations, or only texts with ASFINAG specific annotations. The definition of ASFINAG specific annotations is set by the next option.
  • ASFINAG annotated text are text with at least one keyword, road, or parking annotation. This is selected with the option “Or”. To make a more specific selection, the option “And” selects ASFINAG annotated texts having at least one keyword annotation and one road or parking annotation as well.
  • Finally, the time range of the text’s publication needs to be specified before the texts are retrieved.
  • For each text, the annotations are displayed in a table and highlighted in the text.
  • If annotations in the text match with annotations of a traffic message and the publication time of the text falls into the validity that of said traffic messages, these traffic messages are shown together with the text.
  • If the text is annotated with a road name, the traffic status for this road is linked and can be shown.