Seminario dr. Alonso: Data2Text Systems: “A hundred words are worth a million data”

  • Quando il 27/10/2017 dalle 09:30 alle 11:30 (Europe/Berlin / UTC200)
  • Dove Bari, Italia
  • Partecipanti Prof. Dr. José María Alonso Moral Centro Singular de Investigación en Tecnoloxías da Información (CiTIUS) Universidade de Santiago de Compostela (USC), Spain
  • Aggiungi l'evento al calendario iCal
Data2Text Systems: 
“A hundred words are worth a million data”

In recent years, there has been a huge effort connecting all kind of devices to Internet. From small devices (e.g., e-health monitoring sensors or mobile phones) that we carry daily in what is called the body-area-network, to big devices (such as cars), passing by all devices (e.g., TVs or refrigerators) at home. Some researchers are already talking about the “Internet of Everything” where people, objects and services are connected through the same network. Accordingly, in a highly connected world flooded with data, it is necessary to explore new ways of making information comprehensible to users of applications and/or services. In this context, automatic data-to-text (D2T) systems are increasingly being used as a means to facilitate data understanding by humans. D2T is a part of the wide field of Natural Language Generation, which automatically generates high-quality natural language narratives aimd at summarizing the most relevant information hidden in the data. Such narratives are usually provided as a complementary tool to graphical visualizations, in a form that can be understood directly by humans. This seminar will present the general bases of D2T systems. Special attention will be devoted to the actual role that Computational Intelligence, specially Fuzzy Logic, has played in this field. We will also carry out a review of real D2T systems of high impact, in areas such as robot-journalism, meteorological forecasting, health care or smart cities. In addition, we will give an overview on open source software for D2T. Finally, we will conclude with a discussion about the main open challenges and prospects in this research field.

Azioni sul documento

pubblicato il 12/10/2017 ultima modifica 05/10/2022
Hanno contribuito: claudia.damato