NUK - logo
E-viri
Celotno besedilo
  • Zitnik, Slavko; Draskovic, Drazen; Nikolic, Bosko; Bajec, Marko

    2017 25th Telecommunication Forum (TELFOR), 2017-Nov.
    Conference Proceeding

    Nowadays, there are many tools and applications available that use novel machine learning algorithms, but only few tools exist that bridge the gap between the computer scientists and general public. The latter especially holds for the field of natural language processing. Therefore in this paper we present a new information extraction toolkit, called nutIE, that offers a frontend for end-to-end text analysis using a modern javascript-based web application. The core of the toolkit includes all the high-level information extraction techniques with supporting preprocessing and evaluation methods. All these can be accessed by a Scala programmatic API or an arbitrary third-party application via REST interface. It has already been shown that the integrated toolkit's algorithms achieve state-of-the art results. The toolkit is currently applied at international text processing projects and courses.