Primjena kompozicijske distribucijske semantike u zadatku semantičke sličnosti teksta

master's thesis

• Title (english): Application of Compositional Distributional Semantics for Semantic Text Similarity
• Author: Luka Skukan
• Mentor: Jan Šnajder (mentor)
• Committee member: Jan Šnajder (predsjednik povjerenstva)
• Committee member: Siniša Šegvić (član povjerenstva)
• Committee member: Marko Čupić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2017-02-14, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U zadnjih nekoliko godina, moderni pristupi kompozicijskoj distribucijskoj semantici donijelu su revoluciju u mnoga područja bazirana na semantičkoj sličnosti. U ovom je radu dan pregled tog područja. Izabrano je nekoliko modela i iskušani su na zadacima nadziranog i nenadziranog učenja u području semantike, na tekstovima na hrvatskome jeziku. Posebna pažnja dana je sematičkoj sličnosti teksta. Razvijena je i prikazana pokazna web aplikacija.

Abstract (english)

Modern approaches to compositional distributional semantics have revolutionized many areas based on semantic similarity in recent years. In this thesis, a survey of the field is given. Several of these modern models and approaches are selected and tried on a variety of supervised and unsupervised learning tasks in the field of semantics, on texts in the Croatian language, with attention given to semantic text similarity. A showcase web application is developed and presented.

• Keywords: Kompozicijska distribucijska semantika; distribucijska semantika; obrada prirodnog jezika; duboko učenje; modeli mješanja vektora; sematička sličnost; hrvatski jezik
• Keywords (english): Compositional distributional semantics; distributional semantics; natural language processing; deep learning; vector mixture models; semantic similarity; Croatian language
• Language: english
• URN:NBN: urn:nbn:hr:168:473823
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-14 00:45:23