Metoodika kasvatusväärtuste automaatseks eraldamiseks online-meediatekstidest
Files
Date
2019
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Magistritöö eesmärgiks oli luua metoodika koos tarkvaralahendusega, mis võimaldaks online-meediatekstide hulgast leida üles tekstid, mis sisaldavad kasvatusväärtusi, ning neid tekste analüüsida.Loodud tarkvaralahendus hõlmab andmete leidmist ja allalaadimist Perekooli Koolilapse foorumi meediatekstidest, kasvatusväärtuste sõnastiku loomist, kasvatusväärtusi sisaldavate tekstide filtreerimist ning analüüsi, sh ka tekstide emotsionaalsuse analüüsi. Tarkvaralahenduse loomisel on kasutatud Tartu ülikoolis loodud keeletöötlusteeki EstNLTK ning magistritöö analüüsi käigus saadud andmeid on plaanis kasutada teadusartiklite kirjutamisel. Loodud metoodika alusel on plaanis analüüsida ka teisi meediatekste.
The aim of the present thesis was to create a methodology with a software solution to find texts containing socialization values among online media texts and to analyse the filtered texts.The software solution includes web crawling and downloading the texts from the forum Koolilaps of Perekool, creating a dictionary of socialization values, filtering texts using the dictionary of socialization values and analysing the filtered texts with sentiment analysis.The software package EstNLTK was used to create the software solution. Further plans include using the results of the thesis for writing academic articles and using the methodology and the software solution for analysing other media texts.
The aim of the present thesis was to create a methodology with a software solution to find texts containing socialization values among online media texts and to analyse the filtered texts.The software solution includes web crawling and downloading the texts from the forum Koolilaps of Perekool, creating a dictionary of socialization values, filtering texts using the dictionary of socialization values and analysing the filtered texts with sentiment analysis.The software package EstNLTK was used to create the software solution. Further plans include using the results of the thesis for writing academic articles and using the methodology and the software solution for analysing other media texts.