Mustripõhine informatsiooni eraldamine Eesti kohtulahenditest
Date
2016
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Käesolevas töös rakendatakse mustripõhise informatsiooni eraldamise meetodeid Eesti kohtulahenditele, mis on senimaani vähekasutatud eestikeelne ressurss. Töö eesmärgiks on luua andmebaas, mille põhjal saaks arendada tarkvara, mis võimaldaks efektiivselt relevantseid kohtulahendeid otsida ning otsingute tulemusi analüüsida. Töö käigus loodud andmebaasis on kohtulahenditest eraldatud huvipakkuvat informatsiooni sisaldavad tekstiosad, mille põhjal on kohtuotsuste tulemusi võimalik analüüsida näiteks maakondade ja kohtute lõikes.
This thesis describes how pattern based information extraction methods were applied to Estonian court documents, which is an Estonian language resource that until now has remained mostly unused. The goal of the thesis was to create a database, which would enable developing an application that would allow effective searching of relevant court documents and analyzing the results. The result of this work is a database containing facts of interest that were extracted from the court documents. The database enables analyzing decisions written down in the court documents, for example based on counties and the different Estonian courts.
This thesis describes how pattern based information extraction methods were applied to Estonian court documents, which is an Estonian language resource that until now has remained mostly unused. The goal of the thesis was to create a database, which would enable developing an application that would allow effective searching of relevant court documents and analyzing the results. The result of this work is a database containing facts of interest that were extracted from the court documents. The database enables analyzing decisions written down in the court documents, for example based on counties and the different Estonian courts.