TEXT MININIG TOOLS FOR SLOVENE LANGUAGE

Maks Horvat (2013) TEXT MININIG TOOLS FOR SLOVENE LANGUAGE. EngD thesis.

Preview

Abstract

We introduce the use of various tools for Slovenian language processing and adapt them for NLTK library. To automatically determine the part of speech tags we use algorithms from the NLTK library. From Gigafida corpus we build several taggers: n-gram, Brill, naive Bayes, maximum entropy and hidden Markov model. We measure the accuracy of part of speech tags and time complexity of the taggers. We also incorporated Obeliks program for lemmatization and part of speech tags assignment. For text parsing and identification of named entities we use dependencyParser and SLNER tools. We develop and test a module for information retrieval. We use inverted index, search with boolean operators, vector representation of documents and cosine similarity.

Item Type:

Thesis (EngD thesis)

Keywords:

text mining, natural language processing, Slovenian language, language tools

Number of Pages:

Language of Content:

Slovenian

Mentor / Comentors:

Name and Surname	ID	Function
izr. prof. dr. Marko Robnik Šikonja	276	Mentor

Link to COBISS:

http://www.cobiss.si/scripts/cobiss?command=search&base=50070&select=(ID=9903956)

Institution:

University of Ljubljana

Department:

Faculty of Computer and Information Science

Item ID:

2040

Date Deposited:

17 May 2013 09:27

Last Modified:

11 Jun 2013 11:11

URI:

http://eprints.fri.uni-lj.si/id/eprint/2040

Actions (login required)

View Item