Detecting offensive posts with machine learning methods

Leon Noe Jovan (2013) Detecting offensive posts with machine learning methods. EngD thesis.

Preview

Abstract

The main goal of this thesis was to develop a recognition system for offensive posts on the web. Theoretical backgrounds of machine learning, text mining and text categorization approaches are given for better understanding of this field of computer science. We present a framework of such a system, from text pre-processing, feature selection, term weighting to selection of best classifiers. The results are tested using the data obtained from a related competition on Kaggle. For the purpose of the thesis a database of Slovenian comments was built, which serves as a data set to verify the success of the classification of offensive comments in Slovenian language.

Item Type:

Thesis (EngD thesis)

Keywords:

machine learning, text mining, classification, offensive posts

Number of Pages:

Language of Content:

Slovenian

Mentor / Comentors:

Name and Surname	ID	Function
doc. dr. Zoran Bosnić	3826	Mentor

Link to COBISS:

http://www.cobiss.si/scripts/cobiss?command=search&base=50070&select=(ID=10132564)

Institution:

University of Ljubljana

Department:

Faculty of Computer and Information Science

Item ID:

2148

Date Deposited:

13 Sep 2013 15:42

Last Modified:

24 Sep 2013 09:33

URI:

http://eprints.fri.uni-lj.si/id/eprint/2148

Actions (login required)

View Item