Please use this identifier to cite or link to this item: https://www.um.edu.mt/library/oar/handle/123456789/40252
Title: Building crime profiles from public data
Authors: Bezzina, Desiree
Keywords: Data mining
Crime -- Malta
Natural language processing (Computer science)
Crime analysis -- Malta
Issue Date: 2018
Citation: Bezzina, D. (2018). Building crime profiles from public data (Bachelor's dissertation).
Abstract: Over the past several years, crime in Malta has been on the rise (Formosa, 2017). Crime is a very serious issue and a major problem since it effects society, not only in Malta but every country in the world (Adigun, 2013; Badiora and Afon, 2013). Thus, this study aims to find ways with which public data can be exploited and build crime profiles based on the documents and their entities related. The public data used is online news articles and blogs published by the same websites. With the use of Natural Language Processing techniques articles are filtered out and linked to the crime type or crime types that they are related to. Results show that news online can be biased on what news to report. When comparing the articles reported for the tested crime types some news sources focused to report more crimes then others. The statistics obtained by Formosa, do not reflect the crimes reported. Formosa reported that from the total number of crimes, theft makes up to 51% of all crimes (Formosa, 2017), nevertheless, results showed that in some sources, theft was the least crime reported.
Description: B.SC.SOFTWARE DEVELOPMENT
URI: https://www.um.edu.mt/library/oar//handle/123456789/40252
Appears in Collections:Dissertations - FacICT - 2018
Dissertations - FacICTCIS - 2018

Files in This Item:
File Description SizeFormat 
18BSCITSD03.pdf
  Restricted Access
2.03 MBAdobe PDFView/Open Request a copy


Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.