Investigating crimes using text mining and network analysis
Other Title(s)
التحقيق في الجرائم من خلال تحليل النصوص و الشبكات الاجتماعية
Dissertant
Thesis advisor
Comitee Members
al-Agha, Iyad Muhammad
Abu-Shaban, Yusuf Nabil
University
Islamic University
Faculty
Faculty of Information Technology
University Country
Palestine (Gaza Strip)
Degree
Master
Degree Date
2015
English Abstract
In these days, security of citizens is considered one of the major concerns of any government in the world.
In every country, there is a huge amount of unstructured texts coming from investigating offenders in police departments.
As a result, the importance of crimes analysis is growing day after day.
Criminology is one of the hot areas that focuses on the scientific study of crimes and aims to identify crime characteristics and both criminal behaviors and networks.
This field of study is one of the most intelligent research areas where text mining is used to process unstructured texts and extract meaningful information which is hidden in the unstructured texts.
The knowledge extracted from text mining is very useful to police departments where solving crimes is a very complex task that requires human effort and experience.
There is a little research in methods and techniques that extract criminal networks from unstructured investigations texts especially in Arabic language.
Accordingly, the current research proposes a system to identify networks of criminals, and extract useful information relevant to crimes such as offender’s connection networks and discover a new hidden relationship between offenders by linking investigation documents with each other.
After that, the results of the research are visualized direct and indirect relation between offenders to help policemen find pieces of evidences related to certain crimes and accordingly apply the law.
In our proposed system, we climb three main distinct contributions to discover forensics using investigation documents.
The first by extracting offender names from unstructured text.
Secondly, by constructing a crime network from real Arabic investigation documents.
Finally, we provide analysis of the interaction between offenders in different documents that directly and indirectly related used to discover a new clue used to solve the crime puzzle.
To evaluate the performance and effectiveness of the proposed system, real unstructured documents about investigations are obtained from police departments in the Gaza Strip.
The experimental results show that the proposed system is effective in identifying proper offender person's name from real Arabic Documents.
The average results for our system using the F-measure is 89% also the average of F-measure in a proposed algorithm for discovery hidden relationship arrive to 92%.
In addition, we found that our approach achieves best F-measure results in most cases.
Main Subjects
Sociology and Anthropology and Social Work
Information Technology and Computer Science
Topics
No. of Pages
80
Table of Contents
Table of contents.
Abstract.
Abstract in Arabic.
Chapter One : Introduction.
Chapter Two : Related works.
Chapter Three : Theoretical foundation.
Chapter Four : The proposed crime detection approach.
Chapter Five : Experiments setup.
Chapter Six : Conclusion and future work.
References.
American Psychological Association (APA)
al-Yaziji, Nail Taysir Nimr. (2015). Investigating crimes using text mining and network analysis. (Master's theses Theses and Dissertations Master). Islamic University, Palestine (Gaza Strip)
https://search.emarefa.net/detail/BIM-688572
Modern Language Association (MLA)
al-Yaziji, Nail Taysir Nimr. Investigating crimes using text mining and network analysis. (Master's theses Theses and Dissertations Master). Islamic University. (2015).
https://search.emarefa.net/detail/BIM-688572
American Medical Association (AMA)
al-Yaziji, Nail Taysir Nimr. (2015). Investigating crimes using text mining and network analysis. (Master's theses Theses and Dissertations Master). Islamic University, Palestine (Gaza Strip)
https://search.emarefa.net/detail/BIM-688572
Language
English
Data Type
Arab Theses
Record ID
BIM-688572