Money laundering and economy development case of Lebanon

Dissertant

ٍٍٍSawaya, Shadiyah Malik

Thesis advisor

Abd Allah, Khalid Amin

Comitee Members

al-Sarraf, Zaki Jawad
al-Tarawinah, Muhammad Ahmad
al-Halasa, Awni Ibrahim

University

Arab Academy for Financial and Banking Sciences

Faculty

The Faculty of Banking and Financial Sciences

Department

Department of Business Administration

University Country

Jordan

Degree

Ph.D.

Degree Date

2011

English Abstract

This thesis proposes modeling of artificial emotions through agents based on original symbolic / computational approach.

The symbolic approach utilizes a symbolic emotional rule-based system that continuously interacts with the environment.

It has internal "thinking" machinery that comes as a result of-series-of inferences, evaluation, evolution processes, adaptation, learning besides emotions To this date, the field of Artificial Intelligence has largely ignored the use of artificial emotions and intuition to guide reasoning and decision-making.

Several models of artificial emotions have been proposed, but most of the work in this area is limited to analysis without complex applications.

In our thesis, we extend previous research on computational models of emotions [OCC Model and Bates Reilly Model] by modifying architecture [Nilsson Model], and integrating personality, motivations, mood and sensitivity into software agents.

We propose a general, flexible, and powerful architecture to build software agents that embed artificial emotions.

The propose agent possesses rational knowledge and reactive capabilities, and interacts with the external world, including other agents.

We are aiming to develop artificial mechanisms that can play the role emotions play in natural life.

~We investigate if artificial emotional control can improve performance of agents h some circumstances.

The main contribution of this thesis is to present emotions as valid approach to improve the behavior of artificial agents.

Moreover, a systemati< view of the kinds of emotions that can take place, assuming emotions are processe involving assessment, emotion-signal generation, and emotion-response, are presented.

To come across emotions, an agent needs to incorporate a deeper model of personality, mood and sensitivity.

EMAM (Emotional Agent Model) is proposed for this purpose.

EMAM-agent can generate emotions signals, evaluate and assess events, taking into account the integration of personality, mood, sensitivity, feeling and motivational states.

EMAM-agent takes the proper actions or plans for actions (sequence of actions) to accomplish goal(s).

Two different applications; Material Requirement Planning (as business application) and Orphanage Care Application (as social application) are chosen to demonstrate the use of EMAM emotional agent.

Netlogo is used as an agent programming language.

It is the agent-oriented language used to simulate and implement our proposed models.

The two applications are implemented using both agents types; the one with artificial emotions and the one without emotions.

The two systems are simulated and results are compared.

Our study shows that systems (or agents) with proper model of emotions can perform in many cases better than systems without emotions.

In conclusion, we shed the light here on how artificial emotions can be modeled in an emotional rule-based agent systems and if emotions as they exist in "real intelligence" can be helpful for "artificial intelligence agents".

Consequently, our proposed emotional intelligent agent architectures are presented as an original generic blueprints on which the design of agents of that type could be based.

Main Subjects

Economics & Business Administration

Topics

• Economic conditions
• Lebanon
• Money laundering

No. of Pages

540

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-307186