Agent-Based Hybrid Intelligent Systems: An Agent-Based Fromework for Complex Problem Solving

Zhang Ch.3540246231

Solving complex problems in real-world contexts, such as financial investment planning or mining large data collections, involves many different sub-tasks, each of which requires different techniques. To deal with such problems, a great diversity of intelligent techniques are available, including traditional techniques like expert systems approaches and soft computing techniques like fuzzy logic, neural networks, or genetic algorithms. These techniques are complementary approaches to intelligent information processing rather than competing ones, and thus better results in problem solving are achieved when these techniques are combined in hybrid intelligent systems. Multi-Agent Systems are ideally suited to model the manifold interactions among the many different components of hybrid intelligent systems.This book introduces agent-based hybrid intelligent systems and presents a framework and methodology allowing for the development of such systems for real-world applications. The authors focus on applications in financial investment planning and data mining.

---

Table of contents :
Contents……Page 10
Part I Fundamentals of Hybrid Intelligent Systems and Agents……Page 18
1 Introduction……Page 20
1.1 Hybrid Intelligent Systems Are Essential for Solving Complex Problems……Page 21
1.2 Hybrids Are Complex……Page 22
1.3 Agent Perspectives Are Suitable for Hybrids……Page 24
1.4 Motivation and Targets……Page 28
2 Basics of Hybrid Intelligent Systems……Page 30
2.1 Typical Intelligent Techniques……Page 31
2.2 Advantages and Disadvantages of Typical Intelligent Techniques……Page 38
2.3 Classification of Hybrid Intelligent Systems……Page 40
2.4 Current Practice in Typical Hybrid Intelligent System Development……Page 44
3.1 Concepts of Agents and Multi-agent Systems……Page 46
3.2 Agents as a Paradigm for Software Engineering……Page 47
3.3 Agents and Objects……Page 48
3.5 Approaches to Agentification……Page 50
3.7 Agent-Based Hybrid Systems: State of the Art……Page 52
Part II Methodology and Framework……Page 58
4.1 Traditional Methodologies……Page 60
4.2 Gaia Methodology……Page 61
4.4 Prometheus Methodology……Page 62
4.5 Methodology for Analysis and Design of Agent-Based Hybrids……Page 63
4.6 Summary……Page 72
5.1 A Unifying Agent Framework for Hybrid Intelligent Systems……Page 74
5.2 Issues on Ontologies……Page 76
5.3 Summary……Page 80
6 Matchmaking in Middle Agents……Page 82
6.1 Description of the Problem……Page 83
6.2 Related Work of Matchmaking in Middle Agents……Page 84
6.3 Improvements to Matchmaking Algorithms in Middle Agents……Page 90
6.4 Discussion……Page 107
Part III Application Systems……Page 108
7 Agent-Based Hybrid Intelligent System for Financial Investment Planning……Page 110
7.1 Introduction to Some Models Integrated in the System……Page 111
7.2 Analysis of the System……Page 121
7.3 Design of the System……Page 125
7.4 Architecture of the System……Page 128
7.5 Implementation of the System……Page 129
7.6 Case Study……Page 133
8 Agent-Based Hybrid Intelligent System for Data Mining……Page 144
8.1 Typical Data Mining Techniques……Page 145
8.2 Data Mining Requires Hybrid Solutions……Page 150
8.3 Requirements of the Agent-Based Hybrid Systems for Data Mining……Page 151
8.4 Analysis and Design of the System……Page 152
8.5 Implementation of the System……Page 155
8.6 Case Study……Page 157
Part IV Concluding Remark……Page 160
9.1 Flexibility and Robustness Testing……Page 162
9.2 Future Work……Page 163
Appendix: Sample Source Codes of the Agent-Based Financial Planning System……Page 166
A Source Codes for Data Supply Agent (StockData)……Page 167
B Source Codes of Planning Agent for Portfolio Selection (Stock)……Page 170
C Source Codes for Portfolio Selection Agent Based on Markowitz’s Model (Moki)……Page 176
D Source Codes for Portfolio Selection Agent Based on Fuzzy Logic Model (Fuzz)……Page 179
E Source Codes for Portfolio Selection Agent Based on Possibility Distribution Model (Poss)……Page 182
F Source Codes for Decision Aggregation Agent Based on Ordered Weighted Averaging Operators (Aggr)……Page 185
G Source Codes for Planning Agent of Investment Decision-Making (Invpolicy)……Page 187
H Source Codes for Investment Decision-Making Agent (Invppt)……Page 190
I Source Codes for Interest Prediction Agent Based on Fuzzy Logic and Genetic Algorithms (Flga)……Page 193
J Source Codes for Interest Prediction Agent Based on Neural Networks (Ffin)……Page 196
References……Page 200
C……Page 210
J……Page 211
P……Page 212
Y……Page 213