Muutke küpsiste eelistusi

Agent Intelligence Through Data Mining 1st ed. 2005. Corr. 2nd printing 2005 [Kõva köide]

4.00/5 (4 hinnangut Goodreads-ist)
,
Teised raamatud teemal:
  • Kõva köide
  • Hind: 132,08 €*
  • * hind on lõplik, st. muud allahindlused enam ei rakendu
  • Tavahind: 155,39 €
  • Säästad 15%
  • Raamatu kohalejõudmiseks kirjastusest kulub orienteeruvalt 2-4 nädalat
  • Kogus:
  • Lisa ostukorvi
  • Tasuta tarne
  • Tellimisaeg 2-4 nädalat
  • Lisa soovinimekirja
Agent Intelligence Through Data Mining 1st ed. 2005. Corr. 2nd printing 2005Suurem pilt
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9780387243528.html
Märksõnad:
Knowledge, hidden in voluminous data repositories routinely created and maintained by today's applications, can be extracted by data mining. The next step is to transform this discovered knowledge into the inference mechanisms or simply the behavior of agents and multi-agent systems. Agent Intelligence Through Data Mining addresses this issue, as well as the arguable challenge of generating intelligence from data while transferring it to a separate, possibly autonomous, software entity. This book contains a methodology, tools and techniques, and several examples of agent-based applications developed with this approach. This volume focuses mainly on the use of data mining for smarter, more efficient agents.Agent Intelligence Through Data Mining is designed for a professional audience of researchers and practitioners in industry. This book is also suitable for graduate-level students in computer science.

This book addresses the use of data mining for smarter, more efficient agents, as well as the challenge of generating intelligence from data while transferring it to a separate, possibly autonomous, software entity. Following a brief review of data mining and agent technology fields, the book presents a methodology for developing multi-agent systems, describes available open-source tools, and demonstrates the application of the methodology on three different cases.
Dedication v
List of Figures xiii
List of Tables xvii
Foreword xix
Preface xxi
Acknowledgments xxv
Part I Concepts and Techniques
1. INTRODUCTION
3(8)
1 The Quest for Knowledge
3(1)
2 Problem Description
4(1)
3 Related Bibliography
5(1)
4 Scope of the Book
6(2)
5 Contents of the Book
8(1)
6 How to Read this Book
9(2)
2. DATA MINING AND KNOWLEDGE DISCOVERY: A BRIEF OVERVIEW
11(30)
1 History and Motivation
11(7)
1.1 The Emergence of Data Mining
11(2)
1.2 So, what is Data Mining?
13(1)
1.3 The KDD Process
13(2)
1.4 Organizing Data Mining Techniques
15(3)
2 Data Preprocessing
18(3)
2.1 The Scope of Data Preprocessing
18(1)
2.2 Data Cleaning
18(1)
2.3 Data Integration
19(1)
2.4 Data Transformation
19(1)
2.5 Data Reduction
20(1)
2.6 Data Discretization
20(1)
3 Classification and Prediction
21(5)
3.1 Defining Classification
21(1)
3.2 Bayesian Classification
21(1)
3.3 Decision Trees
22(4)
3.3.1 The ID3 algorithm
24(2)
4 Clustering
26(6)
4.1 Definitions
27(1)
4.2 Clustering Techniques
27(1)
4.3 Representative Clustering Algorithms
28(4)
4.3.1 Partitioning Algorithms
28(1)
4.3.2 Hierarchical Algorithms
29(1)
4.3.3 Density-Based Algorithms
30(2)
5 Association Rule Extraction
32(3)
5.1 Definitions
32(1)
5.2 Representative Algorithms
33(2)
6 Evolutionary Data Mining Algorithms
35(5)
6.1 The Basic Concepts of Genetic Algorithms
35(1)
6.2 Genetic Algorithm Terminology
36(1)
6.3 Genetic Algorithm Operands
37(1)
6.4 The Genetic Algorithm Mechanism
38(1)
6.5 Application of Genetic Algorithms
38(2)
7
Chapter review
40(1)
3. INTELLIGENT AGENTS AND MULTI-AGENT SYSTEMS
41(18)
1 Intelligent Agents
41(7)
1.1 Agent Definition
41(1)
1.2 Agent Features and Working Definitions
42(2)
1.3 Agent Classification
44(1)
1.4 Agents and Objects
45(2)
1.5 Agents and Expert Systems
47(1)
1.6 Agent Programming Languages
47(1)
2 Multi-Agent Systems
48(11)
2.1 Multi-Agent System Characteristics
50(1)
2.2 Agent Communication
51(2)
2.3 Agent Communication Languages
53(2)
2.3.1 KQML
53(1)
2.3.2 KIF
54(1)
2.3.3 FIPA ACL
54(1)
2.4 Agent Communities
55(4)
Part II Methodology
4. EXPLOITING DATA MINING ON MAS
59(12)
1 Introduction
59(4)
1.1 Logic and Limitations
60(2)
1.2 Agent Training and Knowledge Diffusion
62(1)
1.3 Three Levels of Knowledge Diffusion for MAS
63(1)
2 MAS Development Tools
63(3)
3 Agent Academy
66(5)
3.1 AA Architecture
67(1)
3.2 Developing Multi-Agent Applications
68(1)
3.3 Creating Agent Ontologies
68(1)
3.4 Creating Behavior Types
68(1)
3.5 Creating Agent Types
69(1)
3.6 Deploying a Multi Agent System
69(2)
5. COUPLING DATA MINING WITH INTELLIGENT AGENTS
71(22)
1 The Unified Methodology
72(10)
1.1 Formal Model
72(1)
1.1.1 Case 1: Training at the MAS application level
72(1)
1.1.2 Case 2: Training at the MAS behavior level
72(1)
1.1.3 Case 3: Training evolutionary agent communities
72(1)
1.2 Common Primitives for MAS Development
73(3)
1.3 Application Level: The Training Framework
76(1)
1.4 Behavior Level: The Training Framework
77(3)
1.5 Evolutionary Level: The Training Framework
80(2)
2 Data Miner: A Tool for Training and Retraining Agents
82(11)
2.1 Prerequisites for Using the Data Miner
82(1)
2.2 Data Miner Overview
82(3)
2.3 Selection of the Appropriate DM Technique
85(1)
2.4 Training and Retraining with the Data Miner
86(7)
Part III Knowledge Diffusion: Three Representative Test Cases
6. DATA MINING ON THE APPLICATION LEVEL OF A MAS
93(22)
1 Enterprise Resource Planning Systems
93(2)
2 The Generalized Framework
95(14)
2.1 IRF Architecture
97(4)
2.1.1 Customer Order Agent type
98(1)
2.1.2 Recommendation Agent type
99(1)
2.1.3 Customer Profile Identification Agent type
99(1)
2.1.4 Supplier Pattern Identification Agent type
100(1)
2.1.5 Inventory Profile Identification Agent type
100(1)
2.1.6 Enterprise Resource Planning Agent type
100(1)
2.2 Installation and Runtime Workflows
101(2)
2.3 System Intelligence
103(12)
2.3.1 Benchmarking customer and suppliers
103(3)
2.3.2 IPIA products profile
106(1)
2.3.3 RA Intelligence
106(3)
3 An IRF Demonstrator
109(3)
4 Conclusions
112(3)
7. MINING AGENT BEHAVIORS
115(20)
1 Predicting Agent Behavior
115(9)
1.1 The Prediction Mechanism
115(4)
1.2 Applying kappa-Profile on MAS
119(2)
1.3 Modeling Agent Actions in an Operation Cycle
121(1)
1.4 Mapping Agent Actions to Vectors
122(1)
1.5 Evaluating Efficiency
123(1)
1.5.1 Profile efficiency evaluation
123(1)
1.5.2 Prediction system efficiency evaluation
124(1)
2 A Recommendation Engine Demonstrator
124(7)
2.1 System Parameters
125(2)
2.1.1 The fuzzy variable Time
125(1)
2.1.2 The fuzzy variable Frequency
126(1)
2.1.3 The output fuzzy variable Weight
127(1)
2.2 The Rules of the FIS
127(3)
2.3 Browsing through a Web Site
130(1)
3 Experimental Results
131(2)
4 Conclusions
133(2)
8. MINING KNOWLEDGE FOR AGENT COMMUNITIES
135(28)
1 Ecosystem Simulation
135(3)
2 An Overview of Biotope
138(10)
2.1 The Biotope Environment
138(1)
2.2 The Biotope Agents
139(3)
2.2.1 Agent sight
139(1)
2.2.2 Agent movement
139(2)
2.2.3 Agent reproduction
141(1)
2.2.4 Agent communication - Knowledge exchange
141(1)
2.3 Knowledge Extraction and Improvement
142(3)
2.3.1 Classifiers
143(1)
2.3.2 Classifier Evaluation mechanism
143(1)
2.3.3 Genetic Algorithm
144(1)
2.4 The Assessment Indicators
145(3)
2.4.1 Environmental indicators
145(1)
2.4.2 Agent performance indicators
146(2)
3 The Implemented Prototype
148(2)
3.1 Creating a New Simulation Scenario
149(1)
4 Experimental Results
150(10)
4.1 Exploiting the Potential of Agent Communication
151(4)
4.1.1 Specifying the optimal communication rate
152(1)
4.1.2 Agent efficiency and knowledge base size
152(1)
4.1.3 Agent communication and unreliability
153(2)
4.2 GAs in Unreliable Environments
155(3)
4.3 Simulating Various Environments
158(2)
5 Conclusions
160(3)
Part IV Extensions...
9. AGENT RETRAINING AND DYNAMICAL IMPROVEMENT OF AGENT INTELLIGENCE
163(14)
1 Formal Model
163(3)
1.1 Different Retraining Approaches
165(1)
2 Retraining in the Case of Classification Techniques
166(3)
2.1 Initial Training
166(1)
2.2 Retraining an Agent Type
167(1)
2.3 Retraining an Agent Instance
168(1)
3 Retraining in the Case of Clustering Techniques
169(1)
3.1 Initial Training
170(1)
3.2 Retraining
170(1)
4 Retraining in the Case of Association Rule Extraction Techniques
170(1)
4.1 Initial Training
170(1)
4.2 Retraining
170(1)
5 Retraining in the Case of Genetic Algorithms
171(1)
6 Experimental Results
171(4)
6.1 Intelligent Environmental Monitoring System
171(2)
6.2 Speech Recognition Agents
173(1)
6.3 The Iris Recommendation Agent
174(1)
7 Conclusions
175(2)
10. AREAS OF APPLICATION & FUTURE DIRECTIONS
177(12)
1 Areas of Application
177(4)
1.1 Environmental Monitoring Information Systems
177(2)
1.2 Agent Bidding and Auctioning
179(1)
1.3 Enhanced Software Processing
180(1)
2 Advanced AT-DM Symbiosis Architectures
181(2)
2.1 Distributed Agent Training Architectures
181(1)
2.2 Semantically-Aware Grid Architectures
182(1)
3 Summary and Conclusions
183(2)
4 Open Issues and Future Directions
185(4)
References 189(10)
Index 199(2)
About The Authors 201