Bramer Artificial Intelligence in Theory and Practice

IFIP 19th World Computer Congress, TC 12: IFIP AI 2006 Stream, August 21-24, 2006, Santiago, Chile
1. Auflage 2006
ISBN: 978-0-387-34747-9
Verlag: Springer
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, Band 217, 513 Seiten

Reihe: IFIP Advances in Information and Communication Technology

This volume presents proceedings from the 19th IFIP World Computer Congress in Santiago, Chile. The proceedings of the World Computer Congress are a product of the gathering of 2,000 delegates from more than 70 countries to discuss a myriad of topics in the ICT domain. Of particular note, this marks the first time that a World Computer Congress has been held in a Latin American country.

Topics in this series include:

The 4th International Conference on Theoretical Computer Science
Education for the 21st Century- Impact of ICT and Digital Resources
Mobile and Wireless Communication Networks
Ad-Hoc Networking
Network Control and Engineering for QoS, Security, and Mobility
The Past and Future of Information Systems: 1976-2006 and Beyond
History of Computing and Education
Biologically Inspired Cooperative Computing
Artificial Intelligence in Theory and Practice
Applications in Artificial Intelligence
Advanced Software Engineering: Expanding the Frontiers of Software

For a complete list of the more than 300 titles in the IFIP Series, visit springer.com. For more information about IFIP, please visit ifip.org.

Written for:
Researchers and practitioners in Information and Communication Technologies

Autoren/Hrsg.

Bramer, Max

Inhaltsverzeichnis

1;Contents;6
2;Foreword;12
3;Can Common Sense uncover cultural differences in computer applications?;16
3.1;1. Introduction;16
3.2;2. The Open Mind Common Sense Approach for Gathering and Using Common Sense Facts;18
3.3;3. Common Sense and Eating Habits;19
3.4;4 Using Cultural Knowledge in Interactive Applications;20
3.5;5. Conclusions and Future Works;24
3.6;References;25
4;ARTIFICIAL INTELLIGENCE AND KNOWLEDGE MANAGEMENT;26
5;The IONWI Algorithm: Learning when and when not to interrupt;35
5.1;1. Introduction;35
5.2;2. The/OA^fF/Algorithm;36
5.3;3. Experimental Results;41
5.4;4. Related Work;43
5.5;5. Conclusions and Future Work;43
5.6;References;44
6;Formal Analysis of the Communication of Probabilistic Knowledge;45
6.1;1 Introduction;45
6.2;2 Motivation;46
6.3;3 SLP Probabilistic Logic;48
6.4;4 Communication of Probabilistic Knowledge;51
6.5;5 Future Works;54
6.6;7 Acknowledgments;54
6.7;8 References;54
7;Detecting and Repairing Anomalous Evolutions in Noisy Environments: Logic Programming Formalization and Complexity Results;55
7.1;1 Introduction;55
7.2;2 Preliminaries on Extended Logic Programs (ELPs);58
7.3;3 Formal Framework;58
7.4;4 Reasoning with Noisy Sensors;61
7.5;5 Conclusions;63
7.6;References;64
8;Adding Semantic Web Services Matching and Discovery Support to the MoviLog Platform;65
8.1;1 Introduction;65
8.2;2 Semantic Web Services;66
8.3;3 MoviLog;67
8.4;4 Semantic Matching in MoviLog;68
8.5;5 A sample scenario;71
8.6;6 Experimental results;72
8.7;7 Related work;73
8.8;8 Conclusion and future work;73
8.9;References;74
9;Learning Browsing Patterns for Context-Aware Recommendation;75
9.1;1 Introduction;75
9.2;2 Learning and Using Browsing Patterns;76
9.3;3 Experimental Results;80
9.4;4 Related Work;83
9.5;5 Conclusions;84
9.6;References;84
10;Applying Collaborative Filtering to Reputation Domain: a model for more precise reputation estimates in case of changing behavior by rated participants;85
10.1;1 Introduction;85
10.2;2 Proposed Model;87
10.3;3 Experiment;91
10.4;4 Conclusions;94
10.5;5 References;94
11;Combine Vector Quantization and Support Vector Machine for Imbalanced Datasets;95
11.1;1 Introduction;95
11.2;2. Support Vector Machine;96
11.3;3. Vector Quantization;96
11.4;4 VQ-SVM to Rebalance Dataset;98
11.5;5. Experiments;99
11.6;6. Conclusion;101
12;Ontology Support for Translating Negotiation Primitives;103
12.1;1 Introduction;103
12.2;2 Architecture of the Translator;105
12.3;3 Shared Ontology;106
12.4;4 Implementation of the Negotiation System;107
12.5;5 Experimentation;108
12.6;6 Conclusions;111
12.7;References;111
13;Statistical Method of Context Evaluation for Biological Sequence Similarity;113
13.1;1 Introduction;113
13.2;2 Methods;115
13.3;3 Experiments;118
13.4;4 Conclusions;120
14;Biological inspired algorithm for Storage Area Networks (ACOSAN);123
14.1;1 INTRODUCTION;123
14.2;2. FIBRE CHANNEL;126
14.3;3. USES OF ANT COLONY OPTIMIZATION IN ROUTING ALGORITHMS;127
14.4;4. ROUTING ALGORITHM;128
14.5;5. CONCLUSIONS AND FUTURE WORKS;131
14.6;6. ACKNOWLEDGMENTS;131
14.7;7. REFERENCES;131
15;Radial Basis Functions Versus Geostatistics in Spatial Interpolations;133
15.1;1 Introduction;133
15.2;2 Geostatistics in Spatial Interpolations;134
15.3;3 Soft Computing Methods in Spatial Interpolations;134
15.4;4 RBF Versus Geostatistics;136
15.5;5 Study Case: SIC 2004;138
15.6;6 Conclusions;140
15.7;7 References;141
16;Neural Networks applied to wireless communications;143
16.1;1 Introduction;143
16.2;2 Neural network-based modeling;145
16.3;3 TDNN model;146
16.4;4 Measurements and validation results;148
16.5;References;152
17;Anomaly Detection using prior knowledge: application to TCP/IP traffic;153
17.1;1 Introduction;153
17.2;3 Experimental Results;160
17.3;4 Conclusions;161
17.4;5 References;162
18;A study on the ability of Support Vector Regression and Neural Networks to Forecast Basic Time Series Patterns;163
18.1;1 Introduction;163
18.2;2 Modelling SVR and NN for Time Series Prediction;164
18.3;3 Experiments and Results;167
18.4;4 Conclusion;171
18.5;References;172
19;Neural Plasma;173
19.1;1 Introduction;173
19.2;2 Confidence in Classification;175
19.3;3 Jittering;175
19.4;5 Materials and Methods;179
19.5;6 Results;180
20;Comparison of SVM and Some Older Classification Algorithms in Text Classification Tasks;183
20.1;1 Introduction;183
20.2;2 Document Collection, Algorithms and Evaluation Methodology;184
20.3;3 Experimental Results;186
20.4;4 Conclusion;191
20.5;References;192
21;An automatic graph layout procedure to visualize correlated data;193
21.1;1 Introduction;193
21.2;2 Graph Layout Procedure;194
21.3;3 Memetic Algorithm;196
21.4;4 Computational Results;199
21.5;5 Conclusions;201
21.6;References;201
22;Knowledge Perspectives in Data Grids;203
22.1;1 Introduction;203
22.2;2 Knowledge Perspectives;204
22.3;3 Knowledge Perspectives iraplementation in Data Grids;205
22.4;4 An example using Wordnet;207
22.5;5 Related Work;209
22.6;6 Conclusions and Future Work;211
22.7;References;211
23;On t h e Class Distribution Labelling Step Sensitivity of CO-TRAINING;213
23.1;1 Introduction;213
23.2;2 Related Work;214
23.3;5 Experimental Evaluation;218
23.4;6 Conclusions and Future Work;221
23.5;References;221
24;Global Convexity in the Bi-Criteria Traveling Salesman Problem;231
24.1;1 Introduction;231
24.2;2 Multi-Objective Optimization Problems;232
24.3;3 The Multi-Objective TSP;233
24.4;4 Global Convexity;233
24.5;5 Topological Analysis of the Solution Space;234
24.6;6 Conclusions and Future Work;238
24.7;References;240
25;Evolutionary Algorithm for State Encoding;241
25.1;1 Introduction;241
25.2;2 Evolutionary Algorithm;243
25.3;3 Variations of the Evolutionary Algorithm;247
25.4;4 Experimentation;256
25.5;5 In the Practices;258
25.6;6 Conclusion;259
25.7;References;260
26;Multitree-Multiobjective Multicast Routing for Traffic Engineering;261
26.1;1 Introduction;261
26.2;2 Problem Formulation;262
26.3;3 Multiobjective Optimization Problems;263
26.4;4 Proposed Algorithm;264
26.5;5 Testing Scenario;266
26.6;6 Experimental results;268
26.7;7 Conclusion and future work;269
26.8;References;270
27;Learning Discourse-new References in Portuguese Texts;281
27.1;1 Introduction;281
27.2;2 Related Work;282
27.3;3 Classes Description;282
27.4;5 Decision Trees Learning;286
27.5;6 Feature Analysis;287
27.6;7 Evaluation on unseen data;288
27.7;8 Final Remarks;289
27.8;References;290
28;Fuzzy Rule-Based Hand Gesture Recognition*;299
28.1;1 Introduction;299
28.2;2 The Fuzzy Rule Based for Hand Gesture Recognition;301
28.3;3 Case Study: Hand Gestures of LIBRAS;305
28.4;4 Conclusion and Final Remarks;307
28.5;References;308
29;Comparison of distance measures for historical spelling variants;309
29.1;1 Introduction;309
29.2;2 Requirements for hsv-distance measures;311
29.3;3 Comparative study of distance measures;312
29.4;4 Evaluation methodology;314
29.5;5 Results and interpretation;315
29.6;7 Further work and outlook;317
29.7;8 Acknowledgements;317
29.8;References;317
30;Patterns in Temporal Series of Meteorological Variables Using SOM & TDIDT;319
30.1;1. Introduction;319
30.2;2. Problem;320
30.3;3. Proposed Solution;321
30.4;4. Data for experiments;321
30.5;5. Results of experiments;322
30.6;6. Conclusions;325
30.7;7. References;326
31;Applying Genetic Algorithms to Convoy Scheduling;329
31.1;1 Introduction;329
31.2;2 Related Work;331
31.3;3 Our Approach;332
31.4;4 Results;335
31.5;5 Conclusion and Future Work;336
31.6;6 References;337
32;A GRASP algorithm to solve the problem of dependent tasks scheduling in different machines;338
32.1;1 Introduction;338
32.2;2 Proposed GRASP Algorithm;341
32.3;3. Numeric Experiences;345
32.4;4. Conclusions;347
32.5;REFERENCES;347
33;A support vector machine as an estimator of mountain papaya ripeness using resonant frequency or frequency centroid;348
33.1;1.1 Introduction;348
33.2;1.2 Materials and methods;351
33.3;1.3 Results;353
33.4;1.4 Discussion and conclusions;355
33.5;References;357
34;FieldPlan: Tactical Field Force Planning in BT;358
34.1;1 Introduction;358
34.2;2 The BT Field Force Planning Scenario;359
34.3;3 The FieldPlan System;360
34.4;4 Results;365
34.5;5 Conclusions;367
34.6;References;367
35;An Agent Solution to Flexible Planning and Scheduling of Passenger Trips;368
35.1;1 Introduction;368
35.2;2 Transportation Requirements;369
35.3;3 The Agent Architecture;370
35.4;4 Concrete Planning Systems;374
35.5;References;377
36;Facial expression recognition using shape and texture information;378
36.1;1.1 Introduction;378
36.2;1.2 System description;379
36.3;1.3 Texture information extraction;379
36.4;1.4 Shape information extraction;382
36.5;1.5 Fusion of texture and shape information;384
36.6;1.6 Experimental results;385
36.7;1.7 Conclusions;386
36.8;1.8 Acknowledgment;387
36.9;References;387
37;Limited Receptive Area neural classifier for texture recognition of metal surfaces;388
37.1;1 Introduction;388
37.2;2 Metal surface texture recognition;389
37.3;3 The LIRA neural classifier;390
37.4;5 Discussion;396
37.5;6 Conclusion;396
37.6;Acknowledgment;397
37.7;References;397
38;A Tracking Framework for Accurate Face Localization;398
38.1;1 Introduction;398
38.2;2 Tracking Framework;399
38.3;3 A trellis structure for optimal face detection;401
38.4;4 Experiments and results;404
38.5;5 Conclusion;405
38.6;6 Acknowledgement;405
38.7;References;406
39;Three Technologies for Automated Trading;416
39.1;1 Introduction;416
39.2;2 Data Mining;416
39.3;3 Trading Agents;418
39.4;4 Virtual Institutions;422
39.5;5 Conclusions;424
39.6;References;425
40;Modeling Travel Assistant Agents: a graded BDI approach;426
40.1;1 Introduction;426
40.2;2 Graded BDI agent model;428
40.3;3 A Travel Assistant Agent;429
40.4;4 Conclusions and Future Work;434
40.5;References;435
41;e-Tools: An agent coordination layer to support the mobility of persons with disabilities.;436
41.1;1 Introduction;436
41.2;2 e-Tools PROJECT;437
41.3;3 Designing the e-Tools Coordination Layer;439
41.4;4 Study and Conclusions;444
41.5;References;445
42;Conceptualization Maturity Metrics for Expert Systems;446
42.1;1. Metrics;446
42.2;2. Suggested Metrics;446
42.3;3. Applying the Metrics to Real World;453
42.4;4. Some Results;453
42.5;5. Conclusions;455
42.6;6. References;455
43;Toward developing a tele-diagnosis system on fish disease;456
43.1;1 Introduction;456
43.2;2 System architecture and components;457
43.3;3 The diagnosing process;461
43.4;4 Implementation;463
43.5;5 Conclusions;463
43.6;Acknowledgement;464
43.7;References;464
44;A new method for fish-disease diagnostic problem solving based on parsimonious covering theory and fuzzy inference model;466
44.1;1 Introduction;466
44.2;2 The fish-disease diagnostic model based on parsimonious covering theory;467
44.3;3 Diagnosis model based on fuzzy theory and diagnosis algorithm;471
44.4;4 Test results;472
44.5;5 Conclusion;474
44.6;Acknowledgement;475
44.7;Reference;475
45;Effective Prover for Minimal Inconsistency Logic;476
45.1;1 Introduction;476
45.2;2 The mbC Logic;477
45.3;3 A KE System for mbC;479
45.4;4 Problem Families;481
45.5;5 Evaluation;483
45.6;6 Conclusion;484
45.7;References;484
46;Identification of Important News for Exchange Rate Modeling;486
46.1;1 Introduction;486
46.2;2 News classification using a rule-based system;487
46.3;3 Experiment results;490
46.4;4 Conclusions;492
46.5;References;493
47;Autonomous Search and Rescue Rotorcraft Mission Stochastic Planning with Generic DBNs;494
47.1;1 Introduction;494
47.2;2 Hierarchical factored MDP;496
47.3;3 Abstract generic Dynamic Bayesian Network;496
47.4;4 Application to a search and rescue mission;501
47.5;5 Conclusion;503
47.6;References;503
48;Solving multi-objective scheduling problems—An integrated systems approach;504
48.1;1 Introduction;504
48.2;2 A metaheuristic system for multi-objective scheduling;506
48.3;3 Computational results;509
48.4;4 Conclusions and discussion;511
48.5;Acknowledgements;512
48.6;References;512

Leseproben

Biological inspired algorithm for Storage Area Networks (ACOSAN) (p. 109-110)

Anabel Fraga Vazquez'
1 Universidad Carlos III de Madrid
Av. Universidad, 30, Leganes, Madrid, SPAIN [e-mail]

Abstract. The routing algorithms like Storage Area Networks (SAN) algorithms are actually deterministic algorithms, but they may become heuristics or probabilistic just because of applying biological inspired algorithms like Ant Colony Optimization (ACO) of Dorigo. A variant suggested by Navarro and Sinclair in the University of Essex in UK, it is called MACO and it may open new paths for adapting routing algorithms to changes in the environment of any network. A new algorithm is anticipated in this paper to be applied in routing algorithms for SAN Fibre Channel switches, it is called ACOSAN.

1 INTRODUCTION

This paper helps to create new paths for betters routing algorithms based in biological inspired algorithms like ACO (Ant Colony Optimization) of Dorigo [5,18,19]. The base of the paper is to apply these kind of algorithms and variants of that in Fibre Channel's switches (FC), for package routing in an adaptive way. In particular the Multiple Ant Colony Optimization of Navarro and Sinclair and ANTNET of Dorigo are very useful in that matter [11,13,18].

The reminder of this section establishes the history and introduction to the SAN networks, and benefits of this technology. Section 2 explains some key concepts and problems of Fibre Channel networks. Section 3 surveys some related previous work applying ACO to networking problems. Sections 4 describe basis for the algorithms proposed and the algorithm itself. And the final sections cover acknowledgments, future work and conclusions.

1.1. History and definitions

The increasing need year by year to connect disks to computers by SCSI connectors, which is a standard in the eighties in parallel connections, but not so fast as expected because of the problem that parallel connection have in front of serial connections.

Day by day new faster connections are needed and technology does not stop and must not. The Fibre Channel technology is a prove of advance, it starts in the nineties and it has an appreciated speed which is over Gigabits, serial connection, and allows large distances over 10 kilometers.

The external storage is a new discovers, disks not connected anymore point-topoint to servers. Large storage arrays of disks now are the centre of the external storage, it may content five disks or even thousands of disks depending on the size of the company and the data to be storage, terabytes of information are connected by Fibre Channel to servers.

Brocade, an enterprise recognized in the area of Fibre Channel' s switches, is at the moment one of the factories for Fabric topologies in Storage Area Networks (SAN). This company defines SAN as a network for storage and system components, which are all communicated in a Fibre Channel net, used to consolidate and share information, offering high performance links, high availability links, higher speed backups, and support for clustering servers.

1.2. Providers

The main providers of SAN and Fibre Channel spares are placed in Table I. McData is the growing company in the area, followed by Brocade. McData works with IBM and Brocade works with Hewlett Packard (HP). These alliances are strategic for both companies in order to provide a whole package for the customers.