SlideShare ist ein Scribd-Unternehmen logo
1 von 16
An Approach for the Qualitative
Analysis of Open Agent Conversations
Emilio Serrano1, Michael Rovatos2 and Juan A. Botía1
Contact: emilioserra@um.es, michael.rovatsos@ed.ac.uk, juanbot@um.es
1. University of Murcia, Spain / 2. University of Edinburgh, U.K.
Presented in ITMAS 2012,Third InternationalWorkshop on Infrastructures
andTools for Multiagent Systems
1
Content
 Introduction
 Protocols, context, and context models
 Preparing the training data set
 ProtocolMiner
 Case study
 Conclusion and future work
2
Introduction
 Interaction in MASs is essential
 Analysis tools
◦ Analysis of agents' mental states
 Implementation available
◦ Analysis of the interactions among agents
 Fixed syntactic elements
 Quantitative analysis (QuanA)
◦ What a QuanA can offer?
 E.g FIPA contract-net
3
Introduction II
 Qualitative Analysis (QualA)
◦ Generalizing individual cases to build a general
theory
 Context models.
◦ Correlating the status of logical constraints to
perceived agent behaviour
 Contribution: definition, construction, and use (+tool)
◦ Utility
 to make predictions about future behaviour
 to infer the definitions other agents
 to analyse the reliability and trustworthiness
4
Protocols, context, and context models
Context:
Negotiation protocol example:
Messages:
Performative(sender,receiver, content)
Constraints:
Nameevaluator(parameters)
Context + data mining = context model 5
Preparing the training data set
 How exactly training data is constructed?
◦ Dealing with different agents
 An agent only can assure its own context
 Most cautious strategy
 Most trusting: entire path information
◦ Dealing with different paths
 Set of variables contained in these may differ
 Create a different data sets
 Merge data across different paths
 Samples can be “stuffed‘” with “unknown”.
 Path group label: success
◦ Dealing with loops
 Variables used in the loop can have several constants
 N “copies” of each variable
 First/last ground term
6
ProtocolMiner
 Plugin for the OpenKnowledge platform
 Comprehensive functionality for QalA
◦ For human designer or agents
◦ Automates the construction of context models
◦ Definition of protocols by LCC
 Lightweight Coordination Calculus
 E.g. participant role in FIPA Request
 other interaction platforms may be used…
7
ProtocolMiner II
8
Protocol Miner III
 Context model of a single constraint,
acceptable
 Weka algorithms integrated
9
Case study
 Car selling domain
 A requests B offers forT and P
 10 customer agents with different mental
states
10
Case study II
 Protocol outcome prediction
◦ From the point of view of one seller
◦ Average model accuracy across 100 repeated experiments
◦ Evaluated using cross-validation
◦ Three open source data mining techniques
11
Case study III
12
Case study IV
 Agents predicting interactions
◦ Agents build and use context models to choose a good seller.
◦ Compared to a random strategy and a quantitative strategy:
13
Case studyV
14
Conclusion and future work
 Novel mechanism to exploit qualitative
information
◦ logical constraint as semantic annotations
◦ derive knowledge of the internal workings
◦ ProtocolMiner implements the approach
 Future works:
◦ more real-world examples
◦ more advanced machine learning methods
15
Thanks for your attention
Emilio Serrano1, Michael Rovatos2 and Juan A. Botía1
Contact: emilioserra@um.es, michael.rovatsos@ed.ac.uk, juanbot@um.es
1. University of Murcia, Spain / 2. University of Edinburgh, U.K.
16

Weitere ähnliche Inhalte

Was ist angesagt?

A Vague Sense Classifier for Detecting Vague Definitions in Ontologies
A Vague Sense Classifier for Detecting Vague Definitions in OntologiesA Vague Sense Classifier for Detecting Vague Definitions in Ontologies
A Vague Sense Classifier for Detecting Vague Definitions in Ontologies
Panos Alexopoulos
 
Structured Analysis Proposal for Managing HB 4087 Vendor Responses to RFI
Structured Analysis Proposal for Managing HB 4087 Vendor Responses to RFIStructured Analysis Proposal for Managing HB 4087 Vendor Responses to RFI
Structured Analysis Proposal for Managing HB 4087 Vendor Responses to RFI
Susan Tait, CSM
 

Was ist angesagt? (8)

Detecting, Measuring and Representing Vagueness in Ontologies
Detecting, Measuring and Representing Vagueness in OntologiesDetecting, Measuring and Representing Vagueness in Ontologies
Detecting, Measuring and Representing Vagueness in Ontologies
 
A Vague Sense Classifier for Detecting Vague Definitions in Ontologies
A Vague Sense Classifier for Detecting Vague Definitions in OntologiesA Vague Sense Classifier for Detecting Vague Definitions in Ontologies
A Vague Sense Classifier for Detecting Vague Definitions in Ontologies
 
Vagueness in Semantic Information Management
Vagueness in Semantic Information ManagementVagueness in Semantic Information Management
Vagueness in Semantic Information Management
 
Troubleshooting and Optimizing Named Entity Resolution Systems in the Industry
Troubleshooting and Optimizing Named Entity Resolution Systems in the IndustryTroubleshooting and Optimizing Named Entity Resolution Systems in the Industry
Troubleshooting and Optimizing Named Entity Resolution Systems in the Industry
 
Linking Heterogeneous Scholarly Data Sources in an Interoperable Setting: the...
Linking Heterogeneous Scholarly Data Sources in an Interoperable Setting: the...Linking Heterogeneous Scholarly Data Sources in an Interoperable Setting: the...
Linking Heterogeneous Scholarly Data Sources in an Interoperable Setting: the...
 
Amazon seniment
Amazon senimentAmazon seniment
Amazon seniment
 
Structured Analysis Proposal for Managing HB 4087 Vendor Responses to RFI
Structured Analysis Proposal for Managing HB 4087 Vendor Responses to RFIStructured Analysis Proposal for Managing HB 4087 Vendor Responses to RFI
Structured Analysis Proposal for Managing HB 4087 Vendor Responses to RFI
 
CrowdTruth for medical relation extraction - WAI talk
CrowdTruth for medical relation extraction - WAI talkCrowdTruth for medical relation extraction - WAI talk
CrowdTruth for medical relation extraction - WAI talk
 

Andere mochten auch

Desarrollo y evaluación de sistemas de inteligencia ambiental con UbikSim
Desarrollo y evaluación de sistemas de  inteligencia ambiental con UbikSimDesarrollo y evaluación de sistemas de  inteligencia ambiental con UbikSim
Desarrollo y evaluación de sistemas de inteligencia ambiental con UbikSim
Emilio Serrano
 

Andere mochten auch (11)

Creating and validating emergency management services by social simulation a...
Creating and validating emergency management services  by social simulation a...Creating and validating emergency management services  by social simulation a...
Creating and validating emergency management services by social simulation a...
 
Infrastructure for forensic analysis of multi-agent systems
Infrastructure for forensic analysis  of multi-agent systems Infrastructure for forensic analysis  of multi-agent systems
Infrastructure for forensic analysis of multi-agent systems
 
Casos de sistemas inteligentes (only in Spanish)
Casos de sistemas inteligentes (only in Spanish)Casos de sistemas inteligentes (only in Spanish)
Casos de sistemas inteligentes (only in Spanish)
 
Introducción a la simulación social basada en agentes (only in Spanish)
Introducción a la simulación social basada en agentes (only in Spanish)Introducción a la simulación social basada en agentes (only in Spanish)
Introducción a la simulación social basada en agentes (only in Spanish)
 
Articial societies immersed in an Ambient Intelligence Environment
Articial societies immersed in an Ambient Intelligence EnvironmentArticial societies immersed in an Ambient Intelligence Environment
Articial societies immersed in an Ambient Intelligence Environment
 
Un prototipo para el modelado de un sistema de metaheurísticas cooperativa...
Un prototipo para el modelado de un  sistema  de metaheurísticas  cooperativa...Un prototipo para el modelado de un  sistema  de metaheurísticas  cooperativa...
Un prototipo para el modelado de un sistema de metaheurísticas cooperativa...
 
TOWARDS SOCIO-CHRONOBIOLOGICAL COMPUTATIONAL HUMAN MODELS
TOWARDS SOCIO-CHRONOBIOLOGICAL  COMPUTATIONAL HUMAN MODELS   TOWARDS SOCIO-CHRONOBIOLOGICAL  COMPUTATIONAL HUMAN MODELS
TOWARDS SOCIO-CHRONOBIOLOGICAL COMPUTATIONAL HUMAN MODELS
 
Developing social simulations with UbikSim
Developing social simulations with UbikSimDeveloping social simulations with UbikSim
Developing social simulations with UbikSim
 
An Explanation-Based Alert Management Tool for Basic AAL Systems
An Explanation-Based Alert Management Tool for Basic AAL SystemsAn Explanation-Based Alert Management Tool for Basic AAL Systems
An Explanation-Based Alert Management Tool for Basic AAL Systems
 
Desarrollo y evaluación de sistemas de inteligencia ambiental con UbikSim
Desarrollo y evaluación de sistemas de  inteligencia ambiental con UbikSimDesarrollo y evaluación de sistemas de  inteligencia ambiental con UbikSim
Desarrollo y evaluación de sistemas de inteligencia ambiental con UbikSim
 
Investigaciones y análisis de redes sociales (only in Spanish)
Investigaciones y análisis de redes sociales (only in Spanish)Investigaciones y análisis de redes sociales (only in Spanish)
Investigaciones y análisis de redes sociales (only in Spanish)
 

Ähnlich wie An Approach for the Qualitative Analysis of Open Agent Conversations

Software Architects’ Experiences of Quality Requirements: What we Know and ...
Software Architects’ Experiences  of Quality Requirements:  What we Know and ...Software Architects’ Experiences  of Quality Requirements:  What we Know and ...
Software Architects’ Experiences of Quality Requirements: What we Know and ...
Luigi Buglione
 

Ähnlich wie An Approach for the Qualitative Analysis of Open Agent Conversations (20)

Requirements document for big data use cases
Requirements document for big data use casesRequirements document for big data use cases
Requirements document for big data use cases
 
A Context-aware Model for the Analysis of User Interaction and QoE in Mobile ...
A Context-aware Model for the Analysis of User Interaction and QoE in Mobile ...A Context-aware Model for the Analysis of User Interaction and QoE in Mobile ...
A Context-aware Model for the Analysis of User Interaction and QoE in Mobile ...
 
Comparative study of the computer tools functions...
Comparative study of the computer tools functions...Comparative study of the computer tools functions...
Comparative study of the computer tools functions...
 
ST_Short_Story.pptx
ST_Short_Story.pptxST_Short_Story.pptx
ST_Short_Story.pptx
 
Ieee doctoral progarm final
Ieee doctoral progarm finalIeee doctoral progarm final
Ieee doctoral progarm final
 
Software Architects’ Experiences of Quality Requirements: What we Know and ...
Software Architects’ Experiences  of Quality Requirements:  What we Know and ...Software Architects’ Experiences  of Quality Requirements:  What we Know and ...
Software Architects’ Experiences of Quality Requirements: What we Know and ...
 
Taming AI Engineering Ethics and Policy
Taming AI Engineering Ethics and PolicyTaming AI Engineering Ethics and Policy
Taming AI Engineering Ethics and Policy
 
FORMAL MODELING AND VERIFICATION OF MULTI-AGENTS SYSTEM USING WELLFORMED NETS
FORMAL MODELING AND VERIFICATION OF MULTI-AGENTS SYSTEM USING WELLFORMED NETSFORMAL MODELING AND VERIFICATION OF MULTI-AGENTS SYSTEM USING WELLFORMED NETS
FORMAL MODELING AND VERIFICATION OF MULTI-AGENTS SYSTEM USING WELLFORMED NETS
 
Smart Cities UK 2018- Stream 1 Data
Smart Cities UK 2018- Stream 1 DataSmart Cities UK 2018- Stream 1 Data
Smart Cities UK 2018- Stream 1 Data
 
Compliance driven process development with DCR graphs
Compliance driven process development with DCR graphsCompliance driven process development with DCR graphs
Compliance driven process development with DCR graphs
 
A Study On Hybrid System
A Study On Hybrid SystemA Study On Hybrid System
A Study On Hybrid System
 
Privacy Requirements Engineering in Agile Software Development
Privacy Requirements Engineering in Agile Software DevelopmentPrivacy Requirements Engineering in Agile Software Development
Privacy Requirements Engineering in Agile Software Development
 
thesis
thesisthesis
thesis
 
Performance Evaluation of Software Quality Model
Performance Evaluation of Software Quality ModelPerformance Evaluation of Software Quality Model
Performance Evaluation of Software Quality Model
 
In Processes We Trust: Privacy and Trust in Business Processes
In Processes We Trust: Privacy and Trust in Business ProcessesIn Processes We Trust: Privacy and Trust in Business Processes
In Processes We Trust: Privacy and Trust in Business Processes
 
Freenome's Biological Machine Learning Platform
Freenome's Biological Machine Learning PlatformFreenome's Biological Machine Learning Platform
Freenome's Biological Machine Learning Platform
 
The prediction of Bitcoin price through gold price using long short-term memo...
The prediction of Bitcoin price through gold price using long short-term memo...The prediction of Bitcoin price through gold price using long short-term memo...
The prediction of Bitcoin price through gold price using long short-term memo...
 
Process Mining for Process Improvement.pptx
Process Mining for Process Improvement.pptxProcess Mining for Process Improvement.pptx
Process Mining for Process Improvement.pptx
 
Stock Market Analysis and Prediction (1) (2).pdf
Stock Market Analysis and Prediction (1) (2).pdfStock Market Analysis and Prediction (1) (2).pdf
Stock Market Analysis and Prediction (1) (2).pdf
 
A context-aware model for QoE analysis in mobile environments
A context-aware model for QoE analysis in mobile environmentsA context-aware model for QoE analysis in mobile environments
A context-aware model for QoE analysis in mobile environments
 

Kürzlich hochgeladen

Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Sérgio Sacani
 
Seismic Method Estimate velocity from seismic data.pptx
Seismic Method Estimate velocity from seismic  data.pptxSeismic Method Estimate velocity from seismic  data.pptx
Seismic Method Estimate velocity from seismic data.pptx
AlMamun560346
 
Chemical Tests; flame test, positive and negative ions test Edexcel Internati...
Chemical Tests; flame test, positive and negative ions test Edexcel Internati...Chemical Tests; flame test, positive and negative ions test Edexcel Internati...
Chemical Tests; flame test, positive and negative ions test Edexcel Internati...
ssuser79fe74
 
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 bAsymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Sérgio Sacani
 
Bacterial Identification and Classifications
Bacterial Identification and ClassificationsBacterial Identification and Classifications
Bacterial Identification and Classifications
Areesha Ahmad
 

Kürzlich hochgeladen (20)

PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
 
Creating and Analyzing Definitive Screening Designs
Creating and Analyzing Definitive Screening DesignsCreating and Analyzing Definitive Screening Designs
Creating and Analyzing Definitive Screening Designs
 
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
 
Pulmonary drug delivery system M.pharm -2nd sem P'ceutics
Pulmonary drug delivery system M.pharm -2nd sem P'ceuticsPulmonary drug delivery system M.pharm -2nd sem P'ceutics
Pulmonary drug delivery system M.pharm -2nd sem P'ceutics
 
Zoology 4th semester series (krishna).pdf
Zoology 4th semester series (krishna).pdfZoology 4th semester series (krishna).pdf
Zoology 4th semester series (krishna).pdf
 
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43bNightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
 
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRLKochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
 
GBSN - Biochemistry (Unit 1)
GBSN - Biochemistry (Unit 1)GBSN - Biochemistry (Unit 1)
GBSN - Biochemistry (Unit 1)
 
Forensic Biology & Its biological significance.pdf
Forensic Biology & Its biological significance.pdfForensic Biology & Its biological significance.pdf
Forensic Biology & Its biological significance.pdf
 
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 60009654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000
 
Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...
Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...
Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...
 
COST ESTIMATION FOR A RESEARCH PROJECT.pptx
COST ESTIMATION FOR A RESEARCH PROJECT.pptxCOST ESTIMATION FOR A RESEARCH PROJECT.pptx
COST ESTIMATION FOR A RESEARCH PROJECT.pptx
 
Seismic Method Estimate velocity from seismic data.pptx
Seismic Method Estimate velocity from seismic  data.pptxSeismic Method Estimate velocity from seismic  data.pptx
Seismic Method Estimate velocity from seismic data.pptx
 
Botany krishna series 2nd semester Only Mcq type questions
Botany krishna series 2nd semester Only Mcq type questionsBotany krishna series 2nd semester Only Mcq type questions
Botany krishna series 2nd semester Only Mcq type questions
 
Chemical Tests; flame test, positive and negative ions test Edexcel Internati...
Chemical Tests; flame test, positive and negative ions test Edexcel Internati...Chemical Tests; flame test, positive and negative ions test Edexcel Internati...
Chemical Tests; flame test, positive and negative ions test Edexcel Internati...
 
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 bAsymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
 
Animal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptxAnimal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptx
 
Botany 4th semester file By Sumit Kumar yadav.pdf
Botany 4th semester file By Sumit Kumar yadav.pdfBotany 4th semester file By Sumit Kumar yadav.pdf
Botany 4th semester file By Sumit Kumar yadav.pdf
 
Bacterial Identification and Classifications
Bacterial Identification and ClassificationsBacterial Identification and Classifications
Bacterial Identification and Classifications
 
Recombination DNA Technology (Nucleic Acid Hybridization )
Recombination DNA Technology (Nucleic Acid Hybridization )Recombination DNA Technology (Nucleic Acid Hybridization )
Recombination DNA Technology (Nucleic Acid Hybridization )
 

An Approach for the Qualitative Analysis of Open Agent Conversations

  • 1. An Approach for the Qualitative Analysis of Open Agent Conversations Emilio Serrano1, Michael Rovatos2 and Juan A. Botía1 Contact: emilioserra@um.es, michael.rovatsos@ed.ac.uk, juanbot@um.es 1. University of Murcia, Spain / 2. University of Edinburgh, U.K. Presented in ITMAS 2012,Third InternationalWorkshop on Infrastructures andTools for Multiagent Systems 1
  • 2. Content  Introduction  Protocols, context, and context models  Preparing the training data set  ProtocolMiner  Case study  Conclusion and future work 2
  • 3. Introduction  Interaction in MASs is essential  Analysis tools ◦ Analysis of agents' mental states  Implementation available ◦ Analysis of the interactions among agents  Fixed syntactic elements  Quantitative analysis (QuanA) ◦ What a QuanA can offer?  E.g FIPA contract-net 3
  • 4. Introduction II  Qualitative Analysis (QualA) ◦ Generalizing individual cases to build a general theory  Context models. ◦ Correlating the status of logical constraints to perceived agent behaviour  Contribution: definition, construction, and use (+tool) ◦ Utility  to make predictions about future behaviour  to infer the definitions other agents  to analyse the reliability and trustworthiness 4
  • 5. Protocols, context, and context models Context: Negotiation protocol example: Messages: Performative(sender,receiver, content) Constraints: Nameevaluator(parameters) Context + data mining = context model 5
  • 6. Preparing the training data set  How exactly training data is constructed? ◦ Dealing with different agents  An agent only can assure its own context  Most cautious strategy  Most trusting: entire path information ◦ Dealing with different paths  Set of variables contained in these may differ  Create a different data sets  Merge data across different paths  Samples can be “stuffed‘” with “unknown”.  Path group label: success ◦ Dealing with loops  Variables used in the loop can have several constants  N “copies” of each variable  First/last ground term 6
  • 7. ProtocolMiner  Plugin for the OpenKnowledge platform  Comprehensive functionality for QalA ◦ For human designer or agents ◦ Automates the construction of context models ◦ Definition of protocols by LCC  Lightweight Coordination Calculus  E.g. participant role in FIPA Request  other interaction platforms may be used… 7
  • 9. Protocol Miner III  Context model of a single constraint, acceptable  Weka algorithms integrated 9
  • 10. Case study  Car selling domain  A requests B offers forT and P  10 customer agents with different mental states 10
  • 11. Case study II  Protocol outcome prediction ◦ From the point of view of one seller ◦ Average model accuracy across 100 repeated experiments ◦ Evaluated using cross-validation ◦ Three open source data mining techniques 11
  • 13. Case study IV  Agents predicting interactions ◦ Agents build and use context models to choose a good seller. ◦ Compared to a random strategy and a quantitative strategy: 13
  • 15. Conclusion and future work  Novel mechanism to exploit qualitative information ◦ logical constraint as semantic annotations ◦ derive knowledge of the internal workings ◦ ProtocolMiner implements the approach  Future works: ◦ more real-world examples ◦ more advanced machine learning methods 15
  • 16. Thanks for your attention Emilio Serrano1, Michael Rovatos2 and Juan A. Botía1 Contact: emilioserra@um.es, michael.rovatsos@ed.ac.uk, juanbot@um.es 1. University of Murcia, Spain / 2. University of Edinburgh, U.K. 16