• Information Technology
  • Human Resources
  • Careers
  • Giving
  • Library
  • Bookstore
  • Active Living
  • Continuing Education
  • Go Dinos
  • UCalgary Maps
  • UCalgary Directory
  • Academic Calendar
My UCalgary
Webmail
D2L
ARCHIBUS
IRISS
  • Faculty of Arts
  • Cumming School of Medicine
  • Faculty of Environmental Design
  • Faculty of Graduate Studies
  • Haskayne School of Business
  • Faculty of Kinesiology
  • Faculty of Law
  • Faculty of Nursing
  • Faculty of Nursing (Qatar)
  • Schulich School of Engineering
  • Faculty of Science
  • Faculty of Social Work
  • Faculty of Veterinary Medicine
  • Werklund School of Education
  • Information TechnologiesIT
  • Human ResourcesHR
  • Careers
  • Giving
  • Library
  • Bookstore
  • Active Living
  • Continuing Education
  • Go Dinos
  • UCalgary Maps
  • UCalgary Directory
  • Academic Calendar
  • Libraries and Cultural Resources
View Item 
  •   PRISM Home
  • Graduate Studies
  • The Vault: Electronic Theses and Dissertations
  • View Item
  •   PRISM Home
  • Graduate Studies
  • The Vault: Electronic Theses and Dissertations
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Concept Learning Supported Semantic Search Using Multi-Agent System Based on Social Networks

Thumbnail
View
ucalgary_2014_ElSherif_Shimaa.pdf
Download
ucalgary_2014_ElSherif_Shimaa.pdf (3.104Mb)
Advisor
Far, Behrouz
Eberlein, Armin
Author
El_sherif, Shimaa Mansour Moailak
Accessioned
2014-04-17T17:20:33Z
Available
2014-06-16T07:00:30Z
Issued
2014-04-17
Submitted
2014
Other
Multi-Agent System
Social Networks
Concept Learning
Subject
Artificial Intelligence
Computer Science
Engineering--Electronics and Electrical
Type
Thesis
Metadata
Show full item record

Abstract
In this research, we propose an agent-based semantic search system supported by ontological concept learning and contents annotation. Our system consists of a group of multi-agent systems (MASs), each controlling a repository of structured and unstructured documents. Each repository has its own concept hierarchy, i.e. ontology. . Each MAS consists of a group of agents, each with its own responsibilities and an assigned tasks to perform. All agents in each MAS cooperate with each other to perform more general tasks. Agents of different MASs communicate with each other by developing a common understanding of concepts used during communication. Agents communicate with each other via a social network. Strengths of ties between agents in the social network represent how close/far two agents are to each other. In our system, there are two major modules: a semantic search module and a concept learning module. In the semantic search module, MASs cooperate with each other to perform semantic search and return results back to the user. The second major module in our system is the concept learning module. During the semantic search process, a MAS may discover that it needs to learn new concepts. The Concept Learning module helps that MAS to learn the new concepts from other MASs. A social network is used in communication between agents from different MASs. In this research, we define two case studies to test the system. These case studies evaluate the efficiency of using social networks in representing relationships between agents in different MASs in learning new concepts from several teachers. We also introduce a novel approach of calculating tie strengths in social networks using Hidden Markov Model (HMM). The results obtained show that using social networks in communicating between agents in different MASs has a positive effect in our system. During leaning new concepts, using social networks between the learner and the teachers gives betters accuracies in all concepts learnt and with different machine learning techniques used. On the other hand, using social networks decrease the negative effect of increasing number of teachers in the concept learning process.
Corporate
University of Calgary
Faculty
Graduate Studies
Doi
http://dx.doi.org/10.5072/PRISM/25406
Uri
http://hdl.handle.net/11023/1419
Collections
  • The Vault: Electronic Theses and Dissertations

Browse

All of PRISMCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

My Account

LoginRegister

Statistics

Most Popular ItemsStatistics by CountryMost Popular Authors

  • Email
  • SMS
  • 403.220.8895
  • Live Chat

Energize: The Campaign for Eyes High

Privacy Policy
Website feedback

University of Calgary
2500 University Drive NW
Calgary, AB T2N 1N4
CANADA

Copyright © 2017