AN ONTOLOGY DRIVEN E-LEARNING AGENT FOR SOFTWARE

Document Sample
AN ONTOLOGY DRIVEN E-LEARNING AGENT FOR SOFTWARE Powered By Docstoc
					INTERNATIONAL JOURNAL Of ACADEMIC RESEARCH                                              Vol. 3. No. 2. March, 2011, Part I


                    AN ONTOLOGY DRIVEN E-LEARNING AGENT
                       FOR SOFTWARE RISK MANAGEMENT
                                                                 1          2
                                             C.R. Rene Robin , G.V.Uma
                                        1                    2
                                       Research Scholar, Professor & Head,
             Department of Information Science and Technology, Anna University, Tamil Nadu, (INDIA)
                             E-mails: crrenerobin@gmail.com, gvuma@annauniv.edu

         ABSTRACT

         The Ontology based e-learning system provides a critical support mechanism for educational institutions.
This paper delineates ontology is used in e-learning for the organization of teaching materials and semantically
defines concepts, properties and relations, thereby improving the quality of teaching resources. It also provides
semantic search of key terms which is more effective than keyword-based search. This paper will address
theoretical knowledge of the area of interest and related issues of other areas of the domain, examples of usage of
the theoretical knowledge in different contexts of explanations and exercises for all three kinds of educational
objectives such as Knowledge, Comprehension and Application. In this paper the authors focused on the design
and development of ontology-based e-learning system and for the evaluation of the developed system, an ontology
for software risk management (SRMONTO) has been constructed. In this paper, the authors made better use of
SRMONTO for the proposed e-learning system that enables to provide learners with a more effective education
support.

       Key words: Ontology, E-Learning, Software Risk Management, SRMONTO, OWL (Web Ontology
Language), Protege

         1. INTRODUCTION

           The concept of reusing learning objects became very common among teachers of Higher Education. In
our survey with different categories of teachers among various Engineering Colleges affiliated by Anna University,
Chennai, most of them look for learning objects to reuse in their teaching and frequently use search engines to
identify suitable resources on the Web to create new, tailor-made resources. Even though the need is huge, it has
also been observed that a few public-spirited academics make their teaching materials open, by publishing them on
their college web site, and these materials are often much downloaded from outside the institution. That fact is
many teachers are hesitant or unwilling to make their materials open due to so many reasons like quality
judgments of their materials and possible copyright claims against embedded content that they have downloaded
and reused in their resources without the specific permission of the owner or the real publisher.
           The above observation indicates that the colleges and universities are not responding fast enough to the
business and technology changes that have redefined the role of information systems in today's organizations. In
the same time awareness has also been increased to provide the right type of education for future information
systems professionals through effective educational tools. It is very clear that providing semantic-rich e-learning
environments is one essential issue in current computer-based education. The term ’e-learning’ is currently very
used and refers to various notions such as logistic (administrative management), resources (course broadcasting)
or technology (virtual conference tools). Numerous definitions of e-learning have been proposed. They usually put
the emphasis on network using (which explains the ”e” in e-learning) and on Information Technology [Amal Zouaq,
2008].
           E-learning is more and more popularization in many high school educations. It has become an important
teaching model. In developed nations like United States of Americal, United Kingdom, and few European countries,
many educational institutions and organizations are adopting e-learning. Many people recognize that e-learning
can overcome the obstacle of geographical location and time, promote knowledge, skills learning and
personalization. It also offers learning-on-demand and reduces the costs and time of learning [1]. E-learning is
currently a multi-billion dollar worldwide, but still is in its nascent stage in India. It has high scope for development,
since youngsters are aware of the technology and tend to prefer the Internet to books. E-learning is not merely
converting printed pages of the textbook on to the web but uses databases to store course content and GUI for
effective interaction with the learner. Our proposed ontology based e-learning agent could use its ontology
i.e.SRMONTO : An education Ontology for Software Risk Management to enrich the presentation of the resulting
list to the end user, e.g. by replacing the endless list of hits with a navigation structure based on the semantics of
the hits.

         2. RELATED WORK

         Nowadays the popularity of the web encourages the development of hypermedia Systems dedicated to e-
learning. The Web frees the teacher and student of restrictions like space and time while providing a powerful
vehicle to disseminate knowledge. Ontology is needed to share a common understanding of the learning among
learners and software agents, enable re-use of and analysis of the domain knowledge and to separate domain
knowledge from operational knowledge.


30 | www.ijar.lit.az
INTERNATIONAL JOURNAL Of ACADEMIC RESEARCH                                                Vol. 3. No. 2. March, 2011, Part I

           Accoding to Boyce, S., & Pahl, C. (2007) Ontologies have the potential to play an important role in
instructional design and the development of course content. They can be used to represent knowledge about
content, supporting instructors in creating content or learners in accessing content in a knowledge-guided way.
They have concluded that while ontologies exist for many subject domains, their quality and suitability for the
educational context might be unclear, for numerous subjects, ontologies do not exist and presented a method for
domain experts rather than ontology engineers to develop ontologies for use in the delivery of courseware content.
           An Effective information services [Yoshihito Takahashi, et. al., 2005] has be provided for users by
investing meta data in information sources and making software agents understand it. The learning domain is
cryptography in network security. They defined various concepts in the cryptographic domain and the relations
among them as the ontology, and proposed a way of utilizing it in the learning process. The strength of their
ontology-based e-learning system is demonstrated through application examples of the prototype system they have
developed.
           The construction of ontology-base teaching knowledge base can greatly improve the quality of teaching
resources. It resolves the problem that traditional teaching resources neglect to semantic and knowledge concepts.
E-learning has become an important teaching model. Many educational institutions and organizations are adopting
e-learning. More and more people recognize that e-learning can overcome the obstacle of geographical location
and time, promote knowledge, skills learning and personalization. It also offers learning-on-demand and reduces
the costs and time of learning [Zhang D. et. al., 2003]. E-learning is currently a multi-billion dollar worldwide, but still
is in its nascent stage in India. It has high scope for development, since youngsters are aware of the technology
and tend to prefer the Internet to books.
           Finally it is understood that, e-learning is not merely converting printed pages of the textbook on to the
web but uses databases to store course content and GUI for effective interaction with the learner. By considering
all the points discussed above an ontology based e-learning system has been proposed, implemented and
evaluated by SRMONTO.

         3. PROPOSED SYSTEM ARCHITECTURE

         The intelligent ontology based e-learning system is developed as a web service, containing the following
four layers. Figure 1 shows the architecture of the proposed system.

         3.1. Interface Layer
         The interface layer contains the following three application-independent components:

          3.1.1. Representation of Educational Resources
          Java Server Pages are used to represent educational resources in a learner-friendly manner. The list of
topics available for study under the particular learning domain is displayed and the learner can either select a
particular chapter or follow the sequence prescribed. The system keeps track of all the chapters completed by the
learner and displays it to the learner for their reference.

        3.1.2. Knowledge Diagnosis Tools
        Knowledge diagnosis is performed on the newly registered learner to determine his/her knowledge level.
The system presents the learner with a set of predetermined questions of varying difficulty. Based on the
performance of the learner, the system predicts the performance of the learner using a bayesian approach
[Francesco Colace, et. al., 2010].

         3.1.3. Learner Profile Manager
         The learner profile contains personal and professional information about the learner. The learner can view
the information of other learners in the system and update his/her profile, if needed.




                                                                                     B a k u , A z e r b a i j a n | 31
INTERNATIONAL JOURNAL Of ACADEMIC RESEARCH                                           Vol. 3. No. 2. March, 2011, Part I




                         Fig. 1. 4-tire architecture for Ontology based E-Learning System

         3.2. Intelligent Layer

          3.2.1. Support for Study Process
          The subsystem for supporting the study process contains features for annotation and bookmarking certain
topics that the learner may peruse later. When the learner finishes studying a particular topic, he may mark the
topic as ‘finished’ for his future reference.

         3.2.2. Search Logistics
         The learner can search for a particular topic which may or may not be present in the local ontology. The
subsystem of search logistics initially searches for the requested topic in the local ontology. Since it retrieves
content based on semantics rather than keywords, it ensures high precision and recall rate. In case the topic
searched by the learner is unavailable in the local ontology, the subsystem transfers control to the module for
querying remote knowledge bases and updating the system knowledge accordingly. The system keeps track of the
topics searched by various learners, and if a particular topic receives the majority of hits, the ontology manager is
informed to update the local ontology with this information.




                                  Fig. 2. Data Flow Diagram of Deeper Knowledge Layer



32 | www.ijar.lit.az
INTERNATIONAL JOURNAL Of ACADEMIC RESEARCH                                                Vol. 3. No. 2. March, 2011, Part I

          3.2.3. Feedback Generation
          The system is capable of providing intelligent feedback testing, which presents questions dynamically on
the basis of the performance of the learner in the previous questions. The learner is initially presented with a
question of medium-level difficulty. If the question is answered right, then the learner is presented with a question
of high-level difficulty. If not, then the learner is presented with a question of low-level difficulty. Thus the testing
subsystem adapts to the knowledge level of the learner.
          The test results are then displayed to the learner as a table or a bar chart and updated in his/her profile.

          3.3. Learning Management System Layer
          The Learning Management System layer contains the learner’s profiles and learning objects consisting of
the syllabus and courseware. The syllabus is structured based on the underlying ontology to provide effective
retrieval of semantically-related topics. Questions, classified on the basis of their difficulty are also stored.

         3.4. Deeper Knowledge Layer
         The deeper knowledge layer contains the following components:

        3.4.1. Local ontology
        The local ontology contains information about the domain which is required for providing learning services.
The local ontology is integrated with the system by the developer and periodically updated by the ontology
manager.

          3.4.2. Ontology Manager
          The ontology manager keeps track of the various topics searched by the different learners and also
current trends in the learning domain. Updation of the local ontology can be performed by the Ontology Manager
using the Jena Framework.

          3.4.3. Jena Framework
          The Jena framework can be integrated with a Java-based application to access the ontology using the
built-in OWL (Web Ontology Language) API.

      3.4.4. Core
      Core is the component that enables generation of SPARQL [E. Prud’hommeaux, et. al., 2008] queries.
SPARQL is an RDF query language that can be used to query a remote knowledge base.

        3.4.5. Remote knowledge base
        Remote knowledge bases may be similar to DBPedia that contains structured information extracted from
Wikipedia.

         4. METHODOLOGIES USED

        The system is developed using J2EE technology. The user interface is developed using Java Servlets and
JSP technology; database access is provided using SQL Server; ontology is accessed using JAXP (Java for XML
Processing) API, and ontology is developed using Protégé [Horridge, M., 2004] software tool.

         5. SOFTWARE RISK MANAGEMENT

           Software Risk Management [Barry W. Boehm, 1993] is a discipline whose objectives are to identify,
address and eliminate software risk items before they become either threats to successful software operation or
major sources of expensive software rework. In other words, software risk management may be defined as the
activity that identifies a risk, assesses the risk and defines the policies or strategies to alleviate or lessen the risk. It
should be a continuous process of systematically deciding cost effective approaches for minimizing the outcome of
threat realization to the organization.The authors have taken risk management paradigm introduced by Software
Engineering Institute as our standard to construct the sub ontologies for the target ontology i.e. software risk
management ontology. Risk is omnipresent in each and every step of the software development and all the
interactions that software developers carry out. Software development project risk management [Karolak D
W,1996][Fairley, R. (1994)][ Myerson M., 1996] have focused on reduction and prevention of risks, continuously
assess possible problems, define potential risks, determine what risks are important and deal with them. So a
whole project picture is required for successful risk management.

         6. SRMONTO: SOFTWARE RISK MANAGEMENT ONTOLOGY

          Any e-learning application needs the required knowledge to be represented in a good and enough way to
help its users. Without a structural model for a domain, target knowledge will be unstructured pieces of text, which
are difficult to use in applications like knowledge management and e-learning systems. The knowledge domain in
an e-learning application can be structured in many ways: dictionaries, thesaurus, book summary, library catalog,
indexes and metadata, knowledge graphs, ontologies, etc. The tree organization of a knowledge domain is an
important property that can significantly reduce the processing, but it is insufficient to describe the rich network of


                                                                                      B a k u , A z e r b a i j a n | 33
INTERNATIONAL JOURNAL Of ACADEMIC RESEARCH                                             Vol. 3. No. 2. March, 2011, Part I


relations that ties the concept structures. It needs to be complemented with relations between concepts and axioms
in order to sustain more refined mechanism of conceptual matching and inference. With the advancement of
artificial intelligence technologies, ontology technologies enable a linguistic infrastructure to represent relationships
between concepts of a domain and ontology technology is considered to be a highly suitable means of supporting
educational-technology systems [Mizoguchi & Bourdeau, 2000]. Drawing on the previous work [Marvin J. Carr,
1993] and the experience gained in numerous projects, we now present an ontology for software risk identification
that combines the concepts of knowledge, semantic description of each concepts, and properties. It provides ways
to annotate semantically resources in e-learning and knowledge management environments, in particular to define
semantics of individual concepts, prerequisites and goals for activities and resource content, for e-learning and
knowledge management applications.
            The required concepts, the semantic description of the concepts and the interrelationship among the
concepts along with all other ontological components have been collected from various literatures and experience
of the people from software industry. From which, a taxonomy has been constructed by using the property ‘isA’ and
the design architecture for the required ontology has also been sketched out manually with nearly four different
types of properties. In order to reduce implementation efforts, the Protégé platform a scalable and integrated
framework for ontological engineering, has been used to construct the ontology. The constructed ontology has
been represented in owl format, which makes it more machine understandable. Then the semantic representation
of the knowledge [C.R.Rene Robin et.al., 2010] has been made using the OWL document generator, which
automatically generates a set of documents from the ontology. In order to understand the knowledge in more
detailed way again the ontology has been visualized using ontoviz tool. This all about SRMONTO and it has been
constructed specifically for the proposed ontology based e-learning application.


              Risk Control                                                                    Risk Tacking
                                                            SRM


                                     Risk                                    Risk Analysis
           N(C)=40
           N(P)=04               Identification                                                          N(C)=12
                                                                                                         N(P)=04

                                                        Risk Planning
                                                                                   N(C)=48
                                 N(C)=84                                           N(P)=06
                                 N(P) =04

                                                         N(C) =16
                                                         N(P) =04



                                           Fig. 3. The Structure of SRMONTO

         Figure 1 demonstrates five top levels of the developed ontology i.e. SRMONTO. Lower levels trivially
expand the hierarchy therefore we have hidden them. It is an integration of Risk Identification Ontology, Risk
Analysis Ontology, Risk Planning Ontology, Risk Control Ontology and Risk Tracking Ontology where N(C) is the
number of concepts and N(P) is the number of relationships each ontology has.

         7. PROPOSED SYLLABUS FOR SOFTWARE RISK MANAGEMENT

         Educators, for the most part, concur with this vision but have reacted slowly in implementing required
curriculum changes. Bridging the gap between what is practitioners expect of graduates and what graduates have
learned will require a fresh look at the IS curriculum. A key concern of curriculum designers is striking the right
balance between technical and business knowledge. Figure 4 shows the curriculum developed for software risk
management.
       1. Introduction to Software Risk Management
              a. Software Risk Items
              b. Who should participate in Risk Management Process?
              c. Types of Software Risk Management
                                  i. Proactive
                                 ii. Reactive
              d. Software Risk Management Principles
              e. Software Risk Management Process
                               i.    Identification
                              ii.    Analysis and Prioritization
                             iii.    Plan and Schedule
                             iv.     Track and Report
                              v.     Control
              f. Summary




34 | www.ijar.lit.az
INTERNATIONAL JOURNAL Of ACADEMIC RESEARCH                                          Vol. 3. No. 2. March, 2011, Part I

     2. Risk Identification
           a. Where can risk occur?
                             i.  Development Environment
                            ii.  Product Engineering
                           iii.  Program Constraints
           b. Risk Identification Techniques
                          i.     Affinity Grouping Technique
                         ii.     Checklists
                        iii.     Decision Driver Analysis
                       iv.       Assumption Analysis
           c. Risk Identification Guidelines
           d. Summary

       3. Risk Analysis
            a. Categories of Risk
                        i.      Schedule Risk
                       ii.      Budget Risk
                      iii.      Operational Risk
                      iv.       Technical Risk
                       v.       Programmatic Risk
            b. Risk Analysis Techniques
                        i.      Qualitative Techniques
       1. Hazard and Operability Study
       2. Preliminary Risk Analysis
       3. Failure Mode and Effect Analysis
                       ii.      Techniques for Dynamic Systems
       1. Go Method
       2. Digraph/Fault Graph
       3. Markov Modelling
       4. Dynamic Event Logic Analytical Methodology
       5. Dynamic Event Tree Analysis Model
                      iii.      Tree-based Techniques
       1. Fault Tree Analysis
       2. Event Tree Analysis
       3. Cause Consequence Analysis
       4. Management Oversight Risk Tree
       5. Safety Management Organization Review Technique

            c. Summary

       4. Risk Planning
           a. Implementation Factors
                       i.   Availability of Staff
                      ii.   Funds
                     iii.   Impact of Risk
                     iv.    Probability of Risk Occurance
           b. Planning Process
                       i.   Develop Plan
                      ii.   Explore Alternatives
                     iii.   Review Recommendations
           c. Risk Management Strategies
                       i.   Avoid the risk
                      ii.   Accept the risk
                     iii.   Mitigate the risk
                     iv.    Transfer the risk
           d. Case Study
           e. Summary

       5. Risk Tracking
           a. Risk Manager
           b. Risk Status Reporting
           c. Risk Tracking Techniques
                        i.    Risk Matrix
                       ii.    Risk Scale
                      iii.    Risk Assessment
                              1. Identify the hazards
                              2. Decide who might be harmed and how
                              3. Evaluate the risks and decide on precautions
                              4. Record the findings and implement them
                              5. Review the assessment and update
                     iv.      Milestone Tracking
                       v.     Summary
       6. Risk Control
           a. Risk Control Strategies
                        i.    Acceptance
                       ii.    Avoidance


                                                                                B a k u , A z e r b a i j a n | 35
INTERNATIONAL JOURNAL Of ACADEMIC RESEARCH                                         Vol. 3. No. 2. March, 2011, Part I


                          iii.    Contingency
                         iv.      Reduction
                           v.     Transference
             b.    Risk Control Process
                            i.    Prioritize actions
                           ii.    Evaluate Recommended Control Options
                          iii.    Conduct Cost-Benefit Analysis
                         iv.      Select Control
                           v.     Assign Responsibility
                         vi.      Develop a Safeguard-Implementation Plan
                        vii.      Implement Selected Control
             c.    Risk Control Methods
                            i.    Technical Methods
                               1. Support
                               2. Prevent
                               3. Detect and Recover
                        ii.       Non-technical Methods
                               1. Management Methods
                               2. Operational Methods

                             Fig. 4. Proposed Curriculum for Software Risk Management

        8. CONCLUSION AND FURTHER ENHANCEMENTS

         In this paper, a novel method for learning software risk management using an ontology based e-learning
system has been presented. The ontology developed for the system has been effectively represented by OWL
format and it has also been represented by semantically and visually. In order to eliminate the discrepancies, the
curriculum for the software risk management has been developed. In future, a discussion forum may be developed
for enabling interaction between students. Updating the Syllabus may be automated based on changes in the
ontology.

        REFERENCES

             1.    Amal Zouaq and Roger Nkambou, “Building Domain Ontologies from Text for Educational
                   Purposes”, IEEE Transactions on Learning Technologies, VOL. 1, No. 1, January-March 2008,
             2.    Barry W. Boehm, Software Risk Management, IEEE Computer Society Press, 1993,
             3.    Boyce, S., & Pahl, C. (2007). Developing Domain Ontologies for Course Content. Educational
                   Technology & Society, 10 (3), 275-288.Marie-H´el`ene Abel Dominique Lenne Claude Moulin,
                   Using two ontologies to index e-learning resources, Proceedings of the IEEE/WIC/ACM
                   International Conference on Web Intelligence (WI’04), 0-7695-2100-2/04.
             4.    Fairley, R. (1994) Risk management for software projects. IEEE Software 11:3 pp 57-67.
             5.    Francesco Colace, Member, IEEE, and Massimo De Santo, Member, IEEE, Ontology for E-
                   Learning: A Bayesian Approach, IEEE Transactions on Education, Vol. 53, No. 2, May 2010.
             6.    Horridge, M., H. Knublauch, A. Rector, R. Stevens, C. Wroe.: A Practical Guide To Building OWL
                   Ontologies Using the Protégé-OWL Plugin and CO-ODE Tools, Technical Report, Ed. 1.0, The
                   University Of Manchester (2004).
             7.    Karolak D W (1996) Software engineering risk management. IEEE Computer Society Press
             8.    Marvin. J.Carr, S.L.Konda, I.Monarch, F.C.Ulrich, C.F.Walker, “Taxonomy-Based Risk
                   Identification”, Software Engineering Institute Technical Report, CMU/SEI-93-TR-6, Carnegie
                   Mellon University, Pittsburgh, Pennsylvania, 1993.
             9.    Mizoguchi R. & Bourdeau, J(2000) Using ontological engineering to overcome common AI-ED
                   problems. International Journal of Artificial Intelligence in Education, 11(2), 107-121.
             10.   Myerson M. (1996) Risk Management Processes for Software Engineering Models. Artech
                   House
             11.   E.Prud’hommeaux and A.Seaborne. SPARQL Query Language for RDF.
                   http://www.w3.org/TR/2008/REC-rdf-sparql-query-20080115/, January 2008.
             12.   C.R.Rene Robin and G.V. Uma, Ontology Based Semantic Knowledge Representation for
                   Software Risk Management, International Journal of Engineering Science and Technology, Vol.
                   2(10), 2010, 5611-5617.
             13.   Zhang D. and Nunamaker J. F. 2003, “Powering E-Learning In the New Millennium: an Overview
                   of E-Learning and Enabling Technology”, Information Systems Frontiers, vol. 5, no. 2, 2003, pp.
                   207-218.
             14.   Yoshihito Takahashi, Tomomi Abiko et.al., An Ontology-Based e-Learning System for Network
                   Security, 19th International Conference on Advanced Information Networking and Applications
                   (AINA'05), Taipei, Taiwan, March 25-March 30, 2005 Volume 1 ISBN: 0-7695-2249-1.




36 | www.ijar.lit.az

				
DOCUMENT INFO
Shared By:
Tags:
Stats:
views:41
posted:11/27/2011
language:English
pages:7