Content uploaded by SMFD Syed Mustapha
Author content
All content in this area was uploaded by SMFD Syed Mustapha on Oct 11, 2016
Content may be subject to copyright.
M.Gh. Negoita et al. (Eds.): KES 2004, LNAI 3213, pp. 343–349, 2004.
© Springer-Verlag Berlin Heidelberg 2004
Intelligent Conversational Channel for Learning
Social Knowledge Among Communities
S.M.F.D. Syed Mustapha
Faculty of Computer Science and Information Technology, University of Malaya,
50603 Kuala Lumpur, Malaysia
smfd@um.edu.my
http://www.perdana.fsktm.um.edu.my/~symalek
Abstract. Recent studies have shown two approaches in building learning sys-
tem. Each corresponds to the two types of knowledge which are the content
knowledge and social knowledge. The former is knowledge about knowing how
to perform a task while the latter is more about best practices. Intelligent Con-
versational Channel (ICC) is built to support for learning social knowledge. In
this paper, the two types of knowledge are explained and how ICC can be used
to support learning among communities.
1 Introduction
There are numerous types of learning system being used as the technology for learning
such as the Intelligent Tutoring System, Computer Aided Learning, Microworld and
Computer-based Learning [4]. Expert system was also used to train general practitio-
ner to be specialist [1]. These systems support learning for a specific domain of
knowledge. In the last decade, study has shown that learning through social process
has becoming an integral learning method besides the conventional self-learning.
Community of Practice is a social learning theory that describes one’s learning
through participation and reification through the community activities. Collaborative
learning supports learning by sharing knowledge through mutual contribution [2].
Social Knowledge-building (SKB) describes a collaborative knowledge building by a
community [3]. Researchers claim that story-telling as an effective mode of knowl-
edge sharing and knowledge transfer [5]. The type of knowledge that suits learning
using this approach is so-called social knowledge.
Our approach to enabling knowledge sharing for social knowledge has three
prongs which are to facilitate communication through virtual community, to analyze
social interaction through discourse analyzer and building social knowledge through
story-telling. Intelligent Conversational Channel (thereafter, ICC) has been developed
that support this approach through three main components which are the Discourse
Communicator, Hyper-media Learning Space and Discourse Analyzer. These three
components are built on a community channel as the main venue for knowledge shar-
ing. In section 2, a descriptive analysis is given to differentiate between content
knowledge and social knowledge, section 3 describes the ICC components and the life
cycle model of social knowledge and section 4 is the conclusion and future work.
344 S.M.F.D. Syed Mustapha
2 What Is Social Knowledge?
In our daily life activities, there are two types of knowledge frequently used. First is
the content knowledge. Content knowledge is all about learning how to perform cer-
tain tasks in a professional manner. It may be derived from basic principles learned
from formal education such as tertiary institution or learned from an experienced ex-
pert. Many learning tools support the learning of content knowledge as that reflects
one’s in-depth knowledge about his/her skill and professionalism. The current tools
that are known are such as the expert system, intelligent tutoring system, intelligent
computer aided-learning, microworld etc. In a simple example, a medical doctor is
called specialist when he/she embarks on specialized course and training in order to be
an orthopedist or pediatrician. His/her knowledge accumulated after a long years of
experience. This type of knowledge is static, rigid and stable. However, the second
type of knowledge is called social-knowledge (or socially-derivable knowledge) which
may not be obtained though formal learning or experience but rather through commu-
nity interactions. Knowledge about the current epidemics and which medical center
has the best treatment can only be obtained through interaction with the community.
Knowledge about the best practices in conducting staff appraisals by the blue chip
company can be known through social interactions. This type of knowledge is dy-
namic, fluid and unstable in the sense that it may change from time to time and its
validity can easily be superseded by the most current ones. In the other scenario,
Denning [6] describes how the problem in Pakistan’s highway was solved at instant
after contacts with colleagues who had experience solving the similar problems in
South Africa. The knowledge exchange was not on the content knowledge (about
fundamental theories in engineering course) but rather a social knowledge which can
only be derived through acquaintance.
Due to the differences between content knowledge and social knowledge, the de-
velopment tool in facilitating the learning is also different. The content knowledge
which contains facts and fundamental theories can be learned using courseware or
computer-based learning software; while experience can be learned through expert
system or intelligent tutoring system. Nevertheless, social knowledge requires com-
munity as the integral part of knowledge source. The process of building the system
that support learning for social knowledge requires consideration given to the follow-
ing factors [7]:
• Multiplicity in learning objects – knowledge in the real world is delivered or ob-
tained in different forms. The objects, which are used as part of the learning
whether directly or indirectly is called learning, object as described by Community
of Practice [8]. Radio, television or LCD screen used for advertising are examples
of broadcasting system that contribute to one’s knowledge. Newspaper, maga-
zines, leaflets or brochures are pieces of information, which transform into one’s
knowledge when he/she reads them. Other forms of learning objects are the work-
ing colleagues, animated or unanimated artifacts such as the copier machine, pets
at home, video movies and neighbors whom one socialize with. In this respect, the
expert knowledge does not come from a single source as well as the multiplicity in
Intelligent Conversational Channel 345
methodology for delivering the knowledge. Expert’s talk in the open seminars or
television are examples of learning objects.
• Open-world assumptions – assumption is needed when one designs a system to be
used as problem-solver. The assumptions are perspective that draws the boundary
of the intended world in order for the system to work successfully within the speci-
fied limit. In modeling the content-knowledge, close-world assumption is always
used. Unlike the content knowledge, social knowledge does not specify the as-
sumption as the knowledge is not modeled but shared in its original form. The
knowledge contains the description about the real world problems and solution
rather than the hypothesized.
• Rapid knowledge-building – content knowledge requires a system builder to ana-
lyze and study, to model the solution, to build the system and test its performance.
These processes are rather time-consuming and costly. On the other hand, the so-
cial knowledge is built by the community in a progress manner and can be learned
immediately without the need of highly mechanistic and sophisticated process.
Knowledge is presented in a human-readable format rather than machine-readable
format.
• Unorganized, ubiquitous but retrievable – content knowledge built in an expert
system is structurally organized and frequently validated by the truth maintenance
technology. The purpose is to avoid conflict of facts and retain consistencies in
delivering solution. The retrieval of the solution depends on the reasoning tech-
nique employed in the system. Social knowledge is rather unstructured and ubiqui-
tous. The knowledge allows conflict solutions to a single problem as it can be
treated as having choices of different perspectives. Learners are not confined to
solution of a single expert in this case as knowledge is contributed by several ex-
perts or non-experts who is involved in the knowledge construction process. The
social knowledge is retrieved through social interactions and dialogues with the
communities.
In the following section, we discuss the technology built on ICC as a tool in sup-
porting learning social knowledge.
3 Components of Intelligent Conversational Channel
The technology of ICC is built to enable the operation of the upper stream of the
knowledge management which is at the user or community level. There are researches
about building techniques in extracting knowledge from resources such as documents,
images, videos, audio, data warehouse using intelligent information retrieval or human
expert through knowledge acquisition. Our claim is that these systems are not flexible
to allow the knowledge to be shaped up by the community who are the main benefici-
aries of the knowledge. For example, several educational softwares are designed ac-
cording to the specifications of the pedagogy theories which are rather predetermined
by the designer. The design of an expert system takes consideration of small scope of
human users while its application is expected to be wide. In all cases, the design is
known and fixed before its development. ICC approaches towards knowledge shaping
is flexible such that the community will determine what knowledge will be placed on
346 S.M.F.D. Syed Mustapha
the knowledge repository, the content of knowledge is extracted through “mix and
match”1 process by the community, the shaping of knowledge process is resilient and
destined by the responses and arguments posted into the community channel and
knowledge externalization is done through dynamic interaction with the virtual com-
munity. These ideas are illustrated in Fig. 1.
3.1 Community Channel
In the community channel, two forms of knowledge can be presented using narrated
text typed in the story object and also uploading of multimedia objects such as video
clips, images, documents, html files.
Fig. 2 shows a user expressed his/her concern about school delinquency problem
and uses an image file to share the reality. Other members have the choices of replying
to the above message or submit a new story object as shown in Fig. 3. The text on
1 Each member of the community has his/her own way of extracting (match) the gist of know
edge he/she is interested in from a single source. The combination (mixing) of these knowl-
edge collections gradually builds the community knowledge base.
Fig. 1. Components of Intelligent Conversational Channel
Discourse commu-
nicator through virtual
community
Hypermedia learning space
Community
Channel D
Community
Channel C
Community
Channel B
Community Channel A
Company’s annual report
Discourse analyzer
Intelligent Conversational Channel 347
the left box is submitted by the user who wants to start with new subtopic about canning
system practiced in school. The right text box contains responses of another two mem-
bers who respectively support the earlier statement and suggest a new solution. The
taggers <<Support and <<Suggest label the intended semantic meaning of the sentence.
Fig. 3. Members’ responses in the form of support and suggestion
3.2 Discourse Communicator
Discourse communicator provides simulated community discussion using software
agents. Fig. 4 shows an interaction session with agents. A member posted a query
Fig. 2. Community channel that supports two forms of knowledge represen-
tation
Fig. 2. Community channel that supports two forms of knowledge representation
Image file is
uploaded to the
server to augment
the reality of the
written story.
User types in
narrated text about
students’ delin-
quencies and
school bullies
ries.
Canning is not a solution to violence in
schools. It is just coarsening the relation-
ship between the teachers and taught. In
Australia parent and non governmental
bodies are opposed to corporal punishment.
I attached the re
p
ort.
<<Support: I agree with the argument that
canning return bad repercussion. Canning can
be misused as a means to show out of the teach-
ers’ personal anger and not merely as a punish-
ment.
Please click the link for more sto
<<Suggest: There is another alternative to
canning like killing flies. Take a look what
happen in Dutch school.
348 S.M.F.D. Syed Mustapha
“What do you think about canning practice in school?”. Using keyword searching
method, the system activates the discourse block that has the highest relevancy.
3.2 Discourse Analyzer
In order to stimulate the social knowledge building, discourse analyzer plays the role
of monitoring and analyzing the dynamism of group interaction and topic of discourse.
These are done based on four factors which are 1) intensity of participation - measures
the ratio between the numbers of participated activity of a member against the num-
bers of participated activity by all members 2) multiplicity in existence - describes the
versatility of a member in participating different subject matter 3) interactivity of the
subtopic – uses standard deviation to measure the popularity of each subtopic and 4)
social identity recognition – analyze the density of social interaction between all
members (for detail refers to [9]).
3.3 Hypermedia Learning Space
Members build the association between the story objects and the multimedia objects
which are stored in the hypermedia learning space. Therefore, each object is indexed
by the brief text description entered in the story object. They are linked together by
the main keywords to form a complex network association of knowledge units
(knowledge unit is a composition of story object and the multimedia object). The
network of knowledge unit enables the retrieval of a discourse block when one of the
user’s query texts matches the keywords.
4 Conclusion and Future Work
In this paper, it is argued that there are significant differences of knowledge namely
the content knowledge and social knowledge. The content knowledge is very specific
Fig. 4. Interactions with agents session
Intelligent Conversational Channel 349
knowledge that requires a specific design in capturing its content. The type of knowl-
edge representation for content knowledge may be different from domain to domain.
However, for social knowledge, the knowledge can be presented simply using natural
language while for a more elaborative description, it can be presented using several
other knowledge media. While we do not deny the importance of the content knowl-
edge, social knowledge requires equal attention and system that supports the learning
is necessary. ICC has been proposed as the tool for learning social knowledge on the
basis that the virtual community is created as representative to actual community in
catalyzing community interaction, community knowledge is kept in the hypermedia
learning space for future retrieval and reuse, and the community interaction is kept
alive through monitoring members participating profile. The system can be enhanced
in several ways. One of the chosen future works is modeling the agents so that can be
embodied with personal characteristics. At the moment, the agent is generic and its
actions such as gestures are predetermined. The agents are not capable to produce
original response, as they are not equipped with personal knowledge. In order to make
the conversation dynamic and malleable to the user’s response and behavior, the agent
should be allowed to make their own responses and reactions.
References
1. Kazaz, A.: Application of a Expert System on the Fracture Mechanics of Concrete. Artifi-
cial Intelligence Review 19: 177-190, (2003), Kluwer Academics Publishers.
2. Wan, D.: CLAIRE – A Computer-supported Collaborative Learning Environment Based on
the Thematic Structure of Scientific Text. PhD Thesis, University of Hawaii, 1993.
3. Stahl, G.: A Model of Collaborative Knowledge-building. In B. Fishman & S. O’Connor-
Divelbiss (Eds)., Proceedings of the Fourth International Conference of the Learning Sci-
ences, pp 70-77. (2000). Mahwah, NJ:Erlbaum.
4. Chang, C.Y.: A Problem-solving Based Computer Assisted Tutorial for the Earth Sciences.
Journal of Computer Assisted Learning, (2001), 17, pp263 – 274.
5. Brown, J.S, Denning, S., Groh, K., Prusak, L. Story-telling – Passport to the 21st Century.
Available at http://www.creatingthe21stcentury.org/Intro0-table.html.
6. Denning, S.: Story-telling: The art of Springboard story. Available at http:// www. creating-
the21stcentury.org/Steve.html.
7. S.M.F.D Syed Mustapha.: Knowledge construction technology through hypermedia-based
Intelligent Conversational Channel. International Conference of Artificial Intelligence and
Engineering Technology, (2004) – to appear.
8. Wenger, E. Community of Practice – Learning, Meaning and Identity. (1998). Cambridge
University Press.
9. S.M.F.D Syed Mustapha. Agent-mediated for Intelligent Conversational Channel for Social
Knowledge Building in Educational Environment, 5th Int. Conf. on Information Technology
Based Higher Education and Training: ITHET ’04 31th May - 2nd June 2004, Turkey.
