Evolutionary of Building Knowledge Model from content

Evolutionary of Building Knowledge Model – from content knowledge to social knowledge Evolutionary of Building Knowledge Model – Content Knowledge to Social Knowledge S. M. F. D Syed Mustapha, Ph. D Faculty of Information Technology - Universiti Tun Abdul Razak

Outlines n Overview of the talk Ø Ø Ø n n n Motivation Knowledge and its sources – content knowledge and social knowledge Learning system and knowledge building History and evolution Intelligent Computer Aided Instruction (ICAI)/ Intelligent Tutoring System (ITS) Modeling Rheologist’s knowledge Modeling Chemist’s knowledge Modeling Social knowledge Conclusion

Overview of the talk n Motivation Ø Building learning system- • how learners learn, • how knowledge is encoded with the capability to reason, • how pedagogy can be instituted in the learning system • how an individual learning can be supported using computer technology Knowledge and its sources – content knowledge and social knowledge n Learning system and knowledge building n

Overview of the talk n Motivation n Building learning system- how learners learn Individual Inquisitive Peer to Peer Personal Coaching Multiple Guidance

Overview of the talk n Motivation n Building learning system- how knowledge is encoded with the capability to reason Printed materials Learning Systems Human reasoning Multimedia Objects

Overview of the talk n Motivation n Building learning system - how pedagogy can be instituted in the learning system Sage on stage Special Interest Group Learner centered Tutoring

Overview of the talk n Motivation n Building learning system - how an individual learning can be supported using computer technology P 2 P network Video /Audio Conferencing Computer-based

Overview of the talk n Motivation n Knowledge and its sources – • Reasoning • Task • Experience • Observation • Structured Document • Unstructured Document • Semi-structured Document Human Expert Community • Behavior & Culture • Preferences • Social Networking • Crowd movement Document Media • Images • Video (streaming) • Discourse (text/audio) • Social Media Where knowledge can be extracted!!

Overview of the talk n Motivation n Knowledge and its sources – two types of knowledge Content Knowledge Social Knowledge for building learning system

Overview of the talk n Motivation n Knowledge and its sources – content knowledge Printed materials Content Knowledge Institutions Expert • formal knowledge • structured presentation • institutionalized • longer life span • universally accepted • derived from single expert • formally assessed • rigid, stable and consistent

Overview of the talk n Motivation n Knowledge and its sources – social knowledge Social Knowledge E-mails Blogs Discussion • informal knowledge • unstructured presentation • non-institutionalized • contemporary • locally accepted and recognized • derived from multiple experts • not formally assessed • amenable, inconsistent and fluid

Overview of the talk n Motivation n Learning system and knowledge building n io t lu o v E Content knowledge learning system – QALSIC and RAS History of Teaching machine and Computer Aided Instruction (CAI) Social knowledge learning system – ICC (Intelligent Conversational Channel)

History and evolution n Teaching machine Ø Ø Ø n n n B. F Skinner, Ph. D (Harvard) Professor, University of Harvard Professor of Psychology The teaching machine was a mechanical device whose purpose was to administer a curriculum of programmed instruction. It housed a list of questions, and a mechanism through which the learner could respond to each question. Upon delivering a correct answer, the learner would be rewarded. Learning theory: rote and drill Technology: Structured program instruction

History and evolution n Contemporary of Skinner’s invention Ø Norman Crowder • introduced branching program • Programmed Instruction • provides feedback on correctness and incorrectness of the learner’s response • instructions are built into frames • learners who failed on a certain section will be branched back to the earlier frame • beginning of building adaptive system

History and evolution n Reflexive Model Era: The technology adopted in the teaching machine did not allow generation of new materials, required pre-stored instruction, responded based on “correct” or “incorrect” input, had limited learning pathway (i. e. no opportunity for selfexploration by the student) and knowledge content that was delivered was not tailored to the individual leaner.

History and evolution n Computer Aided Instruction (CAI) Ø 1960 – 1970 s. Ø new generation of teaching machine. Ø known as “generative system” or “adaptive system” Ø generates different sets of questions from a test bank and solutions automatically.

History and evolution Generative System Era: The CAI individualized the presentation of teaching materials which was based on the learner’s performance - each learner would have experienced different learning path. n Auto-generated problems and solutions that enriched the learning space. n Learner’s performance was measured as the basis to gauge the student’s level of competence in the subject. n

History and evolution n Weakness of CAI Ø Expert knowledge model – CAI does not possess well-defined knowledge of the subject matter and not able to mimic the actual performance of the human tutor. Ø Student model – CAI measured the student’s performance but it had no means to determine the gap between student’s levels of understanding to the expert’s. The gap fillers differ for each individual student and hence it requires customized approach, socalled “personalized delivery”.

History and evolution n Weakness of CAI Ø Pedagogical model – CAI had very limited teaching strategy as it was not able to analyze the error made by the student that could be used for diagnosis and proper selection of teaching strategy. Ø User interface – the available technology during the development of CAI was still primitive and limited to be able to process complex communication language and symbols.

Intelligent Computer Aided Instruction (ICAI)/ Intelligent Tutoring System (ITS) Pedagogical Module Expert Knowledge Module Student Model Module User Interface Module

Generic ITS Architecture n Why generic ITS structure remains valid until today? Ø Knowledge media and representation - the knowledge media has been greatly influenced by the continuous development of multimedia software tools and objects as well as the hardware devices. Researchers in the field of artificial intelligence started with production rules, semantic network and first order logic and further developed frame-based knowledge representation, case-based reasoning, qualitative models and semantic web technologies.

Generic ITS Architecture n Why generic ITS structure remains valid until today? Ø Communication and information delivery technologies - the communication technologies have changed situated learning like in the class room to mobile learning. Ø Teaching and learning strategies - the traditional “sage on stage” has migrated into student-centered learning mode e. g. problembased learning. The “drill-and-practice” method that is applied to all has to be customized for individualized needs of the learners.

Generic ITS Architecture n Why generic ITS structure remains valid until today? Ø Knowledge type and learning culture and communities - from the early days, knowledge that is stored in the teaching machine or in the intelligent tutoring system is categorized as “content knowledge” which is structured and formal. Today, social computing and the growing interests in social media introduce “social knowledge” which is unstructured and informal. That also changes the culture of learning from modeling knowledge of a single expert to multiple experts.

Modeling Rheologist’s Knowledge Rheometer Advisory System

Modeling Rheologist’s Knowledge n What is rheology? n Study of the flow and deformation of “non-classical materials” (or called Non-Newtonian Fluids)

Modeling Rheologist’s Knowledge Rheometer → Data Capturing → Graph Analysis

Modeling Rheologist’s Knowledge Applying Rheological Model on Rheogram

Modeling Rheologist’s Knowledge Rheometer Selection Geometry Selection Rheological Model Building Sample Preparation Rheogram Visual Reasoning

Modeling Rheologist’s Knowledge Rheometer Advisory System Tutorial Approach Automated Visual Reasoner Automated Model Builder

Modeling Rheologist’s Knowledge – Tutorial Approach

Modeling Rheologist’s Knowledge – Automated Visual Reasoner

Modeling Rheologist’s Knowledge – Automated Visual Reasoner Normal Graph Anomalous Graph

Modeling Rheologist’s Knowledge – Automated Model Builder

RAS – in the perspective of RAS n Expert Module: Ø Automated Visual Reasoning - qualitative description and curve attributes. Ø Tutorial approach - textual and graphic representation. Ø Task Modeling – action structure, actors, tools and information. n Pedagogy Module Ø Mixed initiative approach – fully automated model builder or semi-automated

References n n n Syed Mustapha, S. M. F. D. , Moseley, L. G. , Jones, T. E. R. , Phillips, T. N. and Price, C. J. (1999). Viscometric flow interpretation using qualitative and quantitative techniques, International Journal of Engineering Applications in Artificial Intelligence, Vol 12, pp. 255 – 272. Syed Mustapha, S. M. F. D. , Phillips, T. N. (2002). Rheometer Advisory System – Embedding Mixed-Initiative Approach Into Its Teaching and Learning Strategy, Journal of Applied System Studies, Vol 3, No 1. Syed Mustapha, S. M. F. D. (1998). Visual Reasoning Using Qualitative Interpretation, National Conference in Cognitive Science, The Mines Resort, pp 213 -220.

Modeling Reasoning - QALSIC (Qualitative Analysis and Laboratory Simulation for Inorganic Chemistry)

Modeling Reasoning - QALSIC Work Book – Experimental Procedure Work Book – Report Observation Experiment in Wet Lab Traditional Laboratory Observation

Modeling Reasoning - QALSIC Work Book – Experimental Procedure Work Book – Report Observation Experiment (QALSIC) QALSIC + Wet Laboratory Environment Observation (QALSIC)

Modeling Reasoning - QALSIC Work Book – Experimental Procedure (QALSIC) Work Book – Report Observation (QALSIC) Experiment (QALSIC) QALSIC – Full Lab Environment Observation (QALSIC)

Modeling Reasoning - QALSIC Work book in QALSIC

Modeling Reasoning - QALSIC Simulation in QALSIC Example of equilibrium state H 2 S 2 H+ + S 2 - (black precipitation)

Modeling Reasoning - QALSIC Simulation in QALSIC - system generated explanation The dissolution of H 2 S (reagent) in water produces H+ and S 2 - ions, until equilibrium is reached between H 2 S and both ions. Since only S 2 - is used up to react with Fe 2+ to form the precipitation, H+ ion (which is an ion responsible for acidic environment), remain unused. The accumulation of the H+ ions towards saturation in the system will hinder the dissolution of H 2 S anymore. Thus, there are insufficient amount of S 2 - ions to react with Fe 2+, and precipitation will not be formed. Adding HCl will thus hinder the forward reaction more extensively.

Modeling Reasoning - QALSIC IV 1 IV 2 IV 3 IV 4 VI (View Instance) Flow of Qualitative Process Individual Structure (IS) (preconditions quantity condition)initial ≠ (preconditions quantity condition)current P 1 P 2 P 3 P 4 Process Instance (PI) Process Structure (PS) PI

Modeling Reasoning - QALSIC

Modeling Reasoning - QALSIC Fe. S formed a black precipitation (mixed with water)

Modeling Reasoning – QALSIC H 2 S dissociation is not complete (acidic environment)

QALSIC – in the perspective of ITS n Expert module – knowledge is modeled based on the chemistry principles and basic process. Ø allows reasoning on low level of granularity such that QALSIC is capable to handle unprecedented cases. Ø system supports recommendation on the potential ions that may exist Ø n Pedagogy module – Ø Inquisitive-based learning – learners are allowed to perform individual learning on full QALSIC mode or integrated with traditional practice.

References n n Pang, J. S. , Syed Mustapha, S. M. F. D. , and Zain, S. M. (2001) Preliminary studies on embedding qualitative reasoning into qualitative analysis and laboratory simulaton, The Pacific Asian Conference on Intelligent Systems 2001 (PAIS 2001), pp 230 – 236, Seoul, Korea. Syed Mustapha, S. M. F. D and Esther Gnanamalar Sarojini Daniel. (2005). Implementation of a Piagetian- Vygotskian Inquiry. Based Learning Model through the QALSIC-ICC systems. International Conference in Information Technology and Communication in Management, 23 rd – 25 th May, 2005. Syed Mustapha, S. M. F. D. , Pang J. S. and Zain, S. M (2005). QALSIC: Towards building an articulate educational software using Qualitative Process Theory approach in inorganic chemistry for high school level, International Journal of Artificial Intelligence in Education, 15(3) 2005, pp 229 – 257. Syed Mustapha, S. M. F. D. , Pang J. S. and Zain, S. M. (2002). Application Qualitative Process Theory to Qualitative Simulation and Analysis of Inorganic Chemical Reaction, 16 th International Workshop of Qualitative Reasoning, Barcelona, Spain, pp 177 -184.

Modeling Social Knowledge Intelligent Conversational Channel (ICC)

Modeling Social Knowledge n What is the learning theories now? Ø behaviorist learning – Teaching Machine Ø developmental learning – interaction with objects Ø cognitive learning – regeneration of knowledge Ø socio-cultural learning – community behavior and practices Organizational learning – Community of Practice n Learning in 21 st Century University – J. S Brown n

Modeling Social Knowledge n Social knowledge in Community of Practice Ø Ø Community structure – the community emerges upon specific interest that could be explicit or implicit. In a sponsored Co. P (formally organized), the goal is well-defined and the community member is usually is firmed and well-defined. In the true sense of Co. P, the memberships are formed naturally and not consistent at all times. Learning through participation and reification – participation from each member is expected since s/he establishes clear role and function. This can be achieved through repetitive involvement and deep relationship that subsequently produce unprecedented result. Negotiation of meaning – community artifacts are progressively developed and evolved over several attempts, understandings, discussions and collaborations. In that view, conflicts and resolutions are essential part in the negotiation process. Learning as temporal – the life span of community engagement is not ad hoc but spreads over a period of time. Usually, the time space indicates the learning through a shift of the knowledge content at several points of time scale.

Modeling Social Knowledge n Social knowledge in Community of Practice Ø Ø Ø Boundary objects – learning objects from various members are shared and the multiplicity of the interpretations and understanding of how those objects should be used and applied. The learning objects are defined in broad manner that they range from the abstract concept, ideas, philosophy, formulae, language and theory to the more physical form, such as programming codes, architect miniature and others. Boundary encounters – communities of different group in contact on several occasions to produce a new learning object (refer to boundary objects). Mutual engagement – members appear in a flat structure that each member is equally important with same level of participation. The community members establish agreeable activities among them.

Modeling Social Knowledge n Social knowledge in Community of Practice Ø Ø Ø Joint enterprise – members recognize each other through sense-making of every member’s capability and coordinate accordingly. The ownership of the task is voluntarily taken while maintaining interactivity and link among members. Shared repertoire – identifiable sociable products that are unique to the community but shared exclusively. The tangible and intangible products are formed through deep social engagement and interlocking. Identity – recognizing an individual’s raison d'être in the community based on one’s specific skill, knowledge, talent, contribution, leadership and quality.

Modeling Social Knowledge n John Seely Brown Ø John Seely Brown was the chief scientist and now visiting scholar as well as advisor to Provost at University of Southern California (USC) and also Independent Co-Chairman for Deloitte’s Center for Edge Innovation. He has been deeply involved in the radical innovation, corporate strategy and strategic positioning for Xerox Palo Alto Research Center. Ø pioneers to developing intelligent tutoring system in 1980 s – SOPHIE (I, III), BUGGY, WEST.

Modeling Social Knowledge n John Seely Brown – University in the 21 st Century Ø Ø Ø New vernacular – young generation’s acute interest in social media and learning using digital technologies leads to enculturation of digital vernacular. Independent learning – despite there is growing interest by the public on the importance of education in the new century, many of them are less keen to spend money for it. For that, the students should be given to explore the knowledge they wanted it to be. Are these appropriate – take home assignment, quizzes, “sage on stage”, exams? Fixed and single career – the fast growing industrial change and innovation, graduates have high tendency of working in multiple professions along the career pathway. Most likely that the knowledge acquired in the university is no longer relevant. Systemic problems – most significant problem of tomorrow requires multiple specialties in order to solve it. That means student must be prepared to acquire new set of knowledge that is outside the domain of university studies. Institution of higher learning (IHL) or Learning institution? – students can play the teacher’s role and vice versa. Institution must be the learning organization itself!

Modeling Social Knowledge

Modeling content knowledge and social knowledge for building learning system n Differences Ø Multiplicity in learning objects Ø Open-world assumptions Ø Rapid knowledge-building Ø Unorganized, ubiquitous but retrievable

Modeling Social Knowledge - Intelligent Conversation Channel Community Channel B Community Channel C Community Channel D Company’s annual report Community Channel A Discourse communicator through virtual community Hypermedia learning space Discourse analyzer

Modeling Social Knowledge - Intelligent Conversation Channel Canning is not a solution to violence in schools. It is just coarsening the relationship between the teachers and taught. In Australia parent and non governmental bodies are opposed to corporal punishment. I attached the report. <<Support: I agree with the argument that canning return bad repercussion. Canning can be misused as a means to show out of the teachers’ personal anger and not merely as a punishment. Please click the link for more stories. <<Suggest: There is another alternative to canning like killing flies. Take a look what happen in Dutch school.

Modeling Social Knowledge - Intelligent Conversation Channel Traditional approach Hypermedia approach

Modeling Social Knowledge - Intelligent Conversation Channel

References n n Syed Mustapha, S. M. F. D. (2004 a). Intelligent Conversational Channel for Learning Social Knowledge among Communities. 8 th International Conference on Knowledge-based Intelligent Information & Engineering System, KES 04, New Zealand, Vol I, pp 343 -349, 2004. Syed Mustapha, S. M. F. D. (2004 b). An Algorithm for Avoiding Paradoxical Argument among the Multi-Agent in the Discourse Communicator. 8 th International Conference on Knowledge-based Intelligent Information & Engineering System, KES 04, New Zealand, Vol I, pp 350 – 356, 2004. Syed Mustapha, S. M. F. D. (2004 c) Towards Building Socially Intelligent Knowledge-Building System – in probe to missing component. Technical Report, Department of Information and Communication Engineering, School of Information Science and Technology, The University of Tokyo, EMEDIA 2004, pp 169 -179. Syed Mustapha, S. M. F. D. (2004 d) Agent mediated for intelligent conversational channel for social knowledge-building in educational environment. 5 th Int. Conf. on Information Technology Based Higher Education and Training: ITHET ’ 04, 31 st. May – 2 June, 2004, Turkey, pp 533 – 538.

Conclusion Learners Learning System Pedagogy