Developer Data Modeling Mistakes: From Postgres to NoSQL
Sintec
1. Knowledge-Based e-Learning
The SINTEC Personalized,
Knowledge based systems
Knowledge-Based E-Learning
Student modeling
Environment Reasoning for:
Ştefan Trăuşan-Matu 1,2
Student diagnosis
Valentin Cristea 1
Explanations generation
Octavian Udrea 3
Lesson planning
1Computer Science Department,
Bucharest "Politehnica" University,
Intelligent interfaces
2Romanian Academy Institute for Artificial Intelligence
ROMANIA
3Maryland University, US
1/15/2010 OL-KWM 2005 2
Knowledge Ontologies
Learning is a knowledge centered activity:
One of the main goals of a learning process is the
articulation in the learner’s mind of a body of "An ontology is a specification of a
knowledge for the considered domain. conceptualization....That is, an ontology is
The skeleton of this body is usually a semantic a description (like a formal specification of
network of the main concepts involved in that a program) of the concepts and
domain - ONTOLOGY
relationships that can exist for an agent or
a community of agents" (Gruber)
1/15/2010 OL-KWM 2005 3 1/15/2010 OL-KWM 2005 4
Ontologies used in e-Learning Personalized texts for e-Learning
Are adapted to each users':
Domain knowledge - student model
Tutoring learning style
psychological profile
Human-computer interfacing
goals (e.g. lists of concepts to be learned)
Lexical level (novice, expert)
Upper Level preferences (e.g. style of web pages)
context of interaction
1/15/2010 OL-KWM 2005 5 1/15/2010 OL-KWM 2005 6
1
2. Student model Intelligent e-Learning Projects at
CS-Polytechnic University & ICIA
Keeps track of the concepts known, unknown or
wrongly known by the student (Dimitrova, Self, Brna, MacPAIL
2000)
ITS for programming
Inferred from results at tests or from interaction
(visited web pages, topics searched etc.) WebGen
Is usually defined in relation with the domain LARFLAST
ontology (concept net, Bayesian net)
SINTEC
EU-NCIT
COOPER
1/15/2010 OL-KWM 2005 7 1/15/2010 OL-KWM 2005 8
SINTEC (2002-2003) SINTEC
INFOSOC- Funded Project Collaborative tools for distance and distributed e-Learning
Web services technology for distributed processing
Includes experience from ITS and LARFLAST knowledge (ontologies) from the (Semantic) Web
Partners : Content creation and reuse from the web, according to
metadata standards for e-Learning like IMS, ARIADNE,
CS Dept., “Polytechnica” Univ. Bucharest SCORM, AICC
Romanian Academy Institute for AI ITS technology (student modelling and inference)
Romanian Academy Psychology Institute Text Mining:
SIVECO S.A. Romania intelligent search of learning materials on the web
knowledge extraction
Continued in FP6 SSA EU-NCIT categorization
summarization
1/15/2010 OL-KWM 2005 9 1/15/2010 OL-KWM 2005 10
INSTITUTIONAL PORTAL - WWW
Collaborative Services
Content
Editor Knowledge
Server
Intelligent
CONTENT Search
Interface SERVER
Generator Knowledge
Editor
Content Knowledge Knowledge
Information
Management Extraction Repository Knowledge
Repository
Management
Training Management Web Pages Tests
Generator Answer
Generator
Analysis
Student
Lecture Model Intelligent Tutoring
1/15/2010 Scheduling OL-KWM 2005 11 1/15/2010 OL-KWM 2005 12
e-Learning
2
4. Conclusions
The approach was used for CS students at
PUB
Algernon is not reliable – in future JESS
A lot of psychology work to be done
Difficult to develop the ontology
1/15/2010 OL-KWM 2005 19
4