WPLE Seminar, 1 st November 20061 Issues in Personalised, Adaptive systems for supporting individual learners and communities of learners George Magoulas.

Slides:



Advertisements
Similar presentations
Web Passive Voice Tutor: an Intelligent Computer Assisted Language Learning System over the WWW Maria Virvou & Victoria Tsiriga Department of Informatics,
Advertisements

Performance Assessment
Requirements gathering
MUSHI-Life Presenter Richard Joiner Designer : Chris Quintana.
Experimental Course for Students with LD/ADHD Diana Cassie, Ph.D. Dalhousie University.
TU e technische universiteit eindhoven / department of mathematics and computer science Modeling User Input and Hypermedia Dynamics in Hera Databases and.
TU/e technische universiteit eindhoven Hera: Development of Semantic Web Information Systems Geert-Jan Houben Peter Barna Flavius Frasincar Richard Vdovjak.
Agent-Based Architecture for Intelligence and Collaboration in Virtual Learning Environments Punyanuch Borwarnginn 5 August 2013.
ARCHITECTURES FOR ARTIFICIAL INTELLIGENCE SYSTEMS
CCE GUIDELINES FOR CLASSES 6 TO 8
Data accreditation standard for the IM&T DES12 Sept The IM&T DES Using the tools that support e- audit John Williams & James Barrett.
Chapter 11 user support. Issues –different types of support at different times –implementation and presentation both important –all need careful design.
Bayesian Network and Influence Diagram A Guide to Construction And Analysis.
Towards Adaptive Web-Based Learning Systems Katerina Georgouli, MSc, PhD Associate Professor T.E.I. of Athens Dept. of Informatics Tempus.
Personalized and adaptive eLearning Applications in LSMs
/ Where innovation starts 1212 Technische Universiteit Eindhoven University of Technology 1 Incorporating Cognitive/Learning Styles in a General-Purpose.
Consistency of Assessment
Page16/2/2015 Sirlan Usage and usability considerations for SIRLAN solution success.
Personalized Hypermedia Presentation Techniques for Improving Online Customer Relationships Alfred Kobsa, Jurgen Koenemann and Wolfgang Pohl Presented.
Design Activities in Usability Engineering laura leventhal and julie barnes.
Reproduced with permission from BESTEAMS 2004
Marakas: Decision Support Systems, 2nd Edition © 2003, Prentice-Hall Chapter Chapter 1: Introduction to Decision Support Systems Decision Support.
Marakas: Decision Support Systems, 2nd Edition © 2003, Prentice-Hall Chapter Chapter 7: Expert Systems and Artificial Intelligence Decision Support.
Next Generation eLearning Can Technology Learn from the Learners: The case for Adaptive Learning Objects Vincent Wade Research Director, Knowledge & Data.
Adaptive Hypermedia: What is it and why are we doing it? Dr. Alexandra Cristea
Developing Intelligent Agents and Multiagent Systems for Educational Applications Leen-Kiat Soh Department of Computer Science and Engineering University.
Building Knowledge-Driven DSS and Mining Data
12 -1 Lecture 12 User Modeling Topics –Basics –Example User Model –Construction of User Models –Updating of User Models –Applications.
1. Human – the end-user of a program – the others in the organization Computer – the machine the program runs on – often split between clients & servers.
The 2nd International Conference of e-Learning and Distance Education, 21 to 23 February 2011, Riyadh, Saudi Arabia Prof. Dr. Torky Sultan Faculty of Computers.
Richards Middle School Columbus, Georgia
Enabling Organization-Decision Making
Challenges in Information Retrieval and Language Modeling Michael Shepherd Dalhousie University Halifax, NS Canada.
Interstate New Teacher Assessment and Support Consortium (INTASC)
Domain 1: Preparation and Planning. ElementUnsatisfactoryBasicProficientDistinguished Knowledge of content and the structure of the discipline In planning.
1 Commissioned by PAMSA and German Technical Co-Operation National Certificate in Paper & Pulp Manufacturing NQF Level 3 Collect and use data to establish.
ADALE’06 DublinJune 20, 2006 Issues in developing adaptive learning management systems for higher education institutions Jesus G. Boticario, Olga C. Santos.
University of Dublin Trinity College Localisation and Personalisation: Dynamic Retrieval & Adaptation of Multi-lingual Multimedia Content Prof Vincent.
11 C H A P T E R Artificial Intelligence and Expert Systems.
© 2007 Tom Beckman Features:  Are autonomous software entities that act as a user’s assistant to perform discrete tasks, simplifying or completely automating.
A COMPETENCY APPROACH TO HUMAN RESOURCE MANAGEMENT
1 Commissioned by PAMSA and German Technical Co-Operation National Certificate in Paper & Pulp Manufacturing NQF Level 4 Perform visual inspection tests.
What is a Business Analyst? A Business Analyst is someone who works as a liaison among stakeholders in order to elicit, analyze, communicate and validate.
Information Retrieval Evaluation and the Retrieval Process.
Fundamentals of Information Systems, Third Edition2 Principles and Learning Objectives Artificial intelligence systems form a broad and diverse set of.
RCDL Conference, Petrozavodsk, Russia Context-Based Retrieval in Digital Libraries: Approach and Technological Framework Kurt Sandkuhl, Alexander Smirnov,
Competency Training for aeronautical meteorological forecasters and observers Wang Yong WMO RTC Nanjing, China.
Illustrations and Answers for TDT4252 exam, June
CS2003 Usability Engineering Human-Centred Design Dr Steve Love.
10th International Baltic Conference on Databases and Information Systems July 8-11, 2012, Vilnius, Lithuania Learner Model’s Utilization in the e-Learning.
A Context Model based on Ontological Languages: a Proposal for Information Visualization School of Informatics Castilla-La Mancha University Ramón Hervás.
I Robot.
Screen 1 of 20 Vulnerability Vulnerability Assessment LEARNING OBJECTIVES Define the purpose and scope of vulnerability assessment. Understand how vulnerability.
Fundamentals of Information Systems, Third Edition1 The Knowledge Base Stores all relevant information, data, rules, cases, and relationships used by the.
1 Knowledge Acquisition and Learning by Experience – The Role of Case-Specific Knowledge Knowledge modeling and acquisition Learning by experience Framework.
Harvesting Social Knowledge from Folksonomies Harris Wu, Mohammad Zubair, Kurt Maly, Harvesting social knowledge from folksonomies, Proceedings of the.
1 Commissioned by PAMSA and German Technical Co-Operation National Certificate in Paper & Pulp Manufacturing NQF Level 4 Apply knowledge of statistics.
Jane Reid, AMSc IRIC, QMUL, 30/10/01 1 Information seeking Information-seeking models Search strategies Search tactics.
Augmenting (personal) IR Readings Review Evaluation Papers returned & discussed Papers and Projects checkin time.
1 Reference Model for Evaluating Intelligent Tutoring Systems Esma Aimeur, Claude Frasson Laboratoire HERON Informatique et recherche opérationnelle Université.
ELED 6560 Summer Learning Exercises #10 The Un-Natural Part of Teaching  Five Ways that Teaching Behavior is Un-Natural 1. Helping Others 2.
What is Research?. Intro.  Research- “Any honest attempt to study a problem systematically or to add to man’s knowledge of a problem may be regarded.
User Modeling for the Mars Medical Assistant MCS Project By Mihir Kulkarni.
Certificate IV in Project Management Assessment Outline Course Number Qualification Code BSB41507.
Adaptivity, Personalisation and Assistive Technologies Hugh Davis.
Decision Support and Business Intelligence Systems (9 th Ed., Prentice Hall) Chapter 12: Artificial Intelligence and Expert Systems.
IB Assessments CRITERION!!!.
Software Requirements analysis & specifications
Chapter 11 user support.
Univ. Prof Dr Viktor Jakupec
Presentation transcript:

WPLE Seminar, 1 st November Issues in Personalised, Adaptive systems for supporting individual learners and communities of learners George Magoulas and Alex Poulovassilis

WPLE Seminar, 1 st November Outline of the presentation Common questions about adaptive hypermedia systems Generic Architecture Main components: User Model and Personalisation Engine Open issues, in general and specifically for the Technical-Communicative Skills (TCS) learning which is relevance to the Workplace Personalised Learning Environments (WPLE) project

WPLE Seminar, 1 st November Common questions about Adaptive Hypermedia Systems 1. What are adaptive hypermedia (AH) systems and why is there a need for them? 2. What are the advantages for a user? 3. How can one develop AH systems and what are the major challenges in this area?

WPLE Seminar, 1 st November What are adaptive hypermedia systems and why is there a need for them? Adaptive hypermedia systems have emerged as environments which adapt content, structure, and/or presentation of information to an individual users characteristics and usage of the system. Such systems make adjustments in order to accommodate the diversity of user needs and abilities in order to maintain the appropriate context for interaction of each individual user with the system.

WPLE Seminar, 1 st November An AH system: uses hypermedia allowing navigation through the information hyperspace of the application domain; includes a user model to describe the user, and provides an adaptation mechanism for dynamic adaptation of the hypermedia on the basis of the state of the user model. The adaptation consists of changing the content and/or the presentation of the nodes and links of the hypermedia.

WPLE Seminar, 1 st November What are the advantages for a user? Individuals differ in traits such as skills, aptitudes and preferences for processing information, constructing meaning from information, and applying it to real-world situations. AH systems aim to enhance the mode of delivery to users, according to both content and presentation of information, in order to accommodate the needs and preferences of each individual user.

WPLE Seminar, 1 st November How can one develop AH systems and what are the major challenges? There is a distinction between adaptable systems and adaptive systems: An adaptable system is customisable, allowing the user to configure it by changing some parameters that capture user preferences; and the system then adapts its behaviour accordingly. An adaptive system adapts autonomously to the user by monitoring information about the user, recording this information in a user model and adapting dynamically to the current state of the user model.

WPLE Seminar, 1 st November Recording information about the user may include capturing the users navigation choices through the hypermedia; their other interactions and choices using the system; their answers to on-line forms or questionnaires; and any initial and/or ongoing information they may explicitly provide to the system. Adaptation can be made by either dynamically changing predefined presentation or content, or by dynamically constructing presentation and content out of predefined fragments of information.

WPLE Seminar, 1 st November Developing AH systems presents several challenges: identifying the type of user information that is needed in order to establish meaningful relationships between the users and the adaptive functionalities of a system; user modelling: capturing dynamically the variety and time- varying nature of users interests and characteristics; encoding system content, usage data and user models in such a way that the adaptation mechanism is tractable, and that the personalisation decisions made by the system are readily interpretable. taking into account human factors during design of the adaptive system, including establishing design guidelines for the system development and criteria for both formative and summative system evaluation.

WPLE Seminar, 1 st November Generic architecture Decision making and adaptation engine

WPLE Seminar, 1 st November User models and user data Stereotype models : these are defined by the designer. The system automatically assigns one or more stereotypes to each user. This implies a clear understanding of the different categories of users. Explicit information : observations and interviews with experts; expressing expert knowledge as linguistic rules; information supplied by users which can be used to initialise their user model e.g. via on-line forms or questionnaires. Observation of user behaviour in using the system : analysing the users goals, interests and levels of knowledge e.g. via links followed, pages visited, items selected, tasks undertaken, time spent viewing specific pages or undertaking specific tasks.

WPLE Seminar, 1 st November Examples of behavioural user data related to learning: knowledge data, e.g. the number of correct, incorrect or almost correct answers given in assessments chronometric data, e.g. the time spent to read about a topic, the time taken to supply a correct answer to a question in a test, the total time spent on task, the time of idle intervals try data, e.g. the number of attempts to find a correct solution, the number of times needed to review relevant information before finding a correct solution navigation data, e.g. the number of times a topic, activity, tool, or exercise has been selected, the frequency that specific selections have been made

WPLE Seminar, 1 st November Dimensions of a user model Nine dimensions of a user model can be identified (i) Personal data, such as gender, age, language, culture, affect the perception of the interface layout, and should be taken into account when designing personalisation services. For example, the preferences of males and females differ in terms of navigation support, attitudes, information seeking strategies and media preferences. (ii) Cognitive or learning styles refer to a users information processing habits and have an impact on a users preferred modes of perceiving and processing information, and problem solving. They can be used to personalise navigation support, and the presentation and organisation of information.

WPLE Seminar, 1 st November (iii) Device information concerns the hardware used for access and affects personalisation services in terms of screen layout and bandwidth limitations. (iv) Context-related data capture the physical environment from where the user is accessing the information and can be used to infer information about the users goals. (v) User history data capture past interaction with the system and can be useful under the assumption that users future behaviour will be similar to their recent past behaviours. (vi) User preferences and interests, in the form of keywords or topics of interest for that user.

WPLE Seminar, 1 st November (vii) Goal-related data indicate the reason for which that user is seeking access to information during a particular session. For example, it is not the same to search for information about China as a tourist or as a student writing a school essay. (viii) System experience indicates the knowledge of that particular user about the information system itself, or more generally that type of information system (e.g. digital library, learning environment). It can be used to personalise the navigation support and the presentation and organisation of information, or to provide intelligent help. (ix) Domain expertise relates to the level of understanding of a particular user of the domain knowledge. This can vary with the domain and influences the users behaviour with the system.

WPLE Seminar, 1 st November Types of user models and personalisation services There are generally two types of user model: individual and group models The user model can be used to provide a variety of personalisation services e.g. Prediction: the capability of anticipating user needs using past user behaviour, on the assumption is that a users immediate future will be similar to his/her recent past. Filtering: the selection, and possibly ranking, of a subset of items that are relevant to a user from an original, generic set of items. Recommendation: the capability of suggesting relevant information to a user based on information that is not based on their past behaviour e.g. from the items recommended by a group or community of users, or from the collective behaviour of other users.

WPLE Seminar, 1 st November The Personalisation Engine The aim of the personalisation engine is to support the adaptivity and adaptablity of the system. Adaptivity may be applied through a variety of techniques and aims at the automatic (system-controlled) adaptation of content, presentation and navigation support. Adaptability relates to direct (user-controlled) manipulation of goals, personal traits and preferences, user model or assigned stereotype(s). Users can initiate adaptation, intervene or deactivate personalisation features.

WPLE Seminar, 1 st November What to personalise: Navigation support Generation and presentation of content Filtering and recommendation of content User support Based on what information: knowledge level preferences recent history of user behaviour learning goals personal traits Learning or cognitive style Summary

WPLE Seminar, 1 st November Adaptation techniques

WPLE Seminar, 1 st November Modelling user behaviour Different AI techniques provide different capabilities: Expert systems model knowledge or behaviour by means of logical representations based on symbolic structures. They have problems with handling uncertainty and vagueness. Probability theory allows to model phenomena using probabilities. Statistical techniques (regression analysis, cluster analysis etc) help to analyse observation data. Fuzzy logic is an appropriate approach when knowledge can be expressed using linguistic rules. Neural networks can learn from simulation or the real task.

WPLE Seminar, 1 st November Linguistic rules for relating user behaviour with user characteristics If the time spent to read about a topic is short and the number of correct answers is high, and few attempts to find the correct answers have been made then the student learning rate is fast. If Percentage of Questions the learner has Tried is very low then knowledge is Inadequate. If Percentage of Questions the learner has Tried is high and Percentage of Correct Answers is low then knowledge level is Mediocre. If Percentage of Questions the learner has Tried is high and Percentage of Correct Answers is high then knowledge level is Advanced. If Percentage of Questions the learner has Tried is high and Average Number of Attempts is high then knowledge level is Mediocre. General rule form: IF B 1 is V 1i AND B 2 is V 2f …AND B k is V kf THEN C L is C Lj,

WPLE Seminar, 1 st November Some general open issues 1. Is it possible to infer profiles of typical user behaviours? 2. How can user needs, abilities and changes in behaviour be automatically discovered and represented in user models? 3. Is it possible to aggregate partial user models generated by various services to reuse them in new personalised systems? 4. What is the best way to evaluate adaptive hypermedia systems? 5. How can personalisation technologies be incorporated into authoring/course design tools? 6. Is it possible to develop content independent, reusable pedagogical templates to support authoring of pedagogic models for adaptive/personalised learning environments?

WPLE Seminar, 1 st November Some issues for the WPLE project Structuring of hypermedia for TCS learning? Content design and development for TCS learning? Dimensions of the user model for TCS learning? What types of personalisation are appropriate for TCS learning? Navigation support, generation and presentation of content, filtering and recommendation of content, user support? What types of user information should be gathered, in order to establish meaningful relationships between users and the adaptive functionalities?

WPLE Seminar, 1 st November Some issues for the WPLE project Will there be stereotype user models? For individuals and/or groups of learners? How should users be matched against these stereotypes? Is there expert knowledge available on TCS learning? Can we express this knowledge as linguistic (or other) rules? Identification of appropriate machine learning techniques for the adaptation functionality? Encoding of system content, user data and user models for tractable adaptation and interpretable decisions? Criteria for formative and summative evaluation of the effectiveness of the personalisation?