Personalisation in Self e-Learning Networks Kevin Keenoy Birkbeck College, University of London IST Workshop on Metadata Management in Grid and P2P systems (MMGPS) - Models, Services and Architectures 16th December 2003 The SeLeNe Project (IST-2001-39045) is a one-year Accompanying Measure funded by EU FP5 1st November 2002 to 31st January 2003 16th December 2003 Personalisation in Self e-Learning 2 Networks Overview • Why SeLeNe? - Motivation • What is a SeLeNe? • The SeLeNe information space – Learning Object metadata • Personalisation in SeLeNe – Personal Views – Personalised query results – Notification services 16th December 2003 Personalisation in Self e-Learning 3 Networks Why SeLeNe? • There are a huge number of learning resources now available on the Web • Electronic “textbooks” can now be created collaboratively in ways that were previously impossible • We need tools to allow for the discovery, sharing and collaborative creation of learning resources • Semantic metadata describing these resources can enable advanced services more powerful than traditional Web techniques 16th December 2003 Personalisation in Self e-Learning 4 Networks What is a SeLeNe? • Formed by members of a learning community – instructors, learners and content providers • The community creates a collection of shared Learning Objects (LOs) and their metadata descriptions • Users register and share a LO by providing a metadata description of it; some parts of the metadata can be automatically generated • The descriptions form a metadata repository – it is the descriptions and not the LOs themselves that SeLeNe manages • There are various deployment options, the most general of which allows the repository to be distributed across many Sites 16th December 2003 Personalisation in Self e-Learning 5 Networks IEEE LOM Custom Sch. Program Module Course SITE Lesson SITE Self e-learning PEER SITE Network SITE LO SITE ACM CSS Objectives SITE SITE CSCourses Lesson SITE AICourses ppt SITE Learning Object Metadata • IEEE LOM chosen for SeLeNe • Augment with “customised attributes” such as – Learning styles catered for – Detailed relationships between LOs • Use taxonomies of – Topic/subject Domains (e.g. ACM-CCS) – Learning Objectives (e.g. Bloom’s Taxonomy) – Learning Styles (e.g. Honey and Mumford) • All represented using RDF/S 16th December 2003 Personalisation in Self e-Learning 7 Networks SeLeNe Metadata Schema related to has prerequisite part of Learning Name Topics Taxonomy Learning with subject Role Contributor Object contributed by with goal Organization Learning Objectives Taxonomy Title Language Format ISA Context Level Time Program Course Module Lesson Component LOM-RDF Example “Object Oriented “Data Bases” “Web Programming” title Programming” CS 252 CS 340 title CS 359 title has part of part of has prerequisite prerequisite Object Graphical CS 240 Classes Interfaces CS 252 title has “Data title prerequisite part of part of Structures” “Object Oriented Programming” Event Swing Handling title CS 150 “Introduction to Course Module Lesson Programming” Personalisation • There are many LOs available to users of a SeLeNe; some will be useful for them and others will not • Personalised access to LOs provides learners with tools to aid the discovery of useful LOs: Views: Learner can browse the LO information space according to just the attributes of interest to them personally Search: Learners are presented with LOs relevant to their current educational needs Notification: Learners are notified of the updates and additions to the SeLeNe that are relevant to them 16th December 2003 Personalisation in Self e-Learning 10 Networks Personalised Views • The user’s view of LO descriptions and schemas can be personalised to reflect their perception of the information space • Personalised views can be browsed and queried directly • RDF View Language (RVL) allows definition and population of virtual schemas and LO descriptions 16th December 2003 Personalisation in Self e-Learning 11 Networks Personalised Query Results • These depend on a User Profile, which includes – Some PAPI-Learner elements – Some IMS-LIP and IMS-RCD elements – Additional elements to record learning goals and learning styles • LO descriptions are queried using RQL, generated from keyword-based queries – query generation takes account of the profile as well as the query – e.g. the language of LOs required can be specified • The set of LO descriptions returned by query evaluation are ranked according to the original query and the User Profile 16th December 2003 Personalisation in Self e-Learning 12 Networks The SeLeNe User Profile IMS-LIP:QCL IMS-LIP: SeLeNe: Qualifications Interests History PAPI: Personal Info IMS-RCD ims-lip: Competency competency LEARNER Taxonomy SeLeNe: messages Notifications preferences goals Catalog Entry Updated New LOs LOs Description IMS-LIP: Title Goals New Users goal Priority Date description PAPI & SeLeNE: Preferences Learning Objectives learning Taxonomy style pref. LO Accessibility Providers goal topic Learning Styles Taxonomies Descriptive Annotation Verb Learning Topics Taxonomy Style Taxonomy 16th December 2003 Personalisation in Self e-Learning 13 Networks Personalised Event and Change Notification • Users can register personal Event-Condition-Action (ECA) rules, which act over the RDF repository like traditional database triggers • This enables notification of: – Registration of new LOs of interest to the user – Changes to descriptions of particular LOs • Rules of the form: on event if condition do action are automatically generated from user input to higher-level presentation and application services 16th December 2003 Personalisation in Self e-Learning 14 Networks Open Issues • RDF query processing over P2P systems needed • Investigation of best algorithm for the personalised ranking of query results • Combination of ECA rules with transactions and consistency maintenance in RDF repositories • Design of User Interfaces for access to SeLeNe’s services • Implementation and deployment of our service-based architecture for the system 16th December 2003 Personalisation in Self e-Learning 15 Networks Further details and technical reports available from: http://www.dcs.bbk.ac.uk/selene/ 16th December 2003 Personalisation in Self e-Learning 16 Networks