rec systel 2012 competency based recommendation
TRANSCRIPT
Competency Comparison Relations for Recommendation in Technology Enhanced Learning Scenarios
Gilbert Paquette, Délia Rogozan, Olga Gilbert Paquette, Délia Rogozan, Olga MarinoMarino
www.licef.ca/cice
Canada Research Chair in Instructional and Canada Research Chair in Instructional and Cognitive Enginerring (CICE)Cognitive Enginerring (CICE)
LICEF Research CenterLICEF Research Center
Télé-universitéTélé-université
RecSysTEL Workshop 2012RecSysTEL Workshop 2012Saarbruecken September 18, 2012Saarbruecken September 18, 2012
Background
Add semantic references to scenario components: actors, tasks and resources to educational modeling languages such as IMS-LD (2003)
– Paquette and Marino, 2005
“Include the improved modeling of users and items, and incorporation of the contextual information into the recommendation process”
– Adomavicus and Tuzhilin (2005)
The “Adaptive Semantic Web” opens new approaches for recommenders systems: use of folksonomies and ontological filtering of resources
– Jannach et al, 2011
Recommendation (assistance) Recommendation (assistance) PrinciplesPrinciples
Epiphyte – grafted on the scenario process Epiphyte – grafted on the scenario process
but external to it; no scenario modificationbut external to it; no scenario modification
Multi-agent system: agents are associated to Multi-agent system: agents are associated to
tasks at different levels in the scenariotasks at different levels in the scenario
Flexible association: one, some or all of the Flexible association: one, some or all of the
tasks are assisted.tasks are assisted.
Delegation between a task agent towards its Delegation between a task agent towards its
super tasks agents; tree topologysuper tasks agents; tree topology
InsertionInsertion of recommenders of recommenders (assistance agents): an example(assistance agents): an example
The implemented recommender The implemented recommender modelmodel
Recommender = {rules}Recommender = {rules} Rule = <actor, event, condition, action >Rule = <actor, event, condition, action > Event = Event =
– Activity transition Activity transition (started, terminated, revisited,…)(started, terminated, revisited,…)– Time spent (activity, global …) Time spent (activity, global …) – Resources opened, reopened,…Resources opened, reopened,…
Condition = boolean expression comparing: Condition = boolean expression comparing: – Target actor progress in the scenario + Target actor progress in the scenario + knowledge and knowledge and
competencies acquired + evidence => competencies acquired + evidence => User persistent modelUser persistent model
– Resources: prerequisite and target competenciesResources: prerequisite and target competencies
– Activities: prerequisite and target competenciesActivities: prerequisite and target competencies
Action = advice, notification, model updateAction = advice, notification, model update
Semantic Referencing of Resources
Of what– Actors, activities, documents, tools, models, scenarios …
Why– Help select resources at design time for better quality scenarios
– Inform users of the resources’ content at design or delivery time
– Assist users according to their knowledge and competencies
How– Associate formal semantic descriptors to resources from a
domain ontologies and/or competencies based on ontology references
Knowledge Descriptors
Classes and instances
(From OWL-DL domain ontologies)General properties:
–Domain – Data Properties –Domain – ObjectProperty – Range
Instanciated properties (facts):–Instance – Property–Instance – Property – Value
Competency Descriptors
Knowledge descriptors
Competency descriptors
– (K, S, P) triples(K, S, P) triples K: Knowledge descriptorK: Knowledge descriptor
– From a OWL domain ontologyFrom a OWL domain ontology
S: Generic SkillS: Generic Skill– From a 10-level taxonomy From a 10-level taxonomy (Paquette, 2007)(Paquette, 2007)
P: Performance levelP: Performance level– A combination of P-values A combination of P-values (Paquette, 2007) (Paquette, 2007)
S=ApplyS=ApplyS=ApplyS=Apply
P=ExpertP=ExpertP=ExpertP=Expert
K=PlanetK=PlanetK=PlanetK=Planet
Referencing Process in the TELOS Implementation
OntologyOntologycontructioncontructionor importor import
… and/or competencies
ResourceResourceselectionselection1111 2222
SemanticSemanticReferencingReferencingOf resourcesOf resources
3333
Semantic Search Methods
Type de rechercheType de recherche Type de résultatType de résultat
Simple Using key words from the ontology
AdvancedUsing knowledge and competency Using knowledge and competency boolean queryboolean query
Resource PairingUsing semantic comparison between queried ressource and other resources
→ → Rests on knowledge and competency comparisonRests on knowledge and competency comparison
Exact matchExact match
Exact matchExact match
Semanticallynear match
Semanticallynear match
Exact matchExact match
Knowledge Comparison (K1 et K2)
Based on the Based on the structure of the ontology where the of the ontology where the knowledge descriptors are storedknowledge descriptors are stored
Compare the Compare the neighbourhoods of K1 and K2of K1 and K2
Possible resultsPossible results– K2 K2 near and more and more specialized / / general than K1 than K1
Competency Comparison
Based on knowledge Based on knowledge comparison ((KK))
Base on Base on the distance between skills’ levels (between skills’ levels (HH) ) and and performance levels distances(performance levels distances(PP))
Possible resultsPossible results C2C2 veryNear / Near C1 C1 C2C2 stronger / weaker than C1than C1 C2 more C2 more specialized / general than C1than C1
C1=(K1, S1, P1) et C2=(K2, S2, P2)C1=(K1, S1, P1) et C2=(K2, S2, P2)
Competency ComparisonCompetency Comparison
Competency comparison Competency comparison within rule conditionswithin rule conditions
A competency-based condition is a triple:– ObjectCompetencyList is the list of prerequisite or target
competencies of another actor, a task or a resource to be compared with user’s actual competency list
– Relation is one of the comparison relations : Identical, Near, VeryNear, MoreGeneric, MoreSpecific, Stronger, Weaker, or any combination of these.
– Quantification takes two values: HasOne or HasAll
EX: HasAll /NearMoreSpecific / Target competencies for Essay EX: HasOne/Weaker/Target competency for Build Table activity
Recommendation exampleRecommendation example
Notification exampleNotification example
User model updateUser model update
Achievements in this project
Extension of the TELOS Technical Ontology for semantic referencing of resources, search and recommendation
Definition of a Typology of semantic descriptors (ontology descriptors and competenciers)
Search methods for resources ‘identical’ ou ‘near’ sémantically
Recommendation Model: based on competency comparison between actors, tasks and resources
New integrated suite of tools: Semantic referencer, Semantic search tools, Competency and Ontology editors, Integration to recommanders scenarios, Recomenders’ rule editor.
Future stepsFuture steps
More experimental validation to refine the semantic relations
between OWL-DL references, i.e adding weights to the various
comparison cases
Investigate recommendation methods for groups in collaborative
scenarios (permitted by our model of multi-actor learning scenarios)
Improve the practical use of the approach, partly automate tasks,
improve the ergonomics
Investigate the integration of other recommendation methods (e.g.
user analytics)
“Free” the suite of tools from TELOS to extend its usability on the
Web of data.
Questions ?Comments ?
Gilbert Paquette, Délia Rogozan, Olga Gilbert Paquette, Délia Rogozan, Olga MarinoMarino
www.licef.ca/cice; www.licef.ca/gp
Canada Research Chair in Instructional and Canada Research Chair in Instructional and Cognitive Enginerring (CICE)Cognitive Enginerring (CICE)
LICEF Research CenterLICEF Research Center
Télé-universitéTélé-université
RecSysTEL Workshop 2012RecSysTEL Workshop 2012Saarbruecken September 18, 2012Saarbruecken September 18, 2012