Campus Sorocaba

Luciana A M Zaina

Adaptive learning in the educational e-LORS Adaptive learning in the educational e-LORS system: an approach based on preference system: an approach based on preference categories categories

Available in:1)Complete work: http://www.icmc.usp.br/~junio/PublishedPapers/IJLT060402%20ZAINA.pdf (in English) http://dx.doi.org/10.5753/RBIE.2012.20.1.04 (in Portuguese)

2)Partial contribution: http://dx.doi.org/10.1145/1878450.1878488 (in English)

Campus Sorocaba

IntroductionIntroduction The recommendation process occurs through the

investigation of the user’s preferences: Based on information obtained from the student explicit and

implicit learning practices

The dynamic linkage between contents and student learning profiles may enhance the adequacy of the learning objects that will be offered to the students

The observation of learning styles provides users with different teaching strategies, meeting the student’s individual needs: learning style should be observed through different dimensions

achieving diverse aspects of her/his preferences.

Campus Sorocaba

Paper ObjectivePaper Objective An approach to design the student interaction based on

the dynamic recommendation of e-learning objects. The recommendation is based on:

the theme of learning the student learning profile: are described by a set of

preference categories that describe the student learning preferences.

A relationship between the learning profiles and the learning objects is drawn upon the linkage of the preference categories to the metadata fields of the learning objects.

The approach was validated in a regular college course on Computer Engineering Data Structures.

Campus Sorocaba

Related ConceptsRelated Concepts

Important issues to support the approach: Recommendation strategies Learning profiles Learning objects

Campus Sorocaba

Recommendation StrategiesRecommendation Strategies In general, the recommendation process is accomplished

by the following recommendation approaches: collaborative filtering: provides personal recommendation

on a group-based fashion, adjusting to sets of people with similar preferences and interests.

content-based filtering: concerns the content rather than the users. To do so, it learns about the most relevant contents based on the features derived from the objects that the user has accessed.

rule-based filtering: concerns in the definition of rules to control the adaptation process.

hybrid filtering: combinations of other techniques.

Campus Sorocaba

Learning ProfileLearning Profile The learning style involves the strategies that a student tends to

apply frequently to a given teaching situation. The Felder-Silverman Learning Style Model is describe by

dimensions of Learning and Teaching Styles, creating a relationship to learning styles and teaching strategies that could be adopted to support the student learning style.

The Felder-Silverman model was selected to this work, because it's close relationship to learning styles and teaching strategies, resulting in an adherence between these aspects.

Campus Sorocaba

Dimensions of Felder-Silverman Dimensions of Felder-Silverman Learning Style ModelLearning Style Model

Learning Style Teaching Strategies

Features

sensory concrete It is related with the perception of content.intuitive abstract

visual visual It is related with the format of content presentation.auditory verbal

active active It is related with the student participation in the activities.reflective passive

sequential sequential It is related with the best order to present the content: step-by-step progression or a overview first of content.

global global

Campus Sorocaba

Learning ObjectLearning Object It can be defined as an entity to be applied in a teaching-

learning process. e-learning: the aim is to create contents in digital formats.

Metadata usually is adopted to organize learning objects, improving their reuse.

The LOM (Learning Object Metadata) standard of the Institute of Electrical and Electronics Engineers – IEEE is the metadata specification used in the area of learning objects. It has a structure that describes learning objects through

descriptor categories.

Campus Sorocaba

LOM CategoriesLOM CategoriesLOM Category LOM Field Characterization

General Identifier, Type, Title, Language, Description and Keywords.

General description.

Technical Media Format (video type, sound), Size, Physical location, Requirements (object use: software version, for example).

Technical features description.

Educational Interactive type (active, expositive) Educational function

and pedagogicalcharacteristics object description.

Learning Resource Type (exercise, simulation, questionnaire)

Campus Sorocaba

e-LORS Featurese-LORS Features

e-Learning Object Recommendation. The student learning profile is:

a part of student model and splits the student learning profile into three categories, based on Felder and Silverman dimensions: Perception, Participation e Presentation Format

we called Preference Categories: its goal is to detect clusters of preferences that reflect different data perspectives caught during the tracking of learning styles.

The main goal is to recommend learning objects according to student preferences.

Campus Sorocaba

e-LORS Overviewe-LORS Overview

retrieve LO references (theme)

check the learning profile

LOMRepository

e-LORS

Concept-based

filtering

Learning profile-based

filtering

references of retrieved LOs

request LO references (theme)

recommended LOs references

Categories of Preferences

Perception

Presentation Format

Student Participation

A

B

C

E

D

Campus Sorocaba

Step A – Request Step A – Request

e-LORS starts the recommendation process by receiving the theme parameter.

This parameter describes the topic of interest according to the metadata defined by the LOM standard.

The theme is used to determine which LOs are to be considered for matching the learners’ profiles.

Campus Sorocaba

Step B and C – Step B and C – Concept-based Concept-based filtering filtering

Concept-based filtering searching the LO references in the LOM-based repository.

The system seeks for learning objects that match the theme parameter according to the fields of title, description and keywords, which belong to the LOM’s General Category.

Step C presents the references of retrieved LO.

LOM Category LOM Field Characterization

GeneralIdentifier, Type, TitleTitle, Language, Description Description and KeywordsKeywords.

General description.

Campus Sorocaba

Step D and E– Step D and E– Learning profile-based Learning profile-based filtering (1)filtering (1)

Consider the first outcome of references achieved in the previous step.

The student Preference Categories reported in the learning profiles are compared to the Interactive and Learning Resources found in the Educational LOM standard category

Campus Sorocaba

Step D and E– Step D and E– Learning profile-based Learning profile-based filtering (2)filtering (2)

LOM – Educational Field

Educational Field Values

Teaching- Learning

correspondence

Preference Category

InteractivityActive Concrete-Sensing

PerceptionPerceptionExpositive Abstract-Intuitive

Learning ResourceType

Figure, Video, Film, and others

Visual-VisualPresentation-Presentation-

FormatFormatText, Sound, and others

Verbal-Auditory

Practical Exercise, Experiment, and

othersActive-Active

Student Student ParticipationParticipation

Questionnaire and Readings

Passive-Reflective

Campus Sorocaba

e-LORS Validation (1)e-LORS Validation (1)

Regular college course on Computer Engineering Data Structures with 50 students.

To identify the students’ profiles (Preference Categories) we applied the questionnaire of Soloman e Felder.

Steps of Validation:1. we have cataloguedcatalogued several learning objects learning objects for the topics on

Data Structures, what allowed us to attend different learning styles during the experience.

2. e-LORS in Action – concept and learning profile-based filtering process.

Campus Sorocaba

e-LORS Validation (2)e-LORS Validation (2)

Result: system e-LORS returns the list of recommended learning objects and the process is concretized with the building of the student workplace with the recommended objects.

User Satisfaction Evaluation: the evaluation of the students’ perception of the system indicated the student satisfaction with the recommended learning

objects. the students reported that the workplaces achieved with the use of

e-LORS were, in fact, more adequate than what would be achieved with casual browsing.

Campus Sorocaba

Conclusions and future works Conclusions and future works The development of flexible educational environments

that are adaptable has become an important requisite within the teaching-learning process.

The association between learning profiles and learning objects metadata grants dynamism in the content retrieval process.

Future directions: One important subject for future work is to extend

the proposal to considering to the retrieval mechanism the features of mobile learning as differences between devices.

Campus Sorocaba

Thanks!Thanks!

lzaina@ufscar.brlzaina@ufscar.br