The 5R Adaptation Framework for Location-Based Mobile Learning Systems Kinshuk, PhD Associate Dean, Faculty of Science & Technology Professor, School of Computing and Information Systems NSERC/iCORE/Xerox/Markin Industrial Research Chair for Adaptivity and Personalization in Informatics Athabasca University, Canadakinshuk@ieee.orghttp://kinshuk.athabascau.ca (jointly with Qing Tan, Xiaokun Zhang and Rory McGreal)
Overall research direction • Individualised learning in increasingly global educational environment • Bridging the gap among different types of learners • Support for: • Mobile and life-long learners • Just-in-time and on-demand learning • Context adaptation
Vision ~ Learning omnipresent and highly contextual ~ So how do we do it?
Adaptivity in ubiquitous learning Extensive modelling of learner’s actions, interactions, “mood”, trends of preferences, skill & knowledge levels, implicit and explicit changes in skill & knowledge levels Real-time monitoring of learner’s location, technology use, and change of situational aspects
Learner awareness • Personalization of learning experience through the dynamic learner modeling • Performance based model • Cognitive trait model • Learning styles
Dynamic learner modeling Mining of historical and real time data for real-time adaptivity • Learning activities • Learning style • Interests & knowledge • Problem solving activities • Learning object/activity usage • Social activities • Learner location • Location related activities
Technology awareness • Personalization of learning experience through the identification of technological functionality • Identifying various device functionality • Dynamically optimize the content to suit the functionality • Display capability, Audio and video capability, Multi-language capability, Memory, Bandwidth, Operation platform
Location awareness • Personalization of learning experience through the use of location modeling • Location based optimal grouping • Location based adaptation of learning content • Location based collaborative creation of authentic content
Location aware dynamic grouping Location Grouping Mobile Learner’s Address Mobile Learner’s Cellular Data Mobile Learner’s GPS Coordinates Mobile Learner’s Other Location Info Mobile Virtual Campus Mobile Learner’s Learning Profile Mobile Learner’s Learning Style Mobile Learner’s Learning Interests
Real-life physical objects • Personalization of learning experience as per surrounding environment • Public databases of POIs • QR Codes • Wi-Fi & Bluetooth Access Point identification • Active and Passive RFIDs
Surrounding context • Personalization of learning experience through the use of surrounding context • Identifying specific context-aware knowledge structure among different domains • Identify the learning objective(s) that the learner is really interested in • Propose learning activities to the learner • Lead the learner around the learning environment
Introduction of framework • A conceptual framework for the implementation of Adaptive Mobile Learning systems. • An ontology model of the framework in which the factors of Learner, Location, Time, and Mobile Device are considered in generating Personalized Learning Contents
Introduction of framework (cont. 1) • Challenge of facilitating mobile learning and ensuring learners’ performance: Presenting or generating personalized learning contents and instructions dynamically Dynamic Contents Learning environment and mobile device Mobile Device The context of learning process and instruction Context Aware Appropriately identifying characters of particular learner. Learner Identification
Introduction of framework (cont. 2) The challenge facing the development of location-based adaptive learning applications is the ability to deal with these contexts from Learning Perspective. One of the key strategies is to: identify and normalize context information based on efficient context-aware data fusion. semantic-based context constraints using composable ontology models.
Introduction of framework (cont. 3) The Ontology-based approach: Uses predefined metadata models of the learning contents, learner models, context information of the learning activities, and mobile device, etc. Retrieve structured and unstructured learning materials and generate personalized, just-in-time, and location-aware learning contents or adaptive “filter” that directs mobile learner to access right contents.
Introduction of framework (cont. 4) The First approach: To create semantic learning contents manually. • The Second approach: • To take advantage of pre-existing learning objects. • To develop shareable ontologies, publishing learning objects standard, and reward mobile service system to make the learning objects widely accessible. Our Research Aim: To conduct bottom-up development of the ontology for the personalized learning objectives, learning context information and proposed 5R constraint information.
Introduction of framework (cont. 5) The Third approach: To develop software and knowledge retrieval mechanism that automatically identifies appropriate learning components and extracts structural knowledge from unstructured learning contents. Learning contents are pre-developed and stored in the learning contents repository of the learning management system. Our Research Aim: To build and manipulate adaptive “filter” to direct just-in-time retrieval paths during the mobile learning processes.
5R Adaptation Framework • The Right Time: Factors: the Date-Time and the Learning Progress Timing Match! Show Contents! Alberta Legislature: Open: 09:00 AM Close: 04:30 PM Device Date Time: 03:25 PM
5R Adaptation Framework • The Right Location: The learner’s current geographic location GPS Coordination Match! Show Contents!
5R Adaptation Framework • The Right Device: • Text Contents • HD Video Contents • Flash VideoContents • Web Page Contents • Audio Contents
5R Adaptation Framework • The Right Contents: Learning objects, learning activities, and leaning instruction LO A1 Back Home Screen Shot Design Manner of Legislature Building Take Picture Screen Shot Back Home LO H3 LO P2 LO A1 Take Picture
5R Adaptation Framework • The Right Learner: • Practical English • Computer Science • Physical Education • Art • Mathematics
Implementation of framework The First layer: Consists of “Location”, “Time”, “Learner”,“Device”, and “LearningContents”, respectively representing the five adaptation inputs. The Second layer: Furtherdescription of information or data of each adaptation input. The ontology scheme, namely, the relationshipsamong the adaptation inputs,which illustrates why the inputs need to be described and how the inputs are interconnected.
Framework application scenario: Field trip Location-based mobile fieldtrip applications at a zoo
Framework application scenario: Field trip Location-based mobile fieldtrip applications using visualized interaction with dynamic geospatial data
5R adaptation features in the fieldtrip scenario Location-based experiment Lab interface and visualized interaction on mobile devices. Dynamic annotation or blog on the visualized semantic physical object model. FieldtripScenario Adaptive learning content retrieval constrained by the location and ongoing fieldtrip activities. Real-time sharing experience between students and others who are in the field or in remote areas via visualized virtual interaction interface. Visualized fieldtrip plan, real-time activity collaboration and monitoring during the fieltrip
Framework application scenario: RFID Classroom System Screenshot: User Login • System Screenshot: Learning Object - RoundExicter @ Location code 80
Framework application scenario: RFID Classroom • System Screenshot: Learning Object - RectangularExicter @ Location code 3F