More thinking about xApi and IMS Caliper - Structural/Syntactic & Ontological Mapping
3 likes1,907 views
The document discusses the interoperability between xAPI and IMS Caliper regarding learning analytics, focusing on structural, syntactic, and ontological mapping. It outlines the implications of both specifications for capturing learning experience data and the challenges in mapping them without proper ontology guidelines. Additionally, it raises questions for stakeholders on how to effectively design and profile learning data for better interoperability.
More thinking about xApi and IMS Caliper - Structural/Syntactic & Ontological Mapping
- 1. More thinking about xAPI and IMS Caliper - Structural/Syntactic & Ontological Mapping - Korea Education & Research Information Service Yong-Sang Cho, Ph.D zzosang@keris.or.kr FB: /zzosang Twitter: @zzosang
- 2. This slides is following thinking from “quick review for xAPI and IMS Caliper” (ISO/IEC JTC1 SC36/WG8 first webinar in Nov. 11, 2015) http://www.slideshare.net/zzosang/quick-review-xapi-and-ims-caliper-princi ple-of-both-data-capturing-technologies
- 3. xAPI Transcript/learning data can be delivered to LMSs, LRSs or reporting tools Experience data LMS: Learning Management System LRS: Learning Record Store
- 4. IMS Caliper <Source: New Architect for Learning (Rob Abel, 2014) http://www.slideshare.net/JEPAslide/day3-edupub-tokyoims?qid=76ce5d4a-1ccf-468f-a428-c652584c395a&v=default&b=&from_search=4>
- 5. Principle of both specifications Subject Triple Bindings Predicate Object With contexts information Learning Applications Generated (objects) Outcomes Courseware Group Timestamp
- 6. “The Experience API is a service that allows for statements of experience to be delivered to and stored securely in a Learning Record Store (LRS). These statements of experience are typically learning experiences, but the API can address statements of any kind of experience. The Experience API is dependent on Activity Providers to create and track these learning experiences“ < from the specification of xAPI, https://github.com/adlnet/xAPI-Spec/blob/master/xAPI.md#roleofxapi >
- 7. Implications of xAPI are (from the perspective of learning analytics interoperability) • xAPI is an well designed structural/syntactic architecture for learning experience (or activity) data • Predicates (a.k.a “verb” in xAPI) may be profiled (a.k.a recipe) in application domain, because the xAPI does not specify any particular Verbs • Object is well designed for SCORM object and interaction based on interactions of cmi data model • But, to use Object in more diverse learning situations this needs to be profiled for specific purposes with controlled vocabulary of object types • xAPI data model is well relevant to JSON binding like IMS Caliper -> It seems not difficult to make interoperable data between specs in terms of structural/syntactic mapping. Note: From xAPI to IMS Caliper mapping may not be easy without ontology mapping
- 8. “The purpose of the IMS Caliper project is to define a standard for enabling the collection of rich contextual data about learning interactions and a Sensor API™ for capturing and reporting this data. This work will enable learning environments to capture data from learning interactions and share it with other learning environments and consumers of learning analytics. “ < from IMS Caliper Implementation Guide, http://www.imsglobal.org/caliper/caliperv1p0/ims-caliper-analytics-implementation-guide#1 >
- 9. Implications of IMS Caliper are (from the perspective of learning analytics interoperability) • IMS Caliper is started with minimum metric profile for learning activity like a lean-startup -> It may be extended to be wider due to feed back from adopters • It may be met with intersection point between too much detail (or complex) and simple (or ambiguous) • Sensor APIs per learning activity need to be well combined with existing learning environments to generate value of properties, such as ‘generated’ or ‘target’ • Caliper data model is well relevant to JSON binding like xAPI -> It seems not difficult to make interoperable data between specs in terms of structural/syntactic mapping as well. Note: From IMS Caliper to xAPI in terms of downcasting mapping may be easy without ontology mapping
- 10. Event Store Learning Record Store IMS Caliper Sensor APIs xAPIs Mapped to JSON data Instance idea for interoperable data between IMS Caliper and xAPI Phase 1. Structural/Syntactic mapping rule Phase 2. Ontological mapping rule
- 11. • How to mapped between xAPI and IMS Caliper data? i.e. see next slide • Is there any principle (or guideline) for design (or profiling) of learning data in terms of learning analytics interoperability? i.e. to use xAPI for specific purpose or adopt IMS Caliper profiles • How to make ISO/IEC 20748 Learning analytics interoperability to be practical standards (TRs) for diverse stakeholders? - Part 1: Reference model - Part 2: System requirements - … - Part x: Principle of data design enabling learning analytics interoperability - Part y: Guideline for mapping learning activity (experience) data Question list to Study Group of SC36/WG8:
- 12. P1. Potential example for structural/syntactic mapping rule between specs <IMS Caliper properties of assignable> <xAPI Statement properties>
- 13. P2 (a). Potential example for ontological mapping rule between specs <IMS Caliper> <xAPI + Recipes> Class Class http://www.imsglobal.org/caliper/ http://adlnet.gov/expapi/Entities … Concept tree Property/relation Property/relation Concept detail tree {actor, action, event, target, generated, etc…} {actor, verb, object, context, etc…} Instance Instance { “action”: “completed” } { “verb”: “finished” } Instance Table - ontology mapping rule (?) Structural/ Syntactic Mapping Semantic Mapping (under assumption xAPI’s recipes are looked as single form)
- 14. P2 (b). Potential example for ontological mapping rule between specs Semantic Filter/ Mapper IMS Caliper Sensor APIs xAPI – recipe (a) xAPI – recipe (b) xAPI – recipe (c) … Ontology Repo (for common sense) (under assumption xAPI’s recipes are looked differently)
- 15. Question list to Stakeholders related to xAPI and IMS Caliper: • Can you make use cases to exchange interoperable data? i.e. from IMS Caliper to xAPI or reverse or both? • How much SC36/WG8 experts make detail guideline(s) to describe interoperable data exchange or flows? i.e. what kinds of items for the works will be included in the scope or TR(s) • How do stakeholders understand xAPI recipes? Is it single form or different? It cause critical factor to decide for the direction of guideline. Note: this question lists need to be completed at the WG8 meeting
- 16. More Questions? Korea Education & Research Information Service Yong-Sang CHO, Ph.D zzosang@gmail.com FB: /zzosang Twitter: @zzosang