Learning using QS Learning

What is Q-Sphere Learning?

Q-Sphere Learning is a comprehensive global, distributed, collaborative, multi-media, and real-time content and process (workflow) management system, with significant content aggregation and dissemination features, configured to support personalised learning on behalf of all categories of Internet users.

Key Assumptions

Our solution starts from the following assumptions:

  • Every student is unique and would benefit from personalised learning.
  • Learning is a lifelong process present in most spheres of our lives.
  • Omnipresence of computing power and communication bandwidth allows actors in learning, e.g. professors, students, authors, publishers, operational people etc. to work as efficiently, or even better, on distributed basis as on co-located basis.
  • Course material could be made so it is context-aware and adapts, to a degree, to student abilities and environment.
  • Course workflow is not a static schema even for a single student but could adapt to learning progress or be refactored as a result of acquired volumetric and meta-data.

System abstract from eLearning perspective

Q-Sphere Learning is a revolutionary learning appliance based upon strong support for an innovative cognitive tutor. It integrates teaching, learning, content provision and learning evaluation, and is capable of personalising, optimising, self-organising and self-healing.

Q-Sphere Learning focuses on specific objectives:

  • Addresses specific social and learning problems that occur during cognitive apprenticeship
  • Handles specific tasks that impose a high cognitive load, e.g., cognitive information overload during apprenticeship
  • Inherently uses a market-based model to facilitate exploitation, in fact, an innovative flexibly managed market based model is one of the base models that is used to organise and interlink learning resources

  • Highly innovative features

    Q-Sphere Learning prides itself on continues innovation and the following is a selection of innovative features used in this service:

    • Flexible interlinking decoupled from organisation of distributed Smart Objects, representing content fragments and other learning resources based upon a market-based model.
    • Course fragments can be optimally interlinked through informed hints offered to users or informed decision that can be made automatically.
    • Self-reflection of distributed operation can be used for usability adjustment, assessment, auditing and copyright management.
    • Flexible interactive and self-adapting content discovery, structuring and auto-generation.
    • Multilateral support for the main phases of cognitive apprenticeship (modelling, coaching, scaffolding etc.) is based upon semantic analysis.
    • Multi-modal support for eLearning is integrated with context-aware blended learning and personalised learning.
    • Learning resources can be personalised at multiple levels to individuals, groups and societies.

    Challenges that Q-Sphere Learning addresses:

    The core challenges that Q-Sphere Learning addresses fall into two categories:

    • Cognitive apprenticeship related
      • Undetermined level of Intelligibility of Learners hence Intelligibility of Learning Resources not matched to Learner’s level
      • No optimised personalised coaching plan to raise learner’s knowledge & ability
      • There is little explicit system support for the use of Peer-to-Peer interaction to boost learner performance
    • High cognitive load related
      • Freely available glut of information.
      • Learners often don’t understand how different concepts relate to each other or how these relationships may need to change as the teaching content gets modified and expands in content and richness.
      • Users must manually filter, organise the wealth teaching resources and be able to do this according to their individual needs

    Cognitive Apprenticeship Challenges

    • Modelling
      • QSL exposes conceptual models of domains by explicitly modelling relationships, by exposing these relationship using a powerful interactive editor and allowing users to browse and configure conceptual relationships and enabling the system to display relationships that it has discovered such as associative relationships and semantic relationships.
      • The system also builds conceptual models that are both implicit formed through monitoring user interaction and explicit models through forming user queries and quizzes for direct user input. The system can thus look at how the user’s conceptual model aligns with the domain conceptual model that is linked and derived from the learning resources. Thus the system can assess how intelligible the learning resource conceptual model is in relation to the user’s conceptual model and coach how to harmonise these.
      • QSL supports both passive and active coaching that can filter and stage the learning resources and deliver these in relationship to the user’s level (the current model of the user.) This can be used to complement lectures to provide targeted help and to allow the use to catch up and make the lectures more intelligible to users.
    • Scaffolding and fading
      • Through tracking and monitoring how the user’s conceptual models changes QSL can, at the start of an apprenticeship, give more explanations, more details about conceptual relationships (more scaffolding to support the user) but as the system detects and assesses that the user’s conceptual model is at an appropriate level it can fade this scaffolding and challenge the user more, e.g., through queries and quizzes, to expand their domain understanding with less system support.
      • QSL supports Reflection so that a learner may compare his user model with recalled domain models and set and examine plans to progress his or her performance as expressed in his user model.
      • QSL offers powerful support for exploration though directing students into problem-solving situations where the path to solution is not clearly labelled and where guidance is sparse.
    • Undetermined level of Intelligibility to Learners that is not matched to the level of the Learning Resources
      • Because does not model and cannot determine the mental models of Learners at the beginning, it cannot fine tune how it offers learning resources and interaction to users in order to deliver an optimum learning experience. In this direction, QSL adopts measurement of service and usage characteristics in the broadest sense, from big service modules down to individual user interactivity with smart objects or learning resources.

    High Cognitive Load Challenges

    • Freely Available Information Glut
      • Flood of a freely available quantity but not necessarily high quality online resources and content makes it easier for students to access content but it also makes it easier for students to surface scrape knowledge avoiding gathering a deeper understanding of knowledge, to replay other’s knowledge, to be overwhelmed by the amount of knowledge that needs to process to gain a basic understanding.
      • In this direction, QSL will use filtering and prioritising techniques and processes for making the best use of the vast availability of online resources, while also providing the necessary knowledge for deeper understanding.
    • Undirected Information Gathering and Low-valued Stored Data
      • Key-word syntactic searches will generate many results, many of which are considered irrelevant and if automatically stored will store huge amounts of data of low value. For managing this issue, QSL will deploy electronic market places of Intelligent Objects and other indexed learning resources with meta-data driven exchange characteristics that will be able to control the amount of data stored and assure their added value.
    • Insufficient Organisation and Interlinking of Data
      • Learning resources are often first developed as a closed resource to support particular learning outcomes. They are often be organised linearly and along a single dimension. This type of organisation cannot scale up and is hard to maintain as new multi-sourced learning material gets added which may appear to be contradictory or appear to offer multiple paths to explain ideas and multiple ways to organise learning resources to support a flow of ideas.
      • For addressing and managing the issue of insufficient organisation and interlinking of data, QSL offers to the users the ability for richer personalised organisation that goes beyond merely ordering and prioritising but includes interlinking. Furthermore, the addition, removal, and modification of knowledge may impact the knowledge hence metadata links may themselves self-organise.

    Existing modules and services from Q-Sphere portfolio used by Q-Sphere Learning:

    • An active document editor, where a document is an aggregation of document fragments from local or global network, some simple, some multi-media, some intelligent.
    • A distributed, real-time process (workflow) editor.
    • A global, real-time exchange.
    • A real-time messaging platform with extreme scalability and resilience, including self-healing.
    • A multi-modal communicator.
    • A finely-grained access rights module.
    • A distributed identity management module.
    • An audit module.
    • A meta-data and ontology management module.