A Design for a Hypermedia-Based Learning Environment


Ossi Nykänen

Department of Mathematics, Digital Media Institute

Tampere University of Technology


Email: ossi.nykanen@cc.tut.fi

Martti Ala-Rantala

Department of Mathematics, Digital Media Institute

Tampere University of Technology


Email: martti.ala-rantala@iki.fi




Our paper presents a design for a Hypermedia-Based Learning Environment, HBLE for short. It is an environment for teaching and learning mathematical sciences with distributed interactive hypermedia systems. HBLE offers tools and methods for course development, teaching, maintenance and different learner-centered studying strategies. The system also has information acquisition functionality for research purposes.

HBLE may be used either as learning material to supplement a regular university course or by an independent student. In either case, graded examinations will still retain their traditional form. The system is integrated with our university’s student registration and course enrollment databases.

Material Structure

As well as pedagogical and technical viewpoints of a learning system, the appropriate structure of the course material is one of the key factors in succesful learning environment design. Good material structure is the essential basis for creating versatile content materials and functionality suitable for both distance learning and tutored classroom studying.

From the viewpoint of content structuring, the basic properties of any good (computerized) learning environment are quite clear. The three key elements are rich and yet logical content of the content material, smart interaction mechanism supporting the working style and the level of knowldge of the individual student and motivating response from the whole learning process itself.

We have decided to formulate the overall course structure using graph data model with fixed link semantics where each node of the graph equals to some hypermedia content. This approach has some very nice implications that can be efficiently used in adaptive hypermedia in, e.g., student modeling.

From the learner's point of view the learning material is presented as a directed acyclic graph, knowledge graph. Learning material is divided into cells. Each cell introduces a single topic consisting one simple issue to be studied, e.g., a definition or a theorem. Arcs define the prerequisite relations between cells; to fully understand the content of a cell one must understand all prerequisite cells of that cell.

Each topic is presented as a small hypermedia document consisting of static multimedia elements (text, graphics, audio and video materials) and interactive hypermedia elements (interactive examples and exercises). Cells provide both the content and means for studying with theory, example, exercise and test elements.

In order to provide context and order to studying, individual cells may also be grouped as virtual sections. Virtual sections present no new information but define logical groups of topics instead. They bind the relevant topics together in a meaningful fashion providing overview of the topics discussed with general background, examples, exercises and tests. Virtual sections may be combined into even larger entities.

An approach using relatively small, linked knowledge elements has many advantages. First, it makes possible to effectively define specific goals and strategies for studying. Second, each student's knowledge of each topic can be estimated separately enabling automated navigation guidance and tutoring and third, the hierarchical structure of information is stated explicitly.

The graph-like structure enhanced with student modeling provides also the basis for adaptive hypermedia techniques. Link-level adaptation may be achieved quite straightforwardly from the graph structure itself and content-level adaptation can be implemented using, e.g., appropriately graded elements in hypermedia contents.

System Architecture

The high-level architectural design of HBLE is a distributed and modular object-oriented framework facilitating future extensions like additional media types, enhancements to the graph theory and teaching subjects other than mathematics.

The user interface is implemented on top of a Web browser with Java 1.1 and frames support (Netscape Communicator 4.0 or Microsoft Internet Explorer 4.0). In addition to the user interface, a major part of the intelligence and functionality of the system will be located in the client end, realized as Java applets. A more conventional implementation using CGI scripts on the server side would make the system very slow and sensitive to network problems.

Extended HTML is used for storing the learning material components on the server side. The extensions are needed because HTML mostly concerns layout matters instead of semantic meanings. Various parsers and generators create the run-time presentation of the material by filtering the static components through a student model. Exercises and complex examples are implemented on top of a Java applet framework. Maple engine, wrapped in a CORBA object, is used in the learning material for checking the answers to exercises and creating new instances of exercises. The data collected about the learners and the exercise material are stored in an external database, which facilitates effective information acquisition and makes the use of the collected material for research purposes easy and straightforward.

Some of the tools intended for users other than students (administrators, authors) are located on the server side and cannot be used via Web. This concerns mostly tools that allow destructive or otherwise critical operations..

From software engineering point of view, the learning environment is based on a domain object model that is distributed via Internet using CORBA. This architecture is reminiscent of the three-tier architecture well-known among distributed database application designers. The model consists of the student model, learning material and administrative information. The object model is distributed over the network with in such a manner that only operations dealing with object persistence and building the learning material from the components stored in the databases are performed on the server side. Plain HTTP does not offer adequate platform for this type of architecture; therefore CORBA must be used. HTTP and regular Web operations are used to transfer the hypermedia material over Internet.

The server runs in a Sun Unix workstation. It consists of the implementations of CORBA objects written mostly in Java and SQL except the Maple engine interface that is realized in C. The learning material and other information handled by the system are stored in part in Oracle database tables and in part in Unix files.