HE institutions need to determine how best to serve this new demand. The willingness to accommodate the changing student population can be seen in the increased number of modes within which a student can study: full-time, part-time and by distance learning. However, lifelong learning demands a flexible mode of delivery which spans all of these. Although distance learning has had considerable success, many students fail to complete courses. A key factor in this is the sense of isolation amongst distance learners and a perceived lack of support.
Internet technologies offer a new range of opportunities to educators for the teaching and learning process. In this paper we discuss one approach, using the VALIENT environment, which is being built specifically to support flexible and distance learning. This environment aims to offer the maximum flexibility to learners whilst maintaining a degree of support and interaction that will reduce the sense of isolation and alleviate the problems of studying primarily alone.
Entity Relationship (ER) Modelling (Chen, 1976) is a fundamental technique of database design, used to capture the data requirements of an information system. However, it is an activity where novices exhibit considerable difficulties both in performing the task and understanding the procedures used for it. We have developed an intelligent learning environment that provides students with the opportunity for practicing their database design skills (Hall and Gordon, 1998; Gordon and Hall, 1998). This environment is interactive, on-line, real-time and text based. It provides a virtual learning environment where learners exist within a textual scenario (typically a description of a business area) from which they need to develop an ER model.
The learner is supported within this environment by the existence of intelligent agents who provide direct, immediate feedback to the learner whilst they are attempting to perform their tasks. In addition, the learner is monitored to seek to identify what aspects the student does not appear to understand in relation to the task. Figure 1 illustrates our ER modelling environment running on top of a web browser.
A key problem lies in determining which information to provide, and how to provide that information. In an environment where the learning is student-centred or independent, and where educator support will be minimal (in that the learning is remote), it is essential that technological means are used to ensure that students are presented with appropriate, timely, relevant information.
Our approach to this problem is based on observing the learner's behaviour within the Intelligent Learning Environment. By doing this, the learner can be presented with information, via the WWW, which is precisely tailored to their current tasks, and to the current problems that they are experiencing in trying to perform these tasks. The learner may, for example, be experiencing difficulties with certain aspects of the ER modelling task (e.g. misidentifying potential entities). Once these problems have been identified, the learner can be directed to explanatory material, worked examples of similar (or simpler) modelling problems, or to fuller explanations and support materials concerning the errors that they have been identified as making in the ILE.
VALIENT is essentially based around an intelligent web server system that replaces the functionality of conventional web servers and coordinates the presentation of learning material to a learner. The web server provided in VALIENT is intelligent, and adaptive. As detailed, the presented material is individually tailored to students needs. Figure 2 identifies this architecture, showing the communications between the web and the intelligent learning environment.
A second level of support is provided through the use of bulletin boards that enable students to post queries relating to aspects of the course that they do not fully understand. Although a staff member monitors these boards, the aim of their use is for other students to respond to their peers. This can provoke discussion, albeit asynchronous, and also helps to develop a feeling of community amongst separate learners. The educator will respond to queries if it is clear that students do not understand certain topics, or again they can be directed to the specially constructed mini-tutorials.
Our own work and that of previous researchers has clearly indicated that each of the component technologies that make up VALIENT are useful in their own right. However, we believe that the synergy resulting from the integration of these technologies will provide learners with a much richer learning experience.
Although VALIENT is being tested in the domain of database design, it appears that it will be useful across a number of different domains. VALIENT is itself domain independent and provides a framework within which to base the learning experience. We believe that this environment will have considerable utility and will provide a new and effective way for students to learn.
Chen, PP (1976) The entity-relationship model - towards a unified view of data. ACM transactions on database systems 1(1) pp 9-36
Dillenbourg, P., Mendelsohn, P., Schneider, D.(1994). The Distribution of Pedagogical Roles in a Multi-agent Learning Environment, in Lewis, R. & Mendelsohn, P., (Eds): Lessons from Learning, Amsterdam: North-Holland.
Gordon, A. and Hall, L. (1998b). A Collaborative Learning Environment for Data Modelling. In Proceedings of the 1998 Florida Artificial Intelligence Research Symposium. Sanibael Island, FA: pp. 158-62
Hall, L., and Gordon, A. (1998) A Virtual Learning Environment for Entity Relationship Modelling. In Proceedings of the ACM SIGCSE Conference. Atlanta: GA, pp. 345-49.