During my final year of my BSc Software Engineering degree at Durham University, I undertook a project investigating distributed networks within the field of profile-based e-Learning, culminating in the creation of a test network, and the publishing of my findings.
This page is still under construction – please visit again soon for more content!
The field of e-Learning has a wide variety of different concepts, some of which I amalgamated together in forming this project. My examination of the field can be read in my Literature Survey, where I draw detail from a variety of papers written by many authors (some of which also influenced the final paper)
Full details of the design of my distributed profile-based e-learning network can be found in my Design Report. Put simply – I set out to create a scalable model for content sharing between a series of servers, which could be queried by clients, all based around the principles of user profiling, even after the content is distributed.
My implementation of a distributed e-Learning network was achieved through a series of servers running a PHP script listening for incoming RPC messages, which were then interpreted and dealt with accordingly. The specification for the XML-RPC methods involved can all be found in my final paper (see below). These servers then interfaced with each other through more RPC methods in order to form the network (this is examined in more detail in the paper, and other ways of interfacing are discussed), which shared the e-Learning content and implemented content selection algorithms based upon user profiles. This was achieved by use of ‘content tagging’ over the network, something which is discussed in detail within the design report and final paper.
In order to interact with this network, I also wrote a client program (again using PHP) that implemented the RPC methods on the server in order to search and display content.
The final paper I wrote on the topic, giving detail of my implementation, as well as suggesting future directions that this work could take, is available here.