and Importance

Traditional forms of learning are static, which provides learners with a one-size-fits-all set of learning materials. This means that all learners receive the same learning materials, regardless of the fact that learners understand various subjects at different paces and require different levels of depth for different subtopics.

This results in the rise of Adaptive Learning systems. Adaptive Learning systems aim to provide learners with a personalised set of learning materials that best suit their needs. However, current Adaptive Learning systems are not without their flaws.



GALMAT is an Adaptive Learning system that gives learners a personalised learning experience, by providing them with tailored learning materials as opposed to one general set of learning materials. This is achieved by providing the learner with a set of questions and then using their answers to the questions to determine their abilities in various subjects. Unique learning materials will then be generated for each learner, dependent on these abilities. This is a repeated process that ensures that the learning materials continuously reflect a learner's current knowledge and abilities.


Automatic Question Generation

The AQG component automatically generates a set of questions for a learner, by using the learner's knowledge of certain topics as a guide as to what questions the system should ask the learner. The AQG aims to ask high-quality questions in terms of grammaticality, understandability, adaptability, and suitability; and these questions should assist in ensuring that the system is able to identify a learner's weaker subjects so that it can provide extensive learning materials for that subject.


Adaptive Learning System

The Adaptive Learning System's primary objective is to identify any potential knowledge gaps learners' might have within a specific domain. It achieves this by evaluating a learner's proficiency across a range of topics through the application of Item Response Theory, a mathematical framework. The system's ultimate goal is to not only identify these knowledge gaps but also to actively address them by offering tailored, personalised learning materials. This approach is designed to enhance both the efficiency and engagement of the learning process.


Natural Language Generation Algorithm

The NLGA component generates the project’s final output which is an informative document comprising comprehensive notes on subjects that the system identifies as being the learner’s weak points. This final section of the project aims at producing detailed and human-like sentences. The amount of details and context to give is also varied based on how much the learner knows on those specific subjects so that the document is more tailored to the learner.