The potential for student performance prediction in small cohorts with minimal available attributes
British Journal of Educational Technology
June 25, 2019
The measurement of student performance during their progress through university
study provides academic leadership with critical information on each student’s
likelihood of success. Academics have traditionally used their interactions with
individual students through class activities and interim assessments to identify those “at
risk” of failure/withdrawal. However, modern university environments, offering easy
on-line availability of course material, may see reduced lecture/tutorial attendance,
making such identification more challenging. Modern data mining and machine
learning techniques provide increasingly accurate predictions of student examination
assessment marks, although these approaches have focussed upon large student
populations and wide ranges of data attributes per student. However, many university
modules comprise relatively small student cohorts, with institutional protocols limiting
the student attributes available for analysis. It appears that very little research attention
has been devoted to this area of analysis and prediction. We describe an experiment
conducted on a final-year university module student cohort of 23, where individual
student data are limited to lecture/tutorial attendance, virtual learning environment
accesses and intermediate assessments. We found potential for predicting individual
student interim and final assessment marks in small student cohorts with very limited
attributes and that these predictions could be useful to support module leaders in
identifying students potentially “at risk”.
Tags: AI, Edtech, Predictive Analytics
"Developing an Agent-Based Simulation Model of Software Evolution"
Journal of Information and Software Technology
April 08, 2018
Context: In attempt to simulate the factors that affect the software evolution behaviour and possibly predict it, several simulation models have been developed recently. The current system dynamic (SD) simulation model of software evolution process was built based on actor-network theory (ANT) of software evolution by using system dynamic environment, which is not a suitable environment to reflect the complexity of ANT theory. In addition the SD model has not been investigated for its ability to represent the real-world process of software evolution.
Objectives: This paper aims to re-implements the current SD model to an agent-based simulation environment ‘Repast’ and checks the behaviour of the new model compared to the existing SD model. It also aims to investigate the ability of the new Repast model to represent the real-world process of software evolution.
Methods: a new agent-based simulation model is developed based on the current SD model's specifications and then tests similar to the previous model tests are conducted in order to perform a comparative evaluation between of these two results. In addition an investigation is carried out through an interview with an expert in software development area to investigate the model's ability to represent real-world process of software evolution.
Results: The Repast model shows more stable behaviour compared with the SD model. Results also found that the evolution health of the software can be calibrated quantitatively and that the new Repast model does have the ability to represent real-world processes of software evolution.
Conclusion: It is concluded that by applying a more suitable simulation environment (agent-based) to represent ANT theory of software evolution, that this new simulation model will show more stable behaviour compared with the previous SD model; And it will also shows the ability to represent (at least quantitively) the real-world aspect of software evolution.
Tags: Design Thinking, Emerging Technology