Towards The Development of an Engine that Marks Descriptive Examinations For E-Learning

Abstract

Traditional assessment of skills uses sets of human markers to mark candidates’ scripts by hand. This may lead to inaccurate results and slow feedback to the students as this task requires significant amount of time, effort and concentration by the marker. The results are also inconsistent when marked by a group of examiners rather than a single marker. Automation of the marking can improve the accuracy and consistency of the results and automation of the management of the process improves the quality of the assessment. This research aimed at identifying the means by which an automated E-marking engine can be generated specifically to mark descriptive examinations. Key words were employed using pattern matching that enabled this automation to be implemented. Each keyword was designated a value and total marks for each examination done were obtained