Comparison Analysis of Rabin-Karp and Winnowing Algorithms in Automated Essay Answer Assessment System

The role of technology during the Covid-19 pandemic has been evidently dominant, particularly in the teaching and learning process. The application of learning from home, enables E-Learning to serve as the primary learning medium. E-learning requires an automatic essay answer assessment feature to improve the objectives and efficiency of online learning activities. This study proposes the automated assessment solutions utilizing Rabin-Karp and Winnowing algorithms. The types of essay questions handled in this study include the free description and limited description types. The answer key data and answer data will be further preprocessed to be similarly calculated to the two algorithms in the process. Algorithm accuracy assessment is conducted by converting the value into a human rates version compared with the evaluation conducted by lecturers manually. This study indicated that Rabin-Karp’s Algorithm had better accuracy with the slightest difference of 27.81%, compared to Winnowing’s Algorithm accuracy with a more significant distinction of 44.32%.