Automatic essay exam scoring system: a systematic literature review.

Area of Science:

• Educational Technology
• Natural Language Processing
• Computer-Assisted Assessment

Background:

• The COVID-19 pandemic necessitated a shift to distance learning globally, impacting the education sector significantly.
• Traditional face-to-face learning was replaced by remote educational practices, creating challenges for assessment.
• The need for efficient and scalable assessment methods became apparent, particularly for essay-based evaluations.

Purpose of the Study:

• To conduct a systematic literature review on automated essay answer assessment.
• To synthesize existing research on essay answer scoring systems and their methodologies.
• To provide insights for future research in automated essay exam scoring systems.

Main Methods:

• Systematic literature review of existing studies on automated essay scoring.
• Synthesis and analysis of various methods developed for essay answer scoring.
• Identification of common themes and variations in dataset formats used in prior research.

Main Results:

• A comprehensive overview of diverse methods employed in automated essay scoring.
• Identification of trends and challenges in developing and implementing automated scoring systems.
• Understanding the landscape of datasets utilized for training and evaluating these systems.

Conclusions:

• Automated essay scoring systems offer a viable solution for remote assessment needs.
• Further research is needed to refine methods and explore optimal dataset configurations.
• This review contributes to advancing the field of automated essay examination scoring.