Exam Eval: An Automated Model for Correcting Exam Papers Using OCR and Text Summarization
Keywords:
NLP (Natural Language Processing), NLTK (Natural language Tool Kit), CV (Computer Vision), OCR (Optical Character Recognition), Automated ModelAbstract
In the realm of education, grading and correcting exam papers can be a time-consuming and often daunting task for teachers. To streamline this process and provide more efficient feedback to students, this project aims to develop a model that leverages advanced techniques in natural language processing (NLP) and computer vision. The primary objective is to correct exam papers by summarizing the handwritten answers provided by students, incorporating Optical Character Recognition (OCR) through OpenCV for handling handwritten scripts. This model addresses the challenge of grading handwritten scripts by automatically converting the handwritten text into a machine-readable format and generating concise summaries that align with the corresponding teacher's answer summary. By harnessing OCR capabilities, the model extracts text from handwritten images and preprocesses it to improve readability and accuracy. Subsequently, text summarization techniques are applied to condense the extracted text into coherent and informative summaries. The integration of OCR with text summarization provides an efficient and automated solution for correcting handwritten exam papers, ultimately saving time and effort for teachers. This model aims to enhance the grading process, provide consistent and unbiased evaluations, and offer valuable insights into students' performance by matching their answer summaries with those prepared by the teacher.
References
Liu M, Li Y, Xu W, Liu L. Automated essay feedback generation and its impact on revision. IEEE Trans Learning Technologies. 2016 Sep 22; 10(4): 502–13.
Kumar M, Jindal MK, Sharma RK. Review on OCR for handwritten Indian scripts character recognition. In International Conference on Digital Image Processing and Information Technology 2011 Sep 23 (pp. 268-276). Berlin, Heidelberg: Springer Berlin Heidelberg.
Balyan R, Crossley SA, Brown III W, Karter AJ, McNamara DS, Liu JY, Lyles CR, Schillinger D. Using natural language processing and machine learning to classify health literacy from secure messages: The ECLIPPSE study. PloS one. 2019 Feb 22; 14(2): e0212488.
Memon J, Sami M, Khan RA, Uddin M. Handwritten optical character recognition (OCR): A comprehensive systematic literature review (SLR). IEEE Access. 2020 Jul 28; 8: 142642–68.
Armbruster BB, Anderson TH, Ostertag J. Does text structure/summarization instruction facilitate learning from expository text? Read Res Q. 1987 Jul; 1: 331–346.
Mostafa A, Mohamed O, Ashraf A, Elbehery A, Jamal S, Salah A, Ghoneim AS. An End-to-End OCR Framework for Robust Arabic-Handwriting Recognition using a Novel Transformers-based Model and an Innovative 270 Million-Words Multi-Font Corpus of Classical Arabic with Diacritics. arXiv preprint arXiv:2208.11484. 2022 Aug 20.
Alomari A, Idris N, Sabri AQ, Alsmadi I. Deep reinforcement and transfer learning for abstractive text summarization: A review. Comput Speech Lang. 2022 Jan 1; 71: 101276.
Vithlani P, Kumbharana CK. Comparative study of character recognition tools. Int J Comput Appl. 2015 Jan 1; 118(9).
Névéol A, Dalianis H, Velupillai S, Savova G, Zweigenbaum P. Clinical natural language processing in languages other than English: opportunities and challenges. J Biomed Semantics. 2018 Dec; 9(1): 1–3.
Gui T, Wang X, Zhang Q, Liu Q, Zou Y, Zhou X, Zheng R, Zhang C, Wu Q, Ye J, Pang Z. Textflint: Unified multilingual robustness evaluation toolkit for natural language processing. arXiv preprint arXiv:2103.11441. 2021 Mar 21.
