One-Word Answer Correction using Deep Learning Models and OCR
K. P. K Devan1, Sruthi Prabakaran P2, Tamizhazhagan S3, Vaishnavi S4

1K. P. K. Devan, Associate Professor, CSE Department, Easwari Engineering College, Chennai, Tamil Nadu.
2Sruthi Prabakaran P, CSE Department, Easwari Engineering College, Chennai, Tamil Nadu.
3Tamizhazhagan S, CSE Department, Easwari Engineering College, Chennai, Tamil Nadu.
4Vaishnavi S, CSE Department, Easwari Engineering College, Chennai, Tamil Nadu.

Manuscript received on May 25, 2020. | Revised Manuscript received on June 29, 2020. | Manuscript published on July 30, 2020. | PP: 679-682 | Volume-9 Issue-2, July 2020. | Retrieval Number: B3849079220/2020©BEIESP | DOI: 10.35940/ijrte.B3849.079220
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Examinations/Assessments are a way to assess the understanding of a student on a particular subject. Even today many educational organizations prefer to conduct exams by offline mode (pen and paper). And evaluating them is a time-consuming process. There is no effectual model to evaluate Offline descriptive answers automatically. The traditional method involves staff assessing the content manually. In place of this process, a new approach using image captioning by using deep learning algorithms can be implemented. Handwritten Text Recognition (HTR) can be used to evaluate descriptive answers. One-word Answers captured as images are pre-processed to extract the text features using deep learning models and pytesseract. This paper presents a comparison between the CNN-RNN hybrid model and Optical Character Recognition (OCR) to predict a score for one-word answers. 
Keywords: Convolutional Neural Network (CNN), Handwritten Text Recognition (HTR), Optical Character Recognition (OCR), Recurrent Neural Network (RNN).