Intelli-Write is an intelligent system designed to streamline academic evaluations by automatically checking handwritten assignments. It features handwriting recognition, plagiarism detection, and automated grading, making the assessment process faster, more accurate, and efficient for educators.
Objectives
To develop an intelligent system that automates the evaluation of handwritten assignments by integrating handwriting recognition, plagiarism detection, and automated grading, thereby enhancing the efficiency, accuracy, and fairness of academic assessments.
Socio-Economic Benefit
Improved Educational Equity: Automated grading ensures consistent and unbiased evaluation, promoting fairness for students from diverse backgrounds.
Time and Cost Efficiency: Reduces the workload of teachers, saving time and administrative costs in manual assessment and plagiarism checking.
Enhanced Learning Quality: Quick feedback helps students identify their mistakes and improve their understanding more effectively.
Support for Remote and Rural Education: Enables schools with limited teaching resources to manage large volumes of handwritten assignments efficiently.
Methodologies
Data Collection and Preprocessing:
First, we collected a diverse set of handwritten assignment samples. These were preprocessed using image enhancement techniques like noise reduction, grayscale conversion, and binarization to ensure better accuracy during text recognition.
Handwriting Recognition:
We used OCR technology combined with machine learning models—such as Convolutional Neural Networks (CNNs)—to accurately convert the handwritten content into digital text. The model was trained and fine-tuned using labeled handwriting datasets.
Plagiarism Detection:
Once the text was digitized, we applied Natural Language Processing techniques to detect similarities with existing content. We used methods like cosine similarity and semantic analysis to flag potentially plagiarized sections.
Automated Grading:
For grading, we designed a rubric-based system that evaluates assignments on multiple criteria like relevance, grammar, and structure. We also experimented with AI-based scoring models to ensure fairness and consistency in evaluation.
Outcome
The Intelli-Write system successfully automates the evaluation of handwritten assignments by accurately recognizing handwriting, detecting plagiarism, and providing consistent, rubric-based grading. It significantly reduces manual workload for educators, ensures fair and timely feedback for students, and promotes academic integrity. The project demonstrates the practical integration of OCR, NLP, and AI technologies in real-world educational assessment
