Beyond OCR: Intelligent Transcript Analysis

Erscheinungsjahr 2027
ISBN: 978-3-032-41789-3
Verlag: Springer

Buch, Englisch, 169 Seiten, Format (B × H): 155 mm x 235 mm

Reihe: Studies in Big Data

This book provides comprehensive technical frameworks, implementation strategies, and evidence-based guidance for deploying scalable assessment technologies. It serves as an essential resource for AI researchers, educational technology professionals, graduate students, and institutional decision-makers. The book addresses needs of educators navigating AI adoption, software engineers building educational platforms, and policymakers considering ethical AI implementation in academic settings. The book demonstrates how contemporary AI technologies, particularly large language models like Gemini 2.0 Flash, can achieve near-human accuracy in optical character recognition while maintaining the pedagogical value and personal touch that handwritten assessments provide. By bridging the gap between traditional educational practices and modern technological capabilities, this book offers educators, administrators, and technology developers a practical roadmap for implementing automated grading systems that preserve academic integrity while dramatically improving efficiency and consistency.

Zielgruppe

Professional/practitioner

Autoren/Hrsg.

Sarkar, Sujit

Panday, Mrityunjoy

Sharma, Neha

Fachgebiete

• Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz

Inhaltsverzeichnis

Introduction The Evaluation Crisis.- The State of Assessment: A Multi-Institutional Survey and Field Analysis.- Literature Review From Automated Essay Scoring to Multimodal Assessment.- Recognition Handwriting, OCR Models, and the Transcription Layer.- Question–Answer Mapping Bridging Transcribed Text to Evaluation Criteria.- Reasoning Under Evaluation Prompting Strategies, LLM Benchmarking, and RMSE Reduction.- Experimental Evaluation and Results.- Institutional Deployment Adoption, Ethics, and Governance.- Conclusion and Future Directions.

Über Autor(innen)

Sujit Sarkar is Data Scientist specializing in large language models (LLMs), machine learning pipelines, and agentic AI systems. He holds a master of engineering in data science from the Army Institute of Technology, Pune, and a bachelor of technology in mechatronics engineering from SRM Institute of Science and Technology, Chennai. His research focuses on the development of interpretable, scalable AI architectures for intelligent automation, and data-driven decision-making. Mr. Sarkar has published his work in reputed venues such as IEEE PuneCon and has a forthcoming monograph with Springer. His notable research projects include AutoGrader AI, an intelligent evaluation framework for handwritten answer sheets, Birthweight Prediction using Advanced Machine Learning Models on PMNS Data, aiming to improve early risk assessment in maternal and child health care. The motivation behind this book originates from his vision to enhance transparency and reliability in AI-assisted evaluation systems.

Mrityunjoy Panday is Principal Architect of AI and Advanced AI Competency Lead, where he spearheads research and development in agentic AI, causal modeling, and quantum-inspired algorithms. He holds an M.S. in software engineering from BITS Pilani and a B.Sc. in computer science from RKMRC Narendrapur. His work bridges deep learning, natural language processing, and unsupervised learning, with a focus on scalable, interpretable systems for enterprise AI. Mr. Panday is Inventor of two patents, including innovations in data analytics assessment and hierarchical time series forecasting. His publications in Springer volumes and award-winning papers reflect his sustained contribution to AI-driven automation and forecasting. The motivation behind this book stems from his pursuit of explainable, human-aligned intelligence systems that merge analytical rigor with operational practicality. Based in Kolkata, India, he balances his professional pursuits with a keen interest in quantum computing and scientific communication.

Dr. Neha Sharma is Distinguished Academic and Industry Professional with over twenty-six years of experience in data science, artificial intelligence, and analytics. She currently holds the position of Consultant at Tata Consultancy Services and is Founder President of Bharat Research in AI and NextGen Foundations, as well as Founder Secretary of the Society for Data Science. She earned her Ph.D. from the Indian Institute of Technology (Indian School of Mines), Dhanbad, for her pioneering work on applying data mining techniques for early detection and prevention of oral cancer. An ACM Distinguished Speaker and IEEE Senior Member, Dr. Sharma has edited and authored numerous Springer volumes, including the Data Management, Analytics, and Innovation series, and has been recognized with multiple national and international awards for her contributions to research and education. Based in Pune, India, she continues to advance interdisciplinary research bridging sustainability, health care, and intelligent systems.