Automatic extraction of relevant knowledge to domain-specific questions from Optical Character Recognition (OCR) documents is critical for developing intelligent systems, such as document search engines, sentiment analysis, and information retrieval, since hands-on knowledge extraction by a domain expert with a large volume of documents is intensive, unscalable, and time-consuming. There have been a number of studies that have automatically extracted relevant knowledge from OCR documents, such as ABBY and Sandford Natural Language Processing (NLP). Despite the progress, there are still limitations yet-to-be solved. For instance, NLP often fails to analyze a large document. In this thesis, we propose a knowledge extraction framework, which takes domain-specific questions as input and provides the most relevant sentence/paragraph to the given questions in the document. Overall, our proposed framework has two phases. First, an OCR document is reconstructed into a semi-structured document (a document with hierarchical structure of (sub)sections and paragraphs). Then, relevant sentence/paragraph for a given question is identified from the reconstructed semi structured document. Specifically, we proposed (1) a method that converts an OCR document into a semi structured document using text attributes such as font size, font height, and boldface (in Chapter 2), (2) an image-based machine learning method that extracts Table of Contents (TOC) to provide an overall structure of the document (in Chapter 3), (3) a document texture-based deep learning method (DoT-Net) that classifies types of blocks such as text, image, and table (in Chapter 4), and (4) a Question & Answer (Q&A) system that retrieves most relevant sentence/paragraph for a domain-specific question. A large number of document intelligent systems can benefit from our proposed automatic knowledge extraction system to construct a Q&A system for OCR documents. Our Q&A system has applied to extract domain specific information from business contracts at GE Power.
","abstract_html":"<p>Automatic extraction of relevant knowledge to domain-specific questions from Optical Character Recognition (OCR) documents is critical for developing intelligent systems, such as document search engines, sentiment analysis, and information retrieval, since hands-on knowledge extraction by a domain expert with a large volume of documents is intensive, unscalable, and time-consuming. There have been a number of studies that have automatically extracted relevant knowledge from OCR documents, such as ABBY and Sandford Natural Language Processing (NLP). Despite the progress, there are still limitations yet-to-be solved. For instance, NLP often fails to analyze a large document. In this thesis, we propose a knowledge extraction framework, which takes domain-specific questions as input and provides the most relevant sentence/paragraph to the given questions in the document. Overall, our proposed framework has two phases. First, an OCR document is reconstructed into a semi-structured document (a document with hierarchical structure of (sub)sections and paragraphs). Then, relevant sentence/paragraph for a given question is identified from the reconstructed semi structured document. Specifically, we proposed (1) a method that converts an OCR document into a semi structured document using text attributes such as font size, font height, and boldface (in Chapter 2), (2) an image-based machine learning method that extracts Table of Contents (TOC) to provide an overall structure of the document (in Chapter 3), (3) a document texture-based deep learning method (DoT-Net) that classifies types of blocks such as text, image, and table (in Chapter 4), and (4) a Question & Answer (Q&A) system that retrieves most relevant sentence/paragraph for a domain-specific question. A large number of document intelligent systems can benefit from our proposed automatic knowledge extraction system to construct a Q&A system for OCR documents. Our Q&A system has applied to extract domain specific information from business contracts at GE Power.</p>","abstract_has_math":false,"creators":["Kosaraju, Sai"],"institution":null,"degree_name":"Master of Science in Computer Science (MSCS)","degree_level":"Thesis","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Mingon Kang","Chih-Cheng Hung"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2019,"date_issued":"2019-11-08T08:00:00Z","date_published":"2019-11-08T08:00:00Z","updated_at":"2026-07-24T02:43:33Z","subjects":["Document Layout Analysis","Convolutional Neural Network","Dilated Layers","Question and Answers","Computer and Systems Architecture","Other Computer Engineering"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://digitalcommons.kennesaw.edu/cs_etd/28","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Mingon Kang","Chih-Cheng Hung"]},{"key":"dc:creator","label":"Author","values":["Kosaraju, Sai"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2020-11-08T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science in Computer Science (MSCS)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Document Layout Analysis","Convolutional Neural Network","Dilated Layers","Question and Answers","Computer and Systems Architecture","Other Computer Engineering"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.kennesaw.edu/cs_etd/28"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Automatic extraction of relevant knowledge to domain-specific questions from Optical Character Recognition (OCR) documents is critical for developing intelligent systems, such as document search engines, sentiment analysis, and information retrieval, since hands-on knowledge extraction by a domain expert with a large volume of documents is intensive, unscalable, and time-consuming. There have been a number of studies that have automatically extracted relevant knowledge from OCR documents, such as ABBY and Sandford Natural Language Processing (NLP). Despite the progress, there are still limitations yet-to-be solved. For instance, NLP often fails to analyze a large document. In this thesis, we propose a knowledge extraction framework, which takes domain-specific questions as input and provides the most relevant sentence/paragraph to the given questions in the document. Overall, our proposed framework has two phases. First, an OCR document is reconstructed into a semi-structured document (a document with hierarchical structure of (sub)sections and paragraphs). Then, relevant sentence/paragraph for a given question is identified from the reconstructed semi structured document. Specifically, we proposed (1) a method that converts an OCR document into a semi structured document using text attributes such as font size, font height, and boldface (in Chapter 2), (2) an image-based machine learning method that extracts Table of Contents (TOC) to provide an overall structure of the document (in Chapter 3), (3) a document texture-based deep learning method (DoT-Net) that classifies types of blocks such as text, image, and table (in Chapter 4), and (4) a Question & Answer (Q&A) system that retrieves most relevant sentence/paragraph for a domain-specific question. A large number of document intelligent systems can benefit from our proposed automatic knowledge extraction system to construct a Q&A system for OCR documents. Our Q&A system has applied to extract domain specific information from business contracts at GE Power.
"]},{"key":"dc:title","label":"Title","values":["Document Layout Analysis and Recognition Systems"]}]}],"canonical_facts":{"dc:contributor":["Mingon Kang","Chih-Cheng Hung"],"dc:creator":["Kosaraju, Sai"],"dc:date.available":["2020-11-08T08:00:00Z"],"dc:description.abstract":["Automatic extraction of relevant knowledge to domain-specific questions from Optical Character Recognition (OCR) documents is critical for developing intelligent systems, such as document search engines, sentiment analysis, and information retrieval, since hands-on knowledge extraction by a domain expert with a large volume of documents is intensive, unscalable, and time-consuming. There have been a number of studies that have automatically extracted relevant knowledge from OCR documents, such as ABBY and Sandford Natural Language Processing (NLP). Despite the progress, there are still limitations yet-to-be solved. For instance, NLP often fails to analyze a large document. In this thesis, we propose a knowledge extraction framework, which takes domain-specific questions as input and provides the most relevant sentence/paragraph to the given questions in the document. Overall, our proposed framework has two phases. First, an OCR document is reconstructed into a semi-structured document (a document with hierarchical structure of (sub)sections and paragraphs). Then, relevant sentence/paragraph for a given question is identified from the reconstructed semi structured document. Specifically, we proposed (1) a method that converts an OCR document into a semi structured document using text attributes such as font size, font height, and boldface (in Chapter 2), (2) an image-based machine learning method that extracts Table of Contents (TOC) to provide an overall structure of the document (in Chapter 3), (3) a document texture-based deep learning method (DoT-Net) that classifies types of blocks such as text, image, and table (in Chapter 4), and (4) a Question & Answer (Q&A) system that retrieves most relevant sentence/paragraph for a domain-specific question. A large number of document intelligent systems can benefit from our proposed automatic knowledge extraction system to construct a Q&A system for OCR documents. Our Q&A system has applied to extract domain specific information from business contracts at GE Power.
"],"dc:identifier":["https://digitalcommons.kennesaw.edu/cs_etd/28"],"dc:subject":["Document Layout Analysis","Convolutional Neural Network","Dilated Layers","Question and Answers","Computer and Systems Architecture","Other Computer Engineering"],"dc:title":["Document Layout Analysis and Recognition Systems"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["Master of Science in Computer Science (MSCS)"]},"updated_at":"2026-07-24T02:43:33Z"}