A Collaborative Framework for Structure Identification over Print Documents

We describe Texture, a framework for data extraction over print documents that allows end-users to construct data extraction rules over an inferred document structure. To effectively infer this structure, we enable developers to contribute multiple heuristics that identify different structures in English print documents, crowd-workers and annotators to manually label these structures, and end-users to search and decide which heuristics to apply and how to boost their performance with the help of ground-truth data collected from crowd-workers and annotators. Texture's design supports each of these different user groups through a suite of tools. We demonstrate that even with a handful of student-developed heuristics, we can achieve reasonable precision and recall when identifying structures across different document collections.

• Publication year

  2019

• Venue

  HILDA@SIGMOD

• Publication date

  2019-07-05

• Fields of study

  Computer Science

• Identifiers
  DOI 10.1145/3328519.3329131
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Interactive Repair of Tables Extracted from PDF Documents on Mobile Devices

  2019cited by this paper
• Snorkel: Rapid Training Data Creation with Weak Supervision

  2017influential reference
• Fonduer: Knowledge Base Construction from Richly Formatted Data

  2017cited by this paper
• SEER: Auto-Generating Information Extraction Rules from User-Specified Examples

  2017cited by this paper
• Multi-Scale Multi-Task FCN for Semantic Page Segmentation and Table Detection

  2017cited by this paper
• PDFFigures 2.0: Mining figures from research papers

  2016cited by this paper
• Looking Beyond Text: Extracting Figures, Tables and Captions from Computer Science Papers

  2015cited by this paper
• TEXUS: A Task-based Approach for Table Extraction and Understanding

  2015cited by this paper
• FlashExtract: a framework for data extraction by examples

  2014cited by this paper
• Automatic web spreadsheet data extraction

  2013cited by this paper
• Understanding Tables in Context Using Standard NLP Toolkits

  2013cited by this paper
• Shreddr: pipelined paper digitization for low-resource organizations

  2012influential reference
• DeepDive: Web-scale Knowledge-base Construction using Statistical Learning and Inference

  2012cited by this paper
• Extracting general lists from web documents: a hybrid approach

  2011cited by this paper
• A Table Detection Method for Multipage PDF Documents via Visual Seperators and Tabular Structures

  2011cited by this paper
• Wrangler: interactive visual specification of data transformation scripts

  2011cited by this paper
• Structure extraction from PDF-based book documents

  2011cited by this paper
• SystemT: An Algebraic Approach to Declarative Information Extraction

  2010cited by this paper
• Information extraction by finding repeated structure

  2010cited by this paper
• Geometric Layout Analysis Techniques for Document Image Understanding: a Review

  2008cited by this paper
• Automatic extraction of titles from general documents using machine learning

  2005cited by this paper
• Extraction,layout analysis and classification of diagrams in PDF documents

  2003cited by this paper
• Table extraction using conditional random fields

  2003cited by this paper
• Document structure analysis algorithms: a literature survey

  2003cited by this paper
• A brief survey of web data extraction tools

  2002cited by this paper
• Automatic Data Extraction from Lists and Tables in Web Sources

  2001cited by this paper
• Layout and Language: Integrating Spatial and Linguistic Knowledge for Layout Understanding Tasks

  2000cited by this paper
• Extraction Patterns for Information Extraction Tasks : A Survey

  1999cited by this paper
• Logical structure analysis of book document images using contents information

  1997cited by this paper

• No citing papers are available for this paper.