Article

Meta-algorithmic systems for document classification

Authors:

Steven J. Simske,

David W. Wright,

Margaret SturgillAuthors Info & Claims

DocEng '06: Proceedings of the 2006 ACM symposium on Document engineering

Pages 98 - 106

https://doi.org/10.1145/1166160.1166190

Published: 10 October 2006 Publication History

Get Access

Abstract

To address cost and regulatory concerns, many businesses are converting paper-based elements of their workflows into fully electronic flows that use the content of the documents. Scanning the document contents into workflows, however, is a manual, error-prone, and costly process especially when the data extraction process requires high accuracy. These manual costs are a primary barrier to widespread adoption of distributed capture solutions for business critical workflows such as insurance claims, medical records, or loan applications. Software solutions using artificial intelligence and natural language processing techniques are emerging to address these needs, but each have their individual strengths and weaknesses, and none have demonstrated a high level of accuracy across the many unstructured document types included in these business critical workflows. This paper describes how to overcome many of these limitations by intelligently combining multiple approaches for document classification using meta-algorithmic design patterns. These patterns explore the error space in multiple engines, and provide improved and "emergent" results in comparison to voting schemes and to the output of any of the individual engines. This paper considers the results of the individual engines along with traditional combinatorial techniques such as voting, before describing prototype results for a variety of novel metaalgorithmic patterns that reduce individual document error rates by up to 13% and reduce system error rates by up to 38%.

References

[1]

Simske, S and Lin, X. "Creating digital libraries: content generation and re-mastering", HP Labs Technical Report 2003-259, posted at: http://www.hpl.americas.hp.net/techreports/2003/HPL-2003-259.pdf.

Google Scholar

[2]

Mohomine text classifier, subsequently purchased by Kofax (ww.kofax.com) and integrated into Indicius, http://www.kofax.com/products/indicius/index.asp.

Google Scholar

[3]

"Divmod Reverend" at http://divmod.org/trac/wiki/DivmodReverend.

Google Scholar

[4]

The University of California, Irvine Knowledge Discovery in Databases Archive, 20 Newsgroups data set, http://kdd.ics.uci.edu/databases/20newsgroups/20newsgroups.html.

Google Scholar

[5]

Lin, X, Yacoub, S, Burns, J and Simske, S. "Performance analysis of pattern classifier combination by plurality voting". Pattern Recognition Letters 24:1959--1969 (2003).

Digital Library

Google Scholar

[6]

Sebastiani, F. "Machine learning in automated text categorization". ACM Comput. Surv. 34(1):1--47 (2002).

Digital Library

Google Scholar

[7]

Ruta, D, Gabrys, B. "An overview of classifier fusion methods." Computing and Information Systems 7(1):1--10 (2000).

Google Scholar

[8]

Freund, Y, Schapire, RE. "A decision-theoretic generalization of on-line learning and an application to boosting," Proc. Second European Conf Learning Theory, LNCS (1995).

Digital Library

Google Scholar

[9]

Baoli L, Qin L, Shiwen Y. "An adaptive k-nearest neighbor text categorization strategy." ACM Trans Asian Language Info Proc (TALIP), 3:215--226 (2004).

Digital Library

Google Scholar

[10]

Zhang, D, Chen, X, Lee, WS. "Text classification with kernels on the multinomial manifold." Proc. 28th Ann. Intl. ACM SIGIR Conf on Res Dev in IR, 266--273 (2005).

Digital Library

Google Scholar

Cited By

View all

Markoulidakis IMarkoulidakis G(2024)Probabilistic Confusion Matrix: A Novel Method for Machine Learning Algorithm Generalized Performance AnalysisTechnologies10.3390/technologies1207011312:7(113)Online publication date: 13-Jul-2024
https://doi.org/10.3390/technologies12070113
Sturgill MSimske SSablatnig RHassan T(2016)Mass Serialization Method for Document Encryption Policy EnforcementProceedings of the 2016 ACM Symposium on Document Engineering10.1145/2960811.2967166(193-196)Online publication date: 13-Sep-2016
https://dl.acm.org/doi/10.1145/2960811.2967166
Simske S(2015)The rationale for ensemble and meta-algorithmic architectures in signal and information processingAPSIPA Transactions on Signal and Information Processing10.1017/ATSIP.2015.104Online publication date: 2-Sep-2015
https://doi.org/10.1017/ATSIP.2015.10
Show More Cited By

Index Terms

Meta-algorithmic systems for document classification

Recommendations

A semantic weighting method for document classification based on Markov logic networks
RACS '14: Proceedings of the 2014 Conference on Research in Adaptive and Convergent Systems

This paper proposes a semantic weighting method to classify textural documents. Human lives in the world where web documents have a great potential and the amount of valuable information has been consistently growing over the year. There is a problem ...
A feature mining based approach for the classification of text documents into disjoint classes

This paper proposes a new approach for classifying text documents into two disjoint classes. The new approach is based on extracting patterns, in the form of two logical expressions, which are defined on various features (indexing terms) of the ...
Document Classification and Page Stream Segmentation for Digital Mailroom Applications
ICDAR '13: Proceedings of the 2013 12th International Conference on Document Analysis and Recognition

In this paper we present a method for the segmentation of continuous page streams into multipage documents and the simultaneous classification of the resulting documents. We first present an approach to combine the multiple pages of a document into a ...

Reviews

Reviewer: Amos O Olagunju

Industry is integrating more paperwork activities into electronic information storage and retrieval systems, in response to regulatory and cost-saving mandates. For decades, automatic indexing and document classification efforts have focused on individual techniques [1,2]. The incorporation of multiple document classification algorithms and incongruent indexing schemes into one engine is a very tricky issue. With fewer correlated classifiers, document classification error rates ought to reduce. However, creating a document classification system to simultaneously reduce the classification error rate and diminish errors in the indexing scheme is not easy. Simske et al. advocate an automated, transcending algorithmic pattern design for minimizing manual human intervention in document processing, to drastically curtail the operational costs of document workflows. The document classifiers investigated include an open source Bayesian spam filter, the Hewlett-Packard engine that normalizes term frequency by the distribution frequency, a commercial neural net engine, and clever combinations of these algorithmic systems. Multifarious blends of statistical techniques were applied to various fusions of document classifier engines that significantly shrunk classification error rates. The authors presented a variety of transformed algorithmic patterns that greatly improved the initial and final classifications of sets of documents in the presence of preliminary group errors in the indexing designs. Although the investigation of interactions between highly correlated document classifiers with obscure or unknown probability spaces lingers as a problem, this paper offers insights for the design and performance evaluation of future multiengine document retrieval systems. Online Computing Reviews Service

Access critical reviews of Computing literature here

Become a reviewer for Computing Reviews.

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

DocEng '06: Proceedings of the 2006 ACM symposium on Document engineering

October 2006

232 pages

ISBN:1595935150

DOI:10.1145/1166160

General Chair:
Dick Bulterman
CWI, Netherlands
,
Program Chair:
David F. Brailsford
University of Nottingham, UK

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 10 October 2006

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Article

Conference

DocEng06

Sponsor:

DocEng06: ACM Symposium on Document Engineering

October 10 - 13, 2006

Amsterdam, The Netherlands

Acceptance Rates

Overall Acceptance Rate 194 of 564 submissions, 34%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

9
Total Citations
View Citations
492
Total Downloads

Downloads (Last 12 months)5
Downloads (Last 6 weeks)1

Reflects downloads up to 01 Jan 2025

Other Metrics

View Author Metrics

Citations

Cited By

View all

Markoulidakis IMarkoulidakis G(2024)Probabilistic Confusion Matrix: A Novel Method for Machine Learning Algorithm Generalized Performance AnalysisTechnologies10.3390/technologies1207011312:7(113)Online publication date: 13-Jul-2024
https://doi.org/10.3390/technologies12070113
Sturgill MSimske SSablatnig RHassan T(2016)Mass Serialization Method for Document Encryption Policy EnforcementProceedings of the 2016 ACM Symposium on Document Engineering10.1145/2960811.2967166(193-196)Online publication date: 13-Sep-2016
https://dl.acm.org/doi/10.1145/2960811.2967166
Simske S(2015)The rationale for ensemble and meta-algorithmic architectures in signal and information processingAPSIPA Transactions on Signal and Information Processing10.1017/ATSIP.2015.104Online publication date: 2-Sep-2015
https://doi.org/10.1017/ATSIP.2015.10
Simske S(2013)Second‐Order Meta‐Algorithmics and their ApplicationsMeta‐Algorithmics10.1002/9781118626719.ch8(272-309)Online publication date: 27-May-2013
https://doi.org/10.1002/9781118626719.ch8
Simske S(2013)Introduction to Meta‐AlgorithmicsMeta‐Algorithmics10.1002/9781118626719.ch6(175-240)Online publication date: 27-May-2013
https://doi.org/10.1002/9781118626719.ch6
Simske S(2013)The FutureMeta‐Algorithmics10.1002/9781118626719.ch11(360-368)Online publication date: 27-May-2013
https://doi.org/10.1002/9781118626719.ch11
Gaubatz MSimske S(2009)Printer-scanner identification via analysis of structured security deterrents2009 First IEEE International Workshop on Information Forensics and Security (WIFS)10.1109/WIFS.2009.5386463(151-155)Online publication date: Dec-2009
https://doi.org/10.1109/WIFS.2009.5386463
Simske S(2009)Dynamic biometrics: The case for a real-time solution to the problem of access control, privacy and security2009 First IEEE International Conference on Biometrics, Identity and Security (BIdS)10.1109/BIDS.2009.5507535(1-10)Online publication date: Sep-2009
https://doi.org/10.1109/BIDS.2009.5507535
Sturgill MSimske Sde Graça Pimentel MBulterman DGomes Soares L(2008)An optical character recognition approach to qualifying thresholding algorithmsProceedings of the eighth ACM symposium on Document engineering10.1145/1410140.1410197(263-266)Online publication date: 16-Sep-2008
https://dl.acm.org/doi/10.1145/1410140.1410197

View Options

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Cited By

Index Terms

Recommendations

A semantic weighting method for document classification based on Markov logic networks

A feature mining based approach for the classification of text documents into disjoint classes

Document Classification and Page Stream Segmentation for Digital Mailroom Applications

Reviews

Access critical reviews of Computing literature here