Files in this item



application/pdfTestingNerOnMetadata201002041250.pdf (232kB)
Main reportPDF


Title:Testing the Extracting Metadata for Preservation project's Named-Entity Recognizer on metadata
Author(s):Jackson, Larry S.
Subject(s):named entity recognition
metadata generation
Abstract:Named entity recognition software shown to perform well in free-text performed reasonably well in textual descriptive metadata, but experienced considerably more errors in dealing with titles. Capitalization conventions and fragmentary sentences used in titles may require deliberate compensatory additions to the training data of the machine learning software behind the recognizer.
Issue Date:2010-02-04
Citation Info:Jackson, L. S. (2010). Testing the Extracting Metadata for Preservation project's Named‐Entity Recognizer on metadata. Technical Report #ISRN UIUCLIS‐‐2010/1+EAP. Graduate School of Library and Information Science, University of Illinois at Urbana-Champaign.
Genre:Technical Report
Other Identifier(s):ISRN UIUCLIS‐‐2010/1+EAP
Publication Status:unpublished
Peer Reviewed:not peer reviewed
Sponsor:Library of Congress / NDIIPP-2 A6075
Date Available in IDEALS:2010-04-13

This item appears in the following Collection(s)

Item Statistics