IDEALS Home University of Illinois at Urbana-Champaign logo The Alma Mater The Main Quad

Automatic Metadata Extraction Using Machine Learning

Show full item record

Bookmark or cite this item: http://hdl.handle.net/2142/9139

Files in this item

File Description Format
Microsoft PowerPoint automaticmetada ... inese_biodiversity_Qin.ppt (3MB) Automatic Metadata Extraction Using Machine learning Microsoft PowerPoint
Other Available Formats
PDF automaticmetada ... e_biodiversity_Qin.ppt.pdf (533KB) Automatically converted using OpenOffice.org PDF
Title: Automatic Metadata Extraction Using Machine Learning
Author(s): Wei, Qin, Heidorn, P. Bryan
Subject(s): Metadata Extraction Machine Learning
Abstract: The information in museum specimen labels is not well recognized and used. The reason is mainly because this information is not part of the information retrieval systems. It is very expensive to have human to hand input the data to the system. We are using machine learning tools to automatically extract Darwin Core (DwC) and other metadata from these labels processed through Optical Character Recognition (OCR). The DwC is a metadata profile describing the core set of access points for search and retrieval of natural history collections and observation databases. Using the HERBIS Learning System (HLS) we extract 74 independent elements from these labels. The automated text extraction tools are provided as a web service so that users can reference digital images of specimens and receive back an extended Darwin Core XML representation of the content of the label. This automated extraction task is made more difficult by the high variability of museum label formats, OCR errors and the open class nature of some elements. Here, we introduce our overall system architecture, and variability robust solutions including, the application of Hidden Markov and Naïve Bayes machine learning models, data cleaning, use of field element identifiers, and specialist learning models. The techniques developed here could be adapted to any metadata extraction situation with noisy text and weakly ordered elements.
Issue Date: 2008-10-17
Genre: Presentation / Lecture / Speech
Type: Text
Language: English
URI: http://hdl.handle.net/2142/9139
Publication Status: published or submitted for publication
Peer Reviewed: is peer reviewed
Date Available in IDEALS: 2008-10-26
 

This item appears in the following Collection(s)

Show full item record

Item Statistics

  • Total Downloads: 456
  • Downloads this Month: 6
  • Downloads Today: 0

Browse

My Account

Information

Access Key