Extraction of Features with Unstructured Representation From HTML Documents

Ray R. Hashemi; Charles W. Ford; Tim Vamprooyen; John R. Talburt

Extraction of Features with Unstructured Representation From HTML Documents

Ray R. Hashemi, Charles W. Ford, Tim Vamprooyen, John R. Talburt

Computer Science

Research output: Contribution to book or proceeding › Chapter

Abstract

The goal of this research effort is to mine web pages for unstructured features (Names, Titles, and their associations.) Unstructured features are not easily identifiable because they lack the presence of obvious patterns in their ASCII representations. In addition, the crucial process of establishing associations among the extracted features adds another level of complexity to the mining process. The results obtained from the application of our methodology to a test bed of 20 URLs with 500 total pages revealed: (a) the measures of recovery and accuracy of the extracted Name and Title features are quite satisfactory, and (b) the proposed methodology is highly effective.

Original language	American English
Title of host publication	Proceedings of the International Conference on WWW/Internet
State	Published - Nov 2002

Disciplines

Engineering
Computer Sciences

Keywords

Extraction
Features
HTML documents
Mining names from text
Structured features
Text mining
Unstructured features
Unstructured representation
Web mining

Cite this

@inbook{36ece8ed547e42c3b987534f57c25820,

title = "Extraction of Features with Unstructured Representation From HTML Documents",

abstract = " The goal of this research effort is to mine web pages for unstructured features (Names, Titles, and their associations.) Unstructured features are not easily identifiable because they lack the presence of obvious patterns in their ASCII representations. In addition, the crucial process of establishing associations among the extracted features adds another level of complexity to the mining process. The results obtained from the application of our methodology to a test bed of 20 URLs with 500 total pages revealed: (a) the measures of recovery and accuracy of the extracted Name and Title features are quite satisfactory, and (b) the proposed methodology is highly effective.",

keywords = "Extraction, Features, HTML documents, Mining names from text, Structured features, Text mining, Unstructured features, Unstructured representation, Web mining",

author = "Hashemi, {Ray R.} and Ford, {Charles W.} and Tim Vamprooyen and Talburt, {John R.}",

note = "The goal of this research effort is to mine web pages for unstructured features (Names, Titles, and their associations.) Unstructured features are not easily identifiable because they lack the presence of obvious patterns in their ASCII",

year = "2002",

month = nov,

language = "American English",

booktitle = "Proceedings of the International Conference on WWW/Internet",

}

TY - CHAP

T1 - Extraction of Features with Unstructured Representation From HTML Documents

AU - Hashemi, Ray R.

AU - Ford, Charles W.

AU - Vamprooyen, Tim

AU - Talburt, John R.

N1 - The goal of this research effort is to mine web pages for unstructured features (Names, Titles, and their associations.) Unstructured features are not easily identifiable because they lack the presence of obvious patterns in their ASCII

PY - 2002/11

Y1 - 2002/11

N2 - The goal of this research effort is to mine web pages for unstructured features (Names, Titles, and their associations.) Unstructured features are not easily identifiable because they lack the presence of obvious patterns in their ASCII representations. In addition, the crucial process of establishing associations among the extracted features adds another level of complexity to the mining process. The results obtained from the application of our methodology to a test bed of 20 URLs with 500 total pages revealed: (a) the measures of recovery and accuracy of the extracted Name and Title features are quite satisfactory, and (b) the proposed methodology is highly effective.

AB - The goal of this research effort is to mine web pages for unstructured features (Names, Titles, and their associations.) Unstructured features are not easily identifiable because they lack the presence of obvious patterns in their ASCII representations. In addition, the crucial process of establishing associations among the extracted features adds another level of complexity to the mining process. The results obtained from the application of our methodology to a test bed of 20 URLs with 500 total pages revealed: (a) the measures of recovery and accuracy of the extracted Name and Title features are quite satisfactory, and (b) the proposed methodology is highly effective.

KW - Extraction

KW - Features

KW - HTML documents

KW - Mining names from text

KW - Structured features

KW - Text mining

KW - Unstructured features

KW - Unstructured representation

KW - Web mining

UR - http://www.iadisportal.org/digital-library/extraction-of-features-with-unstructured-representation-from-html-documents

M3 - Chapter

BT - Proceedings of the International Conference on WWW/Internet

ER -

Extraction of Features with Unstructured Representation From HTML Documents

Abstract

Disciplines

Keywords

Other files and links

Fingerprint

Cite this