Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Collaborative Research and Development Grants

Simon Fraser University

$14,500.00

Extracting and Tagging Tabular Data on the Web

FY 2009-10

Summary
Many Web pages contain tabular data, such as a Web page on a shopping Web site using a table to compare multiple products. Relatively, tabular data has better structure and quality. Therefore, extracting and labeling tabular data on the Web plays an important role in Web search. Many social search software products on Internet and enterprise intranets need a powerful module in extracting and tagging tabular data. However, extracting tabular data is far from straightforward since table tags in HTML Web pages are frequently used for layout formatting and some other purposes. Although there exist some studies on extracting tabular data on the Web, three important issues remain unsolved. First, there is no large benchmark data sets to objectively evaluate the performance of tabular data extraction methods. Second, while most of the existing methods use simple heuristics and employ supervised learning approaches, classification models learned from a specific training data set may likely carry bias on the learning instances, and thus may not be robust to extracting tabular data from other Web pages that are not represented well in the training set. Last, there is no existing work on automatically tagging tabular data at both the table level and the column level. In this project, we will develop effective techniques to extract and tag tabular data on the Web. First, we will provide several labeled benchmark tabular data sets crawled from many Web sites. The data sets will consist of Web pages containing both data tables and layout tables. Moreover, the data sets will cover many domains as well as Web sites with various popularity of tabular data. Second, we will develop a tabular data extraction method and a prototype system to automatically capture tabular data components on Web pages and transform tabular data into spreadsheets. Different from the existing methods, our approach is not a supervised learning approach. Instead, we want to explore the language model of a table as well as the language models of the columns. Third, we will develop a tagging method to generate tags for an extracted table as well as the columns in the table, which can be used in advanced Web search and data integration.
Program
Collaborative Research and Development Grants
Location
British Columbia, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
380433-2008
Name nom
Pei, Jian
Department département
Computing Science
Organization id
5
Institution établissement
Simon Fraser University
Province en
British Columbia
Province fr
Colombie-Britannique
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2009
Competition year année de concours
2008
Award amount
14500
Program id
CRDPJ
Program nane en
Collaborative Research and Development Grants
Program name fr
Subventions de recherche et développement coopérative
Group en
Research partnerships programs
Group fr
Programmes de partenariat
Committee code
96
Committee name en
RPP Internal Decision Cttee
Committee name fr
Cté de décision interne - PPR
Area of application code
801
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Computer software
Area of application fr
Logiciel
Research subject code
2711
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Database management
Research subject fr
Gestion de bases de données
Partie
1
Nb partie
1
Application title
Extracting and Tagging Tabular Data on the Web
Keyword
NA
Application summary
Many Web pages contain tabular data, such as a Web page on a shopping Web site using a table to compare multiple products. Relatively, tabular data has better structure and quality. Therefore, extracting and labeling tabular data on the Web plays an important role in Web search. Many social search software products on Internet and enterprise intranets need a powerful module in extracting and tagging tabular data. However, extracting tabular data is far from straightforward since table tags in HTML Web pages are frequently used for layout formatting and some other purposes. Although there exist some studies on extracting tabular data on the Web, three important issues remain unsolved. First, there is no large benchmark data sets to objectively evaluate the performance of tabular data extraction methods. Second, while most of the existing methods use simple heuristics and employ supervised learning approaches, classification models learned from a specific training data set may likely carry bias on the learning instances, and thus may not be robust to extracting tabular data from other Web pages that are not represented well in the training set. Last, there is no existing work on automatically tagging tabular data at both the table level and the column level. In this project, we will develop effective techniques to extract and tag tabular data on the Web. First, we will provide several labeled benchmark tabular data sets crawled from many Web sites. The data sets will consist of Web pages containing both data tables and layout tables. Moreover, the data sets will cover many domains as well as Web sites with various popularity of tabular data. Second, we will develop a tabular data extraction method and a prototype system to automatically capture tabular data components on Web pages and transform tabular data into spreadsheets. Different from the existing methods, our approach is not a supervised learning approach. Instead, we want to explore the language model of a table as well as the language models of the columns. Third, we will develop a tagging method to generate tags for an extracted table as well as the columns in the table, which can be used in advanced Web search and data integration.
Simon Fraser University - $14,500 Grant (2009-10) | Canada Spends