{"id":167873,"date":"2015-05-01T00:00:00","date_gmt":"2015-05-01T00:00:00","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/msr-research-item\/tegra-table-extraction-by-global-record-alignment\/"},"modified":"2021-06-25T12:27:27","modified_gmt":"2021-06-25T19:27:27","slug":"tegra-table-extraction-by-global-record-alignment","status":"publish","type":"msr-research-item","link":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/tegra-table-extraction-by-global-record-alignment\/","title":{"rendered":"TEGRA: Table Extraction by Global Record Alignment"},"content":{"rendered":"<div class=\"asset-content\">\n<p>It is well known today that pages on the Web contain a large number of content-rich relational tables. Such tables have been systematically extracted in a number of efforts to empower important applications such as table search and schema discovery. However, a significant fraction of relational tables are \\emph{not} embedded in the standard HTML table tags, and are thus difficult to extract. In particular, a large number of relational tables are known to be in a &#8220;list&#8221; form, which contains a list of clearly separated rows that are not separated into columns.<\/p>\n<p>In this work, we address the important problem of automatically extracting multi-column relational tables from such lists. Our key intuition lies in the simple observation that in correctly-extracted tables, values in the same column are \\emph{coherent}, both at a syntactic and at a semantic level. Using a background corpus of over 100 million tables crawled from the Web, we quantify semantic coherence based on a statistical measure of value co-occurrence in the same column from the corpus. We then model table extraction as a principled optimization problem &#8212; we allocate tokens in each row sequentially to a fixed number of columns, such that the sum of coherence across all pairs of values in the same column is maximized. Borrowing ideas from $A^\\star$ search and metric distance, we develop an efficient 2-approximation algorithm. We conduct large-scale table extraction experiments using both real Web data and proprietary enterprise spreadsheet data. Our approach considerably outperforms the state-of-the-art approaches in terms of quality, achieving over 90\\% F-measure across many cases.<\/p>\n<p>Our benchmark data has been made available on GitHub <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/github.com\/Yeye-He\/TEGRA-Table-Segmentation\">https:\/\/github.com\/Yeye-He\/TEGRA-Table-Segmentation<span class=\"sr-only\"> (opens in new tab)<\/span><\/a> to facilitate future research.<\/p>\n<\/div>\n<p><!-- .asset-content --><\/p>\n","protected":false},"excerpt":{"rendered":"<p>It is well known today that pages on the Web contain a large number of content-rich relational tables. Such tables have been systematically extracted in a number of efforts to empower important applications such as table search and schema discovery. However, a significant fraction of relational tables are \\emph{not} embedded in the standard HTML table [&hellip;]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13563],"msr-publication-type":[193716],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-field-of-study":[],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-167873","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-data-platform-analytics","msr-locale-en_us"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2015-5-1","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"204374","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"file","viewUrl":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/Main.pdf","id":"204374","title":"Main.pdf","label_id":"243109","label":0},{"type":"file","viewUrl":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/Webtables_10k_testset_Readme.txt","id":"204376","title":"Webtables_10k_testset_Readme.txt","label_id":"243109","label":0},{"type":"file","viewUrl":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/Webtables_10k_testset.txt","id":"204375","title":"Webtables_10k_testset.txt","label_id":"243109","label":0}],"msr_related_uploader":"","msr_attachments":[{"id":204376,"url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/Webtables_10k_testset_Readme.txt"},{"id":204375,"url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/Webtables_10k_testset.txt"},{"id":204374,"url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/Main.pdf"}],"msr-author-ordering":[{"type":"text","value":"Xu Chu","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Yeye He","user_id":34992,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Yeye He"},{"type":"user_nicename","value":"Kaushik Chakrabarti","user_id":32503,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Kaushik Chakrabarti"},{"type":"user_nicename","value":"Kris Ganjam","user_id":32579,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Kris Ganjam"}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[957177],"msr_project":[169513],"publication":[],"video":[],"download":[],"msr_publication_type":"inproceedings","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/167873"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":3,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/167873\/revisions"}],"predecessor-version":[{"id":756937,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/167873\/revisions\/756937"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=167873"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=167873"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=167873"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=167873"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=167873"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=167873"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=167873"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=167873"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=167873"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=167873"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=167873"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=167873"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=167873"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=167873"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=167873"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}