{"id":162888,"date":"2012-07-01T00:00:00","date_gmt":"2012-07-01T00:00:00","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/msr-research-item\/polarity-inducing-latent-semantic-analysis\/"},"modified":"2018-10-16T21:11:08","modified_gmt":"2018-10-17T04:11:08","slug":"polarity-inducing-latent-semantic-analysis","status":"publish","type":"msr-research-item","link":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/polarity-inducing-latent-semantic-analysis\/","title":{"rendered":"Polarity Inducing Latent Semantic Analysis"},"content":{"rendered":"<div class=\"asset-content\">\n<p>Existing vector space models typically map synonyms and antonyms to similar word vectors, and thus fail to represent antonymy. We introduce a new vector space representation where antonyms lie on opposite sides of a sphere: in the word vector space, synonyms have cosine similarities close to one, while antonyms are close to minus one.<\/p>\n<p>We derive this representation with the aid of a thesaurus and latent semantic analysis (LSA). Each entry in the thesaurus \u2013 a word sense along with its synonyms and antonyms \u2013 is treated as a \u201cdocument,\u201d and the resulting document collection is subjected to LSA. The key contribution of this work is to show how to assign signs to the entries in the co-occurrence matrix on which LSA operates, so as to induce a subspace with the desired property.<\/p>\n<p>We evaluate this procedure with the Graduate Record Examination questions of (Mohammed et al., 2008) and find that the method improves on the results of that study. Further improvements result from refining the subspace representation with discriminative training, and augmenting the training data with general newspaper text. Altogether, we improve on the best previous results by 11 points absolute in F measure.<\/p>\n<\/div>\n<p><!-- .asset-content --><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Existing vector space models typically map synonyms and antonyms to similar word vectors, and thus fail to represent antonymy. We introduce a new vector space representation where antonyms lie on opposite sides of a sphere: in the word vector space, synonyms have cosine similarities close to one, while antonyms are close to minus one. We [&hellip;]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13556,13554],"msr-publication-type":[193716],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-162888","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-artificial-intelligence","msr-research-area-human-computer-interaction","msr-locale-en_us"],"msr_publishername":"ACL\/SIGPARSE","msr_edition":"Experimental Methods in Natural Language Processing 2012","msr_affiliation":"","msr_published_date":"2012-07-01","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"Experimental Methods in Natural Language Processing 2012","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":"205933","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"file","title":"pilsa2.pdf","viewUrl":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/pilsa2.pdf","id":205933,"label_id":0}],"msr_related_uploader":"","msr_attachments":[{"id":205933,"url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/pilsa2.pdf"}],"msr-author-ordering":[{"type":"user_nicename","value":"scottyih","user_id":33556,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=scottyih"},{"type":"user_nicename","value":"jplatt","user_id":32416,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=jplatt"},{"type":"user_nicename","value":"gzweig","user_id":31938,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=gzweig"}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[],"msr_project":[171065,170884],"publication":[],"video":[],"download":[],"msr_publication_type":"inproceedings","related_content":{"projects":[{"ID":171065,"post_title":"Recurrent Neural Networks for Language Processing","post_name":"recurrent-neural-networks-for-language-processing","post_type":"msr-project","post_date":"2012-11-23 11:45:31","post_modified":"2019-08-19 14:55:37","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/recurrent-neural-networks-for-language-processing\/","post_excerpt":"This project focuses on advancing the state-of-the-art in language processing with recurrent neural networks. We are currently applying these to language modeling, machine translation, speech recognition, language understanding and meaning representation. A special interest in is adding side-channels of information as input, to model phenomena which are not easily handled in other frameworks. A toolkit for doing RNN language modeling with side-information is in the associated download. Sample word vectors for use with this toolkit&hellip;","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/171065"}]}},{"ID":170884,"post_title":"MSR Sentence Completion Challenge","post_name":"msr-sentence-completion-challenge","post_type":"msr-project","post_date":"2011-12-08 11:11:26","post_modified":"2017-05-31 16:13:49","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/msr-sentence-completion-challenge\/","post_excerpt":"The MSR sentence completion challenge is intended to stimulate research in the area of semantic modeling. The challenge set consists of fill-in-the-blank questions similar to those found on the widely used Scholastic Aptitude Test. The sentence completion questions we focus on test the students ability to select words which are meaningful and coherent in the the context of a complete sentence. In general, this determination cannot be made on the basis of grammatical correctness alone.&hellip;","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/170884"}]}}]},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/162888","targetHints":{"allow":["GET"]}}],"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":2,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/162888\/revisions"}],"predecessor-version":[{"id":533631,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/162888\/revisions\/533631"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=162888"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=162888"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=162888"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=162888"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=162888"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=162888"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=162888"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=162888"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=162888"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=162888"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=162888"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=162888"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=162888"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=162888"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=162888"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=162888"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}