{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/90799"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/90799","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Multi-theme sentiment analysis with sentiment shifting","abstract":"Business reviews contain rich sentiment on multiple themes, disclosing more interesting information than the overall polarities of documents. When it comes to fine-grained sentiment analysis, given any segment of text, we are not only interested in overall polarity of such segment, but also the sentiment words play major effects. However, sentiment analysis at the word level poses significant challenges due to the complexity of reviews, the inconsistency of sentiment in different themes, and the sentiment shifting resulting from linguistic patterns---contextual valence shifters. To simultaneously resolve the multi-theme and sentiment shifting dilemma, a unified explainable sentiment analysis model, MTSA, is proposed in this paper, which enables both classification of sentiment polarity and discovery of quantified sentiment-shifting patterns. MTSA formulates multi-theme sentiment by learning embeddings (i.e., vector representations) for both themes and words, and derives the shifter effect learning algorithm by modeling the shifted sentiment in a logistic regression model. Extensive experiments have been conducted on Yelp business reviews and IMDB movie reviews. The improvement of sentiment polarity classification demonstrates the effectiveness of MTSA at rectifying word feature representations of reviews, and the human evaluation shows its successful discovery of multi-theme sentiment words and automatic effect quantification of contextual valence shifters.","abstract_html":"Business reviews contain rich sentiment on multiple themes, disclosing more interesting information than the overall polarities of documents. When it comes to fine-grained sentiment analysis, given any segment of text, we are not only interested in overall polarity of such segment, but also the sentiment words play major effects. However, sentiment analysis at the word level poses significant challenges due to the complexity of reviews, the inconsistency of sentiment in different themes, and the sentiment shifting resulting from linguistic patterns---contextual valence shifters. To simultaneously resolve the multi-theme and sentiment shifting dilemma, a unified explainable sentiment analysis model, MTSA, is proposed in this paper, which enables both classification of sentiment polarity and discovery of quantified sentiment-shifting patterns. MTSA formulates multi-theme sentiment by learning embeddings (i.e., vector representations) for both themes and words, and derives the shifter effect learning algorithm by modeling the shifted sentiment in a logistic regression model. Extensive experiments have been conducted on Yelp business reviews and IMDB movie reviews. The improvement of sentiment polarity classification demonstrates the effectiveness of MTSA at rectifying word feature representations of reviews, and the human evaluation shows its successful discovery of multi-theme sentiment words and automatic effect quantification of contextual valence shifters.","abstract_has_math":false,"creators":["Yu, Hongkun"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"M.S.","degree_level":"Thesis","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Han, Jiawei"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2016,"date_issued":"2016-07-07T20:27:48Z","date_published":"2016-07-07T20:27:48Z","updated_at":"2026-07-22T22:26:34Z","subjects":["Sentiment Analysis","Multi-theme","Natural language processing (NLP)","Data Mining"],"languages":["en"],"rights":["Copyright 2016 Hongkun Yu"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/90799","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Han, Jiawei"]},{"key":"dc:creator","label":"Author","values":["Yu, Hongkun"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2016-07-07T20:27:48Z","2018-07-08T09:15:16Z","2016-04-21","2016-05"]},{"key":"dc:type","label":"Dc Type","values":["text"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Sentiment Analysis","Multi-theme","Natural language processing (NLP)","Data Mining"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2016 Hongkun Yu"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/90799"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Business reviews contain rich sentiment on multiple themes, disclosing more interesting information than the overall polarities of documents. When it comes to fine-grained sentiment analysis, given any segment of text, we are not only interested in overall polarity of such segment, but also the sentiment words play major effects. However, sentiment analysis at the word level poses significant challenges due to the complexity of reviews, the inconsistency of sentiment in different themes, and the sentiment shifting resulting from linguistic patterns---contextual valence shifters. To simultaneously resolve the multi-theme and sentiment shifting dilemma, a unified explainable sentiment analysis model, MTSA, is proposed in this paper, which enables both classification of sentiment polarity and discovery of quantified sentiment-shifting patterns. MTSA formulates multi-theme sentiment by learning embeddings (i.e., vector representations) for both themes and words, and derives the shifter effect learning algorithm by modeling the shifted sentiment in a logistic regression model. Extensive experiments have been conducted on Yelp business reviews and IMDB movie reviews. The improvement of sentiment polarity classification demonstrates the effectiveness of MTSA at rectifying word feature representations of reviews, and the human evaluation shows its successful discovery of multi-theme sentiment words and automatic effect quantification of contextual valence shifters.","Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2018-05-01","The student, Hongkun Yu, accepted the attached license on 2016-04-20 at 13:24.","The student, Hongkun Yu, submitted this Thesis for approval on 2016-04-20 at 13:27.","This Thesis was approved for publication on 2016-04-21 at 08:52.","DSpace SAF Submission Ingestion Package generated from Vireo submission #9349 on 2016-07-07 at 13:50:14","Made available in DSpace on 2016-07-07T20:27:48Z (GMT). No. of bitstreams: 2 YU-THESIS-2016.pdf: 1377626 bytes, checksum: fe33b7692b15580351e718e199b44e59 (MD5) LICENSE.txt: 4207 bytes, checksum: 6ecb6be099535d4749440bbf2fcd7721 (MD5) Previous issue date: 2016-04-21","Embargo set by: Seth Robbins for item 93151 Lift date: 2018-07-07T20:28:14Z Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD system","Embargo set by: Seth Robbins for item 93151 Lift date: 2018-07-07T20:35:34Z Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD system","U of I Only Restriction Lifted for Item 93151 on 2018-07-08T09:15:16Z."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Multi-theme sentiment analysis with sentiment shifting"]}]}],"canonical_facts":{"dc:contributor":["Han, Jiawei"],"dc:creator":["Yu, Hongkun"],"dc:date":["2016-07-07T20:27:48Z","2018-07-08T09:15:16Z","2016-04-21","2016-05"],"dc:description":["Business reviews contain rich sentiment on multiple themes, disclosing more interesting information than the overall polarities of documents. When it comes to fine-grained sentiment analysis, given any segment of text, we are not only interested in overall polarity of such segment, but also the sentiment words play major effects. However, sentiment analysis at the word level poses significant challenges due to the complexity of reviews, the inconsistency of sentiment in different themes, and the sentiment shifting resulting from linguistic patterns---contextual valence shifters. To simultaneously resolve the multi-theme and sentiment shifting dilemma, a unified explainable sentiment analysis model, MTSA, is proposed in this paper, which enables both classification of sentiment polarity and discovery of quantified sentiment-shifting patterns. MTSA formulates multi-theme sentiment by learning embeddings (i.e., vector representations) for both themes and words, and derives the shifter effect learning algorithm by modeling the shifted sentiment in a logistic regression model. Extensive experiments have been conducted on Yelp business reviews and IMDB movie reviews. The improvement of sentiment polarity classification demonstrates the effectiveness of MTSA at rectifying word feature representations of reviews, and the human evaluation shows its successful discovery of multi-theme sentiment words and automatic effect quantification of contextual valence shifters.","Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2018-05-01","The student, Hongkun Yu, accepted the attached license on 2016-04-20 at 13:24.","The student, Hongkun Yu, submitted this Thesis for approval on 2016-04-20 at 13:27.","This Thesis was approved for publication on 2016-04-21 at 08:52.","DSpace SAF Submission Ingestion Package generated from Vireo submission #9349 on 2016-07-07 at 13:50:14","Made available in DSpace on 2016-07-07T20:27:48Z (GMT). No. of bitstreams: 2 YU-THESIS-2016.pdf: 1377626 bytes, checksum: fe33b7692b15580351e718e199b44e59 (MD5) LICENSE.txt: 4207 bytes, checksum: 6ecb6be099535d4749440bbf2fcd7721 (MD5) Previous issue date: 2016-04-21","Embargo set by: Seth Robbins for item 93151 Lift date: 2018-07-07T20:28:14Z Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD system","Embargo set by: Seth Robbins for item 93151 Lift date: 2018-07-07T20:35:34Z Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD system","U of I Only Restriction Lifted for Item 93151 on 2018-07-08T09:15:16Z."],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/90799"],"dc:language":["en"],"dc:rights":["Copyright 2016 Hongkun Yu"],"dc:subject":["Sentiment Analysis","Multi-theme","Natural language processing (NLP)","Data Mining"],"dc:title":["Multi-theme sentiment analysis with sentiment shifting"],"dc:type":["text"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["M.S."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:26:34Z"}