{"id":{"repo_id":"whiterose","oai_identifier":"oai:etheses.whiterose.ac.uk:1570"},"canonical_url":"https://search.dev.ndltd.org/etd/whiterose/oai:etheses.whiterose.ac.uk:1570","repository":{"repo_id":"whiterose","name":"White Rose University Consortium","base_url":"https://etheses.whiterose.ac.uk/cgi/oai2"},"display":{"title":"Opinion Analysis through Constraint Optimisation","abstract":"Opinion lexicon plays a vital role in sentiment classifi�cation. A previous study shows that a compositional model can be e�ective in sentiment classifi�cation. But such a model has been only applied using hand-crafted composition rules. The need for hand-crafted rules arise when dealing with conflicting polarity values within the same phrase. In this thesis, we show that an alternative is to employ a weighted polarity lexicon. There are several key advantages of a weighted polarity lexicon. Firstly, compositionality rules simply become linear sums without requiring conflict resolution rules. Secondly, a weighted polarity lexicon can be automatically learnt from review data using constraint optimisation. Thirdly, instead of providing just a binary positive or negative output, our model can be used to provide a graded overall sentiment. Our experiments show that our model provides state-of-the-art opinion classi�cation.","abstract_html":"Opinion lexicon plays a vital role in sentiment classifi�cation. A previous study shows that a compositional model can be e�ective in sentiment classifi�cation. But such a model has been only applied using hand-crafted composition rules. The need for hand-crafted rules arise when dealing with conflicting polarity values within the same phrase. In this thesis, we show that an alternative is to employ a weighted polarity lexicon. There are several key advantages of a weighted polarity lexicon. Firstly, compositionality rules simply become linear sums without requiring conflict resolution rules. Secondly, a weighted polarity lexicon can be automatically learnt from review data using constraint optimisation. Thirdly, instead of providing just a binary positive or negative output, our model can be used to provide a graded overall sentiment. Our experiments show that our model provides state-of-the-art opinion classi�cation.","abstract_has_math":false,"creators":["Pandey, Suraj Jung"],"institution":"University of York","degree_name":"M.Sc by research","degree_level":"masters","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Manandhar, Suresh"],"committee_chairs":[],"committee_members":[],"year":2011,"date_issued":"2011-01","date_published":"2011-01","updated_at":"2026-07-24T06:04:57Z","subjects":[],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Manandhar, Suresh"]},{"key":"dc:creator","label":"Author","values":["Pandey, Suraj Jung"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2011-01-04"]},{"key":"dc:date.issued","label":"Date","values":["2011-01"]},{"key":"dc:publisher.commercial","label":"Dc Publisher Commercial","values":["University of York"]},{"key":"dc:publisher.department","label":"Dc Publisher Department","values":["Computer Science (York)"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["University of York"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://etheses.whiterose.ac.uk/id/eprint/1570/"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["masters"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["M.Sc by research"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://etheses.whiterose.ac.uk/id/eprint/1570/1/thesis_suraj.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Opinion lexicon plays a vital role in sentiment classifi�cation. A previous study shows that a compositional model can be e�ective in sentiment classifi�cation. But such a model has been only applied using hand-crafted composition rules. The need for hand-crafted rules arise when dealing with conflicting polarity values within the same phrase. In this thesis, we show that an alternative is to employ a weighted polarity lexicon. There are several key advantages of a weighted polarity lexicon. Firstly, compositionality rules simply become linear sums without requiring conflict resolution rules. Secondly, a weighted polarity lexicon can be automatically learnt from review data using constraint optimisation. Thirdly, instead of providing just a binary positive or negative output, our model can be used to provide a graded overall sentiment. Our experiments show that our model provides state-of-the-art opinion classi�cation."]},{"key":"dc:format","label":"Dc Format","values":["text"]},{"key":"dc:title","label":"Title","values":["Opinion Analysis through Constraint Optimisation"]}]}],"canonical_facts":{"dc:contributor.advisor":["Manandhar, Suresh"],"dc:creator":["Pandey, Suraj Jung"],"dc:date":["2011-01-04"],"dc:date.issued":["2011-01"],"dc:description.abstract":["Opinion lexicon plays a vital role in sentiment classifi�cation. A previous study shows that a compositional model can be e�ective in sentiment classifi�cation. But such a model has been only applied using hand-crafted composition rules. The need for hand-crafted rules arise when dealing with conflicting polarity values within the same phrase. In this thesis, we show that an alternative is to employ a weighted polarity lexicon. There are several key advantages of a weighted polarity lexicon. Firstly, compositionality rules simply become linear sums without requiring conflict resolution rules. Secondly, a weighted polarity lexicon can be automatically learnt from review data using constraint optimisation. Thirdly, instead of providing just a binary positive or negative output, our model can be used to provide a graded overall sentiment. Our experiments show that our model provides state-of-the-art opinion classi�cation."],"dc:format":["text"],"dc:identifier.uri":["https://etheses.whiterose.ac.uk/id/eprint/1570/1/thesis_suraj.pdf"],"dc:publisher.commercial":["University of York"],"dc:publisher.department":["Computer Science (York)"],"dc:publisher.institution":["University of York"],"dc:relation.isreferencedby":["https://etheses.whiterose.ac.uk/id/eprint/1570/"],"dc:title":["Opinion Analysis through Constraint Optimisation"],"dc:type":["Thesis"],"dc:type.qualificationlevel":["masters"],"dc:type.qualificationname":["M.Sc by research"]},"updated_at":"2026-07-24T06:04:57Z"}