{"id":{"repo_id":"bu","oai_identifier":"oai:open.bu.edu:2144/39604"},"canonical_url":"https://search.ndltd.org/etd/bu/oai:open.bu.edu:2144/39604","repository":{"repo_id":"bu","name":"Boston University","base_url":"https://open.bu.edu/oai/request"},"display":{"title":"Empirical studies of factors affecting opinion dynamics","abstract":"The advent of new online services has an enormous potential to impact the opinion of users. Two main drivers of this impact are crowdsourced evaluations and ratings, and algorithmically-chosen recommendations. However, understanding the relationship between these systems and their impacts is very challenging due the complex nature of recommender systems and due to the heterogeneous nature of crowdsourced reviews. In this thesis, we explore how these two drivers affect opinion dynamics with respect to two potential impacts: reliability of information and polarization of user opinion. First, we analyze the reliability of online ratings. By performing an empirical analysis of a large corpus of online ratings, we point out how different influences such as shifts in population or platform characteristics are correlated with changes in the perception of an item over time. Second, we investigate polarization in the context of recommender systems. We define three metrics - intensity, simplification, and divergence - to capture essential traits of user opinions and explore how they vary in a closed-loop with recommender systems. Finally, we examine reliability in recommendations via an empirical exploration on YouTube. We quantify changes in the nature of the recommended content, and we show how YouTube recommendations lead users - especially privacy-seeking users - away from reliable information. Taken together, these studies shed light on important factors that affect how user opinion is shaped by online systems.","abstract_html":"The advent of new online services has an enormous potential to impact the opinion of users. Two main drivers of this impact are crowdsourced evaluations and ratings, and algorithmically-chosen recommendations. However, understanding the relationship between these systems and their impacts is very challenging due the complex nature of recommender systems and due to the heterogeneous nature of crowdsourced reviews. In this thesis, we explore how these two drivers affect opinion dynamics with respect to two potential impacts: reliability of information and polarization of user opinion. First, we analyze the reliability of online ratings. By performing an empirical analysis of a large corpus of online ratings, we point out how different influences such as shifts in population or platform characteristics are correlated with changes in the perception of an item over time. Second, we investigate polarization in the context of recommender systems. We define three metrics - intensity, simplification, and divergence - to capture essential traits of user opinions and explore how they vary in a closed-loop with recommender systems. Finally, we examine reliability in recommendations via an empirical exploration on YouTube. We quantify changes in the nature of the recommended content, and we show how YouTube recommendations lead users - especially privacy-seeking users - away from reliable information. Taken together, these studies shed light on important factors that affect how user opinion is shaped by online systems.","abstract_has_math":false,"creators":["Pinheiro Spinelli, Larissa"],"institution":"Boston University","degree_name":"Doctor of Philosophy","degree_level":"doctoral","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":[],"advisors":["Crovella, Mark"],"committee_chairs":[],"committee_members":[],"year":2019,"date_issued":"2019","date_published":"2019","updated_at":"2026-11-19T02:41:25Z","subjects":["Computer science","Crowdsourced reviews","Data science","Opinion dynamics","Polarization","Recommender systems"],"languages":["en_US"],"rights":["Attribution 4.0 International"],"rights_urls":["http://creativecommons.org/licenses/by/4.0/"],"identifier_entries":[],"person_identifier_entries":[{"key":"dc:identifier.orcid","label":"ORCID","values":["0000-0001-7886-523X"],"render_values":[{"text":"0000-0001-7886-523X","href":"https://orcid.org/0000-0001-7886-523X","code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/2144/39604","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"https://hdl.handle.net/2144/39604","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://open.bu.edu/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Aopen.bu.edu%3A2144%2F39604","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Crovella, Mark"]},{"key":"dc:creator","label":"Author","values":["Pinheiro Spinelli, Larissa"]},{"key":"dc:identifier.orcid","label":"ORCID","values":["0000-0001-7886-523X"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2020-03-02T16:57:35Z"]},{"key":"dc:date.embargo","label":"Dc Date Embargo","values":["2020-08-24T00:00:00Z"]},{"key":"dc:date.issued","label":"Date","values":["2019"]},{"key":"dc:date.updated","label":"Dc Date Updated","values":["2020-02-24T20:02:45Z"]},{"key":"dc:type","label":"Dc Type","values":["Thesis/Dissertation"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["doctoral"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Boston University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Computer science","Crowdsourced reviews","Data science","Opinion dynamics","Polarization","Recommender systems"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en_US"]},{"key":"dc:rights","label":"Dc Rights","values":["Attribution 4.0 International"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://creativecommons.org/licenses/by/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/2144/39604"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["The advent of new online services has an enormous potential to impact the opinion of users. Two main drivers of this impact are crowdsourced evaluations and ratings, and algorithmically-chosen recommendations. However, understanding the relationship between these systems and their impacts is very challenging due the complex nature of recommender systems and due to the heterogeneous nature of crowdsourced reviews. In this thesis, we explore how these two drivers affect opinion dynamics with respect to two potential impacts: reliability of information and polarization of user opinion. First, we analyze the reliability of online ratings. By performing an empirical analysis of a large corpus of online ratings, we point out how different influences such as shifts in population or platform characteristics are correlated with changes in the perception of an item over time. Second, we investigate polarization in the context of recommender systems. We define three metrics - intensity, simplification, and divergence - to capture essential traits of user opinions and explore how they vary in a closed-loop with recommender systems. Finally, we examine reliability in recommendations via an empirical exploration on YouTube. We quantify changes in the nature of the recommended content, and we show how YouTube recommendations lead users - especially privacy-seeking users - away from reliable information. Taken together, these studies shed light on important factors that affect how user opinion is shaped by online systems."]},{"key":"dc:title","label":"Title","values":["Empirical studies of factors affecting opinion dynamics"]}]}],"canonical_facts":{"dc:contributor.advisor":["Crovella, Mark"],"dc:creator":["Pinheiro Spinelli, Larissa"],"dc:date.accessioned":["2020-03-02T16:57:35Z"],"dc:date.embargo":["2020-08-24T00:00:00Z"],"dc:date.issued":["2019"],"dc:date.updated":["2020-02-24T20:02:45Z"],"dc:description.abstract":["The advent of new online services has an enormous potential to impact the opinion of users. Two main drivers of this impact are crowdsourced evaluations and ratings, and algorithmically-chosen recommendations. However, understanding the relationship between these systems and their impacts is very challenging due the complex nature of recommender systems and due to the heterogeneous nature of crowdsourced reviews. In this thesis, we explore how these two drivers affect opinion dynamics with respect to two potential impacts: reliability of information and polarization of user opinion. First, we analyze the reliability of online ratings. By performing an empirical analysis of a large corpus of online ratings, we point out how different influences such as shifts in population or platform characteristics are correlated with changes in the perception of an item over time. Second, we investigate polarization in the context of recommender systems. We define three metrics - intensity, simplification, and divergence - to capture essential traits of user opinions and explore how they vary in a closed-loop with recommender systems. Finally, we examine reliability in recommendations via an empirical exploration on YouTube. We quantify changes in the nature of the recommended content, and we show how YouTube recommendations lead users - especially privacy-seeking users - away from reliable information. Taken together, these studies shed light on important factors that affect how user opinion is shaped by online systems."],"dc:identifier.orcid":["0000-0001-7886-523X"],"dc:identifier.uri":["https://hdl.handle.net/2144/39604"],"dc:language.iso":["en_US"],"dc:rights":["Attribution 4.0 International"],"dc:rights.uri":["http://creativecommons.org/licenses/by/4.0/"],"dc:subject":["Computer science","Crowdsourced reviews","Data science","Opinion dynamics","Polarization","Recommender systems"],"dc:title":["Empirical studies of factors affecting opinion dynamics"],"dc:type":["Thesis/Dissertation"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_level":["doctoral"],"thesis:degree_name":["Doctor of Philosophy"],"thesis:institution_name":["Boston University"]},"updated_at":"2026-11-19T02:41:25Z"}