{"id":{"repo_id":"carleton","oai_identifier":"oai:carleton.scholaris.ca:20.500.14718/45136"},"canonical_url":"https://search.dev.ndltd.org/etd/carleton/oai:carleton.scholaris.ca:20.500.14718/45136","repository":{"repo_id":"carleton","name":"Carleton University","base_url":"https://carleton.scholaris.ca/server/oai/request"},"display":{"title":"A Taxonomic Framework for the Classification of Wildfire Social Media Data in Canada","abstract":"This study introduces WildFireCan-MMD, a novel multimodal dataset containing annotated posts from X from recent Canadian wildfires across twelve key themes, addressing the need for building cost-effective real-time information systems during wildfire events. It evaluates vision-language models, deep learning, and traditional classifiers on this dataset, finding that custom deep learning models significantly outperform other classifiers, with the best achieving an f1-score of 84.48±0.69\\%. The model’s capability to identify trends in large unlabeled wildfire data highlights the value of tailored datasets for diverse disaster response needs. Additionally, this research tackles the critical trust issue in deploying deep learning models in high-stakes wildfire scenarios by emphasizing multimodal explainability. A proposed evaluation framework assesses the faithfulness of explanations from both black-box and white-box methods applied to the WildFireCan-MMD classifier. Using a standardized faithfulness metric, this framework facilitates systematic benchmarking and supports responsible, transparent deployment of automated tools in wildfire response efforts.","abstract_html":"This study introduces WildFireCan-MMD, a novel multimodal dataset containing annotated posts from X from recent Canadian wildfires across twelve key themes, addressing the need for building cost-effective real-time information systems during wildfire events. It evaluates vision-language models, deep learning, and traditional classifiers on this dataset, finding that custom deep learning models significantly outperform other classifiers, with the best achieving an f1-score of 84.48±0.69\\%. The model’s capability to identify trends in large unlabeled wildfire data highlights the value of tailored datasets for diverse disaster response needs. Additionally, this research tackles the critical trust issue in deploying deep learning models in high-stakes wildfire scenarios by emphasizing multimodal explainability. A proposed evaluation framework assesses the faithfulness of explanations from both black-box and white-box methods applied to the WildFireCan-MMD classifier. Using a standardized faithfulness metric, this framework facilitates systematic benchmarking and supports responsible, transparent deployment of automated tools in wildfire response efforts.","abstract_has_math":false,"creators":["Sherritt, Braeden Alexander"],"institution":"Carleton University","degree_name":"Master of Applied Science (M.App.Sc.)","degree_level":"Master's","degree_discipline":"Engineering, Electrical and Computer","degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026","date_published":"2026","updated_at":"2026-07-24T01:34:30Z","subjects":[],"languages":["en"],"rights":["Copyright © 2026 the author(s). Theses may be used for non-commercial research, educational, or related academic purposes only. Such uses include personal study, distribution to students, research and scholarship. Theses may only be shared by linking to the Carleton University Institutional Repository and no part may be copied without proper attribution to the author; no part may be used for commercial purposes directly or indirectly via a for-profit platform; no adaptation or derivative works are permitted without consent from the copyright owner."],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.doi","label":"DOI","values":["10.22215/etd/2026-16979"],"render_values":[{"text":"10.22215/etd/2026-16979","href":"https://doi.org/10.22215/etd/2026-16979","code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/20.500.14718/45136","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["Sherritt, Braeden Alexander"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-07-22T18:38:37Z"]},{"key":"dc:date.issued","label":"Date","values":["2026"]},{"key":"dc:publisher","label":"Institution","values":["Carleton University"]},{"key":"dc:type","label":"Dc Type","values":["thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Engineering, Electrical and Computer"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Master's"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Applied Science (M.App.Sc.)"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright © 2026 the author(s). Theses may be used for non-commercial research, educational, or related academic purposes only. Such uses include personal study, distribution to students, research and scholarship. Theses may only be shared by linking to the Carleton University Institutional Repository and no part may be copied without proper attribution to the author; no part may be used for commercial purposes directly or indirectly via a for-profit platform; no adaptation or derivative works are permitted without consent from the copyright owner."]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.doi","label":"DOI","values":["10.22215/etd/2026-16979"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/20.500.14718/45136"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This study introduces WildFireCan-MMD, a novel multimodal dataset containing annotated posts from X from recent Canadian wildfires across twelve key themes, addressing the need for building cost-effective real-time information systems during wildfire events. It evaluates vision-language models, deep learning, and traditional classifiers on this dataset, finding that custom deep learning models significantly outperform other classifiers, with the best achieving an f1-score of 84.48±0.69\\%. The model’s capability to identify trends in large unlabeled wildfire data highlights the value of tailored datasets for diverse disaster response needs. Additionally, this research tackles the critical trust issue in deploying deep learning models in high-stakes wildfire scenarios by emphasizing multimodal explainability. A proposed evaluation framework assesses the faithfulness of explanations from both black-box and white-box methods applied to the WildFireCan-MMD classifier. Using a standardized faithfulness metric, this framework facilitates systematic benchmarking and supports responsible, transparent deployment of automated tools in wildfire response efforts."]},{"key":"dc:title","label":"Title","values":["A Taxonomic Framework for the Classification of Wildfire Social Media Data in Canada"]}]}],"canonical_facts":{"dc:creator":["Sherritt, Braeden Alexander"],"dc:date.accessioned":["2026-07-22T18:38:37Z"],"dc:date.issued":["2026"],"dc:description.abstract":["This study introduces WildFireCan-MMD, a novel multimodal dataset containing annotated posts from X from recent Canadian wildfires across twelve key themes, addressing the need for building cost-effective real-time information systems during wildfire events. It evaluates vision-language models, deep learning, and traditional classifiers on this dataset, finding that custom deep learning models significantly outperform other classifiers, with the best achieving an f1-score of 84.48±0.69\\%. The model’s capability to identify trends in large unlabeled wildfire data highlights the value of tailored datasets for diverse disaster response needs. Additionally, this research tackles the critical trust issue in deploying deep learning models in high-stakes wildfire scenarios by emphasizing multimodal explainability. A proposed evaluation framework assesses the faithfulness of explanations from both black-box and white-box methods applied to the WildFireCan-MMD classifier. Using a standardized faithfulness metric, this framework facilitates systematic benchmarking and supports responsible, transparent deployment of automated tools in wildfire response efforts."],"dc:identifier.doi":["10.22215/etd/2026-16979"],"dc:identifier.uri":["https://hdl.handle.net/20.500.14718/45136"],"dc:language.iso":["en"],"dc:publisher":["Carleton University"],"dc:rights":["Copyright © 2026 the author(s). Theses may be used for non-commercial research, educational, or related academic purposes only. Such uses include personal study, distribution to students, research and scholarship. Theses may only be shared by linking to the Carleton University Institutional Repository and no part may be copied without proper attribution to the author; no part may be used for commercial purposes directly or indirectly via a for-profit platform; no adaptation or derivative works are permitted without consent from the copyright owner."],"dc:title":["A Taxonomic Framework for the Classification of Wildfire Social Media Data in Canada"],"dc:type":["thesis"],"thesis:degree_discipline":["Engineering, Electrical and Computer"],"thesis:degree_level":["Master's"],"thesis:degree_name":["Master of Applied Science (M.App.Sc.)"]},"updated_at":"2026-07-24T01:34:30Z"}