{"id":{"repo_id":"calpoly","oai_identifier":"oai:digitalcommons.calpoly.edu:theses-5001"},"canonical_url":"https://search.ndltd.org/etd/calpoly/oai:digitalcommons.calpoly.edu:theses-5001","repository":{"repo_id":"calpoly","name":"Cal Poly","base_url":"https://digitalcommons.calpoly.edu/do/oai/"},"display":{"title":"Developing a Humpback Whale Vocalization Detector Using Machine Learning Models","abstract":"Humpback whale songs are notoriously complex. Identification of humpback whale song units requires bioacousticians to tediously listen, analyze, and annotate collected sound data. Even sparse data requires listening to the entirety of the collected acoustic data. In this study, three hours of audio containing over one-thousand humpback whale song units was collected in Monterey Bay, California. Prior studies have seen success using convolutional neural networks by performing image classification on hundreds of hours worth of spectrograms. Our study uses traditional machine learning models, as they are less computationally demanding, and require less data. We use time splitting and Mel-frequency cepstrum to transform complex, continuous sound data into a set of trainable, numerical coefficients. We capture the complexity of humpback songs by creating observations based on the evolution of a song, called context windows. We apply two non-parametric machine learning models, k-nearest neighbors, and random forests, to our created observations. Results are tested in two manners, per-context-window and per-vocalization. Context windows with more time steps avoid overfitting, produces better per-window metrics, and is more successful at identifying individual calls in a high presence dataset. Model selection, parameter tuning, and context window adjusting yield observation-based F1 scores up to 0.824, and identify over 98% of vocalizations.","abstract_html":"Humpback whale songs are notoriously complex. Identification of humpback whale song units requires bioacousticians to tediously listen, analyze, and annotate collected sound data. Even sparse data requires listening to the entirety of the collected acoustic data. In this study, three hours of audio containing over one-thousand humpback whale song units was collected in Monterey Bay, California. Prior studies have seen success using convolutional neural networks by performing image classification on hundreds of hours worth of spectrograms. Our study uses traditional machine learning models, as they are less computationally demanding, and require less data. We use time splitting and Mel-frequency cepstrum to transform complex, continuous sound data into a set of trainable, numerical coefficients. We capture the complexity of humpback songs by creating observations based on the evolution of a song, called context windows. We apply two non-parametric machine learning models, k-nearest neighbors, and random forests, to our created observations. Results are tested in two manners, per-context-window and per-vocalization. Context windows with more time steps avoid overfitting, produces better per-window metrics, and is more successful at identifying individual calls in a high presence dataset. Model selection, parameter tuning, and context window adjusting yield observation-based F1 scores up to 0.824, and identify over 98% of vocalizations.","abstract_has_math":false,"creators":["Kantorowski, Lucas"],"institution":null,"degree_name":"MS in Statistics","degree_level":null,"degree_discipline":"Statistics","degree_department":null,"school":null,"contributors":["Kelly Bodwin","Statistics","College of Science and Mathematics"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T07:12:51Z","subjects":["Humpback whales","Monterey Bay","Mel frequency cepstrum coefficients (MFCC)","random forest","k-nearest neighbors","machine learning","Applied Statistics","Biostatistics","Data Science","Statistical Methodology","Statistics and Probability"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://digitalcommons.calpoly.edu/theses/3311","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://digitalcommons.calpoly.edu/theses/3311","record_label":"Repository record","record_kind":"record","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://digitalcommons.calpoly.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adigitalcommons.calpoly.edu%3Atheses-5001","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Kelly Bodwin","Statistics","College of Science and Mathematics"]},{"key":"dc:creator","label":"Author","values":["Kantorowski, Lucas"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2026-06-11T07:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2026-06-01T07:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Statistics"]},{"key":"thesis:degree_name","label":"Degree Name","values":["MS in Statistics"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Humpback whales","Monterey Bay","Mel frequency cepstrum coefficients (MFCC)","random forest","k-nearest neighbors","machine learning","Applied Statistics","Biostatistics","Data Science","Statistical Methodology","Statistics and Probability"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.calpoly.edu/theses/3311"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Humpback whale songs are notoriously complex. Identification of humpback whale song units requires bioacousticians to tediously listen, analyze, and annotate collected sound data. Even sparse data requires listening to the entirety of the collected acoustic data. In this study, three hours of audio containing over one-thousand humpback whale song units was collected in Monterey Bay, California. Prior studies have seen success using convolutional neural networks by performing image classification on hundreds of hours worth of spectrograms. Our study uses traditional machine learning models, as they are less computationally demanding, and require less data. We use time splitting and Mel-frequency cepstrum to transform complex, continuous sound data into a set of trainable, numerical coefficients. We capture the complexity of humpback songs by creating observations based on the evolution of a song, called context windows. We apply two non-parametric machine learning models, k-nearest neighbors, and random forests, to our created observations. Results are tested in two manners, per-context-window and per-vocalization. Context windows with more time steps avoid overfitting, produces better per-window metrics, and is more successful at identifying individual calls in a high presence dataset. Model selection, parameter tuning, and context window adjusting yield observation-based F1 scores up to 0.824, and identify over 98% of vocalizations."]},{"key":"dc:title","label":"Title","values":["Developing a Humpback Whale Vocalization Detector Using Machine Learning Models"]}]}],"canonical_facts":{"dc:contributor":["Kelly Bodwin","Statistics","College of Science and Mathematics"],"dc:creator":["Kantorowski, Lucas"],"dc:date.available":["2026-06-11T07:00:00Z"],"dc:date.created":["2026-06-01T07:00:00Z"],"dc:description.abstract":["Humpback whale songs are notoriously complex. Identification of humpback whale song units requires bioacousticians to tediously listen, analyze, and annotate collected sound data. Even sparse data requires listening to the entirety of the collected acoustic data. In this study, three hours of audio containing over one-thousand humpback whale song units was collected in Monterey Bay, California. Prior studies have seen success using convolutional neural networks by performing image classification on hundreds of hours worth of spectrograms. Our study uses traditional machine learning models, as they are less computationally demanding, and require less data. We use time splitting and Mel-frequency cepstrum to transform complex, continuous sound data into a set of trainable, numerical coefficients. We capture the complexity of humpback songs by creating observations based on the evolution of a song, called context windows. We apply two non-parametric machine learning models, k-nearest neighbors, and random forests, to our created observations. Results are tested in two manners, per-context-window and per-vocalization. Context windows with more time steps avoid overfitting, produces better per-window metrics, and is more successful at identifying individual calls in a high presence dataset. Model selection, parameter tuning, and context window adjusting yield observation-based F1 scores up to 0.824, and identify over 98% of vocalizations."],"dc:identifier":["https://digitalcommons.calpoly.edu/theses/3311"],"dc:subject":["Humpback whales","Monterey Bay","Mel frequency cepstrum coefficients (MFCC)","random forest","k-nearest neighbors","machine learning","Applied Statistics","Biostatistics","Data Science","Statistical Methodology","Statistics and Probability"],"dc:title":["Developing a Humpback Whale Vocalization Detector Using Machine Learning Models"],"thesis:degree_discipline":["Statistics"],"thesis:degree_name":["MS in Statistics"]},"updated_at":"2026-11-19T07:12:51Z"}