{"id":{"repo_id":"kennesaw","oai_identifier":"oai:digitalcommons.kennesaw.edu:dataphd_etd-1000"},"canonical_url":"https://search.dev.ndltd.org/etd/kennesaw/oai:digitalcommons.kennesaw.edu:dataphd_etd-1000","repository":{"repo_id":"kennesaw","name":"Kennesaw State University","base_url":"https://digitalcommons.kennesaw.edu/do/oai/"},"display":{"title":"Deep Embedding Kernel","abstract":"Kernel methods and deep learning are two major branches of machine learning that have achieved numerous successes in both analytics and artificial intelligence. While having their own unique characteristics, both branches work through mapping data to a feature space that is supposedly more favorable towards the given task. This dissertation addresses the strengths and weaknesses of each mapping method through combining them and forming a family of novel deep architectures that center around the Deep Embedding Kernel (DEK). In short, DEK is a realization of a kernel function through a newly deep architecture. The mapping in DEK is both implicit (like in kernel methods) and learnable (like in deep learning). Prior to DEK, we proposed a less advanced architecture called Deep Kernel for the tasks of classification and visualization. More recently, we integrate DEK with the novel Dual Deep Learning framework to model big unstructured data. Using DEK as a core component, we further propose two machine learning models: Deep Similarity-Enhanced K Nearest Neighbors (DSE-KNN) and Recurrent Embedding Kernel (REK). Both models have their mappings trained towards optimizing data instances' neighborhoods in the feature space. REK is specifically designed for time series data. Experimental studies throughout the dissertation show that the proposed models have competitive performance to other commonly used and state-of-the-art machine learning models in their given tasks.","abstract_html":"Kernel methods and deep learning are two major branches of machine learning that have achieved numerous successes in both analytics and artificial intelligence. While having their own unique characteristics, both branches work through mapping data to a feature space that is supposedly more favorable towards the given task. This dissertation addresses the strengths and weaknesses of each mapping method through combining them and forming a family of novel deep architectures that center around the Deep Embedding Kernel (DEK). In short, DEK is a realization of a kernel function through a newly deep architecture. The mapping in DEK is both implicit (like in kernel methods) and learnable (like in deep learning). Prior to DEK, we proposed a less advanced architecture called Deep Kernel for the tasks of classification and visualization. More recently, we integrate DEK with the novel Dual Deep Learning framework to model big unstructured data. Using DEK as a core component, we further propose two machine learning models: Deep Similarity-Enhanced K Nearest Neighbors (DSE-KNN) and Recurrent Embedding Kernel (REK). Both models have their mappings trained towards optimizing data instances' neighborhoods in the feature space. REK is specifically designed for time series data. Experimental studies throughout the dissertation show that the proposed models have competitive performance to other commonly used and state-of-the-art machine learning models in their given tasks.","abstract_has_math":false,"creators":["Le, Linh"],"institution":null,"degree_name":"Doctor of Philosophy in Analytic and Data Science","degree_level":"Dissertation","degree_discipline":"Statistics and Analytical Sciences","degree_department":null,"school":null,"contributors":["Dr. Ying Xie","Dr. Jennifer Priestley","Dr. Erik Westlund","Dr. Meng Han","Dr. Michael McBurnett"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2019,"date_issued":"2019-04-18T07:00:00Z","date_published":"2019-04-18T07:00:00Z","updated_at":"2026-07-24T02:43:33Z","subjects":["Deep Embedding Kernel","Deep Kernel","Deep Learning","Machine Learning","Kernel Methods","Supervised Learning","Computer Sciences","Statistics and Probability"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://digitalcommons.kennesaw.edu/dataphd_etd/1","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Dr. Ying Xie","Dr. Jennifer Priestley","Dr. Erik Westlund","Dr. Meng Han","Dr. Michael McBurnett"]},{"key":"dc:creator","label":"Author","values":["Le, Linh"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2019-04-30T07:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Statistics and Analytical Sciences"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy in Analytic and Data Science"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Deep Embedding Kernel","Deep Kernel","Deep Learning","Machine Learning","Kernel Methods","Supervised Learning","Computer Sciences","Statistics and Probability"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.kennesaw.edu/dataphd_etd/1"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Kernel methods and deep learning are two major branches of machine learning that have achieved numerous successes in both analytics and artificial intelligence. While having their own unique characteristics, both branches work through mapping data to a feature space that is supposedly more favorable towards the given task. This dissertation addresses the strengths and weaknesses of each mapping method through combining them and forming a family of novel deep architectures that center around the Deep Embedding Kernel (DEK). In short, DEK is a realization of a kernel function through a newly deep architecture. The mapping in DEK is both implicit (like in kernel methods) and learnable (like in deep learning). Prior to DEK, we proposed a less advanced architecture called Deep Kernel for the tasks of classification and visualization. More recently, we integrate DEK with the novel Dual Deep Learning framework to model big unstructured data. Using DEK as a core component, we further propose two machine learning models: Deep Similarity-Enhanced K Nearest Neighbors (DSE-KNN) and Recurrent Embedding Kernel (REK). Both models have their mappings trained towards optimizing data instances' neighborhoods in the feature space. REK is specifically designed for time series data. Experimental studies throughout the dissertation show that the proposed models have competitive performance to other commonly used and state-of-the-art machine learning models in their given tasks."]},{"key":"dc:title","label":"Title","values":["Deep Embedding Kernel"]}]}],"canonical_facts":{"dc:contributor":["Dr. Ying Xie","Dr. Jennifer Priestley","Dr. Erik Westlund","Dr. Meng Han","Dr. Michael McBurnett"],"dc:creator":["Le, Linh"],"dc:date.available":["2019-04-30T07:00:00Z"],"dc:description.abstract":["Kernel methods and deep learning are two major branches of machine learning that have achieved numerous successes in both analytics and artificial intelligence. While having their own unique characteristics, both branches work through mapping data to a feature space that is supposedly more favorable towards the given task. This dissertation addresses the strengths and weaknesses of each mapping method through combining them and forming a family of novel deep architectures that center around the Deep Embedding Kernel (DEK). In short, DEK is a realization of a kernel function through a newly deep architecture. The mapping in DEK is both implicit (like in kernel methods) and learnable (like in deep learning). Prior to DEK, we proposed a less advanced architecture called Deep Kernel for the tasks of classification and visualization. More recently, we integrate DEK with the novel Dual Deep Learning framework to model big unstructured data. Using DEK as a core component, we further propose two machine learning models: Deep Similarity-Enhanced K Nearest Neighbors (DSE-KNN) and Recurrent Embedding Kernel (REK). Both models have their mappings trained towards optimizing data instances' neighborhoods in the feature space. REK is specifically designed for time series data. Experimental studies throughout the dissertation show that the proposed models have competitive performance to other commonly used and state-of-the-art machine learning models in their given tasks."],"dc:identifier":["https://digitalcommons.kennesaw.edu/dataphd_etd/1"],"dc:subject":["Deep Embedding Kernel","Deep Kernel","Deep Learning","Machine Learning","Kernel Methods","Supervised Learning","Computer Sciences","Statistics and Probability"],"dc:title":["Deep Embedding Kernel"],"thesis:degree_discipline":["Statistics and Analytical Sciences"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Doctor of Philosophy in Analytic and Data Science"]},"updated_at":"2026-07-24T02:43:33Z"}