{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/116137"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/116137","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Multi-network association: Algorithms and applications","abstract":"Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-11-15 without embargo terms","abstract_html":"Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-11-15 without embargo terms","abstract_has_math":false,"creators":["Du, Boxin"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Tong, Hanghang","Banerjee, Arindam","Zhai, Chengxiang","Tang, Jian"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2022,"date_issued":"2022-08","date_published":"2022-08","updated_at":"2026-07-22T22:24:55Z","subjects":["multi-network association","graph mining","machine learning","algorithm","application"],"languages":["en","eng"],"rights":["Copyright 2022 Boxin Du"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/2142/116137","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Tong, Hanghang","Banerjee, Arindam","Zhai, Chengxiang","Tang, Jian"]},{"key":"dc:creator","label":"Author","values":["Du, Boxin"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2022-08","2022-06-01"]},{"key":"dc:type","label":"Dc Type","values":["text","Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph.D."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["multi-network association","graph mining","machine learning","algorithm","application"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en","eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2022 Boxin Du"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://hdl.handle.net/2142/116137"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-11-15 without embargo terms","The student, Boxin Du, accepted the attached license on 2022-04-26 at 19:09.","The student, Boxin Du, submitted this Dissertation for approval on 2022-04-26 at 19:26.","This Dissertation was approved for publication on 2022-06-01 at 15:20.","DSpace SAF Submission Ingestion Package generated from Vireo submission #17971 on 2022-11-15 at 17:37:23","Networks extracted from multiple sources and platforms or from multiple instances of identical domains form the multi-network, such as large social networks collected from Facebook and Instagram, medium-scaled networks of chemical compound and proteins extracted from chemical/protein interaction, etc. Multi-network association refers to the node associations or proximities in a multi-network model, which goes beyond the boundary of node associations of a single simple network. Multi-network association offers a fundamental primitive for mining multi-networks, in the sense that it reveals the unique, collective relations among node sets, which can not be captured by mining individual networks separately. Although network mining has become a ubiquitous tool of knowledge discovery for researchers and practitioners in diverse application domains, research in the multi-network association is still relatively limited, owing to the following three major challenges. First ( Problem formulation), how do we explicitly formulate the multi-network association inference problem in various multi-network scenarios, such as multiple plain/attributed networks, multi-layered networks, hypergraphs, etc.? How do we implicitly preserve multi-network association in an embedding model which is targeted at multi-network mining tasks? Second (Computational complexity), how can we develop efficient algorithms for mitigating the high complexity of the problems defined on multi-networks, in terms of both time and space complexity? Third (Application), how will the multi-network association empower or enable novel applications on multi-network data? To what extent can the multi-network association based methods boost the classic multi-network mining tasks? In this Ph.D. thesis, an in-depth study of the multi-network association is formally discussed and analyzed to jointly tackle the aforementioned challenges. Specially, the research works from this thesis are organized based on the taxonomy of the network associations (i.e. pairwise vs. high-order association), and the taxonomy of the core algorithmic techniques (i.e. numerical vs. neural methods). First (pairwise association with numerical techniques), we develop a family of fast solvers (FASTEN) for the Sylvester equation, which lays the foundation of numerous multi-network mining tasks. We further introduce a novel application, namely interactive subgraph matching, empowered by the Sylvester equation. Second (pairwise association with neural techniques), we extend the boundary of the numerical techniques for pairwise association and design Sylvester Multi-Graph Neural Network model. As a generalization of traditional Sylvester equation, its flexible architecture could incorporate numerical features, and it is able to be adapted to various downstream tasks. We then show that how such technique could be successfully applied on the application of social recommendation, to achieve up to 30% improvement over baseline methods. Third (high-order association with numerical techniques), we design a family of algorithms (i.e., SyTE) for multi-way association problem on both plain and attributed networks. It shows applicability in a variety of multi-network mining tasks, such as multi-network alignment. Forth (high-order association with neural techniques), we develop an unsupervised multi-resolution multi-network embedding model to simultaneously embed network elements of different resolutions and different networks into the same embedding space. We also present a hypergraph representation learning model via pre-training strategy, with a real-world case study on the inconsistent variation family detection problem for Amazon selection and catalog system."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Multi-network association: Algorithms and applications"]}]}],"canonical_facts":{"dc:contributor":["Tong, Hanghang","Banerjee, Arindam","Zhai, Chengxiang","Tang, Jian"],"dc:creator":["Du, Boxin"],"dc:date":["2022-08","2022-06-01"],"dc:description":["Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-11-15 without embargo terms","The student, Boxin Du, accepted the attached license on 2022-04-26 at 19:09.","The student, Boxin Du, submitted this Dissertation for approval on 2022-04-26 at 19:26.","This Dissertation was approved for publication on 2022-06-01 at 15:20.","DSpace SAF Submission Ingestion Package generated from Vireo submission #17971 on 2022-11-15 at 17:37:23","Networks extracted from multiple sources and platforms or from multiple instances of identical domains form the multi-network, such as large social networks collected from Facebook and Instagram, medium-scaled networks of chemical compound and proteins extracted from chemical/protein interaction, etc. Multi-network association refers to the node associations or proximities in a multi-network model, which goes beyond the boundary of node associations of a single simple network. Multi-network association offers a fundamental primitive for mining multi-networks, in the sense that it reveals the unique, collective relations among node sets, which can not be captured by mining individual networks separately. Although network mining has become a ubiquitous tool of knowledge discovery for researchers and practitioners in diverse application domains, research in the multi-network association is still relatively limited, owing to the following three major challenges. First ( Problem formulation), how do we explicitly formulate the multi-network association inference problem in various multi-network scenarios, such as multiple plain/attributed networks, multi-layered networks, hypergraphs, etc.? How do we implicitly preserve multi-network association in an embedding model which is targeted at multi-network mining tasks? Second (Computational complexity), how can we develop efficient algorithms for mitigating the high complexity of the problems defined on multi-networks, in terms of both time and space complexity? Third (Application), how will the multi-network association empower or enable novel applications on multi-network data? To what extent can the multi-network association based methods boost the classic multi-network mining tasks? In this Ph.D. thesis, an in-depth study of the multi-network association is formally discussed and analyzed to jointly tackle the aforementioned challenges. Specially, the research works from this thesis are organized based on the taxonomy of the network associations (i.e. pairwise vs. high-order association), and the taxonomy of the core algorithmic techniques (i.e. numerical vs. neural methods). First (pairwise association with numerical techniques), we develop a family of fast solvers (FASTEN) for the Sylvester equation, which lays the foundation of numerous multi-network mining tasks. We further introduce a novel application, namely interactive subgraph matching, empowered by the Sylvester equation. Second (pairwise association with neural techniques), we extend the boundary of the numerical techniques for pairwise association and design Sylvester Multi-Graph Neural Network model. As a generalization of traditional Sylvester equation, its flexible architecture could incorporate numerical features, and it is able to be adapted to various downstream tasks. We then show that how such technique could be successfully applied on the application of social recommendation, to achieve up to 30% improvement over baseline methods. Third (high-order association with numerical techniques), we design a family of algorithms (i.e., SyTE) for multi-way association problem on both plain and attributed networks. It shows applicability in a variety of multi-network mining tasks, such as multi-network alignment. Forth (high-order association with neural techniques), we develop an unsupervised multi-resolution multi-network embedding model to simultaneously embed network elements of different resolutions and different networks into the same embedding space. We also present a hypergraph representation learning model via pre-training strategy, with a real-world case study on the inconsistent variation family detection problem for Amazon selection and catalog system."],"dc:format":["application/pdf"],"dc:identifier":["https://hdl.handle.net/2142/116137"],"dc:language":["en","eng"],"dc:rights":["Copyright 2022 Boxin Du"],"dc:subject":["multi-network association","graph mining","machine learning","algorithm","application"],"dc:title":["Multi-network association: Algorithms and applications"],"dc:type":["text","Thesis"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Ph.D."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:24:55Z"}