[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-128801-105":59,"doc-detail-128801-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","latent-associations-in-signed-graphs-doctor-of-philosophy-thesis","Latent Associations in Signed Graphs - Doctor of Philosophy Thesis","","Signed graphs model attractive and divisive relations through positive and negative links, challenging conventional network-science assumptions where proximity often implies similarity. Negative edges introduce antagonistic proximity, where nodes that are structurally close may be functionally distant. The thesis studies how to adapt methods for signed graphs to infer implicit connections from explicitly observed structure. It investigates latent associations locally and at macro-scale, showing that interaction type and occurrence context jointly reveal hidden relationships, and proposes embedding and prediction algorithms plus new real-world datasets.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/latent-associations-in-signed-graphs-doctor-of-philosophy-thesis/128801/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/latent-associations-in-signed-graphs-doctor-of-philosophy-thesis/128801.png","ImageObject",300,407,{"name":92,"@type":93},"Violet","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-09-19","2026-08-06",true,{"@type":102,"interactionType":103,"userInteractionCount":34},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What is antagonistic proximity in signed graphs?","Question",{"text":112,"@type":113},"It describes situations where two nodes may be close in the graph structure yet remain functionally distant due to the presence of negative edges and their interpreted interaction patterns.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does the thesis infer latent associations in signed graphs?",{"text":117,"@type":113},"It uses explicit signed-graph connections to uncover implicit relationships, considering both the sign (positive/negative) of interactions and the contexts where they occur, at local and macro scales.",{"name":119,"@type":110,"acceptedAnswer":120},"What are the thesis’s main methodological contributions?",{"text":121,"@type":113},"It proposes a sign-aware, edge-attribute-informed embeddings model; shows triadic closure with multimodal cues can predict edge signs; and generalises the Louvain algorithm to optimize modularity efficiently on real-world signed graphs. It also releases three large open-source signed graph datasets.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},128801,1786003540,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":34,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":129,"read_time":144},1099523885336,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","Latent Associationsin Signed Graphs  \nJohn Ngan POUGU ´E BIYONG  \nLincoln College University of Oxford  \nA thesis submitted for  \nthe degree of Doctor of Philosophy in Mathematics  \nAbstract  \nSigned graphs model attractive and divisive relations with positive and negative links. While graph connections have historically represented similarity, the presence of negative edges in signed graphs questions the conventional paradigm in network science, and introduces the challenging concept of antagonistic proximity: two nodes close in the graph may be functionally distant. A growing body of work has studied the adaptation of conventional methods to understand structural and diffusive patterns in signed graphs.  \nThis thesis investigates the connections explicitly present in signed graphs to infer implicit ones. We seek to uncover latent associations, not only locally but also at the macro-scale, between regions of the graph. We find that the nature of the interactions (positive and negative) and the contexts in which these interactions occur are both important to unveil latent connections.  \nOur contributions are four-fold. First, we design a novel graph embeddings model informed by edge-attribute and sign-aware random walks. The model is able to infer node positions with respect to contexts they have never engaged in. Second, we show that triadic closure, combined with multi-modal cues, is effective to predict edge signs. Third, we generalise the Louvain algorithm, a popular and efficient method for modularity optimisation in unsigned graphs. Our method is able to increase modularity with a competitive computational speed on real-world signed graphs. Fourth, alongside our methods, we build and open-source three large realworld signed graphs to foster future research. Overall, we propose notable contributions using intuitive principles that can be engineered for further improvements.  \nD´edi´e `a Tony. PFM.  \nAcknowledgements  \nIn Spring 2019, having earned a Distinction in the M.Sc. in Mathematical Modelling and Scientific Computing (MMSC), Dr Kathryn Gillow, Course Director, encouraged me to apply for the DPhil.  \nIn an emotional moment, I called my mother to tell her that I was offered the opportunity to do the DPhil, but would not accept. At the time, I felt a duty to relieve our family’s financial difficulties. I had been interviewing for big tech companies with the feeling that it was not where I wanted tobe but also the firm belief that it was where I had to go. She told me not to worry about her. She would do her part: tighten the belt, and protect me with her prayers, and I should do my part: open the doors to something much bigger for our family.  \nAs part of my Masters degree, I had already been working on a Networks project with Professor Renaud Lambiotte. Renaud had made me feel comfortable from the start and I knew that if I were to do a doctoral program, it had to be with him. After noticing the departmental deadline the night prior, Professors Renaud Lambiotte and J. Doyne Farmer interviewed me on a Monday at 10pm. For some reason, this confirmed to me that I was in the right place. Renaud and Doyne promised to help me secure funding and shared some relevant literature. We finished the call minutes to midnight, right before the department’s deadline.  \nThe department granted me an EPSRC scholarship and assigned me the supervisors I had requested. It also approved my request to defer the start of the DPhil to 2020: the school year had been mentally exhausting and I had not had a break since the passing of my uncle Papa during the summer of 2018, just before my MSc course began. In particular, I thank Sandhya Patel, Graduate Studies Administrator, whose support and unspoken understanding were crucial at key moments during my time at the Mathematical Institute. My experience in the department has never been the same since she left.  \nRenaud understood from the start that I was committed to making the most out of the pr","cbCaihcg84t0FnRl","https://ap.wps.com/l/cbCaihcg84t0FnRl","pdf",5357120,114,"English","# Abstract\n## Signed graphs and antagonistic proximity\n## Inferring latent associations\n## Contributions: embeddings, edge-sign prediction, generalized Louvain\n## Open-source signed graph datasets\n# Acknowledgements\n## Funding, supervision, and research development\n## Collaborations and dataset work","[{\"question\":\"What is antagonistic proximity in signed graphs?\",\"answer\":\"It describes situations where two nodes may be close in the graph structure yet remain functionally distant due to the presence of negative edges and their interpreted interaction patterns.\"},{\"question\":\"How does the thesis infer latent associations in signed graphs?\",\"answer\":\"It uses explicit signed-graph connections to uncover implicit relationships, considering both the sign (positive/negative) of interactions and the contexts where they occur, at local and macro scales.\"},{\"question\":\"What are the thesis’s main methodological contributions?\",\"answer\":\"It proposes a sign-aware, edge-attribute-informed embeddings model; shows triadic closure with multimodal cues can predict edge signs; and generalises the Louvain algorithm to optimize modularity efficiently on real-world signed graphs. It also releases three large open-source signed graph datasets.\"}]","Latent Associations in Signed Graphs - Doctor of Philosophy Thesis | PDF",287]