[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120739-en":3,"doc-seo-120739-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},120739,549758146520,"Patrick","https://ap-avatar.wpscdn.com/avatar/80002397d8c0411e94?_k=1775819394049821470",8,"Research & Report","Modelling the Relationships Between Ground and Buildings Using 3D Architectural Topological Models - A Graph Machine Learning Approach","The document presents a proof-of-concept workflow for modelling the building–ground relationship using 3D architectural topological representations and graph machine learning. It reframes the task as a geometry–topology problem where topology preserves connectivity while abstracting form and physical distance. The approach generates a large synthetic dataset by encoding building/ground grammar rules with Topologic, embedding semantic information into non-manifold topologies, and converting models into graphs for end-to-end prediction with graph neural networks.","Modelling the Relationships Between Ground and Buildings Using 3D Architectural Topological Models Utilising Graph Machine Learning  \nAbdulrahman Alymani, Wassim Jabi, and Padraig Corcoran  \n1 Introduction  \nThe built environment may be considered a collection of three-dimensional (3D) elements that are typically topologically connected. This may be represented by a graph of connected vertices and edges, where a vertex represents an element, and an edge represents the relationship between two or more adjacent elements. One of the crucial relationships that architects and urban designers must acknowledge and analyse is the connection between a building and its surrounding ground (Berlanda, 2014; Porter, 2015, 2017). Understanding this building/ground relationship is inherently a geometrical and topological problem. Computational/algorithmic methods may assist the process and reflect on both geometry and their associated topological graphs using combinational logic.  \nThe principal distinction between geometry and topology is that the latter abstracts away the concepts of form and physical distance yet retains the notion of connectivity. Consequently, complex designs may be identified more efficiently and analysed at far higher levels of abstraction than would be possible solely through geometry (Jabi, 2015) . Prior published research has illustrated the ability to capture several 2D photographic image snapshots of 3D models, after which they are matched to an image-based query (Kasaei, 2019; Sarkar et al., 2017) (see Fig. 1). Nevertheless, these approaches do not comprehensively capture the 3D and topological information embedded in the data.  \nEven if 3D datasets are available, it may prove challenging to recognise them because of their varied formats in terms of appropriateness, usability, and licencing. To address these shortcomings, this paper concentrates on methods from a recent branch of Machine Learning, called Graph Machine Learning (GML), which can  \nFig. 1 Most machine learning systems rely on 20 pixel-based image recognition  \nclassify an entire graph, predict node properties or predict inter-node connectivity. Several studies have investigated this GML promising approach (Kriege & Mutzel, 2012; Orsini et al., 2015; Vishwanathan et al., 2010). Nevertheless, numerous strategies have been hindered by the fact that they must break graphs down into smaller substructures, called paths and walks, then draw similarities based on a summary of the graphs' characteristics. Such restrictions are circumvented by DGCNNs, which categorise graph-based information using end-to-end deep learning (Zhang et al., 2018) .  \nIt is pervasively acknowledged that Machine Leaming models require extensive data for training. Therefore, this paper proposes a novel proof-of-concept workflow in the generation of a sizeable synthetic dataset with embedded semantic information for 3D prototype building. Relationships were encoded using a software programme known as Topologic, which is underpinned by graph theory.  \nThe proposed workflow focuses on enhancing the representation of 3D models based on non-manifold topology (NMT) and embedded semantic information, which may be adopted in relation to different Machine Leaming models. The process is initiated by using shape grammars (Stiny, 1980) to create a building/ground relationship grammar rule. Subsequently, a formal mechanism for defining languages used in 3D spatial designs is established. During the next phase of the process, the Topologic software library automatically and generatively creates a large synthetic dataset of building/ground precedents. Following this, the models are labelled and transformed into topological graphs, comprising comprehensive semantic data that are then passed on to a Graph Machine Leaming system.  \nThe remainder of this paper is organised as follows. Section 2 summarises the historical architectural approach of the building and ground relationship as well as the Toma","cbCaisjCGQSUPFPP","https://ap.wps.com/l/cbCaisjCGQSUPFPP","pdf",1675421,1,19,"English","en",105,"# Introduction\n## Geometry vs. topology in architectural modelling\n## Graph machine learning for graph-based representation\n## Proposed synthetic-data workflow\n# Motivation\n# Building and Ground Relationship","[{\"question\":\"Why is the building–ground relationship treated as a geometrical and topological problem?\",\"answer\":\"The relationship can be represented as a graph where vertices model elements and edges model adjacency relationships. Topology keeps connectivity information even when form and physical distance are abstracted away.\"},{\"question\":\"What role does Graph Machine Learning play in the proposed workflow?\",\"answer\":\"Graph machine learning is used to classify graphs, predict node properties, and infer inter-node connectivity. The workflow provides end-to-end graph representations suitable for graph neural network models.\"},{\"question\":\"How is the synthetic dataset for training created and prepared?\",\"answer\":\"A shape-grammar based rule is defined for building/ground precedents, then the Topologic library generates a large synthetic dataset. Models are labelled and transformed into topological graphs with embedded semantic data before being fed into a GML system.\"}]","Modelling the Relationships Between Ground and Buildings Using 3D Architectural Topological Models - A Graph Machine Learning Approach | PDF",1785731783,48,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"modelling-the-relationships-between-ground-and-buildings-using-3d-architectural-topological-models-a-graph-machine-learning-approach","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/modelling-the-relationships-between-ground-and-buildings-using-3d-architectural-topological-models-a-graph-machine-learning-approach/120739/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is the building–ground relationship treated as a geometrical and topological problem?","Question",{"text":75,"@type":76},"The relationship can be represented as a graph where vertices model elements and edges model adjacency relationships. Topology keeps connectivity information even when form and physical distance are abstracted away.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What role does Graph Machine Learning play in the proposed workflow?",{"text":80,"@type":76},"Graph machine learning is used to classify graphs, predict node properties, and infer inter-node connectivity. The workflow provides end-to-end graph representations suitable for graph neural network models.",{"name":82,"@type":73,"acceptedAnswer":83},"How is the synthetic dataset for training created and prepared?",{"text":84,"@type":76},"A shape-grammar based rule is defined for building/ground precedents, then the Topologic library generates a large synthetic dataset. Models are labelled and transformed into topological graphs with embedded semantic data before being fed into a GML system.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},"General","general"]