Ernst, Sebastian2023-09-212023-09-212023Ernst S., Graph-Supported Preparation of GIS Machine Learning Datasets. W: Progress in Polish Artificial Intelligence Research 4, Wojciechowski A. (Ed.), Lipiński P. (Ed.)., Seria: Monografie Politechniki Łódzkiej Nr. 2437, Wydawnictwo Politechniki Łódzkiej, Łódź 2023, s. 107-112, ISBN 978-83-66741-92-8, doi: 10.34658/9788366741928.15.978-83-66741-92-8http://hdl.handle.net/11652/4790https://doi.org/10.34658/9788366741928.15The paper presents an approach to preparing spatial (GIS) datasets for machine learning models, and using graph structure to materialise and utilise the results. The presented work is based on the Spatially-Triggered Graph Transformations (STGT) methodology, previously used for many realworld applications, e.g. in the area of smart cities. A workflow using OSM data is presented, aimed at improving the granularity and semantic annotation of map features.enDla wszystkich w zakresie dozwolonego użytkuFair use conditiongraph transformationsMLclassificationGISsmart citiestransformacje grafówMLklasyfikacjaMLinteligentne miastaRozdział - monografiaLicencja PŁLUT License10.34658/9788366741928.15