2026-07-23T05:29:48Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:18948 2024-03-06T11:03:02Z ddc:004 open_access

Exploring the Potential of Machine and Deep Learning Models for OpenStreetMap Data Quality Assessment and Improvement (Short Paper) Miloudi, Salim Meguenni, Bouhadjar OpenStreetMap (OSM) Volunteered Geographic Information (VGI) Machine Learning (ML) Deep Learning (DL) Quality Assessment (QA) Building Footprint Detection Semantic Segmentation The OpenStreetMap (OSM) project is a widely-used crowdsourced geographic data platform that allows users to contribute, edit, and access geographic information. However, the quality of the data in OSM is often uncertain, and assessing the quality of OSM data is crucial for ensuring its reliability and usability. Recently, the use of machine and deep learning models has shown to be promising in assessing and improving the quality of OSM data. In this paper, we explore the current state-of-the-art machine learning models for OSM data quality assessment and improvement as an attempt to discuss and classify the underlying methods into different categories depending on (1) the associated learning paradigm (supervised or unsupervised learning-based methods), (2) the usage of extrinsic or intrinsic-based metrics (i.e., assessing OSM data by comparing it against authoritative external datasets or via computing some internal quality indicators), and (3) the use of traditional or deep learning-based models for predicting and evaluating OSM features. We then identify the main trends and challenges in this field and provide recommendations for future research aiming at improving the quality of OSM data in terms of completeness, accuracy, and consistency. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Salim Miloudi and Bouhadjar Meguenni 2023 Is Part Of LIPIcs, Volume 277, 12th International Conference on Geographic Information Science (GIScience 2023) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/LIPIcs.GIScience.2023.53 urn:nbn:de:0030-drops-189486 https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.GIScience.2023.53 eng https://creativecommons.org/licenses/by/4.0/legalcode