Converting OpenStreetMap (OSM) Data to Functional Road Networks for Downstream Applications

In this work, we study the OpenStreetMap (OSM) data that contains Extensible Markup Language (XML) formatted data. OpenStreetMap data has many different formats. OSM XML format is one of them. OSM data has information in the form of nodes (points), ways (lines and boundaries), and relations (relationships between two or more nodes or ways). Here, we preprocess OSM XML data to extract the ways and nodes information using python to get the whole map of the streets for the Memphis area. We parse the OSM data in such a way that gives us the whole map of the Memphis area. We can further use this map for different Neural Networks (NN) and Machine learning (ML) applications. The steps that are included in this work downloading the Memphis area OSM data, understanding and parsing the OSM XML file, converting the nodes and ways information into the Pandas DataFrame, and visualizing these data into the whole map by using python's available data visualization libraries.