Classifying Roads with Multi-Step Graph Embeddings
Abstract: Machine learning-based road-type classification is pivotal in intelligent road network systems, where accurate network modelling is crucial. Graph embedding methods have emerged as the leading paradigm for capturing the intricate relationships within road networks. However, their effectiveness hinges on the quality of input features. This paper introduces a novel two-stage graph embedding approach used to classify road-type. The first stage employs Deep Autoencoders to produce compact representation of road segments. This compactified representation is then used, in the second stage, by graph embedding methods to generate an embedded vectors, leveraging the features of neighbouring segments. Results achieved, with experiments on realistic city road network datasets, show that the proposed method outperforms existing approaches with respect to classification accuracy.
| Main Authors: | , |
|---|---|
| Format: | Digital revista |
| Language: | English |
| Published: |
Instituto Politécnico Nacional, Centro de Investigación en Computación
2024
|
| Online Access: | http://www.scielo.org.mx/scielo.php?script=sci_arttext&pid=S1405-55462024000100257 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract: Machine learning-based road-type classification is pivotal in intelligent road network systems, where accurate network modelling is crucial. Graph embedding methods have emerged as the leading paradigm for capturing the intricate relationships within road networks. However, their effectiveness hinges on the quality of input features. This paper introduces a novel two-stage graph embedding approach used to classify road-type. The first stage employs Deep Autoencoders to produce compact representation of road segments. This compactified representation is then used, in the second stage, by graph embedding methods to generate an embedded vectors, leveraging the features of neighbouring segments. Results achieved, with experiments on realistic city road network datasets, show that the proposed method outperforms existing approaches with respect to classification accuracy. |
|---|