| Paper Abstract and Keywords |
| Presentation |
2024-06-06 13:45
GSP-Traffic Dataset: Graph Signal Processing Dataset Based on Traffic Simulation
Rui Kumagai (Osaka Univ), Hayate Kojima (Tokyo Univ of Agriculture and Technology), Hiroshi Higashi, Yuichi Tanaka (Osaka Univ) |
| Abstract |
(in Japanese) |
(See Japanese page) |
| (in English) |
Data analysis on graphs, including graph signal processing and graph neural networks, is an emerging research field in signal processing and machine learning. To quantitatively compare the performance of these methods, reliable datasets containing graphs as well as graph signals are required. However, such datasets are currently limited, and many studies use synthetic graphs and/or graph signals for experiments. In this paper, we propose GSP-Traffic Dataset, a large-scale time-varying graph signal dataset on simulated traffic networks. Our dataset utilizes a traffic flow simulator SUMO and it contains graph signals across multiple cities, facilitating easy comparison of graph signal properties and features. It also utilizes actual road networks: This makes the dataset reliable. This dataset is beneficial for the quantitative comparison of methods of data analysis on graphs within the dataset. Moreover, we made an experiment using our dataset and confirmed the potential for a wide range of applications. The dataset will be publicly available at https://github.com/rukumagai/GSP-Traffic-Dataset. |
| Keyword |
(in Japanese) |
(See Japanese page) |
| (in English) |
Graph Signal Processing / Dataset / Reconstruction of time-varying graph signal / Simulation of Urban MObility (SUMO) simulation of urban mobility (SUMO) / / / / |
| Reference Info. |
ITE Tech. Rep. |
| Paper # |
|
| Date of Issue |
|
| ISSN |
Online edition: ISSN 2424-1970 |
| Download PDF |
|