I'm searching for my thesis about the time-series (non-video) datasets for the purpose of clustering (and classification). By non-video I mean the dataset should not be video data. Any idea of the dataset or where I can get such data? Any suggestions are welcome.
Add a comment
|
2 Answers
Check out enigma.io's 'Public Data Explorer' and search for business license permits. This will result in a long list of locations that you can the use the service's filtering feature to hone in on a small enough area to provide a dataset that is manageable and not massive in size. After doing that, you'll have the address data of license permit issuances and the date they were first issued. That will then allow you to plot urban development over time geospatially.
There will likely be such high-density of permits you'll have an example ripe for clustering via space on map and classification by permit type.
The Monash Time Series Forecasting Repository contains multiple different time series datasets from various domains. These have been collected from forecasting competitions or other previous forecasting use, so you may also be able to learn something from previous work that used a particular dataset. Some of these data should also have been used in clustering previously.
An introduction and description of the repository, along with more information about the datasets, is given by Godahewa et al. (2021).
