I am looking for a dataset possibly with more than 5 classes and 5000 to 10,000 features. Can anyone suggest or provide me a link of a high dimensional multi-class dataset for the dimensionality reduction task?
If you don't work on image, signals maybe will be a good suggestion
You can exploit a dataset of human activity recognition using smartphone to classify six activities: performed were as follows: Walking, Walking Upstairs, Walking Downstairs, Sitting, Standing, Laying. The data collected is the x, y, and z accelerometer data (linear acceleration) and gyroscopic data (angular velocity) Observations were recorded at 50 data points per second
Farzana Anowar https://archive.ics.uci.edu/ml/datasets/human+activity+recognition+using+smartphones
You can see :
https://www.analyticsvidhya.com/blog/2018/08/dimensionality-reduction-techniques-python/
During my master program I studied Markov blanket techniques with mematic algorithm to make dimensionality reduction for several types of DNA-microarrays.
You can make search for
"Markov blanket: Efficient strategy for feature subset selection method for high dimensional microarray cancer datasets"
check out this dataset it is contained 84 features contains with high data cardinality:https://www.unb.ca/cic/datasets/ddos-2019.html
you can use PCA,tSNE...etc
if you want more than 84 features I can share you medical dataset.
You should mention in your question what specific task are you going to perform? Then we will be able to recommend any of dataset for you.
You can try out ECG signal dataset from UCI and MIT-BH databases
- Badges
- Science topic