I want to find four different database that have more than 6 dimensions to test clustering methods. I found an image and a time series data set. but what do you suggest for other databases?
Text corpora are interesting as well -- using unigram-, bigram-, trigram-, n-gram- bag-of-word document vectorization models, the number of dimensions easily winds up in the hundreds of thousands to millions.
The Reuters corpora: http://www.ai.mit.edu/projects/jmlr/papers/volume5/lewis04a/lyrl2004_rcv1v2_README.htm and the "20 newsgroups" corpus: https://archive.ics.uci.edu/ml/machine-learning-databases/20newsgroups-mld/ are classics to consider.
At this moment iam workind on human activity recognition from sensory data (infrastructure sensors) there is a variety of labeled datasets including many classes here:http://ailab.wsu.edu/casas/datasets/
I don't have a good database for clustering in mind but I can suggest a very good and difficult data set: The IRIS data set.
I believe the best data set to begin with testing your algorithm(s) is the IRIS data set. There is a significant overlapping between the points that represent two of the three classes involved, so this is a god tool to evaluate your algorithm's ability to deal with uncertainty.
This website has many databases for all sorts of problems:
https://archive.ics.uci.edu/ml/datasets.html
The Iris data set suggested by Professor N.B. Karayiannis is fine, but if you want an even more difficult data set with 6 or more attributes as you asked, I suggest the equally difficult Wine data set. (you can find it on the website I linked)
https://archive.ics.uci.edu/ml/datasets.html
http://archive.ics.uci.edu/ml/datasets.html?format=&task=clu&type=&sort=nameUp&view=table
http://sci2s.ugr.es/keel/category.php?cat=uns
http://www.ncdc.noaa.gov/
and
https://archive.ics.uci.edu/ml/datasets/Iris
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA