The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
The label propagation procedure can be used to predict cluster membership for new, unlabeled observations based on their similarity to previously labeled observations. These unlabeled observations could be a held out test set from your original sample or a new sample entirely.
The process involves the following steps:
There is a lot of room for flexibility in how steps 1 and 2 are conducted. SNF is not necessary at any part of the process. For example, step one could be done by assigning clusters in your training set manually or by a simple clustering method like k-means. Step two could be done just by calculating the euclidean distances across all the training and testing observations for a small subset of features. The features used to calculate the similarities in step 2 don’t necessarily need to be the same ones used to derive the cluster solution in the training set either.
A worked example of the label propagation process can be found at the end of the complete example vignette.
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.