I'm curious as to whether research been done into random forests that combine unsupervised with supervised learning in a way allowing a single algorithm to find patterns in, and work with, multiple different data sets. I have googled every possible way to find research on this, and have come up empty. Can anyone point me in the right direction?
1 Answer
Semi-Supervised Learning
The combination of unsupervised learning and supervised learning is referred to as semi-supervised learning
, which is the concept that I believe you are searching for.
Label propagation
is often cited when outlining the heuristics of semi-supervised learning
. The essence is to employ clustering, but to use a tiny set of known cases in order to derive (or propogate) the labels of the clusters. Hence one is able to use a small set of labeled cases to classify a much larger set of unsupervised data.
Here are some references:
- Wikipedia has an entry on the
semi-supervised learning
. - The scikit learn User Guide is often a useful starting point and has a label propogation routine.
- There are, in fact, papers treating
semi-supervised
random forest
models. - Another one here
Hope this helps!
-
1$\begingroup$ This is genius! Thank you so much. I am new to data science (as a university student) and I had no idea where to start to learn for myself. Those links are perfect. $\endgroup$ Dec 31, 2015 at 0:25