Hi,
I'm new in OV- I made myself a Little bit familiar with it, but have still problems in understanding the implications of the terms STIMULATION, CLASSIFICATION and FEATURES.
Where can I find any documentation which explains these terms in detail?
Thank you in advance for any advice
Peter
Implication of STIMULATION, CLASSIFICATION and FEATURES
-
jtlindgren
- Posts: 775
- Joined: Tue Dec 04, 2012 3:53 pm
- Location: INRIA Rennes, FRANCE
Re: Implication of STIMULATION, CLASSIFICATION and FEATURES
Hello Peter,
There might not be documentation summarizing these terms, but you can see how they are used in the various BCI scenarios included with openvibe.
In short, stimulation is a bit odd name for discrete event markers. Basically a stimulation means that some event happened in some point in time. If some segment of data is associated with at certain brain state, such as the user imagining right hand movement, we can send a stimulation designating that at the presumed start of the segment. This is also called a label for this segment. Features, on the other hand, is a common machine learning term for describing some presentation of the data that is associated with some state. In openvibe, features are vectors of numeric values. Once the data segment has been massaged into a feature vector in some manner, it can be used as part of training data set to a train a classifier or to make a prediction with a trained classifier. The term classification means turning these features into predictions of the user state. It's these predictions that then control the application in BCI.
Aside from the concept of stimulation, the two others should be very well described in machine learning tutorials you can find from the net.
Hope this helps,
Jussi
There might not be documentation summarizing these terms, but you can see how they are used in the various BCI scenarios included with openvibe.
In short, stimulation is a bit odd name for discrete event markers. Basically a stimulation means that some event happened in some point in time. If some segment of data is associated with at certain brain state, such as the user imagining right hand movement, we can send a stimulation designating that at the presumed start of the segment. This is also called a label for this segment. Features, on the other hand, is a common machine learning term for describing some presentation of the data that is associated with some state. In openvibe, features are vectors of numeric values. Once the data segment has been massaged into a feature vector in some manner, it can be used as part of training data set to a train a classifier or to make a prediction with a trained classifier. The term classification means turning these features into predictions of the user state. It's these predictions that then control the application in BCI.
Aside from the concept of stimulation, the two others should be very well described in machine learning tutorials you can find from the net.
Hope this helps,
Jussi
Re: Implication of STIMULATION, CLASSIFICATION and FEATURES
Thanks a lot, Jussi- yes, that helped very much.
Peter
Peter