How to build a decision-tree / an neural network to classify frames of accelerometer-data provided by a smartphone.

KNIME Analytics Platform
1

Hi there,

I am new to Knime but I want to use it as part of my Bachelor-Thesis.

It is my goal to train a decision-tree / an neural network / or whatever achives the best results in classifiying frames of accelerometer-data provided by a smartphone.

 

Basicly I have something like the attachment and I want to know/train wether there is some specific movement.

 

Do you have some tipps and tricks, how I should build it, which functions are helpfull and maybe some usefull tutorials or something like that.

 

Thanks a lot.

Jan

2

Hi Jan, 

This is pretty basic but to start, have a good look at the Lag Column node in KNIME.  This will let you effectively look back in time for each row in your data.  From there, it should be a normal(ish) machine learning problem. 

 

3