Hello together,
I am still new to KNIME, so please excuse me if the question is about “Basic”.
I would like to “feed” an AI with data based on recorded motion data (including: longitude, latitude, altitude, speed), so that the AI is finally able to recognize the type of transport (e.g. car, bicycle, running, scooter, …).
The motion data are available as CSV files. In the columns are the above mentioned data, in the rows the values for each second during the measurement.
Can I create a program in KNIME that automatically reads all CSV files in a folder?
Or do I have to load each CSV file individually into the program to train the AI?
Can an AI, for example as a decision tree or with a clustering method, work with several tables?
Furthermore, I have the following question: Is it possible to calculate the acceleration at any time based on speed and time, in other words, to make a kind of derivation?
You can read many files into a KNIME workflow. There are several ways to accomplish this. You can search through the KNIME Hub for examples you can download and play with it. Heres a few:
I’m not really sure I understand what the second question is asking, but I’ll take a stab at it. It is possible to calculate a specific point in time calculation if you have the data, with specific nodes like Math Formula, Rule Engines, and Column Expression Nodes for example. You can use Create Date & Time Range to either get a specific date time or the range of times.
Regarding the second question. Here I may have expressed myself a bit unclearly:
I have been giving the speed of a vehicle at certain points in time (as shown in the attachment). From this information and the change in time I would like to calculate the acceleration.
Do you understand what I mean?
Yes I think there are a couple of ways you could do this. One that comes to mind would be to use the Lag Column to get the previous record’s value, then use the Math Formula to perform the calculation.
