Problems with KNIME for clusteranalysis

Hello,
I’m a beginner with KNIME and found with tutorials and other stuff no solution for my problem. That´s why I´m searching here for help.

I tried different ways to find a solution by myself but right now now I´m lost.

I have the problem that I can’t use my table .xls or as .csv in KNIME via the reader.
I am new to KNIME and wanted to analyze my data and examine it for patterns.
Even with simple numeric tables I can’t manage to get a process for analysis to run.

Can anyone help me? Does anyone have maybe an example to be able to trace content?

Thanks for any hints or advices :slight_smile:
Bernd

Hi there @Bernd_Student,

welcome to KNIME Community!

not sure what problems are you having? Reading your data into KNIME or something else?

Br,
Ivan

Hi @ipazin,

both is correct.
My first problem was to read my data into KNIME. First I attemted to read a diffrent excel sheed into KNIME and now there was the problem that everytime I try to read another sheet, the first try is shown. With adjust settings and select another sheet (not ) it worked. So I hope that is working now.

The other problem is a bit tricky to explain. I tried now different knots but get everytime an error. So maybe I can explain it what I want to do.
I want find out if there’s a pattern when I analyze data. As an example the supplement management. Each supplement has different positions. Supplement A has 4 different items that all have to meet certain criteria to be approved. (Columns with words and numbers in the same row). What mostly has to be fulfilled for approval. Do you think it is possible with KNIME?

BR
Bernd

Hi there,

so I guess you have your data in KNIME now. That’s good. Considering KNIME is a tool for data analysis I would say that most probably you can use it in your case but still not sure what case that is. Can you share part of your data and expected result?

Br,
Ivan

Hi,
yes I try to share some data parts. Hopefully you can imagine what I want to do.

I want to proof if there is a correlation between supplements provided and approved supplements.
Target:
Show samples under Recognized Supplements and Rejected Supplements. To identify particularities of a quantitative nature.

Implementation:
Perform classification and use a decision tree to clarify which constellation leads to acceptance/approval and which constellation leads to rejection.

Target:
Are there outliers in supplements that require special attention?
Implementation:
Cluster analysis for visual detection of outliers.excel sheett.xlsx (14.5 KB)

Thank you so much for your time!

Br,
Bernd

Hi there @Bernd_Student,

this seems like a student assignment, right? If so I suppose you are the one that should do it then :wink:

Anyways I can share some links that can help you.

For decision tree model in KNIME check this example: https://kni.me/w/gxrPOF2R8QCJCBk0

For different outlier detection techniques check this example: https://kni.me/w/jP4OxhLiSQZmgFo6

For cluster analysis check this example: https://kni.me/w/QqKlIQmDnCAg0lEQ

All examples are from KNIME Hub. There you can find more example. If any particular problem with certain part of your assignment or node feel free to ask here :slight_smile:

Good luck!

Br,
Ivan

1 Like

Hey @ipazin,

yes right. I just thougt that someone have maybe some good hints for me.

Thanks maybe the links can help.

Br
Bernd

1 Like

This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.