MATLAB code to cluster categorical data

Hello everyone,

I am new to MATLAB programming. I want to cluster my categorical data. I have six columns which five of which are non-numeric data, and one column is numeric data as follows:

• Market (non-numeric data)
• Vet Drug (non-numeric data)
• Published Commodity (non-numeric data)
• Index Commodity Group (non-numeric data)
• Index Commodity (non-numeric data)
• MRL Amount (ppb) (numeric data)

There are 1545 data (number of rows). I am attaching part of the data here. Is there any suggestion or recommendation for the best and appropriate method to cluster my categorical data so that we get useful information by looking at the results and plots? Any help would be appreciated.

Thanks,

[Mahesh Taparia]

Hi

Do you have the associated labels of your data? If not, do you know the number of categories the dataset have?

[N Z]

Hello Mahesh,

Thank you for your message and reply. My data does not have any labels, and I want to use an unsupervised learning method to classify my data into optimal classes. I do not know the number of categories the dataset have in advance; however, we can categorize our dataset into 3.

I really appreciate any help you can provide,

Answer 1

[Mahesh Taparia]

Hi

You can refer this documentation of clustering. Hope it will help!