Hi Splunk Community,
Sorry if my question is basic but I am new to ML usage in Splunk. I saw an inbuilt example of Splunk ML model which is to predict the presence of malware. So my idea was to use the the same model which contains email data with status of "malicious" or "not".
I used around 2000 malicious emails and 2500 thousand non malicious emails.
When I am using that CSV file in Predict Categorical Fields using any ML model like logistic-categorical or random-forest I am getting error of "No valid fields to fit or apply mode to.
Here I am trying to predict the status field.
SPL Query:
| inputlookup email_data.csv | head 5000| fit LogisticRegression fit_intercept=true "status" from "fromAddress" "messageid" "senderIP" "senderdomain" "subject" into "example_malware"
Any help of if anyone implemented this and can suggest me with example would be really helpful.
