How to group strings based on similarities in the string

Question

I have a splunk query that produces a summarises errors by frequency

index="pc_1" LogLevel=ERROR 
   | eval Message=split(_raw,"|") 
   | stat