DSXchange

How do you see the rows which have fallen into unhandled patterns?

What do you exactly mean by unhandled patterns?

"Unhandled pattern" is a column in the report from an Investigation. If that field (unhandled pattern) is populated, then the row has an unhandled pattern. Note that there is also "unhandled string" which can be useful in deciding whether or not you need to create a new rule or a rule override.

My Bad. Didn't realise that the post was related to Quality Stage and hence the confusion.

ray and kris, thank you very much for your posts. ray, can you tell me how I can find out which rows have unhandled patterns? For example, if there are 1 million rows and 3000 of them fall in a pattern, how do I look at those 3000 only? The PAT, FRQ and SRT files don't have that info.

Is there a technique that you gurus use that you can share? Thank you very much.

The SRT file is sorted by pattern. But that doesn't mean you can't re-sort it by unhandled pattern, or run it through a filter that eliminates those rows that have an empty (all spaces) unhandled pattern field.

So really there is no way other than manually going through the data?

Correct. In the fullness of time, as you tune your rule set and rule overrides, you expect to have fewer unhandled patterns. You will always, of course, have unclassified tokens, but these can still be handled in patterns. For example in the ?,FI pattern the unknown token in a name rule set is going to be placed into the LN (primary name) bucket.

You can use a DCT field to store a value which time a pattern is used.

But to achive this you will need to change your pattern file.

At the end, the DCT field will have a value match to the pattern that treat which record. If its empty, no pattern had match.