No matter how good recognition engines are, sometimes they make
mistakes. There are times when a recognizer will generate results that,
while having a "high confidence" level, are very wrong. To catch these
instances before they become problems,
What do you expect the value to be in any field? Is it always a number? Can it only be 10 characters long? Should it always lie between a minimum and maximum value? All of these are logical rules that can be made to a field value, either after recognition, or after manual data entry. Logic checks of this kind are applied according to the underlying business rules that describe the desired data. To be effective, the rules processing approach must allow as few or as many edits to be applied to a field, mirroring the business rules governing the data.
For example, a US Social Security number may have these rules applied to it:
Must be all digits - Allow leading zeros - Must be exactly nine characters long.
One of the best ways to validate data is to consult a definitive list of valid values. When that list is stored separately from the validation rule itself, then the rule performs a "lookup" to see if there is a match. To use the example above of a Social Security number validation, if the expected value was, for example, all taxpayers in a given state, then the validation rule might look like:
Must be all digits - Allow leading zeros - Must be exactly nine characters long - Must match list in large database table
If the lookup "fails," that is, no match is found, then the field will fail the validation, just as if it had only been eight characters long. Advanced lookup rules will not only match values, but can also pre-fill other fields based on data that it retrieves from the matching record, for example the taxpayer's name and address after matching the Social Security number.
Most applications have many nuances - you want to validate your data according to your own business rules. That's where scripting comes in. With a script in the right place you can do exactly what your application requires. But scripting requires a programmer. If you could get the power of business rules without scripting, then it would be much easier to build your capture app. Taskmaster Rules are scripts, but you don't need to do scripting to use them. That's because Datacap provides hundreds of "actions" that can be assembled into business rules. Each action is a script, so savvy users can enhance existing actions or create new ones. But an experienced business analyst can also create his or her own actions for more nuanced rules. Learn how Taskmaster Rules work.
back to top | request more information | contact Datacap | site map