Considering all of the ways that businesses rely on data, the benefits of increasing the accuracy of information and processing it more quickly can be huge. From improving internal and external transactions to enhancing customer service; increasing collections and saving time, money and the hassle of researching incorrect data, most organizations stand to gain both monetary benefits and increased business efficiencies. In some industries, such as in healthcare, where disease codes are used frequently, and in banking, where each correct digit counts, it can have an even greater impact.
Context and business rules are two tools companies can use to process information faster and improve the quality of data. These magic bullets perform some of the thinking or logic for organizations processing data. They determine how information should be entered increasing the accuracy of data recognized by software, speeding up the process and reducing the amount of manual data entry that will be required. These rules allow more information to be recognized automatically during the first pass of processing, transfer the same context to keyers during the second phase of data entry, and help to identify and verify low confidence fields during the final validation stage.
Context plays a significant role in the recognition process by helping to explain the characteristics or properties of a field. When humans read handwriting or print they look at entire words and even the entire document to correctly identify what is written. Knowing a range of probable meanings makes the task of reading much easier. This is why recognition engines use context as an effective and flexible tool to compensate for the inherent ambiguity of handwriting and to improve recognition accuracy.
Rules for context can be applied seamlessly during recognition (for instance, to tell the engine that the next character must be a 0 and not an O), during data entry (with the same numeric rule for keyers to follow, or by having the system set up so that individuals cant enter anything to the contrary) and in validation to check one last time for accuracy. In many programs, a letter or number in a field may be highlighted to prompt verification of questionable information.
Business Rules, on the other hand, are referred to as the logic. They employ if/then thinking or scenarios–such as if this field is X, then the answer must be Z–to determine the next valid answer or to confirm accuracy of one or more fields. Business rules help to ensure that the data meets certain criteria. They can be used to automatically populate fields with database lookup (such as, if the code is 00123, then the name, address and phone are automatically x, y, z) and also to eliminate errors (such as when an area code/address combination is known and can be used to check/repair a digit on an answer to determine if it should be a 3 or an 8). Other common uses include matching ZIP codes with appropriate mailing addresses for address recognition, or verifying the numeric amount (i.e., $108.35) on a check with the alphanumeric amount (i.e., One hundred eight and 35/100) for check processing applications.
Both context and business rules can be used to increase accuracy during each stage of data capturein character recognition, to determine answers, increase accuracy and reduce keying; during keying, with sample tests, and finally in validation. Especially when used together, context and business rules offer greater speed and efficiency and reduce costs and time. Looking at their own data, most organizations can determine numerous ways to employ context and business rules for process improvement.
Learn the latest in automated handwriting recognition technology in the free white paper Automating Handwriting Recognition Not all ICR software is created equal.