Introduction

We’ve discussed automating various documents using paper, web and mobile capture. We switch our attention to processing forms and surveys, common surveys such as patient/customer/staff feedback, comment cards and any sort of research that requires information collection for a select target audience.Clinical trials can be aided by data capture

How it works

We will not going into the details of the techniques of automation as this has been covered and addressed in invoice processing and managing business process, when it comes to automating form and surveys the process is the same.

Instead we’ll be focus on the extraction techniques. The type of data held in the form will determine the extraction technique. There are four different extraction techniques. Just to recap that before the extraction stage, the forms have already been scanned. The methods vary as follows.

Optical Mark Recognition

OMR is the extraction technique used to automate data from quantitative sources i.e. multiple choice questions where either a tick or a cross is required. As the forms are being scanned, the powerful extraction tool will pick up the areas where there is a mark such as a tick and extract the data. The problem that can arise with this method is that sometimes creases in paper are assumed to be a mark and in cases where respondent has changed their answer hence two marks are picked up. It is at this point highly skilled verifiers take control and apply common sense where necessary.

Intelligent Character Recognition

ICT extracts hand written entries. It is important to distinguish between handwritten entries and free text. The former is written within constrained boxes i.e. each letter/number separated in each box whereas the latter is completed unstructured enabling the respondent to write as they want.

Intelligent Word Recognition

As mentioned the different between handwritten and free text, IWR is responsible for extraction free text.

Optical Character Recognition

Lastly, there is OCR, responsible for extracting machine printed text such as barcodes, serial numbers and so forth

Once the data is extracted as per the appropriate technique, it is verified to ensure that it is accurate and exported to the desired destination.

Conclusion

It is important to bear in mind that as sophisticated as these extraction techniques are, they are prone to error-like any technology, that is why we have a team of highly skilled verifiers at hand who override these errors to ensure the data you have received is accurate with no compromises.

Any organisation who is embarking in data collection wants the data as quickly as accurately as possible. If you’d like further information please do not hesitate to contact us. Alternatively, book your free demo. The purpose of the demo is to enable a first-hand understanding of the solution. Furthermore, we’re able to demo the solution using your own documents as that will facilitate an even better understanding.

Like it? Share it!