FlexiScribe Data Processing System

Data entry, and data processing, have always been a tedious, time-consuming manual process, subject to errors at every step, especially when extracting data from scanned images or handwritten documents. The vast variety of document types, project requirements, and data formats has prevented the development of general purpose automation; until now.

MBS-NL, a company with deep experience in both application software development and data processing, is proud to announce the birth of FlexiScribe, a flexible and powerful application employing a revolutionary concept of handling input and processing information.

FlexiScribe combines automated document recognition, data capture, QA, and output data transformation and formatting into a single modularized process, automatically routing items to human operators when required for interpretation or verification. Multiple modules of any type can be easily configured for specific purposes, and keyed to the Document Recognition module for identification and appropriate sequencing. Thus, any number of projects large or small can be worked and intermixed within a single operating system with automatic compensation for volume fluctuations while providing specific processing sequences and tools for each unique requirement.

FlexiScribe allows data capture and data entry for a stream of incoming documents. The Manager (or the Administrator) can define different data processing tasks for different types of documents. In general case this will be processing of scanned images, submitted by the customer. Depending on the type of images and the tasks required the incoming images are automatically routed to different services.

Document Definition and Classification: This is the beginning of processing when documents are formed from received files, grouped together, and assigned the task to be performed and the services, which will perform this task.

Document Identification: This service is used for automatic identification of the type and the areas where the necessary data will be captured.

Automatic Data Capture: This service employs an OCR engine to capture meaningful data from the images and convert the data into electronic form filling in all possible fields.

Manual Data Capture and Verification: This service requires employing human operators to either capture data from an unstructured input or to verify the results of automatic data capture (or other operators)

Automatic QA: This service allows to perform automatic quality assurance of the data either entered manually or captured automatically. Having such service allows decreasing the number of mistakes in the final output.

Data Transformation and Output: The core of this service is conversion of the captured data into any format required by the customer, and store or ship it depending on the request.

FlexiScribe is not just a Data entry utility. This is a very flexible and powerful application that employs a revolutionary concept of handling input information and processing.

As the name implies, the system is FLEXIBLE. The core of the system is the Database and File Repository, which serve as the file storage and the control centre. Apart from the core, which is stationary and stable, the rest of the system is flexible and adaptive, meaning any number of any services can be added to the system.
For example, unique data capture requirements can be accommodated with additional Automatic Data Capture services tailored to the requirement. The number of services can be increased or decreased at any time.

On the other hand any new service can be introduced to the system without affecting the work already being done. The service will start its work and will wait until the documents ready to be processed appear in the system. New documents, which require processing by the new server, will be routed to it and processed.

Another important feature of FlexiScribe is handling of incoming files. Whereas most of existing systems are tuned for processing of one or several types of image files to produce the data output, FlexiScribe is not constrained by the type of incoming files. Every file comes to the system’s input is regarded not as an “image file” but as a Data Container. This allows to eliminate the restrictions on processing only images.

This flexible and configurable architecture permits addition of any number of services tailored for any kind of data, language, document format, image file, or information structure. As an example, a single project can be configured to combine information from intermixed WinWord docs, .tif images, and EDI files, and output it to a client’s proprietary database format.

FlexiScribe introduces an entirely new level of flexibility and response to MBS-NL’s worldwide Data Processing division; and represents another targetted success for MBS-NL’s Software Development team.