Document Data Capture 

Document data capture is the process of extracting and keying important document data in order to create a repository that will enable end-users to access, identify and process documents and information quickly. SVI has developed its own method of data capture that consists of technical abstraction and document indexing.

Technical Abstraction

Data abstraction is the process of identifying and marking data fields for use in intelligent business applications (i.e. account management, litigation support, healthcare information management). Our two-step process involves highlighting and keying of specified fields on the document images.  Specialized knowledge workers best fitted to identify the data fields highlight these on the document images.  Professional encoders then encode the highlighted fields. This two-step process guarantees that the correct fields are encoded and are of the highest accuracy.

Documents for function-specific applications are best understood by specially trained professionals (ie. accountants, lawyers, doctors). SVI employs experts in the field to analyze and examine the documents, and to perform data abstraction.

Document Indexing

Document indexing is the process of keying documents in order to create identifier fields of documents in the database.  These indexes are used to search for specific documents with ease and allow the user to retrieve the correct document image within seconds.

Indexing may be done manually or software-assisted via Optical Character Recognition (OCR).

Manual Indexing

SVI offers manual indexing services that guarantees 99.95% accuracy rate.  This service is designed to meet critical business indexing needs that requires the highest accuracy rating. Our manual indexing services are able to achieve your index accuracy requirements even when indexing unstructured documents and even handwritten documents.  This is achieved through our double blind two-pass indexing process, which guarantees accurate indexes as well as the confidentiality of your data.  To fulfill your mission critical indexing needs, the indexing process can be customized to produce 100% accuracy.

Software-Assisted Indexing via Optical Character Recognition (OCR) 

OCR is an automated conversion process of scanned/electronic images to text files to allow the manipulation of data contained in the images. OCR serves as a complement to manual document indexing.

SVI utilizes parallel computing technologies for OCR to drastically shorten the amount of time required to process large volumes of documents. To ensure quality, output is checked using Word Confidence Rating (WCR) standards.