Lexis LAW comes with a separate application (or module) called Early Data Analyzer. EDA can be used to filter out or cull data, and reduce the overall cost of electronic discovery, based on several criteria:
1. File hashes - for files that do not contain user related data.
3. Date ranges.
4. File types.
5. Email sender domain.
Note that EDA makes it possible to export email threads without attachments that are filtered out.
EDA has settings for analysis which give you the option to use either MD5 or SHA1 either across custodian sources, and within individual custodians. Text extraction can be skipped in order to improve performance. Language identification analysis can be restricted to only a specified number of KB from the beginning of a file, or restricted to only 'common languages'.
The Indexing features can be modified to ignore or include accents, make text case sensitive, or insert spaces between Asian language characters. Email addresses, dates, and credit card numbers can also be automatically identified.
During the import process LAW will generate an exception log which will indicate which files it could not recognize or process.
EDA has a Global Dashboard which allows it to take advantage of the processing power of multiple networked computers.
In addition to endorsing documents with bates numbers, and indexing text extracted from native files, LAW EDA has two different programs which can render OCR, Abbyy FineReader and ExperVision OpenRTK. As noted in the Tip of the Night for May 16, 2015 different kinds of OCR software will generate different results. Abbyy is 2 to 3 times slower than OpenRTK but it supports Asian language characters, can run multiple instances on a single PC, and can generate PDF/A files. See the Tip of the Night for September 25, 2015. EDA can be configured to detect multiple columns of text on a page while OCRing, and also be modified to deal differently with dot matrix, typewritten, or laser & inkjet printed pages.