Text mining - recognizing text from unstructured data

It understand what the context of the text

The algorithm, based on the overall analysis of the document (and not just the keywords selected), assumes the classifying process by assigning a “subject” of the document from a fixed list and directing it to further processing according to the accepted pattern.

Reads data in a relational structure

Data originally written implicitly, in the natural text of the document (eg, based on the text of the agreement, associative algorithms can read the subject matter of the contract, date, counterparties’ data, value and send this data to relational database of CRM or ERP system).

Detects the mutual context of information occurring in large collections

The algorithm detects that a given text pattern (return term) occurs in certain configurations with other patterns (for example, that the address appearing on the contract is in 60% of sales invoices).

Menerva is able to detect the fact that a given text pattern is present in certain documents in relation to other patterns, for example, that an address in a given contract, also occurs in 60% of sales invoices.