Document classification and extraction of the desired data to process them in a structured way. These documents can be forms or less structured documents such as powers of attorney or land registry certificates.
Search and selection of relevant or interesting data
It allows locating public documents of interest, classifying them, and extracting relevant data beyond what a simple search would find. For example, locating addresses from official gazettes. Also applicable to the location of bid specifications and tenders.
Customer support development
Creation of knowledge base
A knowledge base is generated out of product manuals, guarantee conditions, or policies. It extracts and relate entities such as products, brands, contract expiry dates, etc. With graph search or a query, it can find the answer more effectively than a mere indexed search engine.
It can work with PDF documents, scanned documents, photos, cards, forms and plain text.
It can classify the type and subtypes of the document to treat each case differently.
It can be implemented as a search engine or a query application with all the documents processed and stored.
The system can detect the language of the document automatically.
It allows to locate and extract the phrases and clauses that have been trained.
IMPLEMENTATION IN WEB SERVICE
It can be implemented as an API that can be called, document to document, for integration with your existing applications.
If necessary, the system can translate the content of the document.
It can locate and extract brands, companies and organizations regulations, names of people, dates, locations, city and country, as well as any interesting data liable to be trained.