Documents can be retrieved from email inboxes or remote locations automatically
Intelligent Data Recognition
Reduce time needed for analyzing documents and emails. Automate user tasks in your systems or build a standalone solution.
Reduce time needed for analyzing documents and emails. Automate user tasks in your systems or build a standalone solution.
IDR is a tool that helps you:
Extract data from documents for further processing
Anonymise documents automatically
Categorise emails to assign departments or to start according processes
Documents can be retrieved from email inboxes or remote locations automatically
Any on-premise or cloud OCR engine can be used:
Preset document types (invoices, orders, CMR etc.)
CustomDoc builder to define your own document types and data to be retrieved
Build in searchers based on: Context, Location, Dictionaires, Regex, External databases
Build your own searcher
Optional user verification helps IDR to learn and be more precise in the future
CustomDoc builder allows you to create analysing strategy for every type of document and test it in real-time on a set of documents.
The verification is easy and intuitive, IDR indicates where the user should focus on verifying data. The user can visually mark the location of the data if there is no other way of analysing the content.
Documents can be retrieved from email inboxes or remote locations automatically
Any on-premise or cloud OCR engine can be used:
IDR defines the type for each data on the document: ID, Licence plate, Date, Price, Medical terms etc.
Based on the type of the document IDR chooses which data should be masked and creates a suggestion map
The user verifies the suggestions . Any corrections are implemented into the algorithms automatically.
This step is optional and can be removed of full automation of the process
IDR prepares a suggestion map of the user to verify. Any unnecessary masks can be easily removed.
If there’s a missed word that should be anonymised user can easily drag the pointer across the word to instruct the system to include in the anonymisation process. IDR will then ask if the system should search the entire document for this word and adjust the masking.
As emails are unstructured the algorithms are not defined manually. Based on previously categorised emails IDR uses machine learning to define the algorithms automatically.
Every e-mail is being analysed by IDR and assigned to users responsible for the category. If the assignment is wrong, the user picks the correct category and IDR automatically updates the algorithms.
Allows you to use multiple sources of data input but also to use the data to provide your systems with extracted data or start workflows.
Any of your systems can be integrated directly with IDR, whether it’s a Ticketing solution, CRM, ERP or just an email inbox.
IDR is being constantly updated with new searchers and features that increase efficiency.
e-MSI IT Systems LLC
Level 36, Etihad Towers - Tower 3 Corniche Road, Abu Dhabi - UAE