Axis AI is Software as a Service

Clients essentially send us their documents electronically, Axis-AI and our quality control team process, extract, analyze, validate and return the essential data back to the client – in a matter of hours. This process may consist of just a few dozen files or even hundreds of thousands of pages of information.

The result is faster, more accurate data extraction than the typical results found from manual human data entry, and at a fraction of the cost.

“Show us your Data, We’ll do the Rest.”

Here’s How It’s Done.

The Documents Are Processed Through a
Multi-Step Automated Data Extraction Workflow.

Process Flow

There Are Four Main Steps To Process The Documents

Capture the Documents – Import Images into the System

Classify/Extract the Data – OCR and Natural Language Processing

Validate the Data Quality – Automated and Human Quality Control

Deliver Indexes to Client – Release Data to Client

Capture Steps
Document Conversion

Documents Are First Sent To Axis For Processing.

This is typically performed by electronically transferring images, in a PDF or TIFF format, of the documents via FTP. Document are typically scanned at the clients offices. Electronic files like emails, Office documents, fax images can also be managed.

At this stage documents are converted and processed with highly advanced Optical Character Recognition (OCR) algorithms to a standard format designed so the software can read and analyze the data.

The Files Are Next Identified And Classified As Unique Types Of Documents.

A batch of files many include one unique document type or a complete document package, like a loan including a Deed, Tax Forms, Appraisal, etc.

In many cases documents are delivered to Axis as a multipage/multi-document type file, a loan package is a great example. If this is the case Axis AI automatically reviews the original document, discovers the different document types within the file and separates and identifies the file into the various document types. This could be a simple as a few pages or even hundreds of pages and document types.

These identified document types can result in a single or multi-page documents. No separator sheets are required by the client to identify the document types the Axis AI software automatically performs document Classification and Separation.

.

AI Extraction Engine
Post Document Processing

Data and Index Extraction and Processing.

Once the documents have been identified the software distinguishes what rules to apply to abstract the specific data and indexes requested by the client.

In some cases a single index many be the solitary required field, other documents many have dozens or even exceed a hundred essential indexes. There is no limit to the number and length of the indexes configured for extraction processing.

Post Extraction Business Rules Can Be Applied To Clean Up Or Convert The Data.

For example dates maybe converted to a common format for example 1/1/15 converted to January 1st, 2015. In other cases text maybe parsed from a sentence to create a new index field, for example a “Legal Description” field maybe parsed into fields including Lot Number, Block ID, Plat ID and Subdivision.

Data Extraction
Data Validation

Axis AI Automatically Runs Validation Rules Against The Data For Accuracy.

Business rules help validate the extracted data. Validation can include look-ups to other database systems, for example you could validate a zip code against a city name database to confirm the quality of the data.

There are also cases of a potential OCR error, our validation/QA staff work within our validation module where they are triggered to review even the smallest questionable error like the number 8 being mistaken for the letter B. Axis AI’s configuration allows tuning to allow different levels of validation and quality analysis, from field all the way down to character level correction.

Retraining and New Document Assessments.

At any anytime during the capture process if the software finds new document types without extraction configurations or we find the configuration requires modification or tuning, the operator can retrain Axis AI on these changes and reprocess the document.

Documents are not equal
Release Data to LOB Systems

Delivery of Data and Indexes to the Client.

Upon completion of the data extraction and validation process the obtained data is then formatted and configured for delivery back to the client.

Various methods for deliver can be configured with processes as simple as a FTP file upload or a more complex direct communication between another system using XML or Web Services.

Finally A Solution For Your Most Challenging Data Capture Initiatives.

Axis AI processes documents faster and more accurately than typical human manual data entry. It performs this processing on some of the most complex documents ever created. Axis AI does it faster, cheaper and more accurately than traditional methods therefore increasing productivity while still reducing operating costs.

Product Overview
Business Benefits