Evolution AI extracts data from financial statements with human-like accuracy.

Asset managers and banks require precise and up-to-date financial data to make investment decisions and report on performance.

In Partnership with

The Challenge

Extracting financial data requires reading and understanding financial statements such as balance sheets, cash flow and income statements. Automating this process requires an understanding of finance terminology and the ability to read complex financial tables.

The Solution

Our solution extracts data from financial documents with 99.5% accuracy. Evolution Transcribe reads complex financial tables by interpreting the visual cues on the page much like a human would. Our algorithms have the flexibility to understand when a field is named in different ways: the field ‘tangible assets’ can appear as 'machinery', 'plant', 'property', 'inventory', or many other variations, but it nevertheless needs to be extracted correctly.

Complex tables including nested headers and multi-page tables.

Currency and units tracked, even when not explicitly stated.

Comprehensive QA workflow and audit log, tracking each data point to its source within a document.

Post processing rules and accountancy calculations.

All data can be validated by a human annotator.

Features

Complex tables including nested headers and multi-page tables.

Currency and units tracked, even when not explicitly stated.

Comprehensive QA workflow and audit log, tracking each data point to its source within a document.

Post processing rules and accountancy calculations.

All data can be validated by a human annotator.

Functionality

We provide specific functionality to handle units and currency for each extracted monetary value along with sophisticated date handling. Complex multipage tables are also properly parsed and extracted.