Dynamic Data Capture
Kefron AP automates invoice data extraction using Dynamic Data Capture technology.
Our Machine Learning enabled extraction forms part of a 3-step process to ensure the most accurate and efficient data capture as possible.
How it works
The system collects different parameters that are used to classify each document.
The collected data is prepared for the next steps
Choosing a Model
The documents are classified and processed through relevant extraction models depending on the relevant information
Training the Model
The system is trained to recognise different document formats
Evaluating the Model
The model is tested to suit different scenarios
The parameters are customised to improve the output results
The model gains independence from human interference and draws its own conclusion on the basis of its data sets and training
Our machine learning data extraction model combines the knowledge of millions of captured historical invoice data with continuous learning of how your business processes invoices.
The result of which is a reduction in exceptions, and fast end-to-end invoice processing capabilities.
Fewer exceptions mean faster invoice approval times.
While exceptions cannot be completely eradicated, Kefron AP is unique in offering an inhouse data engineering specialist team that will assist with reducing exceptions.
With access to millions of previously scanned invoices, our machine learning model is continuously improving its ability to extract data without manual intervention.
DID YOU KNOW?
Our innovative machine learning technology replaces the need to build invoice templates.
- Non-Template Extraction
- Managed Exceptions Service
- Flexible Extraction Rules
E-tags are machine learning data tags created to tell the extraction engines what to extract and what to ignore
Our machine learning technology replaces the need to build invoice templates
Our In-House data extraction team manages any exceptions that the technology could not pick up.
Extraction is further enhanced by the use of Intelligent Data Tagging to extract data from the data layer of machine readable pdfs
Including multiple data fields like reference numbers, and site locations
Want to learn more?
Talk to an AP Specialist
- Calculate the true cost of your current AP process
- Identify bottlenecks
- Assess if your AP process is suitable for automation
- Calculate the real business case for AP Automation