As part of the Research and Innovation (R&I) FUSION Programme, the Malta Council for Science and Technology (MCST) will be funding the Automated Document Analysis and Classification for Enhanced Enterprise Efficiency (ADACE3) project – a joint effort between the University of Malta and PTL Limited.
Although we live in a digital age, paperwork and tedious manual interventions remain ubiquitous in business, having a knock-on effect on efficiency and ultimately revenue and profits. A variety of business documents exist, and are used by organisations to track and record internal or external transactions, such as quotations, invoices, receipts and purchase orders.
Even though e-invoicing has been around for quite a while, most companies, especially the smaller ones still make use of either paper, email or pdf invoices, where the semantic information is lost, from the point of view of the machine, limiting further useful processing. It is therefore desirable to have systems that extract the semantic information from these documents. In this respect, off-the-shelf commercial products are either characterised by limited functionality, increasing the incidence of human intervention, or require periodic higher order skills to re-train and maintain a working system.
The project plans to leverage state of the art artificial intelligence algorithms, namely self-supervised learning, joint vision-language models and never-ending learning techniques, to develop a system that minimises human intervention and is potentially more attractive to the small company market, in terms of cost and ease of use.
PTL will develop a minimum viable product software package that integrates the AI models developed by the University of Malta with a user-centric designed interface, ready for commercial exploitation. To guarantee the usefulness of the product, external stakeholders will be engaged during the design and user acceptance test periods.
More information can be found here.