On the use of Deep Learning for Automation of Accounting

Artificial Intelligence takes accounting offices Artificial intelligence has in recent years had a major impact on several industries and areas. We have seen examples of intelligent systems that are better suited to diagnosing patients than doctors, are safer drivers, and are better chess players than humans. By using methods based on the same principles as our own brain, artificial intelligence has been able to analyze and understand complicated relationships in complicated data. The financial sector, with its strong dependence on large amounts of numbers and data, has undergone a digitalization change, including the use of electronic trading formats. Artificial intelligence is ready to lead to a paradigm shift in the accounting offices. This project aims to further develop methods and algorithms to automate and improve several aspects of the financial and accounting sectors. The project will be based on a type of artificial intelligence algorithm which has received much attention the only years, Deep Learning. The project will further develop variations of deep learning, especially with regard to increased insight and deep understanding of accounting data. Central to the project is the development of algorithms for classification problems, such as understanding trends in the accounting data and predicting problems as well as notifying about the next best action. The algorithm will, based on historical data, understand when a business is likely to get a revenue loss and propose actions that improve and, if possible, avoid the situation. Furthermore, it will also develop methods to identify and predict deviations and trends in large amounts of data where conventional methods do not extend. The research will result in a system of automatic deep understanding of the field, which will provide an intelligent form of customer support.

Deep neural networks exploit composition hierarchies of data, which makes it particularly suiting for finding trends in complex accounting records. This project will explore the potential of deep neural networks for automating big accounting data, particularly: (1) Generative adversarial networks to generate accounting data including invoices. (2) Image segmentation for invoices for quality assurance. (3) Pattern recognition for invoice classification. The candidate will specifically work with automated data generation to increase the accuracy of pattern recognition tasks. The candidate is expected to explore existing techniques, specifically in combinations of UBL and EHF, and develop and verify new variants of deep neural networks with real customer data. The candidate will develop a fully functional prototype. Goal 1: Automatic generation of accounting data for text using deep generative adversarial network. Goal 2: Automatic detection of crucial information from invoices including account number, recipient, and so on using deep convolutional neural networks. Goal 3: Automatic classification of invoices using a combination for real data and generated data (from goal 1).