Use of Machine Learning in tax administrations – IOTA workshop in Budapest

Ágnes Kővágó's picture
Published by:

55 participants from 31 IOTA member countries came to Budapest to discuss a very special and innovative subject: the use of Machine learning methods in tax administrations in IOTA “special interest workshop” on 2-4 October.


Our special interest workshops focus on a specific area related to the operation of tax administration. Machine learning turned to be a very appealing subject as it attracted the highest number of participants of this type of workshops.


Machine learning is a branch of Artificial Intelligence based on the idea that systems can learn from data, identify patterns and even make decisions. It is in initial stage in tax administrations but it has been used by private sector (banks, insurance companies, tax consulting firms, etc.) with success for years.


Participants of the workshop shared their experience in the use of Machine learning such as debt collection, VAT audits, communications, etc. Tax officials responsible for development of Machine learning projects, IT specialists, analysts, data scientists demonstrated advantages and possible disadvantages and risks, as well as best practices related to the introduction and use of such systems. Participants could listen to the presentations of country examples from Austria, Belgium, France, Germany, Norway, Sweden and the United Kingdom and discuss further those in group work sessions. Representatives of business sector (Vertex, PwC) also showed their solutions and how these can be used in tax administrations.


The workshop was useful both for experienced countries, both for countries just in the beginning of introducing Machine learning methods. Experienced countries could get inspiration from other examples, and inexperienced countries could choose from a wide “menu” of the use of several ways of Machine learning.


The workshop was organised with the contribution of experts from Norway, Sweden and the UK.


More information such as all presentations, a post-event report and more photos of the event is available HERE for IOTA members.