Fraud management: AI at the service of people01 Décembre 2021
The COVID-19 pandemic and the resulting recession have increased the level of risk, which is expected to lead to a dramatic acceleration in fraud cases in the coming months according to a recent study led by an international institute for training and professionalisation concerning the fight against fraud.
Since 2015, BlueLink has already detected and managed cases of fraud linked to the loyalty programme of a large hotel group, whereas no tool on the market has offered a solution tailored to the specific problems of loyalty. Based on the twin observation that the number of acts of fraud in loyalty programmes was steadily rising and that the number of transactions to be analysed had doubled, we offered support based on artificial intelligence combined with the expertise of the teams in order to optimise the work done by our experts and so limit the risks. By developing and testing a combination of several supervised and unsupervised algorithms, we were able to detect a greater number of anomalies in transactions and uncover new patterns of fraud. To visualise and process the different cases, we developed a prototype case-management tool according to the business rules defined by our experts. The use of AI systems combining high-performance algorithms, high computing power, machine learning capabilities – deep learning – makes it possible to automate those frequently repetitive and sometimes thankless actions that were previously carried out by employees. Able to process a very large number of cases, or to perform certain tasks automatically, in a way that is much faster and more reliable than humans, artificial intelligence now allows them to focus on their true strengths and on missions with higher added value. This virtuous combination reinforces our operational excellence.
In the detection and management of fraud, managing human and artificial intelligence together and in a synergistic manner provides an opportunity to counter the increase in risks but also to empower teams and restore meaning to work. In short, it is all about artificial intelligence reading the lines and humans reading between the lines.