Digital transformation in action – a financial services case study

Digital transformation in action – 2 financial services case studies

November 30, 2021

The true is not all digital transformations are the same. This makes it particularly difficult for financial institutions to identify the AI-based solution that fits best their business. In a previous article, we listed 3 key points which, in our experience, make a difference to the digital transformation of financial services: 1. valuing domain expertise, 2. staying human-centric, 3. respecting privacy regulations. These are the principles we follow every time we help a business approach and integrate eXplainable AI into its decision-making process:

1. Rulex platform seamlessly integrates data-driven insight with human expertise expressed through heuristic rules, to reach optimum results.

2. It supports users through its clear box technology, providing human understandable reasons for each prediction, so users are always fully in control of the process.

3. It employs eXplainable AI to produce predictive models in the form of self-explanatory logical if-then rules (see here for more details), which can be fully understood and explained by business users, making it GDPR compliant by design.

Digital transformation in action: ad-hoc financial services solutions 

Rulex offers a suite of advanced solutions for the financial sector, from compliance and data governance to lending and credit risk, and marketing & commerce. In this article, we see Rulex technology in action, considering two case studies.

How we manage false positives in fraud detention

Detecting fraud is statistically complex, due to the limited samples of fraud, which correspond to approximately 1% overall. Consequently, the risk of creating false positives is very high.

Rulex technology found a solution to this problem and was key to building a successful anti-fraud solution together with official partner GFT, for the banking and insurance sector. How does this solution work?

Rulex’s innovative LLM algorithm (characterized by a high level of explainability) is able to identify patterns from the limited sample data of actual fraud and build reliable and accurate predictive models. The output is a fraud shortlist, which can subsequently be tailored to the anti-fraud team needs, integrating their domain business knowledge. Finally, a score is added to each record in the shortlist, which ranks the likelihood of fraud.

How to evaluate the best course of action for NPLs?

The logical approach of many credit institutes is to pursue legal actions to recover funds from non-performing loans. However, this is an expensive and time-consuming strategy, the highest cost derives from initiating and pursuing legal actions to force repayment, also considering that many litigated loans are lost to default nonetheless.

Rulex’s NPL solution is able to distinguish between cases where it is worth starting a legal action to retrieve funds, and cases where a legal action would simply represent a further loss of money.

For those credits where collection is recommended, Rulex’s NPL solution also suggests the most efficient course of action in terms of success and cost of resources. It provides a list of operational actions that guide the operator in the credit recovery decision process, thus providing not only tactical, but also strategical information to empower the decision process.

The next step: how Rulex can help your business

The use of eXplainable artificial intelligence can help organizations combat fraud, estimate risks, and recover funds. But what can Rulex technology do for your business? Visit our dedicated Financial Services page to explore all our case-studies, find more information regarding our solutions, and get in touch with our experts.





Senior Digital Learning Specialist • Training and Documentation

Related Posts