Multiplica talks for OpenFinance on how Machine Learning is revolutionizing the world of finance in terms of customer interaction and knowledge, as well as risk assessment.
Once again, Multiplica gave a webinar for OpenFinance’s audience, in which Jordi Navarro, Head of Customer Intelligence, presented an interesting talk on predictive models applied to the financial industry.
According to Gartner, 70% of technology projects in the financial services industry will be employing artificial intelligence by 2024; the use of machine learning will be an increasingly common tool that will optimize processes, allowing to generate better user experiences, and therefore, better business results.
Machine learning in the financial industry
“Machine learning is a scientific discipline in the field of artificial intelligence that creates systems that learn automatically.”
According to what we discussed with Jordi, algorithms are classified according to their complexity and the value contribution they have for the business; starting with descriptive analytics, which allows us to understand what has happened before, to later enter into a deeper analysis in which machine learning will enable us to understand in depth three types of analytics:
- Diagnostic: It indicates the causes and their weight when we see the consequences and their effects.
- Predictive: Indicates probabilities that something may happen.
- Prescriptive: It tells us what recommendations can be made based on the user’s characteristics and preferences.
All of the above will give us the guideline to reach a final state of deep learning in which, through cognitive analytics, we will be able to identify the true value that a situation or customer has for the business.
Use cases in the financial sector
During the talk, the expert pointed out some use cases where we can make use of machine learning in the financial industry:
- Personalization: Everything related to improving the customer experience, personalized offers and prices, and product recommendations.
- Scoring: Refers to giving a value to understand credit risk issues, probability of default and how much can be foreseen in monetary terms that a customer will represent for the business (Customer LifeTime Value).
- Fraud detection: Helps detect activities such as money laundering, financing of illicit activities and market abuse surveillance.
Conclusions
According to what was discussed during the talk, the projects that achieve the best results are those that focus on automating repetitive micro-decisions that until now should be undertaken by one person, clearly contribute to revenue generation and demonstrate a clear ROI on Data & Analytics initiatives.