In today’s highly competitive market, 88% of financial services institutions fear they are losing ground to innovators.1 Faced with growing volumes of data, they are turning to advanced analytics to help them regain their momentum by turning data into actionable insights.
Across the financial services industry, many leading institutions are already seeing the benefits of advanced analytics. From solutions that pool vast and disparate data sources in a single interconnected platform to those deploying artificial intelligence (AI) to better identify and control risks, advanced analytics is breaking new boundaries in efficiency, innovation, and risk mitigation.
From our work at Intel, here are two best-practice case studies of organizations that have transformed their customer experience and business outcomes with advanced analytics.
Generating new business insights with big data platforms
The rise of online and mobile banking has brought with it a huge increase in data for many financial organizations. One leading Spanish financial group in retail banking and insurance embraced this data deluge as an opportunity to innovate.
By implementing a big data solution based on Oracle Appliance*, Exalytics*, and Exadata*, and powered by the Intel® Xeon® processor E7 family, the bank gained the processing power it needed for real-time responses, while also taking in larger data volumes.2 It pulled all its data together into an interconnected platform, enabling it to use whatever data, in whichever environment or format the business needs. It can even build its own applications that draw on this rich data resource.
Having achieved a cohesive overview of its data, the bank applies advanced analytics to streamline operations and build predictive models that help deliver more personalized services and improve risk management procedures.
Mitigating risk with machine learning
A China-based international financial institution, which handles up to 20 billion payments every year, wanted to close the loopholes in traditional security risk models that could be exploited by criminals.
The bank had previously used a rules-based model to assess its risk profile, which evaluated pre-set user behavior characteristics and historic risk data. However, this became unsustainable as transaction volumes grew.
By moving to a neural-network model based on Apache Spark* computing clusters and powered by Intel® technology, the bank can now:
- Identify non-linear patterns in large data sets, which, combined with automatic updates, ensure new information can be added to improve risk prediction accuracy.3
- Use machine learning to update its system in tandem with new, evolving risks. This introduces more flexibility than a rules-based system, which takes a binary view as to whether a set of criteria have been met.4
- Swiftly analyze, aggregate, and correlate data through machine learning and an evaluation model to evaluate risks in real time.5
The bank has achieved up to 60% greater accuracy using this new solution, compared to its previous rules-based systems.6 This has also enabled its in-house team to generate insights that can help apply analytical tools and data science practices to its raw data.
Find out more about how advanced analytics can drive business insights across a range of industries by reading Intel’s new cross-sector storybook on the Business Impact of Advanced Analytics.
6 62% precision indicates the level of precision observed when scoring a set of unlabeled test data that was not used in the development of the model. For further details see: https://www.intel.com/content/www/us/en/financial-services-it/union-pay-case-study.html.