Dr. Reuven Shnaps is the Chief Analytics Officer at Earnix, a leading provider of mission-critical systems for global insurers and banks.
What initially attracted you to data science and AI?
I have always had a fascination with math, data, and their potential to solve business challenges. Throughout my academic studies and career, I have sought out opportunities to learn about statistics, economics, and how to apply these fields to understand consumer behavior. I admire data scientists, modern statisticians and econometricians, who have the unique ability to analyze vast amounts of data and address real-world business problems. I have dedicated my career to data science and combining traditional statistical methodologies, emerging technologies, new machine learning (ML) algorithms, and the latest artificial intelligence (AI) applications to create business solutions that deliver long-term value for our customers.
What does Earnix do?
Earnix is a global provider of software solutions that empower insurers and banks to provide faster, smarter and safer rates, prices and personalized product offerings to consumers. Our system is powered by AI & ML , and includes a wide array of analytical modeling tools, applications and advanced algorithms. Recently, Earnix made the list of “Insurtechs to watch in 2021” in the U.S. and was recognized as a market leader in predictive analytics by CB Insights, an analysis and research company in the technology sector. Earnix leverages innovative technology to help insurers and banks meet consumer needs in real-time.
Earnix recently wrote an article for us on the importance of Explainability in AI. How important do you believe this Explainability in AI is?
Explainability is a trending topic in AI and data analytics. It affects companies across industries, whether they use AI or not.
Most companies experience a trade-off between their level of control over and the effectiveness of AI. For businesses to trust and adopt the use of “black box” AI, there needs to be a mechanism that provides experts and stakeholders the ability to interpret complex AI decision-making processes and ensure adherence to regulatory demands. Consumers alike can benefit from it by being able to understand, and potentially navigate key drivers behind pricing, credit or underwriting decisions. Explainability is the bridge that makes complicated AI more understandable and transparent. With the ability to translate advanced ML models, analytics professionals do not have to sacrifice more advanced algorithms because of a lack of understanding. Explainability minimizes the trade-off between control and value while maximizing the benefits of AI.
Could you discuss some of the machine learning technologies that are used at Earnix?
Earnix enables insurers and banks to deploy the latest AI in an enterprise-wide, scalable, cloud-based rating engine. Our fully automated rating engine operationalizes models into production to offer real-time, highly personalized insurance and banking products that are aligned with consumer needs. Our open platform capabilities allow integration with a broad range of ML models and platforms such as H20, Data Robot and Python. Earnix developed a unique hybrid model to effectively leverage the combined power of traditional statistical modeling techniques and tree-based ML models for pricing decisions
We are always on a mission to discover new ways to leverage machine learning capabilities that insurers and banks can use to enhance their businesses in a way that directly benefits the consumer. We are currently exploring applications of smart monitoring and online learning, Contextual Multi-Arm Bandits (MAB) for designing efficient package tests and tailoring personalized product/package offerings to customers, as well as deep learning algorithms for analyzing high-frequency data. This way, we help insurers and banks offer faster, smarter and safer products, rates and prices.
How is AI modernizing banks and insurance companies?
Traditional insurers and banks who rely on legacy systems and siloed, manual processes are being challenged by AI-driven companies. AI has created an environment where insurance companies and banks have to innovate faster, process a vast amount of data across multiple sources & types, automate manual operational processes, streamline and enhance customer journey and engagement, offer more personalized products and rates, and achieve immediate time to market value. It has redefined how insurers and banks operate, and incorporating AI is integral to staying competitive and giving consumers what they want.
For instance, consumers are increasingly turning to usage-based and behavior-based insurance (UBI/BBI) options because they want their auto insurance rates to be determined by their usage and driving behavior, rather than by their demographics alone. With our automation, ML and analytical capabilities, insurers can now operationalize the vast telematics data, achieve fast deployment of offers, re-assess consumer risk, as well as adjust pricing and product offerings accordingly.
Is there anything else that you would like to share about Earnix?
There is a lot of research about how the use of AI improves business processes and outcomes, but there is no playbook on how to combine the power of AI and its potential into business operations to deliver value immediately and propel business growth. Many financial services organizations struggle with connecting AI to their operations. Earnix’s end-to-end technology addresses this acute need and pain point. I am proud to be working for a company that uses analytics to find new ways to foster a profound trust-oriented relationship between a consumer and the banking and insurance industries.
Thank you for the great responses, readers who wish to learn more should visit Earnix.
- Pexip Collaborating with NVIDIA to Create Immersive Video Meeting Experiences
- Sean Byrnes, Co-founder and CEO at Outlier – Interview Series
- Microsoft Buys Nuance For $19.7 billion
- Deep Neural Network Can Screen for Skin Disease on Laptop
- AI Systems Might Prefer Human Language Instead of Numerical Data