Proprietary research for smarter rating, pricing & personalization
Our Analytics team is continuously researching and exploring state-of-the-art tools, solutions, methodologies & algorithms to embed the latest science into the Earnix products and solutions.
Current areas of focus
The urgent need for explaining black-box models has led to the emergence of a new research field: Explainable Artificial Intelligence. Explainable AI is one of the hottest topics in the field of Machine Learning these days. Researchers in this field are developing tools that enable us to peek inside the black box and unravel the magic, at least partly. The Earnix Analytics team is exploring various explainability tools, such as Permutation Feature Importance and Shap, and assess their suitability for usage in the financial services world.
Online learning will enable insurers and banks to continuously and automatically adjust key operational models and deploy them, almost instantaneously, once the business is ready.
Contextual Multi-Arm Bandits (MAB)
The MAB methodology is a smart extension of the A/B testing framework. We are currently exploring its applicability for designing efficient package tests and tailoring personalized product/package offerings to customers. The MAB framework is an outstanding tool when it comes to the launch of new products or when wanting to quickly assess & respond to changes in consumer behavior.
Early “detection/warning” systems
Earnix Analytics is exploring the development of propensity models utilizing scores or features which are continuously updated through streaming data such as customer transaction and digital activity data. Of particular focus is Long Short Term Memory (LSTM) networks capable of learning long-term dependencies. The advantage of this methodology upon more traditional ML models is that it can be adapted to predicting different event types without the need to engage in a lengthy feature engineering process.
Business Benefits Of Better Science
Earnix technology provides the ability to create and reuse processes in an automated manner – from strategy creation to deployment. The result is time and cost savings that other pricing and product personalization software doesn’t provide.
With accurate insight coming from world-class data science, analytical modeling, and integrated machine learning, the most complex business challenges can be addressed.
The Earnix technology enables shorter time-to-market. Earnix real-time capabilities enable these insights to be delivered to customer interaction channels in milliseconds.
Earnix runs on Amazon Web Services, which provides the security, scalability, availability, and performance that financial institutions require. REST API functionality enables Earnix offers to be delivered to online channels with real-time speed.
Earnix is an end-to-end system that operates as one solution, and not a conglomeration of products. Structuring technology in this manner, versus a series of integrations, allows for total control over the pricing, personalization, feedback, monitoring, and governance processes.
Wait, there is more...
The Analytics-driven Bank
Becoming fully analytics-driven is the baseline survival criteria for banks. But is that an achievable goal for all financial institutions?Read Now
Webinar: The Fully Automated Ratemaking Process
The insurance industry has been heavily investing in process innovations. But what about the process of ratemaking; what is our next innovation there?Watch Now
Analytics – Banking’s New Best Friend
Normal was and is a luxury when it comes to credit and pricing decisions. A rigid organization is a toothless tiger, ridden of its main strength.Read Now