Relativity6 Research Paper on Maximizing Insurance Customer Lifetime Value Accepted at International Conference on Machine Learning

CAMBRIDGE, Mass., Sept. 24, 2019 /PRNewswire/ -- Relativity6, Inc., a provider of artificial intelligence and machine learning predictive analytics, today announced that a paper authored by the company's data science team has been accepted for presentation at the 18^th IEEE International Conference on Machine Learning and Applications (ICMLA 2019).

Relativity6's paper entitled "Maximizing Customer Lifetime Value using Stacked Neural Networks: An Insurance Industry Application" proposes a two-stage neural network architecture: Stage-I uses a self-attention mechanism and collaborative metric learning to generate product recommendations, and Stage-II uses a neural network-based survival analysis to infer insurance product recommendations that maximize customer lifetime. The full paper may be accessed here.

"Our team is very excited to participate in ICMLA 2019," said Ismael Moreno, Chief Data Scientist of Relativity6. "Customer lifetime value is widely used as a key performance metric to evaluate retention strategies in insurance industry, but many retention products do not seek to maximize customer lifetime value by layering on additional patterns. We were keen to explore temporal patterns and excited about, not just total customer retention, but maximizing customer lifetime value on an individual basis." Moreno co-authored the paper with Gadiel Desirena, Jalil Desirena, Armando Diaz, and Daniel Garcia.

ICMLA 2019 will be held in Boca Raton, Florida, on December 16-19, 2019. Its goal is to bring together researchers and practitioners to present the latest achievements and innovations in the area of machine learning.

About Relativity6

Relativity6 is a big data analytics company that uses artificial intelligence and machine learning to help insurance companies retain their customers. For more information, please visit www.relativity6.com.

Media contact: Sam Breen
617-595-7055
[email protected] 

SOURCE Relativity6

Related Links

https://www.relativity6.com