Deep learning is a key component of artificial intelligence, with particularly important implications for insurers

There is a lot of excitement in the insurance industry about the potential impact of artificial intelligence on areas including the customer experience, underwriting, distribution and claims processing.  One of the problems with separating the hype from the reality is that “artificial intelligence” is a broad term that encompasses many different types of technologies.

Insurers talking about artificial intelligence, for example, lump machine learning and deep learning together, but deep learning is a type of machine learning that uses multi-layered neural networks. This means, essentially, that machines can learn new things as they operate, such as how to recognize images.

Deep learning is a difficult undertaking.  According to the Nikkei Asian Review, Google, one of the pioneers in deep learning, had to input 10 million cat images from the Internet before its artificial neural network could automatically detect cats in pictures.

With huge volumes of structured (internal) and unstructured (external) data coming in from a variety of sources – including newly accessible third-party databases and the Internet of Things — insurers can use deep learning not only to identify images but to recognize patterns. This can have an enormous impact in claims as deep learning applications help insurers, not only to assess claims at high speed, but to identify anomalies indicating potential claims fraud.

At Accenture, we have already incorporated machine learning into traditional claims dashboards designed to model the expertise of individual agents, freeing them up to spend time on the those claims where human expertise is needed.   Now, we are working within an ecosystem of partners to explore how deep learning can further transform claims processing along with other core insurance functions such as underwriting and customer service.  We are just at the beginning of a very exciting journey for insurers, and will update you periodically on new developments on this front.

Submit a Comment

Your email address will not be published. Required fields are marked *