Machine learning in practice
Machine learning will inevitably bring about changes. Our industry experts provide their assessments of how various sectors and business models can profit from it.
Insurance & Health
Machine learning is already used in the (health) insurance industry for many use cases – at varying stages of maturity. At an early stage for instance, is a self-learning algorithm that makes decisions about liability questions based on claims data records and submitted photos and detects cases of fraud. Self-learning chatbots are also increasingly used for generating offers, even if they are currently still rather formbased. Their use in complex consultations tends to be moderate so far as the data to properly train the bots is lacking. However, improvement is in sight. Sentiment analysis already works rather well in English and holds great potential for quickly classifying matters.
Among banks, the onboarding process can be simplified through machine learning. It is primarily used in risk recognition and scoring. The validation of scans and photos is also helpful, which is automatically and accurately accomplished by machines.
In the retail industry, machine learning is first and foremost used for automatic classification. Particularly frequently asked questions or social media posts can be quickly allocated to the right topics and agents. It also serves as an early warning system – keywords: food scandals or consumer protection campaigns. I also see major potential when it comes to “next best action” or “next best offer”.
In the pharmaceutical industry, and especially when it comes to clinical trials, machine learning is already partially being used to predict trial outcomes based on existing clinical data. In the future, it could potentially support and optimize the risk-based monitoring of trial centers.