Machine learning methods for automation of the processing for complex claims:
With our client we started the initial proof of concept, through the definition of the business case and the implementation of the machine learning algorithm, all the way to the deployment and the operationalisation of the new system.
State of the art NLP methods:
We used word2vec and sent2vec to define a similarity measure and learn a sparse space of more than 100’000 features. Additionally we worked together with end users to build an intuitive validation framework, thus building trust in the approach.
- Automate claims processing on top of rule engine based approach with Machine Learning Algorithm
- Cover 50% of previously uncovered claims with first developed algorithm
- D ONE co-created with data scientists and claim experts of client
- SDS2018 /Strata London