Description
Part 1 – Digitizing documents with powerful python libraries
So.. in the digital era you suddenly realise that until very recently
not all valuable information kept is/was machine friendly, there are
tons of word and pdf documents with reports, meetings minutes,
contracts and so-on waiting to be re-discovered. With python libraries
PdfMiner, PyPDF2 among others, we will unlock the power of text data
in pdf format.
Part 2 – Optimise risk assessment with advanced machine learning models Traditional risk assessment in insurance process can be based on creating a set of rules to classify cases that are likely to default. With powerful machine learning libraries such as SCIKIT-LEARN and XGBOOST, a probability of default model is trained using historical data to predict the future cases. In order to understand the decision making process, we can evaluate the output using Eli5 libraries and create visualisation tools for the end users.