Legal documents parser
The business goal
Our client is a fintech company from London. They specialize in providing their customers with information regarding complex financial products like RMBS, ABS, CLO. For these purposes they built a software named CF engine.
The business goal of the project was to create a platform (‘CF Engine’) to model complex financial products (RMBS, ABS, CLO, etc.). Our part was the development of a Machine Learning module with the features that allow users to review the related legal documents based on the information from the model.
The developed model needs to check if a specific document corresponds to one of the created models. For example, if the processed document is a mortgage, the process would be the following:
– parses the mortgage document (from PDF, Word, plain text format);
– checks if the document contains all required information (all parties are specified and described correctly, property is described, interest rate is specified, all information required by law is provided and so on);
– if the document fits the model then the system extracts the important information (parties, property description, interest rates and so on) and provides it as a summary for the user to review.
The system supports different formats of input documents and different types of documents, such as mortgages, car loans, commercial loans and so on. Also the system supports different countries of operating, i.e. different structures of documents for each country and different languages.
We applied Natural Language Processing and Machine Learning respectively.
The result of the project was a successful implementation of an ability to review the legal documents and automatically highlighting the most important information.