"Data journalism solution for analysing all of the UK House of Commons debates since 1988. Big data semantics similarity search with Machine Learning APIs"

The official brief changed and the machine learning aspect of the project has been take out. We are also including all the debates that have been recorded since 1802 rather than 1988 onwards. Our API will be accessible by the public.

Fundamentally, our system is to be made in the form of a web application and it would be hosted on the web, which would be accessed by users through a web page. Our system has to somehow extract text from the Hansard database and display search results when queried. We also need to provide the public with API, which could be in the form of RESTful or Java API.
On top of the fundamental requirements, the system should also provide at least a few important features to provide better usability to the users. The users should be able to search by topic, either by choosing from a given list or by filling a form field. After a search has been performed, the system should also suggest texts that are similar to the one currently presented. The users should also be able to choose the size of the texts being displayed. This can range from individual sentences to whole debates.