Documents (placeholder)

This is a feature/design note.

This page/view will focus on documents tabled in parliament such as reports, petitions and the text of bills. These are ancillary to the main debates but since debates are often about these documents, clearly it's desirable to view them in context.

In another respect, access to parliamentary documents is a stand-alone use case that would be a significant improvement on the accessibility of these documents to the wider public.

Parliamentary documents are invariably PDF files. The Hansard/document feature requires a data pipeline to extract text and metadata from PDF files andcalculate word embeddings for chunks of text to support advanced search and retrieval features.

We also anticipate creating an AI-powered document query feature to allows users to asks questions about documents in plain language, e.g. to ask for a summary of a document, or to ask a specific question relating to a document.

Australian Research Data Commons
Australian Digital Observatory
Queensland University of Technology