Hansard

Citation Author(s):
Stephanie
Ng
Submitted by:
Stephanie Ng
Last updated:
Tue, 06/20/2023 - 18:32
DOI:
10.21227/53t5-fw17
Data Format:
*.csv (zip); *.json (zip); *.pickle (zip); *.npz (zip);
License:
Creative Commons Attribution
60 Views
Categories:
Machine Learning
Keywords:
Stance Classification
0
0 ratings - Please login to submit your rating.
ACCESS DATASET CITE SHARE/EMBED

Abstract 

This research studies the stance classification task of parliamentary debates with the aims to analyse how parliamentarians argue on different debate topic, what is their political stance, and the impact of homophily with respect to their party affiliation. A state-level Australian Hansard data is collected focusing on debates related to obesity and food marketing policies in Australia. It covers 6 states and 1 territory (NT is excluded) from the period 1/1/2000 to 1/1/ 2022. The raw data is stored in .jsonlist for each of the 6 states and 1 territory after data extraction from PDF and HTML files, resulting in the length of  645,594. 

 

Instructions: 

The data contains the following attributes: data, fileIdx, idx, keywords, filename, state, title, house, utterance, speaker, text, source, party.

Dataset Files

LOGIN TO ACCESS DATASET FILES

 

QUESTIONS?