Prime Minister Speech Analysis

Project Info

Team Name


Angry Anguses


Team Members


Ben

Project Description


Using sentiment analysis and word groupings we can understand how sentiment of our prime ministers has changed, and how they're language might correlate with ongoing events in Australian History


#prime ministers #australia #leadership #sentiment analysis

Data Story


Prime Ministers throughout Australian history are there to lead us through difficult and prosperous times. This is usually done by being a source of positive action, or by going on the attack and trying to mitigate weaknesses by being a source of power

The Data

The data used is transcripts from the Prime Minister's speeches as well as press releases and interviews throughout their terms.

How it's used

This data was run through a sentiment analysis using monkey learns API to get an idea of whether or not a Prime Minister's talking points were generally negative, neutral or positive

The interpretation

Viewing this data on a timeline gives us an idea of how each Prime Minister went about their term, and can be linked to major events in Australian history (e.g. wars, recession, terrorism etc.). We could then correlate events to the sentiment, or get an idea of how the Prime Minister at the time went about their leadership. As well as this, comparing speeches with all available data including live interviews, we can see if there's a difference in how each Prime Minister went about scripted speeches versus his demeanor on more ad hoc and casual appearances


Evidence of Work

Video

Homepage

Team DataSets

PM Speeches

Description of Use Using to run through sentiment analysis and come up with a general sentiment for each PM

Data Set

Challenge Entries

The language of leadership

In times of crisis words can inspire and unite us, but they can also provoke division and conflict. How has the language of Australia’s leaders changed over time? How can we represent these changes in public discourse within a historical timeline?

Go to Challenge | 7 teams have entered this challenge.