Twitter Reacts to Budget 2017 Announcement

20 January 2021
Twitter Reacts to Budget 2017 Announcement

Posted: 11/10/16

It has been widely talked about in recent weeks but as budget 2017 was announced Twitter activity steadily increased as the people reacted to Budget 2017. Using the main hashtags (#Budget2017 and #Budget17) along with keywords and trending topics as they emerged, the ADAPT Centre for Digital Content Technology tracked the volume, sentiment and frequency of budget related tweets.  Here’s how it all went down…

In the run up to this year’s Budget announcement Twitter already had over 7000 unique tweets linked to the Irish Budget 2017.  That number increased dramatically today as over 6,000 tweets from Ireland and overseas were harvested about Budget 2017, involving thousands of unique Twitter accounts.

Childcare and the sugar tax dominated the conversation on Twitter.

Childcare is the top budget issue being discussed on Twitter.

The ADAPT Centre has been monitoring the conversation and data harvested also shows that health related to the sugar tax is the second most discussed topic, followed by tobacco and income tax.

The graph produced by the ADAPT Centre represents the breakdown of budget topics mentioned in conversations and a word cloud highlights the frequency of the main budget related keywords.

The frequency of twitter topics only gives us one part of the story.  In order to understand how the public felt about the Budget announcement the ADAPT Centre also looked at the sentiment in those tweets. The overall sentiment of tweets became more polarized with topics such as the pension levy and USC trending as negative once the Budget announcement began. The topics which prompted a positive response were education, childcare and social welfare.

The team at ADAPT worked to gauge people’s opinion, attitudes and reactions towards Budget 2017 by performing sentiment analysis using machine learning techniques on annotated twitter data sets developed and explored at the ADAPT Centre.  Tweets were ‘cleaned’ to prevent noise in the system and then classified for sentiment using opinion lexicons that associate sentiment polarity for words. Sentiment is classified as positive, negative or neutral using computer science techniques and opinion mining research.

Share this article: