Eurovision 2019 Statistics — NLP Analytic Tool Reveals that the Audience Mood on Social Media Reflects the Public Vote

Image for post

That’s it, the 2019 Eurovision Song Contest is over.

Congratulation to Duncan Laurence for winning the 2019 Eurovision trophy with his song Arcade. And good “morning after” to the Netherlands, keep enjoying the glee of victory 😊

At Communit360, the voting process was a nail-biting experience.

Reading Social Media Engagement

Unlike most viewers, Communit360 team was not watching just to enjoy the show. The social reputation and social media engagement monitoring and analytics AI software was hard at work analyzing the audience mood based on the content of their public interactions on social media designed. We had partnered with KannTV to follow the popularity of the Eurovision songs in real time on social media. KannTV, an Israeli digital TV channel was broadcasting from their studio, providing ongoing commentary throughout the competition, relied on Communit360 to gauge the mood of the audience and the popularity of the songs, and our proprietary Natural Language Programming based AI solution raised to the challenge.

Predicting Who Will Win

So, on the big night, as the points granted by the professional judges were being announced, the mood at Communit360 was one of growing doom. Our system had Norway as the winner, and it barely raked 57 points from the professional judges, trailing very low on the scoreboard.

When Norway received a whopping 291 points from the public vote, the relief at Communit360 HQ was palpable. Public opinion was backing our findings. It was the judges who had opinions differing from that of the audience, and our algorithm was not built to second guess Eurovision judges, only to analyze the trends derived from social media activity.

Image for post
Image for post
First Eurovision 2019 Semi-Finals Results and Communit360 Estimates

Still, we had to wait until all the public votes were counted to get confirmation that our system was indeed extremely accurate in evaluating the audience’s mood from data collected on social media and analyzed by our AI.

Image for post
Image for post
Second Eurovision 2019 Semi-Finals Results and Communit360 Estimates

As Communit360 had predicted, Norway would have won the Eurovision had the results been based exclusively on the audience’s vote.

This confirmed our team’s hard work in building the predictive algorithms, as, during the first and second semi-finals, Communit360 had 80% correct prediction rate in identifying the contestants who would proceed to the finals.

Image for post
Image for post
Eurovision 2019 Final — Winner of the popular vote and Number 1 in Communit360 Social Media Stats.

Communit360 is not at all designed to be a predictive tool, but rather to measure brand’s reputation, identify terms that are commonly used to refer to the brand or compare different brands, so we were elated by this unexpected predictive success.

Spotting Unexpected Trends

Communit360 is more geared to identify unexpected trends. For example, in Norway’s case, amongst the predictable mentions of the song and the contest, we noticed numerous mentions of “bald guy”, both positive, neutral and negative

So, it seems that Fred Buljo’s baldness was a feature gathering considerable notice. Trying to understand what is behind the interest for the baldness of one of the band members, we quickly glanced at all the times it is referred to, either positively, negatively or neutrally.

Looking at these tweets, we immediately see that the “feature” bald guy is used as a way to identify a specific singer in Keiino, an easy way to refer to a specific singer in the band.

This kind of data is of high value for brands and trend discovery, as such insight about an unexpectedly salient feature could be leveraged for marketing purposes. For the band choreographer, stylist or other show managers this is valuable information about where the audience’s attention is directed. It would help them to design the set or the costume to maximize or minimize the impact of the baldness of the “bald guy” depending on the effect they want to achieve.

Accurate information is at the root of management, and accessing such fine-grained data in realtime is of immense value to adjust behavior, design, course of action, feature development, marketing campaigns and more.

There are many other features based on Communit360 AI, such as comparing between countries or brands, getting data about the exact number of people commenting on any topic and their combined reach, the most engaged tweets and much more, however, for now, it is time to recover from the exciting night following the Eurovision contest.

Reveling in the Popularity of Communit360’s Live Analytics

Another highlight of the Eurovision for our team was the popularity of our live statistics on the Twittersphere. Sooo many chose Communit360 public real-time analytic dashboard to share their excitement about the songs they supported!

So, again, congratulations to the Netherlands for their victory, and heartfelt congratulations to Norway for their victory with the public vote!

And to Communit360 for its accuracy ;-)

Communit360 is a social reputation, engagement, and risk management tool for brand reputation supervision and management. It is powered by Natural Language Programming (NLP) responsible for analyzing and mapping social media posts and their semantic and contextual relations. It is designed to provide real-time insights regarding social trends while maintaining scalability and flexibility to suit each end-user’s market and needs.
Follow Communit360 on Twitter and Facebook


Social Reputation, Engagement & Risk Management For Brands

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch

Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore

Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store