What People Have Felt and Thought About The 2021 UN Climate Change Conference (COP26)

April 14, 2020
Short Studies
Andy Clarke
Andy Clarke

We collected every tweet published about the 2021 UN Climate Change Conference (COP26) - 31.10.2021 - 12.11.2021 - to study how people have engaged with events during the summit. Using topic analysis and emotion detection, our project dives into people's visceral reactions to agreements on deforestation, commitments between China and the USA and the appearance of Barack Obama.

On top of this, we're loosely looking into whether we can use AI - social media monitoring + Advanced NLP - to monitor the feelings of society in short bursts.

Explore The ProjectWe collected every tweet published about the 2021 UN Climate Change Conference (COP26) to study how people have engaged with events during the summit.

November 12th 2020. The final day of the COP26 UN Climate Summit in Glasgow. For two weeks, world leaders, activists and international delegates have been meeting in Scotland's biggest city. The agenda? Climate change.

But people's feelings about the health of the earth - and what to do about it - is as mixed as the weather in a Scottish summer. For the optimistic amongst us, COP26 offered the promise of progress. For despondents, it represented nothing more than a social gathering for the world's elite. The targets have been ambitious and leaders have a lot to lose. The question we're interested in; Has the latest UN Climate Conference made us more or less hopeful about the future of our planet?

The Method

Collecting every tweet about the conference since the 31st October with Tractor, we set out to find the main topics of conversation surrounding COP26 and analyze how people have thought and felt about them.

The main topics in our data ranged from discussions about the meat industry and methane emissions to debates on the value of the deforestation pledges made by world leaders.

#COP26 Emotion AnalysisUsing topic analysis and emotion detection, our project dives into people's visceral reactions to agreements on deforestation, commitments between China and the USA and the appearance of Barack Obama.

An Emotional Overview

Before we take a closer look at specific emotions associated with conversation topics, it is worth mentioning that almost half of the tweets were classified by our model as featuring a caring tone - an indication of the way the conference has intensified people's custodial attitude towards the earth.

The model we used classified each tweet with 1 labelled emotion out of a possible 28. To run a similar analysis - check the video at the top of this article or copy and paste the following code snippet into your project's code editor!

classify_text(ds.text, {"model": "joeddav/distilbert-base-uncased-go-emotions-student" }) -> (ds["Emotion Detected"])

A Closer Look

To inspect which emotions are prevalent within each conversation topic - we're using a sorting method called term frequency - inverse document frequency or TF-IDF. This works by reflecting the importance of a single category or text value in relation to the entire set values belonging to a variable. In short - TF-IDF picks out over-represented emotions within conversation topics.

Young People

Led by Greta Thunberg, the youth march in Glasgow took place on Friday 5th October and saw young people from all over the world walk from the west end of the city to the centre.

Conversations surrounding young people at COP26 were closely connected to conversations about protesting and have remained largely positive. Gratitude was detected more within this conversation topic than in any other along with a prevailing sense of excitement.

Boris Johnson

The British Prime Minister has been travelling to and from Scotland  - via train (and sometimes more controversially via plane) - to welcome leaders from around the world to the 2021 UN Climate Conference. As well as opening the conference with celebrated naturalist David Attenborough, Boris Johnson has attended key talks on transport and warming pledges.

But his commitment haven't necessarily gone down well with members of the public. The emotion detection analysis we ran drew attention to an overrepresentation of negative emotions surrounding the politician. Annoyanceamusementconfusion and disapproval were not uncommon in conversations about Boris.

Farming, Agriculture & The Meat Industry

Strong and clear communication about the effects of bad farming practices on the heath of our planet has grown substantially in the past 2 years.

Perhaps unsurprisingly then, conversations surrounding the meat industry and farming have been a significant part of the online debate at COP26. Approvalcuriosity and realization are top of our ranking list here. But although these emotions points towards a desire for change - its good to be aware that people tweeting about farming at COP26 are far more likely to already be engaged in this debate.

Explore the project for yourself to learn more about how people have spoken about different conversation topics surrounding the COP26 summit in Glasgow over the past two weeks!

We collected every tweet published about the 2021 UN Climate Change Conference (COP26) - 31.10.2021 - 12.11.2021 - to study how people have engaged with events during the summit. Using topic analysis and emotion detection, our project dives into people's visceral reactions to agreements on deforestation, commitments between China and the USA and the appearance of Barack Obama.

On top of this, we're loosely looking into whether we can use AI - social media monitoring + Advanced NLP - to monitor the feelings of society in short bursts.

Explore The ProjectWe collected every tweet published about the 2021 UN Climate Change Conference (COP26) to study how people have engaged with events during the summit.

November 12th 2020. The final day of the COP26 UN Climate Summit in Glasgow. For two weeks, world leaders, activists and international delegates have been meeting in Scotland's biggest city. The agenda? Climate change.

But people's feelings about the health of the earth - and what to do about it - is as mixed as the weather in a Scottish summer. For the optimistic amongst us, COP26 offered the promise of progress. For despondents, it represented nothing more than a social gathering for the world's elite. The targets have been ambitious and leaders have a lot to lose. The question we're interested in; Has the latest UN Climate Conference made us more or less hopeful about the future of our planet?

The Method

Collecting every tweet about the conference since the 31st October with Tractor, we set out to find the main topics of conversation surrounding COP26 and analyze how people have thought and felt about them.

The main topics in our data ranged from discussions about the meat industry and methane emissions to debates on the value of the deforestation pledges made by world leaders.

#COP26 Emotion AnalysisUsing topic analysis and emotion detection, our project dives into people's visceral reactions to agreements on deforestation, commitments between China and the USA and the appearance of Barack Obama.

An Emotional Overview

Before we take a closer look at specific emotions associated with conversation topics, it is worth mentioning that almost half of the tweets were classified by our model as featuring a caring tone - an indication of the way the conference has intensified people's custodial attitude towards the earth.

The model we used classified each tweet with 1 labelled emotion out of a possible 28. To run a similar analysis - check the video at the top of this article or copy and paste the following code snippet into your project's code editor!

classify_text(ds.text, {"model": "joeddav/distilbert-base-uncased-go-emotions-student" }) -> (ds["Emotion Detected"])

A Closer Look

To inspect which emotions are prevalent within each conversation topic - we're using a sorting method called term frequency - inverse document frequency or TF-IDF. This works by reflecting the importance of a single category or text value in relation to the entire set values belonging to a variable. In short - TF-IDF picks out over-represented emotions within conversation topics.

Young People

Led by Greta Thunberg, the youth march in Glasgow took place on Friday 5th October and saw young people from all over the world walk from the west end of the city to the centre.

Conversations surrounding young people at COP26 were closely connected to conversations about protesting and have remained largely positive. Gratitude was detected more within this conversation topic than in any other along with a prevailing sense of excitement.

Boris Johnson

The British Prime Minister has been travelling to and from Scotland  - via train (and sometimes more controversially via plane) - to welcome leaders from around the world to the 2021 UN Climate Conference. As well as opening the conference with celebrated naturalist David Attenborough, Boris Johnson has attended key talks on transport and warming pledges.

But his commitment haven't necessarily gone down well with members of the public. The emotion detection analysis we ran drew attention to an overrepresentation of negative emotions surrounding the politician. Annoyanceamusementconfusion and disapproval were not uncommon in conversations about Boris.

Farming, Agriculture & The Meat Industry

Strong and clear communication about the effects of bad farming practices on the heath of our planet has grown substantially in the past 2 years.

Perhaps unsurprisingly then, conversations surrounding the meat industry and farming have been a significant part of the online debate at COP26. Approvalcuriosity and realization are top of our ranking list here. But although these emotions points towards a desire for change - its good to be aware that people tweeting about farming at COP26 are far more likely to already be engaged in this debate.

Explore the project for yourself to learn more about how people have spoken about different conversation topics surrounding the COP26 summit in Glasgow over the past two weeks!

Aim

The Data

185,736 tweets about #COP26 

Explore Yourself

#COP26 | Emotion Analysis

Key Variables

Type of Analysis

Relevant Industries

Other stories

Ready To Get Started?

Ready To Get Started?

Let's dive into your data with Graphext. It's super simple, and you'll get your project ready in a few minutes.