Today ended the three-day Google Earth Engine User Summit 2018. On social media channels such as Twitter and YouTube, could non-participants indirectly follow the progress and news around the Earth Engine technology. I think @eMapR_Lab describes the situation quite well.

This blogpost serves as a small personal summary of what I have heard from Dublin in the last three days.

With the help of the rtweet package I downloaded all tweet information with the hashtag #eeus19 and #EEUS18. I analysed the data mainly with the R packages dplyr, tidytext and created plot with ggplot.

Twitter Activity Timeline

Here I show the Twitter activity timeline, which also included the days before the summit. However, the highest Twitter activity was during the days of the Summit from June 12 until June 14. The time of the Twitter data download was at 2018-06-14 17:51:40.

Top Contributors - Original Tweets

I have labelled the following particularly active twitter users (5 or more original tweets) as top contributors.

Tweet Engagement

Engagement rate is the total number of times a user has interacted with a Tweet. This includes all clicks anywhere on the Tweet (including hashtags, links, avatar, username, and Tweet expansion), retweets, replies, follows, and likes (Info: Twitter). Here I considered only retweets and likes as engagement.

Add Score Metric

Dean Attali introduced in his blogpost a simple score metric for how successful a tweet is, using the very little bit of information we have. This is of course very arbitrary. He chose to score a tweet’s success as a linear combination of its “# likes” and “# retweets”. Tweets during the Google Earth Engine User Summit 2018 had roughly four times as many likes as retweets, therefore retweets get four times the weight. By means of the scorer points we find the top 10 tweets with the greatest engagement.

Top 10 - Score Tweets

Count of unique word found in tweets

So what do the conference participants actually tweet about? Here I use the R code from Lesson 3. Text Mining Twitter Data With TidyText in R from www.earthdatascience.org and count unique words found in tweets. I exclude obvious buzzwords like eeus18, earth, engine, googleearth, summit, google, earthengine, googleearthengine, 2018, but also words like https, t.co.

Sentiment Analysis of Google Earth Engine User Summit 2018 Tweets

Sentiment analysis classifies words as positive or negative. Here I use the ‘bing’ sentiment data explained in Lesson 6 from https://www.earthdatascience.org and plot top words, grouped by positive vs. negative sentiment.

The mentioned negative sentiments refer to illegal logging and cloud detection and removal in satellite imagery. These tasks are hard and complex. However, its seem that everybody is super happy & excited to be in Dublin at the Summit.

Given the three day summit tweet time information, I plot the data over time to see how sentiment changed over time.

Wordclouds of Twitter hashtags used during the summit

Hashtags used during the Google Earth Engine User Summit - Day 1: 2018-06-12 Hashtags used during the Google Earth Engine User Summit - Day 1: 2018-06-12
Hashtags used during the Google Earth Engine User Summit - Day 2: 2018-06-13 Hashtags used during the Google Earth Engine User Summit - Day 2: 2018-06-13
Hashtags used during the Google Earth Engine User Summit - Day 3: 2018-06-14 Hashtags used during the Google Earth Engine User Summit - Day 3: 2018-06-14

Bottom line

This blog post is a summary of the Google Earth Engine User Summit 201 tweets from Dublin, Ireland. If you have any questions, suggestions or spotted a mistake, please use the comment function at the bottom of this page.

Previous blog posts are available within the blog archive. Feel free to connect or follow me on Twitter - @Mixed_Pixels.