Observations from TED Talks

Posted on May 25, 2018

Background: TED is a not-for-profit organization devoted to spreading ideas via short talks. Please see ted.com for more details. The dataset used in this blog was obtained from Kaggle. The blog analyzes all the talks published from 2006 to 2017. An associate shiny web app for this blog can be found here

This blog and the web app aim to, (1) understand people's overall engagement with talks, and (2) provide recommendations to potential viewers by providing popular talks and relevant ratings.

Figures 1, 2, and 3 provide insight into people’s level of engagement, as demonstrated by the number of views, comments, and the ratio of comments per ‘1000-views’. [Note: the ratios are typically less than 1% for a social media metrics like comments per views. For example, a 0.5% comments to views ratio can be considered good for a youtube video (http://tubularinsights.com/3-metrics-youtube-success/), this ratio translated into 5 comments per 1000 views. Therefore, for the analyses in this blog, the number of comments are divided by ‘1000 views’. ]

Figure 1. The number of published talks peaked in 2012 and has decreased slightly since then.

Figure 2. Left Panel: the distribution of the number of views per talk has remained more or less the same from 2006 to 2017. Right Panel: the distribution of the number of comments per talk remained steady from 2006 to 2013 and then has gone down significantly.

Figure 3. People's engagement increased sharply from the beginning of the publications of talks to the year 2010, but there has been a sharp decline in people's engagement since 2010. The left panel shows cumulative numbers for all talks, whereas the right panel shows the mean and median values of 'comments per 1000 views' for talks for all the years.

How are TED talks described?

Viewers describe the talks as mostly inspiring, informative, fascinating, persuasive, and beautiful. Please see the full list of descriptions in Figure 4. For each talk, every one of the 14 descriptions was selected by 0 or more viewers. The box plot shows the votes for all descriptions for all talks. For example: the median for the rating 'inspiring' for a talk was 220, meaning, on average 220 people rated a talk as 'inspiring', and half of the talks were rated 'inspiring' by more than 220 people, and half of the talks were rated “inspiring” by less than 220 people.

Figure 4.

Figure 5. Top three talks for each rating descriptions.


Figure 6. Ken Robinson’s “Do schools kill creativity” has the highest views, 47 Million. 24924 people found the talk to be inspiring. The word cloud shows how viewers rated it.

Figure 7. The figure shows the greatest percentage of ratings relative to the total number of views for a specific description. Although some of these talks were viewed much lesser times, these were best received regarding rankings by viewers. For example, 'Building a park in the sky' by Robert Hammond was rated beautiful by the highest % of viewers.


TED talks continue to be well received, but the level of viewers' engagement as demonstrated by their comments has steeply gone down in recent years. Analyzing the youtube statistics for these talks can provide additional insight into the questions probed in this blog.

About Author

Leave a Comment

No comments found.

View Posts by Categories

Our Recent Popular Posts

View Posts by Tags

#python #trainwithnycdsa 2019 airbnb Alex Baransky alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus API Application artist aws beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep Bundles California Cancer Research capstone Career Career Day citibike clustering Coding Course Demo Course Report D3.js data Data Analyst data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization Deep Learning Demo Day Discount dplyr employer networking feature engineering Finance Financial Data Science Flask gbm Get Hired ggplot2 googleVis Hadoop higgs boson Hiring hiring partner events Hiring Partners Industry Experts Instructor Blog Instructor Interview Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter lasso regression Lead Data Scienctist Lead Data Scientist leaflet linear regression Logistic Regression machine learning Maps matplotlib Medical Research Meet the team meetup Networking neural network Neural networks New Courses nlp NYC NYC Data Science nyc data science academy NYC Open Data NYCDSA NYCDSA Alumni Online Online Bootcamp Open Data painter pandas Part-time Portfolio Development prediction Prework Programming PwC python python machine learning python scrapy python web scraping python webscraping Python Workshop R R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn Selenium sentiment analysis Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau team TensorFlow Testimonial tf-idf Top Data Science Bootcamp twitter visualization web scraping Weekend Course What to expect word cloud word2vec XGBoost yelp