Read on Twitter
1/ Text Mining Podcasts: Emotion Classification + Sentiment Analysis
Austen Allred tweeted this a few weeks ago, and I was curious to see what people were thinking
Turns out, they were thinking about @naval https://twitter.com/Austen/status/1305210103730073600
Austen Allred tweeted this a few weeks ago, and I was curious to see what people were thinking
Turns out, they were thinking about @naval https://twitter.com/Austen/status/1305210103730073600
2/ I mined everything in Excel
These were the top podcasts by recommendation number (n = 208)
These were the top podcasts by recommendation number (n = 208)
3/ These were the top *very specific combination* podcasts. Naval dominated here.
4/ In fact, Naval was a ~quarter of all interviwee-specific recommended podcasts
5/ This makes sense.
Naval is fascinating, and he has an incredible worldview. I find him very inspiring.
@eriktorenberg just released The Almanack of Naval Ravikant and there are hundreds of threads about Naval on this site.
Naval is fascinating, and he has an incredible worldview. I find him very inspiring.
@eriktorenberg just released The Almanack of Naval Ravikant and there are hundreds of threads about Naval on this site.
6/ This is a wordcloud of Naval's transcript from The Knowledge Project w Shane Parrish
The word ‘think’ is mentioned most often. The word ‘can’ is mentioned > 100 times, tied w ‘people’ and ‘read’.
Those seem to encapsulate a large portion of what he seems to value (I think)
The word ‘think’ is mentioned most often. The word ‘can’ is mentioned > 100 times, tied w ‘people’ and ‘read’.
Those seem to encapsulate a large portion of what he seems to value (I think)
7/ I pulled the Naval’s interview transcript from The Knowledge Project Podcast, The Tim Ferris Podcast, and Joe Rogan’s podcast.
I used the sentimentr + tm in R to analyze sentiment, emotion, and word association w EmoLex via Saif Mohammad
I used the sentimentr + tm in R to analyze sentiment, emotion, and word association w EmoLex via Saif Mohammad
8/ (Links to aforementioned pods here)
https://fs.blog/knowledge-project/naval-ravikant/
https://www.stitcher.com/podcast/the-joe-rogan-experience/e/61689613 https://tim.blog/2016/01/30/naval-ravikant-on-happiness-hacks/?_ga=2.213633949.866953269.1600970467-1375831015.1600033130
https://fs.blog/knowledge-project/naval-ravikant/
https://www.stitcher.com/podcast/the-joe-rogan-experience/e/61689613 https://tim.blog/2016/01/30/naval-ravikant-on-happiness-hacks/?_ga=2.213633949.866953269.1600970467-1375831015.1600033130
9/ Naval's conversation w @ShaneAParrish was 'trustful' + relatively positive sentiment (although most negative of 3 at 0.07)
Trust words ~ wisdom, wealth, thoughtful,etc.
Most pos sentence (Shane): "More right... more rational... you're going to get nonlinear returns in life"
Trust words ~ wisdom, wealth, thoughtful,etc.
Most pos sentence (Shane): "More right... more rational... you're going to get nonlinear returns in life"
10/ @joerogan and Naval: Much more anticipation relative to trust. The most positive podcast at 0.098.
Most positive sentence: "Very professional, very quick, very thorough but he did more diligence on me than I did on him" (here we see the flaws of sentiment analysis lol)
Most positive sentence: "Very professional, very quick, very thorough but he did more diligence on me than I did on him" (here we see the flaws of sentiment analysis lol)
11/ @tferriss+ Naval. Sentiment= 0.090
Most +: "The ways we try to get peace from mind are indirect... if you understand things if you see things properly you will naturally slowly develop peace from mind"
Most -: “So you have to ruthlessly, ruthlessly disappoint everybody”
Most +: "The ways we try to get peace from mind are indirect... if you understand things if you see things properly you will naturally slowly develop peace from mind"
Most -: “So you have to ruthlessly, ruthlessly disappoint everybody”
12/ What about the individual podcasts?
I just did some wordclouds (I know they're lame, but I think they're cool) of the Knowledge Project, @EconomicsTalks and The Tim Ferris Show, pulling the descriptions + titles
(Joe just lists the interviewee name, so no wordcloud)
I just did some wordclouds (I know they're lame, but I think they're cool) of the Knowledge Project, @EconomicsTalks and The Tim Ferris Show, pulling the descriptions + titles
(Joe just lists the interviewee name, so no wordcloud)
13/ The Knowledge Project: the most common words in the podcast descriptions being ‘author’, ‘making’, ‘can’, ‘learning’, ‘life’ and ‘world’.
14/ Econ Talks was another top podcast, with the word University the most prominent (likely due to the interviewee’s job title) as well as argues, author, book, and policy.
15/ The Tim Ferris Show was lessons, life, master, and building.
16/ Final thoughts: Trust is important.
People like to talk about learning, life, and the world.
Podcasts are a fine line to walk - conveying trust and anticipation seem to be a pretty good model of a successful pod!
People like to talk about learning, life, and the world.
Podcasts are a fine line to walk - conveying trust and anticipation seem to be a pretty good model of a successful pod!
17/ Also here are some more cool pods to check out that offer unique thought and perspective!
Pls add more below!
@vc_she
@associated_pod
@de_havia / Divergent Unicorns
@TheirLeague
@thewaywelead
@basecspodcast
@femstreet
@dig_history
@coolestnerdspod
Pls add more below!
@vc_she
@associated_pod
@de_havia / Divergent Unicorns
@TheirLeague
@thewaywelead
@basecspodcast
@femstreet
@dig_history
@coolestnerdspod
18/ Full blog post here: https://kylascanlon.com/2020/09/24/natural-language-processing-and-naval-the-art-of-the-podcast/
