Many thanks for this, I really appreciated your advice to use word clouds to compare things rather than just show something without any other reference point. I also liked the welcome surprise of that font package, definitely going to use it as well! All the best.
wordclouds are great... although what's been difficult is getting them to reflect nicely in an rmarkdown document. wordcloud2 puts some neat js w/ it while wordcloud makes a static image -- so you may have to utilize htmlwidgets or plotly. what would be great is a series on incorporating libraries for text analysis. it's intuitive to put together an x-y graph with some count of a discrete variable to gain some understanding... but what would be good approaches to get value out of textual data?
I think this is a great question but I don’t know if I can give an exact answer. I think it really depends on what your end goal with the data is. If you’re looking for a more quantitative analysis, it might be worth looking into topic modeling, sentiment analysis (although many may argue isnt that reliable), and n-gram token frequencies (but plotting instead of making a word cloud) just to name a few. These are all things I hope to cover in future videos but there are great resources already out there I’d recommend taking a look at!
when i run this code df = df %>% filter(nchar(as.character(word)) > 2, word != "don’") i get this error below Error in `filter()`: ! Problem while computing `..1 = nchar(as.character(word)) > 2`. Caused by error in `nchar()`: ! invalid multibyte string, element 300 Run `rlang::last_error()` to see where the error occurred.
You deserve an award for this. Best explanation out here. Thanks!
Thank you so much for making this video! It was really well made, and broke things down very effectively!
Glad you enjoyed it!
thank you for also adding the text cleaning process, i've been looking for tutorials of that step
Best video I watched on the topic. Thank you.
These videos are gold!! Keep them coming!
Thank you so much for this video. It is really clear and helpful. Looking forward to ur other tutorials!
Your videos are excellent and really well produced. Looking forward to more!
Love it! You explained it so well! Thank you :)
I really like your explanations. 😌keep going. ❤️
Thank you so much!
Many thanks for this, I really appreciated your advice to use word clouds to compare things rather than just show something without any other reference point. I also liked the welcome surprise of that font package, definitely going to use it as well! All the best.
Glad to hear it! Thanks for watching!
You videos are very informative. I'm eagerly waiting for your upcoming videos
I hope he returns to RUclips 🙏. He's the best in explanation. Or even if he creates a course.
PERFECT.
wordclouds are great... although what's been difficult is getting them to reflect nicely in an rmarkdown document. wordcloud2 puts some neat js w/ it while wordcloud makes a static image -- so you may have to utilize htmlwidgets or plotly.
what would be great is a series on incorporating libraries for text analysis. it's intuitive to put together an x-y graph with some count of a discrete variable to gain some understanding... but what would be good approaches to get value out of textual data?
I think this is a great question but I don’t know if I can give an exact answer. I think it really depends on what your end goal with the data is. If you’re looking for a more quantitative analysis, it might be worth looking into topic modeling, sentiment analysis (although many may argue isnt that reliable), and n-gram token frequencies (but plotting instead of making a word cloud) just to name a few. These are all things I hope to cover in future videos but there are great resources already out there I’d recommend taking a look at!
Is is possible to control the speed at which the words are displayed?
You're a fanomena. Can fit this into 5 classes of r course
when i run this code
df = df %>%
filter(nchar(as.character(word)) > 2,
word != "don’")
i get this error below
Error in `filter()`:
! Problem while computing `..1 = nchar(as.character(word)) > 2`.
Caused by error in `nchar()`:
! invalid multibyte string, element 300
Run `rlang::last_error()` to see where the error occurred.
Cannot thank you enough
Why he speaks so fast?