I should have left a comment: thanks for pointing that out, I made the change indeed!

Thank you for doing that. And also thank you for posting the excellent guide to begin with!

#666 > #333 And #666 is lighter so less contrast. You’re both saying the same thing.

You're right about the graphs. The horizontal axis is the minutes of the movie though (on the second graph). Will get rid of the gradient and add labels!

I love the visualization! Is the website a template? What lib for the graphs?

Just looks some Bootstrap and ChartJS.org.

Good question, I just thought it looked nice but maybe I shouldn't have

Cool! I thought it was an indication of sentiment at first, I was like oh yeah la la land gets sad towards the end. Cool endeavor!

Curious if you investigated other libraries or algorithms. have you seen this - i have been playing with it, seems reliable: https://github.com/cjhutto/vaderSentiment

I don't know this movie but you can try it out by uploading the movies srt file.

Brazil is a dystopian satire. I would wager that GP is asking about scenarios where there is significant ambiguity and/or situational/verbal irony.

I just gave it a try: https://www.crealdo.com/story/movie-sentiment/movies/30 Never seen the movie so I can't judge myself.

That seems like a plausible analysis, the movie becomes darker as it proceeds, though there's a false happy ending (which in some cases of ham-fisted editing is the only ending - I saw it aired on network TV in the US this way) followed by a much more somber closing.

OP here! I was watching the Stanford NLP classes a while back (https://www.youtube.com/playlist?list=PLiNErZ5Bus8qNxNsFZFkh...) and ended up trying the part about sentiment analysis on srt files. The way it works right is quite primitive but you can still see some "trends" on most movies. If there are any suggestions on how I could make this smarter I would love to hear them!

Maybe you could use RNNs to do the sentiment analysis, either Recurrent or Recursive Neural Networks. Some sample code is available in the book "Tensorflow for machine intelligence" in github.

Maybe if you get information about cut frequency you could identify parts that are more actionish / exciting than others.

has anyone tried opensubtitles.org

just found it on google

Try http://www.yifysubtitles.com/

Just sometimes, because is the only place that has subtitles for some old movies. But I personally prefer Subscene.com.

never mind looks kinda scammy

Nah, it's fine, it tries to push you their downloader, but if you avoid the pitfalls the content is good. I've used it many times over the years.

Must be something with your setup, tsc is known to be quite fast. Maybe try using it with the watcher? "tsc -w" and you'll never need to compile more than one file at a time.

How do you market this? Is it mostly SEO or ads as well? Must be expensive if you want to compete with Nike and Adidas for clicks...

At the beginning it was basically word of mouth and niche related forums on the internet.

Then I started reading about SEO and advertising. Organic search more or less work but I got almost no traffic from a couple of banners on related pages during few months. I didn't try advertising networks like AdWords.

I don't see big brands as my competition. I have partnered with some but I don't think they sell much on the internet (typically higher price tags) compared to full equipped city centre stores.

Very cool, I've been thinking of doing an affiliate website for a while. But it seems like it could take a lot of time and patience to get it profitable. Thanks for sharing!

The sooner you start it, the quicker you see some results ;)

Am I the only one that thinks this is a little creepy? I've been meaning to turn this off for a while now but still haven't gotten to it, goes to show how concerned I am about it I guess. (Though I don't use an Android phone anymore).