As a side note, the blog poster, and the blog post he references, characterizing the movie as terrible annoys me and shows that they both absolutely missed the point of the movie.

The movie depicts all sorts of different types of "love". Lust is characterized by the man who cheats on his wife; Sex is characterized by the guy who goes to the US; Platonic Love is the love between the rock star and his manager; First Love is the story of the young boy; Unrequited Love or Unattainable Love is the man who is in love with his best friend's wife, and so forth.

It's not a terrible movie at all, it's freaking brilliant.

It's a pretty screenwriting 101 level of symbolism. Why does that make it brillant?

The poster successfully conveyed the insight that made it brilliant film for them and their enthusiasm makes me wish to see it again.

I don't know what practices are actually taught in 101, but the greatest artists won't be those that think themselves above them, but those who really appreciate them.

You wish to suggest you are superior but multiple stories to give different angles of a concept isn't symbolism, its a theme. The narrative structure is definitely unusual in the genre, frequently described as clever/innovative by reviewers and stands as a substantial achievement to have brought so many stories together successfully. None of that means you have to like it but the cheap middle-brow dismissal fools no-one.

The "how" is more defining than the "what". All bloody kill-em-all movies contain bad one liners, swearing, guts, as-much-almost-nudity-as-allowed-on-tv, etc. But there's a difference between any Tarantino movie and some all-star Rambo-like movie which came out this year.

I don't think that part implied that the movie is brilliant but that the movie is brilliant in itself as a work of art.

Seems like you could get new insights and ideas by getting lost in the graphs (if only I had better recollection of this film!) The interactive graph-timeline is a rather uncommon visualization technique, but I can see it being very useful. It only needs smooth graph transitions while using the timeline scrubber, and it could make a very useful tool with different applications.

And as films go, I think a similar analysis performed on Cloud Atlas would be extremely interesting.

If you like that movie you must read through to the very last sentence of the post. It's a terrible pun? Joke? Reference? Whatever it is, it's funny and a great end to a nifty bit of analysis.

A minor R tip: if you're trying to filter on substrings in dplyr, using grepl() is faster and does not require loading an extra package.

That said, this is a great, clear writeup, and I wasn't aware R had dendrogram capabilities natively. I'll have to spend more time looking into Igraph too.

I find that surprising. Do you have benchmarks?

Huh. Turns out I'm wrong. I ran benchmarking on a 1.8M row dataset of SF crime data w/ 100 replications.

Running grepl() via:

   benchmark(df %>% sample_frac() %>% filter(grepl("ARREST", Resolution)))

completed in 257.58 seconds.

Running str_detect() via:

   benchmark(df %>% sample_frac() %>% filter(str_detect(Resolution, "ARREST")))

completed in 248.917 seconds.

Overall, str_detect() is faster by 3%. The stringr packages has more surprises than I thought!

    For legal reasons I don’t want to host the script file myself, but it’s literally the first Google result for “Love Actually script.” Just copy the .doc contents into a text file called love_actually.txt

Isn't this fair use?

The use of the script for statistical analysis is fair use; distributing it isn't.

nice work