:D

Cool, you should now get a bunch of recs. Try it again.

Sorry, can you tell me your MAL username?

Well, that’s unfortunate. You seem to be unique, so no recommendation can be given. I’ll retrain the system today and see what happens. Sorry for the inconvenience. It might take a while.

Yeah, I’ve set a minimum limit for animelists. Your animelist appears to be smaller than 50 so it won’t be processed. This is not because it can’t, but because I manually set it so that anime newbies don’t get bad quality recommendations. It has something to do with how the algorithm works.

1. I’ll see what I can do.
2. I actually already have an anime recommendation system, but it’s for MAL. And it only works if you already have a bunch of different animelists. I might have to make another one, or have the recommendations be user-submitted like in MAL.

I like that weird smirk. It’s charming in a weird way.

1. I see. That probably could be a solution. Thanks, that gave me an alternative way of doing things. I’ll consider this while making this project and see how I can do it.
2. Yeah, that definitely will be a feature. No worries!

As for point 3 and 4, they most likely will be implemented as well.

Honestly, your repo idea could even make this project not only an anime tracker but basically a tracker for every audiovisual content without much effort.

Unfortunately, this tracker will need to rely on centralized databases such as MAL, anidb and anilist as I can’t really seem to find an easy way to curate new animes without them.

1. I am not very sure about this point honestly, as previously discussed in this thread, having a database that is only maintained by the community has its own disadvantages that definitely outweight its advantages. For example, you would need constant moderator approval, which is slow.

2. I am not sure what you mean by the ability to share them privately. Do you mean like whitelisting certain people that can view your animelist? Like how MAL has a friends only thing?

I mean, it’s a tracker with forums.

Basically the reviews, threads and comments of an anime are going to be federated. So a new anime entry will create a new community in Lemmy for example and threads/reviews are going to be posts in that community which everyone on the fediverse can comment on.

Yes, something similar to Bookwyrm. I decided to use the Jikan API which scrapes the MAL website if I get approval from the MAL team. The migration feature shouldn’t be hard and I’ll implement it for various platforms (MAL, AniList, Kitsu, etc.)

Thanks.

Alright, I’ll see if I can make it exclude already watched animes. The recommendations you get and also the top 3 animes are subject to change as the training is still ongoing. The training should take about 2-3 days maximum.

Thanks for the feedback and have a nice day.

Well, a few more anime probably won’t change your genre combination but if it ever did, it would have to be within a day for you to even notice. The 24 hours expiry limit is configured in the anote.toml configuration file for the genre combination caches. So if something like that happens, your animelist will be calculated again in 24 hours

By the way, what exactly do you mean by “it did not take my completed animes into account”? Do you mean it showed you animes that you’ve already watched?

And also, I recommend you to read the “You should know” section in the README.md file in the GitHub repository as it should explain some stuff that you just asked.

(I don’t really know why that SSL handshake error happens, I will look into it.)

I just released the project, you can find it here: https://github.com/Asudox/anote

Thanks.

https://myanimelist.net/profile/porotoman99

your combo id: 104080

…which translates to:

1. Action
2. Comedy
3. Drama

deleted by creator

That’s okay. I’d also like to test it against small animelists.

The thing is anime.plus is more of a statistic tool than a recommendation algorithm. I can’t comment on their recommendation algorithm since it’s written in PHP and I don’t know PHP at all (my algorithm is written in Rust btw). What I noticed is that they mixed in themes to genres. Probably because of MAL’s poor labeling.

I’ve also been trying to add some rather stalker-like factors into my algorithm just to test, such as your average watch time of 12 episode animes and changing the weights according to that. However, it became complex pretty soon and my algorithm slowed down to 1 millisecond from 500 microseconds, so yeah. It also didn’t have that much of an impact and I figured out that it is rather dumb since liking a certain genre does not mean everyone will binge-watch the anime like I do.