Why did you decide to use a supervised learning approach for this? You're hard-coding the categories in this manner.
You could try using clustering instead (K-means) to group the posts together into their own categories. This is similar to a "trending topics" approach, where the categories could change each day.
On a side note, the domain name is terribly confusing.
I have tried different clustering methods before, but did not receive enough satisfying results. Especially I had a look in what carrot can do, because I find the library pretty cool itself. That its built in Solr and can do everything on the fly was just a bonus. But the entities change too much, there is not enough similar data (each day) which result in poor clustering, the clusters are not meaningful - no matter what I did. Probably I did something wrong. After all, I wanted to look into that a little bit later again. And the main purpose was actually not provide a reader-App with hard-coded categories. The main purpose was a feed which is my personal "top news" based on what I "liked" before - having my own categories. Ping me on thomas@newscombinator dot com and I'll shoot you a beta login, maybe you'll find it useful.
The confusion of the name is already well noted, and a "re-branding" is not excluded. Considering its not even a real product, its weird to talk about a rebranding though :) - Paul forgive me the confusing naming. I had nagrgtr.com before (for News-Aggregator), but I found that domain not-so-good (=sucks) as well. I guess I will sit down again and brainstorm a little bit - suggestions welcome.
you don't build a product around a product with 90% same name, it took me 5 min to figure your blog is not of ycombinator, at first seriously, i rejoiced as a designer, that 'finally' design is showing signs of improvments. but only to find, it was of a side-project built around ycombinators's news section. bad choice of name. project is cool.
I know it does not matter to ycombinator, because they are cool guys. but know the 'consumer confusion' thing?
I had the opposite reaction: good name and idea, but the link takes me to a page with a big image, the link to the reader app is tiny, and the app itself is heavy with JS. I'd rather the design was rethought than the name...
probably being a 34 years old ceo sucks. :) that looks at things a lot differently. and have to deal with things which people call it 'shit happens'. I am only rejoicing when it does not. and it's very rare.
copyright act is entirely and solely built around 'consumer confusion'. anyway ;) there is apple bank https://www.applebank.com/ with apple in their logo too ;) but a slight resemblance in design language resulted in 1 billion fine to samsung. reason., former does not cause consumer confusion and the later did or tried to steal the thunder immorally.
and i did mention, project is cool. it's bookmarked in the same folder where i have hackernews ;)
but generally i tend not to like anything that tries to steal the thunder immorally... in this case, if the OP tries to monetize this app with the same domain name, it would be immoral and would delete it from my bookmarks.
Similar projects (various readers for HN) keep popping up from time to time. I'll put on my startup hat and say that there's a market for it, at least an indication that there's a problem :)
Lets not call it a problem, lets call it an opportunity. Perfection lies in the eye of the beholder...
Everyone feels different about HN. The community is great, the content is awesome, but the topics are too diverse for my own feeling. There seems to be no real "Ycombinator HN is about ..." except of [mixed IT]. You find everything, from Blogs about food to latest 0-day sec-vuln. Lifestyle, Marketing, Management, name it. Its not surprising that there are plenty of readers out there. I don't say I make everything better, best or worse. I just say I make it the way I like it and how I want to read HN. If someone else likes that too, fine. In the meantime I'm happy to have a new side-project in my portfolio which others may find useful as well :).
I had idea, it is basically hackernews feed but with tags or grouped somehow. Can you provide hackernews feed with links being tagged in some nice way ? Might be helpfull to quickly sweep through what is interesting and what not.
Why did you decide to use a supervised learning approach for this? You're hard-coding the categories in this manner.
You could try using clustering instead (K-means) to group the posts together into their own categories. This is similar to a "trending topics" approach, where the categories could change each day.
On a side note, the domain name is terribly confusing.