A brief description #20
Conversation
|
The project can still go into many directions, which is why we've not added a readme just yet. One big theme is that we don't know if we want to keep this around as an example for others to copy of if we want to cookiecutter it and make it more general. Not to mention that I'm working on a variant of this work that explores the ML/active learning a bit more. Time will tell! I'll also gladly leave it up to @victorialslocum to decide if she wants to accept this. |
|
Interesting :) I was at @victorialslocum presentation at pycon and found this project very interesting. Especially the idea of having your own custom point of entry to the internet. Kind of more flexible and intelligent version of getpocket.com |
|
This was exactly the pitch that made us interested in this as well. I should say that the approach seems to fit arxiv very well, mainly because hackernews/reddit and similar sources kind of allready behave as a content-filter based on community likes. The reason why this feels interesting is that arxiv seems like a reasonable source to make personal content-based recommendations based on specific topics. |
|
Alright, I just left the demo link. |
Added a brief description, based on information found in the project. Take or leave, no hard feelings :)