2 ms·
Yeah, I use the incident link graph similarity though. It's been almost ridiculously effective. Not had the same success with word models, although it is proba
by marginalia_nu 4y ago
Yeah, I use the incident link graph similarity though. It's been almost ridiculously effective.
Not had the same success with word models, although it is probably largely a matter of finding an appropriate embedding.
The data is pretty stale, but here is a demo: https://explore2.marginalia.nu/ https://explore2.marginalia.nu/
OP (or anyone else): Shoot me an email and I'll give you a copy of Marginalia's link graph to play with.
- supriyo-biswas 4y agoPlaying around with the explore link, I see most of the popular news websites are excluded (such as wsj.com and nytimes.com)?
- marginalia_nu 4y agoYeah, I excluded them from the computation as a performance optimization. The juice didn't seem worth the squeeze given my work specializes in smaller websites.