Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

If you have a programmatic way of determining the best match for an arbitrary long tail query, then haven't you already solved your search problem?

I suppose if you had a large enough (more or less hand crafted) sample set you could use machine learning techniques to optimize the weighting of each field, but I don't know if that would get you an answer any faster than tweaking human-designed weightings.

Crowdsourcing search result feedback might be another way to make this scale.





Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: