Help get this topic noticed by sharing it on Twitter, Facebook, or email.

Neat tool, but one serious weakness IMHO

I appreciate the challenge of this line of work. As a data analyst formerly of Navteq I can sympathize. But I think there is one problem here, you simply have to solve more effectively for this to really take flight. Of course, I'm talking about the quality of the data. If you are going to rely on data entry from the vast public to tighten up your POI (points of interest) data, it's not going to work.

I've been browsing and reading about your methodology and can see that a lot of good work has been done. But guess what? There are 4 major, popular "amenities" each from a different category within 2 blocks of my house that are missing from the data. I also live 1 block from a very busy rail station and downtown area so I imagine a lot of people examine my neighborhood. I don't understand why the amenities aren't there, but a decent subscription to a POI database provider would handle that for you. More cost perhaps, but much better data - and with regular updates.

Now many users will never notice the data quality problem. And "everything's an approximation" when one has to use creative ways to locate parks and solve other algorithm issues. Maybe that's good enough for your business model. Sometimes we have to look for short cuts and just get something out on the web.
1 person likes
this idea