First time poster on my discovery of the wonderful Echonest.
I'm still researching how all your data gets pulled together to adapt my queries but have discovered a number of spelling errors on song names.
I pulled in a query of all unique song ID's from an artist but noticed there were obvious duplicates where a song as more than one unique ID. I then processed this information on unique song names so that the end result would be a list of all unique songs by an artist.
Due to spelling and Grammar errors I'm still left with a lot of duplicates. For me this isn't really a huge issue as currently it is only reference material but I'm wondering firstly what people do to get around this and secondly what can be done to improve the data.
As I said, I'm still researching but I only found one post about spelling and this wasn't related.
hi Felixkat, we regularly ingest new data and sometimes there will be temporary duplicates in the song/search and artist/songs results. We try to automatically de-duplicate as often as possible. If you can give us a specific example we can investigate if there is something amiss.