http://spark.apache.org/docs/latest/mllib-collaborative-filtering.html https://github.com/apache/spark/blob/master/data/mllib/als/test.data

I want to find missing value in a matrix based on past matrices

But I find that the test.data are not matrices, instead a long list of 3 columns

How do I understand test.data and create my own data?

Which value represent missing value?

and how to understand the result, which value represent result?


This data set is in a sparse format. Most ratings are unknown, which is a common scenario in recommender systems. This data format is much smaller due to leaving out all the unknowns. You can interpret this data as follows:

user, item, rating

What this means is that user U has rated item I with rating R. Every UxI combination not in your dataset is unknown and has to be predicted using collaborative filtering.

  • $\begingroup$ how about missing value in matrix problem? how to apply this? $\endgroup$ – user353573 Jun 6 '16 at 3:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.