Hi
I am working on building a recommender system on a learning content data. My
data format is a user-item matrix of views. Similar to the below one
NS
353 0 0 0 0 0 0 0 0 0
0 ... 0 0 0 0 0 0 0 0
0 0
354 0 0 0 0 0 0 0 0 0
0 ... 0 0 0 0 0 0 0 0
0 0
355 0 0 0 0 0 0 0 0 0
0 ... 0 0 0 0 0 0 0 0
0 0
356 0 0 0 0 0 0 0 0 0
0 ... 0 0 0 0 0 0 0 0
0 0
357 0 0 0 0 0
Where each row is for a user id and each column is all the videos in the
system. The value corresponding to each video column is either 1 if a user
has watched/clicked on the video else 0.
This is an implicit feedback dataset.
Now, I am looking at spark.Mllib package and they seem to give an example
where they are saying the dataframe should be of the form [(userid),
(product),(ratings)]. My dataframe is basically user-video where each column
is different videos and value of those columns is the rating(views in this
case).
I guess this is what the original paper and elsewhere in collaborative
filtering algorithm the data is represented. Am not sure if this format of
data is supported by Spark.mllib or I have to convert it to the one they
have given an example for?. Any idea how to do that from my dataset?
--
View this message in context:
http://apache-spark-user-list.1001560.n3.nabble.com/Data-Format-for-Running-Collaborative-Filtering-in-Spark-MLlib-tp27832.html
Sent from the Apache Spark User List mailing list archive at Nabble.com.
---------------------------------------------------------------------
To unsubscribe e-mail: user-unsubscr...@spark.apache.org
