Hi All,
I was able to successfully run item similarity algorithm on my dataset.My input data had the following format:
userid itemid……
I used the following command:
./mahout itemsimilarity -i /scratch/SimilartyInput -o /scratch/SimilartyOutput -s SIMILARITY_COOCCURRENCE --maxSimilaritiesPerItem 10
Here is the format of the output:
itemid1 itemid2 similarityval
similarityval is a floating point number. Can someone please point me to documentation that says how to interpret this number? Also, is there any tool to visualize this? I have over million users and item ids .
I was able to successfully run item similarity algorithm on my dataset.My input data had the following format:
userid itemid……
I used the following command:
./mahout itemsimilarity -i /scratch/SimilartyInput -o /scratch/SimilartyOutput -s SIMILARITY_COOCCURRENCE --maxSimilaritiesPerItem 10
Here is the format of the output:
itemid1 itemid2 similarityval
similarityval is a floating point number. Can someone please point me to documentation that says how to interpret this number? Also, is there any tool to visualize this? I have over million users and item ids .