0

I have followed the recommendation engine example as given in chapter 6 of the Mahout in Action text book. Now while I get the output for the designated user, Now is it possible to output the item-item similarity matrix computed during the 3rd Map reduce job. On completion of the MapReduce job I find there exists a temp folder which contains a directory called /user/hduser/temp/similarityMatrix however upon executing
hadoop fs -cat temp/similarityMatrix/part-r-00000 I get

SEQ org.apache.hadoop.io.IntWritable%org.apache.mahout.math.VectorWritable�h��8��Oh�pg;Q����@@����@@2����?�����@����?�&����@?�?�&   ����?�@?�&
                      ����?�?�@+�"���������?��D?�?�t�"�����
����@�?��?��?��?�?�����?�?��?��?�|�"�����
                                                    ����@(@�?��"?��?��?��?�
?�J?�΍��?��?�B�"���������?��M?�Ї��?��$?�!?�L�"�����!����@(@"�?�?�&?��@�?��?��@0@!?�)?�� ?��
                                   ?��
                                      ?��?��?�?�k?��?��ӯ�?��?��?�(?�?�F?�'?�!?��
     ?�

Is it possible in any way to output the item-item similarity matrix?

Abhiroop Sarkar
  • 2,251
  • 1
  • 26
  • 45

1 Answers1

0

This is a hadoop sequence file, a binary format. You need to write code to use it. Try mahout seqdumper on it to inspect the contents.

pferrel
  • 5,673
  • 5
  • 30
  • 41