I'm confused by the function "get_relation" defined in "add_relation_label.py". It appears that the authors assume that the song is represented by "s", the user is represented by "u", the artist is represented by "p" and the type of song is represented by "t". However, the song id and user id token directly from the original KG data is something like "++5wYjoMgQHoRuD3GbbvmphZbBBwymzv5Q4l8sywtuU=" or "NYE7sp/xNTq3AMFmhLT9kHdtqItEOA34m7kbGUr4Y0k=" which is quite different from the assumption above. What am I supposed to do? Reindex each user and song? I didn't find any instructions about this problem in "readMe.pdf". Please, help me!
I'm confused by the function "get_relation" defined in "add_relation_label.py". It appears that the authors assume that the song is represented by "s", the user is represented by "u", the artist is represented by "p" and the type of song is represented by "t". However, the song id and user id token directly from the original KG data is something like "++5wYjoMgQHoRuD3GbbvmphZbBBwymzv5Q4l8sywtuU=" or "NYE7sp/xNTq3AMFmhLT9kHdtqItEOA34m7kbGUr4Y0k=" which is quite different from the assumption above. What am I supposed to do? Reindex each user and song? I didn't find any instructions about this problem in "readMe.pdf". Please, help me!