Research Article
Dealing with Pure New User Cold-Start Problem in Recommendation System Based on Linked Open Data and Social Network Features
Table 5
Partial representation of the vector modeling the LOD-based features extracted from DBpedia for the users User 1, User 2, and User 3.
| | Basic LOD-based features | User 1 | User 2 | User 3 |
| | (dbo:birthPlace, dbr:Germany) | 1 | 0 | 0 | | (dbo:birthPlace, dbr:India) | 0 | 1 | 1 | | dbo:birthSign, dbr:Scorpion) | 0 | 0 | 1 | | (dbo:College, dbr:IIT Delhi) | 0 | 1 | 0 | | (dbo:Employer, dbr:Amazon) | 1 | 0 | 0 | | (dbo:age, dbr:36) | 1 | 0 | 1 | | (dbo:age, dbr:22) | 0 | 1 | 1 | | (dbo:Discipline, dbr:Arts) | 1 | 0 | 0 | | (dbo:sex, dbr:Female) | 0 | 0 | 1 | | (dbo:Profession, dbr:Racer) | 1 | 0 | 1 | | (dbo:Profession, dbr:Dancer) | 0 | 1 | 0 | | (dbo:tattoo, dbr:Shiva) | 1 | 0 | 0 | | (dbo:skinColor, dbr:brown) | 0 | 1 | 1 |
|
|
