Abstract | ||
---|---|---|
In the context of semantic knowledge bases, we tackle the problem of ranking resources w.r.t. some criterion. The proposed solution is a method for learning functions that can approximately predict the correct ranking. Differently from other related methods proposed, that assume the ranking criteria to be explicitly expressed (e.g. as a query or a function), our approach is data-driven, being able to produce a predictor detecting the implicit underlying criteria from assertions regarding the resources in the knowledge base. The usage of specific kernel functions encoding the similarity between individuals in the context of knowledge bases allows the application of the method to ontologies in the standard representations for the Semantic Web. The method is based on a kernelized version of the Perceptron Ranking algorithm which is suitable for batch but also online problem settings. Moreover, differently from other approaches based on regression, the method takes advantage from the underlying ordering on the ranking labels. The reported empirical evaluation proves the effectiveness of the method at the task of predicting the rankings of single users in the Linked User Feedback dataset, by integrating knowledge from the Linked Open Data cloud during the learning process. |
Year | DOI | Venue |
---|---|---|
2013 | 10.1145/2480362.2480431 | SAC |
Keywords | Field | DocType |
correct ranking,rank prediction,semantic knowledge base,ranking label,linked user feedback dataset,ranking criterion,knowledge base,implicit underlying criterion,ranking resources w,linked open data cloud,semantically annotated resource,related method,clustering,web services | Information retrieval,Ranking SVM,Ranking,Computer science,Semantic Web,Linked data,Ranking (information retrieval),Artificial intelligence,Knowledge base,Perceptron,Ontology learning,Machine learning | Conference |
Citations | PageRank | References |
1 | 0.35 | 17 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Pasquale Minervini | 1 | 119 | 16.34 |
Nicola Fanizzi | 2 | 1124 | 90.54 |
Claudia D'Amato | 3 | 733 | 57.03 |
Floriana Esposito | 4 | 2434 | 277.96 |