In this work we present an innovative approach to the semantic context elicitation among a set of terms. Topic and context elicitation using semantic features can be applied to query expansion, natural language processing, and multimedia retrieval. Different techniques rely on web objects to extract information considering the direct semantic relationship between the observed objects. In our approach we explore the Wikipedia collaborative network to extract the pairwise semantic chains. The terms that constitute all the pairwise chains will define the context in which the set of terms are immersed. Traversal, graph and Steiner tree analysis are evaluated by experts. Results are encouraging and experts agree that the Steiner tree analysis conveys additional semantic information about the relationship among the words in the context.
Multi-Term Semantic Context Elicitation from Collaborative Networks
Mengoni P.;Milani A.;
2018
Abstract
In this work we present an innovative approach to the semantic context elicitation among a set of terms. Topic and context elicitation using semantic features can be applied to query expansion, natural language processing, and multimedia retrieval. Different techniques rely on web objects to extract information considering the direct semantic relationship between the observed objects. In our approach we explore the Wikipedia collaborative network to extract the pairwise semantic chains. The terms that constitute all the pairwise chains will define the context in which the set of terms are immersed. Traversal, graph and Steiner tree analysis are evaluated by experts. Results are encouraging and experts agree that the Steiner tree analysis conveys additional semantic information about the relationship among the words in the context.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
