A personalized context-dependent web search agent using Semantic Trees

In Web searching applications, contexts and users' preferences are two important factors for internet searches in such a way that results would be much more relevant to users' requests than with current search engines. Researchers had proposed a concept-based search agent which uses Conceptual Fuzzy Set (CFS) for matching contexts-dependent keywords and concepts. In the CFS model, a word exact meaning may be determined by other words in contexts. Due to the fact that numerous combinations of words may appear in queries and documents, it may be difficult to define the relations between concepts in all possible combinations. To solve this issue, we proposed a Semantic Tree (ST) model to define the relations between concepts. Concepts are represented by nodes in the ST, and relations between concepts are defined by the distances between nodes. Moreover, this paper applies users' preferences for personatizing search results. Finally, the fuzzy logic will be used for determining which factor, semantic relations or users' preferences, will dominate results.