A Semantic Retrieval Model with Domain Ontology Based on Wolfberry Disease and Pests

Guan-feng LI,Wei-jun LI

DOI: https://doi.org/10.3969/j.issn.1673-629X.2017.09.011

2017-01-01

Abstract:The traditional information retrieval methods based on keywords and subject are lack of processing and presentation on the se-mantic level and thus lead to incomplete and inaccurate retrieval results,which cannot meet the user' s retrieval needs totally. In order to improve the quality of retrieval system,a model of semantic retrieval based on ontology for wolfberry disease and pests domain has been presented and its key technologies have been investigated in the processing of the introduction of ontology into semantic retrieving. It con-structs the domain ontology of wolfberry diseases and pests,and modifies its inconsistent problem to ensure consistency and accuracy of wolfberry diseases and pests knowledge. Under the guidance of hierarchical tree structure of the domain ontology,a concept similarity method considering semantic distance,superior concepts coincidence degree and depth of concept nodes is proposed. Combined with the semantic inference rules a semantic retrieval model based on domain ontology of wolfberry disease and pests is realized. The experimental results demonstrate that the semantic retrieval model has well overcome the deficiency of the traditional retrieval method and effectively improved the recall and precision of information retrieval.