当前位置: X-MOL 学术Int. J. Inf. Technol. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Query reformulation approach using domain specific ontology for semantic information retrieval
International Journal of Information Technology Pub Date : 2020-05-09 , DOI: 10.1007/s41870-020-00464-2
Navjot Kaur , Himanshu Aggarwal

Conventional search engines provide a lot of irrelevant results to the user’s search query, which consumes a lot of time and effort, as the context and semantics of the request made by the user are not analyzed to the optimum extent. This highlights for the need of a methodology embedded in the search machinery, enabling semantic web search. Ontologies have proven to be an effective technology for such semantic knowledge representation and information retrieval. In this research paper, the researcher has constructed a string ontology in a new domain (the music domain) from scratch using Protégé 5.0 for semantic information retrieval for a query in the search engine. This ontology is further used in the proposed ontology based semantic information retrieval method (OBSIRM), which has been built to refine the web search in the music domain. The researcher has proposed a novel approach to refine the web search in which the query is first replaced with abbreviations along with the use of the multilingual concept to search a query. The results of the proposed method have been compared with Google. The accuracy and efficiency of the proposed method has been measured in terms of precision and recall. The average precision and recall of OBSIRM is 1.43 and 0.7 as compared to generic Google search engine i.e. 1.04 and 0.51. Hence it can safely be concluded that in the domain of research the OBSIRM performs better in terms of users’ expectations than Google (for string instrument domain considered by the researcher) as far as relevant information retrieval is concerned.



中文翻译:

使用特定领域本体进行语义信息检索的查询重构方法

常规的搜索引擎向用户的搜索查询提供了许多不相关的结果,这消耗了大量的时间和精力,因为没有对用户提出的请求的上下文和语义进行最佳分析。这突显了对嵌入在搜索机制中以实现语义Web搜索的方法的需求。本体已被证明是用于这种语义知识表示和信息检索的有效技术。在这篇研究论文中,研究人员使用Protégé5.0从头开始在新领域(音乐领域)中构建了字符串本体,用于在搜索引擎中查询的语义信息。在建议的基于本体的语义信息检索方法(OBSIRM)中进一步使用了该本体,该方法已建立以完善音乐领域中的Web搜索。研究人员提出了一种新颖的方法来完善网络搜索,其中首先使用缩写词代替查询,同时使用多语言概念来搜索查询。所提出的方法的结果已与Google进行了比较。已从精度和召回率方面衡量了所提出方法的准确性和效率。与一般的Google搜索引擎(即1.04和0.51)相比,OBSIRM的平均精度和召回率分别为1.43和0.7。因此可以肯定地得出结论,就相关信息检索而言,在用户期望方面,OBSIRM的性能要优于Google(对于研究人员所考虑的弦乐器领域)。

更新日期:2020-05-09
down
wechat
bug