Improving The Effectiveness of Texts Retrieval using Knowledge-Based Approach
International Journal of Emerging Trends in Engineering Research
A predicate-based document query language is proposed to allow users to define the search criteria precisely and reliably, and their knowledge of the documents to be retrieved. A guided search tool is built as an intelligent user interface oriented to the natural language in order to help users formulate queries. Supported by a generator of intelligent questions, an inference engine, a query base. A problem is faced when using the modern IR systems. It's represented in the vocabulary problem.
... is problem is represented in the inconsistencies between the terms which are used to describe the terms and the documents that are used by the investigator for describing their need for knowledge. The researcher has an automated thesaurus. This device has been designed using the Vector Space Model (VSM). The researcher used the similarity calculation of Cosine in this method. He used the selected 242 abstract Arabic documents in this article. All these abstracts include the information and computer science process. This paper aimed at building and designing automated Arabic thesauri through the use of the term similarity which could be employed in any particular domain or field for improving the process of expansion and obtaining greater number of relevant documents for the user query. In terms of recall and precision rates, it was found that the similar thesaurus is more capable than the conventional information retrieval system to enhance the recall process and precision.