Oriental Journal of Computer Science and Technology

Abstract

In this paper, initially, the description and the processing way of the existing queries and languages in this field is performed and a base is graded for the new method by classifying the present methods and evaluating on strengths and weaknesses of each method. This three-step approach is proposed to improve previous methods. The method involves three steps:

1. First step: Simplifying query and reducing the searching area.

2. Second step: Producing a result table as a guide for the query processor.

3. Third step: Processing the document nodes according to the guide table produced in the previous step.

This three-step method can prove its ability compared to the previous methods. The method is developed so that it has the efficiency for complicated multi-branches queries. Then, it is tried to optimize the method, considering the concept of the extraction points. During the three steps, useless nodes – the nodes that either do not make a response for the user, or return a repetitive one – were used so much; using such nodes is reduced by introducing an index called level index.

Keywords

Full Text