Query Rewriting Berbasis Pengukuran Tingkat Kesamaan Semantik Pada Kata Kunci di Beberapa Search Engine

Ahmad Muhammad Thantawi, Awi (2014) Query Rewriting Berbasis Pengukuran Tingkat Kesamaan Semantik Pada Kata Kunci di Beberapa Search Engine. Doctoral thesis, www.upi-yai.ac.id.

[img] Text
DISERTASI_THANTAWI.pdf
Download (3MB)

Abstract

Information can be obtained by one of them with the help of search engines or search engines on the internet. Currently there are many search engines available in the virtual world, including the popular are Google, Yahoo, Bing, Alta vista and so on . Search engines do not always provide accurate information, despite having promised benefits. Accessing information on the Internet based on the keywords that the search engines can be delivered unsatisfactory results for the user, because the use of keywords that convert into a question (query) may not necessarily result in an answer (response) the expected users. In general, users will create another query using keywords that are the same meaning. In addition to this process, the user also perform queries using different search engines. This is done repeatedly until the user gets the desired information. Algorithms produced and supported by the establishment of a database and database similarity thesaurus is used to support the process of rewriting the query. The algorithm was developed on the query rewriting process is by measuring the level of semantic similarity in the concepts that are the result of keyword extraction process so generated several alternative keywords are submitted to multiple search engines The test is done by comparing the results of relevant information at the time of submitting the keywords on some classic search engines, namely Google, Yahoo and Bing, with a prototype query rewriting process by measuring the degree of semantic similarity in a keyword that is enriched by the use of metasearch engines. Keywords transmitted consists of three categories , namely 1 ) the form of a single word keywords , such as " computer " , " car " , 2) shaped keyword phrases , such as " restaurant " , " four wheel drive vehicle " , and 3 ) the word complex -shaped key words , such as " number of cars in Jakarta ". Testing the accuracy of information obtained from rewriting the query results are shown through the value of precision, recall and F-measure. The results of the testing are contained fixes f-measure value of information retrieval trials using classical search engines compared with tests using prototype search. It is concluded query rewriting based on the semantic similarity measurement level keywords in several search engines give better results can satisfy user desires. Keyword : search engine, query rewriting , semantic similarity, query response

Item Type: Thesis (Doctoral)
Subjects: T Technology > T Technology (General)
Divisions: Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science
Depositing User: Dr Thantawi Ahmad Muhammad
Date Deposited: 24 Aug 2021 01:33
Last Modified: 24 Aug 2021 01:33
URI: http://repository.upi-yai.ac.id/id/eprint/4029

Actions (login required)

View Item View Item
Repository UPI Y.A.I is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.