Somali Language Information Retrieval Using Query Expansion
No Thumbnail Available
Date
2020-06-23
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
In these days the number of documents available online on the Internet is growing at a very rapid speed and has become quite large over the past few years. To search such documents various types of search engines or Information Retrieval systems has been performed for numerous languages. Based on the language the information seeker is using, Information Retrieval allows to find relevant information with respect to the information need of the user. During retrieval many words can be represented by using more than one representation and this can create a confusion in the retrieval process. Modern Information Retrieval systems employ techniques to deal with such confusion in retrieval process.
This research deals with Somali Language Information Retrieval system. Keyword based Information Retrieval systems confirm lower performance compared to the systems with query expansion. To solve such problem, we have implemented the research with query expansion technique using thesaurus. Thesaurus expands original query terms by finding synonyms of terms from manually constructed list of alphabetically ordered words.
A prototype has been developed to test the performance of the system. The prototype is implemented using java programming language with open source library called lucene. A testing collection of about 2007 documents of three different domains collected from various sites used. The experimental result of this study is a two phase, the first phase of the experiment is to test the system without query expansion. The second phase is testing the system with query expansion. Our experiment shows a hopeful result, and notable improvement by recording 53% and 72% of precision and recall respectively. After query expansion technique is applied the system recorded 64% and 84% of precision and recall respectively with unranked set of retrieval.
Description
Keywords
Information Retrieval, Somali Language, Thesaurus, Query Expansion