Systematic review of Query Expansion in Persian Language

Query expansion is considered as an appropriate solution for solving the problem of short and ambiguous user queries. The purpose of this study is to conduct a systematic review of Persian language query expansion.

Current research is done with using a systematic review method based on Okoli & Schabram's Guidance. Searching in the scientific databases with related keywords ​​led to 35 works in Persian language and 18 works in English language. By applying primary refinement, inclusion and exclusion criteria to study and after expert review, six Persian works and eight English works were selected for doing a systematic review. A checklist was designed and needed information extracted from the works. Finally, findings were processed to achieve four goals of the study: identifying methods, knowledge sources, test collections, and research gaps.

The systematic review showed that 14 works deal with query expansion in Persian language. These works fall into four categories based on knowledge sources for term expansion: relevance-based (eight works), knowledge-based (two works), web-based (two works), and combined-based resources (two works). Most of these studies have been done on news test collections and the Hamshahri newspaper corpus has been used almost in the half of researches as a knowledge source for term expansion as well as test collection.

There is much room to research in the field of query expansion in Persian language. Various knowledge sources, especially web-based ontologies and resources should be considered and used for query expansion in Persian language. Besides, the use of a standard test collection provides the researchers with the facility of comparing the different methods.