بررسی مشکلات جستوجو و بازیابی اطلاعات در پایگاههای اطلاعاتی از جنبه ویژگیهای نگارشی زبان فارسی

The present research was carried out with the aim of explicating the major writing and semantic problems of Persian language when using data environments and determining the degree of compatibility and attention to these features in Persian databases. This research is of survey analytical type being conducted through direct observation. Having reviewed the related literature, we kept a checklist of search keywords. Each of these keywords was searched in the databases under study, such as Iranian Research Institute for Information Science and Technology, Regional Centre for Information Science and Technology, NoorMagaz, and Scientific Information Database affiliated with Jahad Daneshgahi, and the number of retrieved findings was recorded. Some of the writing and semantic features of Persian language contribute to problems associated with retrieving information from the selected databases. Some of these features include connected and disconnected forms of writing of derivative, compound, and derivative-compound words, diversity of plural forms, loanwords and their equivalents in writing as well as polysemy, homonymy, etc., in semantics. For instance, retrieving different results for various writing forms of the keywords «فناوری و فن آوری» as derivative-compound words or «پتاسیوم و پتاسیم» as various forms of recording words, or retrieving different findings for keywords «دریای خزر، دریای مازندران و دریای کاسپین»  as well as lack of their appropriate coverage as synonymous words and giving the user information about it in order to improve the exploration process, for it has negative effects on search and retrieval process. Findings indicated that Persian databases do not pay adequate attention to writing and semantic features of Persian language, and disregard many of its features in searching and retrieving information. In connection with the impact of these features on the interaction of users with databases, Persian-speaking users’ need for native exploration tools and databases designed in accordance with the features of their own language have become more and more urgent. The present research has examined the ability of Persian databases in covering some of the features of this language, which have a noticeable impact on the process of searching and retrieval, pinpointing the weak points and strengths of these databases. The results of the present research could be utilized to improve the performance of the above-mentioned databases.