Abstracts of the Articles Challenges of Connected and Disconnected Persian Words in Information Storage and Retrieval

Persian language and script, due to the special features of grammar and writing, create difficulties in information storage and retrieval. Because of the different approaches to writing and because of its inherent complexity. Persain script has created enormous challenges for database designers, indexers, and end-users. This study was conducted using content analysis technique. 100 dissertations were randomly selected as the samples of scholarly Persian texts in the field of LIS in the Iranian Research Institute for Information Science and Technology (IRANDOC) and the Regional Information Center for Science & Technology (RICeST). Using the Style manual adopted by the Academy of Persian Language, words that had been separated were analyzed to identify if the separations were true or false. Next, the items in both databases with different formats were searched and finally, the results were evaluated and compared. The results showed that 70.20 percent of items are correctly written. Also, it was found that 51.6o percent of these words were two-part and 47.50 percent were three-part words and most writers were confused about writing words that are two-part and derivative. In the RICeST database, titles were retrieved only with the exact format entered and in the IRANDOC database, only 58 percent of items were retrieved by changing the format of writing. This study revealed that the IRANDOC database did better in retrieving titles with different variations of writing than RICeST. Paper recommends that thesis writers also need to use a unified national rules with particular emphasis on writing two-part and derivative words.