Farahnaz Naderbeigi; Alireza Esfandiari-Moghadam; Atefeh Zarei; Behrooz Bayat
Alireza Isfandyari Moghaddam; Zohreh Bahari Movaffagh
Volume 9, Issue 2 , May and June 2012
Abstract
Background: Knowing the rate of overlapping in results obtained from information retrieval tools is important to prevent waste of time. Medical and health-related information has an outstanding place among the variety of information available on the Internet. Hence, this research aimed to study the rate ...
Read More
Background: Knowing the rate of overlapping in results obtained from information retrieval tools is important to prevent waste of time. Medical and health-related information has an outstanding place among the variety of information available on the Internet. Hence, this research aimed to study the rate of overlap among medical search results derived from MESH (medical subject headings) in general search engines. Methods: This applied, descriptive, comparative approach survey was conducted in 2010. The research sample included 4 general search engines, i.e. Yahoo, Live Search, Google, and Ask. In order to determine the rate of overlapping among the searches, 10 medical keywords were first selected randomly. The first 10 results retrieved by each search engine were then listed. Finally, the overlap rate was calculated by means of Microsoft Excel. To determine the overlapping search results, 10 common terms were searched in the mentioned search engines. Then, 10 results retrieved in the first page of each search engine were collected and compared. The overlap rate among the retrieved results was finally calculated using a checklist and a researcher-made equation whose validity was confirmed based on observations of the researchers, comments of some expert professors, and especially referring to related literature. In addition, due to the descriptive, comparative approach of the research, descriptive statistics revolving around a statistical equation was utilized. Results: It was found that search engines can retrieve 48% of relevant searches retrieved by other search engines. Yahoo had the highest number of unique results while Google had the highest number of results overlapping with other search engines. The overlap rate of the 4 studied search engines was 11%. Conclusion: The results obtained from different search engines hardly overlap. Therefore, users are suggested to search more than one search engine simultaneously to access more relevant documents. Moreover, they are advised to use specialized search engines to search medical information. Keywords: Internet; Information Retrieval; Keywords; Search Engine
Mohammad Karim Saberi; Alireza Moghaddam
Volume 8, Issue 2 , May and June 2011
Abstract
Introduction: Web citations (URLs) are at risk of decay and can be easily fade because of the variable and unstable nature of the internet. So, the purpose of this research was to assess web citation accessibility and decay of articles published in journal of the medical library association (JMLA).Methods: ...
Read More
Introduction: Web citations (URLs) are at risk of decay and can be easily fade because of the variable and unstable nature of the internet. So, the purpose of this research was to assess web citation accessibility and decay of articles published in journal of the medical library association (JMLA).Methods: In this applied– descriptive study, JMLA articles (231) in a seven year period (from 2002 to 2008) downloaded from magazines archives and the articles that have web citation (175), determined. Then, all of URLs were tested to specify their accessibility and decay. When we failed to directly access the URLs, we used Internet Explorer, Internet Archive and Google search engine to access missing URLs. Obtained data analyzed using Excel software to determine the half-life of web citations and the reasons for the URL failure. Then gathered the result of these tests and analyzed them using Microsoft Excel to distinguish cause of these failure and to count web citations half-life.Results: 76 percent of articles had web citation and the average of web citation per article was 5 URL. From 1049 cited URL, 31 percent was inaccessible, but by searching these URLs, the decay decreased to 13 percent. The majority of errors were the 404 (not found) and the average of web citations half-life was approximately 7 years.Conclusion: The decay of URLs is a big problem that can not be ignored because URLs are increasingly used and lost in scientific articles, the decay of URLs will continue until better preservation policies (such as internet archive) are adopted.Keywords: Internet; Library and Information Science; Citation; Databases.