جستجوی مقالات مرتبط با کلیدواژه « موتورهای کاوش » در نشریات گروه « کتابداری و مدیریت اطلاعات »

The aim of this study was to investigate the image retrieval from selected search engines according to the written and semantic features of Persian language and determine their relevance using recall and precision formulas and to identify the most efficient search engine in retrieving images in Persian and by survey-analytical method. It was done using direct observation technique. After reviewing related researches, search keywords list was formed in the form of a checklist based on the written and semantic features of Persian language. Each of these keywords in the studied search engines, including two general search engines Google and Bing and Duckduckgo semantic search engine, which are among the most used search engines and have also provided the ability to search for images in Persian, search and the number of relevant and unrelated retrieved results were recorded. Then, the recall and precision of search results in each search engine were calculated and the relevance of images based on these features in each of the studied search engines was investigated. A variety of descriptive statistical techniques were applied to analyze the data along with Kolmogorov-Smirnov, Shapiro-Wilk, Kruskal-Wallis and Friedman tests. Findings demonstrated that Google, Bing and Duckduckgo search engines do not pay enough attention to the written and semantic features of Persian language and many of these features are ignored while searching and retrieving images. In the present study, Google search engine had a higher recall and precision than the other two search engines, and despite the claim of semantic search engines to provide better and more relevant information than other search engines, Duckduckgo search engine did not show good performance in retrieving images related to the written and semantic part of Persian language. There is also a significant difference between the recall and the precision of the three studied search engines.

Today, the evaluation of information retrieval systems, especially search engines, has become one of the most important studies in the field of information science. Nevertheless, a research as a review research on the evaluation of search engines was not observed. So, the aim is to analyze the related literature of information retrieval evaluation field using quantitative, qualitative and mixed approaches.

This research is a review study to review the related articles. The terms »arzyabieh motor kavosh«,»arzyabieh nezam bazyabieh ettlaat«,»arzyabieh rabt«,»motor kavosh«,»motor jostejo« in the Magiran and SID database were searched. Also, the terms »evaluation search engine»،«evaluation information retrival»،«method and information retrieval«, »research and search engine« and »relevance« in the google scholar was searched and then the retrieved articles were studied. Finally, the research on the evaluation of general search engines published in Persian and English from 1998 have been studied using a library method and an analytical approach..

The results showed that, in the reviewed studies, the search engines evaluation has conducted through one of the quantitative, qualitative and mixed approaches. In the quantitative approach, there are several research categories such as coverage and overlap, abstracting and indexing quality, retrieval algorithms, recommender systems, interface and document ranking quality. In the qualitative approach, there has been observed two kinds of studies which include ethnography and grounded theory. In the ethnography research, the users’ information retrieval behavior is described. In the other category, namely grounded theory, two research areas may be identified; one is fully committed to the qualitative principles and another semi-committed to them. Hence, Hjorland (2010) has criticized researchers who regard their research as a "functional" and qualitative approach, but did not follow all of their principles.

As in the research methodology, first, a quantitative method and then a qualitative method are proposed and finally, a mixed method is proposed to use the strengths of both methods. It seems that this process of change in research methods from quantitative to qualitative, and then from qualitative to composite has also affected the evaluation of information retrieval, especially search engines. In data retrieval research, a systematic approach has been initially introduced in evaluating information retrieval systems. Then some researchers have criticized this approach, which is the result of a user-oriented approach. In recent years, researchers such as Saracevic (2007) and Thoronley (2012) have considered both approaches as necessary ones and pointed to a dual-axis approach to data retrieval assessment research.

Provide an image indexing template based on search engines to retrieve text-based and content-based indexed images

In terms of purpose,application and type of descriptive research, there is a correlation.The statistical population included experts working in the offices of national newspapers in Tehran,who were familiar with five search engines.Data collection for component design is by a researcher-made questionnaire with 20 items by documentary method.The quality validity coefficient of the content of the questionnaire is higher than 0.78 and the content validity index is higher than 0.79.The reliability of the questionnaire was measured based on Cronbach's alpha coefficient equal to0.936.

In exploratory factor analysis, 20 items studied in four factors with textual indexing factor loads (with factor load range between0.51 to 0.76), image indexing(with factor load range between0.59 to 0.74), content indexing (Loaded with factor load range between 0.57 to 0.78)and effectiveness (with factor load range between0.53 to 0.85).

Internal factor 1(text indexing)with an average factor load of 0.79plays the most important role in constructing the external factor or the overall scale, while internal factor2(image indexing) with an average factor load of0.74 is the second most important component.In the construction, the external factor was general,followed by the internal factor3(content indexing)with an average factor load of0.72; Finally,internal factor4(effectiveness)with a factor load of 0.69 had the least role in making the overall factor.The fit indices of the model, such as relative chi-square and economical indices, showed the fit to be quite appropriate.As a result, the obtained model has a relatively good fit to the data.

This research was conducted to examine the effect of discarding XML declaration and changing the file extension on increasing the index ability and visibility of the element tag names of metadata records based on DCXML, MARCXML, and MODS in the Web search engines environment.
Two groups of metadata records (300 XML-based records as the control group with the normal structure, and 300 XML-based records without XML declaration and with the extension according to the name of related metadata standard in an experimental group) were analyzed through the experimental approach. Two independent websites assigned to each studied group, and they were introduced to Google and Yahoo search engines. Subsequently, using element-based search strategies, the index ability and visibility of the published metadata records on those websites were examined.
Findings: Findings showed that Google and Yahoo indexed all the elements of tagged names of the metadata records relating to the experimental group indexes, and presented them in their search results. But the elements of tagged names in the control group’s metadata records were not indexed by the search engines.
Based on this study, it is possible to retrieve the experimental group's metadata records by their element of tagged names in the search engines. But the records of the control group are accessible by the values of the element only. In the end, , some patterns were suggested to the metadata creators and the search engine developers.

This research was aimed to measure the number of duplicated items in Persian information retrieval from the Web, and to compare search engine's abilities in removing duplicated items from the results.
Quantitative methods And Descriptive Statistics And Evaluation of information retrieval systems.
Findings: The findings showed that in subjects such as law, geography, literature there were more than 70% duplicated items in search results. Google retrieved the lowest number of duplicated items (42.8%) and Bing retrieved the highest number of duplicated items (58.3%).

This paper aimed at investigating the results of searching in title and URL fields carried out by professionals from three areas: Engineering, Pure Sciences and Economics in Hamedan province, Iran. The comparison of the frequency of keywords in title and URLs was also another aim of the study. Overall 90 experts from Hamedan universities from three areas who were involved in research projects were selected. They carried out advanced searches in Google with their keywords (single and multi words) limiting it to the title and URL fields. Then, they indicated the relevance of search results filling in the related checklist. Finally, the researcher classified the indicated relevance which we reported into descriptive and analytical statistics. There was a significant relationship between the quantity of keywords and conformation rate of title and URLs. An increase in keywords co-occurred with decrease in conformation rate and an increase in conformation rate between title and URL resulted in more retrieval relevance. There was also a significant difference between the three groups of experts based on the quantity of keywords and also URLs they used to search.

The aim of this research was to examine the use of data island method for creating metadata records based on DCXML, MARCXML, and MARC21with indexability and visibility of element tag names in the Web search engines environment. 600 metadata records were developed in two groups (300 XML-based records as the control group with the normal structure, and 300 HTML-based records in an experimental group with special structure embedded in the tag of HTML based on the data island strategy). These records were analyzed through an experimental approach. The records of these two groups were published on two independent websites and were introduced to Google and Yahoo! search engines for indexing. Findings show that all the element tag names of the metadata records created based on the data island method relating to the experimental group indexed by Google and Yahoo! were visible in the search results. But the element tag names in the control group’s metadata records were not indexed by the search engines. Based on this study, it is possible to retrieve the metadata records by their element tag name in search engines. But the records of the control group are accessible by the element values only. The research suggests some patterns to metadata creators and end users.

This research is going to compare internet results derived by some users Technical and engineer groups, factual sciences and economics in G.G.O with restriction at title & URL, and also the of keyword's appearance in the internet URL and research group's titles. Survey society are 90 experts of Hamedan G.G.O They searched their desired keywords (one or more words) in the URL and title of web pages and determined the amount of relevance retrieval results in the check list that was provided for this reason. For classifying and data analysis, descriptive statistics (frequency, mean, base of sample and standard deviation) and inferential statistics (univariateKalmvgvrf Smirnov test, Duncan’s post hoc test, chi-square independence test, paired t-test, correlation COEFFICIENT erson and Spearman’s test, ANOVA and chi Krvksa L – Wallis test) used. The results showed that the increase in the number of search keywords and URLs as the correlation decreases and the greater the degree of correspondence between the title and the URL link is higher. the result show up that with increase of search keywords scale in the title & URL will reduce and there are meaningful defference between numbers of present thematic keywords at title & URL researched groups and search engines can increase weighting to present keywords in the URLs rank the more correlative results and provide for users.

The goal of information systems is to retrieve relevant information fulfilling users’ needs. According to this, search engines have offered a new capability as “suggested keywords”. This is an essential problem that how much these suggested keywords have compatibility with users’ needs. The present research aims to investigate the efficacy of Google's suggested keywords in query expansion and retrieval relevance from users’ view. Through a survey method data were collected from 60 postgraduate students in Humanities/ social sciences and Basic/ Engineering Sciences at Ferdowsi University of Mashhad, Iran. These data were collected by using a researcher-made questionnaire. Non-probability sampling –especially purposive sampling- was used as the sampling method. Also, individual interviews were established with each of the 60 students. Findings show that there is a significant difference between the relevance of retrieved results related to primary keyword search and results retrieved through query expansion (suggested keywords). Also there is no significant difference between retrieved results in the two fields of sciences. Finally, the paper suggests some solutions to improve retrieved results related to suggested keywords.

This article has evaluated the impact of using Meta tag on the improvement of information retrieval of the World Wide Web. So it reviewed the research papers that have examined Meta tag as a means to enhance information retrieval. Results showed there were some researches that have confirmed the using of Meta tags was efficient for improving information retrieval of the web, but results of a research that has done by Mohamed (2006) indicated no significant difference in the rank of pages that include metadata and those that don’t include any metadata. Therefore more studies should be done in this field. In general, Meta tags can provide site owners and authors with a degree of control over how a web page is indexed and this is a considerable advantage that it shouldn’t be ignore easily.

Increasing growth of information sources on the internet, coupled with the technical and non-technical drawbacks of search engines, has caused a massive amount of such resources to be kept hidden from users. The phenomenon, referred to as invisible web, is a subject of debate by many information scientists. Presently many libraries, information centers and even researchers tend to digitize resources and put them on the Net. Many of these are one of the kind and in the absence of a suitable access to internet resources, many of human scientific, research, intellectual and cultural heritage would be lost to both present and future generations. Librarians and information officers could play an important role in pushing down the level of information invisibility through boosting user awareness about invisible web and various techniques for accessing and seeking such information. The present paper has endeavored to unlock the concept of invisible web from both a technical and user-oriented perspective. In the conclusion, the role of librarians and information officers for promoting various modes of accessing such information are discussed.