Research Article
Hyperlink Analyses of High School Library Website in China
Reference Room for School of Sciences, Yahshan University of Qinhuangdao, Hebei, 066004, China
Along with internet prevalence, more and more readers are accustomed to gain information via it. As regional literature information center, high school library is the most basic and important information source. High school library is qualified to present all kinds of web information services. Therefore, it has become an issue for many library website constructors to ponder how to construct high school library website, how to evaluate construction quality and how to present readers with satisfying web information service. This study is to analyze the hyperlink of high school library website in China.
RESEARCH SCHEME DESIGN
Research object selection: For convenience, we select libraries of famous high schools, excluding those in Taiwan, Hong Kong and Macao.
Research index selection: Webpage quantity: It generally refers to the searched quantity of web pages searched via a search engine.
Total hyperlink quantity: It generally refers to the total webpages connected with a certain website searched via a search engine
Inner hyperlink quantity: It generally refers to the total web pages, via a search engine, related with a certain website in the range of one website.
Outer hyperlink quantity: It generally refers to the total web pages, via a search engine, related with a certain website out of the range of one website.
Hyperlink measure deviance: theoretically, there exists an equation: Overall hyperlink quantity equals inner hyperlink quantity plus outer hyperlink quantity. But in fact, searching has deviance. Hyperlink measure deviance equals (overall hyperlink quantity minus inner hyperlink quantity plus outer hyperlink quantity) divided by overall hyperlink quantity. But it has not evaluating and analyzing significance in this study, so it is not used.
Web-influence factor: Web-influence factor equals overall hyperlink quantity divided by webpage quantity; Outer web-influence factor equals outer hyperlink quantity divided by webpage quantity (Yangtao, 2003; Junping and Lu, 2003).
Research tools selection: Commercial research tools can be selected from some research engines with advanced searching function such as Google, alltheweb, etc, among which Google is more powerful and convenient. Besides the foregoing web information metrology index, Google can provide other concrete-valued data, such as similar webpage, which is used to search webpage relating to some webpage. The number of similar webpage can reflect character degree of website. Hyperlink is used to search webpage connecting with certain webpage (in this article it refers to webpage connecting with the home webpage of some website), the number of hyperlink can reflect other websites the recognition toward it.
Data processing tools: according to need and convenience, we choose EXCEL.
DATA SOURCE AND SEARCHING METHODS
Data searching methods: Firstly, we search high school website via Google. After filling website address in advanced searching frame, press searching button and acquire the relevant data.
Table 1: | Searching formula list |
Table 2: | Statistics of high school website hyperlink library |
2006.3.6(19:00-21:00) data stastic of the webpage |
Otherwise, we can search by searching formula in Table 1.
Website hyperlink data: Data acquired by the above method are listed in Table 2.
DATA ANALYSIS
Statement on website hyperlink data: In the above statistical table, total hyperlink number is webpage number containing some website URL searched by Google. In fact these web pages have connections with the website, which reflect the number of website hyperlink. Inner hyperlink number is webpage number from the website among all websites searched by Google. Including URL. It reflects the completeness of the inner structure of the website; outer hyperlink number is webpage number out side of the website among all websites searched by Google including URL. It better reflects the quality of the website; webpage number is webpage number of a certain website searched by Google. It reflects the size of the website in some way; web influence factor reflects the overall average level of the webpage is connected; outer web influence factor reflects the overall average level of the webpage connected by the outer, which is regarded as reflecting the level of website better; similar hyperlink number is the webpage which is similar and relevant to some home webpage searched by Google. It reflects the individuality of the home webpage; Hyperlink number is webpage number connected with certain website searched by Google. These can be sorted by different index, just as the following Table 3.
It can be seen from Table 3 that the results are quite different with different indexes and it is quite difficult to tell which index is better. So a synthetical index is needed for taxis. The following is relating taxis.
Hyperlink relevance taxis: For a website, the bigger the above indexes, the better they are, which is opposite to similar webpage. Therefore, a reference row can be inserted in Table 2, where the value corresponding to the column of similar webpage is the minimum among the column which is located in, while other values is the maximum. Then by the point relevance formula:
Table 3: | The result taxis of different index |
Table 4: | Hyperlink relating taxis of the web |
Table 5: | Relating texis of every index |
Letting ≤0.5 and the above formula is
By computing, the relevancy values are filled in the above table and sorted by relevance. The result is Table 4.
Just because of the integration of advantages and features, relevancy taxis is more objective and reliable.
For the data in Table 4, a reference column can be listed by relevancy data and also we can take advantage of relevance formula to compute relevancy which connected with other index data column (which is named index relevancy ) in this way, which index is better can be show and sorted.
In Table 5, index relevancy shows the advantage and disadvantage of other web information counting index. It shows the feasibility of evaluating websites by using outer web influence factor etc and the index reliability shows the quality of sorting different index. The results should have significance for the research of web information computation.