http://hdl.handle.net/11189/1444 Sun, 16 Aug 2026 08:30:24 GMT 2026-08-16T08:30:24Z https://digitalknowledge.cput.ac.za:443/retrieve/3845/ http://hdl.handle.net/11189/1444 http://hdl.handle.net/11189/6020 Title: Search engine optimisation: application to an academic digital library Authors: Dhansay, S; Weideman, Melius Abstract: The purpose of this research was to identify and evaluate techniques used to ensure high listing as well as build up a high rank of an academic digital library on Google. The first step was to identify techniques. They were implemented and their impact on listing and ranking measured.The techniques that were measured and evaluated included, metadata tags, number of pages, impact of in-links, length of time the domain has been available and use of directories. The effects of these techniques were measured using the scoring of the PageRank algorithm as used by Google. Certain limitations were encountered including the limited space of time to build up directory listing and the frequency of Google crawling websites, as this has a direct influence on the results. The research conducted provides evidence that using keywords in the right position as well as gaining links from websites that using keywords in the right position as well as gaining links from websites that contain relevant and quality content improves the listing on Google. Mon, 01 Jan 2007 00:00:00 GMT http://hdl.handle.net/11189/6020 2007-01-01T00:00:00Z http://hdl.handle.net/11189/6019 Title: Information retrieval from the internet for academic use and copyright purposes Authors: Reyneveld, J; Weideman, Melius Abstract: This research study discusses the reason for using search engines vs academic search engines for information retrieval, and the importance of copyright. The internet was used to gather information for the study, in which the relevance of the different search engines for different end users is indicated. This research took into account that every searcher may have a different need, but it focuses more on the need for academics to identify good sources of information. The result shows that academic search engines are better suited if in need of information of an academic nature, be it science or history. It also proved that an academic search could present more accurate information, as previous occasions. The conclusion was drawn that to base one's investigation for information solely on popular search engines could leave one with information that is not viable because anybody could have placed it on the Internet. With academic databases, one can be assured to a great extent that the information retrieved comes from a credible source. Mon, 01 Jan 2007 00:00:00 GMT http://hdl.handle.net/11189/6019 2007-01-01T00:00:00Z http://hdl.handle.net/11189/5050 Title: Rogue’s Gallery – South African university website visibility Authors: Weideman, Melius Abstract: South Africa has 23 universities of a variety of descriptions. The current generation of potential students often use search engines to collect information about universities, which could influence their final choice of where to enroll. The increase in mobile web usage has also contributed to an increase in university web access for timetables, exam results and course details. As a result, university websites have joined commercial websites, fighting for top rankings for certain search queries. The purpose of this research project was to evaluate the 23 South African university homepages using a proven academic model. The most important contributing elements of website visibility were identified on each website, measured, and a scoring system designed. This system enabled the researcher to allocate points on a sliding scale for visibility to search engine crawlers. The results proved that South African university homepages spanned a wide range of scores, from webpages meeting most criteria to those scoring rather low overall. It was clear that in some cases, no effort was made to assist search engine crawlers to find websites with ease and index them well. In these cases, it could have a serious effect on the ease with which the student body’s expectations of relevant search results can be met. Suggestions are made on how to improve the situation. Sat, 01 Jan 2011 00:00:00 GMT http://hdl.handle.net/11189/5050 2011-01-01T00:00:00Z http://hdl.handle.net/11189/5049 Title: A comparative analysis of search engine indexing time Authors: Zuze, Herbert; Weideman, Melius Abstract: Internet usage is increasing dramatically daily, as is Web development which is enhanced by the emergence of new Web technologies. Social networks are continuously dominating our Web interactions through Facebook, Twitter, LinkedIn and MySpace, drawing a considerable number of Internet users. However, the dominance of the Web as the epicentre of both important and useless information has affected competition in industries like e-business. It takes some time for a website to be indexed and appear on the search engine result page. The indexing process involves the reading and recording of the weight-carrying words in a search format to an index file by a crawler. This results in the webpage being discovered by users on a search engine results page. A user can also submit a website manually for indexing. Crawlers should then visit this website, record all the words on the pages, and note links to other sites. The index file is updated regularly, either by human editors or by these crawlers. After an extensive literature survey, empirical evidence indicated that the indexing period for webpages is not fixed. In a recent two phase experiment, five websites were monitored to determine their indexing time for Google, Yahoo! and Bing. The experiment proved that there is a relationship between keyword density and indexing time. Yahoo! and Bing seem to favour sites with high keyword density when indexing. During the first experiment the shortest indexing waiting time was five days and the longest, 33 days. For the second one, the waiting time varied between 19 and 29 days. However, according to this study, a period of approximately 19 days is a reasonable average waiting time. It is possible that the Google sandbox effect plays a role in these experiments, and this was also investigated. Sat, 01 Jan 2011 00:00:00 GMT http://hdl.handle.net/11189/5049 2011-01-01T00:00:00Z