The thesis describes a search engine of Google company and its method of web pages evaluating, which is used for sorting them on the search engine result page. In addition to the description of how search engines work in general, the thesis targets on differencies between Google and other search engines, as well as it mentions characteristics and principles designed by its founders, Larry Page and Sergey Brin, which make it unique. In the second part there are outlined some of many signals which are rated for every single web page. The most concerned signal is PageRank as the leading idea of pages evaluating, based on their link scheme which is founded on principles of citation analysis. Factors for evaluating are analysed using particular examples, and the topic is ended by the description of Google search engines changes from its beginning in 1998 until present.

No description available.

Current Web image search engines, such as Google or Bing Images, adopt a hybrid search approach in which a text-based query (e.g. "apple") is used to retrieve a set of relevant images, which are then refined by the user (e.g. by re-ranking the retrieved images based on similarity to a selected example). This approach makes it possible to use both text information (e.g. the initial query) and image features (e.g. as part of the refinement stage) to identify images which are relevant to the user. One limitation of these current systems is that text and image features are treated as independent components and are often used in a decoupled manner. This work proposes to develop an integrated hybrid search method which leverages the synergies between text and image features. Recently, there has been tremendous progress in the computer vision community in learning models of visual concepts from collections of example images. While impressive performance has been achieved on standardized data sets, scaling these methods so that they are capable of working at web scale remains a significant challenge. This work will develop approaches to visual modeling that can be scaled to address the task of retrieving billions of images on the Web. Specifically, we propose to address two research issues related to integrated text- and image-based retrieval. First, we will explore whether models of visual concepts which are learned from collections of web images can be utilized to improve the image ranking associated with a text-based query. Second, we will investigate the hypothesis that the click-patterns associated with standard web image search engines can be utilized to learn query-specific image similarity measures that support improved query-refinement performance. We will evaluate our research by constructing a prototype integrated hybrid retrieval system based on the data from 300K real-world image queries. We will conduct user-studies to evaluate the effectiveness of our learned similarity measures and quantify the benefit of our method in real world search tasks such as target search.

This thesis focuses on index and retrieval mechanism of an ontology based medical domain search engine. First, indexing techniques and retrieval methods are reviewed. Then, a special indexing and retrieval mechanism are introduced. This thesis also specifies the functional requirements of these mechanisms. Finally, an evaluation is given by indicating the positive and negative aspects of mechanisms.

No description available.

Thesis (M.S.)--State University of New York at Binghamton, Department of Computer Science, Thomas J. Watson School of Engineering and Applied Science, 2006. / Includes bibliographical references.

Thesis (MPhil)--Stellenbosch University, 2004 / ENGLISH ABSTRACT: With the introduction of the Internet and the World Wide Web (www), information can be easily accessed and retrieved from the web using information retrieval systems such as web search engines or simply search engines. There are a number of search engines that have been developed to provide access to the resources available on the web and to help users in retrieving relevant information from the web. In particular, they are essential for finding text information on the web for academic purposes. But, how effective and efficient are those search engines in retrieving the most relevant text information from the web? Which of the search engines are more effective and efficient? So, this study was conducted to see how effective and efficient search engines are and to see which search engines are most effective and efficient in retrieving the required text information from the web. It is very important to know the most effective and efficient search engines because such search engines can be used to retrieve a higher number of the most relevant text web pages with minimum time and effort. The study was based on nine major search engines, four search queries and relevancy judgments as relevant/partly-relevanUnon-relevant. Precision and recall were calculated based on the experimental or test results and these were used as basis for the statistical evaluation and comparisons of the retrieval effectiveness of the nine search engines. Duplicated items and broken links were also recorded and examined separately and were used as an additional measure of search engine effectiveness. A response time was also recorded and used as a base for the statistical evaluation and comparisons of the retrieval efficiency of the nine search engines. Additionally, since search engines involve indexing and searching in the information retrieval processes from the web, this study first discusses, from the theoretical point of view, how the indexing and searching processes are performed in an information retrieval environment. It also discusses the influences of indexing and searching processes on the effectiveness and efficiency of information retrieval systems in general and search engines in particular in retrieving the most relevant text information from the web. / AFRIKAANSE OPSOMMING: Met die koms van die Internet en die Wêreldwye Web (www) is inligting maklik bekombaar. Dit kan herwin word deur gebruik te maak van inligtingherwinningsisteme soos soekenjins. Daar is 'n hele aantal sulke soekenjins wat ontwikkel is om toegang te verleen tot die hulpbronne beskikbaar op die web en om gebruikers te help om relevante inligting vanaf die web in te win. Dit is veral noodsaaklik vir die verkryging van teksinligting vir akademiese doeleindes. Maar hoe effektief en doelmatig is die soekenjins in die herwinning van die mees relevante teksinligting vanaf die web? Watter van die soekenjins is die effektiefste? Hierdie studie is onderneem om te kyk watter soekenjins die effektiefste en doelmatigste is in die herwinning van die nodige teksinligting. Dit is belangrik om te weet watter soekenjin die effektiefste is want so 'n enjin kan gebruik word om 'n hoër getal van die mees relevante tekswebblaaie met die minimum van tyd en moeite te herwin. Heirdie studie is baseer op die sewe hoofsoekenjins, vier soektogte, en toepasliksheidsoordele soos relevant /gedeeltelik relevant/ en nie- relevant. Presiesheid en herwinningsvermoë is bereken baseer op die eksperimente en toetsresultate en dit is gebruik as basis vir statistiese evaluasie en vergelyking van die herwinningseffektiwiteit van die nege soekenjins. Gedupliseerde items en gebreekte skakels is ook aangeteken en apart ondersoek en is gebruik as bykomende maatstaf van effektiwiteit. Die reaksietyd is ook aangeteken en is gebruik as basis vir statistiese evaluasie en die vergelyking van die herwinningseffektiwiteit van die nege soekenjins. Aangesien soekenjins betrokke is by indeksering en soekprosesse, bespreek hierdie studie eers uit 'n teoretiese oogpunt, hoe indeksering en soekprosesse uitgevoer word in 'n inligtingherwinningsomgewing. Die invloed van indeksering en soekprosesse op die doeltreffendheid van herwinningsisteme in die algemeen en veral van soekenjins in die herwinning van die mees relevante teksinligting vanaf die web, word ook bespreek.

Thesis (MSc)--Stellenbosch University, 2003. / ENGLISH ABSTRACT: We present DEADLINER, a special-purpose search engine that indexes conference and workshop announcements, and which extracts a range of academic information from the Web. SVMs provide an efficient and highly accurate mechanism for obtaining relevant web documents. DEADLINER currently extracts speakers, locations (e.g. countries), dates, paper submission (and other) deadlines, topics, program committees, abstracts, and affiliations. Complex and detailed searches are possible on these fields. The niche search engine was constructed by employing a methodology for rapid implementation of specialised search engines. Bayesian integration of simple extractors provides this methodology, that avoids complex hand-tuned text extraction methods. The simple extractors exploit loose formatting and keyword conventions. The Bayesian framework further produces a search engine where each user can control each fields false alarm rate in an intuitive and rigorous fashion, thus providing easy-to-use metadata. / AFRIKAANSE OPSOMMING: Ons stel DEADLINER bekend: 'n soekmasjien wat konferensie en werkvergaderingsaankondigings katalogiseer en wat uiteindelik 'n wye reeks akademiese byeenkomsmateriaal sal monitor en onttrek uit die Web. DEAD LINER herken en onttrek tans sprekers, plekke (bv. landname), datums, o.a. sperdatums vir die inlewering van akademiese verrigtings, onderwerpe, programkomiteë, oorsigte of opsommings, en affiliasies. 'n Grondige soek is moontlik oor en deur hierdie velde. Die nissoekmasjien is gebou deur gebruik te maak van 'n metodologie vir die vinnige oprigting van spesialiteitsoekmasjiene. Die metodologie vermy komplekse instelling m.b.v. hande-arbeid van die teksuittreksels deur gebruik te maak van Bayesiese integrering van eenvoudige ontsluiters. Die ontsluiters buit dan styl- en gewoonte-sleutelwoorde uit. Die Bayesiese raamwerk skep hierdeur 'n soekmasjien wat gebruikers toelaat om elke veld se kans om verkeerd te kies op 'n intuïtiewe en deeglike manier te beheer.

No description available.

This research is an integral part of the effort aimed at overcoming the limitations of the classic search engines. This thesis is concerned with the investigation of the impact of different modes of mediation on the web search process. Conceptually, it is divided into three main parts. The first part details the investigation of methods and mechanisms in user profile generation and in filtering search results. The second part deals with the presentation of an approach and its application in the development of a mediation framework between the user and the classic Web Search engines. This involved the integration of the explicit, implicit and hybrid modes of mediation within a content-based method, and was facilitated by the adoption of the Vector Space Model. The third part presents an extensive comparative evaluation of the impact of the different types of mediation systems on web search, in terms of precision, recall and F-measure. The thesis concludes by identifying the contribution of the research programme and the satisfaction of the stated objectives.