Search Engine Model Research Articles

The Study of the Competition Model of Portal and Search Engine

This paper is to discuss the competition and cooperation of portal and search engine mainly by building stability model. We can use Lotka-Volterra mutualism model to describe and build cooperation model of portal and search engine, and we can use Lotka-Volterra predator – prey model with Logistic items to describe and build competition model of portal and search engine.

Brand Names as Keywords in Sponsored Search Advertising

The business models of major Web search engines depend on online advertising, primarily in the form of keyword advertising. In recent years, a controversy has gained notoriety worldwide, in both the international court systems and the media. It concerns a form of potential “bait and switch” advertising where a consumer, searching using the brand name of one company, is presented with an advertisement by a competitor of the searched-for brand. We refer to this practice as “piggybacking.” In the U.S. in particular, the legality of this practice, and the potential liability of the search engines for contributing to trademark infringement, is unclear. However, the eventual resolutions of the issue could significantly and negatively impact the business model of Internet search engines. In this paper, we investigate the actual prevalence of piggybacking of major brands in U.S. search engines. We submitted 100 search queries consisting of top global brand names to three major search engines. Analysis of 2,350 advertisements from search engine results pages showed that just 4 percent were triggered by competitors’ trademarked terms. There was even lower use of those trademark terms in the ad text. Thus, overall competitive piggybacking does not appear to be a deceptive or widespread phenomenon. Implications for this are discussed, and suggestions for future research are presented.

Language input architecture for converting one text form to another text form with tolerance to spelling, typographical, and conversion errors

A language input architecture converts input strings of phonetic text to an output string of language text. The language input architecture has a search engine, one or more typing models, a language model, and one or more lexicons for different languages. The typing model is configured to generate a list of probable typing candidates that may be substituted for the input string based on probabilities of how likely each of the candidate strings was incorrectly entered as the input string. The language model provides probable conversion strings for each of the typing candidates based on probabilities of how likely a probable conversion output string represents the candidate string. The search engine combines the probabilities of the typing and language models to find the most probable conversion string that represents a converted form of the input string.

A hierarchical model of a FTP search engine with applications

Because the traditional FTP search engines usually adopt centralised spiders to collect data, insufficient temporal effectiveness is their major demerit. For solving this problem, this paper presents an efficient hierarchical FTP search engine model that deploys the spider agent on the node host of some specific network for collecting file data of FTP servers. The key technologies involve a regional responsibility mechanism, a search mechanism based on the asynchronous retrieval technology and a PAT Tree storage mechanism. The simulation shows that the responding time is less than that of the traditional system and the temporal effectiveness arrives at the application level. In addition, the test results show that the architecture has a good scalability.

Effective and efficient classification on a search-engine model

Traditional document classification frameworks, which apply the learned classifier to each document in a corpus one by one, are infeasible for extremely large document corpora, like the Web or large corporate intranets. We consider the classification problem on a corpus that has been processed primarily for the purpose of searching, and thus our access to documents is solely through the inverted index of a large scale search engine. Our main goal is to build the “best” short query that characterizes a document class using operators normally available within search engines. We show that surprisingly good classification accuracy can be achieved on average over multiple classes by queries with as few as 10 terms. As part of our study, we enhance some of the feature-selection techniques that are found in the literature by forcing the inclusion of terms that are negatively correlated with the target class and by making use of term correlations; we show that both of those techniques can offer significant advantages. Moreover, we show that optimizing the efficiency of query execution by careful selection of terms can further reduce the query costs. More precisely, we show that on our set-up the best 10-term query can achieve 93% of the accuracy of the best SVM classifier (14,000 terms), and if we are willing to tolerate a reduction to 89% of the best SVM, we can build a 10-term query that can be executed more than twice as fast as the best 10-term query.

Using the repertory grid and laddering technique to determine the user's evaluative model of search engines

PurposeThe purpose of this research is to explore a method for the determination of users' representations of search engines, formed during their interaction with these systems. Determines the extent to which these elicited “mental models” indicate the system aspects of importance to the user and from this their evaluative view of these tools.Design/methodology/approachThe repertory grid technique is used to elicit a set of constructs that define facets within the mental model of an individual. A related technique of laddering then considers each of the user's constructs to determine the reasons for its importance within the user's mental model.FindingsThe model derived from the qualitative data comprises three hierarchical strata and conveys the interrelations between basic system description, evaluative description, and the key evaluations of ease, efficiency, effort and effectiveness. Two additional layers relating to the perceived process and the experience of emotion are also discussed.Research limitations/implicationsTen participants is considered to be optimum for obtaining constructs in a repertory grid, but limits the findings to the context of the user group and the systems used in this study.Originality/valueThe methodology has not previously been used to determine mental models of search engines and from these to understand users' evaluative view of systems. The resulting model of key evaluations with the conjunctions of procedural elements suggests a framework for further research to evaluate search engines from the user perspective.

A model for a SOft Fusion of Information Accesses on the web

In this paper, the definition of a Meta search engine model named SOFIA (SOft Fusion of Information Access) is proposed that applies a soft and flexible fusion of the ranked lists of documents retrieved by distinct search engines available over the Internet. The peculiarity of the fusion is that the final rank is not determined by a linear combination of the ranks in the lists, as generally happens in most metasearch engines. Instead, a linguistic quantifier modeled by an Induced Ordered Average (IOWA) operator that allows us to realize soft fusions in between that of the intersection and the union of the lists respectively expresses the fusion criterion. Flexibility is obtained by allowing user to specify his/her retrieval attitude that can be either recall or precision oriented, and by computing distinct fitness scores of the search engines based on a relevance feedback mechanism. These scores are used to define the IOWA reorder vector. In this way, the search engines with highest fitness determine more heavily the ranking of the documents in the fused list.

Using the information seeker to elicit construct models for search engine evaluation

AbstractWith the rise of interactive search engines, direct access to information is available to the end user. Evaluation of these systems is clearly important for their comparison and development. From the perspective of the user engaged with the system in some information‐seeking task, the range of relevant evaluation factors could be considerable. This article presents an investigation into the suitability of repertory grid technique eliciting a mental model of search engines in the user. Such a model comprises constructs held important to the user, which can be used to evaluate the system. The design of the repertory grid application is described and the data analyzed in two stages. The important issues in determining the suitability of the method are identified and the constructs are analyzed to determine their association, discriminatory ability, and clustering around a central overall rating. From this, it is concluded that repertory grid technique is appropriate for user‐centered determination of evaluative constructs. Further analysis of the characteristics of the resulting construct set is undertaken towards establishing an underlying user model of the search engines, and further research is identified for its use in system evaluation.

Uptake and use of electronic information services: trends in UK higher education from the JUSTEIS project

The aim of the JUSTEIS project over the first three cycles (1999‐2002) was to examine the uptake and use of electronic information services in higher education in the UK, so that planning of services could be informed by trends in usage and evidence of specific needs. The objectives were to: examine which services were used by students and academic staff; how senior library staff planned services to purchase content and support its use; and examine how library and information services promoted services through their Web pages. Results over the three years explained the growing popularity of electronic journal services, the acceptance of the search engine model for information retrieval and the important role academic staff play in the promotion of electronic information services for student learning. Conclusions and recommendations concern the need for library and information staff to make their approach to integration of information skills into the curriculum appropriate for the discipline, the type of institution, and its strategy for implementation of any virtual or managed learning environment software.

Optimal Allocation Of Heterogeneous Robots In World Wide Web Search Engines

It is a conflicting requirement to reduce network traffic and loads on Web servers caused by the robots in a World Wide Web search engine and to increase the currency of the database in the search engine. This paper addresses the problem of finding optimal allocations of a given number of robots in a search engine to heterogeneous Internet domains such that a cost function which characterizes the combined consideration of network/server loads and database currency can be minimized. The cost function and the optimization problem are formulated based on a queueing model of search engines. Our algorithm to solve the problem uses the depth-first search strategy to traverse an enumeration tree which generates all possible allocations.

Search engines and Web dynamics

In this paper we study several dimensions of Web dynamics in the context of large-scale Internet search engines. Both growth and update dynamics clearly represent big challenges for search engines. We show how the problems arise in all components of a reference search engine model. Furthermore, we use the FAST Search Engine architecture as a case study for showing some possible solutions for Web dynamics and search engines. The focus is to demonstrate solutions that work in practice for real systems. The service is running live at www.alltheweb.com and major portals worldwide with more than 30 million queries a day, about 700 million full-text documents, a crawl base of 1.8 billion documents, updated every 11 days, at a rate of 400 documents/second. We discuss future evolution of the Web, and some important issues for search engines will be scheduling and query execution as well as increasingly heterogeneous architectures to handle the dynamic Web.

Evaluating Search Engine Models for Scholarly Purposes

Evaluating Search Engine Models for Scholarly Purposes