UNT Theses and Dissertations - 211 Matching Results

Search Results

Increasing Telecommunications Channel Capacity: Impacts on Firm Profitability

Description: In calling for the deployment of high-capacity telecommunications infrastructures, the Clinton Administration is relying on market forces to drive demand toward self-sustaining development. There is little doubt that many firms will embrace the new telecommunications services for a variety of reasons including market differentiation, vertical market integration, and other organization-specific factors. However, there is little evidence at the firm level that adopting the use of increased-capacity telecommunications technologies is associated with improvements in firm profitability. This study seeks to identify the presence of impacts on firm income that can be associated with the adoption of T1 telecommunications services.
Date: August 1997
Creator: Clower, Terry L.
Partner: UNT Libraries

An Analysis of the Ability of an Instrument to Measure Quality of Library Service and Library Success

Description: This study consisted of an examination of how service quality should be measured within libraries and how library service quality relates to library success. A modified version of the SERVQUAL instrument was evaluated to determine how effectively it measures library service quality. Instruments designed to measure information center success and information system success were evaluated to determine how effectively they measure library success and how they relate to SERVQUAL. A model of library success was developed to examine how library service quality relates to other variables associated with library success. Responses from 385 end users at two U.S. Army Corps of Engineers libraries were obtained through a mail survey. Results indicate that library service quality is best measured with a performance-based version of SERVQUAL, and that measuring importance may be as critical as measuring expectations for management purposes. Results also indicate that library service quality is an important factor in library success and that library success is best measured with a combination of SERVQUAL and library success instruments. The findings have implications for the development of new instruments to more effectively measure library service quality and library success as well as for the development of new models of library service quality and library success.
Date: December 1999
Creator: Landrum, Hollis T.
Partner: UNT Libraries

A multi-dimensional entropy model of jazz improvisation for music information retrieval.

Description: Jazz improvisation provides a case context for examining information in music; entropy provides a means for representing music for retrieval. Entropy measures are shown to distinguish between different improvisations on the same theme, thus demonstrating their potential for representing jazz information for analysis and retrieval. The calculated entropy measures are calibrated against human representation by means of a case study of an advanced jazz improvisation course, in which synonyms for "entropy" are frequently used by the instructor. The data sets are examined for insights in music information retrieval, music information behavior, and music representation.
Date: December 2005
Creator: Simon, Scott J.
Partner: UNT Libraries

A Conceptual Map for Understanding the Terrorist Recruitment Process: Observation and Analysis of Turkish Hezbollah Terrorist Organizations.

Description: Terrorism is a historical problem; however, it becomes one of the biggest problems in 21st century. September 11 and the following Madrid, Istanbul and London attacks showed that it is the most significant problem threatening world peace and security. Governments have started to deal with terrorism by improving security measurements and making new investments to stop terrorism. Most of the governments' and scholars' focus is on immediate threats and causes of terrorism, instead of looking at long-term solutions such as root causes and underlying reasons of terrorism, and the recruitment style of terrorist organizations If terrorist recruitment does not stop, then it is safe to say terrorist activities cannot be stopped. This study focused on the recruitment process by observing two different terrorist organizations, DHKP/C and Turkish Hezbollah. The researcher brings 13 years of field experience and first-person data gathered from inside the terrorist organizations. The research questions of this study were: (i) How can an individual be prevented from joining or carrying out terrorist activities?; (ii) What factors are correlated with joining a terrorist organization?; (iii) What are the recruitment processes of the DHKP/C, PKK, and Turkish Hezbollah?; (iv) Is there any common process of being a member of these three terrorist organizations?; and (v) What are the similarities and differences these terrorist organizations? As a result of this analysis, a terrorist recruitment process map was created. With the help of this map, social organizations such as family and schools may be able to identify ways to prevent individuals from joining terrorist organizations. Also, this map will also be helpful for government organizations such as counterterrorism and intelligence to achieve the same goal.
Date: August 2007
Creator: Teymur, Samih
Partner: UNT Libraries

Constraints on Adoption of Innovations: Internet Availability in the Developing World.

Description: In a world that is increasingly united in time and distance, I examine why the world is increasingly divided socially, economically, and digitally. Using data for 35 variables from 93 countries, I separate the countries into groups of 31 each by gross domestic product per capita. These groups of developed, lesser developed and least developed countries are used in comparative analysis. Through a review of relevant literature and tests of bivariate correlation, I select eight key variables that are significantly related to information communication technology development and to human development. For this research, adoption of the Internet in the developing world is the innovation of particular interest. Thus, for comparative purposes, I chose Internet Users per 1000 persons per country and the Human Development Index as the dependent variables upon which the independent variables are regressed. Although small in numbers among the least developed countries, I find Internet Users as the most powerful influence on human development for the poorest countries. The research focuses on key obstacles as well as variables of opportunity for Internet usage in developing countries. The greatest obstacles are in fact related to Internet availability and the cost/need ratio for infrastructure expansion. However, innovations for expanded Internet usage in developing countries are expected to show positive results for increased Internet usage, as well as for greater human development and human capital. In addition to the diffusion of innovations in terms of the Internet, the diffusion of cultures through migration is also discussed in terms of the effect on social capital and the drain on human capital from developing countries.
Access: This item is restricted to the UNT Community Members at a UNT Libraries Location.
Date: December 2006
Creator: Stedman, Joseph B.
Partner: UNT Libraries

Detecting the Presence of Disease by Unifying Two Methods of Remote Sensing.

Description: There is currently no effective tool available to quickly and economically measure a change in landmass in the setting of biomedical professionals and environmental specialists. The purpose of this study is to structure and demonstrate a statistical change-detection method using remotely sensed data that can detect the presence of an infectious land borne disease. Data sources included the Texas Department of Health database, which provided the types of infectious land borne diseases and indicated the geographical area to study. Methods of data collection included the gathering of images produced by digital orthophoto quadrangle and aerial videography and Landsat. Also, a method was developed to identify statistically the severity of changes of the landmass over a three-year period. Data analysis included using a unique statistical detection procedure to measure the severity of change in landmass when a disease was not present and when the disease was present. The statistical detection method was applied to two different remotely sensed platform types and again to two like remotely sensed platform types. The results indicated that when the statistical change detection method was used for two different types of remote sensing mediums (i.e.-digital orthophoto quadrangle and aerial videography), the results were negative due to skewed and unreliable data. However, when two like remote sensing mediums were used (i.e.- videography to videography and Landsat to Landsat) the results were positive and the data were reliable.
Date: May 2002
Creator: Reames, Steve
Partner: UNT Libraries

Coyote Ugly Librarian: A Participant Observer Examination of Lnowledge Construction in Reality TV.

Description: Reality TV is the most popular genre of television programming today. The number of reality television shows has grown exponentially over the last fifteen years since the premier of The Real World in 1992. Although reality TV uses styles similar to those used in documentary film, the “reality” of the shows is questioned by critics and viewers alike. The current study focuses on the “reality” that is presented to viewers and how that “reality” is created and may differ from what the participants of the shows experience. I appeared on two reality shows, Faking It and That's Clever, and learned a great deal as a participant observer. Within the study, I outline my experience and demonstrate how editing changed the reality I experienced into what was presented to the viewers. O'Connor's (1996) representation context web serves as a model for the realities created through reality television. People derive various benefits from watching reality TV. Besides the obvious entertainment value of reality TV, viewers also gather information via this type of programming. Viewers want to see real people on television reacting to unusual circumstances without the use of scripts. By surveying reality TV show viewers and participants, this study gives insight into how real the viewers believe the shows are and how authentic they actually are. If these shows are presented as reality, viewers are probably taking what they see as historical fact. The results of the study indicate more must be done so that the “reality” of reality TV does not misinform viewers.
Date: May 2007
Creator: Holmes, Haley K.
Partner: UNT Libraries

Discovering a Descriptive Taxonomy of Attributes of Exemplary School Library Websites

Description: This descriptive study examines effective online school library practice. A Delphi panel selected a sample of 10 exemplary sites and helped to create two research tools--taxonomies designed to analyze the features and characteristics of school library Websites. Using the expert-identified sites as a sample, a content analysis was conducted to systematically identify site features and characteristics. Anne Clyde's longitudinal content analysis of school library Websites was used as a baseline to examine trends in practice; in addition, the national guidelines document, Information Power: Building Partnerships for Learning, was examined to explore ways in which the traditional mission and roles of school library programs are currently translated online. Results indicated great variation in depth and coverage even among Websites considered exemplary. Sites in the sample are growing more interactive and student-centered, using blogs as supplemental communication strategies. Nevertheless, even these exemplary sites were slow to adopt the advances in technology to meet the learning needs and interests of young adult users. Ideally the study's findings will contribute to understanding of state-of-the-art and will serve to identify trends, as well as serving as a guide to practitioners in planning, developing, and maintaining school library Websites.
Date: August 2007
Creator: Valenza, Joyce Kasman
Partner: UNT Libraries

The Effect of Media on Citizens' Fear of Crime in Turkey.

Description: This study was conducted on-site in Istanbul, Turkey, to determine the effects that mass media has on citizens' perceptions about fear of crime, in particular, and fear, in general. Specifically, the study was designed to (1) determine the tendency of citizens' media consumption, (2) determine the level of fear of crime among Turkish citizens, (3) establish the effect of media on citizens' fear of crime, and (4) determine if gender, age, educational level, neighborhood, and monthly income have an independent effect on fear of crime. To achieve this purpose, after administering a survey in Istanbul, the researcher collected appropriate data and then utilized regression analysis to examine the relationship between media variables and fear of crime. A survey consisting of three parts was administered to 545 Turkish citizens over the age of 18 who currently reside in Istanbul, Turkey. In Part I of the survey, respondents were asked to identify their trends in relation to media consumption, and in Part II respondents were asked to report their feelings about fear of crime. Finally, Part III consisted of socio-demographic characteristics including gender, age, marital status, level of education, and income. The media variables used for this study were, general TV viewing, watching crime drama, watching TV news, listening to radio news, reading newspaper news, and reading Internet news. Regarding the independent effects of socio-demographic variables on fear of crime, only gender was found to be significantly related thereby supporting the research hypothesis. From six media variables, only watching crime drama show and reading Internet news found to be related with individuals' fear of crime; however, this relation disappeared after controlling with socio-demographic variables. In addition, no cultivation effect could be found among the sub-groups of sample.
Date: August 2009
Creator: Erdonmez, Erhan
Partner: UNT Libraries

The Effect of Personality Type on the Use of Relevance Criteria for Purposes of Selecting Information Sources.

Description: Even though information scientists generally recognize that relevance judgments are multidimensional and dynamic, there is still discussion and debate regarding the degree to which certain internal (cognition, personality) and external (situation, social relationships) factors affect the use of criteria in reaching those judgments. Much of the debate centers on the relationship of those factors to the criteria and reliable methods for measuring those relationships. This study researched the use of relevance criteria to select an information source by undergraduate students whose task it is to create a course schedule for a semester. During registration periods, when creating their semester schedules, students filled out a two-part questionnaire. After completion of the questionnaire the students completed a Myers-Briggs Type Indicator instrument in order to determine their personality type. Data was analyzed using one-way ANOVAS and Chi-Square. A positive correlation exists between personality type as expressed by the MBTI and the information source selected as most important by the subject. A correlation also exists between personality type and relevance criteria use. The correlation is stronger for some criteria than for others. Therefore, one can expect personality type to have an effect on the use of relevance criteria while selecting information sources.
Date: December 2002
Creator: Sims, Dale B.
Partner: UNT Libraries

Information Structures in Notated Music: Statistical Explorations of Composers' Performance Marks in Solo Piano Scores

Description: Written notation has a long history in many musical traditions and has been particularly important in the composition and performance of Western art music. This study adopted the conceptual view that a musical score consists of two coordinated but separate communication channels: the musical text and a collection of composer-selected performance marks that serve as an interpretive gloss on that text. Structurally, these channels are defined by largely disjoint vocabularies of symbols and words. While the sound structures represented by musical texts are well studied in music theory and analysis, the stylistic patterns of performance marks and how they acquire contextual meaning in performance is an area with fewer theoretical foundations. This quantitative research explored the possibility that composers exhibit recurring patterns in their use of performance marks. Seventeen solo piano sonatas written between 1798 and 1913 by five major composers were analyzed from modern editions by tokenizing and tabulating the types and usage frequencies of their individual performance marks without regard to the associated musical texts. Using analytic methods common in information science, the results demonstrated persistent statistical similarities among the works of each composer and differences among the work groups of different composers. Although based on a small sample, the results still offered statistical support for the existence of recurring stylistic patterns in composers' use of performance marks across their works.
Date: May 2016
Creator: Buchanan, J. Paul
Partner: UNT Libraries

Impetuses for First, Second, and Third Year Law Student Information Seeking Behavior, and Perception of Common Knowledge and Citation

Description: This dissertation examined how previous information literacy training, law student gender, age, and previously obtained education affects first, second, and third year law students selection of information sources, their understanding of common knowledge, and their decision of whether or not to give attribution to these sources. To examine these factors, this study implemented a paradigm called the principle of least effort that contended humans in general tended to complete the least amount of work possible to complete presented tasks. This study sought to discover whether law students follow this same path of completing the least amount of work possible to finish presented tasks, and whether this behavior affects information source selection, citation, and understanding of common knowledge. I performed six focus groups and crafted and disseminated an online survey to examine these factors. Via this data collection, it was discovered that law students do exhibit some differences in understanding of citation and citation behavior based on age and their year in law school. They also exhibited some differences regarding common knowledge based on their year in law school, where they received their information literacy training, and where they attend law school. Yet, no statistically significant differences were discovered regarding where one attends law school and citation and source selection. Further this study revealed law students do follow this paradigm and seek the path of least resistance to accomplish law school assignments.
Date: May 2016
Creator: Helge, Kristyn Swen
Partner: UNT Libraries

Development of an Instrument to Measure the Level of Acceptability and Tolerability of Cyber Aggression: Mixed-Methods Research on Saudi Arabian Social Media Users

Description: Cyber aggression came about as a result of advances in information communication technology and the aggressive usage of the technology in real life. Cyber aggression can take on many forms and facets. However, the main focus of this study is cyberbullying and cyberstalking through information sharing practices that might constitute digital aggressive acts. Human aggression has been extensively investigated. Studies focusing on understanding the causes and effects that can lead to physical and digital aggression have shown the prevalence of cyber aggression in different settings. Moreover, these studies have shown strong relationship between cyber aggression and the physiological and physical trauma on both perpetrators and their victims. Nevertheless, the literature shows a lack of studies that could measure the level of acceptance and tolerance of these dangerous digital acts. This study is divided into two main stages; Stage one is a qualitative pilot study carried out to explore the concept of cyber aggression and its existence in Saudi Arabia. In-depth interviews were conducted with 14 Saudi social media users to collect understanding and meanings of cyber aggression. The researcher followed the Colaizzi’s methods to analyze the descriptive data. A proposed model was generated to describe cyber aggression in social media applications. The results showed that there is a level of acceptance to some cyber aggression acts due to a number of factors. The second stage of the study is focused on developing scales with reliable items that could determine acceptability and tolerability of cyber aggression. In this second stage, the researcher used the factors discovered during the first stage as source to create the scales’ items. The proposed methods and scales were analyzed and tested to increase reliability as indicated by the Cronbach’s Alpha value. The scales were designed to measure how acceptable and tolerable is cyber-bullying, cyber-stalking in Saudi ...
Access: This item is restricted to UNT Community Members. Login required if off-campus.
Date: May 2016
Creator: Albar, Ali A
Partner: UNT Libraries

Poststructuralist Critical Rhetorical Analysis as a Problem Analysis Tool: A Case Study of Information Impact in Denton’s Hydraulic Fracturing Debate

Description: Energy and the natural environment are central concerns among stakeholders across the globe. Decisions on this scale often require interaction among a myriad of institutions and individuals who navigate a complex variety of challenges. In Denton, Texas in 2014, voters were asked to make such a decision when tasked with a referendum to determine whether the city would continue to allow hydraulic fracturing activity within its borders. For social scientists, this situation requires further analysis in an effort to better understand how and why individuals make the decisions they do. One possible approach for exploring this process is a method of poststructuralist critical rhetorical analysis, which is concerned with how individuals’ identities change through interaction with institutions. This study reflects upon the texts themselves through a poststructuralist critical rhetorical analysis of images employed by those in favor of and those against Denton’s ban on hydraulic fracturing in an attempt to identify images that alter the grid of intelligibility for the audience. The paper includes deliberation about the relative merits, subsequent disadvantages, and possible questions for further study as they relate to the theoretical implications of critical rhetorical analysis as information science. Ultimately, the study identifies poststructuralist critical rhetorical analysis as a method for solving information science problems in a way that considers closely the way identity is shaped through engagement with institutions.
Date: May 2016
Creator: Sykes, Jason
Partner: UNT Libraries

Conversational Use of Photographic Images on Facebook: Modeling Visual Thinking on Social Media

Description: Modeling the "thick description" of photographs began at the intersection of personal and institutional descriptions. Comparing institutional descriptions of particular photos that were also used in personal online conversations was the initial phase. Analyzing conversations that started with a photographic image from the collection of the Library of Congress (LC) or the collection of the Manchester Historic Association (MHA) provided insights into how cultural heritage institutions could enrich the description of photographs by using informal descriptions such as those applied by Facebook users. Taking photos of family members, friends, places, and interesting objects is something people do often in their daily lives. Some photographic images are stored, and some are shared with others in gatherings, occasions, and holidays. Face-to-face conversations about remembering some of the details of photographs and the event they record are themselves rarely recorded. Digital cameras make it easy to share personal photos in Web conversations and to duplicate old photos and share them on the Internet. The World Wide Web even makes it simple to insert images from cultural heritage institutions in order to enhance conversations. Images have been used as tokens within conversations along with the sharing of information and background knowledge about them. The recorded knowledge from conversations using photographic images on Social Media (SM) has resulted in a repository of rich descriptions of photographs that often include information of a type that does not result from standard archival practices. Closed group conversations on Facebook among members of a community of interest/practice often involve the use of photographs to start conversations, convey details, and initiate story-telling about objets, events, and people. Modeling of the conversational use of photographic images on SM developed from the exploratory analyses of the historical photographic images of the Manchester, NH group on Facebook. The model was influenced by the ...
Date: May 2016
Creator: Albannai, Talal N.
Partner: UNT Libraries

Assessing Terrorist Cyber Threats: Engineering a Functional Construct

Description: Terrorist organizations and individuals make use of the Internet for supportive activities such as communication, recruiting, financing, training, and planning operations. However, little is known about the level of computer-based (“cyber”) threat such terrorist organizations and individuals pose. One step in facilitating the examination and assessment of the level of cyber threat posed by terrorist organizations and individuals is development of an assessment tool or methodology. This tool would guide intelligence collection efforts and would support and facilitate comparative assessment of the cyber threat posed by terrorist organizations and individuals through the provision of a consistent method of assessment across time, amongst organizations and individuals, and between analysts. This study leveraged the professional experience of experts to engineer a new functional construct – a structured analytical technique designed to assess the cyber threat posed by terrorist entities and individuals. The resultant instrument was a novel structured analytical construct that uses defined indicators of a terrorist organization/individual’s intent to carry out cyber attacks, and their capability to actually do so as measures of an organization/individual’s overall level of cyber threat.
Access: This item is restricted to UNT Community Members. Login required if off-campus.
Date: December 2014
Creator: Morgan, Deanne
Partner: UNT Libraries

Customers' Attitudes toward Mobile Banking Applications in Saudi Arabia

Description: Mobile banking services have changed the design and delivery of financial services and the whole banking sector. Financial service companies employ mobile banking applications as new alternative channels to increase customers' convenience and to reduce costs and maintain profitability. The primary focus of this study was to explore the Saudi bank customers' perceptions about the adoption of mobile banking applications and to test the relationships between the factors that influence mobile banking adoption as independent variables and the action to adopt them as the dependent variable. Saudi customers' perceptions were tested based on the extended versions of IDT, TAM and other diffusion of innovation theories and frameworks to generate a model of constructs that can be used to study the use and the adoption of mobile technology by users. Koenig-Lewis, Palmer, & Moll's (2010) model was used to test its constructs of (1) perceived usefulness, (2) perceived ease of use, (3) perceived compatibility, (4) perceived credibility, (5) perceived trust, (6) perceived risk, and (7) perceived cost, and these were the independent variables in current study. This study revealed a high level of adoption that 82.7% of Saudis had adopted mobile banking applications. Also, the findings of this study identified a statistically significant relationship between all of demographic differences: gender, education level, monthly income, and profession and mobile banking services among adopters and non-adopters. Seven attributes relating to the adoption of mobile banking applications were evaluated in this study to assess which variables affected Saudi banks customers in their adoption of mobile banking services. The findings indicated that the attributes that significantly affected the adoption of mobile banking applications among Saudis were perceived trust, perceived cost, and perceived risk. These three predictors, as a result, explained more than 60% of variance in intention to adopt mobile banking technology in Saudi Arabia. ...
Access: This item is restricted to UNT Community Members. Login required if off-campus.
Date: August 2016
Creator: Alshara, Mohammed Ali
Partner: UNT Libraries

Understanding the Information Seeking of Pre-Kindergarten Students: An Ethnographic Exploration of Their Seeking Behaviors in a Preschool Setting

Description: Although there has been research conducted in the area of information seeking behavior in children, the research focusing on young children, more specifically on pre-kindergarten students, is almost nonexistent. Children at this age are in the preoperational developmental stage. They tend to display curiosity about the world around them, and use other people as a means to gain the information they are seeking. Due to the insistence from President Obama to implement pre-kindergarten programs for all low and middle class children, the need to understand the cognitive, emotional, and physical needs of these children is becoming increasingly imperative. To researchers, the actions displayed by these young children on a daily basis remain vital in determining the methods by which they are categorized, studies, and even taught. This study employed Deci and Ryan's self-determination theory (SDT), Dervin's sense-making theory, Kuhlthau''s information search process model (ISP), and Shenton and Dixon's microcosmic model of information seeking via people to lay the theoretical foundational framework. This ethnographic study aimed to fill the age gap found in information seeking literature. By observing young children in the school setting, I gained insight into how these children seek information. The resulting information collected via field observations and semi-structured interviews were coded based on Shenton and Dixon's model of information seeking via people. The findings, in Chapter 5, revealed emerging codes and trends in the information seeking behaviors of pre-kindergarten students.
Date: August 2016
Creator: Stewart, Sarah Nykole
Partner: UNT Libraries

From the Outside In: A Multivariate Correlational Analysis of Effectiveness in Communities of Practice

Description: Online communities of practice (CoPs) provide social spaces for people to connect, learn, and engage with one another around shared interests and passions. CoPs are innovatively employed within industry and education for their inherent knowledge management characteristics and as a means of improving professional practice. Measuring the success of a CoP is a challenge researchers are examining through various strategies. Recent literature supports measuring community effectiveness through the perceptions of its members; however, evaluating a community by means of member perception introduces complicating factors from outside the community. In order to gain insight into the importance of external factors, this quantitative study examined the influence of factors in the professional lives of educators on their perceptions of their CoP experience. Through an empirical examination of CoPs employed to connect educators and advance their professional learning, canonical correlation analysis was used to examine correlations between factors believed to be influential on the experiences of community members.
Date: August 2016
Creator: Bomar, Shannon Hulbert
Partner: UNT Libraries

Journalist as Information Provider: Examining the One-Voice Model of a Corporate Sports Account

Description: While journalists were once viewed as gatekeepers, dispensing news and information via one-way communication channels, their role as information provider has evolved. Nowhere is this more apparent than on the social networking site Twitter, where information seekers have unprecedented access to information providers. The two-way communication that these information seekers have come to expect can be challenging for organizations such as ESPN who have multiple Twitter accounts and millions of followers. By designating one team of people as responsible for the organization's largest Twitter account, SportsCenter, ESPN has sought to establish manageable methods of interacting with this account's followers, while furthering the goals of the organization and providing sports news around the clock. This study provides a better understanding of the group responsible for ESPN's SportsCenter Twitter account: the motivation and strategies behind the group's Twitter use as well as the dynamics of this network, such as information flow and collaboration. Relying on the Information Seeking and Communication Model, this study also provides a better understanding of information exchanges with those outside the network, specifically a selection of the account's Twitter followers. Additionally, the role of journalist as information provider and certain themes that emerged from the content of the tweets are discussed. The research employed social network analysis and exploratory, descriptive case study methods. The results of this study contribute to social network and information theory as well as to journalistic and information science practice.
Date: August 2016
Creator: Norris, Tiffany D.
Partner: UNT Libraries

The Cluster Hypothesis: A Visual/Statistical Analysis

Description: By allowing judgments based on a small number of exemplar documents to be applied to a larger number of unexamined documents, clustered presentation of search results represents an intuitively attractive possibility for reducing the cognitive resource demands on human users of information retrieval systems. However, clustered presentation of search results is sensible only to the extent that naturally occurring similarity relationships among documents correspond to topically coherent clusters. The Cluster Hypothesis posits just such a systematic relationship between document similarity and topical relevance. To date, experimental validation of the Cluster Hypothesis has proved problematic, with collection-specific results both supporting and failing to support this fundamental theoretical postulate. The present study consists of two computational information visualization experiments, representing a two-tiered test of the Cluster Hypothesis under adverse conditions. Both experiments rely on multidimensionally scaled representations of interdocument similarity matrices. Experiment 1 is a term-reduction condition, in which descriptive titles are extracted from Associated Press news stories drawn from the TREC information retrieval test collection. The clustering behavior of these titles is compared to the behavior of the corresponding full text via statistical analysis of the visual characteristics of a two-dimensional similarity map. Experiment 2 is a dimensionality reduction condition, in which inter-item similarity coefficients for full text documents are scaled into a single dimension and then rendered as a two-dimensional visualization; the clustering behavior of relevant documents within these unidimensionally scaled representations is examined via visual and statistical methods. Taken as a whole, results of both experiments lend strong though not unqualified support to the Cluster Hypothesis. In Experiment 1, semantically meaningful 6.6-word document surrogates systematically conform to the predictions of the Cluster Hypothesis. In Experiment 2, the majority of the unidimensionally scaled datasets exhibit a marked nonuniformity of distribution of relevant documents, further supporting the Cluster Hypothesis. Results of ...
Access: This item is restricted to UNT Community Members. Login required if off-campus.
Date: May 2000
Creator: Sullivan, Terry
Partner: UNT Libraries

Creating a Criterion-Based Information Agent Through Data Mining for Automated Identification of Scholarly Research on the World Wide Web

Description: This dissertation creates an information agent that correctly identifies Web pages containing scholarly research approximately 96% of the time. It does this by analyzing the Web page with a set of criteria, and then uses a classification tree to arrive at a decision. The criteria were gathered from the literature on selecting print and electronic materials for academic libraries. A Delphi study was done with an international panel of librarians to expand and refine the criteria until a list of 41 operationalizable criteria was agreed upon. A Perl program was then designed to analyze a Web page and determine a numerical value for each criterion. A large collection of Web pages was gathered comprising 5,000 pages that contain the full work of scholarly research and 5,000 random pages, representative of user searches, which do not contain scholarly research. Datasets were built by running the Perl program on these Web pages. The datasets were split into model building and testing sets. Data mining was then used to create different classification models. Four techniques were used: logistic regression, nonparametric discriminant analysis, classification trees, and neural networks. The models were created with the model datasets and then tested against the test dataset. Precision and recall were used to judge the effectiveness of each model. In addition, a set of pages that were difficult to classify because of their similarity to scholarly research was gathered and classified with the models. The classification tree created the most effective classification model, with a precision ratio of 96% and a recall ratio of 95.6%. However, logistic regression created a model that was able to correctly classify more of the problematic pages. This agent can be used to create a database of scholarly research published on the Web. In addition, the technique can be used to create a ...
Date: May 2000
Creator: Nicholson, Scott
Partner: UNT Libraries

A Theory for the Measurement of Internet Information Retrieval

Description: The purpose of this study was to develop and evaluate a measurement model for Internet information retrieval strategy performance evaluation whose theoretical basis is a modification of the classical measurement model embodied in the Cranfield studies and their progeny. Though not the first, the Cranfield studies were the most influential of the early evaluation experiments. The general problem with this model was and continues to be the subjectivity of the concept of relevance. In cyberspace, information scientists are using quantitative measurement models for evaluating information retrieval performance that are based on the Cranfield model. This research modified this model by incorporating enduser relevance judgment rather than using objective relevance judgments, and by adopting a fundamental unit of measure developed for the cyberspace of Internet information retrieval rather than using recall and precision-type measures. The proposed measure, the Content-bearing Click (CBC) Ratio, was developed as a quantitative measure reflecting the performance of an Internet IR strategy. Since the hypertext "click" is common to many Internet IR strategies, it was chosen as the fundamental unit of measure rather than the "document." The CBC Ratio is a ratio of hypertext click counts that can be viewed as a false drop measure that determines the average number of irrelevant content-bearing clicks that an enduser check before retrieving relevant information. After measurement data were collected, they were used to evaluate the reliability of several methods for aggregating relevance judgments. After reliability coefficients were calculated, measurement model was used to compare web catalog and web database performance in an experimental setting. Conclusions were the reached concerning the reliability of the proposed measurement model and its ability to measure Internet IR performance, as well as implications for clinical use of the Internet and for future research in Information Science.
Date: May 1999
Creator: MacCall, Steven Leonard
Partner: UNT Libraries