Computer Sciences Commons^™

Overview Of Contrast Data Mining As A Field And Preview Of An Upcoming Book, Guozhu Dong, James Bailey

Kno.e.sis Publications

This report provides an overview of the field of contrast data mining and its applications, and offers a preview of an upcoming book on the topic. The importance of contrasting is discussed and a brief survey is given covering the following topics: general definitions and terminology for contrast patterns, representative contrast pattern mining algorithms, applications of contrast mining for fundamental data mining tasks such as classification and clustering, applications of contrast mining in bioinformatics, medicine, blog analysis, image analysis and subgroup mining, results on contrast based dataset similarity measure, and on analyzing item interaction in contrast patterns, and open research …

Computing Inconsistency Measure Based On Paraconsistent Semantics, Pascal Hitzler, Yue Ma, Guilin Qi

Computer Science and Engineering Faculty Publications

Measuring inconsistency in knowledge bases has been recognized as an important problem in several research areas. Many methods have been proposed to solve this problem and a main class of them is based on some kind of paraconsistent semantics. However, existing methods suffer from two limitations: (i) they are mostly restricted to propositional knowledge bases; (ii) very few of them discuss computational aspects of computing inconsistency measures. In this article, we try to solve these two limitations by exploring algorithms for computing an inconsistency measure of first-order knowledge bases. After introducing a four-valued semantics for first-order logic, we define an …

Stability And Classification Performance Of Feature Selection Techniques, Huanjing Wang, Taghi Khoshgoftaar, Qianhui Liang

Computer Science Faculty Publications

Feature selection techniques can be evaluated based on either model performance or the stability (robustness) of the technique. The ideal situation is to choose a feature selec- tion technique that is robust to change, while also ensuring that models built with the selected features perform well. One domain where feature selection is especially important is software defect prediction, where large numbers of met- rics collected from previous software projects are used to help engineers focus their efforts on the most faulty mod- ules. This study presents a comprehensive empirical ex- amination of seven filter-based feature ranking techniques (rankers) applied to …

Quantifying Computer Network Security, Ian Burchett

Masters Theses & Specialist Projects

Simplifying network security data to the point that it is readily accessible and usable by a wider audience is increasingly becoming important, as networks become larger and security conditions and threats become more dynamic and complex, requiring a broader and more varied security staff makeup. With the need for a simple metric to quantify the security level on a network, this thesis proposes: simplify a network’s security risk level into a simple metric. Methods for this simplification of an entire network’s security level are conducted on several characteristic networks. Identification of computer network port vulnerabilities from NIST’s Network Vulnerability Database …

A Secure And Distributed Framework To Identify And Share Needed Information, Wei Jiang, Bharath K. Samanthula

Computer Science Faculty Research & Creative Works

Information analysis and communication play significant roles in decision making, especially in battle grounds and situations where national security is under threat. in many situations when information in consideration is sensitive/confidential, it is in our best interests to analyze and share only needed information to minimize the potential of security breach regarding other irrelevant but sensitive information. Thus, the goal of this paper is to investigate the required methodologies and propose an advanced communication framework that enables different entities in distributed environments to identify, share and analyze only needed information, without disclosing other unwanted but sensitive information. Such a framework …

Scaling Laws Of Key Predistribution Protocols In Wireless Sensor Networks, Wenjun Gu, Sriram Chellappan, Xiaole Bai, Honggang Wang

Computer Science Faculty Research & Creative Works

Many key predistribution (KP) protocols have been proposed and are well accepted in randomly deployed wireless sensor networks (WSNs). Being distributed and localized, they are perceived to be scalable as node density and network dimension increase. While it is true in terms of communication/computation overhead, their scalability in terms of security performance is unclear. in this paper, we conduct a detailed study on this issue. in particular, we define a new metric called Resilient Connectivity (RC) to quantify security performance in WSNs. We then conduct a detailed analytical investigation on how KP protocols scale with respect to node density and …

Exploring Caching For Efficient Collection Operations, Swetha Surapaneni, Venkata Krishna Suhas Nerella, Sanjay K. Madria, Thomas Weigert

Computer Science Faculty Research & Creative Works

Many useful programs operate on collection types. Extensive libraries are available in many programming languages, such as the C++ Standard Template Library, which make programming with collections convenient. Extending programming languages to provide collection queries as first class constructs in the language would not only allow programmers to write queries explicitly in their programs but it would also allow compilers to leverage the wealth of experience available from the database domain to optimize such queries. This paper describes an approach to reducing the run time of programs involving explicit collection queries by leveraging a cache to store previously computed results. …

Object Recognition Based On Shape And Function, Akihiro Eguchi

Computer Science and Computer Engineering Undergraduate Honors Theses

This thesis explores a new approach to computational object recognition by borrowing an idea from child language acquisition studies in developmental psychology. Whereas previous image recognition research used shape to recognize and label a target object, the model proposed in this thesis also uses the function of the object resulting in a more accurate recognition. This thesis makes use of new gaming technology, Microsoft’s Kinect, in implementing the proposed new object recognition model. A demonstration of the model developed in this project properly infers different names for similarly shaped objects and the same name for differently shaped objects.

Sensitivity Of The Frozen/Melted Basal Boundary To Perturbations Of Basal Traction And Geothermal Heat Flux : Isunnguata Sermia, Western Greenland, Douglas John Brinkerhoff, Toby Warren Meierbachtol, Jesse Johnson, Joel T. Harper

Computer Science Faculty Publications

A full-stress, thermomechanically coupled, numerical model is used to explore the interaction between basal thermal conditions and motion of a terrestrially terminating section of the west Greenland ice sheet. The model domain is a two-dimensional flowline profile extending from the ice divide to the margin. We use data-assimilation techniques based on the adjoint model in order to optimize the basal traction field, minimizing the difference between modeled and observed surface velocities. We monitor the sensitivity of the frozen/melted boundary (FMB) to changes in prescribed geothermal heat flux and sliding speed by applying perturbations to each of these parameters. The FMB …

Modeling 5 Years Of Subglacial Lake Activity In The Macayeal Ice Stream (Antarctica) Catchment Through Assimilation Of Icesat Laser Altimetry, Sasha P. Carter, Helen A. Fricker, Donald D. Blankenship, Jesse V. Johnson, William Lipscomb, Stephen F. Price, Duncan A. Young

Computer Science Faculty Publications

Subglacial lakes beneath Antarctica’s fast-moving ice streams are known to undergo ~1km³ volume changes on annual timescales. Focusing on the MacAyeal Ice Stream (MacIS) lake system, we create a simple model for the response of subglacial water distribution to lake discharge events through assimilation of lake volume changes estimated from Ice, Cloud and land Elevation Satellite (ICESat) laser altimetry. We construct a steady-state water transport model in which known subglacial lakes are treated as either sinks or sources depending on the ICESat-derived filling or drainingrates. The modeled volume change rates of five large subglacial lakes in the downstream portion …

Personality Dimensions And Temperaments Of Engineering Professors And Students – A Survey, Arif Raza, Zaka Ulmustafa, Luiz Fernando Capretz

Electrical and Computer Engineering Publications

This research work aims to study personality profiles and temperaments of Pakistani software engineering professors and students. In this survey we have collected personality profiles of 18 professors and 92 software engineering students. According to the Myers-Briggs Type Indicator (MBTI) instrument, the most prominent personality type among professors as well as among students is a combination of Introversion, Sensing, Thinking, and Judging (ISTJ). The study shows ITs (Introverts and Thinking) and IJs (Introverts and Judging) are the leading temperaments among the professors. About the students’ data, the results of the study indicate SJs (Sensing and Judging) and ISs (Introverts and …

The R Journal (December 2011) 3(2): Complete Issue, The R Foundation

The R Journal

Contributed Research Articles

Creating and Deploying an Application with (R)Excel and R, Thomas Baier, Erich Neuwirth, and Michele De Meo

glm2: Fitting Generalized Linear Models with Convergence Problems, Ian C. Marschner

Implementing the Compendium Concept with Sweave and DOCSTRIP, Michael Lundholm

Watch Your Spelling! Kurt Hornik and Duncan Murdoch

Ckmeans.1d.dp: Optimal k-means Clustering in One Dimension by Dynamic Programming, Haizhou Wang and Mingzhou Song

Nonparametric Goodness-of-Fit Tests for Discrete Null Distributions, Taylor B. Arnold and John W. Emerson

Using the Google Visualisation API with R, Markus Gesmann and Diego de Castillo

GrapheR: a Multiplatform GUI for Drawing Customizable Graphs in …

Personalized Searching For Web Service Using User Interests, Rong Hu, Wanchun Dou, Xiaoqing Frank Liu, Jianxun Liu

Computer Science Faculty Research & Creative Works

Users usually have different prospective even they input a same keyword to search Web services. It is a challenge to personalize web service search engine as more and more keyword-like Web services becoming available on Internet. User interest plays an important role in personalizing search result. Therefore, through interest extraction, Web service search engine is personalized. at last, an experiment is presented to demonstrate the feasibility of the method. © 2011 IEEE.

Wsrank: A Method For Web Service Ranking In Cloud Environment, Rong Hu, Wanchun Dou, Xiaoqing Frank Liu, Jianxun Liu

Computer Science Faculty Research & Creative Works

For services that have similar functionalities, if they are published by different cloud platforms, it is a challenge to evaluate them, for satisfying different end users' personal preferences. in view of this challenge, a method for web service ranking, named WSRank, is investigated in cloud environment in this paper. It aims at ranking different Web services published by different cloud platforms, taking advantage of Page Rank principle. at last, a case study and experiment are presented to demonstrate the feasibility of the method. © 2011 IEEE.

Middleware For Supporting Content Sharing In Dynamic Networks, Mohan Kumar, Sharma Chakravarthy, Sanjay Kumar Madria, Mark Linderman, Waseem Naqvi

Computer Science Faculty Research & Creative Works

This paper proposes a middleware architecture to enable seamless content sharing in highly dynamic networks, such as those involving a number of aerial unmanned autonomous vehicles (UAVs), which are characterized by constantly changing connectivity and network topology. the proposed middleware maintains up-to-date network status as well as metadata information about data objects in the network nodes, masking the content sharing applications from the network dynamics and faults. Mechanisms for node discovery, message passing, and network management are developed. a strategy is also proposed for efficient replication to ensure availability of data in highly mobile environments. the middleware includes a message …

Outlier Detection Using Modified-Ranks And Other Variants, Huaming Huang, Kishan Mehrotra, Chilukuri K. Mohan

Electrical Engineering and Computer Science - Technical Reports

Rank based algorithms provide a promising approach for outlier detection, but currently used rank-based measures of outlier detection suffer from two deficiencies: first they take a large value from an object whose density is high even though the object may not be an outlier and second the distance between the object and its nearest cluster plays a mild role though its rank with respect to its neighbor. To correct for these deficiencies we introduce the concept of modified-rank and propose new algorithms for outlier detection based on this concept.

Data-Intensive Computing For Bioinformatics Using Virtualization Technologies And Hpc Infrastructures, Pengfei Xuan

All Theses

The bioinformatics applications often involve many computational components and massive data sets, which are very difficult to be deployed on a single computing machine. In this thesis, we designed a data-intensive computing platform for bioinformatics applications using virtualization technologies and high performance computing (HPC) infrastructures with the concept of multi-tier architecture, which can seamlessly integrate the web user interface (presentation tier), scientific workflow (logic tier) and computing infrastructure (data/computing tier). We demonstrated our platform on two bioinformatics projects. First, we redesigned and deployed the cotton marker database (CMD) (http://www.cottonmarker.org), a centralized web portal in the cotton research community, using the …

Enhancing Inettest By Improving The Programming Question And Group Grading, Sushil Dosi

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

This report describes an improvement to the Utah State University iNetTest testing system. The iNetTest system allows instructors and/or students to: • Create/take tests with rich sets of question types (multiple choice, essay, true/false, computational programming question, etc.); • Monitor the test takers for cheating; • Auto-grade for many types of questions, as well as group grade for all question types; and • Send scores to students via either email or SMS. Specifically, this report discusses the design and development of an improved computational programming question for the iNetTest system. For programming questions, iNetTest allows for the use of various …

Chernoff Dimensionality Reduction-Where Fisher Meets Fkt, Jing Peng, Guna Seetharaman, Wei Fan, Stefan Robila, Aparna Varde

Department of Computer Science Faculty Scholarship and Creative Works

Well known linear discriminant analysis (LDA) based on the Fisher criterion is incapable of dealing with heteroscedasticity in data. However, in many practical applications we often encounter heteroscedastic data, i.e., within-class scatter matrices can not be expected to be equal. A technique based on the Chernoff criterion for linear dimensionality reduction has been proposed recently. The technique extends well-known Fisher's LDA and is capable of exploiting information about heteroscedasticity in the data. While the Chernoff criterion has been shown to outperform the Fisher's, a clear understanding of its exact behavior is lacking. In addition, the criterion, as introduced, is rather …

Rainbow: An R Package For Visualizing Functional Time Series, Han Lin Shang

The R Journal

Recent advances in computer technology have tremendously increased the use of functional data, whose graphical representation can be infinite-dimensional curves, images or shapes. This article describes four methods for visualizing functional time series using an R add-on package. These methods are demonstrated using age-specific Australian fertility data from 1921 to 2006 and monthly sea surface temperatures from January 1950 to December 2006.

Using The Google Visualisation Api With R, Markus Gesmann, Diego De Castillo

The R Journal

The googleVis package provides an interface between R and the Google Visualisation API to create interactive charts which can be embedded into web pages. The best known of these charts is probably the Motion Chart, popularised by Hans Rosling in his TED talks. With the googleVis package users can easily create web pages with interactive charts based on R data frames and display them either via the local R HTTPhelp server or within their own sites.

Ckmeans.1d.Dp: Optimal K-Means Clustering In One Dimension By Dynamic Programming, Haizhou Wang, Mingzhou Song

The R Journal

The heuristic k-means algorithm, widely used for cluster analysis, does not guarantee optimality. We developed a dynamic programming algorithm for optimal one-dimensional clustering. The algorithm is implemented as an R package called Ckmeans.1d.dp. We demonstrate its ad vantage in optimality and runtime over the standard iterative k-means algorithm.

Implementing The Compendium Concept With Sweave And Docstrip, Michael Lundholm

The R Journal

This article suggests an implementation of the compendium concept by combining Sweave and the LATEX literate programming environment DOCSTRIP.

Content-Based Social Network Analysis Of Mailing Lists, Angela Bohn, Ingo Feinerer, Kurt Hornik, Patrick Mair

The R Journal

Social Network Analysis (SNA) provides tools to examine relationships between people. Text Mining (TM) allows capturing the text they produce in Web 2.0 applications, for example, however it neglects their social structure. This paper applies an approach to combine the two methods named “content-based SNA”. Using the R mailing lists, R-help and R-devel, we show how this combination can be used to describe people’s interests and to find out if authors who have similar interests actually communicate. We find that the expected positive relationship between sharing interests and communicating gets stronger as the centrality scores of authors in the communication …

Watch Your Spelling!, Kurt Hornik, Duncan Murdoch

The R Journal

We discuss the facilities in base R for spell checking via Aspell, Hunspell or Ispell, which are useful in particular for conveniently checking the spelling of natural language texts in package Rd files and vignettes. Spell checking performance is illustrated using the Rd files in package stats. This example clearly indicates the need for a domain-specific statistical dictionary. We analyze the results of spell checking all Rd files in all CRAN packages and show how these can be employed for building such a dictionary.

Nonparametric Goodness-Of-Fit Tests For Discrete Null Distributions, Taylor B. Arnold, John W. Emerson

The R Journal

Methodology extending nonparametric goodness-of-fit tests to discrete null distributions has existed for several decades. However, modern statistical software has generally failed to provide this methodology to users. We offer a revision of R’s ks.test() function and a new cvm.test() function that fill this need in the R language for two of the most popular nonparametric goodness-of-fit tests. This paper describes these contributions and provides examples of their usage. Particular attention is given to various numerical issues that arise in their implementation.

Portable C++ For R Packages, Martyn Plummer

The R Journal

Package checking errors are more common on Solaris than Linux. In many cases, these errors are due to non-portable C++ code. This article reviews some commonly recurring problems in C++ code found in R packages and suggests solutions.

Glm2: Fitting Generalized Linear Models With Convergence Problems, Ian C. Marschner

The R Journal

The R function glm uses step-halving to deal with certain types of convergence problems when using iteratively reweighted least squares to fit a generalized linear model. This works well in some circumstances but non-convergence remains a possibility, particularly with a non standard link function. In some cases this is be cause step-halving is never invoked, despite a lack of convergence. In other cases step-halving is invoked but is unable to induce convergence. One remedy is to impose a stricter form of step halving than is currently available in glm, so that the deviance is forced to decrease in every iteration. …

Creating And Deploying An Application With (R)Excel And R, Thomas Baier, Erich Neuwirth, Michele De Meo

The R Journal

We present some ways of using R in Excel and build an example application using the package rpart. Starting with simple interactive use of rpart in Excel, we eventually package the code into an Excel-based application, hiding all details (including R itself) from the end user. In the end, our application implements a service oriented architecture (SOA) with a clean separation of presentation and computation layer

Grapher: A Multiplatform Gui For Drawing Customizable Graphs In R, Maxime Hervé

The R Journal

This article presents GrapheR, a Graphical User Interface allowing the user to draw customizable and high-quality graphs without knowing any R commands. Six kinds of graph are available: histograms, box-and-whisker plots, bar plots, pie charts, curves and scatter plots. The complete process is described with the examples of a bar plot and a scatter plot illustrating the legendary puzzle of African and European swallows’ migrations.