Background: Computers are everywhere around us today and skills must be acquired in order for a person to use them. However, the topic of computer literacy is not researched enough to specify basic computer skills to consider an individual computer literate. This thesis will contribute to the research gap by investigating the computer skills of the workforce in the IT sector. Purpose: The purpose of this thesis is to examine the connection between computer programming and cyber security skills of the IT professional, e.g. is there a beneficial factor of this connection. Method: For this study the quantitative research method was used to gather data. The authors decided that the best way to reach their target group and answer the research questions was to conduct a survey and pose questions on the topics of computer literacy and cyber security. Conclusion: The results show that there is a statistical significance between the user’s security habits and his or her programming skills (or the absence of them). People who write code, defined as programmers, scored better on security skills survey, whereas their counterparts, the non-programmers, have some knowledge on the topic but they can never be absolutely sure of their cyber safety in the fast changing world of IT.

Algorithms that learn sets of rules describing a concept from its examples have been widely studied in machine learning and have been applied to problems in medicine, molecular biology, planning and linguistics. Many of these algorithms used a separate-and-conquer strategy, repeatedly searching for rules that explain different parts of the example set. When examples are scarce, however, it is difficult for these algorithms to evaluate the relative quality of two or more rules which fit the examples equally well. This dissertation proposes, implements and examines a general technique for modifying rule evaluation in order to improve learning performance in these situations. This approach, called Description-based Evaluation Function Transfer (DEFT), adjusts the way rules are evaluated on a target concept by taking into account the performance of similar rules on a related support task that is supplied by a domain expert. Central to this approach is a novel theory of task similarity that is defined in terms of syntactic properties of rules, called descriptions, which define what it means for rules to be similar. Each description is associated with a prior distribution over classification probabilities derived from the support examples and a rule's evaluation on a target task is combined with the relevant prior using Bayes' rule. Given some natural conditions regarding the similarity of the target and support task, it is shown that modifying rule evaluation in this way is guaranteed to improve estimates of the true classification probabilities. Algorithms to efficiently implement Deft are described, analysed and used to measure the effect these improvements have on the quality of induced theories. Empirical studies of this implementation were carried out on two artificial and two real-world domains. The results show that the inductive transfer of evaluation bias based on rule similarity is an effective and practical way to improve learning when training examples are limited.

<p>A significant part of the software development effort is spent on detecting deviations between software implementations and specifications, and subsequently locating the sources of such errors. This thesis illustrates that is possible to identify a significant number of errors during static analysis of declarative object-oriented equation-based modeling languages that are typically used for system modeling and simulation. Detecting anomalies in the source code without actually solving the underlying system of equations provides a significant advantage: a modeling error can be corrected before trying to get the model compiled or embarking on a computationally expensive symbolic or numerical solution process. The overall objective of this work is to demonstrate that debugging based on static analysis techniques can considerably improve the error location and error correcting process when modeling with equation-based languages.</p><p>A new method is proposed for debugging of over- and under-constrained systems of equations. The improved approach described in this thesis is to perform the debugging process on the flattened intermediate form of the source code and to use filtering criteria generated from program annotations and from the translation rules. Each time when an error is detected in the intermediate code and the error fixing solution is elaborated, the debugger queries for the original source code before presenting any information to the user. In this way, the user is exposed to the original language source code and not burdened with additional information from the translation process or required to inspect the intermediate code.</p><p>We present the design and implementation of debugging kernel prototypes, tightly integrated with the core of the optimizer module of a Modelica compiler, including details of the novel framework required for automatic debugging of equation-based languages.</p><p>This thesis establishes that structural static analysis performed on the underlying system of equations from object-oriented mathematical models can effectively be used to statically debug real Modelica programs. Most of our conclusions developed in this thesis are also valid for other equation-based modeling languages.</p> / Report code: LiU-Tek-Lic-2002:37.

Solution methodologies for mixed integer bilinear problems (MIBLP) are studied in this dissertation. This problem class is motivated using the pooling problem, a multicommodity network flow problem that typically arises in chemical engineering applications. Stronger than previously known results are provided to compare the strengths of polyhedral relaxations of the pooling problem. A novel single node flow relaxation, defined by a bilinear equality constraint and flow balance, is proposed for the pooling problem. Linear valid inequalities in the original space of variables are derived using a well-known technique called lifting. Mixed integer linear (MILP) formulations are proposed for generating feasible solutions to the pooling problem. Some of these MILP models arise from variable discretizations while others possess a network flow interpretation. The effectiveness of these MILP models is empirically validated on a library of medium and large-scale instances. General MIBLPs, not necessarily pooling problems, are solved using extended MILP reformulations. The reformulation is obtained by writing binary representation for each general integer variable. Facet-defining inequalities are provided for the reformulation of each bilinear term. New valid inequalities are also proposed for bilinear terms with a nontrivial upper bound. The proposed reformulation and cutting planes are compared against a global solver on five different classes of MIBLPs.

The subject of this thesis is the design of a FORTRAN compiler for the PDP-11/34 minicomputer owned by the Dept. of Mathematical Sciences of Ball State University, Muncie, Indiana. At the time that this project was undertaken, only the BASIC and ASSEMBLY languages were available on this machine.The problem consisted of writing a two-pass processer. The function of the first pass was to recognize FORTRAN statements, analyze them syntactically and semantically, build symbol tables, and construct an output file containing pointers to the symbol table, flags, and operators. The second pass used the output file of the first part and the symbol tables to build ASSEMBLY language instructions equivalent to the FORTRAN instructions in function.Ball State UniversityMuncie, IN 47306

Computer-Assisted Instruction (CAI) has been in the process of development for over ten years, but the actual usage of the method is still not yet very widespread.The author, however, proposes to use this method to teach COBOL to Business Administration students as an optional course at the College of Arts, Science and Technology in Jamaica and is therefore developing this package for her creative project.The package will require in excess of 4,000 man hours to complete. However, the author will do the necessary research and implement the framework of the software along with some lessons to demonstrate the techniques which will be used throughout the software. A supporting manual is also included in the design of the package.Ball State UniversityMuncie, IN 47306

Machine learning has risen in importance across science, engineering, and business in recent years. Domain experts have begun to understand how their data analysis problems can be solved in a principled and efficient manner using methods from machine learning, with its simultaneous focus on statistical and computational concerns. Moreover, the data in many of these application domains has exploded in availability and scale, further underscoring the need for algorithms which find patterns and trends quickly and correctly. However, most people actually analyzing data today operate far from the expert level. Available statistical libraries and even textbooks contain only a finite sample of the possibilities afforded by the underlying mathematical principles. Ideally, practitioners should be able to do what machine learning experts can do--employ the fundamental principles to experiment with the practically infinite number of possible customized statistical models as well as alternative algorithms for solving them, including advanced techniques for handling massive datasets. This would lead to more accurate models, the ability in some cases to analyze data that was previously intractable, and, if the experimentation can be greatly accelerated, huge gains in human productivity. Fixing this state of affairs involves mechanizing and automating these statistical and algorithmic principles. This task has received little attention because we lack a suitable syntactic representation that is capable of specifying machine learning problems and solutions, so there is no way to encode the principles in question, which are themselves a mapping between problem and solution. This work focuses on providing the foundational layer for enabling this vision, with the thesis that such a representation is possible. We demonstrate the thesis by defining a syntactic representation of machine learning that is expressive, promotes correctness, and enables the mechanization of a wide variety of useful solution principles.

The main purposes of this study is to investigate first year calculus students&rsquo / understanding of formal limit concept and change in their understanding after limit instruction designed by the researcher based on APOS theory. The case study method was utilized to explore the research questions. The participants of the study were 25 students attending first year calculus course in Middle East Technical University in Turkey. Students attended five weeks instruction depending on APOS theory in the fall semester of 2007-2008. Limit questionnaire including open-ended questions was administered to students as a pretest and posttest to probe change in students&rsquo / understanding of limit concept. At the end of the instruction a semi-structured interview protocol developed by the researcher was administered to all of the students to explore students&rsquo / understanding of limit concept in depth. The interview results were analyzed by using APOS framework. The results of the study showed that constructed genetic decomposition was found to be compatible with student data. Moreover, limit instruction was found to play a positive role in facilitating students&rsquo / understanding of limit concept.

Support Vector Machines (SVM) have recently gained prominence in the field of machine learning and pattern classification (Vapnik, 1995, Herbrich, 2002, Scholkopf and Smola, 2002). Classification is achieved by finding a separating hyperplane in a feature space, which can be mapped back onto a non-linear surface in the input space. However, training an SVM involves solving a quadratic optimization problem, which tends to be computationally intensive. Furthermore, it can be subject to stability problems and is non-trivial to implement. This thesis proposes an fast iterative Support Vector training algorithm which overcomes some of these problems. Our algorithm, which we christen Simple SVM, works mainly for the quadratic soft margin loss (also called the l2 formulation). We also sketch an extension for the linear soft-margin loss (also called the l1 formulation). Simple SVM works by incrementally changing a candidate Support Vector set using a locally greedy approach, until the supporting hyperplane is found within a finite number of iterations. It is derived by a simple (yet computationally crucial) modification of the incremental SVM training algorithms of Cauwenberghs and Poggio (2001) which allows us to perform update operations very efficiently. Constant-time methods for initialization of the algorithm and experimental evidence for the speed of the proposed algorithm, when compared to methods such as Sequential Minimal Optimization and the Nearest Point Algorithm are given. We present results on a variety of real life datasets to validate our claims. In many real life applications, especially for the l2 formulation, the kernel matrix K є R n x n can be written as K = Z T Z + Λ , where, Z є R n x m with m << n and Λ є R n x n is diagonal with nonnegative entries. Hence the matrix K - Λ is rank-degenerate, Extending the work of Fine and Scheinberg (2001) and Gill et al. (1975) we propose an efficient factorization algorithm which can be used to find a L D LT factorization of K in 0(nm2) time. The modified factorization, after a rank one update of K, can be computed in 0(m2) time. We show how the Simple SVM algorithm can be sped up by taking advantage of this new factorization. We also demonstrate applications of our factorization to interior point methods. We show a close relation between the LDV factorization of a rectangular matrix and our LDLT factorization (Gill et al., 1975). An important feature of SVM's is that they can work with data from any input domain as long as a suitable mapping into a Hilbert space can be found, in other words, given the input data we should be able to compute a positive semi definite kernel matrix of the data (Scholkopf and Smola, 2002). In this thesis we propose kernels on a variety of discrete objects, such as strings, trees, Finite State Automata, and Pushdown Automata. We show that our kernels include as special cases the celebrated Pair-HMM kernels (Durbin et al., 1998, Watkins, 2000), the spectrum kernel (Leslie et al., 20024, convolution kernels for NLP (Collins and Duffy, 2001), graph diffusion kernels (Kondor and Lafferty, 2002) and various other string-matching kernels. Because of their widespread applications in bio-informatics and web document based algorithms, string kernels are of special practical importance. By intelligently using the matching statistics algorithm of Chang and Lawler (1994), we propose, perhaps, the first ever algorithm to compute string kernels in linear time. This obviates dynamic programming with quadratic time complexity and makes string kernels a viable alternative for the practitioner. We also propose extensions of our string kernels to compute kernels on trees efficiently. This thesis presents a linear time algorithm for ordered trees and a log-linear time algorithm for unordered trees. In general, SVM's require time proportional to the number of Support Vectors for prediction. In case the dataset is noisy a large fraction of the data points become Support Vectors and thus time required for prediction increases. But, in many applications like search engines or web document retrieval, the dataset is noisy, yet, the speed of prediction is critical. We propose a method for string kernels by which the prediction time can be reduced to linear in the length of the sequence to be classified, regardless of the number of Support Vectors. We achieve this by using a weighted version of our string kernel algorithm. We explore the relationship between dynamic systems and kernels. We define kernels on various kinds of dynamic systems including Markov chains (both discrete and continuous), diffusion processes on graphs and Markov chains, Finite State Automata, various linear time-invariant systems etc Trajectories arc used to define kernels introduced on initial conditions lying underlying dynamic system. The same idea is extended to define Kernels on a. dynamic system with respect to a set of initial conditions. This framework leads to a large number of novel kernels and also generalize many previously proposed kernels. Lack of adequate training data is a problem which plagues classifiers. We propose n new method to generate virtual training samples in the case of handwritten digit data. Our method uses the two dimensional suffix tree representation of a set of matrices to encode an exponential number of virtual samples in linear space thus leading to an increase in classification accuracy. This in turn, leads us naturally to a, compact data dependent representation of a test pattern which we call the description tree. We propose a new kernel for images and demonstrate a quadratic time algorithm for computing it by wing the suffix tree representation of an image. We also describe a method to reduce the prediction time to quadratic in the size of the test image by using techniques similar to those used for string kernels.

Thesis (Ph. D.)--University of California, Riverside, 2009. / Includes abstract. Title from first page of PDF file (viewed March 11, 2010). Available via ProQuest Digital Dissertations. Includes bibliographical references (p. 170-172). Also issued in print.