Marcus Brown

User identification via keystroke characteristics of typed names using neural networks

Abstract A method for identifying computer users by analysing keystroking patterns with neural networks and a simple geometric distance is presented. A model of each users normal typing style was created and compared with later typing samples. Preliminary results demonstrate complete exclusion of imposters and a reasonably low false alarm rate when the sample text was limited to the users name.

The literate-programming paradigm

Literate programming, a technique for coding software systems that promotes readability and comprehension, is examined in detail. The current literate-programming paradigm is reviewed by looking at two sample literate programs. A critique of literate programming as it is currently used is presented, and methods for enhancing the process are explored. A number of new facilities are proposed, and restrictions on current literate-programming practices are suggested.<<ETX>>

Towards proactive computer-system forensics

We examine principles and approaches for proactive computer-system forensics. Proactive computer-system forensics is the design, construction and configuring of systems to make them most amenable to digital forensics analyses in the future. The primary goals of proactive computer-system forensics are system structuring and augmentation for automated data discovery, lead formation, and efficient data preservation. We propose: (1) using the Neyman-Pearson Lemma to proactively build online forensics tests with the best possible critical regions for hypothesis testing, and (2) using classical stopping rules for sequential hypothesis testing to determine which users are deviating from standard usage behavior and should be the focus of more investigative resources. Here the focus is on security breaches by the employees or stakeholders of an organization. The main measurements are event-driven logs of program executions.

A practical approach to user authentication

A method for user authentication is presented which analyzes keystroking data as the user types his or her name. This study utilizes the ADALINE (ADAptive LINear Element) and backpropagation neural nets to identify the typing pattern characteristic of a particular user. A simple measure of geometric distance is also used for comparison. This paper provides a brief introduction to this type of neural net. It then describes the research procedure and contrasts the initial and new results, followed by a conclusion with notes concerning future research. For an average 15-character name, a complete exclusion of imposters is obtained from a set of over 5000 imposter samples.<<ETX>>

Comparing mental models of novice programmers when using visual and command line environments

Improving the novices experience with learning to program has been an important research topic for some time. Appropriate programming environments for novices have been one research area. For example, many departments have adopted visual environments to teach programming as opposed to a command line environment at the beginning stages of a CS curriculum. The features in standard command line environments are not as assistive to programmers as visual environments. Novices must learn both language syntax and semantics while navigating the file system and compilation tools. On the other hand, visual environments can enable novices to conduct the same behavior with one or more button clicks. However, the consistency and affordance of certain features in visual environments could cause novices to develop a false perception of programming. In addition, such features may impose the inability for novices to immediately adapt to less assistive environments. In contrast, command line environments may enable novices to develop better mental models for programming because of their limited features, which could also allow them to transition to other environments much easier. A protocol analysis was conducted on seven students from a CS1 course who learned Python using either IDLE or VIM. The objective was to record and analyze each students behavior while they use the opposing environment to write a program. This study showed that students who originally used IDLE were not as equipped for transitioning to VIM as their counterparts who transitioned to IDLE.

A literate programming design language

A literate programming paradigm that provides an environment suitable for documenting and organizing the low-level design of a system is presented. It is shown that a literate program design language (PDL) strengthens the bridge design and implementation activities since the documentation generated during the design phase can be used directly as a shell for program construction. In this type of language, module names and program constructs appear as labeled modules which can be directly expanded into their equivalent code.<<ETX>>

Creating gender parity: an instruction aide's influence

The decline in the number of women in computing disciplines has been attributed to different causes, and research on the decline continues to grow. While there are numerous reasons suggested for the imbalance in these disciplines, the perceptions that women hold about their computing capabilities continue to be of interest. The current study investigates how a vicarious learning experience can be used to influence computer selfefficacy. To implement the vicarious experience, a computer-assisted peer-modeling instruction aide was developed. The study reveals that vicarious learning has a significant impact on female and male computer selfefficacy, and that the impact may be more significant for females.

Studying the Novice’s Perception of Visual Vs. Command Line Programming Tools in CS1

Determining appropriate practices for novice programming has been an important research topic for some time. Programming tools and their effect on novices has been one research area. When learning to program, novices can be exposed to either visual or command line tools. The variation of feature sets between both tool schemes could influence how a novice perceives programming. Visual tools tend to possess features that are familiar, assistive, and user-friendly for novices. Command line tools, on the other hand, usually provide basic essential features for programming. Prior studies have shown the effects of programming tools on novices while learning to program. However, most of these studies primarily focus on visual tools. This article explores visual and command line programming and their effect on novices. Our findings show that visual tools could provide a lower learning curve for novices, while command line tools could broaden their understanding of programming.

Device Fabrication for Data Storage, Semiconductor and MEMS Applications at the University of Alabama Microfabrication Facility

Over the past two years, a Class 100/1000 microfabrication facility set up at the University of Alabama for development of a meso-scale gyroscope has been turned into a central user facility for materials and device research. We will describe a wide range of processes and devices that are being researched, fabricated and optimized in this facility. These include a) a current-perpendicular-to-the-plane giant magnetoresistive (CPP-GMR) sensor, b) precisely shaped silicon tips for atom probe specimens, c) nanosphere lithography for bit-patterned graded media, d) through silicon vias (TSVs) for integration of Si and GaAs-based circuitry, and the original focus of the facility, e) a meso-scale gyroscope for general aviation. These diversely funded efforts represent research programs from the materials and electrical engineering departments, all led by female faculty members and involving several graduate and undergraduate students from under-represented groups. The UA Microfabrication Facility, is successfully incubating and supporting several different areas of funded as well as unfunded research that could lead to significant increases in research expenditures.

Comparing feature sets within visual and command line environments and their effect on novice programming (abstract only)

Improving the novices experience with programming may be key to retaining more students interested in Computer Science. Programming environments is an area that lacks comprehensive study. Some entry-level sequences prefer visual environments while others tout the benefits of command line environments. In either case, there is little more than anecdotal evidence to support either choice. We present a novel categorization of programming tools based on assistive features that predict the usefulness of programming environments for novices. This categorization is supported by a controlled study that measured the differences in engagement, efficiency, comprehension, and usability. We found that the extremes of assistive features are problematic regardless of prior programming experience.