Imputation techniques for multivariate missingness in software measurement data

The problem of missing values in software measurement data used in empirical analysis has led to the proposal of numerous potential solutions. Imputation procedures, for example, have been proposed to ‘fill-in’ the missing values with plausible alternatives. We present a comprehensive study of imputation techniques using real-world software measurement datasets. Two different datasets with dramatically different properties were utilized in this study, with the injection of missing values according to three different missingness mechanisms (MCAR, MAR, and NI). We consider the occurrence of missing values in multiple attributes, and compare three procedures, Bayesian multiple imputation, k Nearest Neighbor imputation, and Mean imputation. We also examine the relationship between noise in the dataset and the performance of the imputation techniques, which has not been addressed previously. Our comprehensive experiments demonstrate conclusively that Bayesian multiple imputation is an extremely effective imputation technique.

A formal framework for scenario development in support of environmental decision-making

Scenarios are possible future states of the world that represent alternative plausible conditions under different assumptions. Often, scenarios are developed in a context relevant to stakeholders involved in their applications since the evaluation of scenario outcomes and implications can enhance decision-making activities. This paper reviews the state-of-the-art of scenario development and proposes a formal approach to scenario development in environmental decision-making. The discussion of current issues in scenario studies includes advantages and obstacles in utilizing a formal scenario development framework, and the different forms of uncertainty inherent in scenario development, as well as how they should be treated. An appendix for common scenario terminology has been attached for clarity. Major recommendations for future research in this area include proper consideration of uncertainty in scenario studies in particular in relation to stakeholder relevant information, construction of scenarios that are more diverse in nature, and sharing of information and resources among the scenario development research community.     

Advances in machine learning applications for scenario intelligence: deep learning

In this paper, we discuss the inextricable link between automating training environment adaptation and deep understanding of the context surrounding specific decisions and actions executed in the performance environment. To enable deep contextual understanding, psychological measurement strategies are needed to more accurately and rapidly model the psychologically meaningful details of the trainee's interactions with events, objects, and people in the training environment. As these interactions often entail complex, nonlinear cue-action relationships, the underlying models must effectively capture the nuance, complexity, and largely intuitive nature of human decision-making. This paper discusses the promise of an emerging field of machine learning – deep neural networks – for supporting this requirement.     

The pairwise attribute noise detection algorithm

Analyzing the quality of data prior to constructing data mining models is emerging as an important issue. Algorithms for identifying noise in a given data set can provide a good measure of data quality. Considerable attention has been devoted to detecting class noise or labeling errors. In contrast, limited research work has been devoted to detecting instances with attribute noise, in part due to the difficulty of the problem. We present a novel approach for detecting instances with attribute noise and demonstrate its usefulness with case studies using two different real-world software measurement data sets. Our approach, called Pairwise Attribute Noise Detection Algorithm (PANDA), is compared with a nearest neighbor, distance-based outlier detection technique (denoted DM) investigated in related literature. Since what constitutes noise is domain specific, our case studies uses a software engineering expert to inspect the instances identified by the two approaches to determine whether they actually contain noise. It is shown that PANDA provides better noise detection performance than the DM algorithm. Jason Van Hulse is a Ph.D. candidate in the Department of Computer Science and Engineering at Florida Atlantic University. His research interests include data mining and knowledge discovery, machine learning, computational intelligence and statistics. He is a student member of the IEEE and IEEE Computer Society. He received the M.A. degree in mathematics from Stony Brook University in 2000, and is currently Director, Decision Science at First Data Corporation. Taghi M. Khoshgoftaar is a professor at the Department of Computer Science and Engineering, Florida Atlantic University, and the director of the Empirical Software Engineering and Data Mining and Machine Learning Laboratories. His research interests are in software engineering, software metrics, software reliability and quality engineering, computational intelligence, computer performance evaluation, data mining, machine learning, and statistical modeling. He has published more than 300 refereed papers in these subjects. He has been a principal investigator and project leader in a number of projects with industry, government, and other research-sponsoring agencies. He is a member of the IEEE, the IEEE Computer Society, and IEEE Reliability Society. He served as the program chair and general chair of the IEEE International Conference on Tools with Artificial Intelligence in 2004 and 2005, respectively. Also, he has served on technical program committees of various international conferences, symposia, and workshops. He has served as North American editor of the Software Quality Journal, and is on the editorial boards of the journals Empirical Software Engineering, Software Quality, and Fuzzy Systems. Haiying Huang received the M.S. degree in computer engineeringfrom Florida Atlantic University, Boca Raton, Florida, USA, in 2002. She is currently a Ph.D. candidate in the Department of Computer Science and Engineering at Florida Atlantic University. Her research interests include software engineering, computational intelligence, data mining, software measurement, software reliability, and quality engineering.     

Physical Properties of Pyridinium Fluorohydrogenate, [pyridine · H+][H2F3]−

Ionic liquids (ILs), also referred to as molten salts, have found application as electrolytes for batteries and super-capacitors, in electroplating baths, as designer solvents, and as reaction media. A few of the desired properties of a super-capacitor electrolyte are nonflammability, thermal stability, and electrochemical stability. ILs containing aromatic cations have been shown to have low viscosity which results in a high electrochemical conductivity. There is a delicate balance between increasing the thermal stability, or decreasing the melting point, and increasing the electrochemical conductivity of the IL. This study focuses on pyridinium fluorohydrogenate, [pyridine · H⁺][H₂F₃]⁻. Pyridinium fluorohydrogenate has been synthesized by the reaction of pyridine and anhydrous hydrofluoric acid. This IL has a relatively high electrical conductivity (~98 mS · cm⁻¹ at 23 °C), a wide electrochemical window, and a boiling point of 186 °C. A stable gel can also be formed by combining [pyridine · H⁺][H₂F₃]⁻ and a super absorbent polymer such as polyacrylic acid. The gel adds mechanical stability to the matrix while not greatly affecting the conductivity of the IL.     

Secure Occupancy: A New Framework for Analysing Security in Rental Housing

Unlike debates about security for owner occupiers which recognise that security is complex and multi-layered, security for renters is often presented as single dimensional and conflated with de jure security of tenure, deriving from a property rights perspective. This article proposes a broader concept of ‘secure occupancy’ to enable a more nuanced understanding of security for tenants. A new framework is developed to enable investigation of the dynamic interactions between legislation/regulation, housing market conditions, public policies and cultural norms around renting, which shape security of occupancy for tenants. The paper illustrates this approach, drawing on a study of the rental systems of nine developed countries, identifying key factors that appear to have a strong bearing on strengthening, and weakening, of security of occupancy for renter households. The article concludes that this approach has the potential to deepen understanding of security for renters and to stimulate new avenues for research.     

Effects of age, system experience, and navigation technique on driving with an advanced traveler information system

This paper explores the effects of age, system experience, and navigation technique on driving, navigation performance, and safety for drivers who used TravTek, an Advanced Traveler Information System. The first two studies investigated various route guidance configurations on the road in a specially equipped instrumented vehicle with an experimenter present. The third was a naturalistic quasi-experimental field study that collected data unobtrusively from more than 1200 TravTek rental car drivers with no in-vehicle experimenter. The results suggest that with increased experience, drivers become familiar with the system and develop strategies for substantially more efficient and safer use. The results also showed that drivers over age 65 had difficulty driving and navigating concurrently. They compensated by driving slowly and more cautiously. Despite this increased caution, older drivers made more safety-related errors than did younger drivers. The results also showed that older drivers benefited substantially from a well-designed ATIS driver interface.     

Pediatric use of intravenous amiodarone: efficacy and safety in critically ill patients from a multicenter protocol

OBJECTIVE: The purpose of this study was to analyze the efficacy and safety of intravenous amiodarone in young patients with critical, drug-resistant arrhythmias. BACKGROUND: Intravenous amiodarone has been investigated in adults since the early 1980s. Experience with the drug in young patients is limited. A larger pediatric study group was necessary to provide responsible guidelines for the drug's use before its market release. METHODS: Eight centers obtained institutional approval of a standardized protocol. Other centers were approved on a compassionate use basis after contacting the primary investigator (J.C.P). RESULTS: Forty patients were enrolled. Standard management in all failed. Many patients had early postoperative tachyarrhythmias (25 of 40), with early successful treatment in 21 (84%) of 25. Twelve patients had ventricular tachyarrhythmias: seven had successful therapy, and six died, none related to the drug. Eleven patients had atrial tachyarrhythmias: 10 of 11 had immediate success, but 3 later died. Fourteen patients had junctional ectopic tachycardia, which was treated with success (sinus rhythm or slowing, allowing pacing) in 13 of 14, with no deaths. Three other patients had supraventricular tachycardias, with success in two and no deaths. The average loading dose was 6.3 mg/kg body weight, and 50% of patients required a continuous infusion. Four patients had mild hypotension during the amiodarone bolus. One postoperative patient experienced bradycardia requiring temporary pacing. There were no proarrhythmic effects. Deaths (9 [23%] of 40) were not attributed to amiodarone. CONCLUSIONS: Intravenous amiodarone is safe and effective in most young patients with critical tachyarrhythmia. Intravenous amiodarone can be lifesaving, particularly for postoperative junctional ectopic tachycardia, when standard therapy is ineffective.     

Steady-state relative bioavailability of three oral ganciclovir dosage regimens delivering 6,000 mg/day in patients with human immunodeficiency virus

This study was designed to determine the steady-state relative bioavailability of ganciclovir after three dosage regimens designed to deliver 6,000 mg/day. The study design was an open-label, randomized, three-treatment crossover design in which 22 human immunodeficiency virus (HIV) and cytomegalovirus (CMV) seropositive patients received in random order multiple oral doses of ganciclovir 1,000 mg six times a day, 1,500 mg four times a day, and 2,000 mg three times a day. Blood samples were obtained on day 3 of each oral regimen over a 24-hour time interval. Mean steady-state average serum concentrations of ganciclovir were greater than 1.0 microgram/mL, which exceeds the median in vitro inhibitory concentration (IC50) of most CMV isolates (0.5-1.0 microgram/mL). All three regimens resulted in values for area under the concentration-time curve from 0 to 24 hours (AUC0-24) that were comparable to those seen after maintenance ganciclovir intravenous infusions of 5 mg/kg/day. The 1,000 mg six times daily regimen resulted in an AUC0-24 that was significantly higher than that of the 1,500 mg four times daily or the 2,000 mg three times daily regimens, although the differences were less than 12.5%.