MuJava: an automated class mutation system

Several module and class testing techniques have been applied to object‐oriented (OO) programs, but researchers have only recently begun developing test criteria that evaluate the use of key OO features such as inheritance, polymorphism, and encapsulation. Mutation testing is a powerful testing technique for generating software tests and evaluating the quality of software. However, the cost of mutation testing has traditionally been so high that it cannot be applied without full automated tool support. This paper presents a method to reduce the execution cost of mutation testing for OO programs by using two key technologies, mutant schemata generation (MSG) and bytecode translation. This method adapts the existing MSG method for mutants that change the program behaviour and uses bytecode translation for mutants that change the program structure. A key advantage is in performance: only two compilations are required and both the compilation and execution time for each is greatly reduced. A mutation tool based on the MSG/bytecode translation method has been built and used to measure the speedup over the separate compilation approach. Experimental results show that the MSG/bytecode translation method is about five times faster than separate compilation. Copyright © 2004 John Wiley & Sons, Ltd.     

Biomechanical and physiological comparison of conventional webbing and the M83 assault vest

This study compared the effect of load distribution using two different webbing designs on oxygen consumption and running kinematics of soldiers. It was hypothesised that running with webbing that distributes the load closer to the body (M83 Assault Vest) would expend less energy compared to running with conventional webbing (CON). Seven soldiers randomly completed three treadmill trials; an unloaded VO(2)max test, and two loaded (8 kg) efficiency tests using either the M83 or CON webbing. The VO(2)max test and the loaded efficiency tests had 4-min stages at 5, 8, 10 and 12 km h(-1). Energy expenditure was measured via indirect calorimetry and video analysis was used to determine stride frequency (SF) and stride length (SL) during each trial. Participants using the M83 webbing expended significantly (p<0.05) less energy at all four running velocities compared to the CON trials. The M83 webbing resulted in smaller changes to SL and SF from the unloaded trial when compared to the CON trial. These results indicate that the M83 vest that is designed to distribute the load closer to the body may have an energy efficiency advantage over conventional webbing when soldiers are running.     

Quantifying DNS namespace influence

Name resolution using the Domain Name System (DNS) is integral to today’s Internet. The resolution of a domain name is often dependent on namespace outside the control of the domain’s owner. In this article we review the DNS protocol and several DNS server implementations. Based on our examination, we propose a formal model for analyzing the name dependencies inherent in DNS. Using our name dependency model we derive metrics to quantify the extent to which domain names affect other domain names. It is found that under certain conditions, more than half of the queries for a domain name are influenced by namespaces not expressly configured by administrators. This result serves to quantify the degree of vulnerability of DNS due to dependencies that administrators are unaware of. When we apply metrics from our model to production DNS data, we show that the set of domains whose resolution affects a given domain name is much smaller than previously thought. However, behaviors such as using cached addresses for querying authoritative servers and chaining domain name aliases increase the number and diversity of influential domains, thereby making the DNS infrastructure more vulnerable.     

Open-Source Change Logs

A recent editorial in Empirical Software Engineering suggested that open-source software projects offer a great deal of data that can be used for experimentation. These data not only include source code, but also artifacts such as defect reports and update logs. A common type of update log that experimenters may wish to investigate is the ChangeLog, which lists changes and the reasons for which they were made. ChangeLog files are created to support the development of software rather than for the needs of researchers, so questions need to be asked about the limitations of using them to support research. This paper presents evidence that the ChangeLog files provided at three open-source web sites were incomplete. We examined at least three ChangeLog files for each of three different open-source software products, namely, GNUJSP, GCC-g++, and Jikes. We developed a method for counting changes that ensures that, as far as possible, each individual ChangeLog entry is treated as a single change. For each ChangeLog file, we compared the actual changes in the source code to the entries in the ChangeLog> file and discovered significant omissions. For example, using our change-counting method, only 35 of the 93 changes in version 1.11 of Jikes appear in the ChangeLog file—that is, over 62% of the changes were not recorded there. The percentage of omissions we found ranged from 3.7 to 78.6%. These are significant omissions that should be taken into account when using ChangeLog files for research. Before using ChangeLog files as a basis for research into the development and maintenance of open-source software, experimenters should carefully check for omissions and inaccuracies.     

Creating non-minimal triangulations for use in inference in mixed stochastic/deterministic graphical models

We demonstrate that certain large-clique graph triangulations can be useful for reducing computational requirements when making queries on mixed stochastic/deterministic graphical models. This is counter to the conventional wisdom that triangulations that minimize clique size are always most desirable for use in computing queries on graphical models. Many of these large-clique triangulations are non-minimal and are thus unattainable via the popular elimination algorithm. We introduce ancestral pairs as the basis for novel triangulation heuristics and prove that no more than the addition of edges between ancestral pairs needs to be considered when searching for state space optimal triangulations in such graphs. Empirical results on random and real world graphs are given. We also present an algorithm and correctness proof for determining if a triangulation can be obtained via elimination, and we show that the decision problem associated with finding optimal state space triangulations in this mixed setting is NP-complete.     

(Approximate) Uncertain Skylines

Given a set of points with uncertain locations, we consider the problem of computing the probability of each point lying on the skyline, that is, the probability that it is not dominated by any other input point. If each point’s uncertainty is described as a probability distribution over a discrete set of locations, we improve the best known exact solution. We also suggest why we believe our solution might be optimal. Next, we describe simple, near-linear time approximation algorithms for computing the probability of each point lying on the skyline. In addition, some of our methods can be adapted to construct data structures that can efficiently determine the probability of a query point lying on the skyline.     

The effect of anisotropic reflectance on imaging spectroscopy of snow properties

How does snow's anisotropic directional reflectance affect the mapping of snow properties from imaging spectrometer data? This sensitivity study applies two spectroscopy models to synthetic images of the spectral hemispherical-directional reflectance factor (HDRF) with prescribed snow-covered area and snow grain size. The MEMSCAG model determines both sub-pixel snow-covered area and the grain size of the fractional snow cover. The Nolin/Dozier model analyzes the ice absorption feature that spans wavelength λ≅1.03 μm to estimate snow grain radius when the pixel is fully snow-covered. Retrievals of subpixel snow-covered area with MEMSCAG are progressively more sensitive to the HDRF as grain size decreases, solar zenith angle increases, and fractional snow cover increases. The model overestimates snow cover in the forward reflectance angles by up to +20% and underestimates it in the backward reflectance angles by as much as −15%. Grain size retrievals from both MEMSCAG and Nolin/Dozier are more sensitive to anisotropy as grain size and solar zenith angle increase. MEMSCAG retrievals of grain size are insensitive to snow-covered area. The largest inferred grain sizes occur around a peak in the backward reflectance angles and the smallest generally occur at the largest view angles in the forward direction. Retrievals of albedo from MEMSCAG and Nolin/Dozier are similarly sensitive to anisotropy, with albedo errors up to 5% for a 30° solar zenith angle and up to 10% at 60°. The albedo differences between the two models are less than 0.015 for all grain sizes and solar zenith angles.     

Maximum entropy direct models for speech recognition

Traditional statistical models for speech recognition have mostly been based on a Bayesian framework using generative models such as hidden Markov models (HMMs). This paper focuses on a new framework for speech recognition using maximum entropy direct modeling, where the probability of a state or word sequence given an observation sequence is computed directly from the model. In contrast to HMMs, features can be asynchronous and overlapping. This model therefore allows for the potential combination of many different types of features, which need not be statistically independent of each other. In this paper, a specific kind of direct model, the maximum entropy Markov model (MEMM), is studied. Even with conventional acoustic features, the approach already shows promising results for phone level decoding. The MEMM significantly outperforms traditional HMMs in word error rate when used as stand-alone acoustic models. Preliminary results combining the MEMM scores with HMM and language model scores show modest improvements over the best HMM speech recognizer.