Discriminative pattern mining in software fault detection

We present a method to enhance fault localization for software systems based on a frequent pattern mining algorithm. Our method is based on a large set of test cases for a given set of programs in which faults can be detected. The test executions are recorded as function call trees. Based on test oracles the tests can be classified into successful and failing tests. A frequent pattern mining algorithm is used to identify frequent subtrees in successful and failing test executions. This information is used to rank functions according to their likelihood of containing a fault. The ranking suggests an order in which to examine the functions during fault analysis. We validate our approach experimentally using a subset of Siemens benchmark programs.

Why high seed densities within buried mesh bags may overestimate depletion rates of soil seed banks

1. Estimates of seed bank depletion rates are essential for modelling and management of plant populations. The seed bag burial method is often used to measure seed mortality in the soil. However, the density of seeds within seed bags is higher than densities in natural seed banks, which may elevate levels of pathogens and influence seed mortality. The aim of this study was to quantify the effects of fungi and seed density within buried mesh bags on the mortality of seeds. Striga hermonthica was chosen as the study species because it has been widely studied but different methods for measuring seed mortality in the soil have yielded contradictory estimates. 2. Seed bags were buried in soil and exhumed at regular time intervals to monitor mortality of the seeds in three field experiments during two rainy seasons. The effect of fungal activity on seed mortality was evaluated in a fungi exclusion experiment. Differences in seed-to-seed interaction were obtained by using two and four densities within the seed bags in consecutive years. Densities were created by mixing 1000 seeds with 0, 10, 100 or 1000 g of coarse sand. 3. The mortality rate was significantly lower when fungi were excluded, indicating the possible role of pathogenic fungi. 4. Decreasing the density of seeds in bags significantly reduced seed mortality, most probably because of decreased seed-to-seed contamination by pathogenic fungi. 5. Synthesis and applications. Models of plant populations in general and annual weeds in particular often use values from the literature for seed bank depletion rates. These depletion rates have often been estimated by the seed bag burial method, yet seed density within seed bags may be unrealistically high. Consequently, estimates of seed mortality rates may be too high because of an overestimation of the effects of soil or seed-borne pathogens. Species that have been classified from such studies as having short-lived seed banks may need to be re-assessed using realistic densities either within seed bags or otherwise. Similarly, models of seed bank dynamics based on such overestimated depletion rates may lead to incorrect conclusions regarding the seed banks and, perhaps, the management of weeds and rare species.

HDC poinsettia tracker: flexible graphical tracking software

Growing pot poinsettia and similar crops involves careful crop monitoring and management to ensure that height specifications are met. Graphical tracking represents a target driven approach to decision support with simple interpretation. HDC (Horticultural Development Council) Poinsettia Tracker implements a graphical track based on the Generalised Logistic Curve, similar to that of other tracking packages. Any set of curve parameters can be used to track crop progress. However, graphical tracks must be expected to be site and cultivar specific. By providing a simple Curve fitting function, growers can easily develop their own site and variety specific ideal tracks based on past records with increasing quality as more seasons' data are added. (C) 2009 Elsevier B.V. All rights reserved.

Teachers as researchers and teachers as software developers: how use-case analysis helps build better educational software

A series of government initiatives has raised both the profile of ICT in the curriculum and the expectation that high quality teaching and learning resources will be accessible across electronic networks. In order for e-learning resources such as websites to have the maximum educational impact, teachers need to be involved in their design and development. Use-case analysis provides a means of defining user requirements and other constraints in such a way that software developers can produce e-learning resources which reflect teachers' professional knowledge and support their classroom practice. It has some features in common with the participatory action research used to develop other aspects of classroom practice. Two case-studies are presented: one involves the development of an on-line resource centred on transcripts of original historical documents; the other describes how 'Learning how to Learn', a major, distributed research project funded under the ESRC Teaching and Learning Research Programme is using use-case analysis to develop web resources and services.

A simulation model of self-organising evolvability in software systems

The evolvability of a software artifact is its capacity for producing heritable or reusable variants; the inverse quality is the artifact's inertia or resistance to evolutionary change. Evolvability in software systems may arise from engineering and/or self-organising processes. We describe our 'Conditional Growth' simulation model of software evolution and show how, it can be used to investigate evolvability from a self-organisation perspective. The model is derived from the Bak-Sneppen family of 'self-organised criticality' simulations. It shows good qualitative agreement with Lehman's 'laws of software evolution' and reproduces phenomena that have been observed empirically. The model suggests interesting predictions about the dynamics of evolvability and implies that much of the observed variability in software evolution can be accounted for by comparatively simple self-organising processes.

Predicting fault inflow in iterative software development processes: an industrial evaluation

This paper addresses the need for accurate predictions on the fault inflow, i.e. the number of faults found in the consecutive project weeks, in highly iterative processes. In such processes, in contrast to waterfall-like processes, fault repair and development of new features run almost in parallel. Given accurate predictions on fault inflow, managers could dynamically re-allocate resources between these different tasks in a more adequate way. Furthermore, managers could react with process improvements when the expected fault inflow is higher than desired. This study suggests software reliability growth models (SRGMs) for predicting fault inflow. Originally developed for traditional processes, the performance of these models in highly iterative processes is investigated. Additionally, a simple linear model is developed and compared to the SRGMs. The paper provides results from applying these models on fault data from three different industrial projects. One of the key findings of this study is that some SRGMs are applicable for predicting fault inflow in highly iterative processes. Moreover, the results show that the simple linear model represents a valid alternative to the SRGMs, as it provides reasonably accurate predictions and performs better in many cases.