Machine learning techniques to improve software quality

A significant proportion of the cost of software development is due to software testing and maintenance. This is in part the result of the inevitable imperfections due to human error, lack of quality during the design and coding of software, and the increasing need to reduce faults to improve customer satisfaction in a competitive marketplace. Given the cost and importance of removing errors improvements in fault detection and removal can be of significant benefit. The earlier in the development process faults can be found, the less it costs to correct them and the less likely other faults are to develop. This research aims to make the testing process more efficient and effective by identifying those software modules most likely to contain faults, allowing testing efforts to be carefully targeted. This is done with the use of machine learning algorithms which use examples of fault prone and not fault prone modules to develop predictive models of quality. In order to learn the numerical mapping between module and classification, a module is represented in terms of software metrics. A difficulty in this sort of problem is sourcing software engineering data of adequate quality. In this work, data is obtained from two sources, the NASA Metrics Data Program, and the open source Eclipse project. Feature selection before learning is applied, and in this area a number of different feature selection methods are applied to find which work best. Two machine learning algorithms are applied to the data - Naive Bayes and the Support Vector Machine - and predictive results are compared to those of previous efforts and found to be superior on selected data sets and comparable on others. In addition, a new classification method is proposed, Rank Sum, in which a ranking abstraction is laid over bin densities for each class, and a classification is determined based on the sum of ranks over features. A novel extension of this method is also described based on an observed polarising of points by class when rank sum is applied to training data to convert it into 2D rank sum space. SVM is applied to this transformed data to produce models the parameters of which can be set according to trade-off curves to obtain a particular performance trade-off.

Service-learning within higher education : rhizomatic interconnections between university and the real world

This paper discusses Service-learning within an Australian higher education context as pedagogy to teach about inclusive education. Using Deleuze and Guattari’s (1987) model of the rhizome, this study conceptualises pre-service teachers’ learning experiences as multiple, hydra and continuous. Data from reflection logs of pre-service teachers highlight how the learning experience allowed them to gain insights in knowledge as socially just, ethical and inclusive. The paper concludes by arguing the need to consider Service-learning as integral to university education for pre-service teachers.

Blackboard as an online learning environment : what do teacher education students and staff think?

Abstract: As online learning environments now have an established presence in higher education we need to ask the question: How effective are these environments for student learning? Online environments can provide a different type of learning experience than traditional face-to-face contexts (for on-campus students) or print-based materials (for distance learners). This article identifies teacher education student and staff perceptions of teaching and learning using the online learning management system, Blackboard. Perceptions of staff and students are compared and implications for teacher education staff interested in providing high quality learning environments within an online space are discussed.

Enhancing professional ethics : service-learning in teacher education

The ethical conduct of professionals has been the focus of increasing scrutiny over the past several decades as members of the public, the media, professional bodies, and legislative authorities have struggled to define ethical behaviour in times of governmental change, increasing internationalisation, globalised communications, threats of terrorism, and the challenges of developments in science and medicine (e.g., Demmke & Bossaert, 2004). National governments and transnational bodies have responded to these concerns about ethics and corruption through measures such as the United Nations Convention Against Corruption (United Nations Office on Drugs and Crime, 2004), Transparency International’s annual corruption index (2010) and Queensland’s Public Sector Ethics Act 1994 (Queensland Parliament 1994). Similarly, academic interest in ethics and its application across a range of domains(e.g., business, health care, social welfare, criminal justice, law, journalism, defence, environment, and media) has also increased. To illustrate, in 1993, a non-partisan, non-profit national umbrella organisation, the Australian Association for Professional and Applied Ethics, was formed following a conference concerned with the teaching of ethics (http://www.arts.unsw.edu.au./aapae/about_aapae/about_aapae.htm), while a recent review of the Excellence in Research for Australian rankings of national and international academic journals revealed that 16 journals related to ethics had received the top ratings of A* or A (Australian Research Council, 2009). In this chapter we examine professional ethics and argue, with specific reference to the context of pre-service teacher education, that Service-learning is one way of enhancing emerging professionals’ understanding of ethics.

Facilitating learning in a real time strategy computer game

The aim of this project was to implement a just-in-time hints help system into a real time strategy (RTS) computer game that would deliver information to the user at the time that it would be of the most benefit. The goal of this help system is to improve the user’s learning in terms of their rate of learning, retention and avoidance of stagnation. The first stage of this project was implementing a computer game to incorporate four different types of skill that the user must acquire, namely motor, perceptual, declarative knowledge and strategic. Subsequently, the just-in-time hints help system was incorporated into the game to assess the user’s knowledge and deliver hints accordingly. The final stage of the project was to test the effectiveness of this help system by conducting two phases of testing. The goal of this testing was to demonstrate an increase in the user’s assessment of the helpfulness of the system from phase one to phase two. The results of this testing showed that there was no significant difference in the user’s responses in the two phases. However, when the results were analysed with respect to several categories of hints that were identified, it became apparent that patterns in the data were beginning to emerge. The conclusions of the project were that further testing with a larger sample size would be required to provide more reliable results and that factors such as the user’s skill level and different types of goals should be taken into account.

Re-engaging adolescent students in science : an experienced teachers' classroom

Adolescents are both aware of and have the impetuous to exploit aspects of Science, Technology, Engineering and Mathematics (STEM) within their personal lives. Whether they are surfing, cycling, skateboarding or shopping, STEM concepts impact their lives. However science, mathematics, engineering and technology are still treated in the classroom as separate fragmented entities in the educational environment where most classroom talk is seemingly incomprehensible to the adolescent senses. The aim of this study was to examine the experiences of young adolescents with the aim of transforming school learning at least of science into meaningful experiences that connected with their lives using a self-study approach. Over a 12-month period, the researcher, an experienced secondary-science teacher, designed, implemented and documented a range of pedagogical practices with his Year-7 secondary science class. Data for this case study included video recordings, journals, interviews and surveys of students. By setting an environment empathetic to adolescent needs and understandings, students were able to actively explore phenomena collaboratively through developmentally appropriate experiences. Providing a more contextually relevant environment fostered meta-cognitive practices, encouraged new learning through open dialogue, multi-modal representations and assessments that contributed to building upon, re-affirming, or challenging both the students' prior learning and the teacher’s pedagogical content knowledge. A significant outcome of this study was the transformative experiences of an insider, the teacher as researcher, whose reflections provided an authentic model for reforming pedagogy in STEM classes.

Outcomes and implications of one teacher's approach to context-based science in the middle years

In Australia, there is a crisis in science education with students becoming disengaged with canonical science in the middle years of schooling. One recent initiative that aims to improve student interest and motivation without diminishing conceptual understanding is the context-based approach. Contextual units that connect the canonical science with the students’ real world of their local community have been used in the senior years but are new in the middle years. This ethnographic study explored the learning transactions that occurred in one 9th grade science class studying a context-based Environmental Science unit for 11 weeks. Outcomes of the study and implications are discussed in this paper.

Neural network learning:theoretical foundations

This important work describes recent theoretical advances in the study of artificial neural networks. It explores probabilistic models of supervised learning problems, and addresses the key statistical and computational questions. Chapters survey research on pattern classification with binary-output networks, including a discussion of the relevance of the Vapnik Chervonenkis dimension, and of estimates of the dimension for several neural network models. In addition, Anthony and Bartlett develop a model of classification by real-output networks, and demonstrate the usefulness of classification with a "large margin." The authors explain the role of scale-sensitive versions of the Vapnik Chervonenkis dimension in large margin classification, and in real prediction. Key chapters also discuss the computational complexity of neural network learning, describing a variety of hardness results, and outlining two efficient, constructive learning algorithms. The book is self-contained and accessible to researchers and graduate students in computer science, engineering, and mathematics

Learning the kernel matrix with semidefinite programming

Kernel-based learning algorithms work by embedding the data into a Euclidean space, and then searching for linear relations among the embedded data points. The embedding is performed implicitly, by specifying the inner products between each pair of points in the embedding space. This information is contained in the so-called kernel matrix, a symmetric and positive semidefinite matrix that encodes the relative positions of all points. Specifying this matrix amounts to specifying the geometry of the embedding space and inducing a notion of similarity in the input space - classical model selection problems in machine learning. In this paper we show how the kernel matrix can be learned from data via semidefinite programming (SDP) techniques. When applied to a kernel matrix associated with both training and test data this gives a powerful transductive algorithm -using the labeled part of the data one can learn an embedding also for the unlabeled part. The similarity between test points is inferred from training points and their labels. Importantly, these learning problems are convex, so we obtain a method for learning both the model class and the function without local minima. Furthermore, this approach leads directly to a convex method for learning the 2-norm soft margin parameter in support vector machines, solving an important open problem.

Multitask learning with expert advice

We consider the problem of prediction with expert advice in the setting where a forecaster is presented with several online prediction tasks. Instead of competing against the best expert separately on each task, we assume the tasks are related, and thus we expect that a few experts will perform well on the entire set of tasks. That is, our forecaster would like, on each task, to compete against the best expert chosen from a small set of experts. While we describe the “ideal” algorithm and its performance bound, we show that the computation required for this algorithm is as hard as computation of a matrix permanent. We present an efficient algorithm based on mixing priors, and prove a bound that is nearly as good for the sequential task presentation case. We also consider a harder case where the task may change arbitrarily from round to round, and we develop an efficient approximate randomized algorithm based on Markov chain Monte Carlo techniques.

A learning-based approach to reactive security

Despite the conventional wisdom that proactive security is superior to reactive security, we show that reactive security can be competitive with proactive security as long as the reactive defender learns from past attacks instead of myopically overreacting to the last attack. Our game-theoretic model follows common practice in the security literature by making worst-case assumptions about the attacker: we grant the attacker complete knowledge of the defender’s strategy and do not require the attacker to act rationally. In this model, we bound the competitive ratio between a reactive defense algorithm (which is inspired by online learning theory) and the best fixed proactive defense. Additionally, we show that, unlike proactive defenses, this reactive strategy is robust to a lack of information about the attacker’s incentives and knowledge.

A unifying view of multiple kernel learning

Recent research on multiple kernel learning has lead to a number of approaches for combining kernels in regularized risk minimization. The proposed approaches include different formulations of objectives and varying regularization strategies. In this paper we present a unifying optimization criterion for multiple kernel learning and show how existing formulations are subsumed as special cases. We also derive the criterion’s dual representation, which is suitable for general smooth optimization algorithms. Finally, we evaluate multiple kernel learning in this framework analytically using a Rademacher complexity bound on the generalization error and empirically in a set of experiments.