International Journal of Software Engineering and Knowledge Engineering

The primary goal of software quality engineering is to produce a high quality software product through the use of some specific techniques and processes. One strategy is applying data mining techniques to software metric and defect data collected during the software development process to identify potential low-quality program modules. In this paper, we investigate the use of feature selection in the context of software quality estimation (also referred to as software defect prediction), where a classification model is used to predict whether program modules (instances) are fault-prone or not-fault-prone. Seven filter-based feature ranking techniques are examined. Among them, six are commonly used, and the other one, named signal to noise ratio (SNR), is rarely employed. The objective of the paper is to compare these seven techniques for various software data sets and assess their effectiveness for software quality modeling. A case study is performed on 16 software data sets, and classification models are built with five different learners and evaluated with two performance metrics. Our experimental results are summarized based on statistical tests for significance. The main conclusion is that the SNR technique performs as well as the best performer of the six commonly used techniques.

Software effort data contains a large amount of missing values of project attributes. The problem of absent features, which occurred recently in machine learning, is often neglected by researchers of software engineering when handling the missingness in software effort data. In essence, absent features (structural missingness) and unobserved values (unstructured missingness) are different cases of missingness although their appearance in the data set are the same. This paper attempts to clarify the root cause of missingness of software effort data. When regarding missingness as absent features, we develop Max-margin regression to predict real effort of software projects. When regarding missingness as unobserved values, we use existing imputation techniques to impute missing values. Then, ε – SVR is used to predict real effort of software projects with the input data sets. Experiments on ISBSG (International Software Benchmarking Standard Group) and CSBSG (Chinese Software Benchmarking Standard Group) data sets demonstrate that, with the tasks of effort prediction, the treatment regarding missingness in software effort data set as unobserved values can produce more desirable performance than that of regarding missingness as absent features. This paper is the first to introduce the concept of absent features to deal with missingness of software effort data.

In the presence of a number of algorithms for classification and prediction in software engineering, there is a need to have a systematic way of assessing their performances. The performance assessment is typically done by some form of partitioning or resampling of the original data to alleviate biased estimation. For predictive and classification studies in software engineering, there is a lack of a definitive advice on the most appropriate resampling method to use. This is seen as one of the contributing factors for not being able to draw general conclusions on what modeling technique or set of predictor variables are the most appropriate. Furthermore, the use of a variety of resampling methods make it impossible to perform any formal meta-analysis of the primary study results. Therefore, it is desirable to examine the influence of various resampling methods and to quantify possible differences. Objective and method: This study empirically compares five common resampling methods (hold-out validation, repeated random sub-sampling, 10-fold cross-validation, leave-one-out cross-validation and non-parametric bootstrapping) using 8 publicly available data sets with genetic programming (GP) and multiple linear regression (MLR) as software quality classification approaches. Location of (PF, PD) pairs in the ROC (receiver operating characteristics) space and area under an ROC curve (AUC) are used as accuracy indicators. Results: The results show that in terms of the location of (PF, PD) pairs in the ROC space, bootstrapping results are in the preferred region for 3 of the 8 data sets for GP and for 4 of the 8 data sets for MLR. Based on the AUC measure, there are no significant differences between the different resampling methods using GP and MLR. Conclusion: There can be certain data set properties responsible for insignificant differences between the resampling methods based on AUC. These include imbalanced data sets, insignificant predictor variables and high-dimensional data sets. With the current selection of data sets and classification techniques, bootstrapping is a preferred method based on the location of (PF, PD) pair data in the ROC space. Hold-out validation is not a good choice for comparatively smaller data sets, where leave-one-out cross-validation (LOOCV) performs better. For comparatively larger data sets, 10-fold cross-validation performs better than LOOCV.

Program matching refers to the mapping between equivalent codes written in different languages — including high-level and low-level languages. This equivalence is useful for some software engineering scenarios such as determining whether rewritten code is correct, which version of a program is being used, and whether a malware is present in the program. In the present work, we propose a novel approach to solve the executable code traceability by using program code analysis and artificial neural networks. From the program code analysis we obtained execution behavior properties of the codes, and from the artificial neural networks we judge about their correspondence. Our evaluation using real code examples shows an acceptable correspondence rate between 62% and 100% with the very low rate of 4% false positives.

Administrative processes are a type of business process commonly used in public institutions and large companies. The workflow definitions of these processes are frequently reused within the organizations. The reuse of these workflow definitions can be easier if they are divided into three separate but related definitions: on the one hand, the definition of the data structures to be managed by the process activities, on the other, the users that can perform each activity and, finally, the process activities together with the relationships among the three definitions. The use of ontologies can facilitate these workflow definitions in the three related parts. In this paper, we describe OntoMetaWorkflow, a generic ontology to represent canonical workflow terms in the domain of administrative processes, and the methods for using it in defining administrative processes. This ontology is the basis of the WEAPON Model, a complete model supported by tools, for defining and managing administrative process workflows.

We demonstrate how a subset of UML statechart diagrams (SCDs) as distinct from statecharts is encoded in the π-calculus as a number of processes which communicate via a channel-passing interaction paradigm. An extension to the original UML semantics for communicating statechart diagrams is presented. Different sorts of equivalences of UML SCDs encompassing isomorphism, strong behavioural equivalence and weak behavioural equivalence are defined. Determining equivalence of any two UML SCDs is transformed to a problem of checking whether their respective π-calculus process expressions are equivalent. The practicality of the approach is exemplified using an industry level case study.

Performance of information retrieval (IR) systems greatly relies on textual keywords and retrieval documents. Inaccurate and incomplete retrieval results are always induced by query drift and ignorance of semantic relationship among terms. Expanding retrieval approach attempts to incorporate expansion terms into original query, such as unexplored words combing from pseudo-relevance feedback (PRF) or relevance feedback documents semantic words extracting from external corpus etc. In this paper a semantic analysis-based query expansion method for information retrieval using WordNet and Wikipedia as corpus are proposed. We derive semantic-related words from human knowledge repositories such as WordNet and Wikipedia, which are combined with words filtered by semantic mining from PRF document. Our approach automatically generates new semantic-based query from original query of IR. Experimental results on TREC datasets and Google search engine show that performance of information retrieval can be significantly improved using proposed method over previous results.