Narrow Results

Machine-learning-based software defect prediction (SDP) methods are receiving great attention from the researchers of intelligent software engineering. Most existing SDP methods are performed under a within-project setting. However, there usually is little to no within-project training data to learn an available supervised prediction model for a new SDP task. Therefore, cross-project defect prediction (CPDP), which uses labeled data of source projects to learn a defect predictor for a target project, was proposed as a practical SDP solution. In real CPDP tasks, the class imbalance problem is ubiquitous and has a great impact on performance of the CPDP models. Unlike previous studies that focus on subsampling and individual methods, this study investigated 15 imbalanced learning methods for CPDP tasks, especially for assessing the effectiveness of imbalanced ensemble learning (IEL) methods. We evaluated the 15 methods by extensive experiments on 31 open-source projects derived from five datasets. Through analyzing a total of 37504 results, we found that in most cases, the IEL method that combined under-sampling and bagging approaches will be more effective than the other investigated methods.

Traditional classifiers are trapped by the class-imbalanced problem due to the fact that they are biased toward the majority class. Oversampling methods can improve imbalanced classification by creating synthetic minority class samples. Noise generation has been a great challenge in oversampling methods. Filtering-based and direction-change methods are proposed against noise generation. Yet, the adopted noise filters in filtering-based methods are biased to the majority class. Besides, the $k$-nearest neighbor (KNN)-based interpolation in filtering-based and direction-change methods is susceptible to abnormal samples (e.g. outliers, noise or unsafe borderline samples). To overcome noise generation while solving the above shortcomings of filtering-based and direction-change methods, this work presents a new synthetic minority oversampling technique based on local means-based KNN (SMOTE-LMKNN). In SMOTE-LMKNN, the local mean-based KNN (LMKNN) is first introduced to describe the local characteristic of imbalanced data. Second, a new LMKNN-based noise filter is proposed to remove noise and unsafe borderline samples. Third, the interpolation between a base sample and its LMKNN is proposed to create synthetic minority class samples. Empirical results of extensive experiments with 18 data sets show that SMOTE-LMKNN is competent compared with seven popular oversampling methods in training KNN classifier and classification and regression tree (CART).

The classification is usually degraded due to the imbalanced class distribution. Synthetic minority oversampling technique (SMOTE) has been successful in improving imbalanced classification and has received great praise. Overgeneralization is one of the most challenges in SMOTE. Although multiple SMOTE-based variations are proposed against overgeneralization, they still have the following shortcomings: (a) creating too many synthetic samples in high-density regions; (b) removing suspicious noise directly instead of modifying them; (c) relying on many parameters. This paper proposes a new SMOTE based on adaptive noise optimization and fast search for local sets (SMOTE-ANO-FLS) to overcome the overgeneralization and the shortcomings of existing works. First, SMOTE-ANO-FLS uses the $k$-D tree to fast search the local sets for each sample. Second, a new noise detection method based on local sets and the imbalanced ratio is proposed to detect suspicious noise. Third, a new adaptive noise optimization method is proposed to modify detected suspicious noise instead of removing them. Finally, a new probability weight based on local sets is proposed to help create more synthetic minority class samples in borderline and sparse regions. The effectiveness of SMOTE-ANO-FLS is proven by employing 7 oversampling methods and random forest on the extensive synthetic and real data sets.

To facilitate software maintenance and save the maintenance cost, numerous machine learning (ML) techniques have been studied to predict the maintainability of software modules or classes. An abundant amount of effort has been put by the research community to develop software maintainability prediction (SMP) models by relating software metrics to the maintainability of modules or classes. When software classes demanding the high maintainability effort (HME) are less as compared to the low maintainability effort (LME) classes, the situation leads to imbalanced datasets for training the SMP models. The imbalanced class distribution in SMP datasets could be a dilemma for various ML techniques because, in the case of an imbalanced dataset, minority class instances are either misclassified by the ML techniques or get discarded as noise. The recent development in predictive modeling has ascertained that ensemble techniques can boost the performance of ML techniques by collating their predictions. Ensembles themselves do not solve the class-imbalance problem much. However, aggregation of ensemble techniques with the certain techniques to handle class-imbalance problem (e.g., data resampling) has led to several proposals in research. This paper evaluates the performance of ensembles for the class-imbalance in the domain of SMP. The ensembles for class-imbalance problem (ECIP) are the modification of ensembles which pre-process the imbalanced data using data resampling before the learning process. This study experimentally compares the performance of several ECIP using performance metrics Balance and g-Mean over eight Apache software datasets. The results of the study advocate that for imbalanced datasets, ECIP improves the performance of SMP models as compared to classic ensembles.

Customer churn prediction is important for bankers to preserve profitable customers. However, the relatively low churn rate makes it become an imbalanced learning problem. We implement the oversampling strategy for four well-known classification algorithms on various criteria, and identify the most suitable techniques for customer churn analysis. The study is based on real world dataset from a commercial bank in China. The experimental results indicate that the synthetic oversampling approach makes notable lifts on the criteria precision, G-mean, and the weighted accuracy. It is suitable for the application areas where the minorities are more concerned.

The stock reviews are important professional advice for stock traders. Based on the massive stock reviews available online, an automatic approach which can identify the emotional tendency hidden in stock reviews will provide tremendous help to stock traders. A sentiment classification strategy is proposed for stock reviews analysis. Firstly, stock reviews are collected from the Internet and construct a labeled corpus. Secondly, a feature extraction approach is designed which can effectively map the corpus into a structured matrix. Finally, four classifiers are evaluated on multiple criteria. The synthetic oversampling approach is utilized to overcome the imbalanced distribution of positive and negative reviews. The experimental results demonstrate that the SVM algorithms coupled with SMOTE can achieve the highest performance for sentiment classification on the real world Chinese stock review corpus.

Brain tumor is a fatal central nervous system disease that occurs in around 250,000 people each year globally and it is the second cause of cancer in children. It has been widely acknowledged that genetic factor is one of the significant risk factors for brain cancer. Thus, accurate descriptions of the locations of where the relative genes are active and how these genes express are critical for understanding the pathogenesis of brain tumor and for early detection. The Allen Developing Mouse Brain Atlas is a project on gene expression over the course of mouse brain development stages. Utilizing mouse models allows us to use a relatively homogeneous system to reveal the genetic risk factor of brain cancer. In the Allen atlas, about 435,000 high-resolution spatiotemporal in situ hybridization images have been generated for approximately 2,100 genes and currently the expression patterns over specific brain regions are manually annotated by experts, which does not scale with the continuously expanding collection of images. In this paper, we present an efficient computational approach to perform automated gene expression pattern annotation on brain images. First, the gene expression information in the brain images is captured by invariant features extracted from local image patches. Next, we adopt an augmented sparse coding method, called Stochastic Coordinate Coding, to construct high-level representations. Different pooling methods are then applied to generate gene-level features. To discriminate gene expression patterns at specific brain regions, we employ supervised learning methods to build accurate models for both binary-class and multi-class cases. Random undersampling and majority voting strategies are utilized to deal with the inherently imbalanced class distribution within each annotation task in order to further improve predictive performance. In addition, we propose a novel structure-based multi-label classification approach, which makes use of label hierarchy based on brain ontology during model learning. Extensive experiments have been conducted on the atlas and results show that the proposed approach produces higher annotation accuracy than several baseline methods. Our approach is shown to be robust on both binary-class and multi-class tasks and even with a relatively low training ratio. Our results also show that the use of label hierarchy can significantly improve the annotation accuracy at all brain ontology levels.