A dynamic over-sampling procedure based on sensitivity for multi-class problems

Classification with imbalanced datasets supposes a new challenge for researches in the framework of machine learning. This problem appears when the number of patterns that represents one of the classes of the dataset (usually the concept of interest) is much lower than in the remaining classes. Thus, the learning model must be adapted to this situation, which is very common in real applications. In this paper, a dynamic over-sampling procedure is proposed for improving the classification of imbalanced datasets with more than two classes. This procedure is incorporated into a memetic algorithm (MA) that optimizes radial basis functions neural networks (RBFNNs). To handle class imbalance, the training data are resampled in two stages. In the first stage, an over-sampling procedure is applied to the minority class to balance in part the size of the classes. Then, the MA is run and the data are over-sampled in different generations of the evolution, generating new patterns of the minimum sensitivity class (the class with the worst accuracy for the best RBFNN of the population). The methodology proposed is tested using 13 imbalanced benchmark classification datasets from well-known machine learning problems and one complex problem of microbial growth. It is compared to other neural network methods specifically designed for handling imbalanced data. These methods include different over-sampling procedures in the preprocessing stage, a threshold-moving method where the output threshold is moved toward inexpensive classes and ensembles approaches combining the models obtained with these techniques. The results show that our proposal is able to improve the sensitivity in the generalization set and obtains both a high accuracy level and a good classification level for each class.     

不同的距离测量方法对人工免疫识别系统的性能影响*

人工免疫识别系统AIRS（Artificial Immune Recognition System）是著名的免疫网络分类器,被成功地应用到大量的分类问题,表现出了良好的性能。为了分析不同的距离测量方法对AIRS的性能影响, 采用三种距离测量方法实现AIRS,这三种方法分别是Euclidean距离、Manhattan距离和RBF核空间距离,并将三种用不同距离测量方法实现的AIRS算法应用于Iris,Heart和Wine数据集的分类测试。所获得的三组数据集分类的准确率和抗体规模进行了相互比较,结果表明采用Manhattan距离AIRS算法获得了对Iris和Heart的最高分类准确率,而采用核空间距离,算法获得了对Wine的最高分类准确率。从抗体群体规模来看,采用核空间距离则能获得最小的抗体群体。从性能比较可知,不同的距离测量方法对AIRS算法的分类性能较大的影响。     

Fast wrapper feature subset selection in high-dimensional datasets by means of filter re-ranking

This paper deals with the problem of supervised wrapper-based feature subset selection in datasets with a very large number of attributes. Recently the literature has contained numerous references to the use of hybrid selection algorithms: based on a filter ranking, they perform an incremental wrapper selection over that ranking. Though working fine, these methods still have their problems: (1) depending on the complexity of the wrapper search method, the number of wrapper evaluations can still be too large; and (2) they rely on a univariate ranking that does not take into account interaction between the variables already included in the selected subset and the remaining ones.Here we propose a new approach whose main goal is to drastically reduce the number of wrapper evaluations while maintaining good performance (e.g. accuracy and size of the obtained subset). To do this we propose an algorithm that iteratively alternates between filter ranking construction and wrapper feature subset selection (FSS). Thus, the FSS only uses the first block of ranked attributes and the ranking method uses the current selected subset in order to build a new ranking where this knowledge is considered. The algorithm terminates when no new attribute is selected in the last call to the FSS algorithm. The main advantage of this approach is that only a few blocks of variables are analyzed, and so the number of wrapper evaluations decreases drastically.The proposed method is tested over eleven high-dimensional datasets (2400-46,000 variables) using different classifiers. The results show an impressive reduction in the number of wrapper evaluations without degrading the quality of the obtained subset.     

基于信息熵的高维分类型数据子空间聚类算法

由于分类型数据相异度度量的局限性以及分类型数据在高维空间中的稀疏性,使得传统的相异度度量在高维分类型数据聚类中失效,针对上述问题,本研究提出了一个基于信息熵的理论高维分类型数据聚类算法。该算法综合考虑对应子空间和噪声空间的维度信息熵设计了一个高效、无监督的子空间搜索对高维数据进行有效降维,同时提出了基于整体数据的平均信息熵的全局优化方法对聚类结果进行迭代寻优。通过用人工数据和Votes、Mushroom和Soybean 3个典型的真实分类数据集试验,与其他分类型聚类算法相比,新算法在聚类准确性、熵值、CU（category utility）以及类个数等指标上有明显提高。     

不平衡数据集的神经网络阈值优化方法

不平衡数据集分类为机器学习热点研究问题之一,近年来研究人员提出很多理论和算法以改进传统分类技术在不平衡数据集上的性能,其中用阈值判定标准确定神经网络中的阈值是重要的方法之一。常用的阈值判定标准存在一定缺点,如不能使少数类及多数类分类精度同时取得最好、过于偏好多数类的精度等。为此提出一种新的阈值判定标准,依据该标准能够使少数类及多数类分类精度同时取得最好而不受样例类别比例的影响。以神经网络与遗传算法相结合训练分类器,作为阈值选择条件和分类器的评价标准,新标准能够得到较好的结果。     

不平衡数据集中的组合分类算法

为提高少数类的分类性能,对基于数据预处理的组合分类器算法进行了研究.利用Tomek links对数据集进行预处理;把新数据集里的多数类样本按照不平衡比拆分为多个子集,每个子集和少数类样本合并成新子集;用最小二乘支持向量机对每个新子集进行训练,把训练后的各个子分类器组合为一个分类系统,新的测试样本的类别将由这个分类系统投票表决.数据试验结果表明,该算法在多数类和少数类的分类性能方面,都优于最小二乘支持向量机过抽样方法和欠抽样方法.     

非平衡技术在高速网络入侵检测中的应用

针对现有的高速网络入侵检测系统丢包率高、检测速度慢以及检测算法对不同类型攻击检测的非平衡性等问题,提出了采用两阶段的负载均衡策略的检测模型。在线检测阶段对网络数据包按协议类型进行分流的检测,离线建模阶段对不同协议类型的数据进行学习建模,供在线部分检测。在讨论非平衡数据处理的各种采样技术基础上,采用改进后的过抽样少数样本合成过采样技术（SMOTE）对网络数据进行预处理,采用AdaBoost 、随机森林算法等进行分类。另外对特征选取等方面进行了实验,结果表明SMOTE过抽样可提高各少数类的检测,随机森林算法分类效果好而且建模所用的时间稳定。     

Comparing alternative classifiers for database marketing: The case of imbalanced datasets

There are various algorithms used for binary classification where the cases are classified into one of two non-overlapping classes. The area under the receiver operating characteristic (ROC) curve is the most widely used metric to evaluate the performance of alternative binary classifiers. In this study, for the application domains where the high degree of imbalance is the main characteristic and the identification of the minority class is more important, we show that hit rate based measures are more correct to assess model performances and that they should be measured on out of time samples. We also try to identify the optimum composition of the training set. Logistic regression, neural network and CHAID algorithms are implemented for a real marketing problem of a bank and the performances are compared.     

基于相关规则的不平衡数据的关联分类

许多研究表明关联分类具有较高的分类准确率,然而,大多数关联分类基于"支持度-置信度"框架,在不平衡数据集中,置信度和支持度都偏向产生多数类的规则,因此,少数类的实例容易被错误分类。针对上述问题,提出了一种基于相关规则的不平衡数据的关联分类算法。该算法挖掘频繁且互关联的项集,在以该项集为前件的分类规则中选取提升度最大的规则。规则按结合了提升度、置信度和补类支持度(CCS)的规则强度进行排序。实验表明,该算法取得了较高的平均分类准确率且在分类少数类的实例时具有更高的准确率。     

Deep learning methods for fingerprint-based indoor positioning: a review

ABSTRACT

Outdoor positioning systems based on the Global Navigation Satellite System have several shortcomings that have deemed their use for indoor positioning impractical. Location fingerprinting, which utilizes machine learning, has emerged as a viable method and solution for indoor positioning due to its simple concept and accurate performance. In the past, shallow learning algorithms were traditionally used in location fingerprinting. Recently, the research community started utilizing deep learning methods for fingerprinting after witnessing the great success and superiority these methods have over traditional/shallow machine learning algorithms. This paper provides a comprehensive review of deep learning methods in indoor positioning. First, the advantages and disadvantages of various fingerprint types for indoor positioning are discussed. The solutions proposed in the literature are then analyzed, categorized, and compared against various performance evaluation metrics. Since data is key in fingerprinting, a detailed review of publicly available indoor positioning datasets is presented. While incorporating deep learning into fingerprinting has resulted in significant improvements, doing so, has also introduced new challenges. These challenges along with the common implementation pitfalls are discussed. Finally, the paper is concluded with some remarks as well as future research trends.