Search

k nearest neighbor (k NN) is a simple and widely used classifier; it can achieve comparable performance with more complex classifiers including decision tree and artificial neural network. Therefore, k NN has been listed as one of the top 10 algorithms in machine learning and data mining. On the other hand, in many classification problems, such as medical diagnosis and intrusion detection, the collected training sets are usually class imbalanced. In class imbalanced data, although positive examples are heavily outnumbered by negative ones, positive examples usually carry more meaningful information and are more important than negative examples. Similar to other classical classifiers, k NN is also proposed under the assumption that the training set has approximately balanced class distribution, leading to its unsatisfactory performance on imbalanced data. In addition, under a class imbalanced scenario, the global resampling strategies that are suitable to decision tree and artificial neural network often do not work well for k NN, which is a local information-oriented classifier. To solve this problem, researchers have conducted many works for k NN over the past decade. This paper presents a comprehensive survey of these works according to their different perspectives and analyzes and compares their characteristics. At last, several future directions are pointed out. [ABSTRACT FROM AUTHOR]

Both cost-sensitive classification and online learning have been extensively studied in data mining and machine learning communities, respectively. However, very limited study addresses an important intersecting problem, that is, “Cost-Sensitive Online Classification”. In this paper, we formally study this problem, and propose a new framework for Cost-Sensitive Online Classification by directly optimizing cost-sensitive measures using online gradient descent techniques. Specifically, we propose two novel cost-sensitive online classification algorithms, which are designed to directly optimize two well-known cost-sensitive measures: (i) maximization of weighted sum of sensitivity and specificity, and (ii) minimization of weighted misclassification cost. We analyze the theoretical bounds of the cost-sensitive measures made by the proposed algorithms, and extensively examine their empirical performance on a variety of cost-sensitive online classification tasks. Finally, we demonstrate the application of the proposed technique for solving several online anomaly detection tasks, showing that the proposed technique could be a highly efficient and effective tool to tackle cost-sensitive online classification tasks in various application domains. [ABSTRACT FROM AUTHOR]

Imbalanced learning is important and challenging since the problem of the classification of imbalanced datasets is prevalent in machine learning and data mining fields. Sampling approaches are proposed to address this issue, and cluster-based oversampling methods have shown great potential as they aim to simultaneously tackle between-class and within-class imbalance issues. However, all existing clustering methods are based on a one-time approach. Due to the lack of a priori knowledge, improper setting of the number of clusters often exists, which leads to poor clustering performance. Besides, the existing methods are likely to generate noisy instances. To solve these problems, this paper proposes a deep instance envelope network-based imbalanced learning algorithm with the multilayer fuzzy c-means (MlFCM) and a minimum interlayer discrepancy mechanism based on the maximum mean discrepancy (MIDMD). This algorithm can guarantee high quality balanced instances using a deep instance envelope network in the absence of prior knowledge. First, the MlFCM is designed for the original minority class instances to obtain deep instances and increase the diversity of instances. Then, the MIDMD is proposed to avoid the generation of noisy instances and maintain the consistency of the interlayers of instances. Next, the multilayer FCM and minimum interlayer discrepancy mechanism are combined to construct a deep instance envelope network – the MlFC&IDMD. Finally, an imbalance learning algorithm is proposed based on the MlFC&IDMD. In the experimental section, thirty-three popular public datasets are used for verification, and over ten representative algorithms are used for comparison. The experimental results show that the proposed approach significantly outperforms other popular methods. • For the first time, a multilayer FCM (MlFCM) algorithm for classification is proposed to mine the instances for more information. • A minimum interlayer discrepancy mechanism (MIDMD) is proposed to make the distribution of instances before and after clustering consistent. • A deep instance envelope network (MlFC&IDMD) is constructed by combining the MlFCM and MIDMD. • An efficient strategy is proposed to determine the number of layers L of the deep instance envelope network. • A new imbalance learning algorithm is proposed based on the MlFC&IDMD. [ABSTRACT FROM AUTHOR]

Abstract: Clustering is a method of unsupervised learning, and a common technique for intelligent data analysis used in many fields. The enlarging volumes of information produced by the progress of technology, makes clustering of very large scale data a challenging task. Therefore, parallel processing of very large scale data set is a good solution to that problem. CLARANS is approved to be an efficient and effective clustering method for spatial data mining. It is insensitive to outliers and has the advantage of processing category data directly compared to other partition method like K-means. However, it is a serial algorithm and cannot process large scale data set efficiently. In this paper, we propose a parallel CLARANS clustering algorithm based on MapReduce, which is a current and powerful parallel programming framework. Experimental results demonstrate that the parallel CLARANS algorithm can handle large scale data set efficiently and effectively. Furthermore, the evaluation criterions of scaleup, speedup and sizeup validate its efficiency. [Copyright &y& Elsevier]

Approximate Nearest Neighbor Search (ANNS) is a fundamental problem in many areas of machine learning and data mining. During the past decade, numerous hashing algorithms are proposed to solve this problem. Every proposed algorithm claims to outperform Locality Sensitive Hashing (LSH), which is the most popular hashing method. However, the evaluation of these hashing article was not thorough enough, and the claim should be re-examined. If implemented correctly, almost all the hashing methods will have their performance improved as the code length increases. However, many existing hashing article only report the performance with the code length shorter than 128. In this article, we carefully revisit the problem of search-with-a-hash-index and analyze the pros and cons of two popular hash index search procedures. Then we proposed a simple but effective novel hash index search approach and made a thorough comparison of eleven popular hashing algorithms. Surprisingly, the random-projection-based Locality Sensitive Hashing ranked the first, which is in contradiction to the claims in all the other 10 hashing article. Despite the extreme simplicity of random-projection-based LSH, our results show that the capability of this algorithm has been far underestimated. For the sake of reproducibility, all the codes used in the article are released on GitHub, which can be used as a testing platform for a fair comparison between various hashing algorithms. [ABSTRACT FROM AUTHOR]