A self-adaptive synthetic over-sampling technique for imbalanced classification

Traditionally, in supervised machine learning, (a significant) part of the available data (usually 50%-80%) is used for training and the rest—for validation. In many problems, however, the data are highly imbalanced in regard to different classes or does not have good coverage of the feasible data space which, in turn, creates problems in validation and usage phase. In this paper, we propose a technique for synthesizing feasible and likely data to help balance the classes as well as to boost the performance in terms of confusion matrix as well as overall. The idea, in a nutshell, is to synthesize data samples in close vicinity to the actual data samples specifically for the less represented (minority) classes. This has also implications to the so-called fairness of machine learning. In this paper, we propose a specific method for synthesizing data in a way to balance the classes and boost the performance, especially of the minority classes. It is generic and can be applied to different base algorithms, for example, support vector machines, k-nearest neighbour classifiers deep neural, rule-based classifiers, decision trees, and so forth. The results demonstrated that (a) a significantly more balanced (and fair) classification results can be achieved and (b) that the overall performance as well as the performance per class measured by confusion matrix can be boosted. In addition, this approach can be very valuable for the cases when the number of actual available labelled data is small which itself is one of the problems of the contemporary machine learning.     

TCBR-HMM: An HMM-based text classifier with a CBR system

This paper presents an innovative solution to model distributed adaptive systems in biomedical environments. We present an original TCBR-HMM (Text Case Based Reasoning-Hidden Markov Model) for biomedical text classification based on document content. The main goal is to propose a more effective classifier than current methods in this environment where the model needs to be adapted to new documents in an iterative learning frame. To demonstrate its achievement, we include a set of experiments, which have been performed on OSHUMED corpus. Our classifier is compared with Naive Bayes and SVM techniques, commonly used in text classification tasks. The results suggest that the TCBR-HMM Model is indeed more suitable for document classification. The model is empirically and statistically comparable to the SVM classifier and outperforms it in terms of time efficiency.     

Multiobjective PSO based adaption of neural network topology for pixel classification in satellite imagery

The proposed work involves the multiobjective PSO based adaption of optimal neural network topology for the classification of multispectral satellite images. It is per pixel supervised classification using spectral bands (original feature space). This paper also presents a thorough experimental analysis to investigate the behavior of neural network classifier for given problem. Based on 1050 number of experiments, we conclude that following two critical issues needs to be addressed: (1) selection of most discriminative spectral bands and (2) determination of optimal number of nodes in hidden layer. We propose new methodology based on multiobjective particle swarm optimization (MOPSO) technique to determine discriminative spectral bands and the number of hidden layer node simultaneously. The accuracy with neural network structure thus obtained is compared with that of traditional classifiers like MLC and Euclidean classifier. The performance of proposed classifier is evaluated quantitatively using Xie-Beni and β indexes. The result shows the superiority of the proposed method to the conventional one.     

Specific emitter identification for satellite communication using probabilistic neural networks

Electromagnetic signal emitted by satellite communication (satcom) transmitters are used to identify specific individual uplink satcom terminals sharing the common transponder in real environment, which is known as specific emitter identification (SEI) that allows for early indications and warning (I&W) of the targets carrying satcom furnishment and furthermore the real time electromagnetic situation awareness in military operations. In this paper, the authors are the first to propose the identification of specific transmitters of satcom by using probabilistic neural networks (PNN) to reach the goal of target recognition. We have been devoted to the examination by exploring the feasibility of utilizing the Hilbert transform to signal preprocessing, applying the discrete wavelet transform to feature extraction, and employing the PNN to perform the classification of stationary signals. There are a total of 1000 sampling time series with binary phase shift keying (BPSK) modulation originated by five types of satcom transmitters in the test. The established PNNs classifier implements the data testing and finally yields satisfactory accuracy at 8 dB(±1 dB) carrier to noise ratio, which indicates the feasibility of our method, and even the keen insight of its application in military.     

Mining citizen emotions to estimate the urgency of urban issues

Crowdsourcing technology offers exciting possibilities for local governments. Specifically, citizens are increasingly taking part in reporting and discussing issues related to their neighborhood and problems they encounter on a daily basis, such as overflowing trash-bins, broken footpaths and lifts, illegal graffiti, and potholes. Pervasive citizen participation enables local governments to respond more efficiently to these urban issues. This interaction between citizens and municipalities is largely promoted by civic engagement platforms, such as See-Click-Fix, FixMyStreet, CitySourced, and OpenIDEO, which allow citizens to report urban issues by entering free text describing what needs to be done, fixed or changed. In order to develop appropriate action plans and priorities, government officials need to figure out how urgent are the reported issues. In this paper we propose to estimate the urgency of urban issues by mining different emotions that are implicit in the text describing the issue. More specifically, a reported issue is first categorized according to the emotions expressed in it, and then the corresponding emotion scores are combined in order to produce a final urgency level for the reported issue. Our experiments use the SeeClickFix hackathon data and diverse emotion classification algorithms. They indicate that (i) emotions can be categorized efficiently with supervised learning algorithms, and (ii) the use of citizen emotions leads to accurate urgency estimates. Further, using additional features such as the type of issue or its author leads to no further accuracy gains.     

ECG heart beat classification method based on modified ABC algorithm

Electrocardiogram is the most commonly used tool for the diagnosis of cardiologic diseases. In order to help cardiologists to diagnose the arrhythmias automatically, new methods for automated, computer aided ECG analysis are being developed. In this paper, a Modified Artificial Bee Colony (MABC) algorithm for ECG heart beat classification is introduced. It is applied to ECG data set which is obtained from MITBIH database and the result of MABC is compared with seventeen other classifier's accuracy.In classification problem, some features have higher distinctiveness than others. In this study, in order to find higher distinctive features, a detailed analysis has been done on time domain features. By using the right features in MABC algorithm, high classification success rate (99.30%) is obtained. Other methods generally have high classification accuracy on examined data set, but they have relatively low or even poor sensitivities for some beat types. Different data sets, unbalanced sample numbers in different classes have effect on classification result. When a balanced data set is used, MABC provided the best result as 97.96% among all classifiers.Not only part of the records from examined MITBIH database, but also all data from selected records are used to be able to use developed algorithm on a real time system in the future by using additional software modules and making adaptation on a specific hardware.     

深度学习识别光网络单元故障的设计与应用

为解决依赖装维上门鉴别光网络单元故障带来的不便,可以从机器视觉入手实现自动化故障识别。近年,ImageNet挑战赛的成功推动了物体识别技术的跨越式发展,特别是基于卷积的深度学习技术在视觉识别方面已经达到人类水平,为光网络单元故障的自动识别提供了技术基础。文章对识别光网络单元的工作状态进行了研究,将设备工作状态分为7个场景,提出了利用手机APP采集图片识别故障的解决方案并投入了实际生产;重点阐述了深度学习模块的设计与实现,提出一种通过算法整合的方式综合运用物体检测和图像分类算法,分3阶段逐步求精,解决了图片过滤,光网络单元型号和状态识别等问题,实现了基于计算机视觉自动识别光网络单元故障。从数据上看产品的端到端准确率超过84%,识别速度达到10 FPS,月均提供服务超过1万人次,在减少用户等待的同时节约了人力资源。     

Efficiency and effectiveness enhancement of an intercooler of two‐stage air compressor by low‐cost Al2O3/water nanofluids

The present study was aimed to utilize low‐cost alumina (Al₂O₃) nanoparticles for improving the heat transfer behavior in an intercooler of two‐stage air compressor. Experimental investigation was carried out with three different volume concentrations of 0.5%, 0.75%, and 1.0% Al₂O₃/water nanofluids to assess the performance of the intercooler, that is, counterflow heat exchanger at different loads. Thermal properties such as thermal conductivity and overall heat transfer coefficient of nanofluid increased substantially with increasing concentration of Al₂O₃ nanoparticles. Specific heat capacity of nanofluids were lower than base water. The intercooler performance parameters such as effectiveness and efficiency improved appreciably with the employment of nanofluid. The efficiency increased by about 6.1% with maximum concentration of nanofluid, that is, 1% at 3‐bar compressor load. It is concluded from the study that high concentration of Al₂O₃ nanoparticles dispersion in water would offer better heat transfer performance of the intercooler.     

Application of Vis–NIR hyperspectral imaging in classification between fresh and frozen-thawed pork Longissimus Dorsi muscles

Fresh and frozen-thawed (F-T) pork meats were classified by Vis–NIR hyperspectral imaging. Eight optimal wavelengths (624, 673, 460, 588, 583, 448, 552 and 609 nm) were selected by successive projections algorithm (SPA). The first three principal components (PCs) obtained by principal component analysis (PCA) accounted for over 99.98% of variance. Gray-level-gradient co-occurrence matrix (GLGCM) was applied to extract 45 textural features from the PC images. The correct classification rate (CCR) was employed to evaluate the performance of the partial least squares-discriminate analysis (PLS-DA) models, by using (A) the reflected spectra at full wavelengths and (B) those at the optimal wavelengths, (C) the extracted textures based on the PC images, and (D) the fused variables combining spectra at the optimal wavelengths and textures. The results showed that the best CCR of 97.73% was achieved by applying (D), confirming the high potential of textures for fresh and F-T meat discrimination.     

因素空间理论下基点分类算法研究

目前，基于因素空间理论的背景基提取算法计算过程复杂，初始化必须依赖各因素极值，基点数量提取冗余等原因，未能在应用中取得很好效果。为此，结合内点判别法和知识可继承、可扩展的思想，提出一种计算简单、初始化独立、基点数量小的改进的背景基提取算法。然后，利用改进的背景基提取算法构造出一种全新的数据分类算法-基点分类算法，基点分类算法以提取每一类样本的背景基为预测模型，再通过新定义的λ-背景基，优化预测模型。数值实验表明:基点分类算法原理简单、构造难度小、分类模型泛化能力强，预测能力准确率高，同时严格的模型限定区域又能为识别新类别提供新方法。