关键词： 数字水印   机器学习   最小二乘支持向量机   奇异值分解   离散小波变换  

摘要： 针对目前数字水印算法存在的不足,本文将离散小波变换和奇异值分解相结合,提出了一种基于机器学习的图像数字水印算法.首先将载体图像进行一级小波变换,提取其低频子带图像对其进行4×4分块处理,然后对每一分块进行奇异值分解后嵌入水印,并提取特征向量用于最小二乘支持向量机的训练,训练好的最小二乘支持向量机用于自适应最大水印嵌入强度的计算以及水印的盲提取.实验选取三张512×512的标准测试图像以及64×64的二值水印图像对算法的透明性与鲁棒性进行测试.实验结果证明,图像具有很好的透明性,PSNR达到了63.71dB,针对旋转、剪切、JPEG压缩、高斯噪声等常规攻击手段时,算法能保持较强的鲁棒性.

关键词： 进化的参与式学习   T-S模糊模型   行程时间   卡尔曼滤波算法   交通流   聚类算法  

摘要： 在常规交通流情况下,应用参与式机器学习模型估算了交通密度的预测值,通过交通流密度与行程时间的关系间接估算行程时间的预测值;对参与式机器学习模型进行了改进,将原模型中的子系统计算方法从最小二乘法改进为适合长时间平稳交通预测的卡尔曼滤波法;在此基础上开发了部署在可变情报板上的交通信息诱导系统,利用观测的交通密度组成的向量作为输入,实时动态估算预测值,最终行程时间估算结果优于其他常见的单一预测方法,表明这种方法对预测行程时间有效,可以应用于交通控制相关的领域.

关键词： Pentatricopeptide repeat   Amino acid   Correlation analysis   Max-relevance-max-distance   Principal component analysis   T test  

摘要： Motivation: Identifying whether a pentatricopeptide repeat (PPR) exists in an amino acid is a significant task in the field of bioinformatics. To address this problem, an identification method that combines an optimal feature set selection framework and machine learning algorithms is proposed to recognize the PPR coding genes and proteins in the sequence of amino acid. The original 188-dimensional (D) features are obtained using a feature extraction method, which is successively optimised through a covariance analysis, max-relevant-max-distance processing, and principal component analysis to reduce it to an optimal feature set that has fewer but more expressive features. Four machine learning methods are then used to serve as the classifiers for the identification task. Results: The final number of feature data dimensions is reduced from 188 to only 10, and according to the experimental results from support vector machine methods, the loss of the AUC and the F-1 values are only 3.26% and 10.1%, respectively. Moreover, after applying the J48, random forest, and naive Bayes methods as classifiers, it was also found that the optimal feature set with 10 dimensions has an almost equivalent performance for a 10-fold validation test. (c) 2020 Elsevier B.V. All rights reserved.

关键词： Residential energy estimation   Machine learning   Predictive models  

摘要： The objective of this study is to identify the key factors that influence residential energy use in energy modeling. In doing so, we explore the impact of data transformations and analysis methods in developing residential energy models using social, economic, and demographic indicators at the zip code level in Atlanta, GA and for the entire state of Georgia. Orthogonalization algorithms, machine learning and variable selection techniques and ordinary least squares (OLS) are used to generate models for annual energy use for each zip code. Using log transformed yearly electricity estimation with orthogonalization yielded better estimates than other transformations [R-2 = 0.80, normalized root mean squared error (NRMSE) = 0.33, parameters = 15] and results for natural gas estimate were better (R-2 = 0.95, NRMSE = 0.15, parameters = 9). As expected, both models showed that sociodemographic factors are significant predictors. For natural gas, income and household make-up are the most important factors while electricity has a broader variety of indicator types. For electricity, despite the model accounting for 80% of electricity variation, the NRMSE was still moderately high (0.33). When electricity use was separated into two clusters (high and low usage), the high use clusters appeared to match the interstate infrastructure morphology. These results show that electricity use, unlike natural gas use, is influenced by the morphology of the interstate roadway infrastructure and other social demographic factors.

关键词： Bibliometric analysis   Excavation risk   Risk assessment methods   Machine learning   Comprehensive risk management  

摘要： This paper presents a brief review on major accidents and conducts bibliometric analysis of risk assessment methods for excavation system in recent year. The summarization of potential risks during excavation provides an important index for establishing an early warning system. The applications of fuzzy set theory and machine learning methods in risk assessment during excavation are presented. A case study of excavation in Guangzhou metro station is used to demonstrate the application of a machine learning method for risk evaluation. The large amount of data collected by 3S techniques (RS, GIS and GPS) and sensors increases accuracy of risk assessment levels in excavation. These procedures, integrated into building information modelling (BIM) management platform, can manipulate dynamic safety risk monitoring, control, and management. Finally, the processing and analysis of big data obtained from 3S techniques and sensors provide promising perspectives for establishing integrated technology system for excavation.

关键词： Covid-19   Convolutional neural networks (CNN)   Deep learning   Lung CT classification   Machine learning   Texture analysis methods  

摘要： The Covid-19 virus outbreak that emerged in China at the end of 2019 caused a huge and devastating effect worldwide. In patients with severe symptoms of the disease, pneumonia develops due to Covid-19 virus. This causes intense involvement and damage in lungs. Although the emergence of the disease occurred a short time ago, many literature studies have been carried out in which these effects of the disease on the lungs were revealed by the help of lung CT imaging. In this study, 1.396 lung CT images in total (386 Covid-19 and 1.010 Non-Covid-19) were subjected to automatic classification. In this study, Convolutional Neural Network (CNN), one of the deep learning methods, was used which suggested automatic classification of CT images of lungs for early diagnosis of Covid-19 disease. In addition, k-Nearest Neighbors (k-NN) and Support Vector Machine (SVM) was used to compare the classification successes of deep learning with machine learning. Within the scope of the study, a 23-layer CNN architecture was designed and used as a classifier. Also, training and testing processes were performed for Alexnet and Mobilenetv2 CNN architectures as well. The classification results were also calculated for the case of increasing the number of images used in training for the first 23-layer CNN architecture by 5, 10, and 20 times using data augmentation methods. To reveal the effect of the change in the number of images in the training and test clusters on the results, two different training and testing processes, 2-fold and 10-fold cross-validation, were performed and the results of the study were calculated. As a result, thanks to these detailed calculations performed within the scope of the study, a comprehensive comparison of the success of the texture analysis method, machine learning, and deep learning methods in Covid-19 classification from CT images was made. The highest mean sensitivity, specificity, accuracy, F-1 score, and AUC values obtained as a result of the study we

关键词： NIR spectroscopy   Model   Bamboo shoot   Pre-processing method   Variable selection  

摘要： The nutrition and quality of bamboo shoots from different species have a large variation, and traditional methods used for the classification of different bamboo shoot species are expensive and time-consuming. Here, the capability of near-infrared reflectance (NIR) spectroscopy to identify bamboo shoot species in a time- and cost-effective manner was examined. The NIR spectra of four bamboo shoot species were collected. Three classification models, a support vector machine (SVM), partial least squares-discriminant analysis (PLSDA), and random forest (RF), were calibrated. Several spectra pre-processing methods and their combination were trained before model calibration for the best classification model collection;each model was run 200 times for the calculation of the prediction error and model stability checking. The SVM model combined with the Det+2nd derivative had the best performance with an overall accuracy of 0.95. The use of less than 16% of the full-length NIR spectra produced a similar high accuracy of 0.91. Eight important regions, 1015, 1135, 1175, 1338, 1380, 1620, 1690, and 1750 nm, were found to be highly related to the classification of bamboo shoots. This study found that NIR spectra combined with SVM methods produced a rapid and non-destructive approach for the classification of bamboo shoot species.

关键词： Electrocardiography   Feature extraction   Picture archiving and communication systems   Support vector machines   Heart rate variability   MIMICs   Premature atrial contraction   premature ventricular contraction   atrial fibrillation   density Poincaré   plot   correlation   wavelet transform   random forest   SVM  

摘要： Objective: Detection of Atrial fibrillation (AF) from premature atrial contraction (PAC) and premature ventricular contraction (PVC) is difficult as frequent occurrences of these ectopic beats can mimic the typical irregular patterns of AF. In this paper, we present a novel density Poincare plot-based machine learning method to detect AF from PAC/PVCs using electrocardiogram (ECG) recordings. Methods: First, we propose the generation of this new density Poincare plot which is derived from the difference of the heart rate (DHR) and provides the overlapping phase-space trajectory information of the DHR. Next, from this density Poincare plot, several image processing domain-based approaches including statistical central moments, template correlation, Zernike moment, discrete wavelet transform and Hough transform features are used to extract suitable features. Subsequently, the infinite latent feature selection algorithm is implemented to rank the features. Finally, classification of AF vs. PAC/PVC is performed using K-Nearest Neighbor, Support vector machine (SVM) and Random Forest (RF) classifiers. Our method is developed and validated using a subset of Medical Information Mart for Intensive Care (MIMIC) III database containing 10 AF and 10 PAC/PVC subjects. Results- During the segment-wise 10-fold cross-validation, SVM achieved the best performance with 98.99% sensitivity, 95.18% specificity and 97.45% accuracy with the extracted features. In subject-wise scenario, RF achieved the highest accuracy of 91.93%. Moreover, we further validated the proposed method using two other databases: wearable armband ECG data and the Physionet AFPDB. 100% PAC detection accuracy was obtained for both databases without any further training. Conclusion: Our proposed density Poincare plot-based method showed superior performance when compared with four existing algorithms;thus showing the efficacy of the extracted image domain-based features. Significance: From intensive care unit's ECG

关键词： HNSCC   Outcomes   Audit   Surgical Margin  

摘要： Most surgical specialties have attempted to address concerns about the unfair comparison of outcomes by 'risk-adjusting' data to benchmark specialty-specific outcomes that are indicative of quality of care. We explore the ability to predict for positive margin status so that effective benchmarking that will account for complexity of case mix is possible. A dataset of care episodes recorded as a clinical audit of margin status after surgery for head and neck squamous cell carcinoma (n=1316) was analysed within the Waikato Environment for Knowledge Analyisis (WEKA) machine learning programme The outcome was a classification model that can predict for positivity of tumour margins (defined as less than lmm) using data on preoperative demographics, operations, functional status, and tumour stage. Positive resection margins of less than lmm were common, and varied considerably between treatment units (19%-29%). Four algorithms were compared to attempt to risk-adjust for case complexity. The 'champion' model was a Naive Bayes classifier (AUROC 0.72) that suggested acceptable discrimination. Calibration was good (Hosmer-Lemershow goodness-of-fit test p=0.9). Adjusted positive margin rates are presented on a funnel plot. Subspecialty groups within oral and maxillofacial surgery are seeking metrics that will allow for meaningful comparison of the quality of care delivered by surgical units in the UK. To enable metrics to be effective, we argue that they can be modelled so that meaningful benchmarking, which takes account of variation in complexity of patient need or care, is possible. (C) 2020 Published by Elsevier Ltd on behalf of The British Association of Oral and Maxillofacial Surgeons.

关键词： Effective doses   Fumonisins   Fungicides   Fusarium spp   Predictive mycology   Zearalenone  

摘要： Fusarium-controlling fungicides are necessary to limit crop loss. Little is known about the effect of antifungal formulations at sub-lethal doses, and their interaction with abiotic factors, on Fusarium culmorum and F. proliferatum development and on zearalenone and fumonisin biosynthesis, respectively. In the present study different treatments based on sulfur, trifloxystrobin and demethylation inhibitor fungicides (cyproconazole, tebuconazole and prothioconazole) under different environmental conditions, in Maize Extract Medium, are assayed in vitro. Several machine learning methods (neural networks, random forest and extreme gradient boosted trees) have been applied for the first time for modeling growth of F. culmorum and F. proliferatum and zearalenone and fumonisin production, respectively. The most effective treatment was prothioconazole, 250 g/L + tebuconazole, 150 g/L. Effective doses of this formulation for reduction or total growth inhibition ranged as follows ED50 0.49-1.70, ED90 2.57-6.02 and ED10 0 4.0-8.0 mg/mL, depending on the species, water activity and temperature. Overall, the growth rate and mycotoxin levels in cultures decreased when doses increased. Some treatments in combination with certain a(w) and temperature values significantly induced toxin production. The extreme gradient boosted tree was the model able to predict growth rate and mycotoxin production with minimum error and maximum R-2 value. (C) 2019 Published by Elsevier Ltd on behalf of British Mycological Society.