关键词： Carbonate reservoirs   Shear wave velocity prediction   Petrophysical model   Wireline logs   LSTM  

摘要： Shear wave velocity (S-wave velocity or Vs) is one of the most critical issues for carbonate reservoirs characterization, because of its complexity of rock compositions and pore structures. The Xu-Payne petrophysical model is a commonly used method to predict S-wave velocity. However, the model excessively depends on the interpretation accuracy of rock compositions, pore structures, fluid properties and fluid saturation, which makes the prediction result potentially uncertain. With the development of intelligent algorithms, a machine learning method (Long-short Term Memory Neural Network, LSTM) is proposed to improve the traditional petrophysical workflow. With the capabilities of deep learning and data mining, the LSTM model can deeply mine the rich information in the wireline logs and then establish the relationships between S-wave velocity and wireline logs. In order to illustrate the prediction effect, the carbonate reservoir in the Majiagou Formation Member 5(Ma 5 Member) which belongs to the Ordovician system in the D area of Sulige gas field is taken as an example. Six kinds of sensitive wireline logs, including compensated acoustic log (AC), natural gamma ray log(GR), photoelectric absorption cross-section log (PE), density log (DEN), deep lateral resistivity log(RLLD), neutron log(CNL), are selected as the input data of the LSTM model. The results reveal that the accuracy of LSTM method is up to 98.9%, whereas the Xu-Payne model is only 73%. With the predicted result of all wells in the areas using LSTM, the sensitive elastic properties, including P-and S-wave velocity, P- and S-wave impendences, ratio of P- and S-wave velocity, Poisson's ratio, lame coefficient and fluid sensitivity factor, are also concluded. Compared with the Xu-Payne model, the newly proposed workflow is proved to be convenient and suitable for the prediction of S-wave velocity in carbonate reservoirs.

关键词： Sustainable smart cities   Security   Traffic   Classification   Data marling   Machine learning   A survey  

摘要： This survey paper describes the significant literature survey of Sustainable Smart Cities (SSC), Machine Learning (ML), Data Mining (DM), datasets, feature extraction and selection for network traffic classification. Considering relevance and most cited methods and datasets of features were identified, read and summarized. As data and data features are essential in Internet traffic classification using machine learning techniques, some well-known and most used datasets with details statistical features are described. Different classification techniques for SSC network traffic classification are presented with more information. The complexity of data set, features extraction and machine learning methods are addressed. In the end, challenges and recommendations for SSC network traffic classification with the dataset of features are presented.

关键词： machine learning   remote sensing   forest stand characteristics   Bayesian additive regression tree  

摘要： The estimation and mapping of forest stand characteristics are vital because this information is necessary for sustainable forest management. The present study considers the use of a Bayesian additive regression trees (BART) algorithm as a non-parametric classifier using Sentinel-2A data and topographic variables to estimate the forest stand characteristics, namely the basal area (m(2)/ha), stem volume (m(3)/ha), and stem density (number/ha). These results were compared with those of three other popular machine learning (ML) algorithms, such as generalised linear model (GLM), K-nearest neighbours (KNN), and support vector machine (SVM). A feature selection was done on 28 variables including the multi-spectral bands on Sentinel-2 satellite, related vegetation indices, and ancillary data (elevation, slope, and topographic solar-radiation index derived from digital elevation model (DEM)) and then the most insignificant variables were removed from the datasets by recursive feature elimination (RFE). The study area was a mountainous forest with high biodiversity and an elevation gradient from 26 to 1636 m. An inventory dataset of 1200 sample plots was provided for training and testing the algorithms, and the predictors were fed into the ML models to compute and predict the forest stand characteristics. The accuracies and certainties of the ML models were assessed by their root mean square error (RMSE), mean absolute error (MAE), and R-squared (R-2) values. The results demonstrated that BART generated the best basal area and stem volume predictions, followed by GLM, SVM, and KNN. The best RMSE values for both basal area (8.12 m(2)/ha) and stem volume (29.28 m(3)/ha) estimation were obtained by BART. Thus, the ability of the BART model for forestry application was established. On the other hand, KNN exhibited the highest RMSE values for all stand variable predictions, thereby exhibiting the least accuracy for this specific application. Moreover, the effectiveness of the na

关键词： Database systems  

摘要： Superhard materials are of great interest in various practical applications, and an increasing number of research efforts are focused on their development. In this article, we demonstrate that machine learning can be successfully applied to searching for such materials. We construct a machine learning model using neural networks on graphs together with a recently developed physical model of hardness and fracture toughness. The model is trained using available elastic data from the Materials Project database and has good accuracy for predictions. We use this model to screen all crystal structures in the database and systematize all the promising hard or superhard materials, and find that diamond (and its polytypes) are the hardest materials in the database. Our results can be further used for the investigation of interesting materials using more accurate ab initio calculations and/or experiments.

关键词： Hyperspectral image   Near infrared   Cannabis sativa Linnaeu   Sparse-Principal Component Analysis   Soft Independent Class Analogy model  

摘要： Remote identification of illegal plantations of Cannabis sativa Linnaeus is an important task for the Brazilian Federal Police. The current analytical methodology is expensive and strongly dependent on the expertise of the forensic investigator. A faster and cheaper methodology based on automatic methods can be useful for the detection and identification of Cannabis sativa L. in a reliable and objective manner. In this work, the high potential of Near Infrared Hyperspectral Imaging (HSI-NIR) combined with machine learning is demonstrated for supervised detection and classification of Cannabis sativa L. This plant, together with other plants commonly found in the surroundings of illegal plantations and soil, were directly collected from an illegal plantation. Due to the high correlation of the NIR spectra, sparse Principal Component Analysis (sPCA) was implemented to select the most important wavelengths for identifying Cannabis sativa L. One class Soft Independent Class Analogy model (SIMCA) was built, considering just the spectral variables selected by sPCA. Sensitivity and specificity values of 89.45% and 97.60% were, respectively, obtained for an external validation set subjected to the s-SIMCA. The results proved the reliability of a methodology based on NIR hyperspectral cameras to detect and identify Cannabis sativa L., with only four spectral bands, showing the potential of this methodology to be implemented in low-cost airborne devices. (C) 2020 Elsevier B.V. All rights reserved.

关键词： Agricultural systems   Change monitoring   Phenological metrics   NDVI time series   Machine learning   MODIS  

摘要： This study aims to provide accurate information about changes in agricultural systems (AS) using phenological metrics derived from the NDVI time series. Use of such information could help land managers optimize land use choices and monitor the status of agricultural lands, under a variety of environmental and socioeconomic conditions. For this purpose, the Moderate Resolution Imaging Spectroradiometer (MODIS) NDVI data were used to derive phenological metrics over the Oum Er-Rbia basin (central Morocco). Random forest (RF), support vector machine (SVM), and K-nearest neighbor (KNN) classifiers were explored and compared on their ability to classify AS classes over the study area. Four main AS classes have been considered: (1) irrigated annual crop (IAC), (2) irrigated perennial crop (IPC), (3) rainfed area (RA), and (4) fallow (FA). By comparing the accuracy of the three classifiers, the RF method showed the best performance with an overall accuracy of 0.97 and kappa coefficient of 0.96. The RF method was then chosen to examine time variations in AS over a 16-year period (2000-2016). The AS main variations were detected and evaluated for the four AS classes. These variations have been found to be linked well with other indicators of local agricultural land management, as well as the historical agricultural drought changes over the study area. Overall, the results present a tool for decision makers to improve agricultural management and provide a different perspective in understanding the spatiotemporal dynamics of agricultural systems.

关键词： Green ports   Energy consumption prediction of ships   Energy consumption reduction strategy   Machine learning  

摘要： Two main contributions of this paper are 1) the energy consumption of ships (ECS) in port is predicted, 2) reduction strategies for energy consumption of ships in port are discussed by the proposed prediction models considering green port. Firstly, 15 characteristics which have impact on the energy consumption of ships are collected by Jingtang Port in China and analysis is conducted. Then, five machine learning models including Gradient Boosting Regression (GBR), Random Forest Regression (RF), BP Network (BP), Liner Regression (LR) and K-Nearest Neighbor Regression (KNN) are developed and 15 features consisting of inherent property of ship and external features of ports are set as inputs. After then, k-folds cross validation is adopted to verify the effectiveness of models. Finally, the feature importance is calculated and the most important features are selected. Besides, experiments are conducted to find the effect of changing several features on energy consumption of ships, and two consumption reduction strategies are discussed. The results show that net tonnage, deadweight tonnage, actual weight and efficiency of facilities are the top 4 features for predicting the energy consumption of ships. In conclusion, when efficiency of facilities is doubled, the energy consumption of ships is reduced by 34.17% at berth and 8.41% in port. The finding of proposed methods and discussed strategies can give references to green port construction. (C) 2020 Elsevier Ltd. All rights reserved.

关键词： Unsupervised learning  

摘要： In general, sea farms are classified into two types depending on their locations, seawater depths, and cultivation species: underwater and above water farms. This study compared the machine learning techniques (random forest, a supervised learning technique, and K-means, an unsupervised learning technique) for mapping both underwater and above water farms using the high-resolution satellite image acquired in the sea farming areas of the South Sea of South Korea through the following steps. First, each machine learning algorithm was separately used in the given high-resolution satellite image to generate a sea surface map. Then underwater and above water farms were detected from each sea surface map. Finally, the accuracy of both the underwater and above water farms detected from the different sea surface maps was assessed. The experimental results led to the following conclusions. First, random forest, a supervised learning technique, has better performance than K-means, an unsupervised learning technique, for detecting both underwater and above water farms from the given high-resolution satellite image. Second, only a few misclassification errors occurred in both the underwater and above water farms detected by the random forest algorithm in the given image, with the underwater farms misclassified as having above water features, or vice versa.

关键词： 键盘输入   机器学习方法   自动更正   训练算法   数据样本   谷歌   卫生系统   保护隐私  

摘要： 为了解决可应用于更广泛患者群体的医学问题,机器学习模型依赖于来自各种机构的大型、多样的数据集。但是,由于法律、隐私和文化方面的限制,卫生系统和医院通常难以共享患者数据。近日发表在Scientific Reports杂志上的一项研究表明,一种称为联合学习(Federated Learning)的新兴技术可以解决这一难题。联合学习是谷歌首创的用于键盘输入自动更正的一种方法,它在多个分散的设备或服务器上训练算法,这些设备或服务器保存本地数据样本,而无需进行数据样本交换。

关键词： 机器学习模型   决策树   K近邻   支持向量机   母猪   产仔数性状  

摘要： 为指导养猪生产者更好地预测母猪的产仔数性状,尽早淘汰繁殖力较差的母猪,提升母猪群体的繁殖潜力,对记录了母猪总产仔数、产活仔数、健仔数、5日龄仔猪数和1 kg以上仔猪数的生产数据进行处理和描述统计,使用R软件中的Boruta包筛选出影响母猪产仔数性状的重要特征如品种、胎次、配种季节等,利用传统回归分析方法(LR)和不同机器学习方法—决策树(decision tree,DT)、K近邻(K-nearest neighbor,KNN)、支持向量机(support vector machine,SVM)对产仔数性状进行回归分析,最后比较机器学习方法与传统回归方法建模的优劣。结果显示,母猪总产仔数、产活仔数、健仔数、5日龄仔猪数和1 kg以上仔猪数不同回归分析方法的R^2均达到0.71以上(0.71~0.88),体现了特征选择的正确性;在预测母猪总产仔数、产活仔数、健仔数、5日龄仔猪数和1 kg以上仔猪数中SVM模型均显著优于其他机器学习模型(P<0.05)并且要优于传统回归方法,而且在以上模型中预测1 kg以上仔猪数的SVM模型最优。因此,在今后的养猪生产中机器学习方法可能会成为养猪生产者早期选育高繁殖力母猪的一种新途径。