关键词： Landslide   Natural hazard   Machine learning   Random forest   Susceptibility  

摘要： Landslides are one of the catastrophic natural hazards that occur in mountainous areas, leading to loss of life, damage to properties, and economic disruption. Landslide susceptibility models prepared in a Geographic Information System (GIS) integrated environment can be key for formulating disaster prevention measures and mitigating future risk. The accuracy and precision of susceptibility models is evolving rapidly from opinion-driven models and statistical learning toward increased use of machine learning techniques. Critical reviews on opinion-driven models and statistical learning in landslide susceptibility mapping have been published, but an overview of current machine learning models for landslide susceptibility studies, including background information on their operation, implementation, and performance is currently lacking. Here, we present an overview of the most popular machine learning techniques available for landslide susceptibility studies. We find that only a handful of researchers use machine learning techniques in landslide susceptibility mapping studies. Therefore, we present the architecture of various Machine Learning (ML) algorithms in plain language, so as to be understandable to a broad range of geoscientists. Furthermore, a comprehensive study comparing the performance of various ML algorithms is absent from the current literature, making an assessment of comparative performance and predictive capabilities difficult. We therefore undertake an extensive analysis and comparison between different ML techniques using a case study from Algeria. We summarize and discuss the algorithm's accuracies, advantages and limitations using a range of evaluation criteria. We note that tree-based ensemble algorithms achieve excellent results compared to other machine learning algorithms and that the Random Forest algorithm offers robust performance for accurate landslide susceptibility mapping with only a small number of adjustments required before training

关键词： drone   anthropogenic debris   OBIA   random forest   support vector machine   k-nearest neighbor  

摘要： Unmanned aerial systems (UASs) have recently been proven to be valuable remote sensing tools for detecting marine macro litter (MML), with the potential of supporting pollution monitoring programs on coasts. Very low altitude images, acquired with a low-cost RGB camera onboard a UAS on a sandy beach, were used to characterize the abundance of stranded macro litter. We developed an object-oriented classification strategy for automatically identifying the marine macro litter items on a UAS-based orthomosaic. A comparison is presented among three automated object-oriented machine learning (OOML) techniques, namely random forest (RF), support vector machine (SVM), and k-nearest neighbor (KNN). Overall, the detection was satisfactory for the three techniques, with mean F-scores of 65% for KNN, 68% for SVM, and 72% for RF. A comparison with manual detection showed that the RF technique was the most accurate OOML macro litter detector, as it returned the best overall detection quality (F-score) with the lowest number of false positives. Because the number of tuning parameters varied among the three automated machine learning techniques and considering that the three generated abundance maps correlated similarly with the abundance map produced manually, the simplest KNN classifier was preferred to the more complex RF. This work contributes to advances in remote sensing marine litter surveys on coasts, optimizing the automated detection on UAS-derived orthomosaics. MML abundance maps, produced by UAS surveys, assist coastal managers and authorities through environmental pollution monitoring programs. In addition, they contribute to search and evaluation of the mitigation measures and improve clean-up operations on coastal environments.

关键词： epoxidation   site of metabolism   machine learning   classification model  

摘要： Epoxidation is one of the reactions in drug metabolism. Since epoxide metabolites would bind with proteins or DNA covalently, drugs should avoid epoxidation metabolism in the body. Due to the instability of epoxide, it is difficult to determine epoxidation experimentally. In silico models based on big data and machine learning methods are hence valuable approaches to predict whether a compound would undergo epoxidation. In this study, we collected 884 epoxidation data manually from various sources, and finally got 829 unique sites of epoxidation. Three types of molecular fingerprints with different lengths (1024, 2048 or 4096 bits) were used to describe the reaction sites. Six machine learning methods were used to build the classification models. The training set and test set were randomly divided into 8 : 2, and 54 models were constructed and evaluated. Four best models were selected for feature selection. The features were then chosen and verified by external validation set. The resulted optimal model had the accuracy and AUC (area under the curve) values at 0.873 and 0.944 for the test set, 0.838 and 0.987 for the external validation set, respectively. The models built in this study could accurately predict whether a compound will undergo epoxidation and which part is most susceptible to epoxidation, which is of great significance for drug design.

关键词： electronic nose   pear brown core   machine learning   neural network   principal component analysis  

摘要： The brown core is an internal disorder that significantly affects the palatability and economic value of Chinese pears. In this study, a framework that includes a back-propagation neural network (BPNN) and extreme learning machine (ELM) (BP-ELMNN) was proposed for the detection of brown core in the Chinese pear variety Huangguan. The odor data of pear were collected using a metal oxide semiconductor (MOS) electronic nose (E-nose). Principal component analysis was used to analyze the complexity of the odor emitted by pears with brown cores. The performances of several machine learning algorithms, i.e., radial basis function neural network (RBFNN), BPNN, and ELM, were compared with that of the BP-ELMNN. The experimental results showed that the proposed framework provided the best results for the test samples, with an accuracy of 0.9683, a macro-precision of 0.9688, a macro-recall of 0.9683, and a macro-F1 score of 0.9685. The results demonstrate that the use of machine learning algorithms for the analysis of E-nose data is a feasible and non-destructive method to detect brown core in pears.

关键词： low-frequency sensor   power transformer   machine learning   low-frequency noise   genetic algorithm  

摘要： The paper proposes a method of automatic detection of parameters of a distribution transformer (model, type, and power) from a distance, based on its low-frequency noise spectra. The spectra are registered by sensors and processed by a method based on evolutionary algorithms and machine learning. The method, as input data, uses the frequency spectra of sound pressure levels generated during operation by transformers in the real environment. The model also uses the background characteristic to take under consideration the changing working conditions of the transformers. The method searches for frequency intervals and its resolution using both a classic genetic algorithm and particle swarm optimization. The interval selection was verified using five state-of-the-art machine learning algorithms. The research was conducted on 16 different distribution transformers. As a result, a method was proposed that allows the detection of a specific transformer model, its type, and its power with an accuracy greater than 84%, 99%, and 87%, respectively. The proposed optimization process using the genetic algorithm increased the accuracy by up to 5%, at the same time reducing the input data set significantly (from 80% up to 98%). The machine learning algorithms were selected, which were proven efficient for this task.

关键词： Supervised machine learning   Wake dynamics   Turbulence  

摘要： We apply supervised machine learning techniques to a number of regression problems in fluid dynamics. Four machine learning architectures are examined in terms of their characteristics, accuracy, computational cost, and robustness for canonical flow problems. We consider the estimation of force coefficients and wakes from a limited number of sensors on the surface for flows over a cylinder and NACA0012 airfoil with a Gurney flap. The influence of the temporal density of the training data is also examined. Furthermore, we consider the use of convolutional neural network in the context of super-resolution analysis of two-dimensional cylinder wake, two-dimensional decaying isotropic turbulence, and three-dimensional turbulent channel flow. In the concluding remarks, we summarize on findings from a range of regression-type problems considered herein.

关键词： Convolutional neural network   Siamese neural network   Trademark similarity assessment   Trademark (TM) infringement   Vector space model  

摘要： The rapid development of consumer products with short life spans, along with fast, global e-commerce and e-marketing distribution of products and services requires greater due diligence to protect intangible assets such as brands and corporate logos which can easily be copied or distributed through grey channels and internet sales sites. Trademarks (TMs) are government registered intellectual property rights (IPRs) used to legally protect a companies' identities and brand equity. The rapid growth of global trademark (TM) registrations and the number of TM infringement cases pose a great challenge for TM owners to detect infringement and take action to protect TMs, consumer trust, and market share. This research develops advanced TM similarity assessment models using machine learning (ML) approaches. Litigation principles over similarity follow US TM laws which are consistent with global TM protection convention under the World Intellectual Property Organization (WIPO). This research covers the similarity analysis of TM spelling, pronunciation, and images, which are most likely to cause TM confusion among customers. The research focuses on deploying machine learning for natural language (spelling and phonetic features) and image similarity analyses. The vector space modeling algorithms are trained and verified for the similarity analysis of TM wordings in both spelling and pronunciation. The convolutional neural network and Siamese neural network models are trained and verified for TM image similarity comparison. The training and testing sets consist of 250,000 and 20,000 different image pairs respectively. This research provides a significant contribution toward implementing intelligent and automated IPR protection. The system solution supports users (companies, TM attorneys, or IP officers) to identify similar registered TMs before registering new TMs ensuring uniqueness to avoid infringement disputes. The solution also supports automatic screening of online conten

关键词： Attack detection   Genetic algorithm   Logistic regression   Artificial neural network   Particle swarm optimization   Feature selection  

摘要： Advances in computer networks led to the generation of much data that computer networks must be capable of transmitting. The security of this volume of data is a major challenge for companies. Intrusion detection systems is one of the solutions that researchers introduced for this challenge. This research aims to introduce a new machine learning model for intrusion detection. The proposed model includes two stages of feature selection and attack identification. The feature selection stage uses genetic algorithm and logistic regression algorithm to find a correlated subset of features. In the attack detection phase, the ANN algorithm is used. ANN is trained by particle optimization (PSO) and gravitational search (GS) algorithms. To evaluate the proposed model, two sets of NSL-KDD and KDD Cup'99 are used and results are compared with ANN based on gradient descent (GD-ANN) and decision tree, ANN based on genetic algorithm (GA-ANN) methods, ANN based on GSPSO (GSPSO-ANN), ANN based on PSO (PSO-ANN) and ANN based on GS (GS-ANN) indicate the superiority of the proposed method.

关键词： deep learning   financial distress prediction   machine learning   management factors   textual factors  

摘要： Financial distress prediction (FDP) has been widely considered as a promising approach to reducing financial losses. While financial information comprises the traditional factors involved in FDP, nonfinancial factors have also been examined in recent studies. In light of this, the purpose of this study is to explore the integrated factors and multiple models that can improve the predictive performance of FDP models. This study proposes an FDP framework to reveal the financial distress features of listed Chinese companies, incorporating financial, management, and textual factors, and evaluating the prediction performance of multiple models in different time spans. To develop this framework, this study employs the wrapper-based feature selection method to extract valuable features, and then constructs multiple single classifiers, ensemble classifiers, and deep learning models in order to predict financial distress. The experiment results indicate that management and textual factors can supplement traditional financial factors in FDP, especially textual ones. This study also discovers that integrated factors collected 4 years prior to the predicted benchmark year enable a more accurate prediction, and the ensemble classifiers and deep learning models developed can achieve satisfactory FDP performance. This study makes a novel contribution as it expands the predictive factors of financial distress and provides new findings that can have important implications for providing early warning signals of financial risk.

关键词： Magnetic resonance imaging   Lung cancer   Radiomics   Machine learning  

摘要： Objectives We develop and validate a radiomics model based on multiparametric magnetic resonance imaging (MRI) in the classification of the pulmonary lesion and identify optimal machine learning methods. Materials and methods This retrospective analysis included 201 patients (143 malignancies, 58 benign lesions). Radiomics features were extracted from multiparametric MRI, including T2-weighted imaging (T2WI), T1-weighted imaging (TIWI), and apparent diffusion coefficient (ADC) map. Three feature selection methods, including recursive feature elimination (RFE), t test, and least absolute shrinkage and selection operator (LASSO), and three classification methods, including linear discriminate analysis (LDA), support vector machine (SVM), and random forest (RF) were used to distinguish benign and malignant pulmonary lesions. Performance was compared by AUC, sensitivity, accuracy, precision, and specificity. Analysis of performance differences in three randomly drawn cross-validation sets verified the stability of the results. Results For most single MR sequences or combinations of multiple MR sequences, RFE feature selection method with SVM classifier had the best performance, followed by RFE with RF. The radiomics model based on multiple sequences showed a higher diagnostic accuracy than single sequence for every machine learning method. Using RFE with SVM, the joint model of T1WI, T2WI, and ADC showed the highest performance with AUC = 0.88 +/- 0.02 (sensitivity 83%;accuracy 82%;precision 91%;specificity 79%) in test set. Conclusion Quantitative radiomics features based on multiparametric MRI have good performance in differentiating lung malignancies and benign lesions. The machine learning method of RFE with SVM is superior to the combination of other feature selection and classifier methods.