关键词： 目标达成情况   土力学   大数据   机器学习方法   课程建设  

摘要： 针对土木工程专业工程教育专业认证要求,以重庆大学土力学课程为例,采用大数据技术分析土力学课程目标达成情况,并根据评价结果提出教学改进建议。实践证明,大数据技术能够为科学评价专业课程目标达成情况,以及推动本科工程教育的高质量发展提供新的思路。

关键词： 洪泽湖   小波分析   长短记忆神经网络   模糊神经网络   耦合应用   综合水质预测  

摘要： 洪泽湖是我国第四大淡水湖,又是“南水北调”东线工程中重要的过水通道,它自身的水环境质量对其周边和北方受水区的供水保证至关重要。建立基于小波分析的长短记忆神经网络水质预测模型(WA-LSTM),首先对淮河入洪泽湖的水质数据进行预处理,将其分解为高频和低频信号输入至LSTM模型中进行单因子水质预测,再通过单因子水质预测结果驱动T-S模糊神经网络方法对洪泽湖水质变化情况进行综合预测和评价。研究结果表明:①小波分析方法可以较好地捕捉水质数据特征,其与LSTM水质预测模型结合能够较大幅度提升水质预测的精度;②WA-LSTM模型与模糊神经网络的综合运用能有效解决单因子预测无法体现水质整体状况的问题。研究方法与结论可以为洪泽湖水质监测管理及水资源调控提供技术支撑和实践参考。

关键词： 入侵检测系统   机器学习   数据集   评估指标  

摘要： 入侵检测方法是基于网络的入侵检测系统的核心,可以是基于特征的,也可以是基于异常的。基于特征的检测方法具有较高的检测率,但不能检测到未知新型攻击;基于异常的检测方法可以检测到新型攻击,但误报率较高。为了降低入侵检测的误报率并提高其检测率,许多机器学习技术被应用到入侵检测系统中。通过对大量带有入侵数据训练样本的学习,构建了一个用于区分正常状态和入侵状态的入侵检测模型。针对目前入侵检测系统存在的高误报率、低检测速度和低检测率等问题,对机器学习技术在入侵检测系统中的的优势、系统检测的通用数据集以及系统评估指标进行了详细阐述,并对未来研究趋势进行了展望。

关键词： Alzheimer's disease   artificial intelligence   drug discovery   machine learning   beta-secretase  

摘要： The beta-site amyloid precursor protein cleaving enzyme 1 (BACE1) is a transmembrane aspartyl-protease, that cleaves amyloid precursor protein (APP) at the beta-site. The sequential proteolytic cleavage of APP, first by beta-secretase and then by gamma-secretase complex, leads to the production and release of amyloid-beta peptide, a pathological hallmark of Alzheimer's disease (AD). BACE1 inhibitors are reported to possess considerable potential in decreasing the level of amyloid-beta in brain and preventing the progression of AD. A classification study has been conducted on 3536 diverse BACE1 inhibitors, obtained from Binding DB database, by extracting two types of descriptors, that is molecular property (Mordred) and fingerprints (Pubchem, MACCS and KRFP). Furthermore, based on the descriptors, various machine learning algorithms such as Naive Bayesian (NB), nearest known neighbours (kNN), support vector machine (SVM), random forest (RF) and gradient-boosted algorithms (XGB) were applied to develop classification models. The performance of models was evaluated by using accuracy, precision, recall and F1 score of test set. The best NB, kNN, SVM, RF and XGB classifiers had F1 score of 0.74, 0.85, 0.86, 0.87 and 0.87, respectively. The diverse 3536 BACE1 inhibitors were clustered into 11 subsets, and the structural features of each subset were evaluated. The important fragments present in active and inactive compounds were also identified. The model developed in the study would serve as a valuable tool for the designing of BACE1 inhibitors, and also in virtual screening of molecules to identify these.

关键词： Bifurcation and chaos   Machine learning   Multilayer perceptrons   2D diagrams   Time series identification  

摘要： This paper proposes new methods of computing 2D bifurcation diagrams for nonlinear time series using MultiLayer Perceptrons (MLPs), LSTM Fully Convolutional Networks (LSTM-FCN), Time Series Forests (TSFs) with entropy, Gini impurity, and K-Nearest Neighbors (KNNs) algorithm with Dynamic Time Warping (DTW). The proposed algorithms can precisely compute 2D bifurcation diagrams for oscillatory time-series (periodic or chaotic) obtained either as solutions of nonlinear systems of ordinary differential equations (ODEs) or measured and recorded when a mathematical model is not known. Illustrative computational examples include chaotic electric arc RLC circuits. The obtained results confirm usefulness of the proposed methods in a creation of 2D bifurcation diagrams - color images representing dynamics of nonlinear processes, circuits or systems. (C) 2021 The Author(s). Published by Elsevier B.V.

关键词： Artificial intelligence   Water quality index   Modelling   Sensitivity analysis   Random forest  

摘要： Groundwater quality appraisal is one of the most crucial tasks to ensure safe drinking water sources. Concurrently, a water quality index (WQI) requires some water quality parameters. Conventionally, WQI computation consumes time and is often found with various errors during subindex calculation. To this end, 8 artificial intelligence algorithms, e.g., multilinear regression (MLR), random forest (RF), M5P tree (M5P), random subspace (RSS), additive regression (AR), artificial neural network (ANN), support vector regression (SVR), and locally weighted linear regression (LWLR), were employed to generate WQI prediction in Illizi region, southeast Algeria. Using the best subset regression, 12 different input combinations were developed and the strategy of work was based on two scenarios. The first scenario aims to reduce the time consumption in WQI computation, where all parameters were used as inputs. The second scenario intends to show the water quality variation in the critical cases when the necessary analyses are unavailable, whereas all inputs were reduced based on sensitivity analysis. The models were appraised using several statistical metrics including correlation coefficient (R), mean absolute error (MAE), root mean square error (RMSE), relative absolute error (RAE), and root relative square error (RRSE). The results reveal that TDS and TH are the key drivers influencing WQI in the study area. The comparison of performance evaluation metric shows that the MLR model has the higher accuracy compared to other models in the first scenario in terms of 1, 1.4572*10-08, 2.1418*10-08, 1.2573*10-10%, and 3.1708*10-08% for R, MAE, RMSE, RAE, and RRSE, respectively. The second scenario was executed with less error rate by using the RF model with 0.9984, 1.9942, 3.2488, 4.693, and 5.9642 for R, MAE, RMSE, RAE, and RRSE, respectively. The outcomes of this paper would be of interest to water planners in terms of WQI for improving sustainable management plans of groundwater

关键词： Gas-liquid flow   LASSO   Random Forest   Regression models   Supervised learning  

摘要： Two Machine Learning algorithms - LASSO and Random Forest - are applied to derive regression models for the prediction of gas bubble diameters using supervised learning techniques. Experimental data obtained from wire-mesh sensor (WMS) measurements in a deionized water/air system serve as the data base. Python libraries are used to extract features characterizing WMS measurement signals of single passing bubbles. Prediction accuracy is largely increased with the obtained regression models, compared to well-established methods to predict bubble sizes based on WMS measurements.

关键词： Heat pipe   Artificial neural network   Nanofluid   Thermal performance  

摘要： Accurate modeling of heat pipes, as a two-phase thermal medium, is the main goal of several research works. Since the effective thermal conductivity of the heat pipes is dependent on different factors, the modeling procedure is complicated. The main techniques presented for modeling the heat pipes are numerical methods, i.e., computational fluid dynamics, and machine learning approaches. Due to the simplicity of the use of machine learning methods, these types of models can be more applicable and interesting for scientists compared with numerical models. In this regard, different types of intelligence methods, including Support Vector Machine and Artificial Neural Network, have been employed for determining and estimating the thermal behavior of various kinds of heat pipes. The precision and applicability of these intelligence models depend on different items, such as the input variable, used algorithm, and structure of the model. In the present work, recent studies performed on the applications of intelligence models in the modeling of heat pipes are reviewed, and their primary outcomes are represented. Based on the findings of the studies, intelligence models can estimate the effective heat transfer coefficients of heat pipes reliably. Also, it is concluded that applying the optimization approach in the structure of models, for minimizing the deviation of the predicted values from the actual ones, results in the accuracy enhancement of the models. Finally, some suggestions are provided for future researches concerning the modeling of heat pipes by employing data-driven approaches.

关键词： Evolutionary algorithms   Algorithm recommendation   Automated machine learning   Classification   Meta-learning  

摘要： This paper presents an experimental comparison among four automated machine learning (AutoML) methods for recommending the best classification algorithm for a given input dataset. Three of these methods are based on evolutionary algorithms (EAs), and the other is Auto-WEKA, a well-known AutoML method based on the combined algorithm selection and hyper-parameter optimisation (CASH) approach. The EA-based methods build classification algorithms from a single machine learning paradigm: either decision-tree induction, rule induction, or Bayesian network classification. Auto-WEKA combines algorithm selection and hyper-parameter optimisation to recommend classification algorithms from multiple paradigms. We performed controlled experiments where these four AutoML methods were given the same runtime limit for different values of this limit. In general, the difference in predictive accuracy of the three best AutoML methods was not statistically significant. However, the EA evolving decision-tree induction algorithms has the advantage of producing algorithms that generate interpretable classification models and that are more scalable to large datasets, by comparison with many algorithms from other learning paradigms that can be recommended by Auto-WEKA. We also observed that Auto-WEKA has shown meta-overfitting, a form of overfitting at the meta-learning level, rather than at the base-learning level.

关键词： flood susceptibility   bivariate statistics models   machine learning models   hybrid models  

摘要： Flash floods are considered to be one of the most destructive natural hazards, and they are difficult to accurately model and predict. In this study, three hybrid models were proposed, evaluated, and used for flood susceptibility prediction in the Dadu River Basin. These three hybrid models integrate a bivariate statistical method of the fuzzy membership value (FMV) and three machine learning methods of support vector machine (SVM), classification and regression trees (CART), and convolutional neural network (CNN). Firstly, a geospatial database was prepared comprising nine flood conditioning factors, 485 flood locations, and 485 non-flood locations. Then, the database was used to train and test the three hybrid models. Subsequently, the receiver operating characteristic (ROC) curve, seed cell area index (SCAI), and classification accuracy were used to evaluate the performances of the models. The results reveal the following: (1) The ROC curve highlights the fact that the CNN-FMV hybrid model had the best fitting and prediction performance, and the area under the curve (AUC) values of the success rate and the prediction rate were 0.935 and 0.912, respectively. (2) Based on the results of the three model performance evaluation methods, all three hybrid models had better prediction capabilities than their respective single machine learning models. Compared with their single machine learning models, the AUC values of the SVM-FMV, CART-FMV, and CNN-FMV were 0.032, 0.005, and 0.055 higher;their SCAI values were 0.05, 0.03, and 0.02 lower;and their classification accuracies were 4.48%, 1.38%, and 5.86% higher, respectively. (3) Based on the results of the flood susceptibility indices, between 13.21% and 22.03% of the study area was characterized by high and very high flood susceptibilities. The three hybrid models proposed in this study, especially CNN-FMV, have a high potential for application in flood susceptibility assessment in specific areas in future studies.