基于信息量与神经网络模型的滑坡易发性评价

针对神经网络模型进行滑坡易发性评价时,传统的随机选取非滑坡单元存在准确性不高的缺点,提出信息量与神经网络结合的易发性评价模型。以江西省上犹县为研究区,首先,基于上犹县滑坡编录与实际调查,选取坡度、高程、坡向、平面曲率、剖面曲率,植被指数(NDVI)、湿度指数(TWI)、距水系距离、距道路距离、土地利用等10个环境因子,其次利用信息量模型对上犹县进行易发性分区,得到上犹县易发性分区图。然后,从信息量模型得出的易发性分区中的低易发区选取非滑坡单元,与滑坡编录中的历史滑坡点组成测试集与训练集,输入神经网络中训练模型,再将上犹县所有栅格输入,预测上犹县栅格的滑坡概率。最后利用自然断点法在上犹县栅格滑坡概率进行分类,得到基于信息量与人工神经网络结合的上犹县易发性分区图。由易发性结果表明:单独的信息量模型的成功率曲线下面积AUC=0.7364,历史灾害点位于高易发区与较高易发区的灾害数占总灾害数的55.6%;基于信息量与神经网络模型的AUC=0.7874;历史灾害点位于高易发区与较高易发区的灾害数占总灾害数的85.8%。信息量–神经网络的评价模型比单独的信息量模型的评价精度提高了5.1%;高易发区与较高易发区所涵盖的灾害数占比高30.2%。信息量–神经网络模型有更好的评价精度,并且证明了在信息量模型中的极低易发区选取非滑坡点具有可行性。     

异质集成学习在滑坡易发性评价中的对比研究

滑坡易发性评价可以为有关部门制定滑坡相关的防灾减灾政策提供技术支撑。目前机器学习已被广泛应用于滑坡易发性评价中，而不同的机器学习模型预测精度各异，为对比分析不同异质集成学习模型在滑坡易发性评价中的精度表现，该文以滑坡多发的甘肃省天水市与陕西省宝鸡市交界处为研究区，采用Stacking、Blending和加权平均三种异质集成学习模型，以随机森林、支持向量机和BP神经网络作为基学习器，对研究区进行滑坡易发性评价对比研究。通过使用准确率、Kappa系数以及ROC曲线指标对三种异质集成模型及基学习器进行模型验证和对比分析，结果表明，Stacking集成模型的各项指标都优于其他对比模型，验证了Stacking集成模型在滑坡易发性评价方面较其他对比模型具有更高的预测精度。     

九寨沟县滑坡灾害易发性快速评估模型对比研究EI北大核心CSCD

通过地理信息系统(GIS)技术,采用信息量(I)、确定性系数(CF)、逻辑回归(LR)、逻辑回归–信息量(LR-I)和逻辑回归–确定性系数(LR-CF)耦合模型的快速评估方法对九寨沟县范围内滑坡灾害易发性评价,并对5种模型进行比较研究。基于历史资料、遥感解译和现场调查,获取九寨沟全县6205个滑坡灾害点作为样本数据库,选取海拔、坡度、坡向、地形曲率、剖面曲率、平面曲率、地形起伏度、地表粗糙度、地表切割度、地层岩性、距断层距离、PGA、降雨、距公路距离和距水系距离共计15项评价指标因子,基于GIS平台提取80%滑坡点作为训练样本,采用I,CF,LR模型建立九寨沟县滑坡灾害易发性评价体系,并将滑坡易发性划分为极低、低、中、高和极高。基于I,CF和LR模型,提出LR-I和LR-CF耦合模型,实现各评价指标因子二次逻辑回归计算,优化了九寨沟地区滑坡灾害易发性区划图。最后利用未参与训练的20%滑坡点作为检验样本,利用频率比和ROC曲线进行精度检验。结果表明:5种评价模型得到的滑坡的高易发区和极高易发区频率比值占总频率比值均超过85%,I,CF,LR,LR-I和LR-CF的AUC评价精度分别为0.762,0.756,0.788,0.838和0.836,表明5种模型均能较好评价九寨沟地区滑坡灾害易发性。LR-I和LR-CF模型与单一的I,CF模型相比能将滑坡易发性评价精度提高约8%;与单一LR模型相比,其精度提高约5%,说明LR-I和LR-CF模型的滑坡预测更优于单一的I,CF和LR模型,为快速建立评价指标体系和区域滑坡易发性提供了可靠途径。     

基于GBDT模型的广东阳春市地质灾害易发性评价研究

本研究旨在利用梯度提升决策树（Gradient Boosting Decision Tree,GBDT）模型,对广东省阳春市滑坡易发性进行评价。通过收集大量地质、地形、气象等相关数据,提取了坡度、坡向、工程岩组等11个地质灾害易发性评价指标,构建了全面的滑坡易发性评价指标体系,并采用GBDT模型进行训练和预测。受试者曲线（Receiver Operator Characteristic,ROC曲线）和AUC值（Area Under Curve,AUC）被用于评估模型的准确性,研究结果表明,模型的AUC值达到了0.9414,说明GBDT模型在阳春市滑坡易发性评价中表现出较高的准确性和可靠性。易发性分区统计结果显示,整个阳春市中,高易发区占4.98%,中易发区占8.42%,低易发区占16.39%,非易发区占70.22%。本文研究方法可为开展区域地质灾害易发性评价提供参考。     

顾及In SAR形变的CNN滑坡易发性动态评估——以刘家峡水库区域为例

已有的滑坡易发性评估方法所采用的滑坡因子多为静态数据(如地形,地质),缺乏动态数据(如地表形变),无法充分挖掘正在变形的滑坡特征,导致滑坡易发性评估可靠性较差。合成孔径雷达干涉测量(synthetic aperture radar interferometry,In SAR)二维In SAR数据可反映滑坡在垂直和水平方向的形变特征。引入二维In SAR形变数据作为动态因子,结合地形、地质、水文以及人文共计16种滑坡影响因子,构建卷积神经网络模型(convolutional neural network,CNN)进行滑坡易发性动态评估;采用多种评价指标来衡量模型精度,同时对比有无顾及二维In SAR因子的滑坡易发性评估结果,并与支持向量机(support vector machines,SVM)方法进行对比,此外,在不同场景进行应用。实验结果表明,顾及In SAR形变动态因子的多种评价指标整体精度有所提高,模型对正处于缓慢变形的滑坡易发性区域的识别具有良好的效果,在训练数据中,识别率由0.78提高至0.93;在验证数据中,识别率提升0.21,揭示出顾及二维In SAR形变因子可以提升滑坡...     

基于信息量法的金沙江峡谷段滑坡易发性评价

文章以云南省迪庆藏族自治州金沙江峡谷区为研究区域,进行滑坡易发性评价。选取坡度、起伏度、植被覆盖指数、地层岩性、构造作用、河流作用、人类工程活动、降雨量八个影响因子。以266个滑坡点为样本,基于信息量法和GIS平台,构建滑坡易发性分区评价指标体系,计算各个影响因子分级状态下的信息量值,绘制各影响因子分级栅格图,通过栅格计算器叠加得到总区域信息量值,采用自然断点法将其分为5级。采用成功率曲线法进行模型评价,曲线下面积(AUC值)为0.7768,准确性较好。     

基于GIS和信息量的滑坡灾害易发性评价——以三峡库区万州区为例

 以滑坡灾害发育较多的三峡库区万州区为研究区,基于指标因素状态分级和因素相关性分析结果,选取坡度、坡向、坡体结构、地层岩性、地质构造、水的作用以及土地利用7项影响因素,以全区700多个滑坡灾害点为样本数据,依据各因素状态下发生的滑坡频率曲线和信息量曲线的突变点为等级划分的临界值来确定因素状态,并在此基础上建立易发性评价指标体系。基于GIS的栅格数据模型,应用信息量理论开展研究区易发性评价,研究结果表明：易发性高和较高的区域主要分布在土地利用总体规划中的建设用地、侏罗系中统上沙溪庙组第二、三段(J2s2,J2s3)、库水变动带和河网影响带以及万州城区。统计结果表明,处在高易发和较高易发区面积为1 210 km2,其中高易发区和较高易发区分别占研究区总面积的9.71%和25.9%,研究区易发性评价精度高达87%。本文完整的论述了县域滑坡灾害易发性评价的理论方法和技术路线,并以三峡库区万州区为例开展滑坡灾害易发性评价、结果分析以及预测精度评价等,为该区域滑坡灾害防治规划与预测预报提供技术支持,为全国范围内县域滑坡灾害易发性评价提供理论指导和技术参考。     

镇康县滑坡易发性评价模型对比研究

准确的滑坡易发性评价结果是滑坡风险评价的关键,对防灾减灾意义重大.为了提高滑坡易发性评价精度,基于地理信息系统(GIS)平台,将镇康县境内150个滑坡灾害点转换为栅格数据作为评价样本,选取高程、坡度、坡向、起伏度、地形曲率、剖面曲率、地层、断层、年均降雨量、河流、土地利用类型、道路12项评价因子,并通过独立性检验,构建...     

汶川县地震滑坡易发性LR与NN评价比较研究

通过地理信息系统(GIS)技术,结合神经网络(NN)和逻辑回归模型(LR)开展汶川县范围内地震诱发滑坡易感性评价,并对两种模型结果进行比较研究。基于2008年5.12 Ms8.0级地震,选取高程、坡度、坡位、坡向、岩性、微地貌、距断层距离、距水系距离、距道路距离、年平均降雨量、归一化植被指数、地震峰值加速度共12个因子作为地震滑坡影响因子,基于ARCGIS10.1平台将这些影响因子专题图层栅格化;采用提取的模型训练样本,由R软件对神经网络(NN)和逻辑回归模型(LR)进行训练;将训练好的模型对整个汶川县地震滑坡易感性进行仿真,并将仿真结果划分为五类滑坡敏感区域:极低,低,中,高和极高,分别得到LR与NN模型仿真的滑坡易发性分区图;根据汶川县实际地震滑坡分布图进行统计分析,以及采用ROC曲线对两种模型的仿真结果进行对比分析,神经网络(NN)和逻辑回归模型(LR)的AUC值分别为0.930和0.941。研究表明两种模型的滑坡易感性评价图与实际滑坡发育基本吻合,评价结果较好,且LR模型预测精度相对较高。     

三峡库区万州区滑坡灾害易发性评价研究

 滑坡灾害易发性研究在滑坡灾害风险管理与城市规划等方面具有非常重要的现实意义。以往的研究中,鲜有对指标因子状态划分作有关深入分析和讨论的。鉴于此,以滑坡灾害频发的三峡库区万州区为研究对象：首先,选取影响滑坡发生的7个致灾因子(地层岩性、地质构造、水系分布、坡度、坡向、坡体结构及土地利用)作为滑坡易发性的评价指标,依据各指标条件下滑坡累计发生频率曲线斜率的变化,并结合滑坡面积比和分级面积比曲线对指标因子的状态进行分级;其次,根据全区655个历史滑坡数据,分别运用信息量模型和逻辑回归模型建立各自的滑坡易发性评价体系;再则,采用快速聚类法(K-means cluster)对以上2种方法所得到的易发性结果进行分级,并基于GIS平台,得到全区滑坡易发性区划图;最后,从模型结果、精度、适用条件等方面对2个模型进行讨论和比较,研究结果表明：信息量模型和逻辑回归模型的预测精度分别为73.0%和54.9%,前者预测能力要优于后者。     

Evaluating landslide susceptibility based on cluster analysis,probabilistic methods,and artificial neural networks

In this study, the cluster analysis (CA), probabilistic methods, and artificial neural networks (ANNs) are used to predict landslide susceptibility. The Geographic Information System (GIS) is used as the basic tool for spatial data management. CA is applied to select non-landslide dataset for later analysis. A probabilistic method is suggested to calculate the rating of the relative importance of each class belonging to each conditional factor. ANN is applied to calculate the weight (i.e., relative importance) of each factor. Using the ratings and the weights, it is proposed to calculate the landslide susceptibility index (LSI) for each pixel in the study area. The obtained LSI values can then be used to construct the landslide susceptibility map. The aforementioned proposed method was applied to the Longfeng town, a landslide-prone area in Hubei province, China. The following eight conditional factors were selected: lithology, slope angle, distance to stream/reservoir, distance to road, stream power index (SPI), altitude, curvature, and slope aspect. To assess the conditional factor effects, the weights were calculated for four cases, using 8 factors, 6 factors, 5 factors, and 4 factors, respectively. Then, the results of the landslide susceptibility analysis for these four cases, with and without weighting, were obtained. To validate the process, the receiver operating characteristics (ROC) curve and the area under the curve (AUC) were applied. In addition, the results were compared with the existing landslide locations. The validation results showed good agreement between the existing landslides and the computed susceptibility maps. The results with weighting were found to be better than that without weighting. The best accuracy was obtained for the case with 5 conditional factors with weighting.

     

A novel hybrid intelligent model of support vector machines and the MultiBoost ensemble for landslide susceptibility modeling

The main aim of this study is to propose a novel hybrid intelligent model named MBSVM which is an integration of the MultiBoost ensemble and a support vector machine (SVM) for modeling of susceptibility of landslides in the Uttarakhand State, Northern India. Firstly, a geospatial database for the study area was prepared, which includes 391 historical landslides and 16 landslide-affecting factors. Then, the sensitivity of different combinations of these factors for modeling was validated using the forward elimination technique. The MBSVM landslide model was built using the datasets generated from the best selected factors and validated utilizing the area under the receiver operating characteristic (ROC) curve (AUC), statistical indexes, and the Wilcoxon signed-rank test. Results show that this novel hybrid model has good performance both in terms of goodness of fit with the training dataset (AUC = 0.972) and the capability to predict landslides with the testing dataset (AUC = 0.966). The efficiency of the proposed model was then validated by comparison with logistic regression (LR), a single SVM, and another hybrid model of the AdaBoost ensemble and an SVM (ABSVM). Comparison results show that the MBSVM outperforms the LR, single SVM, and hybrid ABSVM models. Thus, the proposed model is a promising and good alternative tool for landslide hazard assessment in landslide-prone areas.

     

Enhancing the accuracy of rainfall-induced landslide prediction along mountain roads with a GIS-based random forest classifier

Along mountain roads, rainfall-triggered landslides are typical disasters that cause significant human casualties. Thus, to establish effective mitigation measures, it would be very useful were government agencies and practicing land-use planners to have the capability to make an accurate landslide evaluation. Here, we propose a machine learning methodology for the spatial prediction of rainfall-induced landslides along mountain roads which is based on a random forest classifier (RFC) and a GIS-based dataset. The RFC is used as a supervised learning technique to generalize the classification boundary that separates the input information of ten landslide conditioning factors (slope, aspect, relief amplitude, toposhape, topographic wetness index, distance to roads, distance to rivers, lithology, distance to faults, and rainfall) into two distinctive class labels: ‘landslide’ and ‘non-landslide’. Experimental results with a cross validation process and sensitivity analysis on the RFC model parameters reveal that the proposed model achieves a superior prediction accuracy with an area under the curve  of 0.92. The RFC significantly outperforms other benchmarking methods, including discriminant analysis, logistic regression, artificial neural networks, relevance vector machines, and support vector machines. Based on our experimental outcome and comparative analysis, we strongly recommend the RFC as a very capable tool for spatial modeling of rainfall-induced landslides.

     

基于时间序列与PSO-SVR耦合模型的白水河滑坡位移预测研究

 滑坡位移预测模型的选择及其参数的选取是滑坡位移预测中至关重要的2个问题,以往的模型在预测滑坡位移时具有诸多的限制和不足。以三峡库区白水河滑坡为研究对象,基于时间序列分析方法,分离提取出滑坡趋势性位移与周期性位移。前者主要受控于滑坡的内部因素(物质组成、地质构造、地形地貌等),可用最小二乘法对其进行多项式拟合并预测;后者是由外部影响因素(季节性降雨、库水位变动等)导致,取当月降雨量、双月降雨量、库水位高程、月间库水位变化量、双月库水位变化量和年间累计位移增量作为周期性位移的影响因子,提出采用可优化选参的粒子群优化算法(PSO)与支持向量机回归(SVR)相结合的方法对其进行预测;将各分位移预测值叠加得到累计位移预测值。运用多种方法进行分析对比,结果表明,基于时间序列与PSO-SVR耦合模型的滑坡位移预测精度要明显高于BP神经网络模型、网格搜索法优化的支持向量机模型(GS),其在滑坡位移预测中具有一定的理论基础和良好的应用前景。     

A case study from Koyulhisar (Sivas-Turkey) for landslide susceptibility mapping by artificial neural networks

A case study for the use of an artificial neural network (ANN) model for landslide susceptibility mapping in Koyulhisar (Sivas-Turkey) is presented. Digital elevation model (DEM) was first constructed using ArcGIS software. Relevant parameter maps were created, including geology, faults, drainage system, topographical elevation, slope angle, slope aspect, topographic wetness index, stream power index, normalized difference vegetation index and distance from roads. Finally, a landslide susceptibility map was constructed using the neural networks. The drawbacks of the method are discussed but as the validation procedures used confirmed the quality of the map produced, it is recommended the use of ANN may be helpful for planners and engineers in the initial assessment of landslide susceptibility.      

Exploring spatial non-stationarity in the relationships between landslide susceptibility and conditioning factors: a local modeling approach using geographically weighted regression

Landslide susceptibility is the likelihood of landslide occurrence, in a specific place and time. The identification of the potential relationships between landslide susceptibility and conditioning factors is very important towards landslide hazard mitigation. In this paper, we implement a local statistical analysis model geographically weighted regression, in two catchment areas located in northern Peloponnese, Greece. For this purpose, we examined the following eight conditioning factors: elevation, slope, aspect, lithology, land cover, proximity to the drainage network, proximity to the road network, and proximity to faults. Moreover, the relationship between these factors and landsliding in the study area is examined. The local statistical analysis model was also evaluated by finding its differences with the performance of a standard global statistical model logistic regression. The results indicated that the global statistical model can be enhanced by the application of a local model. The outputs of the proposed approach favored a better understanding of the factors influencing landslide occurrence and may be beneficial to local authorities and decision-makers dealing with the mitigation of landslide hazard.

     

Landslide susceptibility maps using different probabilistic and bivariate statistical models and comparison of their performance at Wadi Itwad Basin,Asir Region,Saudi Arabia

The purpose of the current study is to produce landslide susceptibility maps using different probabilistic and bivariate statistical approaches; namely, frequency ratio (FR), weights-of-evidence (WofE), index-of-entropy (IofE), and Dempster–Shafer (DS) models, at Wadi Itwad, Asir region, in the southwestern part of Saudi Arabia. Landslide locations were identified and mapped from interpretation of high-resolution satellite images, historical records, and extensive field surveys. In total, 326 landslide locations were mapped using ArcGIS and divided into two groups; 75 % and 25 % of landslide locations were used for training and validation of models, respectively. Twelve layers of landslide-related factors were prepared, including altitude, slope degree, slope length, topography wetness index, curvature, slope aspect, distance from lineaments, distance from roads, distance from streams, lithology, rainfall, and normalized difference vegetation index. The relationships between the landslide-related factors and the landslide inventory map were calculated using different statistical models (FR, WofE, IofE, and DS). The model results were verified with landslide locations, which were not used during the model training. In addition, receiver operating characteristic curves were applied, and area under the curve (AUC) was calculated for the different susceptibility maps using the success (training data) and prediction (validation data) rate curves. The results showed that the AUC for success rates are 0.813, 0.815, 0.800, and 0.777, while the prediction rates are 0.95, 0.952, 0.946, and 0.934 for FR, WofE, IofE, and DS models, respectively. Subsequently, landslide susceptibility maps were divided into five susceptibility classes, including very low, low, moderate, high, and very high. Additionally, the percentage of training and validating landslides locations in high and very high landslide susceptibility classes in each map were calculated. The results revealed that the FR, WofE, IofE, and DS models produced reasonable accuracy. The outcomes will be useful for future general planned development activities and environmental protection.     

Predicting profitability of listed construction companies based on principal component analysis and support vector machine—Evidence from China

In order to monitor the operating conditions of the construction industry, this paper incorporates the principal component analysis (PCA) and support vector machine (SVM) to predict the profitability of the construction companies listed on A-share market in China. With annual financial data in 2001–2012, this paper selected six indicators from different profitable perspectives to build a composite profitability index based on the PCA technique, and then established a SVM model to make the corporate profitability prediction of the construction companies in China. The results indicate that, the technical combination of the PCA and SVM can improve the profitability prediction significantly. In 2003–2012, the accuracy of predicting the profitability of the Chinese construction companies exceeded 80% on average. Compared with the artificial neural network (ANN), the SVM model has the superiority in the accuracy prediction of the Chinese construction companies.     

Comparison of machine learning methods for ground settlement prediction with different tunneling datasets

This study integrates different machine learning (ML) methods and 5-fold cross-validation (CV) method to estimate the ground maximal surface settlement (MSS) induced by tunneling. We further investigate the applicability of artificial intelligent (AI) based prediction through a comparative study of two tunnelling datasets with different sizes and features. Four different ML approaches, including support vector machine (SVM), random forest (RF), back-propagation neural network (BPNN), and deep neural network (DNN), are utilized. Two techniques, i.e. particle swarm optimization (PSO) and grid search (GS) methods, are adopted for hyperparameter optimization. To assess the reliability and efficiency of the predictions, three performance evaluation indicators, including the mean absolute error (MAE), root mean square error (RMSE), and Pearson correlation coefficient (R), are calculated. Our results indicate that proposed models can accurately and efficiently predict the settlement, while the RF model outperforms the other three methods on both datasets. The difference in model performance on two datasets (Datasets A and B) reveals the importance of data quality and quantity. Sensitivity analysis indicates that Dataset A is more significantly affected by geological conditions, while geometric characteristics play a more dominant role on Dataset B.     

基于 BAS-SCA-BP 模型的高校建筑工程项目投资估算研究

针对标准BP 神经网络建筑工程项目投资估算模型收敛速度慢、预测精度低的问题,提出融合改进天牛须和正余弦双重优化算法（BAS-SCA）优化BP 神经网络的建筑工程项目投资估算模型。以某市高校建筑工程项目为研究对象,分析相关文献并结合显著性理论初步选择工程造价影响因子,利用粗糙集属性约简算法筛选出关键因素;基于此,通过构建基于BAS-SCA-BP 的神经网络估算模型实现快速、准确的建筑工程投资估算。研究结果表明:基于BAS-SCA-BP 的估算模型较标准BP 神经网络估算模型的估算精度有了大幅提高,与其他智能算法改进的BP 神经网络估算模型的性能相比较,该模型在稳定性和预测精度方面表现更佳。