CN113705607A - Landslide susceptibility evaluation method based on two-step strategy - Google Patents
Landslide susceptibility evaluation method based on two-step strategy Download PDFInfo
- Publication number
- CN113705607A CN113705607A CN202110832986.9A CN202110832986A CN113705607A CN 113705607 A CN113705607 A CN 113705607A CN 202110832986 A CN202110832986 A CN 202110832986A CN 113705607 A CN113705607 A CN 113705607A
- Authority
- CN
- China
- Prior art keywords
- sample
- landslide
- training
- data set
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Pit Excavations, Shoring, Fill Or Stabilisation Of Slopes (AREA)
Abstract
The invention discloses a landslide susceptibility evaluation method based on a two-step strategy, which comprises the following steps of: PU-learning based on a two-step strategy is a learning strategy. In the actual implementation process of the method, a base learner is relied on, and a Support Vector Machine (SVM) is selected as the base learner. In the sample construction, randomly extracting partial sample points P from a positive sample set P, adding the partial sample points P into an unlabeled sample set U to form a new unlabeled sample set U ', taking samples in U' as negative samples to form a training data set D with positive samples in the positive sample set P, and training a naive Bayes classifier; and (3) scoring the samples in the unmarked sample set U by using a trained naive Bayes classifier, setting a threshold value to be 0.5, selecting reliable negative sample, and constructing a reliable negative sample data set RN. In conclusion, the method utilizes the SVM model, utilizes the positive example set P and the reliable negative example sets RN and Q (namely U-RN) to carry out iterative training to obtain a training model, and finally predicts the unmarked sample set U to obtain a landslide susceptibility evaluation result.
Description
Technical Field
The invention belongs to the field of engineering geological research, and particularly relates to a landslide susceptibility evaluation method based on a two-step strategy.
Background
In recent years, with the application of computer technology in various fields and the perfection and maturity of data processing methods based on machine learning theory, many machine learning methods are applied to landslide susceptibility evaluation. Che et al performed landslide susceptibility degree analysis on volcanic terrain areas around the karlon forest mussels (Limbe) by using an information quantity method on the basis of determining data of several environmental control factors in 2012. Von policy and the like take a 4.20 Lushan seismic region as a research region, take gradient, undulation, land type, fault distance and seismic peak acceleration as evaluation indexes based on Remote Sensing (RS) and GIS technology, adopt a logistic regression method to construct an evaluation model to evaluate the landslide susceptibility index of the research region, and verify the effect of the model through a receiver operating characteristic curve (ROC). Liuyanfang and the like take a Ziguoguo county as a research area, the gradient, the elevation, the lithology, the land utilization, the distance from a water system and the like are selected as landslide evaluation factors, sensitivity indexes of various influence factors to landslide occurrence are analyzed by adopting a deterministic coefficient method, and a regional landslide susceptibility region drawing is drawn on the basis. Zhaoling and the like obtain the national landslide incidence index by utilizing the probability ratio method and draw a national incidence chart. In order to solve the problems that a traditional landslide prediction data source is limited, the data updating period is long, rules hidden in a complex landslide system are difficult to find easily and the like, a three gorges reservoir area is taken as a research object, data of landslide influence factors are extracted from multi-source space data, a digital terrain hydrologic analysis method is adopted to divide slope units in the area, the selected environmental influence factors are resampled, two types of support vector machine models are further constructed, the quantitative relation between the various environmental influence factors and landslide is analyzed, and a regional landslide proneness zoning map is generated. Yan occurrence and the like adopt an artificial neural network model (ANN) and a logistic regression model (LR) to evaluate the landslide susceptibility to Yuan-An county in Hubei province, obtain a regional landslide susceptibility region chart, and compare and analyze the prediction result. Research shows that 9 types of index factors related to landslide occurrence are extracted from the co-development historical landslide 177 in the region according to empirical knowledge, elevation factors are removed through correlation analysis, the rest 8 types of factors are finally selected for landslide incidence evaluation, and a landslide incidence region plot of the research region is obtained through analysis and calculation through ArcGIS and SPSS Modler software. And the Xiahui and the like take Wushan county in the three gorges reservoir area as a research area, 9 types of environmental influence factors (elevation, gradient, slope direction, terrain humidity index TWI, surface roughness index TRI, stratum lithology, water system distance, structural distance and vegetation coverage index NDVI) are extracted through analysis, the elevation factors are removed through inter-factor correlation analysis, and disaster points and environmental factors are analyzed through a Support Vector Machine (SVM) and an Artificial Neural Network (ANN) model to obtain a landslide susceptibility area drawing of the research area. Pal et al used an extreme learning machine to predict landslide susceptibility indices in the region near england ritted baud using two remote sensing datasets (one multispectral and one hyperspectral) and draw a landslide susceptibility risk plot.
Because the loess is soft, water and soil loss is easy to occur, and various geological disasters frequently occur on the loess plateau for a long time. Nowadays, with the increasing enhancement of human economy and engineering activities, the types and the number of geological disasters are increasing under the dual influence of human activities and natural transition, wherein the occurrence frequency and the occurrence scale of landslide disasters tend to increase. The special geomorphic environment of the loess plateau provides conditions and a foundation for the generation and inoculation of the loess landslide.
In recent years, geological disasters frequently occur on loess plateau, and the prevention and treatment work of the geological disasters is more and more emphasized by the society in the face of more and more severe disaster situations. In order to safely and effectively plan and construct towns and develop disaster prevention and reduction work of cities, an important reference factor is an area landslide disaster susceptibility area chart.
Disclosure of Invention
The invention aims to overcome the defects and provides a landslide susceptibility evaluation method based on a two-step strategy, which analyzes the contribution degree of each factor to landslide occurrence and inoculation by using a qualitative or quantitative method and finally performs landslide hazard susceptibility evaluation analysis on a research area by comparing and selecting an evaluation model or method suitable for disaster characteristics of the research area.
In order to achieve the above object, the present invention comprises the steps of:
s1, acquiring a plurality of environment factors from a data source according to loess causes, normalizing the grid graphs of all the environment factors into a grid unit set, extracting a plurality of sample data from the grid unit set as an unmarked data set U, and taking the data in the history landslide record as a normal example set P;
s2, constructing a reliable negative sample data set by using a spying method;
s3, training by adopting a base learner to obtain a training model;
and S4, predicting each sample in the unmarked data set U by adopting a training model, determining the category of the unmarked point according to a threshold value, and finally obtaining the landslide susceptibility evaluation result.
At S1, the environmental factors include elevation, grade, slope, curvature, normalized vegetation index NDVI, geological structure, loess erosion strength, and terrain moisture index TWI.
In S1, the grid map of environmental factors is normalized to a 30m × 30m grid cell set.
The specific method of S2 is as follows:
randomly extracting partial sample points P from the positive sample set P, adding the partial sample points P into the unlabeled data set U to form a new unlabeled sample set U ', taking samples in the new unlabeled sample set U' as negative sample, forming a training data set D by the negative sample and the positive sample in the positive sample set P, training, scoring the samples in the unlabeled sample set U by adopting a trained model, selecting the negative sample meeting the requirements, and constructing a reliable negative sample data set RN.
The training adopts a naive Bayes classifier, and the trained model is the trained naive Bayes classifier.
In S3, an SVM model is adopted to carry out iterative training on the set of the positive sample set P, the reliable negative sample data set RN and the unmarked data set U without the reliable negative sample data set RN, and a training model is obtained.
Compared with the prior art, the method comprises the steps of extracting unmarked data sets and positive example sets according to loess causes, constructing reliable negative example sample data sets by a spy method, training by a base learner to obtain a training model, predicting each sample in the unmarked data sets U by the training model, determining the category of unmarked points according to a threshold value, and finally obtaining the landslide susceptibility evaluation result.
Compared with the prior art, the invention has the following beneficial effects:
according to the loess cause, the contribution degree of each environmental factor to landslide occurrence and inoculation is analyzed, and 8 relevant environmental factors are given; a reliable negative sample data set is constructed by utilizing a spying method; and training by adopting a base learner to obtain a training model. It can be seen that the method does not need a large number of known landslide sample points to establish a positive sample data set, so that the robustness of the model is remarkably improved. According to the method, each sample in an unmarked data set U is predicted by adopting a training model, the category of an unmarked point is determined according to a threshold value, and a landslide susceptibility evaluation result is finally obtained. Therefore, the method can make the model more stable and remarkably improve the prediction precision by randomly selecting negative samples for many times and repeatedly training the base learner.
Drawings
FIG. 1 is a grid diagram of historical landslide points and elevations in an embodiment;
FIG. 2 is a grid graph of historical landslide points and slopes in an embodiment;
FIG. 3 is a grid diagram of historical landslide points and slope directions in an embodiment;
FIG. 4 is a graph of historical landslide points and a grid of curvatures in an embodiment;
FIG. 5 is a grid graph of historical landslide points and vegetation coverage indices in an example;
FIG. 6 is a grid diagram of historical landslide points and geological structures in an embodiment;
FIG. 7 is a grid graph of historical landslide points and erosion intensity of yellow map in an example;
FIG. 8 is a grid graph of historical landslide and terrain moisture indices in an embodiment;
FIG. 9 is a PU-two-step method-SVM landslide easy-occurrence zoning map of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
The invention comprises the following steps:
1. constructing an original data set;
according to the research on the cause of loess landslide in the research area and past experience and research, 8 environmental factors are obtained from a data source: taking elevation, gradient, slope direction, curvature, normalized vegetation index (NDVI), geological structure, loess erosion strength and terrain humidity index (TWI) as main environmental factors influencing loess landslide development of a research area, normalizing a grid graph of the 8 environmental factors into a grid unit set of 30m multiplied by 30m, and extracting about 180 ten thousand sample data of the whole area from the grid unit set as an unmarked data set U; 203 historical landslides in the historical landslide catalog are taken as a positive example set P.
2. Constructing a reliable negative sample data set by utilizing a spying method;
and constructing a training data set by utilizing Spy (Spy), namely randomly extracting partial sample points P from the positive example set P, adding the partial sample points P into the unlabeled sample set U to form a new unlabeled sample set U ', taking samples in U' as negative example samples to form a training data set D with positive example samples in the positive example set P, training a naive Bayes classifier, scoring the samples in the unlabeled sample set U by utilizing the trained naive Bayes classifier, setting a threshold value to be 0.5, selecting reliable negative example samples, and constructing a reliable negative example sample data set RN.
The principle of the naive Bayes classifier is to calculate the posterior probability of a sample according to the prior probability of the sample, namely the probability that the sample belongs to a certain class, and finally, select the class with the maximum posterior probability as the class to which the sample belongs. The specific description of the algorithm is as follows:
a) using one n-dimensional feature vector X ═ X per sample1,x2,L,xnRepresents it. The feature vector is used to describe n attributes A1,A2,Λ,AnN dimensions of the sample.
b) Assuming there are n classes, each with C1,C2,Λ,CnAnd (4) showing. Given an unknown sample X, the naive Bayes classifier assigns the unknown sample X to CiClass, then the following must be satisfied:
P(Ci|Xi)>P(Cj|Xj)(1≤j≤n,j≠i)； (1)
wherein P (C)i| X) is the maximum a posteriori probability.
c) At CiIn the case of unknown prior probability of class, P (C)1)＝P(C2)＝Λ＝P(Cn) Otherwise P (C)i)＝siAnd s. Wherein s isiIs CiThe number of training samples in a class, s is the total number of training samples. From Bayesian equation, to calculate P (C)i| X), P (X | C) needs to be calculatedi)P(Ci) And P (X | C)i) The values of (c) can be obtained from a training data set.
3. Training a base learner;
and (3) carrying out iterative training by utilizing an SVM model and utilizing a positive case set P and a reliable negative case set RN and Q (namely U-RN) to obtain a training model.
The invention selects a Support Vector Machine (SVM) as a base learner. The support vector machine SVM is based on the basic principle that an optimal classification hyperplane is found in a sample space, and the distance between a point which is closest to the classification hyperplane and the hyperplane is the largest in two types of sample points. The larger the distance, the better the generalization of the classifier and the lower the error. Given a dataset D { (x)1,y1),(x2,y2),K,(xN,yN) Where i ═ 1, 2, …, N, xiIs the i-th feature vector, yiIs a class label. Given a hyperplane: wxiWhen + b is 0, the geometric spacing of the hyperplane with respect to the sample point is:
the minimum of the geometrical spacing of the hyperplane with respect to all sample points is:
i.e. support vector to hyperplaneThe distance of (c). Order to
Then the solving of the maximum segmentation hyperplane problem of the SVM model is represented as:
this is a convex quadratic programming problem with inequality constraints, introducing lagrange functions:
wherein alpha isiIs a Lagrangian multiplier, and alphaiIs more than or equal to 0. The above formula is subjected to partial derivation to obtain:
substituting into Lagrange function formula to obtain dual problem of original planning problem:
obtaining a model after solving:
in the case of linear divisibility, the support vector machine model is a linear classifier. In the case of linear inseparability, the support vector machine only considers points near the boundary line, i.e. points far from the boundary, during the calculation, and has no effect on the model establishment. The method greatly reduces the calculation amount, and introduces Lagrangian dual when the objective function is determined, so that the mapping calculation of the complex kernel function becomes possible.
4. Predicting sample points in the unmarked data set and obtaining a landslide susceptibility evaluation result;
and predicting each sample in the unmarked data set U by using a training model, and determining the category of the unmarked point according to a threshold value so as to obtain a landslide susceptibility evaluation result.
The method is characterized in that on the basis of the development and distribution characteristics of historical landslide points in a collection area, the geological environment of a research area is combined, the most important disaster-causing factors are screened from numerous environmental factors capable of inducing landslide disasters, the contribution degree of each factor to landslide occurrence and inoculation is analyzed by a qualitative or quantitative method, and finally an evaluation model or method suitable for the disaster characteristics of the research area is selected through comparison, so that landslide disaster susceptibility evaluation and analysis are performed on the research area.
Example (b):
the method comprises the steps of taking a Lingtai county of the plain city of Gansu province as a research area, collecting basic geological data and historical landslide record data of the research area, selecting appropriate environmental factors on the basis of a landslide occurrence mechanism and previous research experiences of a loess plateau where the research area is located, dividing the eight environmental factors of the research area into equal-size grid units by utilizing ArcGIS (geographic information System) drawing software, extracting data of the environmental factors from the grid units, and evaluating the landslide disaster susceptibility of the research area.
Collecting Digital Elevation Model (DEM) data with the resolution of 30m in a research area for extracting topographic and geomorphic information; 1500000 geological map for extracting geological structure information; the 8-month normalized vegetation index map in 2017 is used for extracting vegetation coverage rate information; 2017, extracting loess erosion intensity; landslide record and field data survey are used for obtaining historical landslide data of a research area. According to an expert evaluation method, 8 environmental factors are selected: elevation, slope, curvature, normalized vegetation index (NDVI), geological structure, loess erosion intensity, terrain moisture index (TWI) as influencing factors, as shown in fig. 1-8.
And according to the characteristics of the research area, a regular 30m multiplied by 30m grid unit is selected to divide the research area into about 180 ten thousand grid units, data are extracted from the grid map of the 8 environmental factors collected in the last step, and further normalization is carried out. Because of the particularities of the geological industry, field investigators are generally only interested in historical landslides, and therefore only historical landslide entries and about 180 million unknown units of the study area are collected in the data. Therefore, the problem of how to train the model by using a large number of unlabeled samples and a small number of known landslide samples and then find out the sample with high probability of landslide from the large number of unlabeled samples is not a simple classification problem.
The landslide susceptibility degree of the research area is evaluated by using a model (PU-two-step method-SVM) based on a two-step method strategy, and the result is shown in FIG. 9, so that most of historical landslide points in a Lingtai county landslide susceptibility zoning map obtained by using the PU-two-step method-SVM model fall into areas with high landslide susceptibility and high landslide susceptibility predicted by the model and are consistent with actual historical landslide data.
Claims (6)
1. A landslide susceptibility evaluation method based on a two-step strategy is characterized by comprising the following steps:
s1, acquiring a plurality of environment factors from a data source according to loess causes, normalizing the grid graphs of all the environment factors into a grid unit set, extracting a plurality of sample data from the grid unit set as an unmarked data set U, and taking the data in the history landslide record as a normal example set P;
s2, constructing a reliable negative sample data set by using a spying method;
s3, training by adopting a base learner to obtain a training model;
and S4, predicting each sample in the unmarked data set U by adopting a training model, determining the category of the unmarked point according to a threshold value, and finally obtaining the landslide susceptibility evaluation result.
2. The method of claim 1, wherein in step S1, the environmental factors include elevation, gradient, slope, curvature, normalized vegetation index NDVI, geological structure, loess erosion strength, and terrain moisture index TWI.
3. The method for assessing landslide susceptibility to occurrence according to claim 1 wherein the grid map of environmental factors is normalized to a 30m x 30m grid cell set in S1.
4. The landslide susceptibility assessment method according to claim 1, wherein the specific method of S2 is as follows:
randomly extracting partial sample points P from the positive sample set P, adding the partial sample points P into the unlabeled data set U to form a new unlabeled sample set U ', taking samples in the new unlabeled sample set U' as negative sample, forming a training data set D by the negative sample and the positive sample in the positive sample set P, training, scoring the samples in the unlabeled sample set U by adopting a trained model, selecting the negative sample meeting the requirements, and constructing a reliable negative sample data set RN.
5. The landslide susceptibility assessment method according to claim 4 wherein training employs a naive Bayes classifier and the trained model is a trained naive Bayes classifier.
6. The method for evaluating the landslide susceptibility based on the two-step strategy according to claim 1, wherein in S3, an SVM model is adopted to perform iterative training on a positive sample set P, a reliable negative sample data set RN and a set of unmarked data sets U except the reliable negative sample data set RN to obtain a training model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110832986.9A CN113705607B (en) | 2021-07-22 | 2021-07-22 | Landslide susceptibility evaluation method based on two-step strategy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110832986.9A CN113705607B (en) | 2021-07-22 | 2021-07-22 | Landslide susceptibility evaluation method based on two-step strategy |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113705607A true CN113705607A (en) | 2021-11-26 |
| CN113705607B CN113705607B (en) | 2023-08-22 |
Family
ID=78650316
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110832986.9A Active CN113705607B (en) | 2021-07-22 | 2021-07-22 | Landslide susceptibility evaluation method based on two-step strategy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113705607B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109359738A (en) * | 2018-10-19 | 2019-02-19 | 西南交通大学 | A kind of Landslide hazard appraisal procedure based on QPSO-BP neural network |
| CN111582386A (en) * | 2020-05-11 | 2020-08-25 | 四川师范大学 | Random forest based geological disaster multi-disaster comprehensive risk evaluation method |
| CN111898315A (en) * | 2020-07-27 | 2020-11-06 | 中国科学院空天信息创新研究院 | Landslide susceptibility assessment method based on fractal-machine learning hybrid model |
| WO2020249125A1 (en) * | 2019-06-14 | 2020-12-17 | 第四范式（北京）技术有限公司 | Method and system for automatically training machine learning model |
| CN112966722A (en) * | 2021-02-07 | 2021-06-15 | 南昌大学 | Regional landslide susceptibility prediction method based on semi-supervised random forest model |
-
2021
- 2021-07-22 CN CN202110832986.9A patent/CN113705607B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109359738A (en) * | 2018-10-19 | 2019-02-19 | 西南交通大学 | A kind of Landslide hazard appraisal procedure based on QPSO-BP neural network |
| WO2020249125A1 (en) * | 2019-06-14 | 2020-12-17 | 第四范式（北京）技术有限公司 | Method and system for automatically training machine learning model |
| CN111582386A (en) * | 2020-05-11 | 2020-08-25 | 四川师范大学 | Random forest based geological disaster multi-disaster comprehensive risk evaluation method |
| CN111898315A (en) * | 2020-07-27 | 2020-11-06 | 中国科学院空天信息创新研究院 | Landslide susceptibility assessment method based on fractal-machine learning hybrid model |
| CN112966722A (en) * | 2021-02-07 | 2021-06-15 | 南昌大学 | Regional landslide susceptibility prediction method based on semi-supervised random forest model |
Non-Patent Citations (1)
| Title |
|---|
| 邱维蓉;吴帮玉;潘学树;唐亚明;: "几种聚类优化的机器学习方法在灵台县滑坡易发性评价中的应用", 西北地质, no. 01 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113705607B (en) | 2023-08-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sun et al. | A random forest model of landslide susceptibility mapping based on hyperparameter optimization using Bayes algorithm | |
| Wang et al. | A hybrid model considering spatial heterogeneity for landslide susceptibility mapping in Zhejiang Province, China | |
| Can et al. | Landslide susceptibility mapping at Ovacık-Karabük (Turkey) using different artificial neural network models: comparison of training algorithms | |
| Angra et al. | Machine learning and its applications: A review | |
| Chen et al. | Performance evaluation of GIS-based new ensemble data mining techniques of adaptive neuro-fuzzy inference system (ANFIS) with genetic algorithm (GA), differential evolution (DE), and particle swarm optimization (PSO) for landslide spatial modelling | |
| Tsangaratos et al. | Landslide susceptibility mapping using a modified decision tree classifier in the Xanthi Perfection, Greece | |
| Soma et al. | Optimization of causative factors using logistic regression and artificial neural network models for landslide susceptibility assessment in Ujung Loe Watershed, South Sulawesi Indonesia | |
| Pradhan et al. | Landslide susceptibility assessment and factor effect analysis: backpropagation artificial neural networks and their comparison with frequency ratio and bivariate logistic regression modelling | |
| Su et al. | Mapping of rainfall-induced landslide susceptibility in Wencheng, China, using support vector machine | |
| Eker et al. | Evaluation and comparison of landslide susceptibility mapping methods: a case study for the Ulus district, Bartın, northern Turkey | |
| Wan | Entropy-based particle swarm optimization with clustering analysis on landslide susceptibility mapping | |
| CN111898315B (en) | Landslide susceptibility assessment method based on fractal-machine learning hybrid model | |
| Zhao et al. | Using the rotation and random forest models of ensemble learning to predict landslide susceptibility | |
| Pavel et al. | An analysis of landslide susceptibility zonation using a subjective geomorphic mapping and existing landslides | |
| Niu et al. | Susceptibility assessment of landslides triggered by the Lushan earthquake, April 20, 2013, China | |
| CN113487105B (en) | Landslide geological disaster space prediction method and device and computer equipment | |
| CN114298436A (en) | Landslide susceptibility evaluation method based on Bagging strategy | |
| CN114330812A (en) | Landslide disaster risk assessment method based on machine learning | |
| Mao et al. | Landslide hazards mapping using uncertain Naïve Bayesian classification method | |
| Fekete et al. | Tree segmentation and change detection of large urban areas based on airborne LiDAR | |
| Mao et al. | Landslide susceptibility modelling based on AHC-OLID clustering algorithm | |
| Wen et al. | Rockfall susceptibility mapping using XGBoost model by hybrid optimized factor screening and hyperparameter | |
| Zhao et al. | An improved spatial case-based reasoning considering multiple spatial drivers of geographic events and its application in landslide susceptibility mapping | |
| Conforti et al. | Exploring performance and robustness of shallow landslide susceptibility modeling at regional scale using different training and testing sets | |
| Chen et al. | A one-class-classifier-based negative data generation method for rapid earthquake-induced landslide susceptibility mapping |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |