Search

The development and functional perfection of urban areas have led to increasingly severe fire risks in recent decades. Previous urban fire risk assessment methods relied on subjective judgment, rough data collection, simple linear statistical methods, etc. These drawbacks can lead to low robustness of evaluation and inadequate generalization ability. To resolve these problems, this paper selects the indicator and regression models based on the high-resolution data of the spatial distribution characteristics of Longquanyi distinct in Chengdu, China. and proposes an integrated machine learning algorithm for fire risk assessment. Firstly, the kernel density analysis is used to map the fourteen urban characteristics related to fire risks. The contributions of these indicators (characteristics) to fire risk and its corresponding index are determined by Random Forest (RF), Gradient Boosting Decision Tree (GBDT), and eXtreme Gradient Boosting (XGBoost). Then, the spatial correlation of fire risks is determined through Moran's I, and the spatial distribution pattern of indicator weights is clarified through the raster coefficient space analysis. Finally, with these selected indicators, we test the regression performance with a backpropagation neural network (BPNN) algorithm and a geographically weighted regression (GWR) model. The results indicate that numerical variables are more suitable than dummy variables for estimating micro-scale fire risks. The main factors with a high contribution are all numerical variables, including roads, gas pipelines, GDP, hazardous chemical enterprises, petrol and charging stations, cultural heritage protection units, assembly occupancies, and high-rise buildings. The machine learning algorithm integrating RF and BPNN shows the best performance (R2 = 0.97), followed by the RF-GWR integrated algorithm (R2 = 0.87). Compared with previous methods, this algorithm reduces the subjectivity of the traditional assessment models and shows the ability to automatically obtain the key indicators of urban fire risks. Hence, this new approach provides us with a more robust tool for assessing the future fire safety level in urban areas. [ABSTRACT FROM AUTHOR]

Landslide susceptibility assessment (LSA) based on machine learning methods has been widely used in landslide geological hazard management and research. However, the problem of sample imbalance in landslide susceptibility assessment, where landslide samples tend to be much smaller than non-landslide samples, is often overlooked. This problem is often one of the important factors affecting the performance of landslide susceptibility models. In this paper, we take the Wanzhou district of Chongqing city as an example, where the total number of data sets is more than 580,000 and the ratio of positive to negative samples is 1:19. We oversample or undersample the unbalanced landslide samples to make them balanced, and then compare the performance of machine learning models with different sampling strategies. Three classic machine learning algorithms, logistic regression, random forest and LightGBM, are used for LSA modeling. The results show that the model trained directly using the unbalanced sample dataset performs the worst, showing an extremely low recall rate, indicating that its predictive ability for landslide samples is extremely low and cannot be applied in practice. Compared with the original dataset, the sample set optimized through certain methods has demonstrated improved predictive performance across various classifiers, manifested in the improvement of AUC value and recall rate. The best model was the random forest model using over-sampling (O_RF) (AUC = 0.932). [ABSTRACT FROM AUTHOR]

Geostatistical estimation methods rely on experimental variograms that are mostly erratic, leading to subjective model fitting and assuming normal distribution during conditional simulations. In contrast, Machine Learning Algorithms (MLA) are (1) free of such limitations, (2) can incorporate information from multiple sources and therefore emerge with increasing interest in real-time resource estimation and automation. However, MLAs need to be explored for robust learning of phenomena, better accuracy, and computational efficiency. This paper compares MLAs, i.e., Multiple Linear Regression (MLR) and Random Forest (RF), with Ordinary Kriging (OK). The techniques were applied to the publicly available Walkerlake dataset, while the exhaustive Walker Lake dataset was validated. The results of MLR were significant (p < 10 × 10−5), with correlation coefficients of 0.81 (R-square = 0.65) compared to 0.79 (R-square = 0.62) from the RF and OK methods. Additionally, MLR was automated (free from an intermediary step of variogram modelling as in OK), produced unbiased estimates, identified key samples representing different zones, and had higher computational efficiency. [ABSTRACT FROM AUTHOR]

Landslide susceptibility assessment serves as a critical scientific reference for geohazard control, land use, and sustainable development planning. The existing research has not fully considered the potential impact of the spatial agglomeration and dispersion of landslides on assessments. This issue may cause a systematic evaluation bias when the field investigation data are insufficient, which is common due to limited human resources. Accordingly, this paper proposes two novel strategies, including a clustering algorithm and a preprocessing method, for these two ignored features to strengthen assessments, especially in high-susceptibility regions. Multiple machine learning models are compared in a case study of the city of Bijie (Guizhou Province, China). Then we generate the optimal susceptibility map and conduct two experiments to test the validity of the proposed methods. The primary conclusions of this study are as follows: (1) random forest (RF) was superior to other algorithms in the recognition of high-susceptibility areas and the portrayal of local spatial features; (2) the susceptibility map incorporating spatial feature messages showed a noticeable improvement over the spatial distribution and gradual change of susceptibility, as well as the accurate delineation of critical hazardous areas and the interpretation of historical hazards; and (3) the spatial distribution feature had a significant positive effect on modeling, as the accuracy increased by 5% and 10% after including the spatial agglomeration and dispersion consideration in the RF model, respectively. The benefit of the agglomeration is concentrated in high-susceptibility areas, and our work provides insight to improve the assessment accuracy in these areas, which is critical to risk assessment and prevention activities. [ABSTRACT FROM AUTHOR]

Airborne laser technology produces point clouds that can be used to build 3D models of buildings. However, the work is a laborious process that could benefit from automation. Artificial intelligence (AI) has been widely used in automating building segmentation as one of the initial stages in the 3D modeling process. The algorithms with a high success rate using point clouds for automatic semantic segmentation are random forest (RF) and PointNet++, with each algorithm having its own advantages and disadvantages. However, the training and testing data to develop and test the model usually share similar characteristics. Moreover, producing a good automation model requires a lot of training data, which may become an issue for users with a small amount of training data (limited data). The aim of this research is to test the performance of the RF and PointNet++ models in different regions with limited training and testing data. We found that the RF model developed from a small amount data, in different regions between the training and testing data, performs well compared to PointNet++, yielding an OA score of 73.01% for the RF model. Furthermore, several scenarios have been used in this research to explore the capabilities of RF in several cases. [ABSTRACT FROM AUTHOR]

The indoor location of individuals is a key contextual variable for commercial and assisted location-based services and applications. Commercial centers and medical buildings (e.g., hospitals) require location information of their users/patients to offer the services that are needed at the correct moment. Several approaches have been proposed to tackle this problem. In this paper, we present the development of an indoor location system which relies on the human activity recognition approach, using sound as an information source to infer the indoor location based on the contextual information of the activity that is realized at the moment. In this work, we analyze the sound information to estimate the location using the contextual information of the activity. A feature extraction approach to the sound signal is performed to feed a random forest algorithm in order to generate a model to estimate the location of the user. We evaluate the quality of the resulting model in terms of sensitivity and specificity for each location and we also perform out-of-bag error estimation. Our experiments were carried out in five representative residential homes. Each home had four individual indoor rooms. Eleven activities (brewing coffee, cooking, eggs, taking a shower, etc.) were performed to provide the contextual information. Experimental results show that developing an indoor location system (ILS) that uses contextual information from human activities (identified with data provided from the environmental sound) can achieve an estimation that is 95% correct. [ABSTRACT FROM AUTHOR]

Predicting query cost plays an important role in moving object databases. Accurate predictions help database administrators effectively schedule workloads and achieve optimal resource allocation strategies. There are some works focusing on query cost prediction, but most of them employ analytical methods to obtain an index-based cost prediction model. The accuracy can be seriously challenged as the workload of the database management system becomes more and more complex. Differing from the previous work, this paper proposes a method called CPRQ (Cost Prediction of Range Query) which is based on machine-learning techniques. The proposed method contains four learning models: the polynomial regression model, the decision tree regression model, the random forest regression model, and the KNN (k-Nearest Neighbor) regression model. Using R-squared and MSE (Mean Squared Error) as measurements, we perform an extensive experimental evaluation. The results demonstrate that CPRQ achieves high accuracy and the random forest regression model obtains the best predictive performance (R-squared is 0.9695 and MSE is 0.154). [ABSTRACT FROM AUTHOR]

Enriching and updating maps are among the most important tasks of any urban management organization for informed decision making. Urban cadastral map enrichment is a time-consuming and costly process, which needs an expert's opinion for quality control. This research proposes a smart framework to enrich a cadastral base map using a more up-to-date map automatically by machine learning algorithms. The proposed framework has three main steps, including parcel matching, parcel change detection and base map enrichment. The matching step is performed by checking the center point of each parcel in the other map parcels. Support vector machine and random forest classification algorithms are used to detect the changed parcels in the base map. The proposed models employ the genetic algorithm for feature selection and grey wolf optimization and Harris hawks optimization for hyperparameter optimization to improve accuracy and performance. By assessing the accuracies of the models, the random forest model with feature selection and grey wolf optimization, with an F1-score of 0.9018, was selected for the parcel change detection method. Finally, the detected changed parcels in the base map are deleted and relocated automatically with corresponding parcels in the more up-to-date map by the affine transformation. [ABSTRACT FROM AUTHOR]

Population growth in urban centres and the intensification of segregation phenomena associated with international mobility require improved urban planning and decision-making. More effective planning in turn requires better analysis and geospatial modelling of residential locations, along with a deeper understanding of the factors that drive the spatial distribution of various migrant groups. This study examines the factors that influence the distribution of migrants at the local level and evaluates their importance using machine learning, specifically the variable importance measures produced by the random forest algorithm. It is conducted on high spatial resolution ( 100 × 100 grid cells) register data in Amsterdam and Copenhagen, using demographic, housing and neighbourhood attributes for 2018. The results distinguish the ethnic and demographic composition of a location as an important factor in the residential distribution of migrants in both cities. We also examine whether certain migrant groups pay higher prices in the most attractive areas, using spatial statistics and mapping for 2008 and 2018. We find evidence of segregation in both cities, with Western migrants having higher purchasing power than non-Western migrants in both years. The method sheds light on the determinants of migrant distribution in destination cities and advances our understanding of the application of geospatial artificial intelligence to urban dynamics and population movements. [ABSTRACT FROM AUTHOR]

One of the core tasks in digital soil mapping (DSM) studies is the estimation of the spatial distribution of different soil variables. In addition, however, assessing the uncertainty of these estimations is equally important, something that a lot of current DSM studies lack. Machine learning (ML) methods are increasingly used in this scientific field, the majority of which do not have intrinsic uncertainty estimation capabilities. A solution to this is the use of specific ML methods that provide advanced prediction capabilities, along with innate uncertainty estimation metrics, like Quantile Regression Forests (QRF). In the current paper, the prediction and the uncertainty capabilities of QRF, Random Forests (RF) and geostatistical methods were assessed. It was confirmed that QRF exhibited outstanding results at predicting soil organic matter (OM) in the study area. In particular, R2 was much higher than the geostatistical methods, signifying that more variation is explained by the specific model. Moreover, its uncertainty capabilities as presented in the uncertainty maps, shows that it can also provide a good estimation of the uncertainty with distinct representation of the local variation in specific parts of the area, something that is considered a significant advantage, especially for decision support purposes. [ABSTRACT FROM AUTHOR]

Shallow groundwater is a key resource for human activities and ecosystems, and is susceptible to alterations caused by climate change, causing negative socio-economic and environmental impacts, and increasing the need to predict the evolution of the water table. The main objective of this study is to gain insights about future water level changes based on different climate change scenarios using machine learning algorithms, while addressing the following research questions: (a) how will the water table be affected by climate change in the future based on different socio-economic pathways (SSPs)?: (b) do machine learning models perform well enough in predicting changes of the groundwater in Denmark? If so, which ML model outperforms for forecasting these changes? Three ML algorithms were used in R: artificial neural networks (ANN), support vector machine (SVM) and random forest (RF). The ML models were trained with time-series data of groundwater levels taken at wells in the Hovedstaden region, for the period 1990–2018. Several independent variables were used to train the models, including different soil parameters, topographical features and climatic variables for the time period and region selected. Results show that the RF model outperformed the other two, resulting in a higher R-squared and lower mean absolute error (MAE). The future prediction maps for the different scenarios show little variation in the water table. Nevertheless, predictions show that it will rise slightly, mostly in the order of 0–0.25 m, especially during winter. The proposed approach in this study can be used to visualize areas where the water levels are expected to change, as well as to gain insights about how big the changes will be. The approaches and models developed with this paper could be replicated and applied to other study areas, allowing for the possibility to extend this model to a national level, improving the prevention and adaptation plans in Denmark and providing a more global overview of future water level predictions to more efficiently handle future climate change scenarios. [ABSTRACT FROM AUTHOR]

Based on a study of the spatial distribution of coffee shops in the main urban area of Beijing, the main influencing factors were selected based on the multi-source space data. Subsequently, three regression models were compared, and the best site selection model was found. A comparison was performed between the prediction model functioning with a buffer and without one, and the accuracy of the location model was verified by comparing the actual change trend and the predicted trend in two years. The following conclusions were obtained: (1) coffee shops in the main urban area of Beijing are clustered in an area within 12 km of the main urban center, and also around the core commercial agglomeration area; (2) the random forest (RF) model is the best model in this study, and the accuracy values before and after buffer analysis were 0.915 and 0.929, respectively; and (3) after verifying the accuracy of the model through two years of data, we recommend the establishment of a main road buffer zone for site selection, and the success rate of site selection was found to reach 72.97%. This study provides crucial insight for coffee shop prediction model selection and potential store location selection, which is significant to improving the layout of leisure spaces and promoting economic development. [ABSTRACT FROM AUTHOR]

Black carbon (BC) is a significant source of air pollution since it impacts public health and climate change. Understanding its distribution in the complex urban environment is challenging. We integrated a land use model with four machine learning models to estimate traffic-related BC concentrations in Oakland, CA. Random Forest was the best-performing model, with regression coefficient (R2) values of 0.701 on the train set and 0.695 on the validation set with a root mean square error (RMSE) of 0.210 mg/m3. Vehicle speed and local road systems were the most sensitive variables in estimating BC concentrations. However, this approach was inefficient at identifying hyperlocal hotspots, especially in a complex urban environment where highways and truck routes are significant emission sources. Using the land use method to estimate BC concentrations may lead to underestimating some localized hotspots. This work can improve air quality exposure assessment for vulnerable populations and help emphasize potential environmental justice issues. [ABSTRACT FROM AUTHOR]

Land use and land cover changes driven by urban sprawl has accelerated the degradation of ecosystem services in metropolitan settlements. However, most optimisation techniques do not consider the dynamic effect of urban sprawl on the spatial criteria on which decisions are based. In addition, integrating the current simulation approach with land use optimisation approaches to make a sustainable decision regarding the suitable site encompasses complex processes. Thus, this study aims to innovate a novel technique that can predict urban sprawl for a long time and can be simply integrated with optimisation land use techniques to make suitable decisions. Three main processes were applied in this study: (1) a supervised classification process using random forest (RF), (2) prediction of urban growth using a hybrid method combining an artificial neural network and cellular automata and (3) the development of a novel machine learning (ML) model to predict urban growth boundaries (UGBs). The ML model included linear regression, RF, K-nearest neighbour and AdaBoost. The performance of the novel ML model was effective, according to the validation metrics that were measured by the four ML algorithms. The results show that the Nasiriyah City expansion (the study area) is haphazard and unplanned, resulting in disastrous effects on urban and natural systems. The urban area ratio was increased by about 10%, i.e., from 2.5% in the year 1992 to 12.2% in 2022. In addition, the city will be expanded by 34%, 25% and 19% by the years 2032, 2042 and 2052, respectively. Therefore, this novel technique is recommended for integration with optimisation land use techniques to determine the sites that would be covered by the future city expansion. [ABSTRACT FROM AUTHOR]

Crimes are a common societal concern impacting quality of life and economic growth. Despite the global decrease in crime statistics, specific types of crime and feelings of insecurity, have often increased, leading safety and security agencies with the need to apply novel approaches and advanced systems to better predict and prevent occurrences. The use of geospatial technologies, combined with data mining and machine learning techniques allows for significant advances in the criminology of place. In this study, official police data from Porto, in Portugal, between 2016 and 2018, was georeferenced and treated using spatial analysis methods, which allowed the identification of spatial patterns and relevant hotspots. Then, machine learning processes were applied for space-time pattern mining. Using lasso regression analysis, significance for crime variables were found, with random forest and decision tree supporting the important variable selection. Lastly, tweets related to insecurity were collected and topic modeling and sentiment analysis was performed. Together, these methods assist interpretation of patterns, prediction and ultimately, performance of both police and planning professionals. [ABSTRACT FROM AUTHOR]

Mapping of landslides, conducted in 2021 by the Geological Survey of Denmark and Greenland (GEUS), revealed 3202 landslides in Denmark, indicating that they might pose a bigger problem than previously acknowledged. Moreover, the changing climate is assumed to have an impact on landslide occurrences in the future. The aim of this study is to conduct the first landslide susceptibility mapping (LSM) in Denmark, reducing the geographical bias existing in LSM studies, and to identify areas prone to landslides in the future following representative concentration pathway RCP8.5, based on a set of explanatory variables in an area of interest located around Vejle Fjord, Jutland, Denmark. A subset from the landslide inventory provided by GEUS is used as ground truth data. Three well-established machine learning (ML) algorithms—Random Forest, Support Vector Machine, and Logistic Regression—were trained to classify the data samples as landslide or non-landslide, treating the ML task as a binary classification and expressing the results in the form of a probability in order to produce susceptibility maps. The classification results were validated through the test data and through an external data set for an area located outside of the region of interest. While the high predictive performance varied slightly among the three models on the test data, the LR and SVM demonstrated inferior accuracy outside of the study area. The results show that the RF model has robustness and potential for applicability in landslide susceptibility mapping in low-lying landscapes of Denmark in the present. The conducted mapping can become a step forward towards planning for mitigative and protective measures in landslide-prone areas in Denmark, providing policy-makers with necessary decision support. However, the map of the future climate change scenario shows the reduction of the susceptible areas, raising the question of the choice of the climate models and variables in the analysis. [ABSTRACT FROM AUTHOR]

Poverty statistics are conventionally compiled using data from socioeconomic surveys. This study examines an alternative approach to estimating poverty by investigating whether readily available geospatial data can accurately predict the spatial distribution of poverty in Thailand. In particular, the geospatial data examined in this study include the intensity of night-time light (NTL), land cover, vegetation index, land surface temperature, built-up areas, and points of interest. The study also compares the predictive performance of various econometric and machine-learning methods such as generalized least squares, neural network, random forest, and support-vector regression. Results suggest that the intensity of NTL and other variables that approximate population density are highly associated with the proportion of an area's population that are living in poverty. The random forest technique yielded the highest level of prediction accuracy among the methods considered in this study, primarily due to its capability to fit complex association structures even with small-to-medium-sized datasets. This obtained result suggests the potential applications of using publicly accessible geospatial data and machine-learning methods for timely monitoring of the poverty distribution. Moving forward, additional studies are needed to improve the predictive power and investigate the temporal stability of the relationships observed. [ABSTRACT FROM AUTHOR]

Applications of machine-learning-based approaches in the geosciences have witnessed a substantial increase over the past few years. Here we present an approach that accounts for spatial autocorrelation by introducing spatial features to the models. In particular, we explore two types of spatial features, namely spatial lag and eigenvector spatial filtering (ESF). These features are used within the widely used random forest (RF) method, and their effect is illustrated on two public datasets of varying sizes (Meuse and California housing datasets). The least absolute shrinkage and selection operator (LASSO) is used to determine the best subset of spatial features, and nested cross-validation is used for hyper-parameter tuning and performance evaluation. We utilize Moran's I and local indicators of spatial association (LISA) to assess how spatial autocorrelation is captured at both global and local scales. Our results show that RF models combined with either spatial lag or ESF features yield lower errors (up to 33% different) and reduce the global spatial autocorrelation of the residuals (up to 95% decrease in Moran's I) compared to the RF model with no spatial features. The local autocorrelation patterns of the residuals are weakened as well. Compared to benchmark geographically weighted regression (GWR) models, the RF models with spatial features yielded more accurate models with similar levels of global and local autocorrelation in the prediction residuals. This study reveals the effectiveness of spatial features in capturing spatial autocorrelation and provides a generic machine-learning modelling workflow for spatial prediction. [ABSTRACT FROM AUTHOR]

Climatic changes significantly impact the socio-economic system. Compared with research on the impacts of climate change on the agricultural economic system, researches on the impacts on the industrial economic system are still scarce. This is mainly because of the difficulties in matching climate data with socio-economic data in terms of spatiotemporal resolution, which has greatly limited the exposure degree assessment and the risk assessment of industrial economic systems. In view of this, based on remote sensing inversion and multi-source data fusion, we generated kilometer-grid data of China's industrial output in 2010 and built the spatial distribution model of industrial output, based on random forest, to simulate the spatial distribution of China's industrial output under different climate change scenarios. The results showed that (1) our built spatial distribution simulation model of China's industrial output under different climate change scenarios had an accuracy of up to 93.77%; (2) from 2010 to 2050, the total growth of China's industrial output under scenario RCP8.5 is estimated to be 4.797% higher than that under scenario RCP4.5; and (3) the increasing rate of the average annual growth rate of China's industrial output slows down significantly under both scenarios from 2030 to 2050, and the average annual growth rate will decrease by 7.31 and 6.54%, respectively, under scenarios RCP8.5 and RCP4.5 compared with that from 2010 to 2020. [ABSTRACT FROM AUTHOR]

Academic performance (AP) is explained by a multitude of factors, principally by those related to socioeconomic, cultural, and educational environments. However, AP is less understood from a spatial perspective. The aim of this study was to investigate a methodology using a machine learning approach to determine which answers from a questionnaire-based survey were relevant for explaining the high AP of secondary school students across urban–rural gradients in Ecuador. We used high school locations to construct individual datasets and stratify them according to the AP scores. Using the Boruta algorithm and backward elimination, we identified the best predictors, classified them using random forest, and mapped the AP classification probabilities. We summarized these results as frequent answers observed for each natural region in Ecuador and used their probability outputs to formulate hypotheses with respect to the urban–rural gradient derived from annual maps of impervious surfaces. Our approach resulted in a cartographic analysis of AP probabilities with overall accuracies around 0.83–0.84% and Kappa values of 0.65–0.67%. High AP was primarily related to answers regarding the academic environment and cognitive skills. These identified answers varied depending on the region, which allowed for different interpretations of the driving factors of AP in Ecuador. A rural-to-urban transition ranging 8–17 years was found to be the timespan correlated with achievement of high AP. [ABSTRACT FROM AUTHOR]

We studied trip purpose imputation using data mining and machine learning techniques based on a dataset of GPS-based trajectories gathered in Switzerland. With a large number of labeled activities in eight categories, we explored location information using hierarchical clustering and achieved a classification accuracy of 86.7% using a random forest approach as a baseline. The contribution of this study is summarized below. Firstly, using information from GPS trajectories exclusively without personal information shows a negligible decrease in accuracy (0.9%), which indicates the good performance of our data mining steps and the wide applicability of our imputation scheme in case of limited information availability. Secondly, the dependence of model performance on the geographical location, the number of participants, and the duration of the survey is investigated to provide a reference when comparing classification accuracy. Furthermore, we show the ensemble filter to be an excellent tool in this research field not only because of the increased accuracy (93.6%), especially for minority classes, but also the reduced uncertainties in blindly trusting the labeling of activities by participants, which is vulnerable to class noise due to the large survey response burden. Finally, the trip purpose derivation accuracy across participants reaches 74.8%, which is significant and suggests the possibility of effectively applying a model trained on GPS trajectories of a small subset of citizens to a larger GPS trajectory sample. [ABSTRACT FROM AUTHOR]

A water regime type is a cumulative representation of seasonal runoff variability in a textual, qualitative, or quantitative form developed for a particular period. The assessment of the respective water regime type changes is of high importance for local communities and water management authorities, increasing their awareness and opening strategies for adaptation. In the presented study, we trained a machine learning model—the Random Forest classifier—to predict water regime types in northwest Russia based on monthly climatological hydrographs derived for a historical period (1979–1991). Evaluation results show the high efficiency of the trained model with an accuracy of 91.6%. Then, the Random Forest model was used to predict water regime types based on runoff projections for the end of the 21st century (2087–2099) forced by four different General Circulation Models (GCM) and three Representative Concentration Pathway scenarios (RCP). Results indicate that climate is expected to modify water regime types remarkably. There are two primary directions of projected changes. First, we detect the tendency towards less stable summer and winter flows. The second direction is towards a shift in spring flood characteristics. While spring flooding is expected to remain the dominant phase of the water regime, the flood peak is expected to shift towards earlier occurrence and lower magnitude. We identified that the projected changes in water regime types are more pronounced in more aggressive RCP scenarios. [ABSTRACT FROM AUTHOR]

Slope failure probabilistic models generated using random forest (RF) machine learning (ML), manually interpreted incident points, and light detection and ranging (LiDAR) digital terrain variables are assessed for predicting and generalizing to new geographic extents. Specifically, models for four Major Land Resource Areas (MLRAs) in the state of West Virginia in the United States (US) were created. All region-specific models were then used to predict withheld validation data within all four MLRAs. For all validation datasets, the model trained using data from the same MLRA provided the highest reported overall accuracy (OA), Kappa statistic, F1 Score, area under the receiver operating characteristic curve (AUC ROC), and area under the precision-recall curve (AUC PR). However, the model from the same MLRA as the validation dataset did not always provide the highest precision, recall, and/or specificity, suggesting that models extrapolated to new geographic extents tend to either overpredict or underpredict the land area of slope failure occurrence whereas they offer a better balance between omission and commission error within the region in which they were trained. This study highlights the value of developing region-specific inventories, models, and high resolution and detailed digital elevation data, since models may not generalize well to new geographic extents, potentially resulting from spatial heterogeneity in landscape and/or slope failure characteristics. [ABSTRACT FROM AUTHOR]

The early identification of potential landslide hazards is of great practical significance for disaster early warning and prevention. The study used different machine learning methods to identify potential active landslides along a 15 km buffer zone on both sides of Jinsha River (Panzhihua-Huize section), China. The morphology and texture features of landslides were characterized with InSAR deformation monitoring data and high-resolution optical remote sensing data, combined with 17 landslide influencing factors. In the study area, 83 deformation accumulation areas of potential landslide hazards and 54 deformation accumulation areas of non-potential landslide hazards were identified through spatial overlay analysis with 64 potential active landslides, which have been confirmed by field verification. The Naive Bayes (NB), Decision Tree (DT), Support Vector Machine (SVM) and Random Forest (RF) algorithms were trained and tested through attribute selection and parameter optimization. Among the 17 landslide influencing factors, Drainage Density, NDVI, Slope and Weathering Degree play an indispensable role in the machine learning and recognition of landslide hazards in our study area, while other influencing factors play a certain role in different algorithms. A multi-index (Precision, Recall, F1) comparison shows that the SVM (0.867, 0.829, 0.816) has better recognition precision skill for small-scale unbalanced landslide deformation datasets, followed by RF (0.765, 0.756, 0.741), DT (0.755, 0.756, 0.748) and NB (0.659, 0.659, 0.659). Different from the previous study on landslide susceptibility and hazard mapping based on machine learning, this study focuses on how to find out the potential active landslide points more accurately, rather than evaluating the landslide susceptibility of specific areas to tell us which areas are more sensitive to landslides. This study verified the feasibility of early identification of landslide hazards by using different machine learning methods combined with deformation information and multi-source landslide influencing factors rather than by relying on human–computer interaction. This study shows that the efficiency of potential hazard identification can be increased while reducing the subjective bias caused by relying only on human experts. [ABSTRACT FROM AUTHOR]

Most digital soil mapping (DSM) approaches aim at complete statistical model extraction. The value of the explicit rules of soil delineation formulated by soil-mapping experts is often underestimated. These rules can be used for expert testing of the notional consistency of soil maps, soil trend prediction, soil geography investigations, and other applications. We propose an approach that imitates traditional soil mapping by constructing compact globally optimal decision trees (EVTREE) for the covariates of traditionally used soil formation factor maps. We evaluated our approach by regional-scale soil mapping at a test site in the Belgorod region of Russia. The notional consistency and compactness of the decision trees created by EVTREE were found to be suitable for expert-based analysis and improvement. With a large sample set, the accuracy of the predictions was slightly lower for EVTREE (59%) than for CART (67%) and much lower than for Random Forest (87%). With smaller sample sets of 1785 and 1000 points, EVTREE produced comparable or more accurate predictions and much more accurate models of soil geography than CART or Random Forest. [ABSTRACT FROM AUTHOR]

Complementing information about particular points, places, or institutions, i.e., so-called Points of Interest (POIs) can be achieved by matching data from the growing number of geospatial databases; these include Foursquare, OpenStreetMap, Yelp, and Facebook Places. Doing this potentially allows for the acquisition of more accurate and more complete information about POIs than would be possible by merely extracting the information from each of the systems alone. Problem: The task of Points of Interest matching, and the development of an algorithm to perform this automatically, are quite challenging problems due to the prevalence of different data structures, data incompleteness, conflicting information, naming differences, data inaccuracy, and cultural and language differences; in short, the difficulties experienced in the process of obtaining (complementary) information about the POI from different sources are due, in part, to the lack of standardization among Points of Interest descriptions; a further difficulty stems from the vast and rapidly growing amount of data to be assessed on each occasion. Research design and contributions: To propose an efficient algorithm for automatic Points of Interest matching, we: (1) analyzed available data sources—their structures, models, attributes, number of objects, the quality of data (number of missing attributes), etc.—and defined a unified POI model; (2) prepared a fairly large experimental dataset consisting of 50,000 matching and 50,000 non-matching points, taken from different geographical, cultural, and language areas; (3) comprehensively reviewed metrics that can be used for assessing the similarity between Points of Interest; (4) proposed and verified different strategies for dealing with missing or incomplete attributes; (5) reviewed and analyzed six different classifiers for Points of Interest matching, conducting experiments and follow-up comparisons to determine the most effective combination of similarity metric, strategy for dealing with missing data, and POIs matching classifier; and (6) presented an algorithm for automatic Points of Interest matching, detailing its accuracy and carrying out a complexity analysis. Results and conclusions: The main results of the research are: (1) comprehensive experimental verification and numerical comparisons of the crucial Points of Interest matching components (similarity metrics, approaches for dealing with missing data, and classifiers), indicating that the best Points of Interest matching classifier is a combination of random forest algorithm coupled with marking of missing data and mixing different similarity metrics for different POI attributes; and (2) an efficient greedy algorithm for automatic POI matching. At a cost of just 3.5% in terms of accuracy, it allows for reducing POI matching time complexity by two orders of magnitude in comparison to the exact algorithm. [ABSTRACT FROM AUTHOR]

Up-to-date and reliable land-use information is essential for a variety of applications such as planning or monitoring of the urban environment. This research presents a workflow for mapping urban land use at the street block level, with a focus on residential use, using very-high resolution satellite imagery and derived land-cover maps as input. We develop a processing chain for the automated creation of street block polygons from OpenStreetMap and ancillary data. Spatial metrics and other street block features are computed, followed by feature selection that reduces the initial datasets by more than 80%, providing a parsimonious, discriminative, and redundancy-free set of features. A random forest (RF) classifier is used for the classification of street blocks, which results in accuracies of 84% and 79% for five and six land-use classes, respectively. We exploit the probabilistic output of RF to identify and relabel blocks that have a high degree of uncertainty. Finally, the thematic precision of the residential blocks is refined according to the proportion of the built-up area. The output data and processing chains are made freely available. The proposed framework is able to process large datasets, given that the cities in the case studies, Dakar and Ouagadougou, cover more than 1000 km2 in total, with a spatial resolution of 0.5 m. [ABSTRACT FROM AUTHOR]