Search

In recent years, as a result of climate change as well as rainfall reduction in arid and semi‐arid regions, modelling qualitative and quantitative parameters belonging to aquifers has become crucially important. In Iran, as aquifers are treated as the most commonly used drinking water resources, modelling their qualitative and quantitative parameters is enormously important. In this paper, for the first time, values of salinity, total dissolved solids (TDS), groundwater level (GWL) and electrical conductivity (EC) of the Arak Plain, located in Markazi Province, Iran, are simulated by means of four modern artificial intelligence models including extreme learning machine (ELM), wavelet extreme learning machine (WELM), online sequential extreme learning machine (OSELM) and wavelet online sequential extreme learning machine (WOSELM) as well as the MODFLOW software for a 15‐year period monthly. To develop the hybrid artificial intelligence models, the wavelet is employed. First, the effective lags in estimating the qualitative and quantitative parameters of the groundwater are identified using the autocorrelation function (ACF) and the partial autocorrelation function (PACF) analysis. After that, four different models are developed by the selected input combinations and also the ACF and the PACF in the form of different lags for each of ELM, WAELM, OSELM and WOSELM methods. Then, the superior models in simulating the groundwater qualitative and qualitative parameters are detected by conducting a sensitivity analysis. To forecast the electrical conductivity (EC) by the best WOSELM model, the values of the Nash–Sutcliffe efficiency coefficient (NSC), Mean Absolute Error (MAE) and the scatter index (SI) are obtained to be 0.991, 18.005 and 4.28E‐03, respectively. In addition, the most effective lags in estimating these parameters are introduced. Subsequently, the results found by the MODFLOW model are compared with those of the artificial intelligence models and it is concluded that the latter are more accurate. For instance, the scatter index and Nash–Sutcliffe efficiency coefficient values calculated by WOSELM for TDS, respectively, are 5.34E‐03 and 0.991. Finally, an uncertainty analysis is conducted to evaluate the performance of different numerical models. For example, MODFLOW has an underestimated performance in simulating the salinity parameter. [ABSTRACT FROM AUTHOR]

Abstract: Natural language processing is a composition of several error‐prone and challenging tasks, including part of speech tagging, word sense disambiguation, named entity recognition, and compound verb detection. Studying intrasentence relations and roles is essential to improve the mentioned subtasks. Semi‐automatic schemes such as ontologies can be applied to clarify word's dependencies. This paper presents an ontology that is targeting to improve POS tagging, WSD, NER, and compound verb detection in Persian with extra properties that may ameliorate machine translation. The ontology is tested in combinations with several state‐of‐art algorithms on Dadegan corpus. The results show that coping semantic analysis with machine learning methods enhance relation detection and consequently precision of the mentioned subtasks, which is not widely addressed in Persian. Furthermore, the experimental results declare that the accuracy rate increases between 4.5 and 23% for different tasks. [ABSTRACT FROM AUTHOR]

Floods are both complex and destructive, and in most parts of the world cause injury, death, loss of agricultural land, and social disruption. Flood susceptibility (FS) maps are used by land‐use managers and land owners to identify areas that are at risk from flooding and to plan accordingly. This study uses machine learning ensembles to produce objective and reliable FS maps for the Haraz watershed in northern Iran. Specifically, we test the ability of the support vector regression (SVR), together with linear kernel (LK), base classifier (BC), and hyper‐parameter optimization (HPO), to identify flood‐prone areas in this watershed. We prepared a map of 201 past floods to predict future floods. Of the 201 flood events, 151 (75%) were used for modeling and 50 (25%) were used for validation. Based on the relevant literature and our field survey of the study area, 10 effective factors were selected and prepared for flood zoning. The results show that three of the 10 factors are most important for predicting flood‐sensitive areas, specifically and in order of importance, slope, distance to the river and river. Additionally, the SVR‐HPO model, with area under the curve values of 0.986 and 0.951 for the training and testing phases, outperformed the other two tested models. [ABSTRACT FROM AUTHOR]

An unsolved problem in the digital mapping of categorical soil variables and soil types is the imbalanced number of observations, which leads to reduced accuracy and the loss of the minority class (the class with a significantly lower number of observations compared to other classes) in the final map. So far, synthetic over‐ and under‐sampling techniques have been explored in soil science; however, more efficient approaches that do not have the drawbacks of these techniques and guarantee retention of the minority classes in the produced map are essentially required. Such approaches suggested in the present study for digital mapping of soil classes include machine learning models of ensemble gradient boosting, cost‐sensitive learning and one‐class classification (OCC) of the minority class combined with multi‐class classification. In this regard, extreme gradient boosting (XGB) as an ensemble gradient learner, a cost‐sensitive decision tree (CSDT) within the C5.0 algorithm, and a one‐class support vector machine combined with multi‐class classification (OCCM) were investigated to map eight soil great groups with a naturally imbalanced frequency of observations in northwest Iran. A total of 453 profile data points were used for mapping the soil great groups of the study area. A data split was done manually for each class separately, which resulted in an overall 70% of the data for calibration and 30% for validation. The bootstrapping approach of calibration (with 10 runs) was performed to produce multiple maps for each model. The 10 bootstraps were evaluated against the hold‐out validation dataset. The average values of accuracy measures, including Kappa (K), overall accuracy (OA), producer's accuracy (PA) and user's accuracy (UA), were explored. In addition, the results of this study were compared with a previous study in the same area, in which resampling techniques were used to deal with imbalanced data for digital soil class mapping. The findings show that all three suggested methods can deal well with the imbalanced classification problem, with OCCM showing the highest K (= 0.76) and OA (= 82) in the validation stage. Also, this model can guarantee the retention of the minority classes in the final map. Comparing the present approaches with the previous study approach demonstrates that the three newly suggested methods can remarkably increase both overall and individual class accuracy for mapping. [ABSTRACT FROM AUTHOR]

Copyright of Irrigation & Drainage is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This study was conducted to assess the applicability of novel types of neural network methods (extreme learning machine (ELM), radial basis function (RBF), modified group method of data handling (M‐GMDH)) and M5 tree methods for the prediction of the soil water retention curve (SWRC) and compare their performance with that of derived methods and pedotransfer functions (PTFs) of other studies for soils in Iran. Then, 15 PTFs were developed. Predictions were evaluated by the integral root mean square error (IRMSE), integral mean error (IME), Akaike's information criterion (AIC), and coefficient of determination (R2). The RBF‐based PTFs were better than the M5 tree, M‐GMDH, and ELM‐based PTFs in terms of the IRMSE criterion in the testing step. Also, PTFs and methods developed in the present study were more reliable than other derived PTFs and methods by different researchers. The values of the IRMSE and R2 in the best PTFs (with inputs of sand, clay, total porosity and the moisture content at field capacity and permanent wilting point) of the testing data set of the RBF method were 0.037 cm3cm−3 and 0.988, respectively. For the testing data set, the average values of the IRMSE criterion for all the PTFs of the RBF, ELM, M‐GMDH, and M5 tree methods were 0.051, 0.062, 0.055, and 0.054 cm3cm−3, respectively. Therefore, the differences were considerable only between the ELM and other methods. The IRMSE criterion results of the testing data set showed the suitability of the RBF method in the development of PTFs for the prediction of the SWRC. Key Points: The radial basis function (RBF) network has strong tolerance to input noiseRBF, M5 tree, and modified group method of data handling methods were considerably better than the extreme learning machine methodThe RBF method is suitable in developing pedotransfer functions for the estimation of the soil water retention curve [ABSTRACT FROM AUTHOR]

Most common machine learning (ML) algorithms usually work well on balanced training sets, that is, datasets in which all classes are approximately represented equally. Otherwise, the accuracy estimates may be unreliable and classes with only a few values are often misclassified or neglected. This is known as a class imbalance problem in machine learning and datasets that do not meet this criterion are referred to as imbalanced data. Most datasets of soil classes are, therefore, imbalanced data. One of our main objectives is to compare eight resampling strategies that have been developed to counteract the imbalanced data problem. We compared the performance of five of the most common ML algorithms with the resampling approaches. The highest increase in prediction accuracy was achieved with SMOTE (the synthetic minority oversampling technique). In comparison to the baseline prediction on the original dataset, we achieved an increase of about 10, 20 and 10% in the overall accuracy, kappa index and F‐score, respectively. Regarding the ML approaches, random forest (RF) showed the best performance with an overall accuracy, kappa index and F‐score of 66, 60 and 57%, respectively. Moreover, the combination of RF and SMOTE improved the accuracy of the individual soil classes, compared to RF trained on the original dataset and allowed better prediction of soil classes with a low number of samples in the corresponding soil profile database, in our case for Chernozems. Our results show that balancing existing soil legacy data using synthetic sampling strategies can significantly improve the prediction accuracy in digital soil mapping (DSM). Highlights: Spatial distribution of soil classes in Iran can be predicted using machine learning (ML) algorithms.The synthetic minority oversampling technique overcomes the drawback of imbalanced and highly biased soil legacy data.When combining a random forest model with synthetic sampling strategies the prediction accuracy of the soil model improves significantly.The resulting new soil map of Iran has a much higher spatial resolution compared to existing maps and displays new soil classes that have not yet been mapped in Iran. [ABSTRACT FROM AUTHOR]