Search

Human learning and intelligence work differently from the supervised pattern recognition approach adopted in most deep learning architectures. Humans seem to learn rich representations by exploration and imitation, build causal models of the world, and use both to flexibly solve new tasks. We suggest a simple but effective unsupervised model which develops such characteristics. The agent learns to represent the dynamical physical properties of its environment by intrinsically motivated exploration and performs inference on this representation to reach goals. For this, a set of self-organizing maps which represent state-action pairs is combined with a causal model for sequence prediction. The proposed system is evaluated in the cartpole environment. After an initial phase of playful exploration, the agent can execute kinematic simulations of the environment's future and use those for action planning. We demonstrate its performance on a set of several related, but different one-shot imitation tasks, which the agent flexibly solves in an active inference style. [ABSTRACT FROM AUTHOR]

Circular economy promotes the use of waste materials into new production processes as a key factor for resource efficiency. The construction sector, and specifically the fired clay industry, is able to assimilate large amounts of waste in their processes, without significantly altering the technical properties of products. The introduction of different waste in ceramic products at the laboratory level has been extensively studied in the literature, but most of these studies have not yet been scaled-up to industrial production. Differences in processing with respect to laboratory conditions introduces uncertainty in relation to the expected properties of the final products. This paper uses a Self-Organizing Map (SOM)-based methodology for analysing and assessing the incorporation of industrial waste, Waelz slag (WS) and foundry sand dust (FSD), in ceramic products obtained sequentially at laboratory, semi-industrial and industrial level, over technological properties and metals release. As a result, from the SOM analysis, a clustered map of the samples developed is obtained that highlights the most important parameters affecting the technological and environmental properties to be the type of clay and therefore, the firing temperature; secondly, the Waelz slag content, being independent of the foundry sand addition; and finally, the type and level of processing (laboratory-pressing, semi-industrial extruder, industrial extruder). [ABSTRACT FROM AUTHOR]

In 2019, a nationwide airborne geophysical survey of Sierra Leone was flown at 150 m nominal line spacing and 50 m nominal terrain clearance. Contractual deliverables included magnetic, radiometric and supporting data streams. The primary aim of the geophysical survey was to provide a national geoscientific benchmark for resource management by the National Minerals Agency and to encourage foreign investment in the rich mining potential of Sierra Leone. Geophysical interpretation is often highly subjective. This can be helpful when an interpreter is skilled in the project area and exploration targets being sought but can be constraining and even dangerous when exploring new areas, and for new commodities. Objective classification tools can challenge interpretive bias and enhance geological understanding, both locally and regionally. We present self‐organizing maps and neural network products generated from the Sierra Leone geophysical dataset which both confirm and challenge the validity of conventional, subjective interpretation workflows. We demonstrate the usefulness of self‐organizing maps as a tool for inferring complementary geological information and show some initial results of convolution neural network classification of a reduced‐to‐equator grid to locate magnetic lineaments in plan and in depth. Moreover, we show initial results of self‐organizing map clustering as a tool for joint analysis of multiphysics datasets and construction of a pseudolithological cluster map. [ABSTRACT FROM AUTHOR]

Water quality control and monitoring is an important concern of countries over the world. We present in this work, the use the self-organizing feature maps of Kohonen (SOFM) as features selection technique and advanced classification techniques, such as: Long Short-Term Memory (LSTM) and Support Vector Machines (SVM). This study involved the advanced assessment of surface water quality from Tilesdit dam in Algeria. Typically, water quality status is determined by comparing collected data with water quality standards. LSTM and SVM have been applied with SOFM-based features selection for water quality classification. In this work, the training step is realized using the mentioned approaches to supervise the water quality from several physicochemical parameters. Eleven of them were collected in 4 seasons during the period (2016-2018) from study area. Experiments step using a mentioned dataset in terms of accuracy (training and test), running time and robustness, is carried out. The performance of our approach is optimized by regulating the parameter values using a SFOM based features selection method. The proposed approach outperforms current conventional methods, as this approach is a combination of strong feature selection and classification techniques. Optimal input features are selected directly from the original datasets, aiming to reduce the computational time and complexity. The impact of this result is significant both technically (lower learning time) and economically (reduced the number of sensors) and can improve obviously the performance of our monitoring system. The accuracy is more than 98% in training and testing steps with features selection process for the LSTM and SVM models. The best results of sensitivity, specificity, precision, and F-score of the two proposed models were ranged all between 96,99 % and 100%. In a nutshell, the two comparative machine learning methods provide very high classification accuracy and make a considerable solution for water quality control and monitoring. [ABSTRACT FROM AUTHOR]

The vibrant Mediterranean city of Barcelona, in Spain, welcomed the 13th edition of WSOM + 2019, the Workshop on Self-Organizing Maps and Learning Vector Quantization, Clustering and Data Visualization. A self-organizing world: special issue of the 13th edition of the workshop on self-organizing maps and learning vector quantization, clustering and data visualization, WSOM + 2019 Passive concept drift handling via variations of learning vector quantization. [Extracted from the article]

The purpose of this study is to assess the groundwater evolution in Jerba Island. This paper also focuses on the relationship between mineralization and anthropogenic activities and the identification of processes controlling groundwater geochemistry. An approach combining the use of water dissolved chemical species, isotopic fingerprint, and the self-organizing maps has been used to understand the behavior of a unconfined aquifer system and to determine the origin of its different contamination sources. In the present work, a geochemical characterization of the coastal aquifer from the unconfined Jerba aquifer, Southeast Tunisia, was carried out, using hydrogeochemical and isotopic (2H and 18O) data, correlation matrix, and self-organizing maps. For this aim, 79 wells were sampled and analyzed. Geochemical modeling and isotopic data demonstrated that seawater intrusion, exchange reactions occurring within the aquifer matrix, and anthropogenic activities were the major processes controlling the groundwater contamination in the island. The isotopic tracers indicate that the increasing salinity in the shallow coastal aquifer is caused by seawater intrusion. The methodology appears as a valid tool applicable in other coastal areas to determine the main factors governing the groundwater mineralization. [ABSTRACT FROM AUTHOR]

Human decision-making involves cognitive processes of selection, evaluation, and interpretation among candidate solutions in order to solve decision problems. Nonintelligent decision support systems (DSS) lack automatic interpretations, at least in a low level scale, which can lead to undesired solutions. To tackle this limitation, hence producing enhanced decision making, a hybrid intelligent decision support approach is presented, which combines case-based reasoning cycle, semiotic concepts, and self-organizing maps. In addition, a novel sign deconstruction mechanism is introduced as foundation of the new approach and affords better interpretability and contextualization of candidate solutions without compromising efficiency and precision. The obtained results confirm that our proposed approach has the potential to be readily applicable to decision problems, particularly the ones that are of subjective nature. Moreover, the put forward approach may integrate some unlikely elements of linguistics and cognitive science which could fundamentally help the enhancement of DSS. [ABSTRACT FROM AUTHOR]

Copyright of Shipin Kexue/ Food Science is the property of Food Science Editorial Department 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.)

The vitality of small and medium enterprises (SMEs) is integral to the economic fortification of nations, necessitating refined enhancement mechanisms from governmental bodies. Distinctive in their developmental trajectory, SMEs present a unique challenge in prioritizing interventions. A Decision Support System (DSS), employing a hybridized methodology of Clustering and Analytic Hierarchy Process-Technique for Order Preference by Similarity to Ideal Solution (AHP-TOPSIS), is proposed to facilitate the stratification of SMEs and guide governmental action based on a hierarchical scale of priorities. In this study, K-Means clustering was adopted for the segmentation of SMEs, leveraging its capability to efficiently partition high-dimensional data with minimal error. Subsequently, the TOPSIS method was utilized to rank SMEs within each cluster. However, the critical step of computing criteria weights to ascertain their relative importance was achieved through the AHP method. The latter effectively addresses multivariate considerations encompassing both quantitative and qualitative criteria through pairwise comparison matrices. The research incorporated 11 attributes, encompassing essential characteristics intrinsic to business entities, such as business name, location, operational status, sector, tax identification number, workforce size, average revenue, production costs, operational challenges, credit accessibility, and external financing needs. The clustering process, executed via Self-Organizing Maps (SOM), yielded optimal clusters at 1000 epochs, evidenced by a Davies Bouldin Index (DBI) of 0.74785, translating to an accuracy of 91.2601%. Notably, the SOM's performance in clustering SME data surpassed that of KMeans, as demonstrated by superior results in both Sum of Squared Error (SSE) and DBI metrics, thus showcasing its proficiency in managing data with heterogeneous criteria. The methodology engenders an n-cluster output, from which members are earmarked for priority ranking. AHP-derived weights are calculated, with a Consistency Ratio (CR) exceeding 0 denoting consistency, thereby determining the significance of each SME. TOPSIS calculations then ascribe the final score, delineating the SMEs' standings. This integrated DSS framework presents a robust tool for policymakers, ensuring targeted and efficient allocation of resources towards the advancement of SMEs. [ABSTRACT FROM AUTHOR]

Deep convection initiated under different large-scale environments exhibits different precipitation features and interacts with local meteorology and surface properties in distinct ways. Here, we analyze the characteristics and spatiotemporal patterns of different types of convective systems over southeastern Texas using 13 years of high-resolution observations and reanalysis data. We find that mesoscale convective systems (MCSs) contribute significantly to both mean and extreme precipitation in all seasons, while isolated deep convection (IDC) plays a role in intense precipitation during summer and fall. Using self-organizing maps (SOMs), we found that convection can occur under unfavorable conditions without large-scale lifting or moisture convergence. In spring, fall and winter, frontal-related large-scale meteorological patterns (LSMPs) characterized by baroclinic waves and low-level moisture convergence act as primary triggers for convection, while the remaining storms are associated with an anticyclone pattern and orographic lifting. In summer, IDC are mainly associated with front-related and anticyclones LSMPs, while MCSs occur more in frontal-related LSMPs. We further tracked the lifecycle of MCSs and IDC using the Flexible Object Tracker algorithm over southeastern Texas. MCSs frequently initiate west of Houston, travelling eastward for around 8 hours to southeastern Texas, while IDC initiate locally. The average duration of MCSs in southeastern Texas is 6.1 hours, approximately 4.1 times the duration of IDC. Diurnally, the initiation of convection associated with favorable LSMPs peak at 1100 UTC, 3 hours earlier than those associated with anticyclones. [ABSTRACT FROM AUTHOR]

The selection of better-evaluated genotypes for a target region depends on the characterization of the climate conditions of the environment. With the advancement of computer technology and daily available information about the weather, integrating such information in selection and interaction genotype × environment studies has become a challenge. This article presents the use of the technique of artificial neural networks associated with reaction norms for the processing of climate and georeferenced data for the study of genetic behaviors and the genotype × environment interaction of soybean genotypes. The technique of self-organizing maps (SOM) consists of competitive learning between two layers of neurons; one is the input, which transfers the data to the map, and the other is the output, where the topological structure formed by the competition generates weights, which represent the dissimilarity between the neural units. The methodologies used to classify these neurons and form the target populations of environments (TPE) were the discriminant analysis (DA) and the principal component analysis (PCA). To study soybean genetic behavior within these TPE, the random regression model was adopted to estimate the components of variance, and the reaction norms were adjusted through the Legendre polynomials. The SOM methodology allowed for an explanation of 99% of the variance of the climate data and the formation of well-structured TPE, with the membership probability of the regions within the TPE above 80%. The formation of these TPE allowed us to identify and quantify the response of the genotypes to sensitive changes in the environment. [ABSTRACT FROM AUTHOR]

Copyright of Hacettepe University Journal of Economics & Administrative Sciences / Hacettepe Üniversitesi Iktisadi ve Idari Bilimler Fakültesi Dergisi is the property of Hacettepe University, Faculty of Economic & Administrative Sciences 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.)

Data mining techniques provide a popular and powerful tool set to generate various data-driven classification systems. In this paper, we investigate the combined use of self-organizing maps (SOM) and nonsmooth nonconvex optimization techniques in order to produce a working case of a data-driven risk classification system. The optimization approach strengthens the validity of SOM results, and the improved classification system increases both the quality of prediction and the homogeneity within the risk groups. Accurate classification of prostate cancer patients into risk groups is important to assist in the identification of appropriate treatment paths. We start with the existing rules and aim to improve classification accuracy by identifying inconsistencies utilizing self-organizing maps as a data visualization tool. Then, we progress to the study of assigning prostate cancer patients into homogenous groups with the aim to support future clinical treatment decisions. Using the case of prostate cancer patients grouping, we demonstrate strong potential of data-driven risk classification schemes for addressing the risk grouping issues in more general organizational settings. [ABSTRACT FROM AUTHOR]

Advances in measurement technology are producing increasingly time-resolved environmental exposure data. We aim to gain new insights into exposures and their potential health impacts by moving beyond simple summary statistics (e.g., means, maxima) to characterize more detailed features of high-frequency time series data. This study proposes a novel variant of the Self-Organizing Map (SOM) algorithm called Dynamic Time Warping Self-Organizing Map (DTW-SOM) for unsupervised pattern discovery in time series. This algorithm uses DTW, a similarity measure that optimally aligns interior patterns of sequential data, both as the similarity measure and training guide of the neural network. We applied DTW-SOM to a panel study monitoring indoor and outdoor residential temperature and particulate matter air pollution (PM2.5) for 10 patients with asthma from 7 households near Salt Lake City, UT; the patients were followed for up to 373 days each. Compared to previous SOM algorithms using timestamp alignment on time series data, the DTW-SOM algorithm produced fewer quantization errors and more detailed diurnal patterns. DTW-SOM identified the expected typical diurnal patterns in outdoor temperature which varied by season, as well diurnal patterns in PM2.5 which may be related to daily asthma outcomes. In summary, DTW-SOM is an innovative feature engineering method that can be applied to highly time-resolved environmental exposures assessed by sensors to identify typical diurnal (or hourly or monthly) patterns and provide new insights into the health effects of environmental exposures. [ABSTRACT FROM AUTHOR]

Cellular composition and structural organization of cells in the tissue determine effective antitumor response and can predict patient outcome and therapy response. Here we present Seg-SOM, a method for dimensionality reduction of cell morphology in H&E-stained tissue images. Seg-SOM resolves cellular tissue heterogeneity and reveals complex tissue architecture. We leverage a self-organizing map (SOM) artificial neural network to group cells based on morphological features like shape and size. Seg-SOM allows for cell segmentation, systematic classification, and in silico cell labeling. We apply the Seg-SOM to a dataset of breast cancer progression images and find that clustering of SOM classes reveals groups of cells corresponding to fibroblasts, epithelial cells, and lymphocytes. We show that labeling the Lymphocyte SOM class on the breast tissue images accurately estimates lymphocytic infiltration. We further demonstrate how to use Seq-SOM in combination with non-negative matrix factorization to statistically describe the interaction of cell subtypes and use the interaction information as highly interpretable features for a histological classifier. Our work provides a framework for use of SOM in human pathology to resolve cellular composition of complex human tissues. We provide a python implementation and an easy-to-use docker deployment, enabling researchers to effortlessly featurize digitalized H&E-stained tissue. [ABSTRACT FROM AUTHOR]

Synoptic fingerprinting of flood driving rainfall events over the Hyderabad city, India were assessed through an unsupervised machine learning technique—self‐organizing maps (SOM) for the period 1979–2020. Flood dates were identified using nonscientific data sources, and then large‐scale climate variables, that is, integrated vapour transport (IVT) and geopotential height at 500 hPa (Z500) along with rainfall at daily scale were considered. SOM was applied on IVT and results indicated large spatial scale‐ and season‐specific mechanisms in addition to local circulation systems—this highlights SOM's ability to cluster circulation mechanisms as per the relevance. The rainy seasons, that is, monsoon and postmonsoon showed well‐known moisture pathways. While synoptic systems can be easily linked with monsoon and postmonsoon mechanisms, it is interesting to note that the local systems exist even during the monsoon and postmonsoon season in addition to the summer season. Spatial patterns of Z500 corresponding to SOM nodes of IVT showed the presence of low‐pressure systems (LPSs), tropical cyclones and trough systems. Further analysis was performed, and a total of 35 atmospheric rivers (ARs) were identified. Plots of back trajectories for the selected AR events indicated that Arabian Sea is the main source of moisture in the monsoon season followed by Bay of Bengal for postmonsoon and tropical cyclones in summer season events. Additionally, moisture from land is also observed as another source of moisture. Back‐trajectory analysis further indicated AR‐LPS interaction during the monsoon and ARs feeding moisture to LPSs. Overall, well‐known and season‐specific synoptic systems that brought significant rainfall to the study region were identified by the SOM. The proposed framework is adaptable for different locations, and as floods in cities across the globe are on the rise, these findings will have a prominent role in urban flood forecasting and management. [ABSTRACT FROM AUTHOR]

Supercell thunderstorms develop low-level rotation via tilting of environmental horizontal vorticity (ωh) by the updraft. This rotation induces dynamic lifting that can stretch near-surface vertical vorticity into a tornado. Low-level updraft rotation is generally thought to scale with 0–500 m storm-relative helicity (SRH): the combination of storm-relative flow, |SRF|, |ωh|, and cosϕ (where ϕ is the angle between SRF and ωh). It is unclear how much influence each component of SRH has in intensifying the low-level mesocyclone. This study surveys these three components using self-organizing maps (SOMs) to distill 15 906 proximity soundings for observed right-moving supercells. Statistical analyses reveal the component most highly correlated to SRH and to streamwise vorticity (ωs) in the observed profiles is |ωh|. Furthermore, |ωh| and |SRF| are themselves highly correlated due to their shared dependence on the hodograph length. The representative profiles produced by the SOMs were combined with a common thermodynamic profile to initialize quasi-realistic supercells in a cloud model. The simulations reveal that, across a range of real-world profiles, intense low-level mesocyclones are most closely linked to ωh and SRF, while the angle between them appears to be mostly inconsequential. Significance Statement: About three-fourths of all tornadoes are produced by rotating thunderstorms (supercells). When the part of the storm near cloud base (approximately 1 km above the ground) rotates more strongly, the chance of a tornado dramatically increases. The goal of this study is to identify the simplest characteristic(s) of the environmental wind profile that can be used to forecast the likelihood of strong cloud-base rotation. This study concludes that the most important ingredients for storm rotation are the magnitudes of the horizontal vertical wind shear between the surface and 500 m and the storm inflow wind, irrespective of their relative directions. This finding may lead to improved operational identification of environments favoring tornado formation. [ABSTRACT FROM AUTHOR]

Mesoscale eddies provide critical links between the atmosphere and the ocean interior modulating the exchanges of key climatic tracers across the air‐sea and mixed‐layer interface. Mesoscale eddies, widespread in the Southern Ocean, are unresolved by Earth System Models; to what extent they modify contemporary and future heat and CO2 uptake remains unknown. The missing contribution of mesoscale eddies and their effects on properties and fluxes of heat and CO2 are investigated in the South‐East Atlantic using a mesoscale resolving (1/12°) physical biogeochemical model. Unsupervised Self‐Organizing Maps are used to separate anticyclonic and cyclonic eddies into distinct sub‐categories based on spatial characteristics in relative vorticity. Thermal drivers dominated over non‐thermal on partial pressure of ocean CO2 (pCO2sw) leading to a weaker CO2 sink in anticyclonic eddies and a stronger CO2 sink in cyclonic eddies relative to the climatological mean. Magnitudes and depth extents of tracer anomalies scale with eddy size but are also dependent on background tracer gradients. Resolving only the largest (>50 km radii) eddies leads to underestimations in total magnitude of heat storage by 52% and 77%, and underestimations of 61% and 77% in dissolved inorganic carbon (CT) storage for anticyclonic and cyclonic eddies respectively. The inclusion of smaller eddy features (<50 km) also offsets the asymmetry in heat and CT storage between anticyclonic and cyclonic eddies. Resolving eddies explicitly or improvements in existing parameterizations that fail to sufficiently include eddy effects are crucial to improving estimates of heat and CO2. Plain Language Summary: Mesoscale eddies, abundant in the ocean, are coherent vortices of length scales 10‐100 km and persist for days to months. Eddies modify water properties by trapping water in their core at formation and through horizontal and vertical motions due to their rotation. These features provide links between the atmosphere and ocean influencing local heat and CO2 exchanges between the atmosphere and ocean. Mesoscale eddies are inadequately represented in Earth System Models (ESMs) due to computational expense and parameterizations that fail to capture their effects. Here, we use a high‐resolution ocean model and data analysis techniques to investigate the influence of different eddy types on heat and CO2 in the South–East Atlantic Ocean. Smaller eddies, more numerous than large eddies, were found to produce smaller tracer anomalies extending to shallower depths but the local environment also played a role. Anticyclonic and cyclonic eddies took up less and more CO2, respectively. Only including the effects of larger eddies (>50 km radii) leads to underestimations in heat and CO2 storage as well as an asymmetry between anticyclonic and cyclonic eddies highlighting the need for smaller features to be represented in ESMs for current and future climate projections. Key Points: Larger eddies have stronger and deeper imprints on tracer distributions and stronger influence on the fluxes compared to smaller eddiesResolving only large eddies (>50 km) leads to an underestimation of heat and dissolved inorganic carbon storageThermal controls of pCO2(sw) dominate over non‐thermal in both anticyclonic and cyclonic eddies [ABSTRACT FROM AUTHOR]

Sociological literature on cultural practices seeking to understand the social differentiation of taste pays limited attention to what people avoid consuming, despite its potential as a strategic indicator of taste. Avoidance has special relevance for the understanding of eating and drinking practices which are often characterized by exclusion of items for health, hedonic, reputational, or spiritual reasons. Making use of rich data on twenty-three items commonly consumed by Italian adults, this paper investigates how avoidances—i.e. what people claim never to eat or drink—are clustered, socially patterned and have evolved over time. Methodologically, we propose the novel use and integration of two machine learning techniques—Self-Organizing Maps (SOM) and Boosted Regression Trees (BRT)— to identify nine highly homogeneous avoidance clusters and examine the power of social variables in predicting the probability of individuals' belonging to various clusters and to further characterize them. We conclude by discussing possible rationales behind avoidance. [ABSTRACT FROM AUTHOR]

Forecasting air pollutant levels is essential in regulatory plans focused on controlling and mitigating air pollutants, such as particulate matter. Focusing the forecast on air pollution peaks is challenging and complex since the pollutant time series behavior is not regular and is affected by several environmental and urban factors. In this study, we propose a new hybrid method based on artificial neural networks to forecast daily extreme events of PM 2.5 pollution concentration. The hybrid method combines self-organizing maps to identify temporal patterns of excessive daily pollution found at different monitoring stations, with a set of multilayer perceptron to forecast extreme values of PM 2.5 for each cluster. The proposed model was applied to analyze five-year pollution data obtained from nine weather stations in the metropolitan area of Santiago, Chile. Simulation results show that the hybrid method improves performance metrics when forecasting daily extreme values of PM 2.5 . [ABSTRACT FROM AUTHOR]

The crop water stress index is receiving significant attention these days, especially in arid and semiarid regions, for quantifying water stress and effective irrigation scheduling. Nowadays, machine learning techniques such as neural networks are being widely used to determine CWSI. In the present study, Self-Organizing Maps (SOM) and Feed Forward-Back Propagation Artificial Neural Networks (FF-BP-ANN) are compared while determining the CWSI of rice crop. Irrigation field experiments with varying degrees of irrigation were conducted at the irrigation field laboratory of the Indian Institute of Technology, Roorkee, during the growing season of the rice crop. The CWSI of rice was computed empirically by measuring key meteorological variables (relative humidity, air temperature, and canopy temperature). The empirically computed CWSI was compared with SOM and FF-BP-ANN predicted CWSI. For the lower CWSI baseline of rice, a linear relationship between air and canopy temperature difference (Tc-Ta) and, air vapour pressure deficit (AVPD) was developed, whereas air temperature plus 3°C was taken for the upper CWSI baselines. The performance of SOM and FF-BP-ANN were compared by computing Nash–Sutcliffe efficiency (NSE), index of agreement (d), root mean squared error (RMSE), and coefficient of correlation (R2). It is found that FF-BP-ANN (R2 = 0.97, NSE = 0.92, d = 0.95, RMSE = 0.006) performs better than SOM (R2 = 0.88, NSE = 0.87, d = 0.9, RMSE = 0.075). [ABSTRACT FROM AUTHOR]

To observe the discrepancies between generated output power and load in a system with combined operation of small generation sources and conventional energy sources, the extensive work in this paper is to minimize frequency deviation by incorporating different controllers. Its functioning is verified through hybrid configuration of 3-area interconnected power system with Distributed Generation in area-1 integrated with hybrid sources and energy storage system, integrated with DG, thermal power system in area-2 and area-3. With this, the dynamic response of frequency is improved. Hybrid fuzzy with load frequency control with self-organizing maps with fuzzy controller which is an effective controller provides solution for an intricate effort in view of dynamic characteristics and random load variations are less when compared to the existing ones. The robustness of these controllers is systematically demonstrated under various loading conditions. [ABSTRACT FROM AUTHOR]

Despite the intensifying interest in flash drought both within the United States and globally, moist tropical landscapes have largely escaped the attention of the flash drought community. Because these ecozones are acclimatized to receiving regular, near-daily precipitation, they are especially vulnerable to rapid-drying events. This is particularly true within the Caribbean Sea basin where numerous small islands lack the surface and groundwater resources to cope with swiftly developing drought conditions. This study fills the tropical flash drought gap by examining the pervasiveness of flash drought across the pan-Caribbean region using a recently proposed criterion based on the evaporative demand drought index (EDDI). The EDDI identifies 46 instances of widespread flash drought "outbreaks" in which significant fractions of the pan-Caribbean encounter rapid drying over 15 days and then maintain this condition for another 15 days. Moreover, a self-organizing maps (SOM) classification reveals a tendency for flash drought to assume recurring typologies concentrated in one of the Central American, South American, or Greater Antilles coastlines, although a simultaneous, Caribbean-wide drought is never observed within the 40-yr (1981–2020) period examined. Furthermore, three of the six flash drought typologies identified by the SOM initiate most often during Phase 2 of the Madden–Julian oscillation. Collectively, these findings motivate the need to more critically examine the transferability of flash drought definitions into the global tropics, particularly for small water-vulnerable islands where even island-wide flash droughts may only occupy a few pixels in most reanalysis datasets. Significance Statement: The purpose of this study is to understand if flash drought occurs in tropical environments, specifically the Caribbean. Flash droughts represent a quickly evolving drought, which have particularly acute impacts on agriculture and often catch stakeholders by surprise as conditions evolve rapidly from wet to dry conditions. Our results indicate that flash droughts occur with regular periodicity in the Caribbean. Expansive flash droughts tend to occur in coherent subregional clusters. Future studies will further investigate the drivers of these flash droughts to create early warning systems for flash drought. [ABSTRACT FROM AUTHOR]

Previous findings show that large-scale atmospheric circulation plays an important role in driving Arctic sea ice variability from synoptic to seasonal time scales. While some circulation patterns responsible for Barents–Kara sea ice changes have been identified in previous works, the most important patterns and the role of their persistence remain unclear. Our study uses self-organizing maps to identify nine high-latitude circulation patterns responsible for day-to-day Barents–Kara sea ice changes. Circulation patterns with a high pressure center over the Urals (Scandinavia) and a low pressure center over Iceland (Greenland) are found to be the most important for Barents–Kara sea ice loss. Their opposite-phase counterparts are found to be the most important for sea ice growth. The persistence of these circulation patterns helps explain sea ice variability from synoptic to seasonal time scales. We further use sea ice models forced by observed atmospheric fields (including the surface circulation and temperature) to reproduce observed sea ice variability and diagnose the role of atmosphere-driven thermodynamic and dynamic processes. Results show that thermodynamic and dynamic processes similarly contribute to Barents–Kara sea ice concentration changes on synoptic time scales via circulation. On seasonal time scales, thermodynamic processes seem to play a stronger role than dynamic processes. Overall, our study highlights the importance of large-scale atmospheric circulation, its persistence, and varying physical processes in shaping sea ice variability across multiple time scales, which has implications for seasonal sea ice prediction. Significance Statement: Understanding what processes lead to Arctic sea ice changes is important due to their significant impacts on the ecosystem, weather, and shipping, and hence our society. A well-known process that causes sea ice changes is atmospheric circulation variability. We further pin down what circulation patterns and underlying mechanisms matter. We identify multiple circulation patterns responsible for sea ice loss and growth to different extents. We find that the circulation can cause sea ice loss by mechanically pushing sea ice northward and bringing warm and moist air to melt sea ice. The two processes are similarly important. Our study advances understanding of the Arctic sea ice variability with important implications for Arctic sea ice prediction. [ABSTRACT FROM AUTHOR]

Modern electrical power systems place special demands on the speed and accuracy of transient and steady-state process control. The introduction of renewable energy sources has significantly influenced the amount of inertia and uncertainty of transient processes occurring in energy systems. These changes have led to the need to clarify the existing principles for the implementation of devices for protecting power systems from the loss of small-signal and transient stability. Traditional methods of developing these devices do not provide the required adaptability due to the need to specify a list of accidents to be considered. Therefore, there is a clear need to develop fundamentally new devices for the emergency control of power system modes based on adaptive algorithms. This work proposes to develop emergency control methods based on the use of deep machine learning algorithms and obtained data from synchronized vector measurement devices. This approach makes it possible to ensure adaptability and high performance when choosing control actions. Recurrent neural networks, long short-term memory networks, restricted Boltzmann machines, and self-organizing maps were selected as deep learning algorithms. Testing was performed by using IEEE14, IEEE24, and IEEE39 power system models. Two data samples were considered: with and without data from synchronized vector measurement devices. The highest accuracy of classification of the control actions' value corresponds to the long short-term memory networks algorithm: the value of the accuracy factor was 94.31% without taking into account the data from the synchronized vector measurement devices and 94.45% when considering this data. The obtained results confirm the possibility of using deep learning algorithms to build an adaptive emergency control system for power systems. [ABSTRACT FROM AUTHOR]

The concept of tripartite synapses has revolutionized the world of neuroscience and the way we understand how information is transmitted in the brain. Since its discovery, some research groups have incorporated into connectionist systems classically focused on the development of Artificial Neuron Networks (ANNs) as a single element, artificial astrocytes that try to optimize performance in problem solving.In this systematic review, we searched the ISI Web of Science for papers that focused on the development of such novel models and their comparison with classical ANNs. A total of 22 papers that satisfied the inclusion criteria were analyzed, showing three different ways of applying the neuromodulatory influence of artificial astrocytes on neural networks. Using Multilayer Perceptron Networks, Artificial Neuro-Glial Newtworks and Multilayer Perceptron with Self-Organizing Maps approaches, a detailed analysis of the incorporation of artificial astrocytic networks has been carried out, and the main differences between the different methods have been weighed up. Regardless of the type of inclusion performed, the greater the complexity of the problem to be solved, it has been observed that the influence of artificial astrocytes has improved the performance of classical ANNs, as occurs in the biological brain. [ABSTRACT FROM AUTHOR]

The crop water stress index is receiving significant attention these days, especially in arid and semiarid regions, for quantifying water stress and effective irrigation scheduling. Nowadays, machine learning techniques such as neural networks are being widely used to determine CWSI. In the present study, Self-Organizing Maps (SOM) and Feed Forward-Back Propagation Artificial Neural Networks (FF-BP-ANN) are compared while determining the CWSI of rice crop. Irrigation field experiments with varying degrees of irrigation were conducted at the irrigation field laboratory of the Indian Institute of Technology, Roorkee, during the growing season of the rice crop. The CWSI of rice was computed empirically by measuring key meteorological variables (relative humidity, air temperature, and canopy temperature). The empirically computed CWSI was compared with SOM and FF-BP-ANN predicted CWSI. For the lower CWSI baseline of rice, a linear relationship between air and canopy temperature difference (Tc-Ta) and, air vapour pressure deficit (AVPD) was developed, whereas air temperature plus 3°C was taken for the upper CWSI baselines. The performance of SOM and FF-BP-ANN were compared by computing Nash–Sutcliffe efficiency (NSE), index of agreement (d), root mean squared error (RMSE), and coefficient of correlation (R2). It is found that FF-BP-ANN (R2 = 0.97, NSE = 0.92, d = 0.95, RMSE = 0.006) performs better than SOM (R2 = 0.88, NSE = 0.87, d = 0.9, RMSE = 0.075). [ABSTRACT FROM AUTHOR]

This paper addresses the riverine flood events that have occurred in Greece over the last 136 years (i.e., during the 1886–2022 period), focusing, amongst others, on the case of urban floods. The flood record of various sites of the country has been collected and analyzed to determine their spatial and temporal distribution. Greece is a country where flood data and records are very scarce. Therefore, as there is not an integrated catalog of Greek floods spanning from the 19th century to recently, this is the first attempt to create an integrated catalog for Greece. The sources used include published papers, local and regional newspapers and public bodies (mainly the Ministry of Environment and Energy and the official websites of Greek municipalities). Additionally, the main factors responsible for their occurrence have been issued, regarding the country's climatic, geological and geomorphological setting, as well as human interventions. In addition, the atmospheric circulation driving factors of floods are assessed via an unsupervised neural network approach (i.e., Self-Organizing Maps). Based on the results of this research, an online GIS-based database has been created, depicting the areas that have been struck by riverine floods in Greece. By clicking a flood event in the online database, one can view several characteristics, depending on data availability, such as duration and height of the rainfall that caused them and number of fatalities. Long-term trends of mean and extremes seasonal precipitation also linked to the spatial distribution of floods. Our analysis shows that urban floods are a very large portion of the overall flood record, and they mainly occur in the two large urban centers, Athens and Thessaloniki, as well as near large rivers such as Pineios. Autumn months and mainly November are the periods with higher flood hazards, based on past records and cyclonic atmospheric circulation constitutes the principal driving factor. Our results indicate that a flood catalog at national level is of fundamental importance, as it can provide valuable statistical insights regarding seasonality, spatial distribution of floods, etc., while it can also be used by stakeholders and researchers for flood management and flood risk analysis and modelling. [ABSTRACT FROM AUTHOR]

This study classifies the spatial distribution of heavy precipitation in summer (June–August) from 1979 to 2021 in the three provinces of Northeast China (TPNC) into two patterns by using the self‐organizing maps (SOM) neural network, and then quantitatively analyzes their moisture transport channels and sources using the Lagrangian model. The results show that the summer heavy precipitation in TPNC can be divided into the northern and southern patterns according to the distribution of the heavy precipitation. Both patterns of heavy precipitation are affected by the low‐level vortex west of TPNC, but the strength and shape of the low vortex are different. The northern pattern is mainly influenced by the westerly flow in the vortex in the mid‐high latitudes, which transports moisture from the upstream westerly region into TPNC. The southern pattern is mainly affected by the southerly jet stream southeast of TPNC, which conveys a large amount of moisture from the East Asian summer monsoon region into TPNC. In terms of the summer climatological mean, the northern pattern has a higher precipitation recycling rate, while the southern pattern has a lower recycling rate. [ABSTRACT FROM AUTHOR]

Climate model evaluation work has made progress both in theory and practice, providing strong support for better understanding and predicting climate change. However, at the weather scale, there is relatively little assessment of climate models in terms of daily-scale climate phenomena, such as storm frequency and intensity. These weather-scale variables are of significant importance for our understanding of the impacts of climate change. In order to assess the capability of climate models to simulate weather-scale climate patterns, this study employs Self-Organizing Maps (SOMs) for weather pattern classification. By combining different evaluation metrics, varying the number of SOM types, changing the size of the study area, and altering the reference datasets, the climate models are evaluated to ensure the robustness of the assessment results. The results demonstrate that the size of the study area is positively correlated with observed differences, and there are correlations among different evaluation metrics. The highest correlation is observed between evaluation metrics in large-scale and small-scale spatial domains, while the correlation with SOM size is relatively low. This suggests that the choice of evaluation metrics has a minor impact on model ranking. Furthermore, when comparing the correlation coefficients calculated using the same evaluation metrics for different-sized regions, a significant positive correlation is observed. This indicates that variations in the size of the study area do not significantly affect model ranking. Further investigation of the relationship between model performance and different SOM sizes shows a significant positive correlation. The impact of dataset selection on model ranking is also compared, revealing high consistency. This enhances the reliability of model ranking. Taking into account the influence of evaluation metric selection, SOM size, and reanalysis data selection on model performance assessment, significant variations in model ranking are observed. Based on cumulative ranking, the top five models identified are ACCESS1-0, GISS-E2-R, GFDL-CM3, MIROC4h, and GFDL-ESM2M. In conclusion, factors such as evaluation metric selection, study area size, and SOM size should be considered when assessing model ranking. Weather pattern classification plays a crucial role in climate model evaluation, as it helps us better understand model performance in different weather systems, assess their ability to simulate extreme weather events, and improve the design and evaluation methods of model ensemble predictions. These findings are of great significance for optimizing and strengthening climate model evaluation methods and provide valuable insights for future research. [ABSTRACT FROM AUTHOR]

Energy consumption represents a significant operating expense in the mining and minerals industry. Grinding accounts for more than half of the mining sector's total energy usage, where the semi-autogenous grinding (SAG) circuits are one of the main components. The implementation of control and automation strategies that can achieve production objectives along with energy efficiency is a common goal in concentrator plants. However, designing such controls requires a proper understanding of process dynamics, which are highly complex, coupled, and have non-deterministic components. This complex and non-deterministic nature makes it difficult maintain a set-point for control purposes, and hence operations focus on an optimal control region, which is defined in terms of desirable behavior. This paper investigates the feasibility of employing machine learning models to delineate distinct operational regions within in an SAG mill that can be used in advanced process control implementations to enhance productivity or energy efficiency. For this purpose, two approaches, namely k-means and self-organizing maps, were evaluated. Our results show that it is possible to identify operational regions delimited as clusters with consistent results. [ABSTRACT FROM AUTHOR]

Tool condition monitoring (TCM) is a mean to optimize production systems trying to use cutting tool life at its best. Nevertheless, nowadays available TCM algorithms typically lack robustness in order to be consistently applied in industrial scenarios. In this paper, an unsupervised artificial intelligence technique, based on Growing Self-Organizing Maps (GSOM), is presented in synergy with real-time specific force coefficients (SFC) estimation through the regression of instantaneous cutting forces. The conceived approach allows robustly mapping the SFC, exploiting process parameters and similarity to manage the variability of their estimation due to unmodelled phenomena, like machine dynamics and tool run-out. The devised approach allowed detecting the tool end-of-life in cutting tests with variable lubrication, machine tool and cutting speed, through the adoption of a self-starting control chart running on real-time clustered data. The solution was validated through the comparison of the GSOM framework with respect to the optimized self-starting control chart applied without GSOM clustering. The GSOM reached a root mean squared percentage error (RMSPE) of 13.2% with respect to 56.1% obtained with the analogous control chart in a full-set optimization scenario. When optimised on tests for a unique machine tool and tested on another machine tool, GSOM scored an RMSPE of 34.5%, whereas the optimized control chart scored 64.5%. [ABSTRACT FROM AUTHOR]

Climate model projections of atmospheric circulation patterns, their frequency, and associated temperature and precipitation anomalies under a high-end global warming scenario are assessed over the Pacific Northwest of North America for the final three decades of the twenty-first century. Model simulations are from phase 6 of the Coupled Model Intercomparison Project (CMIP6) and circulation patterns are identified using the self-organizing maps (SOMs) approach, applied to 500-hPa geopotential height (Z500) anomalies. Overall, the range of projected circulation patterns is similar to that in the current climate, especially in winter, whereas in summer the models project a general reduction in the magnitude of Z500 anomalies. Significant changes in pattern frequencies are also projected in summer, with an overall decrease in the frequency of patterns with large Z500 anomalies. In winter, patterns historically associated with anomalously cold weather in northern latitudes are projected to warm the most, and in summer the largest temperature increases are projected over inland areas. Precipitation is found to increase across all seasons and most SOM patterns. However, some summer patterns that are associated with above-average precipitation in the current climate are projected to become significantly drier by the end of the century. [ABSTRACT FROM AUTHOR]

Financial innovation, green investments, or climate change are changing insurers' business ecosystems, impacting their business behaviour and financial vulnerability. Supervisors and other stakeholders are interested in identifying the path toward deterioration in the insurance company's financial health as early as possible. Suptech tools enable them to discover more and to intervene in a timely manner. We propose an artificial intelligence approach using Kohonen's self-organizing maps. The dataset used for development and testing included yearly financial statements with 4058 observations for European composite insurance companies from 2012 to 2021. In a novel manner, the model investigates the behaviour of insurers, looking for similarities. The model forms a map. For the obtained groupings of companies from different geographical origins, a common characteristic was discovered regarding their future financial deterioration. A threshold defined using the solvency capital requirement (SCR) ratio being below 130% for the next year is applied to the map. On the test sample, the model correctly identified on average 86% of problematic companies and 79% of unproblematic companies. Changing the SCR ratio level enables differentiation into multiple map sections. The model does not rely on traditional methods, or the use of the SCR ratio as a dependent variable but looks for similarities in the actual insurer's financial behaviour. The proposed approach offers grounds for a Suptech tool of predictive generation to support early detection of the possible future financial distress of an insurance company. [ABSTRACT FROM AUTHOR]

To investigate patterns of horizontal atmospheric latent energy (LE) transport toward the Arctic, we applied the self-organizing maps (SOM) method to the daily vertically integrated horizontal LE flux from ERA5 in winter (January–March) during 1979–2021. A clear picture depicting the LE transport to the Arctic at a synoptic scale then emerged, with four primary transport pathways identified: the northern Europe, the Davis Strait, the Greenland Sea, and the Bering Strait pathways. The four primary pathways occurred at a comparable frequency, and noticeable interannual variability was observed in their time series of frequency during 1979–2021. Further analysis suggested that the northward LE transport through all these pathways is significantly modulated by cyclones, with the northern Europe and the Greenland Sea pathways being mostly affected. Generally, more frequent and stronger cyclones were observed near the entry regions of LE transport compared to other regions. Moreover, this study provides a comprehensive picture of how atmospheric LE transport is related to air temperature, moisture, surface heat flux, and sea ice anomalies over the Arctic Ocean in winter. Through a thermodynamic perspective, we argue that the deleterious impacts of poleward LE transport on Arctic sea ice are to a large extent attributable to the enhanced local atmosphere–ice interactions, which increase downward longwave radiation (DLR) plus turbulent fluxes, consequently warming the surface and promoting the loss of sea ice. According to the quantitative results, among the four primary pathways, LE transport through the Davis Strait and the Greenland Sea could cause the loss of Arctic sea ice most efficiently. [ABSTRACT FROM AUTHOR]

We investigate the large-scale weather patterns during extreme precipitation (PEx) events over the conterminous United States (CONUS) by applying a version of the quasigeostrophic (QG) omega equation. This work aims to develop a climatology of the weather patterns most related to PEx events during current climate. Extreme events are examined for each of seven regions defined by consistent annual cycles of precipitation and spanning the CONUS. For the CONUS we train several self-organizing maps (SOM) on a pressure–time series of vertical velocity from each of the advective forcing terms in the QG omega equation for each extreme event. The unsupervised learning of the SOM allows us to identify the most descriptive set of nine patterns in vertical velocity associated with precipitation extremes. This method finds multiple frontal- and cyclone-driven patterns while grouping primarily convective events into one pattern. Frontal events include a synoptic pattern consistent with West Coast atmospheric river events as well as pattern groups linked to developing and to mature ("occluded") frontal cyclones. The primary patterns found during PEx events vary seasonally and geographically. Frontal cyclone patterns are most common during PEx events during summer in the part of the Great Plains and during winter for the Northeast, Southeast, Pacific Northwest, and Southwest. Convection is the most common pattern during summer in all regions. Except in the Southeast, the annual cycles of monthly number of PEx events and average precipitation match well, partially validating our choice of regions to aggregate PEx events. [ABSTRACT FROM AUTHOR]

Bio-inspired computing principles are considered as a source of promising paradigms for fault-tolerant computation. Among bio-inspired approaches, neural networks are potentially capable of absorbing some degrees of vulnerability based on their natural properties. This calls for attention, since beyond energy, the growing number of defects in physical substrates is now a major constraint that affects the design of computing devices. However, studies have shown that most neural networks cannot be considered intrinsically fault tolerant without a proper design. In this paper, the fault tolerance of self-organizing maps (SOMs) is investigated, considering implementations targeted onto field programmable gate arrays, where the bit-flip fault model is employed to inject faults in registers. Quantization and distortion measures are used to evaluate performance on synthetic datasets under different fault ratios. Three passive techniques intended to enhance fault tolerance of SOMs during training/learning are also considered in the evaluation. We also evaluate the influence of technological choices on fault tolerance: sequential or parallel implementation, weight storage policies. Experimental results are analyzed through the evolution of neural prototypes during learning and fault injection. We show that SOMs benefit from an already desirable property: graceful degradation. Moreover, depending on some technological choices, SOMs may become very fault tolerant, and their fault tolerance even improves when weights are stored in an individualized way in the implementation. [ABSTRACT FROM AUTHOR]

The use of biota to analyze the distribution pattern of biogeographic regions is essential to gain a better understanding of the ecological processes that cause biotic differentiation and biodiversity at multiple spatiotemporal scales. Recently, the collection of high‐resolution biological distribution data (e.g., specimens) and advances in analytical theory have led to the quantitative analysis and more refined spatial delineation of biogeographic regions. This study was conducted to redefine floristic zones in the southern part of the Korean Peninsula and to better understand the eco‐evolutionary significance of the spatial distribution patterns. Based on 309,333 distribution data of 2,954 vascular plant species in the Korean Peninsula, we derived floristic zones using self‐organizing maps. We compared the characteristics of the derived regions with those of historical floristic zones and ecologically important environmental factors (climate, geology, and geography). In the clustering analysis of the floristic assemblages, four distinct regions were identified, namely, the cold floristic zone (Zone I) in high‐altitude regions at the center of the Korean Peninsula, cool floristic zone (Zone II) in high‐altitude regions in the south of the Korean Peninsula, warm floristic zone (Zone III) in low‐altitude regions in the central and southern parts of the Korean Peninsula, and maritime warm floristic zone (Zone IV) including the volcanic islands Jejudo and Ulleungdo. Totally, 1,099 taxa were common to the four floristic zones. Zone IV showed the highest abundance of specific plants (those found in only one zone), with 404 taxa. Our study improves floristic zone definitions using high‐resolution regional biological distribution data. It will help better understand and re‐establish regional species diversity. In addition, our study provides key data for hotspot analysis required for the conservation of plant diversity. [ABSTRACT FROM AUTHOR]

The emergence of IoT and advanced multimedia information systems have undoubtedly created a proliferation of video sensor data. Although diverse machine learning approaches are utilized to extract useful insights from these data, limitations occur when processing and accommodating the large volumes of video data, which are unlabeled and have previously unseen data structures. This brings out the importance of using self-structuring intelligence that can adapt to the nature of the data and with the ability to learn from multi-modal, spatiotemporal and unstructured data. Encompassing these advances, we propose a recurrent self-structuring machine learning approach for video processing using multi-stream hierarchical recurrent growing self-organizing maps (RGSOM) architecture. We have designed, implemented and evaluated the said approach using a human activity recognition video dataset (Weizmann dataset), achieving state-of-the-art accuracy of 93.5% in the unsupervised domain. We used both spatial and temporal data from the video as separate input feature streams, where RGSOMs were used to self-structure the video data in multi-streams for visual exploratory analysis and video classification. As potential implications, this study can contribute to the existing literature in advancing self-adaptation techniques for video sensor data processing. [ABSTRACT FROM AUTHOR]

The maintenance of genetic diversity is fundamental to ensure the population's viability and to perceive how the evolutionary factors act on these. Using self-organizing maps (SOM) may be interesting to organize the genetic diversity and evidence of the effects caused by dispersive and systematic factors. The objective of this work was to verify if the SOM are able to map the genetic drift, selection, migration and inbreeding effects over generations from the allelic (p and q) and genotypic (D, H and R) frequencies. Populations submitted to these factors were simulated by the software Genes. Afterward, the populations submitted to each factor were mapped by the SOM and analyzed. The topological map showed great efficiency in organizing the populations since it was possible to recognize the expected pattern of each effect tested. Thus, SOM have been shown to be efficient in organizing populations that are subject to a process that reduces variability, such as drift, inbreeding, and selection, and to processes that increase genetic variability, such as migration. [ABSTRACT FROM AUTHOR]

Decadal climate prediction can provide invaluable information for decisions made by government agencies and industry. Modes of internal variability of the ocean play an important role in determining the climate on decadal time scales. This study explores the possibility of using self-organizing maps (SOMs) to identify decadal climate variability, measure theoretical decadal predictability, and conduct decadal predictions of internal climate variability within a long control simulation. SOM is applied to an 11-yr running-mean winter sea surface temperature (SST) in the North Pacific and North Atlantic Oceans within the Community Earth System Model 1850 preindustrial simulation to identify patterns of internal variability in SSTs. Transition probability tables are calculated to identify preferred paths through the SOM with time. Results show both persistence and preferred evolutions of SST depending on the initial SST pattern. This method also provides a measure of the predictability of these SST patterns, with the North Atlantic being predictable at longer lead times than the North Pacific. In addition, decadal SST predictions using persistence, a first-order Markov chain, and lagged transition probabilities are conducted. The lagged transition probability predictions have a reemergence of prediction skill around lag 15 for both domains. Although the prediction skill is very low, it does imply that the SOM has the ability to predict some aspects of the internal variability of the system beyond 10 years. [ABSTRACT FROM AUTHOR]

In fields like security risk analysis, Fast Moving Consumer Goods, Internet of Things, or the banking sector, it is necessary to deal with large datasets containing a great list of variables. In these situations, the analysis becomes intricate and computationally expensive, so data reduction techniques play an important role. Prototype generation methods provide a reduced dataset with the same properties as the original. GSOMs (growing self-organizing maps) reduce the data size without the need for prefixing the number of neurons needed to represent the input space. To the best of the authors' knowledge, this is the first time that the GSOM is applied for reduction and generation of prototypes, posing an advantage over their predecessors, the SOMs (self-organizing maps), which do not have the automatic growth feature. This work addresses the use of a GSOM to reduce the number of prototypes to use in a 1-NN (1 nearest neighbor) classifier. The proposed methodology is applied to an income dataset for testing and a large bank dataset that contain classifications into two different groups. The 1-NN classifier is used to obtain predictions using the nodes of the GSOM as prototypes. This article demonstrates that GSOMs save a significant amount of time in obtaining nearly the same validation results as SOMs by comparing the classifications obtained in the bank dataset. The results show data reductions of more than 99%, and accuracies greater than 80% for the income dataset and 74% for the bank dataset. [ABSTRACT FROM AUTHOR]

Self Organizing Maps (SOM) are unsupervised neural networks suited for visualisation purposes and clustering analysis. This study uses SOM to solve a software engineering problem: detecting the most important (key) classes from software projects. Key classes are meant to link the most valuable concepts of a software system and in general these are found in the solution documentation. UML models created in the design phase become deprecated in time and tend to be a source of confusion for large legacy software. Therefore, developers try to reconstruct class diagrams from the source code using reverse engineering. However, the resulting diagram is often very cluttered and difficult to understand. There is an interest for automatic tools for building concise class diagrams, but the machine learning possibilities are not fully explored at the moment. This paper proposes two possible algorithms to transform SOM in a classification algorithm to solve this task, which involves separating the important classes - that should be on the diagrams - from the others, less important ones. Moreover, SOM is a reliable visualization tool which able to provide an insight about the structure of the analysed projects. [ABSTRACT FROM AUTHOR]

This study identified four patterns of regional extreme precipitation events (REPEs) in Central Asia (CA) and their crucial synoptic systems and multiscale interactions. Four patterns with distinct spatial distributions were identified in: northern Kazakhstan, southern Xinjiang, western CA, and the Tianshan Mountains. Focusing on the three most frequent REPEs, the kinetic energy (KE) cross‐scale transfer from the basic‐to synoptic‐scale windows exhibited a zonal dipole, resulting in the development and enhancement of REPEs in northern Kazakhstan. The available potential energy (APE) cross‐scale transfer exhibited opposing patterns between the upper and lower troposphere, indicating baroclinic instability in the lower troposphere and barotropic instability of the basic flow in the upper troposphere. Both mechanisms enhanced the Central Asian vortices (CAVs) in southern Xinjiang and induced REPEs. Conversely, the energy budgets exhibited baroclinic instability of the basic flow throughout the entire region when the Tianshan Mountains REPEs occurred, providing energy for prevalent CAVs. Plain Language Summary: A self‐organizing map was employed to classify the regional extreme precipitation events (REPEs) in Central Asia (CA) into four patterns with unique spatial distributions (northern Kazakhstan: P1, southern Xinjiang: P2, western CA: P3, and Tianshan Mountains: P4). This study further illuminated the multiscale interaction mechanisms for the development of key synoptic systems that generate REPEs for the three most frequent patterns (P1, P2, and P4). A zonal dipolar kinetic energy (KE) cross‐scale transfer occurred throughout the region in P1, resulting in the development of a long‐wave trough and the enhancement of REPEs. P2 is characterized by a negative eastward shift of the North Atlantic Oscillation‐like in the original field and the reconstructed intraseasonal‐scale field, which promotes the development of Central Asian vortices (CAVs) over southern Xinjiang. The available potential energy (APE) and KE cross‐scale transfer from the basic‐to synoptic‐scale windows indicate that baroclinic and barotropic instability occur in the lower and upper troposphere, respectively. Thus, providing sufficient energy and dynamic conditions for CAVs enhances REPEs. Furthermore, the Silk Road pattern‐like dominates over Eurasia, and CAVs occur over the northwestern Tianshan Mountains, as observed in P4. APE cross‐scale transfer induces strong baroclinic instability throughout the troposphere and provides energy for more prevalent CAVs that intensify the REPEs. Key Points: The long‐wave trough and Central Asia (CA) vortex are the crucial synoptic systems of summer regional extreme precipitation in CAThe crucial synoptic systems are powered by the basic‐ and intraseasonal‐scale baroclinic instabilityThe spatial inconsistency of local multiscale interactions affects the formation of the crucial synoptic systems and extreme precipitation [ABSTRACT FROM AUTHOR]

This study focuses on coalbed methane (CBM) wells in high-ranking coal seam as the research subject. Considering the influence of effective stress and matrix shrinkage, a comprehensive permeability calculation model for CBM reservoirs is established. Based on this model, the variations in pressure and permeability during well production are quantified. By integrating static geological parameters, a finely classified classification of CBM wells is achieved using self-organizing map (SOM) neural network. Subsequently, an analysis of production dynamic characteristics and productivity differences among different types of CBM wells is performed, followed by providing drainage optimization suggestions. The results of SOM analysis show that 7,000 m3/d and 1,500 m3/d can be used as the production boundaries for the wells with different productivity in Block P. The daily gas production of exceptional well exceeds 7,000 m3/d, and the permeability remains relatively stable throughout the drainage process of this well. The daily gas production of the potential well ranges from 1,500 to 7,000 m3/d, and the permeability exhibits a significant decrease during the drainage process. The daily gas production of Inefficient well is consistently below 1,500 m3/d with moderate permeability variation. In addition to well location and structural geology, production variability is also influenced by the matching of reservoir conditions and drainage systems. This is primarily manifested in discontinuous drainage systems and rapid decline in bottom hole pressure (BHP) during early production. The analysis of drainage parameters indicates that in order to achieve optimal production from CBM wells, the BHP should exhibit an initial rapid decline followed by a slowly decrease during the early production period, with an average pressure drop ranging from 0.005 to 0.02 MPa/d. The research findings can offer technical guidance for the future advancement of CBM in the P Block. [ABSTRACT FROM AUTHOR]

Recurrent neural networks (RNNs) are widely utilized in neural network research to capture spatiotemporal features in video data. However, their effectiveness heavily relies on the spatial features upon which they trained. This paper introduces innovative ensembles of features for constructing frame-wise structures by employing impactful neural network models with innovative training pipelines. These features are designed to enhance the recognition of hand gesture videos using RNN by leveraging temporal information. Recognizing hand gestures from videos is a complex task that presents considerable challenges. One notable challenge is the overlap in gesture motion, where different gesture categories exhibit similar hand poses within a single video clip. To overcome this issue, we were motivated to develop extensive and diverse features that offer a more comprehensive description of the gesture video clips, thereby mitigating recognition problems caused by images overlapping. Overall, our efforts to generate diverse features have yielded promising results in enhancing the recognition of hand gestures from videos, particularly in scenarios where overlap poses a significant challenge. We have combined the extracted features from a deep neural network trained from scratch with features obtained from various standard neural networks (Self-Organizing Map, Radial Base Function) that are trained to enhance the deep-trained features. The mutual arrangement for combining the shared features has configured new frame-wise image features. Furthermore, we have provided a performance comparison of the newly constructed frame-wise features through time-sharing to train RNN for recognition. The proposed models have been evaluated on two-hand gesture video datasets, where a preserving gesture sequence is crucial due to overlapping motions. Our work demonstrates a significant improvement in performance for both datasets. [ABSTRACT FROM AUTHOR]

The presence of road cracks is an important indicator of damage. Deep learning is a prevailing method for detecting cracks in road surface images because of its detection ability. Previous research works focused on supervised convolutional neural networks (CNNs) without non‐crack features or unsupervised crack analysis with limited accuracies. The novelty of this study is the addition of background classification. By increasing the number of non‐crack categories, CNNs are driven to learn non‐crack features and improve crack detection performances. Non‐crack images are preprocessed, and their features are extracted in an unsupervised way by a deep convolutional autoencoder. A self‐organizing map clusters features to obtain non‐crack categories. This study focusses on classification though the method can be adopted in parallel with the latest segmentation algorithms. Using common road crack datasets, modified deep CNN models significantly improved accuracy by 1%–4% and f‐measure by 3%–8%, compared to previous models. The modified visual geometry group (VGG) 16 showed the top‐level performance, 96% accuracy and 84%–85% f‐measure. The models drastically reduced false detection cases while maintaining their crack detection abilities. [ABSTRACT FROM AUTHOR]

The infrared (IR) absorption spectral data of 63 kinds of lubricating greases containing six different types of thickeners were obtained using the IR spectroscopy. The Kohonen neural network algorithm was used to identify the type of the lubricating grease. The results show that this machine learning method can effectively eliminate the interference fringes in the IR spectrum, and complete the feature selection and dimensionality reduction of the high-dimensional spectral data. The 63 kinds of greases exhibit spatial clustering under certain IR spectrum recognition spectral bands, which are linked to characteristic peaks of lubricating greases and improve the recognition accuracy of these greases. The model achieved recognition accuracy of 100.00%, 96.08%, 94.87%, 100.00%, and 87.50% for polyurea grease, calcium sulfonate composite grease, aluminum (Al)-based grease, bentonite grease, and lithium-based grease, respectively. Based on the different IR absorption spectrum bands produced by each kind of lubricating grease, the three-dimensional spatial distribution map of the lubricating grease drawn also verifies the accuracy of classification while recognizing the accuracy. This paper demonstrates fast recognition speed and high accuracy, proving that the Kohonen neural network algorithm has an efficient recognition ability for identifying the types of the lubricating grease. [ABSTRACT FROM AUTHOR]

Understanding the dynamics of ecosystem services (ESs) in arid landscapes and socio-ecological systems is crucial for sustainable development and human well-being. This study uses the Invest model to quantify the spatio-temporal changes in four key ecosystems services in Altay from 1990 to 2020: water yield (water yield), carbon stock (carbon stock), soil retention (soil retention), and habitat quality (habitat quality). The trade-offs/synergies between different ESs were investigated via Spearman's correlation analysis. Ecosystem service bundles (ESBs) were mapped using self-organizing mapping (SOM), and the key drivers of ES relationships and the spatio-temporal dynamics of ESBs were revealed through redundancy analysis. The results showed that water yield increased by 33.7% and soil retention increased by 1.2%, while carbon stock and habitat quality decreased by 3.5% and 1.24%, respectively. The spatial distribution pattern had a clear zonal pattern, with the northern mountainous areas higher than the southern desert areas. The six pairs of ESs, in general, showed mainly low trade-off and high synergistic relationships, with trade-offs between water yield and carbon stock, soil retention and habitat quality, and a decreasing trend of trade-offs over time. Four types of ESBs were distinguished, and the compositional differences and spatial distribution within each ESB were determined by interactions between ESs and landscape types. There are complex non-linear relationships between the drivers and the four ESBs in different years. Before 2010, ecological factors were the key drivers influencing the spatio-temporal changes in ESBs, whereas social and environmental factors combined to drive changes in ESB allocations after 2010. Additionally, this study found that the implementation of conservation measures, such as reforestation and sustainable land management practices, positively influenced the provision of ecosystem services in the Altay region. These findings underscore the importance of integrating conservation efforts into land use planning and decision-making processes to ensure the sustainable delivery of ecosystem services in arid landscapes. [ABSTRACT FROM AUTHOR]