Kohonen Self-organizing Map Research Articles

Strategic Formation of Agricultural Market Clusters in Ukraine: Emerging as a Global Player

This study investigates the cluster approach to optimize strategies for agricultural enterprises in Ukraine, emphasizing geographical proximity as a key factor in cluster formation. The research applies Kohonen Self-Organizing Maps (SOMs) and Ward’s hierarchical clustering to classify enterprises based on storage capabilities, transport logistics, crop yields, and military risk exposure. By analyzing these factors, this study identifies distinct patterns of innovation adoption, strategic management, and economic resilience among the clusters. The findings highlight variations in competitiveness and resource efficiency, providing a detailed understanding of regional economic performance. Unlike previous research, this study offers a novel integration of conflict-related risks into the clustering methodology, revealing new insights into how military factors influence cluster dynamics. Comprehensive maps and diagrams illustrate the spatial and economic distribution of clusters, aiding in visual interpretation. The results propose strategic measures tailored to enhance agricultural productivity and competitiveness, particularly in Ukraine’s current military context. This approach offers a more adaptive framework for managing agricultural enterprises, promoting resilience and long-term sustainability in global markets.

Cluster-analysis of human pulse wave by self-organizing Cohonen maps to identify markers of age groups

The present study is methodological in nature and describes an approach to the study of the human cardiovascular system. The assessment of biological age is important for the management of the human aging process. Regular monitoring of this indicator makes it possible to take preventive measures to reduce the likelihood of cardiovascular disasters. The aim of the study is to determine the age groups based on the pulse waves of the subjects using Kohonen neural networks. In this study, Kohonen self-organizing maps were used to cluster pulse waves of people of different ages by observing the dy-namic change of the pulse wave with the selection of parameters of the neural network self-organization algorithm. A data set of different age groups was used. Each group con-tained files of subjects of the same age, each file – pulse waves of these subjects. As a result of experiments with neural network configurations, cluster structures were obtained and geometric characteristics of pulse waves corresponding to age groups were estab-lished. The clustering algorithm used broke the pulse waves into clusters, the reliability of which in some cases reached 91–96 %. The software implementation of the trained neural network and the portability of the pulse wave measuring device allow the proposed tech-nique to be used to monitor the state of the cardiovascular system in people at risk of cardiovascular disasters, as well as people associated with dangerous professions, ath-letes, office workers, etc. Registration of changes in biological age contributes to the adoption of timely preventive measures, decisions on the modification of nutrition, life-style and physical activity. The resulting network can be used in diagnostic systems to support medical decision-making.

Achieving the Best Symmetry by Finding the Optimal Clustering Filters for Specific Lighting Conditions

This article explores the efficiency of various clustering methods for image segmentation under different luminosity conditions. Image segmentation plays a crucial role in computer vision applications, and clustering algorithms are commonly used for this purpose. The search for an adaptive clustering mechanism aims to ensure the maximum symmetry of real objects with objects/segments in their digital representations. However, clustering method performances can fluctuate with varying lighting conditions during image capture. Therefore, we assess the performance of several clustering algorithms—including K-Means, K-Medoids, Fuzzy C-Means, Possibilistic C-Means, Gustafson–Kessel, Entropy-based Fuzzy, Ridler–Calvard, Kohonen Self-Organizing Maps and MeanShift—across images captured under different illumination conditions. Additionally, we develop an adaptive image segmentation system utilizing empirical data. Conducted experiments highlight varied performances among clustering methods under different luminosity conditions. This research enhances a better understanding of luminosity’s impact on image segmentation and aids the method selection for diverse lighting scenarios.

Extraction of Iris Characteristics Using a Combination Transform for Biometric Any individual Identification

The iris of the eye is used as an identifying and affirmative biometric in many vital applications such as banking and airports.And iris identification is one of the most reliable and accurate biometric identification systems available due to its stability and lack of sensitivity to longevity. In this article, authorized standard processes such as iris segmentation, normalization, and feature extraction were utilized to examine the iris of the eye. The circular hough transform is used for iris segmentation, while the daughman rubber-sheet model is used for iris picture normalization. Using a combination transform consisting of four consecutive transforms, feature encoding was employed to restore the most distinguishing aspects of irises. The 2-D fast fourier transform (FFT), radon transform, 1-D inverse fast fourier transform (IFFT), and 1-D discrete multi-wavelet transform are all examples of transforms. For pattern clustering, the kohonen self-organizing feature map (SOFM) was employe to the group that distinguished information vectors of the iris image features extracted using the combination transforms in order to visualize the close proximity of comparable irises features and scatter dissimilar ones. The matching process is carried out between test iris images and target iris photos of 224 people from the IITD database, with the similarity measured using the euclidian distance metric A part investigates the change of the parameters: FRR, FAR, TSR) of different collections of the PID and POD with the threshold levels (90:10,80:20,70:30,60:40,50:50) that resulted EER for the various thresholds are 0.070,0.070,0.080,0.090 and 0.095 respectively. The suggested approach yielded a TSR of 98% when compared to conventional methods.

Reducing Cluster Overlap in Movie Recommendations with IKSOM and Silhouette Clustering

The exponential growth of online material requires the implementation of effective and precise recommendation systems in order to optimize the user experience. Nevertheless, conventional approaches frequently encounter problems such as cluster overlap, which reduces the accuracy of suggestions. This paper presents a new method for minimising the overlap between clusters in movie recommendation systems. It achieves this by combining Improved Kohonen Self-Organizing Maps (IKSOM) with Silhouette Clustering. The proposed method utilises IKSOM to efficiently represent high-dimensional user-item interactions in a two-dimensional space, enabling the formation of distinct and meaningful clusters. Subsequently, Silhouette Clustering is utilised to enhance the separation and cohesion of clusters, hence reducing overlap. The experimental findings show that proposed hybrid model works much better than the baseline techniques, obtaining an RMSE of 0.423 and MAE of 0.216. Additionally, it improves precision (93.6%), recall (94.2%) and F1-score (93.4%). Additionally, the proposed technique demonstrates a high level of accuracy (97.3%) with a precision rate of 95.8%. These results emphasise the method's efficacy in minimising errors and enhancing the overall performance of the recommendation system. The results indicate that combining IKSOM with Silhouette Clustering can improve the precision and dependability of movie recommendation systems by resolving cluster overlap and offering more individualised user experiences. Subsequent research will investigate the implementation of this method in different fields and the integration of supplementary contextual information to enhance the accuracy of recommendations.

Modeling carbon dioxide emissions reduction

The paper proposes a new scientific approach to modeling carbon dioxide (CO2) emissions in individual countries, based on the tools of Kohonen self-organizing maps and the provisions of economic theory. This work is based on conducted a comprehensive analysis of wide set of indicators influencing carbon dioxide emissions and decarbonization processes in 40 selected countries for 10 years time span from 2013 till 2022. During clustering these nations using self-organizing maps, 14 key indicators were chosen that cover economic and demographic growth, energy consumption, CO2 emission data, and includes trade of energy recourses for wider analysis and identification of countries with similar decarbonization potential, economic development and trade energy recourses possibilities. This approach exposed distinct clusters of nations with varying decarbonization capabilities and tracked the progress of analyzed countries in terms of CO2 emission changes in dynamic.The analysis of clustering was focused on identification leaders and followers in decarbonization activities. The leaders cluster includes Sweden, Norway, New Zealand and other countries exhibited leadership in developing a climate-resilient economy among 40 countries. Hence, all countries positioned closer to the leader cluster on the map demonstrated higher efficiency in their decarbonization pathways. Consequently, these properties of Kohonen maps provide a foundation for formulating general recommendations to achieve an efficient and effective low-carbon economy for followers including Ukraine, Kazakhstan and other energy intensive countries.Distinct countries decarbonization profiles resulting from clustering were used to develop practical recommendations towards estimating and establishing targeted CO2 emission levels for analyzed countries, conducting continued promoting renewable energy utilization, enhancing cross-border energy trade, diversifying energy portfolios, etc. Also, the proposed methodology can be utilized to project future emissions trends for each country based on cluster-specific models and facilitate policy decisions making process related to mitigating carbon emissions.

Typology of professional trajectories of gifted individuals using neural network analysis

Based on a data set (100 biographies) created by the authors through content analysis of biographical material about outstanding scientists of the 19th and 20th centuries in the humanities and natural sciences, the clustering of professional trajectories of gifted individuals was carried out. Neural network analysis based on self-organizing Kohonen maps was used as a clustering method. The professional trajectories were formed within the framework of the behavioral model of the linear-stage approach to studying life cycles. Within this approach, career and professional self-realization are understood as a sequence of evolutionary stages fixed in their order of occurrence. Each stage was encoded, and the biographies were transformed into a vector system. In turn, the task of clustering consisted in dividing a hundred vectors into typical groups with several real-valued coordinates. The criteria for the quality of clustering were the minimum sum of quantization errors and the silhouette coefficient. As a result of the study, seven professional trajectories of gifted individuals were identified and interpreted. The analysis of trajectories was carried out from the point of view of the speed of success (average age of success) and those factors and conditions of the life path that could affect either rapid or slow achievement of professional goals and self-realization. This example demonstrates the possibilities and limitations of using neural network analysis for solving similar research tasks, especially when working with complex cluster forms and finding their optimal number.

Credit Risk Assessment Using a Combined Approach of Supervised and Unsupervised Learning

In the financial industry, credit scoring is a crucial tool for assessing credit risk. The study aims to enhance the accuracy and reliability of credit scoring by combining supervised and unsupervised learning methods. We propose an integrated model that combines Kohonen's Self-Organizing Maps (SOM) with the Random Forest algorithm to provide a more comprehensive analysis of credit card user data. Key features for model training were identified through feature selection and extraction. Experimental results show that the integrated model improved the AUC from 0.82 to 0.89, increased user satisfaction from a score of 3.8 to 4.35, and boosted usage rates by 12.5%. Additionally, the integrated model significantly enhanced the discrimination and prediction accuracy of user credit risk. These findings indicate that the combination of unsupervised learning with Kohonen's Self-Organizing Maps and supervised learning with Random Forest can effectively improve the accuracy of credit scoring, providing financial institutions with a more reliable tool for credit risk assessment.

Probabilistic Nested Model Selection in Pharmacokinetic Analysis of DCE-MRI Data in Animal Model of Cerebral Tumor.

Best current practice in the analysis of dynamic contrast enhanced (DCE)-MRI is to employ a voxel-by-voxel model selection from a hierarchy of nested models. This nested model selection (NMS) assumes that the observed time-trace of contrast-agent (CA) concentration within a voxel, corresponds to a singular physiologically nested model. However, admixtures of different models may exist within a voxel's CA time-trace. This study introduces an unsupervised feature engineering technique (Kohonen-Self-Organizing-Map (K-SOM)) to estimate the voxel-wise probability of each nested model. Sixty-six immune-compromised-RNU rats were implanted with human U-251N cancer cells, and DCE-MRI data were acquired from all the rat brains. The time-trace of change in the longitudinalrelaxivity for all animals' brain voxels was calculated. DCE-MRI pharmacokinetic (PK) analysis was performed using NMS to estimate three model regions: Model-1: normal vasculature without leakage, Model-2: tumor tissues with leakage without back-flux to the vasculature, Model-3: tumor vessels with leakage and back-flux. Approximately two hundred thirty thousand (229,314) normalized profiles of animals' brain voxels along with their NMS results were used to build a K-SOM (topology-size: 8×8, with competitive-learning algorithm) and probability map of each model. K-fold nested-cross-validation (NCV, k=10) was used to evaluate the performance of the K-SOM probabilistic-NMS (PNMS) technique against the NMS technique. The K-SOM PNMS's estimation for the leaky tumor regions were strongly similar (Dice-Similarity-Coefficient, DSC=0.774 [CI: 0.731-0.823], and 0.866 [CI: 0.828-0.912] for Models 2 and 3, respectively) to their respective NMS regions. The mean-percent-differences (MPDs, NCV, k=10) for the estimated permeability parameters by the two techniques were: -28%, +18%, and +24%, for , and , respectively. The KSOM-PNMS technique produced microvasculature parameters and NMS regions less impacted by the arterial-input-function dispersion effect. This study introduces an unsupervised model-averaging technique (K-SOM) to estimate the contribution of different nested-models in PK analysis and provides a faster estimate of permeability parameters.

TRANSFORMATIONS OF THE RESOURCE MANAGEMENT STRATEGY OF UKRAINIAN BANKS

This article examines the peculiarities of the management of assets and liabilities of Ukrainian banks in the conditions of significant structural transformations of the resource base during the period of martial law. The analysis is carried out at the level of homogeneous structural and functional groups of banks (SFGBs), which are formed using published reporting data and the application of Kohonen's self-organizing map (SOM). Accumulation of statistical data has been carried out for 5 years, special attention is paid to structural changes in the resource base and directions of placement of bank assets over the past two years.Over the past two years, the bank has been under the influence of shock factors affecting assets and liabilities. At the beginning of 2022, there was an outflow of funds from bank accounts, which was gradually compensated by the inflow of current funds from corporations and the population of individuals. In 2023, the National Bank of Ukraine actively stimulated the development of the term resource base, the basis for ensuring the growth of credit operations. Transactions with state securities continue to grow in the structure of bank assets. The priority task of the banking system remains the financial support of business, but in the conditions of a full-scale war, such development of credit operations is limited. It is expedient to study the strategy of banks by combining the structure of assets and liabilities according to similar characteristics and analyzing the dynamics of groups. Observation of homogeneous groups confirms their stable nature, features of strategy, risk profile and development priorities. It has been proven that banks within homogeneous SFGBs demonstrate similar behaviour in the formation of management strategies and reactions to internal and external shocks. At the macro level, the SOM structure allows you to quantitatively assess the main processes taking place in the banking system, conduct comparisons with maps, and identify problems and priorities in the management of bank assets and liabilities. The SFGB method allows you to evaluate the trajectory of individual banks on the map and develop recommendations for improving the strategy of managing assets and liabilities.

Regionalization of the Onset and Offset of the Rainy Season in Senegal Using Kohonen Self-Organizing Maps

This study explores the spatiotemporal variability of the onset, end, and duration of the rainy season in Senegal. These phenological parameters, crucial for agricultural planning in West Africa, exhibit high interannual and spatial variability linked to precipitation. The objective is to detect and spatially classify these indices across Senegal using different approaches. Daily precipitation data and ERA5 reanalyses from 1981 to 2018 were utilized. The employed method enables the detection of key dates. Subsequently, the Kohonen algorithm spatially classifies these indices on topological maps. The results indicate a meridional gradient of the onset, progressively later from the southeast to the northwest, whereas the end follows a north–south gradient. The duration varies from 45 days in the north to 150 days in the south. The use of self-organizing maps allows for classifying the onset, end, and duration of the season into four zones for the onset and end, and three zones for the duration of the season. They highlight the interannual irregularity of transitions, with both early and late years. The dynamic analysis underscores the complex influence of atmospheric circulation fields, notably emphasizing the importance of low-level monsoon flux. These findings have tangible implications for improving seasonal forecasts and agricultural activity planning in Senegal. They provide information on the onset, end, and duration classes for each specific zone, which can be valuable for planning crops adapted to each region.

Is badfiction processed differently by the human brain? An electrophysical study on reading experience.

Literary reception is a special case of language processing. The judgment of literature reveals deep social patterns with embodied cognition. In this study, we investigate how differences in literary quality resonate in the human brain. Modifying a series of stimuli previously used in studies of the emotional potential of Harry Potter, we alternate passages from the original novels with passages from imitative and intentionally poorly written fanfiction. EEG data shows how the three text types are processed differently by the brain. Comparing the brain activity of the readers for the various text types, we see a difference in the absolute power but not in the relative power of the frequency bands. Reading badfiction evokes the lowest activity. However, the functionality of this activity is the same for all texts since the relative power of the frequency bands does not differ. When comparing the participant groups, we observe the opposite situation. Here, different relative powers of the frequency bands reflect different judgments and reading habits of participants. For example, fans of Harry Potter, regular readers of fantasy texts, and generally frequent readers read the texts more attentively, which is reflected in a pronounced relative activity of the theta and alpha frequency bands. Non-frequent readers and readers who are not devoted to Harry Potter and fantasy in general have increased activity in the delta frequency band. This suggests their saliency detection is more prominent because they are less familiar with reading or the subject matter. To support our findings, we use the EEG data without averaging over stimuli and participants, capturing the participants' responses on the level of individual stimuli. A Kohonen self-organizing map trained on this more extensive data finds reliably detectable differences in the responses to passages from the original Harry Potter novels and fan- and badfiction. Our study allows for an interpretation of an adaptive brain response. Readers who enjoy Harry Potter or have experience with the fantasy genre show different reactions from those who do not. Thus, badfiction appears to be processed differently by the human brain, but not for all readers in the same way.

Modelling of technical, environmental, and economic evaluations of the effect of the organic loading rate in semi-continuous anaerobic digestion of pre-treated organic fraction municipal solid waste

The study concerned technical feasibility, economic profitability, and carbon footprint (CF) analysis of semi-continuous anaerobic digestion (sAD) of organic fraction of municipal solid waste (OFMSW). The research assessed the pre-treatment effect on sAD by varying organic loading rates (OLR) from 3.38 to 6.75 kgvs/m3d. Three sAD configurations were investigated: hydrodynamic-cavitated (HC-OFMSW), enzymatically pre-treated (EN-OFMSW), and non-pre-treated (AD-OFMSW). Principal Component Analysis and Supervised Kohonen's Self-Organizing Maps combined the experimental, economic, and environmental evaluations.The sAD configurations were grouped predominantly according to the OLR however, within each OLR group the configurations were clustered according to the pre-treatments.The finding highlighted that pre-treatments offset inhibition in sAD of OFMSW due to the OLR increase, being economically profitable and CF negative up to 4.50 kgvs/m3d for EN-OFMSW and to 5.40 kgvs/m3d for HC-OFMSW. Whereas sAD-OFMSW remained economically and environmentally viable only up to 3.87 kgvs/m3d.HC-OFMSW reached the highest performance. In detail, for HC-OFMSW the NPV and CF ranged from 17679.30 to 43827.12 euros and from −51.08 to −407.210 kg CO2eq/1 MWh daily produced, by decreasing the OLR from 5.40 to 3.87 kgvs/m3d.These results are fundamental since pre-treatment is usually expensive due to additional energy or chemical requirements.

The ESG Patterns of Emerging-Market Companies: Are There Differences in Their Sustainable Behavior after COVID-19?

We aim to map the ESG patterns of emerging-market companies from 2018 to 2021 in order to determine whether the COVID-19 pandemic exerted any influence on sustainable corporate behavior. Thus, the ESG performances were assessed by employing the Kohonen Self-Organizing Map (also known as the Kohonen neural network) for clustering purposes at three levels: (i) ESG overall, including country and sectoral perspectives; (ii) ESG thematic; and (iii) ESG four-folded (stakeholder, perspective, management, and focus strategic views). Our results show that emerging-market companies focus their ESG efforts on social and governance issues rather than on environmental. However, environmental and social behavior differ more acutely than governance behavior across clusters. The analyses of country-level ESG performance and that of eleven market-based economic sectors corroborate the geographic and sector dependence of ESG performance. The thematic-level analysis indicates that operational activities and community issues received more attention, which suggests that emerging-market companies address distinct ESG topics according to their particularities and competitiveness. Furthermore, our empirical findings provide evidence that the ESG behavior of companies has changed over the course of the COVID-19 pandemic. Thus, our findings are relevant to policy makers involved in regulating ESG disclosure practices, investors focused on enhancing their sustainable investment strategies, and firms engaged in improving their ESG involvement.

Leveraging Machine Learning For Enhanced Database Integration

In this research endeavor, our primary objective revolves around the intricate task of matching schemas across diverse databases. The challenge lies in recognizing the inherent relationships between attributes and employing sophisticated machine learning algorithms as our guide. Doing so aims to identify attributes that harbor comparable or akin values, thus facilitating potential mappings between the two databases for seamless integration. Our investigation delves into the complexities of mapping scenarios, encompassing the conventional one-to-one mappings and the intricate scenarios where a single attribute can map to multiple attributes. Within the scope of this report, an in-depth exploration of one-to-one mapping utilizing the powerful Kohonen Self-Organizing Maps is presented. The theoretical underpinnings of this approach and the experiments conducted to evaluate its efficacy are also elucidated. The outcomes and insights gained from these experiments are meticulously documented and presented. By unraveling the intricate nature of schema matching using machine learning techniques, this research contributes to the ever-growing body of knowledge in this domain. Our findings shed light on the potential of employing advanced algorithms to tackle the challenges of schema matching, paving the way for enhanced database integration.

An R-adaptive algorithm based on self-organizing maps for solving incompressible flows with high-order discontinuous Galerkin methods

Mesh quality is critical for the numerical accuracy of CFD (Computational Fluid Dynamics). Although various techniques have been developed to improve mesh applicability to complex flows, r-adaptive methods have received far too little attention. This paper introduces an r-adaptive algorithm based on the self-organizing maps (SOM) of Kohonen and applies it to unsteady CFD applications using the Discontinuous Galerkin (DG) method. Inspired by the properties of the DG method, the numerical discontinuity over the element interface is quantified as an adaptation indicator. The method requires only one adaptation, followed by solution projection onto the new mesh to save computational costs in unsteady cases. A significant advantage of this method is its higher efficiency without solving complex equations during the adaptation. The performance of the developed mesh adaptation algorithm is tested on the steady laminar viscous flow past a circular cylinder at Re=40 and the unsteady laminar viscous flow past a circular cylinder at Re=100 and two circular cylinders in a side-by-side arrangement at Re=200. Significant improvements in flow contours and coefficients are obtained after adaptation.

Effectiveness of tutoring at school: A machine learning evaluation

Tutoring programs are effective in reducing school failures among at-risk students. However, there is still room for improvement in maximising the social returns they provide on investments.Many factors and components can affect student engagement in a program and academic success. This complexity presents a challenge for Public Administrations to use their budgets as efficiently as possible. Our research focuses on providing public administration with advanced decision-making tools.First, we analyse a database with information on 2066 students of the Programa para la Mejora de Éxito Educativo (Programme for the Improvement of Academic Success) of the Junta de Comunidades de Castilla y Léon in Spain, in 2018–2019, the academic year previous to the pandemic. This program is designed to help schools with students at risk of failure in Spanish, literature, mathematics, and English. We developed a machine learning model (ML) based on Kohonen self-organising maps (SOMs), which are a type of unsupervised (ANN), to group students based on their characteristics, the type of tutoring program in which they were enrolled, and their results in both the completion of the program and the 4th year of Compulsory Secondary Education (ESO).Second, we evaluated the results of tutoring programs and identified and explained how different factors and components affect student engagement and academic success.Our findings provide Public Administrations with better decision-making tools to evaluate and measure the results of tutoring programs in terms of social return on investment, improve the design of these programs, and choose the students to enrol.

Combination of multivariate data analysis and mixing modelling to assess tracer potential of contaminants of emerging concern in aquifers

Collected evidence has shown that contaminants of emerging concern (CECs) in conjunction with more conventional tracers (major ions, nutrients, isotopes etc.) can be used to trace pollution origin in aquatic systems. However, in highly mixed aquifer systems signals obtained from conventional tracers overlap diminishing their potential to be used as tracers. In this study, we present an approach that incorporates multivariate statistical analysis (principal component analysis (PCA) and Kohonen's Self-Organizing Map method (SOM)) and mixing modelling to identify the most suitable CECs to be employed as anthropogenic tracers. The study area is located in the Besòs River Delta (Barcelona, NE Spain) and represents the highly mixed aquifer system. A one-year monthly based monitoring campaign was performed to collect the information about the concentrations of 105 CECs as well as major and minor ions in the river and along the groundwater flow. The dimensionality of the obtained dataset was reduced to 25 CECs, based on their estimated health risk effects, for multivariate data analysis. The obtained results showed the overlap of conventional tracers' signals obtained from PCA. In case of CECs, PCA revealed differences in their distributions allowing the differentiation of the roles of natural attenuation processes, local and regional flows on their occurrence in different parts of the aquifer. This was not possible to do using solely CECs' distribution profiles. SOMs provided the lacking information about the modality of the distribution of each CECs, revealing their ability to represent factors controlling the groundwater hydrochemistry, which assist in defining their tracer potential. Based on the obtained results four identified persistent CECs, two with unimodal (lamotrigine and 5-Desamino-5-oxo-lamotrigine) and two with bimodal (carbamazepine and diazepam (higher modality was not revealed)) distributions, were selected to run a mixing model to compare their tracer performance.

Do IT users behave responsibly in terms of cybercrime protection?

This study aims to analyze the behaviour of IT users regarding their personal protection against potential cybercrimes. The research data set is based on surveys conducted by the European Commission in 2020-2021 for 35 European countries. Canonical analysis revealed that 66.67% of cybercrime cases (Phishing, Pharming, Online identity theft, etc.) determine individuals' choice of personal protection method (using a security token, social media logins, electronic identification, etc.). Kohonen's self-organizing maps were used to form 9 clusters of countries depending on the attitude of IT users to personal cybersecurity. The map results showed that individuals behave less responsibly using a security token, electronic identification certificate or card, pin code list or random characters of a password, and other electronic identification procedures. Users from Denmark, the Netherlands, Iceland, Norway, the UK, Austria, and Finland were the most responsible Europeans in terms of personal protection, while people from Bulgaria, Romania, Serbia, Albania, North Macedonia, Bosnia and Herzegovina were the least conscientious about protection.

Somtimes: self organizing maps for time series clustering and its application to serious illness conversations

There is demand for scalable algorithms capable of clustering and analyzing large time series data. The Kohonen self-organizing map (SOM) is an unsupervised artificial neural network for clustering, visualizing, and reducing the dimensionality of complex data. Like all clustering methods, it requires a measure of similarity between input data (in this work time series). Dynamic time warping (DTW) is one such measure, and a top performer that accommodates distortions when aligning time series. Despite its popularity in clustering, DTW is limited in practice because the runtime complexity is quadratic with the length of the time series. To address this, we present a new a self-organizing map for clustering TIME Series, called SOMTimeS, which uses DTW as the distance measure. The method has similar accuracy compared with other DTW-based clustering algorithms, yet scales better and runs faster. The computational performance stems from the pruning of unnecessary DTW computations during the SOM’s training phase. For comparison, we implement a similar pruning strategy for K-means, and call the latter K-TimeS. SOMTimeS and K-TimeS pruned 43% and 50% of the total DTW computations, respectively. Pruning effectiveness, accuracy, execution time and scalability are evaluated using 112 benchmark time series datasets from the UC Riverside classification archive, and show that for similar accuracy, a 1.8×\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ imes$$\\end{document} speed-up on average for SOMTimeS and K-TimeS, respectively with that rates vary between 1×\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ imes$$\\end{document} and 18×\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ imes$$\\end{document} depending on the dataset. We also apply SOMTimeS to a healthcare study of patient-clinician serious illness conversations to demonstrate the algorithm’s utility with complex, temporally sequenced natural language.