Unsupervised Self-organizing Map Research Articles

Use of Statistical Tools and Artificial Intelligence for Hydrosedimentological Analysis in Experimental Units in Semiarid Brazil

Objective: To compare statistical and artificial intelligence tools to assist in the analysis and influence of hydrosedimentological variables on runoff generation and sediment yield in experimental units in the Brazilian semiarid region. Theoretical framework: The increasing frequency of extreme weather events in areas prone to natural disasters has led to economic, social, and loss of life due to the lack of urban planning, disorderly occupation in densely populated areas, intensive use of natural resources, and difficulty in maintaining and conserving natural ecosystems (CARMO & LIMA, 2020). Continuous monitoring of hydrosedimentological processes is necessary for natural resource planning and management. There is a need for expanding long-term environmental monitoring in Brazil, which is not simple and costly. In hard-to-reach areas, investment in tools and variable analysis for secure estimation of flow and erosion is necessary (AIRES, 2022). Method: Data obtained from the São João do Cariri Experimental Basin - SJCEB (13.6 km²) were used through conventional rain gauges and automatic data collection platforms from three micro-watersheds (0.16 – 1.63 ha) and two experimental plots (100 m²) located in the semiarid region of Paraíba, Northeast Brazil. For the analyses, 147 rain events that generated surface runoff in at least one of the five experimental units of the SJCEB were used, with data on area (m²), vegetation cover (0 no cover and 1 with vegetation cover over the entire area), average slope (%), daily precipitation (mm), period without rainfall (days), runoff depth (mm), and sediment yield (kg.ha⁻¹), from 2001 to 2008. The statistical and artificial intelligence tools used in the analyses were: simple linear regression, Pearson correlation, and Self-Organizing Maps (SOM). The models were implemented in the R programming language. Results and conclusion: In the statistical summary of the studied variables, precipitation ranged from 2.2 mm to 133.0 mm, with a mean of 21.1 mm. The period without rainfall ranged from 0 to 240 days, with an average of 16.2 days. Slope varied from 3.40% to 7.50%, with a mean of 5.7%. The average vegetation cover was 0.44, with a median of 0.60. The Pearson correlation analysis revealed that precipitation showed a positive and significant correlation with surface runoff and sediment yield variables, ranging from moderate to strong. Simple linear regression was unable to provide satisfactory answers to assist in the analysis of runoff and sediment yield, therefore considered inefficient for hydrosedimentological regionalization. The unsupervised SOM network, by presenting complex relationships among the studied variables, detected that sediment yield is particularly influenced by the interactions with precipitation, vegetation cover, and slope. Thus, it can be inferred that models based on unsupervised neural networks are attractive alternatives, as they were able to extract complex and nonlinear relationships among the analyzed variables. When there are complex cause-and-effect relationships between variables of hydrosedimentological processes, the use of nonlinear and multivariate models, such as artificial neural networks. Implications of the research: Manipulations in experimental areas have favored the measurement of flow and quantification of sediment yield. However, the high costs to maintain physical structures and the presence of specialized personnel for monitoring, processing, and dissemination of collected data are factors that may make it impractical to maintain and expand environmental monitoring networks in Brazil. In ungauged areas, the prediction of flow and erosion is necessary, and can be performed through mathematical, statistical, or artificial intelligence tools that facilitate the identification of priority areas for intervention. Manipulations in experimental areas have favored the measurement of runoff and sediment yield, but the high costs of maintaining the physical structures and the presence of specialized personnel to monitor, process and disseminate the collected data are factors that make maintenance and the expansion environmental monitoring networks in Brazil. In non-instrumented areas, forecasting and analysis of runoff and erosion becomes necessary, using mathematical, statistical or artificial intelligence tools that favor the identification of priority areas for intervention. Originality/value: Use of different statistical and artificial intelligence tools for hydrosedimentological regionalization in small experimental units in the Brazilian semiarid region, with emphasis on the application of unsupervised learning neural networks (SOM).

Comparative Analysis of the Clustering Quality in Self-Organizing Maps for Human Posture Classification.

The objective of this article is to develop a methodology for selecting the appropriate number of clusters to group and identify human postures using neural networks with unsupervised self-organizing maps. Although unsupervised clustering algorithms have proven effective in recognizing human postures, many works are limited to testing which data are correctly or incorrectly recognized. They often neglect the task of selecting the appropriate number of groups (where the number of clusters corresponds to the number of output neurons, i.e., the number of postures) using clustering quality assessments. The use of quality scores to determine the number of clusters frees the expert to make subjective decisions about the number of postures, enabling the use of unsupervised learning. Due to high dimensionality and data variability, expert decisions (referred to as data labeling) can be difficult and time-consuming. In our case, there is no manual labeling step. We introduce a new clustering quality score: the discriminant score (DS). We describe the process of selecting the most suitable number of postures using human activity records captured by RGB-D cameras. Comparative studies on the usefulness of popular clustering quality scores-such as the silhouette coefficient, Dunn index, Calinski-Harabasz index, Davies-Bouldin index, and DS-for posture classification tasks are presented, along with graphical illustrations of the results produced by DS. The findings show that DS offers good quality in posture recognition, effectively following postural transitions and similarities.

A Self-Rectifying Synaptic Memristor Array with Ultrahigh Weight Potentiation Linearity for a Self-Organizing-Map Neural Network

Two-terminal self-rectifying (SR)-synaptic memristors are preeminent candidates for high-density and efficient neuromorphic computing, especially for future three-dimensional integrated systems, which can self-suppress the sneak path current in crossbar arrays. However, SR-synaptic memristors face the critical challenges of nonlinear weight potentiation and steep depression, hindering their application in conventional artificial neural networks (ANNs). Here, a SR-synaptic memristor (Pt/NiOx/WO3-x:Ti/W) and cross-point array with sneak path current suppression features and ultrahigh-weight potentiation linearity up to 0.9997 are introduced. The image contrast enhancement and background filtering are demonstrated on the basis of the device array. Moreover, an unsupervised self-organizing map (SOM) neural network is first developed for orientation recognition with high recognition accuracy (0.98) and training efficiency and high resilience toward both noises and steep synaptic depression. These results solve the challenges of SR memristors in the conventional ANN, extending the possibilities of large-scale oxide SR-synaptic arrays for high-density, efficient, and accurate neuromorphic computing.

Analysis of chemotypes and their markers in leaves of core collections of Eucommia ulmoides using metabolomics.

The leaves of Eucommia ulmoides contain various active compunds and nutritional components, and have successively been included as raw materials in the Chinese Pharmacopoeia, the Health Food Raw Material Catalogue, and the Feed Raw Material Catalogue. Core collections of E. ulmoides had been constructed from the conserved germplasm resources basing on molecular markers and morphological traits, however, the metabolite diversity and variation in this core population were little understood. Metabolite profiles of E. ulmoides leaves of 193 core collections were comprehensively characterized by GC-MS and LC-MS/MS based non-targeted metabolomics in present study. Totally 1,100 metabolites were identified and that belonged to 18 categories, and contained 120 active ingredients for traditional Chinese medicine (TCM) and 85 disease-resistant metabolites. Four leaf chemotypes of the core collections were established by integrated uses of unsupervised self-organizing map (SOM), supervised orthogonal partial least squares discriminant analysis (OPLS-DA) and random forest (RF) statistical methods, 30, 23, 43, and 23 chemomarkers were screened corresponding to the four chemotypes, respectively. The morphological markers for the chemotypes were obtained by weighted gene co-expression network analysis (WGCNA) between the chenomarkers and the morphological traits, with leaf length (LL), chlorophyll reference value (CRV), leaf dentate height (LDH), and leaf thickness (LT) corresponding to chemotypes I, II, III, and IV, respectively. Contents of quercetin-3-O-pentosidine, isoquercitrin were closely correlated to LL, leaf area (LA), and leaf perimeter (LP), suggesting the quercetin derivatives might influence the growth and development of E. ulmoides leaf shape.

Target-scale prospectivity modeling for gold mineralization within the Rajapalot Au-Co project area in northern Fennoscandian Shield, Finland. Part 2: Application of self-organizing maps and artificial neural networks for exploration targeting

This paper is part of a two-publication series reporting target-scale prospectivity modeling results for gold enrichment in the Rajapalot project area. The study area is located in the Northern Fennoscandian Shield in Finland and consists of high-grade epigenetic-hydrothermal Au-Co prospects. The first publication of this series described the implementation of knowledge-driven- and hybrid- expert systems. In this second paper we demonstrate the application of machine learning methods such as unsupervised self-organizing maps (SOM) and K-means clustering and supervised artificial neural networks (ANNs). The results from SOM allowed the mapping of deposit-related geological patterns in the evidential layers. K-means clustering of the SOM results identified data clusters favorable for gold enrichment. Quantitative prospectivity values were computed using an ANN model trained on the SOM codebook vectors. This improved the resolution of the prospective-exploration areas mapped within the geospatial domains of the clusters, and reduced the exploration search areas. The ANN result has a significant area under curve value of 0.869 in the receiver operating characteristic plots, confirming the predictive ability of the model. Results identify new areas for exploration, as well as help prioritise drilling targets around the currently less explored prospects. In this second paper we therefore conclude that (1) SOM is an effective method for mapping and visualization of patterns (in the evidential layers) related to mineral deposits at the target scale, and (2) supervised classification of the SOM results, such as using an ANN, can help define new target areas for drilling around the existing prospects. Finally, this two-publication series collectively demonstrates that advanced machine-learning based pattern recognition and prospectivity modeling routines can be used for exploration targeting at camp- and mine- scales. The knowledge-driven methods are efficient at modeling mineralization processes and the data-driven methods are effective at identifying geological attributes related to mineral deposits and occurrences.

GVDeepNet: Unsupervised Deep Learning Techniques for Effective Genetic Variant Classification

Many lives have been lost due to genetic diseases and the inbility to identify them. The genetic disorder is mainly because of the alteration in the common DNA nucleotide sequence, where benign and pathogenic are the common examples of these genetic variants. Deliberate changes for the gene mutation may cause unexpected results; at times, required results occur though it is highly likely to get unexpected outcomes too. In this paper, two unsupervised deep learning classification methods have been proposed to classify these genetic changes. For this work, Self-Organizing Map (SOM) and Autoencoder models have been used. SOM is an unsupervised learning technique used to obtain a low dimensional representation of the data. The SOM has been implemented using MiniSOM library. Autoencoder comprises an encoder and decoder component. The information encoded by encoder is decoded using the decoder component to obtain as close representation to the input as possible. The analysis were performed on ClinVar dataset comprising 6 lac records. The dataset is publicly available. The data was first subjected to pre-processing to handle missing and duplicate values. The result showed the good performance of Autoencoder, where its accuracy is 97% (on Test Data), and SOM has an accuracy of 96% (on Test Data). It has been concluded that unsupervised deep learning models, SOM and Autoencoder, retain enough prediction power to classify and identify if the underline alternation in the gene gives positives changes or the contrary

Machine Learning Analysis of Essential Oils from Cuban Plants: Potential Activity against Protozoa Parasites.

Essential oils (EOs) are a mixture of chemical compounds with a long history of use in food, cosmetics, perfumes, agricultural and pharmaceuticals industries. The main object of this study was to find chemical patterns between 45 EOs and antiprotozoal activity (antiplasmodial, antileishmanial and antitrypanosomal), using different machine learning algorithms. In the analyses, 45 samples of EOs were included, using unsupervised Self-Organizing Maps (SOM) and supervised Random Forest (RF) methodologies. In the generated map, the hit rate was higher than 70% and the results demonstrate that it is possible find chemical patterns using a supervised and unsupervised machine learning approach. A total of 20 compounds were identified (19 are terpenes and one sulfur-containing compound), which was compared with literature reports. These models can be used to investigate and screen for bioactivity of EOs that have antiprotozoal activity more effectively and with less time and financial cost.

Identification of flow regimes in boiling flows in a vertical annulus channel with machine learning techniques

Machine learning techniques are applied to identify the flow regimes of boiling flows in a vertical annulus channel using conductivity probe signals as the input data. The boiling dataset used in this work spans a system pressure of 198.7–907.2 kPa, a heat flux of 68–285 kW/m2, an inlet subcooling of 3.8–11.4 °C, a mass flux of 642.6–2465.1 kg/m2 s, an area-averaged gas superficial velocity of 0.01–2.83 m/s, and an area-averaged liquid superficial velocity of 0.68–2.69 m/s. A two-step approach that uses the unsupervised self-organizing map (SOM) for identifying the local and global flow regimes is proposed. The cumulative distribution functions (CDFs) of bubble durations are used as the input of the SOM. The combinations of local flow regimes that form the global regimes are extensively analyzed. The radial profiles of the global flow regimes are studied where the effects of the heated inner wall and local subcooling are discussed. Furthermore, the supervised support vector machine (SVM) and the K-nearest neighbors (KNN) algorithms are trained with features extracted from the conductivity probe signals for flow regime classification, with the global flow regimes identified by the SOM as the reference. Two sets of features are generated where the first consists of four statistical features of bubble durations and the second consists of six flow features that describe the collective bubble behaviors in the flow. Using the global flow regimes identified by the SOM as the reference and the flow features dataset as the input data, the trained models are able to classify the flow regimes to an accuracy of more than 90%.

NeuroSOFM: A Neuromorphic Self-Organizing Feature Map Heterogeneously Integrating RRAM and FeFET

Many currently available hardware implementations of the unsupervised self-organizing feature map (SOFM) algorithm utilize complementary metal–oxide–semiconductor (CMOS)-only circuits that often compromise key behaviors of the SOFM algorithm due to complexity. We propose a neuromorphic architecture harnessing the unique properties of ferroelectric field-effect transistors (FeFETs) and gated-resistive random access memory (RRAM) for in-memory computing to implement the SOFM algorithm. The FeFET-based synapse, organized in a novel circuit, is able to compute the input-weight Euclidean error in memory via the saturation drain current. The self-decaying states of the gated-RRAM allow for a self-decaying neighborhood and learning rate implementation to allow for convergence and lifelong learning. This novel architecture is able to successfully cluster benchmarks (RGB colors and MNIST handwritten digits) and real-life datasets, such as COVID-19 patient chest X-rays completely unsupervised. The architecture also demonstrates a significant amount of robustness to device variability and damaged neurons. In addition, the proposed architecture is completely parallelized and provides a power-efficient platform for implementing the SOFM algorithm.

Cloud Detection Using an Ensemble of Pixel-Based Machine Learning Models Incorporating Unsupervised Classification

Remote sensing imagery, such as that provided by the United States Geological Survey (USGS) Landsat satellites, has been widely used to study environmental protection, hazard analysis, and urban planning for decades. Clouds are a constant challenge for such imagery and, if not handled correctly, can cause a variety of issues for a wide range of remote sensing analyses. Typically, cloud mask algorithms use the entire image; in this study we present an ensemble of different pixel-based approaches to cloud pixel modeling. Based on four training subsets with a selection of different input features, 12 machine learning models were created. We evaluated these models using the cropped LC8-Biome cloud validation dataset. As a comparison, Fmask was also applied to the cropped scene Biome dataset. One goal of this research is to explore a machine learning modeling approach that uses as small a training data sample as possible but still provides an accurate model. Overall, the model trained on the sample subset (1.3% of the total training samples) that includes unsupervised Self-Organizing Map classification results as an input feature has the best performance. The approach achieves 98.57% overall accuracy, 1.18% cloud omission error, and 0.93% cloud commission error on the 88 cropped test images. By comparison to Fmask 4.0, this model improves the accuracy by 10.12% and reduces the cloud omission error by 6.39%. Furthermore, using an additional eight independent validation images that were not sampled in model training, the model trained on the second largest subset with an additional five features has the highest overall accuracy at 86.35%, with 12.48% cloud omission error and 7.96% cloud commission error. This model’s overall correctness increased by 3.26%, and the cloud omission error decreased by 1.28% compared to Fmask 4.0. The machine learning cloud classification models discussed in this paper could achieve very good performance utilizing only a small portion of the total training pixels available. We showed that a pixel-based cloud classification model, and that as each scene obviously has unique spectral characteristics, and having a small portion of example pixels from each of the sub-regions in a scene can improve the model accuracy significantly.

Improving the reliability of photometric redshift with machine learning

ABSTRACT In order to answer the open questions of modern cosmology and galaxy evolution theory, robust algorithms for calculating photometric redshifts (photo-z) for very large samples of galaxies are needed. Correct estimation of the various photo-z algorithms’ performance requires attention to both the performance metrics and the data used for the estimation. In this work, we use the supervised machine learning algorithm MLPQNA (Multi-Layer Perceptron with Quasi-Newton Algorithm) to calculate photometric redshifts for the galaxies in the COSMOS2015 catalogue and the unsupervised Self-Organizing Maps (SOM) to determine the reliability of the resulting estimates. We find that for zspec < 1.2, MLPQNA photo-z predictions are on the same level of quality as spectral energy distribution fitting photo-z. We show that the SOM successfully detects unreliable zspec that cause biases in the estimation of the photo-z algorithms’ performance. Additionally, we use SOM to select the objects with reliable photo-z predictions. Our cleaning procedures allow us to extract the subset of objects for which the quality of the final photo-z catalogues is improved by a factor of 2, compared to the overall statistics.

Sediment recognition by warp tension monitoring of bottom otter trawling and applying the self-organizing map algorithm

Model towing experiments of a bottom trawl net with hyper-lift trawl door were conducted to investigate the effect of the bottom sediment (concrete, sand, gravel, and rock) on the warp tension of the overall trawl system. The towing speed was from 50 cm/s to 70 cm/s and the ratio of warp length relative to the water depth was within the range of 4–6. Through the signal analysis of time-series warp tension, results reveal that there is a significant dependence of the warp tension on the type of bottom sediment, and the oscillation of warp tension in a frequency range of 1–10 Hz increases in the order of concrete, sand, gravel, and rock. Based on these characterizations, the time-series warp tension is thus represented by the feature vector for the input data of the self-organizing map (SOM) and learning vector quantization (LVQ) neural networks. A clustering method with an unsupervised SOM neural network acting as an updating tool for the bottom sediment database was successfully built using the validation of the prepared sediments. In combination with the output vector of labeled bottom sediment, the supervised LVQ neural network for sediment recognition performed excellently with a high classification accuracy of over 80%.

Optimizing COVID-19 vaccine distribution across the United States using deterministic and stochastic recurrent neural networks.

Optimizing COVID-19 vaccine distribution can help plan around the limited production and distribution of vaccination, particularly in early stages. One of the main criteria for equitable vaccine distribution is predicting the geographic distribution of active virus at the time of vaccination. This research developed sequence-learning models to predict the behavior of the COVID-19 pandemic across the US, based on previously reported information. For this objective, we used two time-series datasets of confirmed COVID-19 cases and COVID-19 effective reproduction numbers from January 22, 2020 to November 26, 2020 for all states in the US. The datasets have 310 time-steps (days) and 50 features (US states). To avoid training the models for all states, we categorized US states on the basis of their similarity to previously reported COVID-19 behavior. For this purpose, we used an unsupervised self-organizing map to categorize all states of the US into four groups on the basis of the similarity of their effective reproduction numbers. After selecting a leading state (the state with earliest outbreaks) in each group, we developed deterministic and stochastic Long Short Term Memory (LSTM) and Mixture Density Network (MDN) models. We trained the models with data from each leading state to make predictions, then compared the models with a baseline linear regression model. We also remove seasonality and trends from a dataset of non-stationary COVID-19 cases to determine the effects on prediction. We showed that the deterministic LSTM model trained on the COVID-19 effective reproduction numbers outperforms other prediction methods.

Self-Organizing Deep Learning (SO-UNet)—A Novel Framework to Classify Urban and Peri-Urban Forests

Forest-type classification is a very complex and difficult subject. The complexity increases with urban and peri-urban forests because of the variety of features that exist in remote sensing images. The success of forest management that includes forest preservation depends strongly on the accuracy of forest-type classification. Several classification methods are used to map urban and peri-urban forests and to identify healthy and non-healthy ones. Some of these methods have shown success in the classification of forests where others failed. The successful methods used specific remote sensing data technology, such as hyper-spectral and very high spatial resolution (VHR) images. However, both VHR and hyper-spectral sensors are very expensive, and hyper-spectral sensors are not widely available on satellite platforms, unlike multi-spectral sensors. Moreover, aerial images are limited in use, very expensive, and hard to arrange and manage. To solve the aforementioned problems, an advanced method, self-organizing–deep learning (SO-UNet), was created to classify forests in the urban and peri-urban environment using multi-spectral, multi-temporal, and medium spatial resolution Sentinel-2 images. SO-UNet is a combination of two different machine learning technologies: artificial neural network unsupervised self-organizing maps and deep learning UNet. Many experiments have been conducted, and the results showed that SO-UNet overwhelms UNet significantly. The experiments encompassed different settings for the parameters that control the algorithms.

Multiscale evaluation of a thin-bed reservoir

A thin-bed laminated shaly-sand reservoir of the Miocene formation was evaluated using two methods: high resolution microresistivity data from the XRMI tool and conventional well logs. Based on high resolution data, the Earth model of the reservoir was defined in a way that allowed the analyzed interval to be subdivided into thin layers of sandstones, mudstones, and claystones. Theoretical logs of gamma ray, bulk density, horizontal and vertical resistivity were calculated based on the forward modeling method to describe the petrophysical properties of individual beds and calculate the clay volume, porosity, and water saturation. The relationships amongst the contents of minerals were established based on the XRD data from the neighboring wells; hence, the high-resolution lithological model was evaluated. Predicted curves and estimated volumes of minerals were used as an input in multimineral solver and based on the assumed petrophysical model the input data were recalculated, reconstructed and compared with the predicted curves. The volumes of minerals and input curves were adjusted during several runs to minimalize the error between predicted and recalculated variables. Another approach was based on electrofacies modeling using unsupervised self-organizing maps. As an input, conventional well logs were used. Then, the evaluated facies model was used during forward modeling of the effective porosity, horizontal resistivity and water saturation. The obtained results were compared and, finally, the effective thickness of the reservoir was established based on the results from the two methods.

Identification of sarcomatoid differentiation in renal cell carcinoma by machine learning on multiparametric MRI

Sarcomatoid differentiation in RCC (sRCC) is associated with a poor prognosis, necessitating more aggressive management than RCC without sarcomatoid components (nsRCC). Since suspected renal cell carcinoma (RCC) tumors are not routinely biopsied for histologic evaluation, there is a clinical need for a non-invasive method to detect sarcomatoid differentiation pre-operatively. We utilized unsupervised self-organizing map (SOM) and supervised Learning Vector Quantizer (LVQ) machine learning to classify RCC tumors on T2-weighted, non-contrast T1-weighted fat-saturated, contrast-enhanced arterial-phase T1-weighted fat-saturated, and contrast-enhanced venous-phase T1-weighted fat-saturated MRI images. The SOM was trained on 8 nsRCC and 8 sRCC tumors, and used to compute Activation Maps for each training, validation (3 nsRCC and 3 sRCC), and test (5 nsRCC and 5 sRCC) tumor. The LVQ classifier was trained and optimized on Activation Maps from the 22 training and validation cohort tumors, and tested on Activation Maps of the 10 unseen test tumors. In this preliminary study, the SOM-LVQ model achieved a hold-out testing accuracy of 70% in the task of identifying sarcomatoid differentiation in RCC on standard multiparameter MRI (mpMRI) images. We have demonstrated a combined SOM-LVQ machine learning approach that is suitable for analysis of limited mpMRI datasets for the task of differential diagnosis.

Displacement Analysis of Geothermal Field Based on PSInSAR And SOM Clustering Algorithms A Case Study of Brady Field, Nevada—USA

The availability of free and high temporal resolution satellite data and advanced SAR techniques allows us to analyze ground displacement cost-effectively. Our aim was to properly define subsidence and uplift areas to delineate a geothermal field and perform time-series analysis to identify temporal trends. A Persistent Scatterer Interferometry (PSI) algorithm was used to estimate vertical displacement in the Brady geothermal field located in Nevada by analyzing 70 Sentinel-1A Synthetic-Aperture Radar (SAR) images, between January 2017 and December 2019. To classify zones affected by displacement, an unsupervised Self-Organizing Map (SOM) algorithm was applied to classify points based on their behavior in time, and those clusters were used to determine subsidence, uplift, and stable regions automatically. Finally, time-series analysis was applied to the clustered data to understand the inflection dates. The maximum subsidence is –19 mm/yr with an average value of –6 mm/yr within the geothermal field. The maximum uplift is 14 mm/yr with an average value of 4 mm/yr within the geothermal field. The uplift occurred on the NE of the field, where the injection wells are located. On the other hand, subsidence is concentrated on the SW of the field where the production wells are located. The coupling of the PSInSAR and the SOM algorithms was shown to be effective in analyzing the direction and pattern of the displacements observed in the field.

Comparison of Spatio-Spectral Properties of Zen-Meditation and Resting EEG Based on Unsupervised Learning

This paper reports distinct spatio-spectral properties of Zen-meditation EEG (electroencephalograph), compared with resting EEG, by implementing unsupervised machine learning scheme in clustering the brain mappings of centroid frequency (BMFc). Zen practitioners simultaneously concentrate on the third ventricle, hypothalamus and corpora quadrigemina touniversalize all brain neurons to construct a detached brain and gradually change the normal brain traits, leading to the process of brain-neuroplasticity. During such tri-aperture concentration, EEG exhibits prominent diffuse high-frequency oscillations. Unsupervised self-organizing map (SOM), clusters the dataset of quantitative EEG by matching the input feature vector Fc and the output cluster center through the SOM network weights. Input dataset contains brain mappings of 30 centroid frequencies extracted from CWT (continuous wavelet transform) coefficients. According to SOM clustering results, resting EEG is dominated by global low-frequency (14.4 Hz); whereas Zen-meditation EEG exhibits globally high-frequency (>16 Hz) activities throughout the entire record. Beta waves with a wide range of frequencies are often associated with active concentration. Nonetheless, clinic report discloses that benzodiazepines, medication treatment for anxiety, insomnia and panic attacks to relieve mind/body stress, often induce beta buzz. We may hypothesize that Zen-meditation practitioners attain the unique state of mindfulness concentration under optimal body-mind relaxation.

Analysis of Spatial Spread Relationships of Coronavirus (COVID-19) Pandemic in the World using Self Organizing Maps.

We describe in this paper an analysis of the spatial evolution of coronavirus pandemic around the world by using a particular type of unsupervised neural network, which is called self-organizing maps. Based on the clustering abilities of self-organizing maps we are able to spatially group together countries that are similar according to their coronavirus cases, in this way being able to analyze which countries are behaving similarly and thus can benefit by using similar strategies in dealing with the spread of the virus. Publicly available datasets of coronavirus cases around the globe from the last months have been used in the analysis. Interesting conclusions have been obtained, that could be helpful in deciding the best strategies in dealing with this virus. Most of the previous papers dealing with data of the Coronavirus have viewed the problem on temporal aspect, which is also important, but this is mainly concerned with the forecast of the numeric information. However, we believe that the spatial aspect is also important, so in this view the main contribution of this paper is the use of unsupervised self-organizing maps for grouping together similar countries in their fight against the Coronavirus pandemic, and thus proposing that strategies for similar countries could be established accordingly.

An Unsupervised Data-Driven Model to Classify Gait Patterns in Children with Cerebral Palsy.

Ankle and foot orthoses are commonly prescribed to children with cerebral palsy (CP). It is unclear whether 3D gait analysis (3DGA) provides sufficient and reliable information for clinicians to be consistent when prescribing orthoses. Data-driven modeling can probe such questions by revealing non-intuitive relationships between variables such as 3DGA parameters and gait outcomes of orthoses use. The purpose of this study was to (1) develop a data-driven model to classify children with CP according to their gait biomechanics and (2) identify relationships between orthotics types and gait patterns. 3DGA data were acquired from walking trials of 25 typically developed children and 98 children with CP with additional prescribed orthoses. An unsupervised self-organizing map followed by k-means clustering was developed to group different gait patterns based on children’s 3DGA. Model inputs were gait variable scores (GVSs) extracted from the gait profile score, measuring root mean square differences from TD children’s gait cycle. The model identified five pathological gait patterns with statistical differences in GVSs. Only 43% of children improved their gait pattern when wearing an orthosis. Orthotics prescriptions were variable even in children with similar gait patterns. This study suggests that quantitative data-driven approaches may provide more clarity and specificity to support orthotics prescription.