Unsupervised Learning Method Research Articles

IoT Enabled Sensory Monitoring System for Fog Optimal Resource Provisioning Method in Health Monitoring System

Fog is data management and analytics service. In this paper gains and most effective novel approach to provide IoT enabled services in healthcare application using Fog Computing. In this research the data is collected from Google Scholar, Science Director and MEDLINE database. IoT based Fog Computing techniques are proposed for delivering quality of services to the user. Optimal Resource Provisioning method is proposed to find edges, service level agreements and administration services for IoT client. The DeepQ residue information processing technique is applied for connecting data centre of the cloud and computing paradigms technique is finding the depth reference of Fog levels. The proposed Optimal resource provisioning algorithm is examining the dataset and TensorFlow tool is used for simulating environment. Fog computing layer consist of IoT sensor data inputs, data centres for the cloud and connected layers for simulations. The Deep belief network is generated based on above inputs using 256 X 256 X 3 layer system and 5000 trained data, 1000 test data are taken for simulations. Each dataset simulation is recording using supervised and unsupervised learning methods. Based on above results IoT enable Fog Computing data management and analytics systems provided 95% accuracy and the compared with existing computing techniques our proposed systems shows better efficiency with respect to safety and convenience.

HFrEF subphenotypes based on 4210 repeatedly measured circulating proteins are driven by different biological mechanisms.

HFrEF is a heterogenous condition with high mortality. We used serial assessments of 4210 circulating proteins to identify distinct novel protein-based HFrEF subphenotypes and to investigate underlying dynamic biological mechanisms. Herewith we aimed to gain pathophysiological insights and fuel opportunities for personalised treatment. In 382 patients, we performed trimonthly blood sampling during a median follow-up of 2.1 [IQR:1.1-2.6] years. We selected all baseline samples and two samples closest to the primary endpoint (PEP; composite of cardiovascular mortality, HF hospitalization, LVAD implantation, and heart transplantation) or censoring, and applied an aptamer-based multiplex proteomic approach. Using unsupervised machine learning methods, we derived clusters from 4210 repeatedly measured proteomic biomarkers. Sets of proteins that drove cluster allocation were analysed via an enrichment analysis. Differences in clinical characteristics and PEP occurrence were evaluated. We identified four subphenotypes with different protein profiles, prognosis and clinical characteristics, including age (median [IQR] for subphenotypes 1-4, respectively:70 [64, 76], 68 [60, 79], 57 [47, 65], 59 [56, 66]years), EF (30 [26, 36], 26 [20, 38], 26 [22, 32], 33 [28, 37]%), and chronic renal failure (45%, 65%, 36%, 37%). Subphenotype allocation was driven by subsets of proteins associated with various biological functions, such as oxidative stress, inflammation and extracellular matrix organisation. Clinical characteristics of the subphenotypes were aligned with these associations. Subphenotypes 2 and 3 had the worst prognosis compared to subphenotype 1 (adjHR (95%CI):3.43 (1.76-6.69), and 2.88 (1.37-6.03), respectively). Four circulating-protein based subphenotypes are present in HFrEF, which are driven by varying combinations of protein subsets, and have different clinical characteristics and prognosis. ClinicalTrials.gov Identifier: NCT01851538https://clinicaltrials.gov/ct2/show/NCT01851538. EU/EFPIA IMI2JU BigData@Heart grant n°116074, Jaap Schouten Foundation and Noordwest Academie.

An unsupervised deep learning model for dense velocity field reconstruction in particle image velocimetry (PIV) measurements

Supervised deep learning methods reported recently have shown promising capability and efficiency in particle image velocimetry (PIV) processes compared to the traditional cross correlation and optical flow methods. However, the deep learning-based methods in previous reports require synthesized particle images and simulated flows for training prior to applications, conflicting with experimental scenarios. To address this crucial limitation, unsupervised deep learning methods have also been proposed for flow velocity reconstruction, but they are generally limited to rough flow reconstructions with low accuracy in velocity due to, for example, particle occlusion and out-of-boundary motions. This paper proposes a new unsupervised deep learning model named UnPWCNet-PIV (an unsupervised optical flow network using Pyramid, Warping, and Cost Volume). Such a pyramidical network with specific enhancements on flow reconstructions holds capabilities to manage particle occlusion and boundary motions. The new model showed comparable accuracy and robustness with the advanced supervised deep learning methods, which are based on synthesized images, together with superior performance on experimental images. This paper presents the details of the UnPWCNet-PIV architecture and the assessments of its accuracy and robustness on both synthesized and experimental images.

GraphPMU: Event Clustering via Graph Representation Learning Using Locationally-Scarce Distribution-Level Fundamental and Harmonic PMU Measurements

This paper is concerned with the complex task of identifying the type and cause of the events that are captured by distribution-level phasor measurement units (D-PMUs) in order to enhance situational awareness in power distribution systems. Our goal is to address two fundamental challenges in this field: a) scarcity in measurement locations due to the high cost of purchasing, installing, and streaming data from D-PMUs; b) limited prior knowledge about the event signatures due to the fact that the events are diverse and infrequent, and have unknown characteristics. To tackle these challenges, we propose an unsupervised graph-representation learning method, called GraphPMU, to significantly improve the performance in event clustering under locationally-scarce data availability by proposing the following two new directions: 1) using the topological information about the relative location of the few available phasor measurement units on the graph of the power distribution network; 2) utilizing not only the commonly used fundamental phasor measurements, bus also the less explored harmonic phasor measurements in the process of analyzing the signatures of various events. Through a detailed analysis of several case studies, we show that GraphPMU can highly outperform the prevalent methods in the literature.

Identifying low-performing regions in Moroccan education: A deep learning approach using the PISA dataset

This study highlights the ongoing nature of the school reform movement, emphasizing the need for continuous attention and action. Despite this effort, academic performance has exhibited relative stability in recent years, while significant regional performance disparities persist. Addressing these inequalities requires novel approaches to enhance educational quality. Past research has explored clustering algorithms in developed countries, providing insights into personalized teaching strategies based on students' learning style preferences. In response, our research aims to identify underperforming regions in Morocco, necessitating attention and intervention. We employ an unsupervised deep learning method called "deep embedding clustering" to group Moroccan students based on their performance. The results are subsequently visualized on a choropleth map, revealing intricate patterns and trends in educational performance that might not be immediately apparent. The analysis employs the comprehensive program for international student assessment (PISA) dataset, encompassing individual students' responses and plausible values reflecting cognitive abilities. The findings indicate that the "Guelmim-Oued Noun" region exhibits the highest performance level among all regions, while "Dakhla-Oued Eddahab," "Béni Mellal-Khénifra," and "Oriental" regions display lower performance levels. As a result, this study urges policymakers to incorporate tailored measures into regional policies to improve students' educational outcomes.

Exploratory analysis of a real wind turbine dataset, using AI tools to cluster and classify data, for condition monitoring and fault detection

In recent years there has been an increase in wind generation, driven by environmental factors and the incentive offered for the development of clean and sustainable technologies for energy generation. However, due to the rapid growth of this technology, concerns about the safety and reliability of wind turbines are increasing, especially due to the associated risks and financial costs. Therefore, health monitoring and fault detection for wind turbines has become an important research focus. Thus, the objective of this work was to realize an exploratory study of real data from a wind turbine, using AI tools that help to group the different behaviors, according to the similarity of resources and characteristics of the data. For this, unsupervised learning methods were used to cluster the data and a model was proposed to train and test, using a multilayer perceptron network, to classify these clusters. The differential of this work is the use of real data from CHESF's wind turbines. Another important contribution is in relation to permanent magnet wind turbines, as there are not many studies in this field, therefore a great potential to be explored.

Robust detection of unknown DoS/DDoS attacks in IoT networks using a hybrid learning model

The fourth industrial revolution is marked by the rapid growth of Internet of Things (IoT) technology, leading to an increase in the number of IoT devices. Unfortunately, this also makes these devices more susceptible to cyber threats, especially DoS/DDoS attacks. While supervised learning models have been adopted to detect and mitigate these threats, they have limitations in detecting unknown attacks that can cause severe consequences. This research aims to address those limitations and provide better protection for IoT networks against DoS/DDoS attacks. We propose a new approach that combines a soft-ordering convolutional neural network (SOCNN) model with local outlier factor (LOF) and isolation-based anomaly detection using nearest-neighbor ensembles (iNNE) models that use both supervised and unsupervised learning methods. We evaluated our approach on three benchmark datasets with varying unknown attack scenarios, and our hybrid model achieved high accuracy in detecting unknown attacks with an average F1-score of 98.94%, 91.68%, and 96.07%, respectively, on BoT-IoT, CIC-IDS-2017, and CIC-IDS-2018 datasets, outperforming state-of-the-art competitors. Our model also showed resilience against adversarial attacks such as the fast gradient sign method (FGSM) and Carlini Wagner (CW) adversarial attacks, highlighting the potential of our approach to enhance IoT network security against DoS/DDoS attacks in unknown attack scenarios.

Leveraging and Refining Image Recognition Technology for Intelligent Logistics Sorting Systems

Obstacles in the realm of target detection, tracking accuracy, real-time performance, and robustness have been identified within the context of intelligent logistics sorting systems, which incorporate image recognition technology. Furthermore, the critical role of detecting anomalous states in augmenting sorting efficiency and curtailing errors is underscored. Image recognition technologies of a conventional nature tend to suffer from limitations in their applicability and robustness. In practical working environments, a paucity of abnormal data from logistics sorting targets is observed, which inhibits the application of supervised learning methods of deep learning. Addressing these challenges, an unsupervised deep learning method is introduced for the detection of anomalous states in logistics sorting targets. This approach reinterprets the detection of logistics sorting targets as an anomaly detection problem and utilizes Variational AutoEncoders (VAE) for modeling the distribution of normal data. This method's dependency rests exclusively on normal data for training, thereby circumventing the need for a substantial quantity of abnormal samples. In practical deployments, the anomalous state of logistics sorting targets is discerned by the method through the computation of similarity and implementation of labeling algorithms, evidencing robustness, generalizability, and adaptability. Overall, this method is presented as an effective solution for the detection of anomalous states within intelligent logistics sorting scenarios, serving to decrease labeling costs, enhance detection accuracy and efficiency, and satisfying the practical requisites of logistics sorting systems for abnormal state detection.

Relationship between learner profiles and learner–content interaction in online learning: Exploring implications for learning experience design

The current study explored the relationship between learner profiles and the nature of their interaction with content in massive, open, and online learning environments. The research was conducted on the Anadolu University Open Education System, and data from 597,164 learners enrolled in 86 different degree programs were analyzed by unsupervised machine learning methods. Cluster analysis was used to identify learner profile groups and association rules were applied to identify learner-content interaction patterns. As a result of the analyses, five clusters were obtained, and it was determined that the attribute with the highest discrimination in determining the clusters was the learners’ semester grade point average. The clusters were named according to learner-content interactions and the learners’ semester grade point average. Analysis of the association rules revealed that various learner-content interactions emerged in the context of profile groups.

Fruit type classification using deep learning and feature fusion

Machine and deep learning applications play a dominant role in the current scenario in the agriculture sector. To date, the classification of fruits using image features has attained the researcher’s attraction very much from the last few years. Fruit recognition and classification is an ill-posed problem due to the heterogeneous nature of fruits. In the proposed work, Convolution neural network (CNN), Recurrent Neural Network (RNN), and Long-short Term Memory (LSTM) deep learning methods are used to extract the optimal image features, and to select features after extraction, and finally, use extracted image features to classify the fruits. To evaluate the performance of the proposed approach, the Support vector machine (SVM) unsupervised learning method, Artificial neuro-fuzzy inference system (ANFIS), and Feed-forward neural network (FFNN) classification results are compared, and observed that the proposed fruit classification approach results are quite efficient and promising.

I’m Me, We’re Us, and I’m Us: Tri-directional Contrastive Learning on Hypergraphs

Although machine learning on hypergraphs has attracted considerable attention, most of the works have focused on (semi-)supervised learning, which may cause heavy labeling costs and poor generalization. Recently, contrastive learning has emerged as a successful unsupervised representation learning method. Despite the prosperous development of contrastive learning in other domains, contrastive learning on hypergraphs remains little explored. In this paper, we propose TriCL (Tri-directional Contrastive Learning), a general framework for contrastive learning on hypergraphs. Its main idea is tri-directional contrast, and specifically, it aims to maximize in two augmented views the agreement (a) between the same node, (b) between the same group of nodes, and (c) between each group and its members. Together with simple but surprisingly effective data augmentation and negative sampling schemes, these three forms of contrast enable TriCL to capture both node- and group-level structural information in node embeddings. Our extensive experiments using 14 baseline approaches, 10 datasets, and two tasks demonstrate the effectiveness of TriCL, and most noticeably, TriCL almost consistently outperforms not just unsupervised competitors but also (semi-)supervised competitors mostly by significant margins for node classification. The code and datasets are available at https://github.com/wooner49/TriCL.

Unsupervised Deep Video Denoising with Untrained Network

Deep learning has become a prominent tool for video denoising. However, most existing deep video denoising methods require supervised training using noise-free videos. Collecting noise-free videos can be costly and challenging in many applications. Therefore, this paper aims to develop an unsupervised deep learning method for video denoising that only uses a single test noisy video for training. To achieve this, an unsupervised loss function is presented that provides an unbiased estimator of its supervised counterpart defined on noise-free video. Additionally, a temporal attention mechanism is proposed to exploit redundancy among frames. The experiments on video denoising demonstrate that the proposed unsupervised method outperforms existing unsupervised methods and remains competitive against recent supervised deep learning methods.

Beyond Smoothing: Unsupervised Graph Representation Learning with Edge Heterophily Discriminating

Unsupervised graph representation learning (UGRL) has drawn increasing research attention and achieved promising results in several graph analytic tasks. Relying on the homophily assumption, existing UGRL methods tend to smooth the learned node representations along all edges, ignoring the existence of heterophilic edges that connect nodes with distinct attributes. As a result, current methods are hard to generalize to heterophilic graphs where dissimilar nodes are widely connected, and also vulnerable to adversarial attacks. To address this issue, we propose a novel unsupervised Graph Representation learning method with Edge hEterophily discriminaTing (GREET) which learns representations by discriminating and leveraging homophilic edges and heterophilic edges. To distinguish two types of edges, we build an edge discriminator that infers edge homophily/heterophily from feature and structure information. We train the edge discriminator in an unsupervised way through minimizing the crafted pivot-anchored ranking loss, with randomly sampled node pairs acting as pivots. Node representations are learned through contrasting the dual-channel encodings obtained from the discriminated homophilic and heterophilic edges. With an effective interplaying scheme, edge discriminating and representation learning can mutually boost each other during the training phase. We conducted extensive experiments on 14 benchmark datasets and multiple learning scenarios to demonstrate the superiority of GREET.

Supervised and unsupervised machine learning approaches using Sentinel data for flood mapping and damage assessment in Mozambique

Natural hazards, such as flooding, have been negatively impacting developed and emerging economies alike. The effects of floods are more prominent in countries of the Global South, where large parts of the population and infrastructure are insufficiently protected from natural hazards. From this scope, a lot of effort is required to mitigate these impacts by continuously providing new and more reliable tools to aid in mitigation and preparedness, during or after a flood event. Flood mapping followed by damage assessment plays an important role in all these stages. In this work we investigate a new dataset provided by DrivenData Labs based on Sentinel-1 (S1) imagery (VH, VV imagery and labels) to help map floods in the city of Beira in Mozambique. Exploiting Google Earth Engine (GEE), we deployed supervised and unsupervised machine learning (ML) methods on a dataset comprising imagery from 13 countries worldwide. We first mapped the floods country-by-country including Mozambique. This first part was helpful to understand the sensitivity of each method when applied to data from different regions and with different polarizations. We then trained the supervised model globally (in all 13 countries) and used it to predict floods in Beira. To assess the accuracy of the experiments we used the intersection over the union (IoU) metric, results of which we compared with the benchmark IoU achieved by the winner in the DrivenData competition for flood mapping in 2021. The implementation of unsupervised and supervised ML using VH and VV+VH produced satisfactory results, and showed to be better than using VV imagery; in Cambodia and Bolivia with VH polarization yielded IoUs values ranging from 0.819 to 0.856 which is above the benchmark (0.8094). The predictions in Beira using VH imagery yielded IoU of 0.568, which is a reasonable outcome. The proposed approach is a reliable alternative for flood mapping, especially in Mozambique due to its low cost and time effectiveness as even with unsupervised approaches, relatively high-quality results are yielded in near real-time. Finally, we used Sentinel-2 (S2) imagery for a land cover classification to perform damage assessment in Beira and integrated population data from Beira to enhance the quality the results. The results show that 20% of agricultural area and about 10% of built up area were flooded. Flooded built up area includes highly populated neighborhoods such as Chaimite and Ponta Gea that are located in the center of the city.

Snake net with a neural network for detecting multiple phases in the phase diagram.

Unsupervised machine learning applied to the study of phase transitions is an ongoing and interesting research direction. The active contour model, also called the snake model, was initially proposed for target contour extraction in two-dimensional images. In order to obtain a physical phase diagram, the snake model with an artificial neural network is applied in an unsupervised learning way by the authors of [Phys. Rev. Lett. 120, 176401 (2018)0031-900710.1103/PhysRevLett.120.176401]. It guesses the phase boundary as an initial snake and then drives the snake to convergence with forces estimated by the artificial neural network. In this work we extend this unsupervised learning method with one contour to a snake net with multiple contours for the purpose of obtaining several phase boundaries in a phase diagram. For the classical Blume-Capel model, the phase diagram containing three and four phases is obtained. Moreover, a balloon force is introduced, which helps the snake to leave a wrong initial position and thus may allow for greater freedom in the initialization of the snake. Our method is helpful in determining the phase diagram with multiple phases using just snapshots of configurations from cold atoms or other experiments without knowledge of the phases.

An unsupervised learning-based generalization of Data Envelopment Analysis

In this paper, we introduce an unsupervised machine learning method for production frontier estimation. This new approach satisfies fundamental properties of microeconomics, such as convexity and free disposability (shape constraints). The new method generalizes Data Envelopment Analysis (DEA) through the adaptation of One-Class Support Vector Machines with piecewise linear transformation mapping. The new technique aims to reduce the overfitting problem occurring in DEA. How to measure technical inefficiency through the directional distance function is also introduced. Finally, we evaluate the performance of the new technique via a computational experience, showing that the mean squared error in the estimation of the frontier is up to 83% better than the standard DEA in certain scenarios.

Anti-money laundering supervision by intelligent algorithm

As global anti-money laundering (AML) protocols and technologies strengthen, traditional laundering methods increasingly face stringent scrutiny. Concurrently, the allure of virtual currencies for laundering illegal funds is magnified, owing to their immunity from seizure, elusive traceability, and seamless cross-border transactions. This necessitates the design of specialized AML strategies for virtual currencies to counter this burgeoning avenue of illicit money circulation. Our study introduces a groundbreaking two-tier algorithm, marrying heuristic rules and integrated learning modalities, designed to enable risk-based detection of virtual currency money laundering. By employing data science methodologies, we scrutinize the distribution of statistical risk attributes within existing transaction samples to institute heuristic rules. By training an amalgamated model of Long Short-Term Memory (LSTM) and Graph Convolutional Neural Network (GCN), we can effectively discern conspicuous anomalous money laundering patterns across diverse transaction datasets. The integrated learning segment deploys a hard voting mechanism, bolstering the recognition of laundering techniques by synergizing multiple anomaly detection classifiers such as Histogram-Based Outlier Scoring (HBOS) and Isolation Forest. This methodology facilitates the detection of laundering activities in unlabeled data, generating explicit anomalous transaction data and concurrently capturing implicit abnormal transactions undetected by heuristic rules. Our approach alleviates the difficulty of procuring labeled money laundering data for AML identification, simultaneously improving the generalization and accuracy of unsupervised learning methods for money laundering detection.

OTHR-07. CIRCULATING MIRNAS AS A NON-INVASIVE MONITORING TOOL FOR PEDIATRIC BRAIN TUMORS

Abstract Current management of pediatric brain tumor patients relies on imaging as the only non-invasive tool available for monitoring therapeutic response and tumor progression. Results are often inconclusive and unable to capture biological changes that presage progression on imaging. Circulating miRNA provides an attractive non-invasive platform because miRNA’s essential role in regulating gene expression, stability in bio-fluids, selective over-expression in tumors, and subsequent active release from tumors into the extracellular environment. We aimed to establish the prognostic and monitoring value of miRNA biomarkers for pediatric brain tumors. We performed miRNA profiling of CSF (n=33) and plasma (n=53) samples across a cohort of 54 pediatric brain tumors from 6 histologies (low grade glioma, ependymoma, germ cell tumor, medulloblastoma, atypical teratoid rhabdoid tumor and high-grade glioma). We utilized unsupervised consensus clustering, differential expression, and machine learning methods to explore miRNA biomarkers. Plasma miRNA signatures were correlated with WHO Grade while CSF signatures revealed a correlation with leptomeningeal disease and higher disease burden (p<0.05). The expression of circulating miRNA revealed histology-specific signatures for low grade glioma, medulloblastoma and germinoma in plasma and CSF (p<0.05). We identified miRNA signatures of significant translational value: (1) A medulloblastoma specific miRNA in CSF can serve as a potential biomarker for tumor monitoring (medulloblastoma specificity was confirmed in independent tissue-based dataset); (2) Pilocytic astrocytoma can be distinguished from high grade glioma in CSF; (3) Most intriguingly, in plasma, miRNA expression identified pilocytic astrocytoma that had relapsed in several months after initial intervention (p<0.05). These results support circulating miRNA in plasma and CSF as biomarkers of diverse clinical diagnostic and prognostic measures. Further utilization of this approach provides unique platforms to inform liquid biopsy-based management in the clinical setting.

Unsupervised Spectral Analysis of Bio-Dyed Textile Samples

Article Unsupervised Spectral Analysis of Bio-Dyed Textile Samples Zong-Yue Li 1,*, Joni Hyttinen 1, Riikka Räisänen 2, Xiao-Zhi Gao 1, and Markku Hauta-Kasari 1 1 School of Computing, Faculty of Science, Forestry and Technology, University of Eastern Finland, Joensuu, 80101, Kuopio, 70211, Finland 2 Department of Education, Faculty of Educational Sciences, University of Helsinki, Helsinki, 00014 , Finland * Correspondence: winstonli711@gmail.com Received: 12 March 2023 Accepted: 23 March 2023 Abstract: Natural compounds such as biological colorants (biocolorants) have long been employed as crucial ingredients for dying textile in the textile industry. As one part of the BioColour Consortium project, our goal is to take advantage of machine learning (in cluster analysis) to discover possible clusters of bio-dyed textile in the absence of ground truth labels or other knowledge of expert domains. Specifically, we use unsupervised learning methods of agglomerative clustering, fuzzy c-means, ordering points to identify the clustering structure (OPTICS) and self-organizing maps (SOMs), resulting in an investigation that combines data visualization and cluster analysis. In summary, we apply some selected data mining methods to 1) discover hidden clusters emerging among products that are colored with biocolorant (specifically bio-dyed textile samples), and 2) show the potentials of clustering techniques in the case study.

Mutated Aquila Optimizer for assisting brain tumor segmentation

Image processing is a crucial effort to discover suspicious regions and robust features from magnetic resonance imaging (MRI). Many segmentation algorithms have been proposed to increase the accuracy of brain tumor detection. However, brain image segmentation is considered a complex and challenging step in medical image processing. K-means is an unsupervised learning method for MRI image clustering and other tasks such as image segmentation and pattern identification of interest. However, the K-means algorithm still has many shortcomings, including a low convergence rate, a tendency to get stuck in local minima, and sensitivity to initialization. An improved version of Aquila Optimizer (AO) boosted with an alternative selection scheme and Cauchy mutation, called Mutated Aquila Optimizer (MAO), and the K-Means algorithm were combined to create a new dynamic and intelligent clustering method for brain tumor segmentation. The proposed MAO-Kmeans approach aims to increase the ability of K-Means to automatically extract the correct number and location of cluster centers and the number of pixels in each cluster in abnormal MRI images (multiple sclerosis lesions). The experimental results on the Brain Tumor Segmentation dataset (BraTS 2020) demonstrated the effectiveness of the MAO-Kmeans method in improving the performance of conventional K-means clustering. Furthermore, qualitative and quantitative comparisons between MAO-Kmeans and the other 12 state-of-the-art segmentation techniques (which include machine learning, artificial neural network, and deep learning-based methods) demonstrate the superiority of our proposed algorithm.