Pattern Recognition Learning Research Articles

Inhibition SNN: unveiling the efficacy of various lateral inhibition learning in image pattern recognition

Inhibition SNN: unveiling the efficacy of various lateral inhibition learning in image pattern recognition

Using Explainable Artificial Intelligence to Interpret Remaining Useful Life Estimation with Gated Recurrent Unit

In engineering, prognostics can be defined as the estimation of the remaining useful life of a system given current and past health conditions. This field has drawn attention from research, industry, and government as this kind of technology can help improve efficiency and lower the costs of maintenance in a variety of technical applications. An approach to prognostics that has gained increasing attention is the use of data-driven methods. These methods typically use pattern recognition and machine learning to estimate the residual life of equipment based on historical data. Despite their promising results, a major disadvantage is that it is difficult to interpret this kind of methodologies, that is, to understand why a certain prediction of remaining useful life was made at a certain point in time. Nevertheless, the interpretability of these models could facilitate the use of data-driven prognostics in different domains such as aeronautics, manufacturing, and energy, areas where certification is critical. To help address this issue, we use Local Interpretable Model-agnostic Explanations (LIME) from the field of eXplainable Artificial Intelligence (XAI) to analyze the prognostics of a Gated Recurrent Unit (GRU) on the C-MAPSS data. We select the GRU as this is a deep learning model that a) has an explicit temporal dimension and b) has shown promising results in the field of prognostics and c) is of simplified nature compared to other recurrent networks. Our results suggest that it is possible to infer the feature importance for the GRU both globally (for the entire model) and locally (for a given RUL prediction) with LIME.

A histogram-based approach to calculate graph similarity using graph neural networks

Deep learning has revolutionized the field of pattern recognition and machine learning by exhibiting exceptional efficiency in recognizing patterns. The success of deep learning can be seen in a wide range of applications including speech recognition, natural language processing, video processing, and image classification. Moreover, it has also been successful in recognizing structural patterns, such as graphs. Graph Neural Networks (GNNs) are models that employ message passing between nodes in a graph to capture its dependencies. These networks memorize a state that approximates graph information with greater depth compared to traditional neural networks. Although training a GNN can be challenging, recent advances in GNN variants, including Graph Convolutional Neural Networks, Gated Graph Neural Networks, and Graph Attention Networks, have shown promising results in solving various problems. In this work, we present a GNN-based approach for computing graph similarity and demonstrate its application to a classification problem. Our proposed method converts the similarity of two graphs into a score, and experiments on state-of-the-art datasets show that the proposed technique is effective and efficient. Results are summarized using a confusion matrix and mean square error metric, demonstrating the accuracy of our proposed technique.

On the decision-making framework for policing: A proposal for improving police decision-making

This article introduces the Decision-Making Framework for Policing (DMFP), a comprehensive tool designed to enhance the decision-making understanding of police officers. The DMFP considers the principles of heuristic, naturalistic, and rational decision-making along a fluid cognitive continuum to create a framework that addresses the limitations of the existing police National Decision Model (NDM). It achieves this by including 10 proposed typologies of police decision-making including: Routine, Tactical, Operational, Crisis, Investigative, Ethical, Interpersonal, Administrative, Managerial, and Strategic. These are integrated alongside existing and adapted decision-making models which are presented using a mnemonic letter strategy. Although the DMFP is theoretical, and its utility is presently untested in comparison to the existing NDM, it is presented to provide a tool to help improve officers' tacit knowledge, pattern recognition, and experiential learning through provision of easily recallable mnemonic decision-models. Thereby fostering a deeper understanding of cognitive processes and the factors influencing police decisions, potentially increasing consistency in reasoning, reducing decision errors, and enhancing policing outcomes.

Design of a differentiable L-1 norm for pattern recognition and machine learning

In various applications of pattern recognition, feature selection, and machine learning, L-1 norm is used as either an objective function or a regularizer. Mathematically, L-1 norm has unique characteristics that make it attractive in machine learning, feature selection, optimization, and regression. Computationally, however, L-1 norm presents a hurdle as it is non-differentiable, making the process of finding a solution difficult. Existing approach therefore relies on numerical approaches. In this work we designed an L-1 norm that is differentiable and, thus, has an analytical solution. The differentiable L-1 norm removes the absolute sign in the conventional definition and is everywhere differentiable. The new L-1 norm is almost everywhere linear, a desirable feature that is also present in the conventional L-1 norm. The only limitation of the new L-1 norm is that near zero, its behavior is not linear, hence we consider the new L-1 norm quasi-linear. Being differentiable, the new L-1 norm and its quasi-linear variation make them amenable to analytic solutions. Hence, it can facilitate the development and implementation of many algorithms involving L-1 norm. Our tests validate the capability of the new L-1 norm in various applications.

Research on flexible antenna and distributed deep learning pattern recognition for partial discharge monitoring of transformer

Research on flexible antenna and distributed deep learning pattern recognition for partial discharge monitoring of transformer

Survey of source code vulnerability analysis based on deep learning

Amidst the rapid development of the software industry and the burgeoning open-source culture, vulnerability detection within the software security domain has emerged as an ever-expanding area of focus. In recent years, the rapid advancement of artificial intelligence, particularly the notable progress in deep learning for pattern recognition and natural language processing, has catalyzed a surge in research endeavors exploring the integration of deep learning for the enhancement of vulnerability detection techniques. In this paper, we investigate contemporary deep learning-based source code analysis methods, with a concentrated emphasis on those pertaining to static code vulnerability detection. We categorize these methods based on various representations of source code employed during the preprocessing stage, including token-based and graph-based representations of source code, and further subdivided based on the types of deep learning algorithms or graph representations employed. We summarize the basic processes of model training and vulnerability detection under these different representation formats. Furthermore, we explore the limitations inherent in current approaches and provide insights into future trends and challenges for research in this field.

Sangerbox 2: Enhanced functionalities and update for a comprehensive clinical bioinformatics data analysis platform.

In recent years, development in high-throughput sequencing technologies has experienced an increasing application of statistics, pattern recognition, and machine learning in bioinformatics analyses. SangeBox platform to meet different scientific demands. The new version of Sangs is a widely used tool among many researchers, which encourages us to continuously improve the plerBox 2 (http://vip.sangerbox.com) and extends and optimizes the functions of interactive graphics and analysis of clinical bioinformatics data. We introduced novel analytical tools such as random forests and support vector machines, as well as corresponding plotting functions. At the same time, we also optimized the performance of the platform and fixed known problems to allow users to perform data analyses more quickly and efficiently. SangerBox 2 improved the speed of analysis, reduced resource required for computer performance, and provided more analysis methods, greatly promoting the research efficiency.

Integration of deep neutral network modeling and LC-MS-based pseudo-targeted metabolomics to discriminate easily confused ginseng species

Metabolomics covers a wide range of applications in life sciences, biomedicine, and phytology. Data acquisition (to achieve high coverage and efficiency) and analysis (to pursue good classification) are two key segments involved in metabolomics workflows. Various chemometric approaches utilizing either pattern recognition or machine learning have been employed to separate different groups. However, insufficient feature extraction, inappropriate feature selection, overfitting, or underfitting lead to an insufficient capacity to discriminate plants that are often easily confused. Using two ginseng varieties, namely Panax japonicus and P. japonicus var. major, containing the similar ginsenosides, we integrated pseudo-targeted metabolomics and deep neural network (DNN) modeling to achieve accurate species differentiation. A pseudo-targeted metabolomics approach was optimized through data acquisition mode, ion pairs generation, comparison between multiple reaction monitoring (MRM) and scheduled MRM, and chromatographic elution gradient. In total, 1980 ion pairs were monitored within 23 min, allowing for the most comprehensive ginseng metabolome analysis. The established DNN model demonstrated excellent classification performance (in terms of accuracy, precision, recall, F1 score, area under the curve, and receiver operating characteristic) using the entire metabolome data and feature-selection dataset, exhibiting superior advantages over random forest (RF), support vector machine (SVM), extreme gradient boosting (XGBoost), and multilayer perceptron (MLP). Moreover, DNNs were advantageous for automated feature learning, nonlinear modeling, adaptability, and generalization. This study confirmed practicality of the established strategy for efficient metabolomics data analysis and reliable classification performance even when using small-volume samples. This established approach holds promise for plant metabolomics and is not limited to ginseng.

CTRNet: An Automatic Modulation Recognition Based on Transformer-CNN Neural Network

Deep learning (DL) has brought new perspectives and methods to automatic modulation recognition (AMR), enabling AMR systems to operate more efficiently and reliably in modern wireless communication environments through its powerful feature learning and complex pattern recognition capabilities. However, convolutional neural networks (CNNs) and recurrent neural networks (RNNs), which are used for sequence recognition tasks, face two main challenges, respectively: the ineffective utilization of global information and slow processing speeds due to sequential operations. To address these issues, this paper introduces CTRNet, a novel automatic modulation recognition network that combines a CNN with Transformer. This combination leverages Transformer’s ability to adequately capture the long-distance dependencies between global sequences and its advantages in sequence modeling, along with the CNN’s capability to extract features from local feature regions of signals. During the data preprocessing stage, the original IQ-modulated signals undergo sliding-window processing. By selecting the appropriate window sizes and strides, multiple subsequences are formed, enabling the network to effectively handle complex modulation patterns. In the embedding module, token vectors are designed to integrate information from multiple samples within each window, enhancing the model’s understanding and modeling ability of global information. In the feedforward neural network, a more effective Bilinear layer is employed for processing to capture the higher-order relationship between input features, thereby enhancing the ability of the model to capture complex patterns. Experiments conducted on the RML2016.10A public dataset demonstrate that compared with the existing algorithms, the proposed algorithm not only exhibits significant advantages in terms of parameter efficiency but also achieves higher recognition accuracy under various signal-to-noise ratio (SNR) conditions. In particular, it performs relatively well in terms of accuracy, precision, recall, and F1-score, with clearer classification of higher-order modulations and notable overall accuracy improvement.

Feasibility of Using Wavelet Analysis and Machine Learning Method in Technical Diagnosis of Car Seats

This paper presents the results of preliminary research aimed at developing a method for rapid, noncontact diagnostics of the electric drive of car seats. The method is based on the analysis of acoustic signals produced during the operation of the drive. Pattern recognition and machine learning processes were used in the diagnosis. A method of feature extraction (diagnostic symptoms) using wavelet decomposition of acoustic signals was developed. The discriminative properties of a set of diagnostic symptoms were tested using the “Classification Learner” application available in MATLAB. The obtained results confirmed the usefulness of the developed method for the technical diagnostics of car seats.

Artificial Intelligence in Forensic Sciences Revolution or Invasion? Part II

Aim: The study is on the emerging role of artificial intelligence in the forensic sciences. After clarifying the basic concepts and a brief historical overview, the possibilities of using AI in various forensic fields are discussed: genetics, pattern recognition, chemistry, toxicology, anthropology, forensic medicine, and scene reconstruction. Methodology: The study synthesises several recently published international papers. Findings: The penetration of the application of artificial intelligence into some fields of science is undoubtedly an ongoing process. Most of the varied forensic fields also cannot avoid this development. Analysing large databases unmanageable with traditional methods, pattern recognition, and machine learning can all be important tools for forensic science. However, an important conclusion is that AI is a supporter of human expert work, not a substitute. Value: In the field of forensic sciences, no such detailed summary article has been published in Hungarian so far.

Mesterséges intelligencia a törvényszéki tudományokban Revolúció vagy invázió? II. rész

Aim: The study is on the emerging role of artificial intelligence in the forensic sciences. After clarifying the basic concepts and a brief historical overview, the possibilities of using AI in various forensic fields are discussed: genetics, pattern recognition, chemistry, toxicology, anthropology, forensic medicine, and scene reconstruction. Methodology: The study synthesises several recently published international papers. Findings: The penetration of the application of artificial intelligence into some fields of science is undoubtedly an ongoing process. Most of the varied forensic fields also cannot avoid this development. Analysing large databases unmanageable with traditional methods, pattern recognition, and machine learning can all be important tools for forensic science. However, an important conclusion is that AI is a supporter of human expert work, not a substitute. Value: In the field of forensic sciences, no such detailed summary article has been published in Hungarian so far.

An analytic formulation of convolutional neural network learning for pattern recognition

Training convolutional neural networks (CNNs) using back-propagation (BP) is a time-consuming and resource-intensive process, primarily due to the need to iterate over the dataset multiple times. In contrast, analytic learning aims to train neural networks in a single epoch, offering a potential solution to these challenges. However, existing studies of analytic learning have been limited to multilayer perceptrons (MLPs). In this article, we propose an analytic formulation for convolutional neural network learning (ACnnL), which represents a significant advancement towards non-iterative learning paradigms for CNNs. Our formulation demonstrates that ACnnL extends the principles of MLP regularization constraints. From the implicit regularization and network interpretability viewpoints, we provide insights into why CNNs often exhibit superior generalization capabilities. The ACnnL is validated by conducting classification tasks on benchmark datasets such as MNIST, FashionMNIST, CIFAR10, CIFAR100 and Tiny-ImageNet. It is encouraging that the ACnnL trains CNNs in a significantly fast manner with reasonably close prediction accuracies to those using BP. In particular, a 5-layer vanilla CNN trained by ACnnL gave an accuracy of 0.9931, 0.9155, 0.7049 and 0.4628 for these datasets. The ACnnL achieves training speeds that are approximately 17 times faster than BP on GPU and 113 times faster than BP on CPU, while maintaining competitive prediction accuracies. Moreover, our experiments disclose a unique advantage of ACnnL under the small-sample scenario when training data are scarce or expensive. In a nutshell, an analytic method which deals well with small-sample size data has been put forward for the first time for fast CNN training with inherent network interpretability.

Aggregated Pattern Classification Method for improving neural disorder stage detection.

Neurological disorders pose a significant health challenge, and their early detection is critical for effective treatment planning and prognosis. Traditional classification of neural disorders based on causes, symptoms, developmental stage, severity, and nervous system effects has limitations. Leveraging artificial intelligence (AI) and machine learning (ML) for pattern recognition provides a potent solution to address these challenges. Therefore, this study focuses on proposing an innovative approach-the Aggregated Pattern Classification Method (APCM)-for precise identification of neural disorder stages. The APCM was introduced to address prevalent issues in neural disorder detection, such as overfitting, robustness, and interoperability. This method utilizes aggregative patterns and classification learning functions to mitigate these challenges and enhance overall recognition accuracy, even in imbalanced data. The analysis involves neural images using observations from healthy individuals as a reference. Action response patterns from diverse inputs are mapped to identify similar features, establishing the disorder ratio. The stages are correlated based on available responses and associated neural data, with a preference for classification learning. This classification necessitates image and labeled data to prevent additional flaws in pattern recognition. Recognition and classification occur through multiple iterations, incorporating similar and diverse neural features. The learning process is finely tuned for minute classifications using labeled and unlabeled input data. The proposed APCM demonstrates notable achievements, with high pattern recognition (15.03%) and controlled classification errors (CEs) (10.61% less). The method effectively addresses overfitting, robustness, and interoperability issues, showcasing its potential as a powerful tool for detecting neural disorders at different stages. The ability to handle imbalanced data contributes to the overall success of the algorithm. The APCM emerges as a promising and effective approach for identifying precise neural disorder stages. By leveraging AI and ML, the method successfully resolves key challenges in pattern recognition. The high pattern recognition and reduced CEs underscore the method's potential for clinical applications. However, it is essential to acknowledge the reliance on high-quality neural image data, which may limit the generalizability of the approach. The proposed method allows future research to refine further and enhance its interpretability, providing valuable insights into neural disorder progression and underlying biological mechanisms.

Diagnosis And Accurate Classification of Apple Leaf Diseases Using Vision Transformers

In agriculture, plant pests and diseases are one of the major factors that reduce fruit and vegetable yields. Early detection of foliar diseases, which cause significant yield losses in apple production, is of great importance for the sustainability of agricultural production. The subjective and inefficient nature of traditional manual and tool-based diagnostic methods is insufficient to provide solutions that meet scientific and production standards. In this context, pattern recognition and deep learning techniques offer more effective alternatives for automatic image analysis and classification. In this study, Vision Transformers are used for disease classification from apple leaf images. The dataset consists of images of healthy apple leaves and apple leaves with different diseases. The results show that Vision Transformers is effective in classifying apple leaf diseases with high accuracy. This study makes an important contribution to increasing the use of digital technologies in agriculture and plant health monitoring.

NMNN: Newtonian Mechanics-based Natural Neighbor algorithm

Natural neighbor (NaN) algorithm, as a parameter-free alternative to KNN, is widely used in various fields such as pattern recognition and machine learning. However, the original NaN algorithm only takes the Euclidean distance of the samples as the only criterion for similarity calculation. It does not reasonably reflect the relationship between the samples and affects convergence and accuracy. Newtonian mechanics considers the effect of forces from many perspectives. Consequently, in this paper, Newtonian mechanics is introduced to calculate the relationship between samples, then the Newtonian mechanics-based Natural Neighbor algorithm (NMNN) is proposed. At first, it introduces Newton's first, second, and third laws and the idea of gravity in the natural neighbor algorithm. Then, it computes the mass of the samples by the natural neighbor density. Moreover, it updates the similarity matrix through Newtonian mechanics, which makes the calculation results more accurate. The formation process of the natural neighbor can be regarded as the spontaneous mutual attraction of samples based on traction. Finally, the experimental results show that the convergence performance of the NMNN algorithm is better than that of the NaN algorithm, and it can achieve better results in applications such as unbalanced data, clustering, and outlier detection.

Determination of early breeder in goldfish (Carassius auratus Linn.) with learning vector quantization, probabilistic and pattern recognition neural networks

It is common practice to categorize growing fish according to quality in fish farming centers. For such an application, experienced and estimating people are needed. By pre-selecting the right fish for the ornamental fish industry, it gets ahead of the competitors. This study deals with the classification of early breeder determination in goldfish using three different Artificial Neural Network (ANN) techniques. This classification model can help the fish industry assess classification risks and make the right decision. The used dataset was derived from the results of the classification section of 120 goldfish. It consisted of 7 input parameters (day, live weight, body length, head height, head width, body height, and current class). During trial, all goldfish fed by a diet contained 360 g crude protein and 4449.85 kcal metabolizable energy (kg / dry matter). The important types of classification ANNs, namely Learning Vector Quantization Neural Network (LVQNN), Probabilistic Neural Network (PNN), and Pattern Recognition Neural Network (PRNN) were employed for the machine learning scheme. The training and test performances of the ANN models were compared with the correct prediction ratio. They showed that all of the proposed ANN techniques were well at classification. However, the PRNN model was better than the LVQNN and PNN as a classifier for the breeder selection of goldfish. Therefore, the results of PRNN model such as histogram, Receiver Operating Characteristic (ROC) curve, regression, confusion matrix, accuracy, sensitivity, precision, and F1 score were given and discussed. Furthermore, the classification of early breeder determination in goldfish were examined for days with the best PRNN.

Predictive energy control for grid-connected industrial PV-battery systems using GEP-ANFIS

Rising energy costs in Uganda's industrial sector have significantly increased manufacturing expenses and led to the closure of potential enterprises, underscoring the urgent need for efficient energy management solutions. The study presents an optimization-based predictive energy control model that integrates Gene Expression Programming (GEP) and Adaptive Network-based Fuzzy Inference Systems (ANFIS) for grid-connected solar PV-battery systems. The hybrid GEP-ANFIS model leverages GEP's strength in complex pattern recognition and ANFIS's adaptive learning capability to manage nonlinear data, resulting in more accurate and reliable energy predictions. The model utilizes historical load patterns, grid data, and solar PV/battery inputs to forecast load demand and solar power generation, optimizing energy usage based on Time of Use (TOU) pricing. The approach addresses the limitations of existing energy management techniques by enhancing predictive accuracy and ensuring seamless integration with industrial applications. The present work demonstrates that the proposed GEP-ANFIS double diode model consistently outperforms its counterparts, achieving a mean absolute percentage error (MAPE) of 7.25 %, a mean absolute deviation (MAD) of 2.95 %, and a root mean square error (RMSE) of 8.42 %. Moreover, the model exhibits superior accuracy in predicting short-term load demand, with a MAPE of 2.24 %, a MAD of 0.13 %, and an RMSE of 0.18 %. Additionally, the model results in an average energy cost reduction of 6.7 % for hybrid grid-connected solar PV/battery configurations, highlighting its economic benefits. The study's key contributions include the creation of a novel hybrid predictive model, significant cost savings, improved load prediction accuracy, and the integration of renewable energy sources into industrial energy management. The model's inferences can effectively address the challenges of rising energy costs, providing a robust framework for sustainable and efficient energy management in industrial settings.

Combinatorial metabolomic and transcriptomic analysis of muscle growth in hybrid striped bass (female white bass Morone chrysops x male striped bass M. saxatilis)

BackgroundUnderstanding growth regulatory pathways is important in aquaculture, fisheries, and vertebrate physiology generally. Machine learning pattern recognition and sensitivity analysis were employed to examine metabolomic small molecule profiles and transcriptomic gene expression data generated from liver and white skeletal muscle of hybrid striped bass (white bass Morone chrysops x striped bass M. saxatilis) representative of the top and bottom 10 % by body size of a production cohort.ResultsLarger fish (good-growth) had significantly greater weight, total length, hepatosomatic index, and specific growth rate compared to smaller fish (poor-growth) and also had significantly more muscle fibers of smaller diameter (≤ 20 µm diameter), indicating active hyperplasia. Differences in metabolomic pathways included enhanced energetics (glycolysis, citric acid cycle) and amino acid metabolism in good-growth fish, and enhanced stress, muscle inflammation (cortisol, eicosanoids) and dysfunctional liver cholesterol metabolism in poor-growth fish. The majority of gene transcripts identified as differentially expressed between groups were down-regulated in good-growth fish. Several molecules associated with important growth-regulatory pathways were up-regulated in muscle of fish that grew poorly: growth factors including agt and agtr2 (angiotensins), nicotinic acid (which stimulates growth hormone production), gadd45b, rgl1, zfp36, cebpb, and hmgb1; insulin-like growth factor signaling (igfbp1 and igf1); cytokine signaling (socs3, cxcr4); cell signaling (rgs13, rundc3a), and differentiation (rhou, mmp17, cd22, msi1); mitochondrial uncoupling proteins (ucp3, ucp2); and regulators of lipid metabolism (apoa1, ldlr). Growth factors pttg1, egfr, myc, notch1, and sirt1 were notably up-regulated in muscle of good-growing fish.ConclusionA combinatorial pathway analysis using metabolomic and transcriptomic data collectively suggested promotion of cell signaling, proliferation, and differentiation in muscle of good-growth fish, whereas muscle inflammation and apoptosis was observed in poor-growth fish, along with elevated cortisol (an anti-inflammatory hormone), perhaps related to muscle wasting, hypertrophy, and inferior growth. These findings provide important biomarkers and mechanisms by which growth is regulated in fishes and other vertebrates as well.