Pattern Classification Applications Research Articles

CogniFiber: Harnessing Biocompatible and Biodegradable 1D Collagen Nanofibers for Sustainable Nonvolatile Memory and Synaptic Learning Applications.

Here, resistive switching (RS) devices are fabricated using naturally abundant, nontoxic, biocompatible, and biodegradable biomaterials. For this purpose, 1D chitosan nanofibers (NFs), collagen NFs, and chitosan-collagen NFs are synthesized by using an electrospinning technique. Among different NFs, the collagen-NFs-based device shows promising RS characteristics. In particular, the optimized Ag/collagen NFs/fluorine-doped tin oxide RS device shows a voltage-tunable analog memory behavior and good nonvolatile memory properties. Moreover, it can also mimic various biological synaptic learning properties and can be used for pattern classification applications with the help of the spiking neural network. The time series analysis technique is employed to model and predict the switching variations of the RS device. Moreover, the collagen NFs have shown good cytotoxicity and anticancer properties, suggesting excellent biocompatibility as a switching layer. The biocompatibility of collagen NFs is explored with the help of NRK-52E (Normal Rat Kidney cell line) and MCF-7 (Michigan Cancer Foundation-7 cancer cell line). Additionally, the biodegradability of the device is evaluated through a physical transient test. This work provides a vital step toward developing a biocompatible and biodegradable switching material for sustainable nonvolatile memory and neuromorphic computing applications.

A generalized Hellinger distance for multisource information fusion and its application in pattern classification

A generalized Hellinger distance for multisource information fusion and its application in pattern classification

Higher Order Fractal Belief Rényi Divergence With Its Applications in Pattern Classification.

Information can be quantified and expressed by uncertainty, and improving the decision level of uncertain information is vital in modeling and processing uncertain information. Dempster-Shafer evidence theory can model and process uncertain information effectively. However, the Dempster combination rule may provide counter-intuitive results when dealing with highly conflicting information, leading to a decline in decision level. Thus, measuring conflict is significant in the improvement of decision level. Motivated by this issue, this paper proposes a novel method to measure the discrepancy between bodies of evidence. First, the model of dynamic fractal probability transformation is proposed to effectively obtain more information about the non-specificity of basic belief assignments (BBAs). Then, we propose the higher-order fractal belief Rényi divergence (HOFBReD). HOFBReD can effectively measure the discrepancy between BBAs. Moreover, it is the first belief Rényi divergence that can measure the discrepancy between BBAs with dynamic fractal probability transformation. HoFBReD has several properties in terms of probability transformation as well as measurement. When the dynamic fractal probability transformation ends, HoFBReD is equivalent to measuring the Rényi divergence between the pignistic probability transformations of BBAs. When the BBAs degenerate to the probability distributions, HoFBReD will also degenerate to or be related to several well-known divergences. In addition, based on HoFBReD, a novel multisource information fusion algorithm is proposed. A pattern classification experiment with real-world datasets is presented to compare the proposed algorithm with other methods. The experiment results indicate that the proposed algorithm has a higher average pattern recognition accuracy with all datasets than other methods. The proposed discrepancy measurement method and multisource information algorithm contribute to the improvement of decision level.

Learning smooth dendrite morphological neurons for pattern classification using linkage trees and evolutionary-based hyperparameter tuning

The current learning approach for smooth dendrite morphological neurons (DMNs) determines dendrite parameters using k-means clustering, which is non-reproducible due to its stochastic nature, risking falling into local minima. To overcome this issue, we introduce a DMN learning approach based on a deterministic hierarchical clustering method, which builds a linkage tree for each class of patterns. In addition, a micro genetic algorithm automatically tunes the cut-off points in the linkage trees hierarchy to create suitable clusters of dendrites. The classification experiments consider 40 real-world datasets. The proposed approach outperforms three DMN models in classification performance and is quite competitive with a hybrid morphological-linear perceptron, multilayer perceptron, random forest, and support vector machine. Therefore, the proposed method is a suitable alternative for pattern classification applications.

Higher order belief divergence with its application in pattern classification

Multi-source information fusion is a sophisticated estimation process that generates a unified profile to assess complex situations. Dempster–Shafer evidence theory (DSET) is a practical theory in handling uncertain information in multi-source information fusion. However, highly-conflicting evidence may cause the combination rule of Dempster to provide counterintuitive results. Thus, how to effectively reconcile highly-conflicting evidence in DSET is still an open issue. To address this problem, in this paper, a novel belief divergence, higher order belief Jensen-Shannon divergence is proposed to measure the discrepancy between BPAs in DSET. The proposed higher order belief Jensen-Shannon divergence is the first method to dynamically measure the discrepancy between BPAs over the time evolution, i.e., to measure the discrepancy between BPAs with different time scale in the future. Besides, the proposed higher order belief Jensen-Shannon divergence has benefits from the perspective of measurement. It satisfies the properties of nonnegativeness and nondegeneracy, symmetry, and the triangle inequality of root form. Based on the proposed higher order belief Jensen-Shannon divergence, a novel multi-source information fusion algorithm is proposed. Eventually, the proposed algorithm is applied to a pattern classification experiment with real-world datasets.

A novel distance measure for intuitionistic fuzzy sets and its application in pattern classification, medical diagnosis, and career determination

A novel distance measure for intuitionistic fuzzy sets and its application in pattern classification, medical diagnosis, and career determination

Fractal Belief Rényi Divergence with Its Applications in Pattern Classification

Fractal Belief Rényi Divergence with Its Applications in Pattern Classification

A cerebellum inspired spiking neural network as a multi-model for pattern classification and robotic trajectory prediction.

Spiking neural networks were introduced to understand spatiotemporal information processing in neurons and have found their application in pattern encoding, data discrimination, and classification. Bioinspired network architectures are considered for event-driven tasks, and scientists have looked at different theories based on the architecture and functioning. Motor tasks, for example, have networks inspired by cerebellar architecture where the granular layer recodes sparse representations of the mossy fiber (MF) inputs and has more roles in motor learning. Using abstractions from cerebellar connections and learning rules of deep learning network (DLN), patterns were discriminated within datasets, and the same algorithm was used for trajectory optimization. In the current work, a cerebellum-inspired spiking neural network with dynamics of cerebellar neurons and learning mechanisms attributed to the granular layer, Purkinje cell (PC) layer, and cerebellar nuclei interconnected by excitatory and inhibitory synapses was implemented. The model's pattern discrimination capability was tested for two tasks on standard machine learning (ML) datasets and on following a trajectory of a low-cost sensor-free robotic articulator. Tuned for supervised learning, the pattern classification capability of the cerebellum-inspired network algorithm has produced more generalized models than data-specific precision models on smaller training datasets. The model showed an accuracy of 72%, which was comparable to standard ML algorithms, such as MLP (78%), Dl4jMlpClassifier (64%), RBFNetwork (71.4%), and libSVM-linear (85.7%). The cerebellar model increased the network's capability and decreased storage, augmenting faster computations. Additionally, the network model could also implicitly reconstruct the trajectory of a 6-degree of freedom (DOF) robotic arm with a low error rate by reconstructing the kinematic parameters. The variability between the actual and predicted trajectory points was noted to be ± 3 cm (while moving to a position in a cuboid space of 25 × 30 × 40 cm). Although a few known learning rules were implemented among known types of plasticity in the cerebellum, the network model showed a generalized processing capability for a range of signals, modulating the data through the interconnected neural populations. In addition to potential use on sensor-free or feed-forward based controllers for robotic arms and as a generalized pattern classification algorithm, this model adds implications to motor learning theory.

New Pythagorean fuzzy-based distance operators and their applications in pattern classification and disease diagnostic analysis

New Pythagorean fuzzy-based distance operators and their applications in pattern classification and disease diagnostic analysis

A Framework for Text Classification Using Evolutionary Contiguous Convolutional Neural Network and Swarm Based Deep Neural Network.

To classify the texts accurately, many machine learning techniques have been utilized in the field of Natural Language Processing (NLP). For many pattern classification applications, great success has been obtained when implemented with deep learning models rather than using ordinary machine learning techniques. Understanding the complex models and their respective relationships within the data determines the success of such deep learning techniques. But analyzing the suitable deep learning methods, techniques, and architectures for text classification is a huge challenge for researchers. In this work, a Contiguous Convolutional Neural Network (CCNN) based on Differential Evolution (DE) is initially proposed and named as Evolutionary Contiguous Convolutional Neural Network (ECCNN) where the data instances of the input point are considered along with the contiguous data points in the dataset so that a deeper understanding is provided for the classification of the respective input, thereby boosting the performance of the deep learning model. Secondly, a swarm-based Deep Neural Network (DNN) utilizing Particle Swarm Optimization (PSO) with DNN is proposed for the classification of text, and it is named Swarm DNN. This model is validated on two datasets and the best results are obtained when implemented with the Swarm DNN model as it produced a high classification accuracy of 97.32% when tested on the BBC newsgroup text dataset and 87.99% when tested on 20 newsgroup text datasets. Similarly, when implemented with the ECCNN model, it produced a high classification accuracy of 97.11% when tested on the BBC newsgroup text dataset and 88.76% when tested on 20 newsgroup text datasets.

A novel belief χ2 ${\chi }^{2}$ divergence for multisource information fusion and its application in pattern classification

Dempster–Shafer (D-S) evidence theory is invaluable in the domain of multisource information fusion for handing uncertainty problems. However, there may be counter-intuitive phenomenon when facing highly conflicting information. In this paper, a novel symmetric enhanced belief χ 2 ${\chi }^{2}$ divergence measure, called S E B χ 2 $SEB{\chi }^{2}$ , is proposed to measure the discrepancy between basic probability assignments (BPAs). The S E B χ 2 $SEB{\chi }^{2}$ divergence consider the features of BPAs as the influence of both single-element subsets and multielement subsets is taken into account. Furthermore, the S E B χ 2 $SEB{\chi }^{2}$ divergence is proven to be symmetric, nonnegative and nondegenerate, which are desirable properties for conflict management. Then, a new algorithm for multisource information fusion based on the S E B χ 2 $SEB{\chi }^{2}$ divergence measure is derived. Finally, an application for pattern classification is used to illustrate the superiority of the proposed S E B χ 2 $SEB{\chi }^{2}$ divergence measure-based fusion method over other existing well-known and recent related works with a better classification accuracy of 94.39%.

A Complex Weighted Discounting Multisource Information Fusion With Its Application in Pattern Classification

Complex evidence theory (CET) is an effective method for uncertainty reasoning in knowledge-based systems with good interpretability that has recently attracted much attention. However, approaches to improve the performance of uncertainty reasoning in CET-based expert systems remains an open issue. One key to performance improvement is the adequate management of conflict from multisource information. In this paper, a generalized correlation coefficient, namely, the complex evidential correlation coefficient (CECC), is proposed for the complex mass functions or complex basic belief assignments (CBBAs) in CET. On this basis, a complex conflict coefficient is proposed to measure the conflict between CBBAs; when CBBAs turn into classic BBAs, the complex correlation and conflict coefficients will degrade into traditional coefficients. The complex conflict coefficient satisfies nonnegativity, symmetry, boundedness, extreme consistency, and insensitivity to refinement properties, which are desirable for conflict measurement. Several numerical examples validate through comparisons the superiority of the complex conflict coefficient. In this context, a weighted discounting multisource information fusion algorithm, which is called the CECC-WDMSIF, is designed based on the CECC to improve the performance of CET-based expert systems. By applying the CECC-WDMSIF method to the pattern classification of diverse real-world datasets, it is demonstrated that the proposed CECC-WDMSIF outperforms well-known related approaches with higher classification accuracy and robustness.

Constructing artificial neural networks using genetic circuits to realize neuromorphic computing

With the advent of the era of big data, existing computing systems limit the development of new technologies such as cloud computing and artificial intelligence. As the demand for high-performance computing continues to grow, traditional computing models are facing unprecedented challenges. Neural mimicry computing based on artificial neural networks provides a potential solution, and biological computing with advantages such as low energy consumption parallelization is very important for its research, in which gene circuitry will be the key to the construction of artificial neural networks. In this study, we used biological elements to construct gene circuits to form neural network structures to realize neural mimicry calculations. Based on the similarity between the structure of gene circuits and neural networks, we describe a neural network design based on the logic gate calculation of gene circuits, and realize the applications of linear classification, nonlinear classification and pattern classification. In the linear classification of NAND, the four datasets in the space are divided into two categories by hyperplane. The structural design of the neural network for linear classification can be realized by using the genetic circuit, which includes five biological elements and the biological elements between different layers are connected by gene regulation. The simulation results of linear classification showed that the gene circuit realizes the function of NAND by means of the neural network and the output fluorescence value as the detection standard. In the nonlinear classification of XOR, the four datasets in the space are divided into two categories. The structural design of the neural network for nonlinear classification can be realized by using the genetic circuit, which includes six biological elements. The simulation results of nonlinear classification showed that the gene circuit realizes the function of XOR by means of the neural network and the output fluorescence value as the detection standard. In the pattern classification, the neural network with five layers and eight neurons was constructed by a genetic circuit with eight biological elements connected by gene regulation. The simulation results of pattern classification showed that the gene circuit realizes the function of pattern classification by means of the neural network and the output fluorescence value as the detection standard. This study realizes neural mimicry computing by constructing neural network through engineering gene circuits. The integrated molecular computing system is expected to be applied to the manufacturing of artificial intelligence chips, and further applied to aerospace, information security and national defense construction, along with other fields.

Application of a standardized EEG pattern classification in the assessment of neurological prognosis after cardiac arrest: A retrospective analysis

IntroductionElectroencephalogram (EEG) is used in the neurological prognostication after cardiac arrest. “Highly malignant” EEG patterns classified according to Westhall have a high specificity for poor neurological outcome when applied within protocols of recent studies. However, their predictive performance when applied in everyday clinical practice has not been investigated. We studied the prognostic accuracy and the interrater agreement when standardized EEG patterns were analysed and compared to neurological outcome in a patient cohort at a tertiary centre not involved in the original study of the standardized EEG pattern classification. MethodsComatose patients treated for out-of-hospital cardiac arrest were included. Poor outcome was defined as Cerebral Performance Category 3–5. Two senior consultants and one resident in clinical neurophysiology, blinded to clinical data and outcome, independently reviewed their EEG registrations and categorised the pattern as “highly malignant”, “malignant” or “benign”. These categories were compared to neurological outcome at hospital discharge. Interrater agreement was assessed using Cohen’s Kappa. ResultsIn total, 62 patients were included. The median (IQR) time to EEG was 59 (42–91) h after return of spontaneous circulation. Poor outcome was found in 52 (84%) patients. In 21 patients at least one of the raters considered the EEG to contain a “highly malignant” pattern, all with poor outcome (42% sensitivity, 100% specificity). The interrater agreement varied from kappa 0.62 to 0.29. Conclusion“Highly malignant” patterns predict poor neurological outcome with a high specificity in everyday practice. However, interrater agreement may vary substantially even between experienced EEG interpreters.

Can One Design a Series of Fast, Energy - Efficient “Brains” based on Neuromorphic Computing to Solve Complex Pattern Recognition Problem?

Background: In this paper, we present a theoretical discussion on neuromorphic computing circuit dynamic and its relations with AI deep learning neural networks. The hardware implememtations of neuromorphic computing and AI deep learning neuronal networks are discussed. Objective: The investigation is motivated by the design of a feasible fast and energy-efficient circuit device as well as an efficient training computation method to solve complex classification problems using AI neuronal networks. Methods: We focus on the investigations of solving pattern classification and recognition problems in real applications from the perspectives of both logic computation view and physical circuit device views. FPGA approaches are considered and a mapping from logic level to physical level is proposed. Results: A pragmatic mapping method is derived. FPGA method is proposed. Conclusion: Thus we propose in this paper an approach to solve complex classification problems. First, the neuromorphic computing as a new research area is introduced, including physical circuit properties, memristive device physical properties and the circuit dynamics described by the temporal and spatial (Maxwell) differential equations. Secondly, we show that by using AI deep learning neural networks to train AI neural networks we are able to derive the optimal AI neuron network weights. Last but not the least, we brief a mapping method and show in general how the neuromorphic circuit will work in practice after mapping the weights from AI deep learning neural networks into the neuromorphic circuit synapses/memristors. We also devote our discussions to the physical device feasibility and related matters. The method proposed in this paper is pragmatic and constructive.

Law of large numbers for a two-dimensional class cover problem

We prove a Law of Large Numbers (LLN) for the domination number of class cover catch digraphs (CCCD) generated by random points in two (or higher) dimensions. DeVinney and Wierman (2002) proved the Strong Law of Large Numbers (SLLN) for the uniform distribution in one dimension, and Wierman and Xiang (2008) extended the SLLN to the case of general distributions in one dimension. In this article, using subadditive processes, we prove a SLLN result for the domination number generated by Poisson points in ℝ2. From this we obtain a Weak Law of Large Numbers (WLLN) for the domination number generated by random points in [0, 1]2 from uniform distribution first, and then extend these result to the case of bounded continuous distributions. We also extend the results to higher dimensions. The domination number of CCCDs and related digraphs have applications in statistical pattern classification and spatial data analysis.

Binary Whale Optimization Algorithm for Dimensionality Reduction

Feature selection (FS) was regarded as a global combinatorial optimization problem. FS is used to simplify and enhance the quality of high-dimensional datasets by selecting prominent features and removing irrelevant and redundant data to provide good classification results. FS aims to reduce the dimensionality and improve the classification accuracy that is generally utilized with great importance in different fields such as pattern classification, data analysis, and data mining applications. The main problem is to find the best subset that contains the representative information of all the data. In order to overcome this problem, two binary variants of the whale optimization algorithm (WOA) are proposed, called bWOA-S and bWOA-V. They are used to decrease the complexity and increase the performance of a system by selecting significant features for classification purposes. The first bWOA-S version uses the Sigmoid transfer function to convert WOA values to binary ones, whereas the second bWOA-V version uses a hyperbolic tangent transfer function. Furthermore, the two binary variants introduced here were compared with three famous and well-known optimization algorithms in this domain, such as Particle Swarm Optimizer (PSO), three variants of binary ant lion (bALO1, bALO2, and bALO3), binary Dragonfly Algorithm (bDA) as well as the original WOA, over 24 benchmark datasets from the UCI repository. Eventually, a non-parametric test called Wilcoxon’s rank-sum was carried out at 5% significance to prove the powerfulness and effectiveness of the two proposed algorithms when compared with other algorithms statistically. The qualitative and quantitative results showed that the two introduced variants in the FS domain are able to minimize the selected feature number as well as maximize the accuracy of the classification within an appropriate time.

Vectorization of persistence barcode with applications in pattern classification of porous structures

Persistence barcode is a topological summary for persistent homology to exhibit topological features with different persistence. Persistence rank function (PRF), derived from persistence barcode, organizes persistence Betti numbers in the form of an integer-valued function. To obtain topological patterns of objects such as point clouds represented by finite-dimensional vectors for machine learning classification tasks, the vectorizing representations of barcodes is generated via decomposing PRF on a system of Haar basis. Theoretically, the generated vectorizing representation is proved to have 1-Wasserstein stability. In practice, to reduce training time and achieve better results, a technique of dimensionality reduction through out-of-sample mapping in supervised manifold learning is used to generate a low-dimensional vector. Experiments demonstrate that the representation is effective for capturing the topological patterns of data sets. Moreover, the classification of porous structures has become an essential problem in the fields such as material science in recent decades. The proposed method is successfully applied to distinguish porous structures on a novel data set of porous models.

A method for embedding a computer vision application into a wearable device

Pattern classification applications can be found everywhere, especially the ones that use computer vision. What makes them difficult to embed is the fact that they often require a lot of computational resources. Embedded computer vision has been applied in many contexts, such as industrial or home automation, robotics, and assistive technologies. This work performs a design space exploration in an image classification system and embeds a computer vision application into a minimum resource platform, targeting wearable devices. The feature extractor and the classifier are evaluated for memory usage and computation time. A method is proposed to optimize such characteristics, leading to a reduction of over 99% in computation time and 92% in memory usage, with respect to a standard implementation. Experimental results in an ARM Cortex-M platform showed a total classification time of 0.3 s, maintaining the same accuracy as in the simulation performed. Furthermore, less than 20 KB of data memory was required, which is the most limited resource available in low-cost and low-power microcontrollers. The target application, used for the experimental evaluation, is a crosswalk detector used to help visually impaired persons.

Iterative Deep Neighborhood: A Deep Learning Model Which Involves Both Input Data Points and Their Neighbors.

Deep learning models, such as deep convolutional neural network and deep long-short term memory model, have achieved great successes in many pattern classification applications over shadow machine learning models with hand-crafted features. The main reason is the ability of deep learning models to automatically extract hierarchical features from massive data by multiple layers of neurons. However, in many other situations, existing deep learning models still cannot gain satisfying results due to the limitation of the inputs of models. The existing deep learning models only take the data instances of an input point but completely ignore the other data points in the dataset, which potentially provides critical insight for the classification of the given input. To overcome this gap, in this paper, we show that the neighboring data points besides the input data point itself can boost the deep learning model's performance significantly and design a novel deep learning model which takes both the data instances of an input point and its neighbors' classification responses as inputs. In addition, we develop an iterative algorithm which updates the neighbors of data points according to the deep representations output by the deep learning model and the parameters of the deep learning model alternately. The proposed algorithm, named “Iterative Deep Neighborhood (IDN),” shows its advantages over the state-of-the-art deep learning models over tasks of image classification, text sentiment analysis, property price trend prediction, etc.