Search

Speech is an effective way for communicating and exchanging complex information between humans. Speech signal has involved a great attention in human-computer interaction. Therefore, emotion recognition from speech has become a hot research topic in the field of interacting machines with humans. In this paper, we proposed a novel speech emotion recognition system by adopting multivariate time series handcrafted feature representation from speech signals. Bidirectional echo state network with two parallel reservoir layers has been applied to capture additional independent information. The parallel reservoirs produce multiple representations for each direction from the bidirectional data with two stages of concatenation. The sparse random projection approach has been adopted to reduce the high-dimensional sparse output for each direction separately from both reservoirs. Random over-sampling and random under-sampling methods are used to overcome the imbalanced nature of the used speech emotion datasets. The performance of the proposed parallel ESN model is evaluated from the speaker-independent experiments on EMO-DB, SAVEE, RAVDESS, and FAU Aibo datasets. The results show that the proposed SER model is superior to the single reservoir and the state-of-the-art studies. [ABSTRACT FROM AUTHOR]

The appearance of a human face changes with the change in body weight and age. With varying lifestyle choices, it is hard to imagine the appearance of a given human face in years to come. Future self-perception is highly associated with one's emotional state, as well as health behavior. Negative future self-perception can cause negative lifestyle choice and negative health behavior, leading to depression and eating disorder. In this paper, a new methodology is introduced for future self-face image synthesis using age and weight, resulting in visualization of future face image derived from given weight category and age. A Constrained Local Model is first used for weight progressed future face image synthesized and then age-progressed future face image is generated using Conditional Adversarial Auto Encoder. In the final step, both weight progressed and age-progressed face images fed to face morphing module which synthesized future face image by keeping natural looks. Experimental results show the advantages of proposed method with promising results. [ABSTRACT FROM AUTHOR]

This paper attempts to solve the typical problems of self-organizing growing network models, i.e. (a) an influence of the order of input data on the self-organizing ability, (b) an instability to high-dimensional data and an excessive sensitivity to noise, and (c) an expensive computational cost by integrating Kernel Bayes Rule (KBR) and Correntropy-Induced Metric (CIM) into Adaptive Resonance Theory (ART) framework. KBR performs a covariance-free Bayesian computation which is able to maintain a fast and stable computation. CIM is a generalized similarity measurement which can maintain a high-noise reduction ability even in a high-dimensional space. In addition, a Growing Neural Gas (GNG)-based topology construction process is integrated into the ART framework to enhance its self-organizing ability. The simulation experiments with synthetic and real-world datasets show that the proposed model has an outstanding stable self-organizing ability for various test environments. [ABSTRACT FROM AUTHOR]

This paper presents a new neural network for online topological map building inspired by beta oscillations and hippocampal place cell learning. The memory layer represents the hippocampus, the input layer represents the entorhinal, and the ρ is the orientation system. In this model, multiple-scale entorhinal grid cell activations form the input layer feature patterns, which are categorized by hippocampal place cells (nodes) and act as spatial categories in the memory layer. Top-down attentive matching and mismatch-mediated reset (beta oscillations), which are triggered by the orientation system, overcome the stability-plasticity dilemma and prevent the catastrophic forgetting of place cell maps. In our proposed method, nodes in the topological map represent place cells (robot location), while edges connect nodes and store robot action (i.e., orientation, direction). Our method is based upon a multi-channel Adaptive Resonance Associative Memory (ARAM) network architecture to obtain multiple sensory sources for topological map building. It comprises two layers: input and memory. The input layer collects sensory data and incrementally clusters the obtained information into a set of topological nodes. In the memory layer, the clustered information is used as a topological map where nodes are associated with actions. The advantages of the proposed method are: (1) it does not require high-level cognitive processes and prior knowledge to make it work in a natural environment; and (2) it can process multiple sensory sources simultaneously in continuous space, which is crucial for real-world robot navigation. Thus, we combine our Topological Gaussian ARAM method (TGARAM) with incremental principle component analysis to constitute a basis for topological map building. Lastly, the proposed method was validated using several standardized benchmark datasets. [ABSTRACT FROM AUTHOR]

Adaptive Resonance Theory (ART) is one of the successful approaches to resolving “the plasticity–stability dilemma” in neural networks, and its supervised learning model called ARTMAP is a powerful tool for classification. Among several improvements, such as Fuzzy or Gaussian based models, the state of art model is Bayesian based one, while solving the drawbacks of others. However, it is known that the Bayesian approach for the high dimensional and a large number of data requires high computational cost, and the covariance matrix in likelihood becomes unstable. This paper introduces Kernel Bayesian ART (KBA) and ARTMAP (KBAM) by integrating Kernel Bayes’ Rule (KBR) and Correntropy Induced Metric (CIM) to Bayesian ART (BA) and ARTMAP (BAM), respectively, while maintaining the properties of BA and BAM. The kernel frameworks in KBA and KBAM are able to avoid the curse of dimensionality. In addition, the covariance-free Bayesian computation by KBR provides the efficient and stable computational capability to KBA and KBAM. Furthermore, Correntropy-based similarity measurement allows improving the noise reduction ability even in the high dimensional space. The simulation experiments show that KBA performs an outstanding self-organizing capability than BA, and KBAM provides the superior classification ability than BAM, respectively. [ABSTRACT FROM AUTHOR]

The decision making process in communication is affected by internal and external factors from dynamic environments. Humans can perform a variety of behaviors in a similar situation, unlike robots. This paper discusses human psychological phenomena during communication from the point of view of internal and external factors, such as perception, memory, and emotional information. Based on these, we introduce the personality affected robotic emotional model and the emotion affected associative memory model for the robot. We organize an interactive robot system to provide suitable decisions for the robot. Results from interactive communication experiments indicate that the robot is able to perform different actions based on internal and external factors. [ABSTRACT FROM AUTHOR]

Mirror neurons are visuo-motor neurons found in primates and thought to be significant for imitation learning. The proposition that mirror neurons result from associative learning while the neonate observes his own actions has received noteworthy empirical support. Self-exploration is regarded as a procedure by which infants become perceptually observant to their own body and engage in a perceptual communication with themselves. We assume that crude sense of self is the prerequisite for social interaction. However, the contribution of mirror neurons in encoding the perspective from which the motor acts of others are seen have not been addressed in relation to humanoid robots. In this paper we present a computational model for development of mirror neuron system for humanoid based on the hypothesis that infants acquire MNS by sensorimotor associative learning through self-exploration capable of sustaining early imitation skills. The purpose of our proposed model is to take into account the view-dependency of neurons as a probable outcome of the associative connectivity between motor and visual information. In our experiment, a humanoid robot stands in front of a mirror (represented through self-image using camera) in order to obtain the associative relationship between his own motor generated actions and his own visual body-image. In the learning process the network first forms mapping from each motor representation onto visual representation from the self-exploratory perspective. Afterwards, the representation of the motor commands is learned to be associated with all possible visual perspectives. The complete architecture was evaluated by simulation experiments performed on DARwIn-OP humanoid robot. [ABSTRACT FROM AUTHOR]

Imitation learning through self-exploration is an essential mechanism in developing sensorimotor skills for human infants as well for robots. We assume that a primitive sense of self is the prerequisite for successful social interaction rather than an outcome of it. During imitation learning, a crucial element of conception involves segmenting the continuous flow of motion into simpler units – motion primitives – by identifying the boundaries of an action. Secondly, in realistic environment the robot must be able to learn the observed motion patterns incrementally in a stable adaptive manner without corrupting previously learned information. In this paper, we propose an on-line and unsupervised motion segmentation method allowing the robot to imitate and perform actions by observing the motion patterns performed by other partner through Incremental Kernel Slow Feature Analysis. The segmentation model directly operates on the images acquired from the robots vision sensor (camera) without requiring any kinematic model of the demonstrator. After segmentation, the spatio-temporal motion sequences are learned incrementally through Topological Gaussian Adaptive Resonance Hidden Markov Model. The learning model dynamically generates the topological structure in a self-organizing and self-stabilizing manner. Each node represents the encoded motion element (i.e. joint angles). The complete architecture was evaluated by simulation experiments performed on DARwIn-OP humanoid robot. [ABSTRACT FROM AUTHOR]

This work proposes a novel unsupervised self-organizing network, called the Self-Organizing Convolutional Echo State Network (SO-ConvESN), for learning node centroids and interconnectivity maps compatible with the deterministic initialization of Echo State Network (ESN) input and reservoir weights, in the context of human action recognition (HAR). To ensure stability and echo state property in the reservoir, Recurrent Plots (RPs) and Recurrence Quantification Analysis (RQA) techniques are exploited for explainability and characterization of the reservoir dynamics and hence tuning ESN hyperparameters. The optimized self-organizing reservoirs are cascaded with a Convolutional Neural Network (CNN) to ensure that the activation of internal echo state representations (ESRs) echoes similar topological qualities and temporal features of the input time-series, and the CNN efficiently learns the dynamics and multiscale temporal features from the ESRs for action recognition. The hyperparameter optimization (HPO) algorithms are additionally adopted to optimize the CNN stage in SO-ConvESN. Experimental results on the HAR problem using several publicly available 3D-skeleton-based action datasets demonstrate the showcasing of the RPs and RQA technique in examining the explainability of reservoir dynamics for designing stable self-organizing reservoirs and the usefulness of implementing HPOs in SO-ConvESN for the HAR task. The proposed SO-ConvESN exhibits competitive recognition accuracy. [ABSTRACT FROM AUTHOR]

A small change in image will cause a dramatic change in signals. Visual system is required to be able to ignore these changes, yet specific enough to perform recognition. This work intends to provide biological-backed insights into 2D translation and scaling invariance and 3D pose-invariance without imposing strain on memory and with biological justification. The model can be divided into lower and higher visual stages. Lower visual stage models the visual pathway from retina to the striate cortex (V1), whereas the modeling of higher visual stage is mainly based on current psychophysical evidences. [ABSTRACT FROM AUTHOR]

Abstract: Nowadays, most factories rely on machines to help boost up their production and process. Therefore, an effective machine condition monitoring system plays an important role in these factories to ensure that their production and process are running smoothly all the time. In this paper, a new and effective machine condition monitoring system using log-polar mapper, quaternion based thermal image correlator and max-product fuzzy neural network classifier is proposed. Two classification characteristics, namely peak to sidelobe ratio (PSR) and real to complex ratio of the discrete quaternion correlation output (p-value) are applied in this proposed machine condition monitoring system. Large PSR and p-value showed a good match among correlation of the input thermal image with a particular reference image, but reversely for small PSR and p-value match. In the simulation, log-polar mapping is found to have solved the rotation and scaling invariant problems in quaternion based thermal image correlation. Besides, log-polar mapping can possess two fold data compression capability. Log-polar mapping helps smoothen up the output correlation plane, hence making better measurement for PSR and p-values. The simulation results have also proven that the proposed system is an efficient machine condition monitoring system with an accuracy of more than 94%. [ABSTRACT FROM AUTHOR]

Lifelong learning is a long-standing aim for artificial agents that act in dynamic environments in which an agent needs to accumulate knowledge incrementally without forgetting previously learned representations. Contemporary methods for incremental learning from images are predominantly based on frame-based data recorded by conventional shutter cameras. We investigate methods for learning from data produced by event cameras and compare techniques to mitigate forgetting while learning incrementally. We propose a model that is composed of both, feature extraction and incremental learning. The feature extractor is utilized as a self-supervised sparse convolutional neural network that processes event-based data. The incremental learner uses a habituation-based method that works in tandem with other existing techniques. Our experimental results show that the combination of different existing techniques with our proposed habituation-based method can help avoid catastrophic forgetting even more, while learning incrementally from the features provided by the extraction module. [ABSTRACT FROM AUTHOR]

This paper proposes a multi-step prediction model for time series prediction, i.e. Meta-cognitive Recurrent Kernel Online Sequential Extreme Learning Machine with Drift Detector Mechanism (Meta-RKOS-ELM ALD). Recurrent multi-step algorithm is applied to release the limitation in the number of prediction steps, and Drift Detector Mechanism (DDM) is used to overcome the problem of concept drift in the prediction model. The new meta-cognitive strategy decides the way of the incoming data during training, which decreases the training computation of prediction model and solves the parameter dependency. In our evaluation, we use a total of six artificial data sets and three real-world data sets (Standard & Poor's 500 Index, Shanghai Stock Exchange Composite Index, and Ozone Concentration in Toronto) to prove the ability of kernel filters, the detecting ability of concept drift detector, and situation of applying meta-cognitive strategy in our proposed model. Experiments results indicate that the Meta-KOS-ELM ALD with DDM has better forecasting ability in various predicting periods with the shortest learning time, as compared with other algorithms. [ABSTRACT FROM AUTHOR]

Highlights • A hairstyle recommender system that follows an expert's guidelines using face shape. • A set of geometric features for face shape identification is introduced. • A new face shape open-dataset is created. • Classifying face shape using a pre-trained model from a face recognition task. • Fusing hand-crafted and deep features by multiple kernel learning to improve results. Abstract Identifying human face shape is the first and the most vital process prior to choosing the right hairstyle to wear on according to guidelines from hairstyle experts, especially for women. This work presents a novel framework for a hairstyle recommender system that is based on face shape classifier. This framework enables an automatic hairstyle recommendation with a single face image. This has a direct impact on beauty industry service providers. It can simulate how the user looks like when she is wearing the chosen hairstyle recommended by the expert system. The model used in this framework is based on Support Vector Machine. The framework is evaluated on hand-crafted, deep-learned (VGG-face) features and VGG-face fine-tuned version for the face shape classification task. In addition to evaluating these individual features by a well-designed framework, we attempted to fuse these three descriptors together in order to improve the performance of the classification task. Two combination techniques were employed, namely: Vector Concatenation and Multiple Kernel Learning (MKL) techniques. All the hyper-parameters of the model were optimised by using Particle Swarm Optimisation. The results show that combining hand-crafted and VGG-face descriptors with MKL yielded the best results at 70.3% of accuracy which was statistically significantly better than using individual features. Thus, combining multiple representations of the data with MKL can improve the overall performance of the expert system. In addition, this proves that hand-crafted descriptor can be complementary to deep-learned descriptor. [ABSTRACT FROM AUTHOR]

In this study, fractal dimension approaches were used for analyzing EEG to determine the optimum electrode placement to distinguish between concentration and relaxation states. The EEG of participants was recorded from multiple electrode placements while under relaxed state and while in a state of heightened mental activity. Higuchi and Katz algorithms were applied to extract the fractal dimension FD indices of windowed segments of the EEG. These were then plotted in a graph, and a simple threshold was applied to best divide the indices of relaxation and concentration. The Higuchi algorithm was found to be better than Katz at distinguishing between relaxation and concentration, and P3 was found to be the best position to measure concentration, with P8 being a close contender. [ABSTRACT FROM AUTHOR]

This paper presents the concept of Simultaneous Localization and Multi-Mapping (SLAMM). It is a system that ensures continuous mapping and information preservation despite failures in tracking due to corrupted frames or sensor’s malfunction; making it suitable for real-world applications. It works with single or multiple robots. In a single robot scenario the algorithm generates a new map at the time of tracking failure, and later it merges maps at the event of loop closure. Similarly, maps generated from multiple robots are merged without prior knowledge of their relative poses; which makes this algorithm flexible. The system works in real time at frame-rate speed. The proposed approach was tested on the KITTI and TUM RGB-D public datasets and it showed superior results compared to the state-of-the-arts in calibrated visual monocular keyframe-based SLAM. The mean tracking time is around 22 milliseconds. The initialization is twice as fast as it is in ORB-SLAM, and the retrieved map can reach up to 90 percent more in terms of information preservation depending on tracking loss and loop closure events. For the benefit of the community, the source code along with a framework to be run with Bebop drone are made available at . [ABSTRACT FROM AUTHOR]

Entropy, the complexity measures for time series, has found numerous successful applications in brain signal analysis such as detection of epileptic seizure and monitoring the depth of anesthesia. Renyi entropy generalizes the well-known Shannon entropy, and hence providing better flexibility in application to real data. The objective of this paper is to evaluate the effectiveness of Renyi entropy as feature extraction method for motor imagery (MI)-based brain-computer interface (BCI). In this paper, Renyi entropy has been implemented in MI systems of various settings using BCI competition data sets. The classification accuracy of Renyi entropy in all data sets is benchmarked against common spatial pattern (CSP), the state-of-the-art feature extraction method. For common binary class data sets, Renyi entropy achieves an accuracy of approximately 3.4 % higher for BCI Competition II Data Set III and 0.87 % lower for BCI Competition III Data Set I, and there is no difference in accuracy for BCI Competition III Data Set IVc when compared against conventional CSP. In small sample setting, the average classification accuracy using Renyi entropy is approximately 3.4 and 0.3 % higher as compared to conventional CSP and best performing variant of regularized CSP, respectively. In addition, Renyi entropy is also compared to other chaos-inspired feature extraction methods, namely Katz and Higuchi which are implemented for MI systems by earlier researchers. The effect of implementing Renyi entropy on multiple narrower frequency sub-bands is also investigated in this study. In multi-class setting, Renyi entropy shows no statistical difference ( p = 0.6022, paired t test) in accuracy when compared to the algorithm of the BCI competition winner. However, Renyi entropy has an advantage of requiring only one single application of feature extraction regardless of the number of classes, while multiple implementation of CSP is required as CSP is originally designed for binary class problem. The successful application of Renyi entropy in all the aforementioned settings indicates that Renyi entropy is a viable feature extraction alternative for MI-based BCI systems. [ABSTRACT FROM AUTHOR]

Estrus detection is essential for dairy farms to take cows for artificial insemination promptly. Conventional approaches for detecting estrus cows use electronic devices attached to cows to gather data for software analysis. However, electronic devices can be costly and make a cow moody and uncomfortable while moving. In a common approach, observers detect estrus cows by observing their behaviors. However, continuous observation can easily lead to errors due to the observer's fatigue. Therefore, we designed CowXNet, an automatic estrus detection system for cows, to assist farmers to detect estrus cows. CowXNet requires only a camera attached in a pen and a computer to analyze recorded videos. CowXNet analyzes the estrus behaviors of each cow in a pen and helps farmers to identify estrus cows. To develop and evaluate CowXNet efficiently and effectively, we collected data from Chokchai Farm, the biggest dairy farm in Asia (14. 65483 ∘ N, 101. 34853 ∘ E). CowXNet has four modules: (i) cow detection uses YOLOv4 to detect cows in recorded videos; (ii) body part detection uses a convolutional neural network to estimate locations of body parts of detected cows; (iii) estrus behavior detection uses body part coordinates to extract a set of discriminative features, and a classification algorithm to detect estrus behaviors, and (iv) behavior analysis module displays estrus behavior for analysis purposes. We evaluated CowXNet for two instances: module-independent evaluation and end-to-end framework evaluation. Overall, CowXNet was promising; it correctly detected estrus behavior interval of cows 83% of cases. • CowXNet: an automated estrus detection system without attached electronic devices. • A set of discriminative features for estrus detection was found. • A new benchmark dataset of cows with annotations is available. • The nose was the key indicator of estrus behavior, especially sniffing. • Cows behave differently when they are in estrus. [ABSTRACT FROM AUTHOR]

As different region of the brain is associated with different mental activity, channel selection is commonly used to enhance the performance of multi-electrode electroencephalography (EEG) system by removing task-irrelevant and redundant channels. Various channel selection methods are successfully implemented in Brain–Computer Interface (BCI) system that uses one type of brain activity by earlier researchers. Upon realizing the limitation of conventional BCI systems, there has been increasing number of hybrid BCI systems. These hybrid systems use combinations of two brain activity patterns to enhance the functionality of a system. In this paper, three multi-objective genetic algorithms (GAs) are proposed to optimize the number of channels selected and system accuracy. The objective of this research is to investigate the optimal tradeoff between the classification accuracy of a BCI system and the number of selected channels. This tradeoff is important because different BCI applications have different priorities; some implementations prefer minimum number of channels while others favor the classification accuracy. The second objective of this research is to investigate the effectiveness of the GAs adopted as a channel selection method for BCI systems based on different brain activity. Three BCI Competition data sets are used to evaluate the performance of the proposed GAs. Non-parametric Friedman test ( p -value=0.635) is also conducted and the result reveals that the significant reduction in number of channels does not have significant impact in the classification accuracy on the evaluation data. This confirms the validity of genetic algorithms as a channel selection method for both P300 and motor imagery data. [ABSTRACT FROM AUTHOR]

Cognitive load theory suggests that overloading of working memory may negatively affect the performance of human in cognitively demanding tasks. Evaluation of cognitive load is a difficult task; it is often assessed through feedback and evaluation from experts. Cognitive load classification based on Functional Near-InfraRed Spectroscopy (fNIRS) is now one of the key research areas in recent years, due to its resistance of artefacts, cost-effectiveness, and portability. To make fNIRS more practical in various applications, it is necessary to develop robust algorithms that can automatically classify fNIRS signals and less reliant on trained signals. Many of the analytical tools used in cognitive sciences have used Deep Learning (DL) modalities to uncover relevant information for mental workload classification. This review investigates the research questions on the design and overall effectiveness of DL as well as its key characteristics. We have identified 45 studies published between 2011 and 2023, that specifically proposed Machine Learning (ML) models for classifying cognitive load using data obtained from fNIRS devices. Those studies were analyzed based on type of feature selection methods, input, and DL model architectures. Most of the existing cognitive load studies are based on ML algorithms, which follow signal filtration and hand-crafted features. It is observed that hybrid DL architectures that integrate convolution and LSTM operators performed significantly better in comparison with other models. However, DL models especially hybrid models have not been extensively investigated for the classification of cognitive load captured by fNIRS devices. The current trends and challenges are highlighted to provide directions for the development of DL models pertaining to fNIRS research. [ABSTRACT FROM AUTHOR]

A novel model called Instinct Plasticity Echo State Network with New Weights Selection Method, which is Optimized by Genetic Algorithm, is proposed (IP-NESN-GA). There are three proposed methods that are employed in the conventional ESN. New weights selection method provides a novel approach to replace the random weights of ESN, which obtains much better performance than the conventional ESN. At the same time, Instinct Plasticity is successfully applied in ESN, which enhances the connection among reservoir states and increase the accuracy. Finally, Genetic Algorithm is applied to seek the most suitable parameters. The detecting ability of our proposed model is assessed on two kinds of distinct grammatical constructions data. It not only performs significantly better than the baselines in the meaning error and sentence error, but it also visualizes the expected probabilities for each of the possible thematic roles. [ABSTRACT FROM AUTHOR]

Abstract: Theoretical and simulational evidence, as well as experimental indications, are accumulating that quantum associative memory and imaging are possible. We compare these data with biological evidence, since we find them to a significant extent compatible. This paper presents a computationally implementable integrative model of appearance-based viewpoint-invariant recognition of objects. The neuro-quantum hybrid model incorporates neural processing up to V1 and quantum associative processing in V1, achieving together an object-recognition result in V2 and ITC. Results of our simulation of the central quantum-like parts of the bio-model, receiving neurally pre-processed inputs, are presented. This part contains our original simulated storage by multiple quantum interference of image-encoding Gabor wavelets done in a Hebbian way, especially using the Griniasty et al. pose-sequence learning rule. [Copyright &y& Elsevier]

In this paper, two real-world medical classification problems using electrocardiogram (ECG) and auscultatory blood pressure (Korotkoff) signals are examined. A total of nine machine learning models are applied to perform classification of the medical data sets. A number of useful performance metrics which include accuracy, sensitivity, specificity, as well as the area under the receiver operating characteristic curve are computed. In addition to the original data sets, noisy data sets are generated to evaluate the robustness of the classifiers against noise. The 10-fold cross validation method is used to compute the performance statistics, in order to ensure statistically reliable results pertaining to classification of the ECG and Korotkoff signals are produced. The outcomes indicate that while logistic regression models perform the best with the original data set, ensemble machine learning models achieve good accuracy rates with noisy data sets. [ABSTRACT FROM AUTHOR]

Highlights: [•] An ensemble of hybrid intelligent models is applied to condition monitoring tasks. [•] A review on condition monitoring of induction motors is presented. [•] The proposed model is able to learn incrementally and explain its predictions. [•] The results ascertain the usefulness of the model for condition monitoring tasks. [•] The knowledge base is presented as a decision tree for interpretation by users. [ABSTRACT FROM AUTHOR]

The article discusses the use of principal component analysis (PCA) technique for data glove-based signature identification and forgery detection. The PCA has been applied to condense the volume and enrich the operational data by noise reduction without affecting the efficiency. To project the advantage of the proposed method, calculation of the sum of mean squares of Euclidean distance has been used. The benefits of signal processing over the conventional image processing techniques are utilised by the emerging trend in signature verification. Benefits include reduction of equipment complexity, increased robustness and independence of media.

Currently, thanks to the rapid development of wireless sensor networks and network traffic monitoring, the data stream is gradually becoming one of the most popular data generating processes. The data stream is different from traditional static data. Cluster analysis is an important technology for data mining, which is why many researchers pay attention to grouping streaming data. In the literature, there are many data stream clustering techniques, unfortunately, very few of them try to solve the problem of clustering data streams coming from multiple sources. In this article, we present an algorithm with a tree structure for grouping data streams (in the form of a time series) that have similar properties and behaviors. We have evaluated our algorithm over real multivariate data streams generated by smart meter sensors—the Irish Commission for Energy Regulation data set. There were several measures used to analyze the various characteristics of a tree-like clustering structure (computer science perspective) and also measures that are important from a business standpoint. The proposed method was able to cluster the flows of data and has identified the customers with similar behavior during the analyzed period. [ABSTRACT FROM AUTHOR]