Search Results (13)

Understanding and interpreting human emotions through neurophysiological signals has become a central goal in affective computing. This paper presents a focused survey of recent advances in emotion recognition using tensor factorization techniques specifically applied to functional Near-Infrared Spectroscopy (fNIRS) data. We examine how tensor-based frameworks have been leveraged to capture the temporal, spatial, and spectral characteristics of fNIRS brain signals, enabling effective dimensionality reduction and latent pattern extraction. Focusing on third-order tensor constructions (trials × channels × time), we compare the use of Canonical Polyadic (CP) and Tucker decompositions in isolating components representative of emotional states. The review further evaluates the performance of extracted features when classified by conventional machine learning models such as Random Forests and Support Vector Machines. Emphasis is placed on comparative accuracy, interpretability, and the advantages of tensor methods over traditional approaches for distinguishing arousal and valence levels. We conclude by discussing the relevance of these methods for the development of real-time, explainable, emotion-aware systems in wearable neurotechnology, with a particular focus on medical applications such as mental health monitoring, early diagnosis of affective disorders, and personalized neurorehabilitation. Full article

In structural health monitoring (SHM), ensuring data completeness is critical for enhancing the accuracy and reliability of structural condition assessments. SHM data are prone to random missing values due to signal interference or connectivity issues, making precise data imputation essential. A latent factorization of tensor (LFT)-based method has proven effective for such problems, with optimization typically achieved via stochastic gradient descent (SGD). However, SGD-based LFT models and other imputation methods exhibit significant sensitivity to learning rates and slow tail-end convergence. To address these limitations, this study proposes an RMSprop-incorporated latent factorization of tensor (RLFT) model, which integrates an adaptive learning rate mechanism to dynamically adjust step sizes based on gradient magnitudes. Experimental validation on a scaled bridge accelerometer dataset demonstrates that RLFT achieves faster convergence and higher imputation accuracy compared to state-of-the-art models including SGD-based LFT and the long short-term memory (LSTM) network, with improvements of at least 10% in both imputation accuracy and convergence rate, offering a more efficient and reliable solution for missing data handling in SHM. Full article

Generative models based on Variational Autoencoders (VAEs) represent an important area of research in Controllable Text Generation (CTG). However, existing approaches often do not fully exploit the potential of latent variables, leading to limitations in both the diversity and thematic consistency of the generated text. To overcome these challenges, this paper introduces a new framework based on Hierarchical Latent Modulation (HLM). The framework incorporates a hierarchical latent space modulation module for the generation and embedding of conditional modulation parameters. By using low-rank tensor factorization (LMF), the approach combines multi-layer latent variables and generates modulation parameters based on conditional labels, enabling precise control over the features during text generation. Additionally, layer-by-layer normalization and random dropout mechanisms are employed to address issues such as the under-utilization of conditional information and the collapse of generative patterns. We performed experiments on five baseline models based on VAEs for conditional generation, and the results demonstrate the effectiveness of the proposed framework. Full article

Background/Objectives: The integration of microbiome and metabolome data could unveil profound insights into biological processes. However, widely used multi-omic data analyses often employ a stepwise mining approach, failing to harness the full potential of multi-omic datasets and leading to reduced detection accuracy. Synergistic analysis incorporating microbiome/metabolome data are essential for deeper understanding. Method: This study introduces a Coupled Matrix and Tensor Factorization (CMTF) framework for the joint analysis of microbiome and metabolome data, overcoming these limitations. Two CMTF frameworks were developed to factorize microbial taxa, functional pathways, and metabolites into latent factors, facilitating dimension reduction and biomarker identification. Validation was conducted using three diverse microbiome/metabolome datasets, including built environments and human gut samples from inflammatory bowel disease (IBD) and COVID-19 studies. Results: Our results revealed biologically meaningful biomarkers, such as Bacteroides vulgatus and acylcarnitines associated with IBD and pyroglutamic acid and p-cresol associated with COVID-19 outcomes, which provide new avenues for research. The CMTF framework consistently outperformed traditional methods in both dimension reduction and biomarker detection, offering a robust tool for uncovering biologically relevant insights. Conclusions: Despite its stringent data requirements, including the reliance on stratified microbial-based pathway abundances and taxa-level contributions, this approach provides a significant step forward in multi-omics integration and analysis, with potential applications across biomedical, environmental, and agricultural research. Full article

Since multi-view learning leverages complementary information from multiple feature sets to improve model performance, a tensor-based data fusion layer for neural networks, called Multi-View Data Tensor Fusion (MV-DTF), is used. It fuses M feature spaces ${\mathcal{X}}_1,\cdots ,{\mathcal{X}}_M$, referred to as views, in a new latent tensor space, $\mathcal{S}$, of order P and dimension $J_1\times \cdots \times J_P$, defined in the space of affine mappings composed of a multilinear map $\mathcal{T}:{\mathcal{X}}_1\times \cdots \times {\mathcal{X}}_M\to \mathcal{S}$—represented as the Einstein product between a $(P+M)$-order tensor $\mathcal{A}$ anda rank-one tensor, $\mathcal{X}={\mathbf{x}}^{(1)}\otimes \cdots \otimes {\mathbf{x}}^{(M)}$, where ${\mathbf{x}}^{(m)}\in {\mathcal{X}}_m$ is the m-th view—and a translation. Unfortunately, as the number of views increases, the number of parameters that determine the MV-DTF layer grows exponentially, and consequently, so does its computational complexity. To address this issue, we enforce low-rank constraints on certain subtensors of tensor $\mathcal{A}$ using canonical polyadic decomposition, from which M other tensors ${\mathcal{U}}^{(1)},\cdots ,{\mathcal{U}}^{(M)}$, called here Hadamard factor tensors, are obtained. We found that the Einstein product $\mathcal{A}{⊛}_M\mathcal{X}$ can be approximated using a sum of R Hadamard products of M Einstein products encoded as ${\mathcal{U}}^{(m)}{⊛}_1{\mathbf{x}}^{(m)}$, where R is related to the decomposition rank of subtensors of $\mathcal{A}$. For this relationship, the lower the rank values, the more computationally efficient the approximation. To the best of our knowledge, this relationship has not previously been reported in the literature. As a case study, we present a multitask model of vehicle traffic surveillance for occlusion detection and vehicle-size classification tasks, with a low-rank MV-DTF layer, achieving up to $92.81\%$ and $95.10\%$ in the normalized weighted Matthews correlation coefficient metric in individual tasks, representing a significant $6\%$ and $7\%$ improvement compared to the single-task single-view models. Full article

Functional dissociation of brain neural activity induced by opening or closing the eyes has been well established. However, how the temporal dynamics of the underlying neuronal modulations differ between these eye conditions during movement-related behaviours is less known. Using a robotic-assisted motor imagery brain-computer interface (MI BCI), we measured neural activity over the motor regions with electroencephalography (EEG) in a stroke survivor during his longitudinal rehabilitation training. We investigated lateralized oscillatory sensorimotor rhythm modulations while the patient imagined moving his hemiplegic hand with closed and open eyes to control an external robotic splint. In order to precisely identify the main profiles of neural activation affected by MI with eyes-open (MIEO) and eyes-closed (MIEC), a data-driven approach based on parallel factor analysis (PARAFAC) tensor decomposition was employed. Using the proposed framework, a set of narrow-band, subject-specific sensorimotor rhythms was identified; each of them had its own spatial and time signature. When MIEC trials were compared with MIEO trials, three key narrow-band rhythms whose peak frequencies centred at ∼8.0 Hz, ∼11.5 Hz, and ∼15.5 Hz, were identified with differently modulated oscillatory dynamics during movement preparation, initiation, and completion time frames. Furthermore, we observed that lower and higher sensorimotor oscillations represent different functional mechanisms within the MI paradigm, reinforcing the hypothesis that rhythmic activity in the human sensorimotor system is dissociated. Leveraging PARAFAC, this study achieves remarkable precision in estimating latent sensorimotor neural substrates, aiding the investigation of the specific functional mechanisms involved in the MI process. Full article

The problem of multispectral and hyperspectral image fusion (MHF) is to reconstruct images by fusing the spatial information of multispectral images and the spectral information of hyperspectral images. Focusing on the problem that the hyperspectral canonical polyadic decomposition model and the Tucker model cannot introduce the physical interpretation of the latent factors into the framework, it is difficult to use the known properties and abundance of endmembers to generate high-quality fusion images. This paper proposes a new fusion algorithm. In this paper, a coupled non-negative block-term tensor model is used to estimate the ideal high spatial resolution hyperspectral images, its sparsity is characterized by adding 1-norm, and total variation (TV) is introduced to describe piecewise smoothness. Secondly, the different operators in two directions are defined and introduced to characterize their piecewise smoothness. Finally, the proximal alternating optimization (PAO) algorithm and the alternating multiplier method (ADMM) are used to iteratively solve the model. Experiments on two standard datasets and two local datasets show that the performance of this method is better than the state-of-the-art methods. Full article

Condition monitoring and fault diagnosis are topics of growing interest for improving the reliability of modern industrial systems. As critical structural components, anti-friction bearings often operate under harsh conditions and are contributing factors of system failures. Efforts have been cast on bearing diagnostics under the sensor fusion and machine learning framework, whilst challenges remain open on the identification of incipient faults. In this paper, exploiting multi-way representations and decompositions of measured vibration data, a novel band separation method based on the factorization of spectrogram tensors using the low rank approximated nonnegative Tucker decomposition (LRANTD) is proposed and applied to identify detailed fault signatures from the spectral, temporal, and spatial dimensions, flexible for extracting multi-sensor features and multi-dimensional correlations. With the proposed method, informative frequency bands of the latent vibrational components can be automatically extracted, in accordance with the inherent temporal patterns that can be conveniently fed for spectral analysis and fault discrimination. Furthermore, an improved cross-spectrum can be calculated from multi-channel vibrations via LRANTD with enhanced fault features. Based on the real-world vibration data of the accelerated bearing life tests, detailed experimental studies and thorough comparisons to the conventional benchmarks have verified the effectiveness of the reported diagnostic methodology. The proposed method significantly improves the presence of the bearing frequency peaks distinctly over the background noises in the spectrum and hence improves the bearing defect detection process. Full article

Unsupervised topic extraction is a vital step in automatically extracting concise contentual information from large text corpora. Existing topic extraction methods lack the capability of linking relations between these topics which would further help text understanding. Therefore we propose utilizing the Decomposition into Directional Components (DEDICOM) algorithm which provides a uniquely interpretable matrix factorization for symmetric and asymmetric square matrices and tensors. We constrain DEDICOM to row-stochasticity and non-negativity in order to factorize pointwise mutual information matrices and tensors of text corpora. We identify latent topic clusters and their relations within the vocabulary and simultaneously learn interpretable word embeddings. Further, we introduce multiple methods based on alternating gradient descent to efficiently train constrained DEDICOM algorithms. We evaluate the qualitative topic modeling and word embedding performance of our proposed methods on several datasets, including a novel New York Times news dataset, and demonstrate how the DEDICOM algorithm provides deeper text analysis than competing matrix factorization approaches. Full article

For a long time, various epidemics, such as lower respiratory infections and diarrheal diseases, have caused serious social losses and costs. Various methods for analyzing infectious disease occurrences have been proposed for effective prevention and proactive response to reduce such losses and costs. However, the results of the occurrence analyses were limited because numerous factors affect the outbreak of infectious diseases and there are complex interactions between these factors. To alleviate this limitation, we propose a cluster-based analysis scheme of infectious disease occurrences that can discover commonalities or differences between clusters by grouping elements with similar occurrence patterns. To do this, we collect and preprocess infectious disease occurrence data according to time, region, and disease. Then, we construct a tensor for the data and apply Tucker decomposition to extract latent features in the dimensions of time, region, and disease. Based on these latent features, we conduct k-means clustering and analyze the results for each dimension. To demonstrate the effectiveness of this scheme, we conduct a case study on data from South Korea and report some of the results. Full article

The incipient damages of mechanical equipment excite weak impulse vibration, which is hidden, almost unobservable, in the collected signal, making fault detection and failure prevention at the inchoate stage rather challenging. Traditional feature extraction techniques, such as bandpass filtering and time-frequency analysis, are suitable for matrix processing but challenged by the higher-order data. To tackle these problems, a novel method of impulse feature extraction for vibration signals, based on sparse non-negative tensor factorization is presented in this paper. Primarily, the phase space reconstruction and the short time Fourier transform are successively employed to convert the original signal into time-frequency distributions, which are further arranged into a three-way tensor to obtain a time-frequency multi-aspect array. The tensor is decomposed by sparse non-negative tensor factorization via hierarchical alternating least squares algorithm, after which the latent components are reconstructed from the factors by the inverse short time Fourier transform and eventually help extract the impulse feature through envelope analysis. For performance verification, the experimental analysis on the bearing datasets and the swashplate piston pump has confirmed the effectiveness of the proposed method. Comparisons to the traditional methods, including maximum correlated kurtosis deconvolution, singular value decomposition, and maximum spectrum kurtosis, also suggest its better performance of feature extraction. Full article

Hyperspectral and light detection and ranging (LiDAR) data fusion and classification has been an active research topic, and intensive studies have been made based on mathematical morphology. However, matrix-based concatenation of morphological features may not be so distinctive, compact, and optimal for classification. In this work, we propose a novel Coupled Higher-Order Tensor Factorization (CHOTF) model for hyperspectral and LiDAR data classification. The innovative contributions of our work are that we model different features as multiple third-order tensors, and we formulate a CHOTF model to jointly factorize those tensors. Firstly, third-order tensors are built based on spectral-spatial features extracted via attribute profiles (APs). Secondly, the CHOTF model is defined to jointly factorize the multiple higher-order tensors. Then, the latent features are generated by mode-n tensor-matrix product based on the shared and unshared factors. Lastly, classification is conducted by using sparse multinomial logistic regression (SMLR). Experimental results, conducted with two popular hyperspectral and LiDAR data sets collected over the University of Houston and the city of Trento, respectively, indicate that the proposed framework outperforms the other methods, i.e., different dimensionality-reduction-based methods, independent third-order tensor factorization based methods, and some recently proposed hyperspectral and LiDAR data fusion and classification methods. Full article

Short-term load forecasting is very important for power systems. The load is related to many factors which compose tensors. However, tensors cannot be input directly into most traditional forecasting models. This paper proposes a tensor partial least squares-neural network model (TPN) to forecast the power load. The model contains a tensor decomposition outer model and a nonlinear inner model. The outer model extracts common latent variables of tensor input and vector output and makes the residuals less than the threshold by iteration. The inner model determines the relationship between the latent variable matrix and the output by using a neural network. This model structure can preserve the information of tensors and the nonlinear features of the system. Three classical models, partial least squares (PLS), least squares support vector machine (LSSVM) and neural network (NN), are selected to compare the forecasting results. The results show that the proposed model is efficient for short-term load and daily load peak forecasting. Compared to PLS, LSSVM and NN, the TPN has the best forecasting accuracy. Full article