Data Dimensionality Reduction Research Articles

PPLC: Data-driven offline learning approach for excavating control of cutter suction dredgers

Cutter suction dredgers (CSDs) play a very important role in the construction of ports, waterways and navigational channels. Currently, most of CSDs are mainly manipulated by human operators, and a large amount of instrument data needs to be monitored in real time in case of unforeseen accidents. In order to reduce the heavy workload of the operators, we propose a data-driven offline learning approach, named Preprocessing-Prediction-Learning Control (PPLC), for obtaining the optimal control policy of the excavating operation of CSDs. The proposed framework consists of three modules, i.e., a data preprocessing module, a dynamics prediction module realized by a Convolutional Neural Network (CNN), and a deep reinforcement learning based control module. The first module is responsible for filtering out irrelevant variables through correlation analysis and dimensionality reduction of raw data. The second module works as a state transition function that provides the dynamics prediction of the excavating operation of a CSD. To realize the learning control, the third module employs the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm to control the swing speed during the excavating operation. The simulation results show that the proposed framework can provide an effective and reliable solution to the automated excavating control of a CSD.

An adaptive indoor localization model based on Cauchy-PSO optimized BPNN

Received Signal Strength Indication (RSSI) fluctuates with the change of indoor noise, resulting in a large positioning error of the trained Back Propagation Neural Network (BPNN). An adaptive indoor positioning model based on Cauchy particle swarm optimization (Cauchy-PSO) BPNN is proposed to solve the problem. In the off-line training phase, the signal with less noise intensity acquired in a good environment is selected as the original training set in the localization phase. The variance of the received set of signals is used as a measure of the noise intensity of the current environment. In the localization phase, the variance of each set of signals received is calculated at equal intervals. If the variance of adjacent intervals differs significantly, the system adjusts the original training set data according to the current noise intensity and re-trains the BP model online. Meanwhile, the particle swarm optimization algorithm using Cauchy variance to optimize the BP network tends to fall into the disadvantage of local optimum. Considering that the collected fingerprint database may generate “high-dimensional disasters”, Principal Component Analysis (PCA) is used to select and downscale the features of the wireless Access Point (AP). The proposed adaptive localization model can be trained online. The improved Cauchy-PSO algorithm and data dimensionality reduction can further improve the localization accuracy and training speed of the BP model. The experimental results show that the adaptive indoor localization model has strong adaptive capability in a noise-varying environment.

DEEP ASSOCIATION RULE-BASED DIMENSIONALITY REDUCTION FOR BIG DATA ANALYTICS UNCOVERING COMPLEX PATTERNS AND IMPROVING EFFICIENCY

In the realm of big data analytics, reducing data dimensionality is crucial for efficient processing and gaining meaningful insights. This paper introduces a novel approach that leverages deep association rule mining to uncover intricate patterns and associations within large datasets. By incorporating deep learning models, such as neural networks, our method captures complex and non-linear relationships that traditional techniques often overlook. The proposed framework involves preprocessing the data, creating embedding representations, training a deep learning model, extracting association rules, filtering and selecting relevant rules, and applying dimensionality reduction. The selected association rules serve as the basis for reducing the dimensionality of the original dataset, either by removing irrelevant variables or consolidating them into higher-level features. The efficacy of our approach is evaluated through experiments, showcasing improved efficiency and the preservation of meaningful information. This research presents a promising avenue for reducing data dimensionality in big data analytics, enhancing the scalability and interpretability of analysis outcomes.

PCAfold 2.0—Novel tools and algorithms for low-dimensional manifold assessment and optimization

We describe an update to our open-source Python package, PCAfold, designed to help researchers generate, analyze and improve low-dimensional data manifolds. In the current version, PCAfold 2.0, we introduce novel tools and algorithms for assessing and optimizing low-dimensional manifolds. This includes a method that generates a “map” of local feature sizes that can help pinpoint researchers to problematic regions on a manifold. We introduce a novel cost function that characterizes the quality of a manifold topology with a single number. We develop two algorithms for feature selection based on principal component analysis (PCA) that use the cost function as an objective function to minimize. We introduce a quantity of interest (QoI)-aware dimensionality reduction strategy where data projections are computed using an artificial neural network and are directly optimized towards representing various projection-independent and projection-dependent QoIs. We also introduce an implementation of partition of unity networks (POUnets) for efficient reconstruction of QoIs from low-dimensional manifolds based on combining neural network classification with localized polynomial regression. Our software can be broadly applicable in all domains of science and engineering that aim to reduce data dimensionality, as well as in the fundamental research on representation learning.

Automatic Categorization of Multi Marketplace FMCGs Products using TF-IDF and PCA Features

The use of technology in line with the increasing number of internet users has caused a shift in the product sales ecosystem to the realm of electronic commerce (electronic commerce). A total of 73.23 customers made purchase transactions using e-commerce and the most purchased products were products classified as Fast Moving Consumer Goods (FMCGs). The increasingly varied FMCGs data coupled with the increasing number of marketplaces is felt to need to be broken down into specific groups. The process is carried out by analyzing e-commerce product information, especially product names, and descriptions. In this study, we propose an automatic categorization of multiple marketplaces using data from multiple marketplaces. Data text is converted into structured data with a series of preprocessing, and comprehensive experiments are carried out to see the extraction performance of variables including TF-IDF, BOW, and N-Gram. All three methods are used to validate text data sets with K-Means grouping results used with the help of PCA to reduce data dimensions. The results show that the performance of the TF-IDF algorithm with a dimension reduction value of 70 and the use of Python can provide optimal results for the percentage of grouping data.

Mutual Information Based Fusion Model (MIBFM): Mild Depression Recognition Using EEG and Pupil Area Signals

The detection of mild depression is conducive to the early intervention and treatment of depression. This study explored the fusion of electroencephalography (EEG) and pupil area signals to build an effective and convenient mild depression recognition model. We proposed Mutual Information Based Fusion Model (MIBFM), which innovatively used pupil area signals to select EEG electrodes based on mutual information. Then we extracted features from EEG and pupil area signals in different bands, and fused bimodal features using the denoising autoencoder. Experimental results showed that MIBFM could obtain the highest accuracy of 87.03%. And MIBFM exhibited better performance than other existing methods. Our findings validate the effectiveness of the use of pupil area as signals, which makes eye movement signals can be easily obtained using high resolution camera, and the EEG electrode selection scheme based on mutual information is also proved to be an applicable solution for data dimension reduction and multimodal complementary information screening. This study casts a new light for mild depression recognition using multimodal data of EEG and pupil area signals, and provides a theoretical basis for the development of portable and universal application systems.

Internet of Things for Diagnosis of Alzheimer’s Disease: A Multimodal Machine Learning Approach Based on Eye Movement Features

Alzheimer’s disease (AD) is a degenerative neurological disease that occurs in the elderly with typical symptoms of decline in cognition, manifested by eye movement behaviors. The key to AD treatment requires early detection of cognitive impairment, which relies on frequent medical screening. This paper proposes an Internet of Things (IoT) architecture constructed with eye-tracker (ET) nodes and cloud-based diagnosis enabled by machine learning (ML), which can provide convenient screening of oculomotor abnormalities and automatic identification of early-stage AD. The bespoke ET nodes collect three-dimensional (3-D) oculomotor responses from diverse stereo video stimulation trials and transmit data into a dedicated multimodal machine learning (MMML) algorithm in the cloud. The algorithm incorporates multimodal features extracted from diverse oculomotor types to enhance the classification accuracy and optimized data dimension reduction in feature fusion to improve the classifier’s performance. From evaluation, the proposed method can distinguish AD patients from the control group with 86% accuracy (ACC), 78% true positive rate (TPR), and 90% positive predictive value (PPV). The results confirm the effectiveness of our MMML algorithm in AD diagnosis with the fusion of multimodal oculomotor features and prove the feasibility of our IoT-powered eye-tracking solution for AD screening.

Homomorphic filtering in digital multichannel image processing

Purpose. The purpose of this article is to develop a preprocessing method for digital multispectral remote sensing images obtained through optical and infrared means in the electromagnetic spectrum. The method aims to ensure invariance with respect to positional formation conditions that determine spatial and radiometric resolution. By implementing homomorphic filtering in this method, we can significantly increase the informative value of processed imagery. Methodology. The problem solving, including the development of the spatial and radiometric resolution increase ways for multispectral geospatial data are based on the methods of brightness spatial distribution fusion, methods of data dimension reduction, de-correlation techniques and geometric correction of image spatial distributions. Findings. The method of preprocessing digital remote sensing data has been developed, which is a component of the methodology for identifying geometric shapes (GS) of objects in multi-channel aerospace images, allowing for a significant improvement in their recognition efficiency when noise is present. Originality. The method of preprocessing photogrammetric scenes using homomorphic filtering to enhance their informational significance is proposed. The method ensures invariance to positional conditions of fixation, improves the accuracy of further recognition, eliminates the drawbacks of known methods associated with the existence of parametric uncertainty dependence, the features of fixation of species information, low values of information indices of synthesized images, and computational process peculiarities. Practical value. Practical value is consists in improving of identification accuracy of objects GS in digital geospatial data, in significant increasing of raster multispectral images information value and in rising of automated image processing efficiency. The use of the method can greatly enhance the value and usefulness of multispectral photogrammetric images in a wide range of applications, from environmental monitoring to urban planning.

Construction of Building Energy Consumption Prediction Model Based on Multi-Optimization Model

This study explores the utilization of the Relevance Vector Machine (RVM) model, optimized using the Sparrow Search Algorithm (SSA), Simulate Anneal Arithmetic (SAA), Particle Swarm Optimization (PSO), and Bayesian Optimization Algorithm (BOP), to construct an energy dissipation model for public buildings in Wuhan City. Energy consumption data and influential factors were collected from 100 public buildings, yielding 15 input variables, including building area, personnel density, and supply air temperature. Energy dissipation served as the output scalar indicator. Through correlation analysis between input and output variables, it was found that building area, personnel density, and supply air temperature significantly impact energy dissipation in public buildings. Principal component analysis (PCA) was employed for data dimensionality reduction, selecting seven main influential factors along with energy dissipation values as the dataset for the predictive model. The BOP-RVM model showed superior performance in terms of R2 (0.9523), r (0.9761), and low RMSE (5.3894) and SI (0.056). These findings hold substantial practical value for accurately predicting building energy consumption and formulating effective energy management strategies.

PARETO PRINCIPLE TO IMPROVE ANOMALY DETECTION ON SOFTWARE ASSET MANAGEMENT

Software Asset Management (SAM) is essential for a large company with a centralized software distribution system. Unfortunately, the operationalization of SAM has various problems. These problems become even more complicated when it comes to managing big data. This research proposes the Pareto Principle to reduce data dimensions to solve the problem of large data sizes without losing dataset characteristics before conducting anomaly detection. This anomaly detection is mandatory to identify and reduce invalid data due to misalignment or misclassification. Therefore, this study compares the state-of-arts anomaly detection algorithms: I-forest, KNN, and SVM. As a result, we found that SVM is the best algorithm, with an accuracy rate of 78.4%. In addition, using Pareto for the total population and software name variation effectively reduces the number of observations to 20% data instances with only 16.5% features without compromising the dataset characteristics. In the best algorithm based on experimental results, the use of Pareto increases accuracy by 3.2% and processing time efficiency by 20%.

On the Effect of Emotion Identification from Limited Translated Text Samples Using Computational Intelligence

Emotion identification from text data has recently gained focus of the research community. This has multiple utilities in an assortment of domains. Many times, the original text is written in a different language and the end-user translates it to her native language using online utilities. Therefore, this paper presents a framework to detect emotions on translated text data in four different languages. The source language is English, whereas the four target languages include Chinese, French, German, and Spanish. Computational intelligence (CI) techniques are applied to extract features, dimensionality reduction, and classification of data into five basic classes of emotions. Results show that when English text is translated to French, classification accuracy is higher than others, i.e., 99.04%. Whereas, when the same is translated to Chinese language, its detection rate is lowest among target languages. It is concluded that emotions remain preserved after translation to some extent. Framework consists of TFIDF features. PCA and Discriminant Analysis perform good to detect emotions from translated data.

A privacy-aware framework for detecting cyber attacks on internet of medical things systems using data fusion and quantum deep learning

Internet of Medical Things (IoMT) devices and systems are often designed without adequate security, leaving them highly susceptible to cyber threats. Unlike other IoT applications and devices, cyber attacks against IoMT devices will pose significant risks to patient safety. As a result of the progress made in attack detection through machine learning, numerous applications of machine learning are now involved in IoMT. However, how to securely collect and fuse the data from heterogeneous IoMT devices and systems and ensure efficient cyber-attack detection while avoiding the privacy disclosure of the dataset remains challenging for IoMT systems. To fill this gap, a new privacy-aware framework is advised for storing the collected IoMT data in distributed cloud nodes and providing data fusion of heterogeneous IoMT data using differential privacy and deep learning and efficient attack detection using quantum deep learning with maintaining data privacy. Specifically, the proposed privacy-aware framework provides sensitive data privacy protection at multi-levels starting from storing data in distributed cloud nodes with privacy-preserving and then fusing IoMT heterogeneous data using deep contractive autoencoder with differential privacy for protecting the sensitive data through the learning process and then using the output of data fusion which is reduced data dimension as input for a quantum neural network for identifying cyber-attacks. Our experimental results indicate that the proposed framework is highly effective in detecting cyber-attacks.

Quantitative diagnosis of loose piston rod threads in reciprocating compressors for hydrogen storage and transport

Fracture of the piston rod yields catastrophic accidents in reciprocating hydrogen compressors, resulting in costly and time-consuming maintenance and developing the latest challenge to hydrogen safety and economy. The fracture-prone section is mostly located at the connecting thread of the crosshead or of the piston, mainly due to thread loosening and hereby insufficient preload followed by fatigue and fracture. Here, a novel quantitative diagnosis method for thread loosening of hydrogen reciprocating compressor piston rods was proposed. First, a finite element model of a single-throw reciprocating compressor was established to analyse the dynamic behaviour of the operating posture and load pattern at the crosshead and piston rod connecting thread. Further, the impact events triggered by crosshead displacement and rod load reversal events were identified to enhance the interpretability of the signal, thereby obtaining a loose-sensitive frequency band of 2–15 kHz and specific phase angle features. Next, the multi-band feature spectrum was formed using variational modal decomposition and data dimensionality reduction based on the Hilbert transform, which demonstrates satisfactory performance in the feature reservation as well as spectrum compression of acceleration signals in both the time and frequency domain. The proposed methodology was validated by experiments with various levels of thread loosening. The result indicated the multi-band feature spectrum feature was effective in quantitative loosening diagnosis with accurate and stable classification.

Aerodynamic Optimization Framework for a Three-Dimensional Nacelle Based on Deep Manifold Learning-Assisted Geometric Multiple Dimensionality Reduction

As a core component of an aero-engine, the aerodynamic performance of the nacelle is essential for the overall performance of an aircraft. However, the direct design of a three-dimensional (3D) nacelle is limited by the complex design space consisting of different cross-section profiles and irregular circumferential curves. The deep manifold learning-assisted geometric multiple dimensionality reduction method combines autoencoders (AE) with strong capabilities for non-linear data dimensionality reduction and class function/shape function transformation (CST). A novel geometric dimensionality reduction method is developed to address the typical constraints of nacelle parameterization. Low-dimensional latent variables are extracted from the high-dimensional design space to achieve a parametric representation of 3D nacelle manifolds. Compared with traditional parametric methods, the proposed geometric dimensionality reduction method improves the accuracy and efficiency of geometric reconstruction and aerodynamic evaluation. A multi-objective optimization framework is proposed based on deep manifold learning to increase the efficiency of 3D nacelle design. The Pareto front curves under drag divergence constraints reveal the correlation between the geometry distribution and the surface isentropic Mach number distribution of 3D nacelles. This paper demonstrates the feasibility of the proposed geometric dimensionality reduction method for direct multi-objective optimization of 3D nacelles.

Classification of Lubricating Oil Types Using Mid-Infrared Spectroscopy Combined with Linear Discriminant Analysis–Support Vector Machine Algorithm

To realize the classification of lubricating oil types using mid-infrared (MIR) spectroscopy, linear discriminant analysis (LDA) was used for the dimensionality reduction of spectrum data, and the classification model was established based on the support vector machine (SVM). The spectra of the samples were pre-processed by interval selection, Savitzky–Golay smoothing, multiple scattering correction, and normalization. The Kennard–Stone algorithm (K/S) was used to construct the calibration and validation sets. The percentage of correct classification (%CC) was used to evaluate the model. This study compared the results obtained with several chemometric methods: PLS-DA, LDA, principal component analysis (PCA)-SVM, and LDA-SVM in MIR spectroscopy applications. In both calibration and verification sets, the LDA-SVM model achieved 100% favorable results. The PLS-DA analysis performed poorly. The cyclic resistance ratio (CRR) of the calibration set was classified via the LDA and PCA-SVM analysis as 100%, but the CRR of the verification set was not as good. The LDA-SVM model was superior to the other three models; it exhibited good robustness and strong generalization ability, providing a new method for the classification of lubricating oil types by MIR spectroscopy.

A Multistep Interval Prediction Method Combining Environmental Variables and Attention Mechanism for Egg Production Rate

The egg production rate is a crucial metric in animal breeding, subject to biological and environmental influences and exhibits characteristics of small sample sizes and non-linearity. Currently, egg production rate prediction research predominantly focuses on single-step point prediction, lacking multistep and interval prediction exploration. To bridge these gaps, this study proposes a recursive, multistep interval prediction method for egg production rates, integrating environmental variables and attention mechanisms. Initially, this study employed three gradient boosting tree models (XGBoost, LightGBM, CatBoost) and the recursive feature elimination (RFE) method to select critical environmental variables and reduce data dimensionality. Subsequently, by scaling the time scale of important environmental variables and utilizing the variational modal decomposition improved by the grey wolf optimization (GWO-VMD) method for time-series decomposition, the volume of important environmental variable data is augmented and its complexity is reduced. Applying the long short-term memory (LSTM) neural network to obtain direct multistep predictions on IMFs, the predicted outcomes are averaged daily to yield the environmental variables for the upcoming two days. Finally, a multistep interval prediction model based on Seq2seq-Attention and Gaussian distribution is proposed in this study, and parameter optimization is carried out using the multi-objective grey wolf optimization algorithm (MOGWO). By inputting the historical egg production rate data and environmental variables into the proposed model, it is possible to achieve multistep point and interval prediction of egg production rates. This method was applied to analyze a dataset of egg production rates of waterfowl. The study demonstrated the feasibility of the recursive multistep prediction approach combined with environmental variables and guides egg production estimation and environmental regulation in animal husbandry.

Low-rank Representation with Adaptive Dimensionality Reduction via Manifold Optimization for Clustering

The dimensionality reduction techniques are often used to reduce data dimensionality for computational efficiency or other purposes in existing low-rank representation (LRR)-based methods. However, the two steps of dimensionality reduction and learning low-rank representation coefficients are implemented in an independent way; thus, the adaptability of representation coefficients to the original data space may not be guaranteed. This article proposes a novel model, i.e., low-rank representation with adaptive dimensionality reduction (LRRARD) via manifold optimization for clustering, where dimensionality reduction and learning low-rank representation coefficients are integrated into a unified framework. This model introduces a low-dimensional projection matrix to find the projection that best fits the original data space. And the low-dimensional projection matrix and the low-rank representation coefficients interact with each other to simultaneously obtain the best projection matrix and representation coefficients. In addition, a manifold optimization method is employed to obtain the optimal projection matrix, which is an unconstrained optimization method in a constrained search space. The experimental results on several real datasets demonstrate the superiority of our proposed method.

Deciphering ligand–receptor-mediated intercellular communication based on ensemble deep learning and the joint scoring strategy from single-cell transcriptomic data

Background:Cell–cell communication in a tumor microenvironment is vital to tumorigenesis, tumor progression and therapy. Intercellular communication inference helps understand molecular mechanisms of tumor growth, progression and metastasis. Methods:Focusing on ligand–receptor co-expressions, in this study, we developed an ensemble deep learning framework, CellComNet, to decipher ligand–receptor-mediated cell–cell communication from single-cell transcriptomic data. First, credible LRIs are captured by integrating data arrangement, feature extraction, dimension reduction, and LRI classification based on an ensemble of heterogeneous Newton boosting machine and deep neural network. Next, known and identified LRIs are screened based on single-cell RNA sequencing (scRNA-seq) data in certain tissues. Finally, cell–cell communication is inferred by incorporating scRNA-seq data, the screened LRIs, a joint scoring strategy that combines expression thresholding and expression product of ligands and receptors. Results:The proposed CellComNet framework was compared with four competing protein–protein interaction prediction models (PIPR, XGBoost, DNNXGB, and OR-RCNN) and obtained the best AUCs and AUPRs on four LRI datasets, elucidating the optimal LRI classification ability. CellComNet was further applied to analyze intercellular communication in human melanoma and head and neck squamous cell carcinoma (HNSCC) tissues. The results demonstrate that cancer-associated fibroblasts highly communicate with melanoma cells and endothelial cells strong communicate with HNSCC cells. Conclusions:The proposed CellComNet framework efficiently identified credible LRIs and significantly improved cell–cell communication inference performance. We anticipate that CellComNet can contribute to anticancer drug design and tumor-targeted therapy.

Rapid detection of lignin content in corn straw based on Laplacian Eigenmaps

Lignin is an essential components of corn stalk and has a wide range of application. To realize the rapid detection of lignin content in corn straw and increase the detection accuracy, we initially preprocess the gathered corn straw near-infrared spectral data with Standard Normal Variable transformation (SNV). After that, the nonlinear dimensionality reduction method Laplacian Eigenmaps (LE) and Local Tangent Space Alignment (LTSA) are applied independently to reduce the dimension of spectral data. principal component analysis (PCA), a linear dimensionality reduction method, is also utilized for spectral data dimensionality reduction. Finally, models for Partial Least Squares Regression (PLSR) and Support Vector Regression (SVR) are constructed. According to the model findings, LE-SVR model offers the best prediction accuracy and stability. The determination coefficient and root mean square error of the training and tests sets are 97.17%, 0.1875 and 96.25%, 0.2718 respectively. Furthermore, the relative analytical error is 5.0776. In addition, the study findings show that the number of neighbor points k has no discernible effect on the model performance. According to the findings, nonlinear modeling using LE-SVR for NIR spectral data of corn stover can lower model complexity, while improving model prediction accuracy and stability. NIR spectroscopy may be used to determine lignin content in corn straw. At the same time, the technique in this work offers a novel approach for the rapidly detecting lignin content in other crops straw.

Overview of Logistics Demand Forecasting Methods

Accurate forecasting of logistics demand is of great theoretical significance and practical application value for the formulation of national policies and the satisfaction of actual demand in the logistics industry. From the modeling form, the existing logistics demand forecasting methods are divided into four categories: single traditional forecasting method, single intelligent forecasting method, combined forecasting method and mixed forecasting method. Among them, single traditional forecasting methods mainly include simple time series method, regression analysis, mathematical and statistical methods, etc.; single intelligent forecasting methods mainly involve gray forecasting method, neural network, support vector machine and their improved forms; combined forecasting methods are mainly summarized into three combined forms: linear combination of single forecasting results, nonlinear combination of single forecasting results, modified single forecasting results; mixed forecasting methods are mainly summarized into three hybrid forms: hybrid intelligent optimization algorithms with single prediction methods, hybrid data dimensionality reduction techniques with intelligent prediction methods, and hybrid data mining techniques with intelligent prediction methods. The four major types of forecasting methods are reviewed, and each forecasting model in the four major types of methods is evaluated in terms of modeling principles, advantages and disadvantages, and applicability, in order to find forecasting methods suitable for different logistics demand forecasting tasks for logistics demand researchers.