Dimensionality Reduction Scheme Research Articles

Hypothesis Tests for Structured Rank Correlation Matrices

Joint modeling of a large number of variables often requires dimension reduction strategies that lead to structural assumptions of the underlying correlation matrix, such as equal pair-wise correlations within subsets of variables. The underlying correlation matrix is thus of interest for both model specification and model validation. In this article, we develop tests of the hypothesis that the entries of the Kendall rank correlation matrix are linear combinations of a smaller number of parameters. The asymptotic behavior of the proposed test statistics is investigated both when the dimension is fixed and when it grows with the sample size. We pay special attention to the restricted hypothesis of partial exchangeability, which contains full exchangeability as a special case. We show that under partial exchangeability, the test statistics and their large-sample distributions simplify, which leads to computational advantages and better performance of the tests. We propose various scalable numerical strategies for implementation of the proposed procedures, investigate their behavior through simulations and power calculations under local alternatives, and demonstrate their use on a real dataset of mean sea levels at various geographical locations. Supplementary materials for this article are available online.

A time-triggered dimension reduction algorithm for the task assignment problem

The task assignment problem is fundamental in combinatorial optimisation, aiming at allocating one or more tasks to a number of agents while minimizing the total cost or maximizing the overall assignment benefit. This problem is known to be computationally hard since it is usually formulated as a mixed-integer programming problem. In this paper, we consider a novel Time-Triggered Dimension Reduction Algorithm (TTDRA). We propose convexification approaches to convexify both the constraints and the cost function for the general non-convex assignment problem. The computational speed is accelerated via our time-triggered dimension reduction scheme, where the triggering condition is designed based on the optimality tolerance and the convexity of the cost function. Optimality and computational efficiency are verified via numerical simulations on benchmark examples.

Item Response Theory True Score Equating for the Bifactor Model Under the Common-Item Nonequivalent Groups Design.

Applying item response theory (IRT) true score equating to multidimensional IRT models is not straightforward due to the one-to-many relationship between a true score and latent variables. Under the common-item nonequivalent groups design, the purpose of the current study was to introduce two IRT true score equating procedures that adopted different dimension reduction strategies for the bifactor model. The first procedure, which was referred to as the integration procedure, linked the latent variable scales for the bifactor model and integrated out the specific factors from the item response function of the bifactor model. Then, IRT true score equating was applied to the marginalized bifactor model. The second procedure, which was referred to as the PIRT-based procedure, projected the specific dimensions onto the general dimension to obtain a locally dependent unidimensional IRT (UIRT) model and linked the scales of the UIRT model, followed by the application of IRT true score equating to the locally dependent UIRT model. Equating results obtained with the two equating procedures along with those obtained with the unidimensional three-parameter logistic (3PL) model were compared using both simulated and real data. In general, the integration and PIRT-based procedures provided equating results that were not practically different. Furthermore, the equating results produced by the two bifactor-based procedures became more accurate than the results returned by the 3PL model as tests became more multidimensional.

Churn prediction in telecommunication industry using kernel Support Vector Machines.

In this age of fierce competitions, customer retention is one of the most important tasks for many companies. Many previous works proposed models to predict customer churn based on various machine learning techniques. In this study, we proposed an advanced churn prediction model using kernel Support Vector Machines (SVM) algorithm for a telecom company. Baseline SVM models were initially built to find out the most suitable kernel types and will be used to make comparison with other approaches. Dimension reduction strategies such as Sequential Forward Selection (SFS) and Sequential Backward Selection (SBS) were applied to the dataset to find out the most important features. Furthermore, resampling techniques to deal with imbalanced data such as Synthetic Minority Oversampling Technique Tomek Link (SMOTE Tomek) and Synthetic Minority Oversampling Technique ENN (SMOTE ENN) were used on the dataset. Using the above-mentioned techniques, we have obtained better results compared to those obtained from previous works, we achieved an F1-score and accuracy of 99% and 98.9% respectively.

Dimensionality Reduction and Extraction of Engineering Remote Sensing Data Based on Building Information Modeling and Geographical Information System

The high dimensionality of the modern remote sensing data of construction land makes it complicated to extract image data. This paper proposes a dimensionality reduction and extraction strategy for the remote sensing data of construction land, with the aid of building information modeling (BIM) and geographical information system (GIS). Firstly, the BIM was employed to reduce the size of the remote sensing data of construction land and to obtain the information of each element. Next, the remote sensing data of construction land were parsed, and the key BIM elements were extracted through semantic filtering. In addition, the remote sensing data were converted into a triangulated irregular network (TIN), which can be processed by the geographical information system (GIS). In the end, random projection was utilized to reduce the dimensionality and compress the remote sensing data, and realize the data extraction. Experimental results show that our approach can compress and extract the information from construction land images in the remote sensing data with a high accuracy.

k Nearest Neighbor Similarity Join Algorithm on High-Dimensional Data Using Novel Partitioning Strategy

k nearest neighbor similarity join on high-dimensional data has broad applications in many fields; several key challenges still exist for this task such as “curse of dimensionality” and large scale of the dataset. A new dimensionality reduction scheme is proposed by using random projection technique, then we design two novel partition strategies, including equal width partition strategy and distance split tree-based partition strategy, and finally, we propose k nearest neighbor join algorithm on high-dimensional data based on the above partition strategies. We conduct comprehensive experiments to test the performance of the proposed approaches, and the experimental results show that the proposed methods have good effectiveness and performance.

DGCyTOF: Deep learning with graphic cluster visualization to predict cell types of single cell mass cytometry data.

Single-cell mass cytometry, also known as cytometry by time of flight (CyTOF) is a powerful high-throughput technology that allows analysis of up to 50 protein markers per cell for the quantification and classification of single cells. Traditional manual gating utilized to identify new cell populations has been inadequate, inefficient, unreliable, and difficult to use, and no algorithms to identify both calibration and new cell populations has been well established. A deep learning with graphic cluster (DGCyTOF) visualization is developed as a new integrated embedding visualization approach in identifying canonical and new cell types. The DGCyTOF combines deep-learning classification and hierarchical stable-clustering methods to sequentially build a tri-layer construct for known cell types and the identification of new cell types. First, deep classification learning is constructed to distinguish calibration cell populations from all cells by softmax classification assignment under a probability threshold, and graph embedding clustering is then used to identify new cell populations sequentially. In the middle of two-layer, cell labels are automatically adjusted between new and unknown cell populations via a feedback loop using an iteration calibration system to reduce the rate of error in the identification of cell types, and a 3-dimensional (3D) visualization platform is finally developed to display the cell clusters with all cell-population types annotated. Utilizing two benchmark CyTOF databases comprising up to 43 million cells, we compared accuracy and speed in the identification of cell types among DGCyTOF, DeepCyTOF, and other technologies including dimension reduction with clustering, including Principal Component Analysis (PCA), Factor Analysis (FA), Independent Component Analysis (ICA), Isometric Feature Mapping (Isomap), t-distributed Stochastic Neighbor Embedding (t-SNE), and Uniform Manifold Approximation and Projection (UMAP) with k-means clustering and Gaussian mixture clustering. We observed the DGCyTOF represents a robust complete learning system with high accuracy, speed and visualization by eight measurement criteria. The DGCyTOF displayed F-scores of 0.9921 for CyTOF1 and 0.9992 for CyTOF2 datasets, whereas those scores were only 0.507 and 0.529 for the t-SNE+k-means; 0.565 and 0.59, for UMAP+ k-means. Comparison of DGCyTOF with t-SNE and UMAP visualization in accuracy demonstrated its approximately 35% superiority in predicting cell types. In addition, observation of cell-population distribution was more intuitive in the 3D visualization in DGCyTOF than t-SNE and UMAP visualization. The DGCyTOF model can automatically assign known labels to single cells with high accuracy using deep-learning classification assembling with traditional graph-clustering and dimension-reduction strategies. Guided by a calibration system, the model seeks optimal accuracy balance among calibration cell populations and unknown cell types, yielding a complete and robust learning system that is highly accurate in the identification of cell populations compared to results using other methods in the analysis of single-cell CyTOF data. Application of the DGCyTOF method to identify cell populations could be extended to the analysis of single-cell RNASeq data and other omics data.

Scheme dependence of two-loop hard thermal loop perturbation theory resummed SYM4,4 thermodynamics

The resummed thermodynamics of ${\cal N}=4$ supersymmetric Yang-Mills theory in four space-time dimensions ($\text{SYM}_{4,4}$) has been calculated previously to two loop order within hard thermal loop perturbation theory (HTLpt) using the canonical dimensional regularization (DRG) scheme. Herein, we revisit this calculation using the regularization by dimensional reduction (RDR) scheme. Since the RDR scheme manifestly preserves supersymmetry it is the preferred scheme, however, it is important to assess if and by how much the resummed perturbative results depend on the regularization scheme used. Comparing predictions for the scaled entropy obtained using the DRG and RDR schemes we find that for $\lambda \lesssim 6$ they are numerically very similar. We then compare the results obtained in both schemes with the strict perturbative result, which is accurate up to order $\lambda^2$, and a generalized Pad\'{e} approximant constructed from the known large-$N_c$ weak- and strong-coupling expansions. Comparing the strict perturbative expansion of the two-loop HTLpt result with the perturbative expansion to order $\lambda^2$, we find that both the DRG and RDR HTLpt calculations result in the same scheme-independent predictions for the coefficients at order $\lambda$, $\lambda^{3/2}$, and $\lambda^2 \log\lambda$, however, at order $\lambda^2$ there is a residual regularization scheme dependence.

Face Recognition Based on SRC Combined with Sparse Embedding Dimensionality Reduction

<p>Sparse representation-based classification (SRC) method has achieved good recognition results and shown strong robustness for face recognition, especially when the face image is affected by illumination variations, expression changes and occlusion. SRC method simply uses the training set as a dictionary to encode test samples. However, the high-dimensional training face data usually contain a large amount of redundant information, which will increase the complexity of this method. Therefore, the image dimensionality reduction procedure is separately performed by most of the existing methods before SRC is launched, but this may not be able to make full use of the discriminative information of the training samples. In this paper, based on the efficient SRC method, a sparse embedding dimensionality reduction strategy is combined with to achieve a face recognition method. For the proposed method, a projection matrix is used to project high-dimensional data into a low-dimensional space. At the same time, a discriminative coefficient constraint term in the objective function is introduced to reduce the classification residual of the sample through the distance relationship between all coefficients. Then the label information of the sample is used to iteratively update the projection matrix and coefficient representation. Finally, the test samples are projected into the low-dimensional space for classification. A large number of experimental results on three widely used face datasets show that the proposed method improves the discrimination of face images in low-dimensional space and can achieve better face recognition results.</p> <p> </p>

Differential contributions of serotonergic and dopaminergic functional connectivity to the phenomenology of LSD

RationaleLSD is the prototypical psychedelic. Despite a clear central role of the 5HT2a receptor in its mechanism of action, the contributions of additional receptors for which it shows affinity and agonist activity remain unclear.ObjectivesWe employed receptor-enriched analysis of functional connectivity by targets (REACT) to explore differences in functional connectivity (FC) associated with the distributions of the primary targets of LSD—the 5HT1a, 5HT1b, 5HT2a, D1 and D2 receptors.MethodsWe performed secondary analyses of an openly available dataset (N = 15) to estimate the LSD-induced alterations in receptor-enriched FC maps associated with these systems. Principal component analysis (PCA) was employed as a dimension reduction strategy for subjective experiences associated with LSD captured by the Altered States of Consciousness (ASC) questionnaire. Correlations between these principal components as well as VAS ratings of subjective effects with receptor-enriched FC were explored.ResultsCompared to placebo, LSD produced differences in FC when the analysis was enriched with each of the primary serotonergic and dopaminergic receptors. Altered receptor-enriched FC showed relationships with the subjective effects of LSD on conscious experience, with serotonergic and dopaminergic systems being predominantly associated with perceptual effects and perceived selfhood as well as cognition respectively. These relationships were dissociable, with different receptors showing the same relationships within, but not between, the serotonergic and dopaminergic systems.ConclusionsThese exploratory findings provide new insights into the pharmacology of LSD and highlight the need for additional investigation of non-5HT2a-mediated mechanisms.

Distributed Evolutionary Feature Selection for Big Data Processing

Feature selection has become a powerful dimensional reduction strategy and an effective tool in handling high-dimensional data. Feature selection aims to reduce the dimension of the feature space, to speed up and reduce the cost of the learning model and that by selecting the most relevant feature subset to data mining and machine learning tasks. The selection of optimal feature subset is an optimization problem that proved to be NP-hard. Metaheuristics are traditionally used to deal with NP-hard problems since they are well known for solving complex and real-world problems in reasonable period of time. Genetic algorithm (GA) is one of the most popular metaheuristics algorithms, which proved to be effective for an accurate feature selection task. However, in the last few decades, data have become progressively larger in both numbers of instances and features. This paradigm is being popularly termed as Big Data. With the tremendous growth of dataset sizes, most current feature selection algorithms and exceptionally GA become unscalable. To improve the scalability of a feature selection algorithm on big data, the distributed computing strategy is always adopted such as MapReduce model and Hadoop system. In this paper, we first present a review for the most recent works which handle the use of Parallel Genetic algorithm in large datasets. Then, we will propose a new Parallel Genetic algorithm based on the Coarse-grained parallelization model (island model). The parallelization of the process and the distribution of the partitioning of data will be performed using Hadoop system with an Amazon cluster. The performance and the scalability of the proposed method were theoretically and empirically compared to the existing feature selection methods when handling large-scale datasets and results confirm the effectiveness of our proposed method.

Novel Modeling and Vibration Analysis Method on a Helicopter Drive Train System

When it comes to analyzing the dynamic characteristics of a geared system, it is common practice to create an equivalent mathematical model for which the rotational velocity is the same for all the elements. In contrast to creating the equivalent system firstly, a novel strategy based on the transfer matrix method is developed, which can directly obtain the equation of motion according to the system topology figure. Actually, any kind of topology of the system, e.g., geared branch or geared closed-loop systems, can be decoupled into a combination of a series of independent chain systems via introducing the virtual branched-gear transfer matrix. The overall transfer equation of the system can be efficiently obtained because the transfer matrix method of a chain system is simply matrix multiplication. The transfer matrices of typical gear types, e.g., reduction gear and planetary gear, are derived. In particular, a simple and aesthetic transfer matrix of rotating beam is derived to solve the rotor blade flapping or lead-lag vibration problems. A dimension reduction strategy is also introduced to address the dimension unmatched issue, further reducing the scale of the overall transfer matrix of the system. At last, an application on modeling an analysis of the coupled flexible rotor blade/engine/tail rotor/drive train system of a helicopter is presented.

Solving High-Dimensional Optimal Stopping Problems Using Optimization Based Model Order Reduction

Solving optimal stopping problems by backward induction in high dimensions is often very complex since the computation of conditional expectations is required. Typically, such computations are based on regression, a method that suffers from the curse of dimensionality. Therefore, the objective of this paper is to establish dimension reduction schemes for large-scale asset price models and to solve related optimal stopping problems (e.g., Bermudan option pricing) in the reduced setting, where regression is feasible. The proposed algorithm is based on an error measure between linear stochastic differential equations. We establish optimality conditions for this error measure with respect to the reduced system coefficients and propose a particular method that satisfies these conditions up to a small deviation. We illustrate the benefit of our approach in several numerical experiments, in which Bermudan option prices are determined.

Reliable metrics‐based linear regression model for multilevel privacy measurement of face instances

Social networking sites have made photo sharing convenient and, consequently, users of those sites frequently share photos. The proliferation of these social images throughout the Internet has inadvertently exposed an increasing number of personally identifiable information, particularly from the visual information of the faces in the images. Most existing methods to de-identify faces results in an excessive loss of visual information. To solve this problem, this paper proposes a reliable metrics-based linear regression model for multilevel privacy measurement of face instances using the size information of face instances. The proposed privacy measurement model provides a novel instance-level-based solution to measure privacy levels. The paper also establishes a scientific relationship between the size information of face instances and their privacy levels, quantifying the degree to which face instances need to be de-identified. Finally, the paper proposes a novel k-Same-DT de-identification method to provide reliable metrics for a linear regression model. It is a real-time k-same-related de-identification algorithm that combines the PCA dimensionality reduction strategy and the Delaunay triangle-based face alignment algorithm.The proposed k-Same-DT method creates a de-identified face with a lower identifiable rate and higher structural similarity, and it can provide reliable de-identification metrics for privacy measurement. Experiments using the classical face dataset demonstrates the effectiveness of the proposed de-identification method. Extensive experiments and surveys on real-world social images were also conducted to verify the proposed measurement model.

Realization of In-Plane Polarized Light Detection Based on Bulk Photovoltaic Effect in A Polar Van Der Waals Crystal.

2D van der Waals materials are widely explored for in-plane polarized light detection owing to their distinctive in-plane anisotropic feature. However, most of these polarized light-sensitive devices root in their low symmetry of in-plane structure and work depending on external power sources, which greatly impedes the simplification of integrated devices and sustainable development. Bulk photovoltaic effect (BPVE), which separates photoexcited carriers via built-in electric field without an external power source and shows an angle-dependence on light polarization, is promising for self-powered polarized light detection to break through the restriction of in-plane anisotropy. Herein, a 2D lead-free van der Waals perovskite (Cl-PMA)2 CsAgBiBr7 (1, Cl-PMA = 4-Chlorobenzylamine) is successfully designed through the dimension reduction strategy. 1 exhibits BPVE with an open-circuited photovoltage up to ≈0.5V. Driven by the BPVE, self-powered in-plane polarized light detection with a large polarization ratio of 1.3 is obtained for 1. As far as it is known, the first in-plane polarized light detection in hybrid perovskites based on BPVE is realized here. This work highlights the strategy of designing lead-free hybrid perovskite with BPVE and opens an avenue for exploiting in-plane highly sensitive polarized light detection in 2D van der Waals materials.

Deciphering potential pharmacological mechanisms of Danhong injection to treat chronic stable angina based on drug response-related modules and genes

Ethnopharmacological relevanceDanhong injection (DHI), a traditional Chinese medicine (TCM) injection that has been widely used to treat coronary heart disease and angina pectoris. However, its underlying pharmacological mechanisms have not been fully elucidated. Not all patients benefit from DHI to the same extent. We attempted to explore the characteristics of potential therapeutic targets in different responsive populations. Aim of the studyThis study aimed to reveal the potential molecular mechanisms of DHI in treating chronic stable angina and identify potential therapeutic targets for DHI. Materials and methodsBased on a previous phase IV clinical trial of DHI in treating chronic stable angina, drug response modules were identified through structural entropy and similarity. Drug response-related genes were screened out based on the correlations between drug response module/module-related genes and clinical features and were assessed using a random forest model. Further validation was conducted using a hypoxia/reoxygenation (H/R) model. ResultsSeven DHI-related response modules were identified. Eight drug response-related genes were screened out, and principal component analysis showed that DHI responders were distinguished from responders in the control group based on their expression values. The combination of the two most important genes, SHC4 and PIP5K1P1, discriminated between responders and nonresponders with an area under the receiver operating characteristic curve (AUC) of 0.714; however, no significant difference was found in the AUC between the combination and a single gene. Reverse transcription–polymerase chain reaction showed that middle-dose DHI treatment significantly decreased SHC4 mRNA expression compared with that in the H/R group (P = 0.026), a finding consistent with our previous analysis of differentially expressed genes. ConclusionsDHI comprehensively exerted a therapeutic effect by acting on multiple response modules related to angina pectoris and drug response-related genes. Our findings indicate that the dimensionality reduction strategy based on the target network-drug response module-therapeutic targets can contribute to revealing the mechanism of action of TCM compounds and guiding precise clinical medication.

Quantization-Aware Processing for Massive MIMO Uplink Cloud RAN

As the deployment of a large number of antennas and more dense networks, the degradation brought by the finite fronthaul capacity needs to be taken into account in uplink cloud radio access networks (RANs). This letter proposes dimensionality reduction schemes to mitigate the degradation induced by quantization noise. The key idea is to transform observations at radio heads (RHs) in a reduced size, leading to less distorted quantized signals to be sent to the central processor (CP). By intensively using the quantization resource on these punctured observations, the decoding performance can be enhanced at the CP, especially for low-fronthaul capacity links. In the Gaussian source and Gaussian quantization setup, we prove that our scheme achieves a higher sum rate than conventional schemes. This gain is also confirmed by simulations.

Interactive Dimensionality Reduction for Comparative Analysis.

Finding the similarities and differences between groups of datasets is a fundamental analysis task. For high-dimensional data, dimensionality reduction (DR) methods are often used to find the characteristics of each group. However, existing DR methods provide limited capability and flexibility for such comparative analysis as each method is designed only for a narrow analysis target, such as identifying factors that most differentiate groups. This paper presents an interactive DR framework where we integrate our new DR method, called ULCA (unified linear comparative analysis), with an interactive visual interface. ULCA unifies two DR schemes, discriminant analysis and contrastive learning, to support various comparative analysis tasks. To provide flexibility for comparative analysis, we develop an optimization algorithm that enables analysts to interactively refine ULCA results. Additionally, the interactive visualization interface facilitates interpretation and refinement of the ULCA results. We evaluate ULCA and the optimization algorithm to show their efficiency as well as present multiple case studies using real-world datasets to demonstrate the usefulness of this framework.

An Optimal Dimension Reduction Strategy and Experimental Evaluation for Parkinson’s Disease Classification

An Optimal Dimension Reduction Strategy and Experimental Evaluation for Parkinson’s Disease Classification

An optimal dimension reduction strategy and experimental evaluation for Parkinson's disease classification

The amount of data streamed and generated through various healthcare systems is exponentially increasing day by day. Applying traditional data mining algorithms on this massive sized data to construct automated decision support systems is a tedious and time consuming task. In recent years, there has been increasing interest in the development of telediagnosis and telemonitoring systems for Parkinson's disease (PD). Parkinson's disease is a progressive neurodegenerative disease which affect the movement characteristics. PD patients commonly face vocal impairments during the early stages of the disease. This work proposes a computationally efficient method for dimension reduction and classification of healthcare related data. The devised framework is capable to deal with the data having discrete as well as continuous natured features. The experimental evaluation is performed on Parkinson's disease classification database (Sakar et al., 2018). The statistical performance metrices used are - validation and test accuracy, precision, recall, F1-score, etc. There will be computational complexity advantages when this reduced dimension data is further processed for modelling and building prediction system. In order to prove the optimality of proposed framework, comparative analysis is performed with the significant existing approaches.