Traditional Principal Component Analysis Algorithm Research Articles

Improved secure PCA and LDA algorithms for intelligent computing in IoT‐to‐cloud setting

AbstractThe rapid development of new technologies such as artificial intelligence and big data analysis requires the simultaneous development of cloud computing technology. The application of IoT‐to‐cloud setting has been fully applied in various industry sectors, such as sensor‐cloud system which is composed of wireless sensor network and cloud computing technology. With the increasing amount and types of collected data, companies need to reduce the dimension of massive data in cloud servers for obtaining data analysis reports rapidly. Due to frequent cloud server data leaks, companies must adequately protect the privacy of some confidential data. To this end, we designed a dimension reduction method for ciphertext data in the sensor‐cloud system based on the CKKS encryption scheme, principal component analysis (PCA) and linear discriminant analysis (LDA) dimension reduction algorithm. As data cannot be directly calculated using traditional PCA and LDA algorithm after encryption, we add some interactive operations and iterative calculations to replace some steps in traditional algorithms. Finally, we select the classification dataset IRIS which is commonly used in machine learning, and screen out the best encryption and calculation parameters, and efficiently realize the dimension reduction method of ciphertext data through a large number of experiments.

Wine component tracing method based on near infrared spectrum fusion machine learning

An intelligent wine detection and traceability method based on infrared spec-troscopy and machine learning is proposed, in order to meet the needs of online rapid nondestructive testing of wine. On the basis of extracting infrared spectrum of wine, the principal component analysis (PCA) – support vector machine (SVM) model was modified by chemometrics. A total of 300 grape wine samples were collected from six production areas. The composition of the samples was analyzed by ultra performance liquid chromatography-quadrupole time-of-flight mass spectrometry (UPLC-Q-TOF-MS). According to the experimental results, indole, sulfacetamide and caffeine were selected as characteristics of different origins. Near infrared spectral wavelengths of wine samples were compressed between 900 and 2,500 nm. The ranges of 1,000 nm ~ 1,400 nm and 1,500 nm ~ 1800 nm were selected for PCA principal component analysis and key spectral wavelengths were extracted. The unsupervised learning model of SVM is used to classify and identify key spectral wavelengths. The experimental results show that the algorithm has higher classification accuracy than traditional PCA-LDA, PCA and other algorithms. The classification accuracy of the algorithm is improved from 98.3 to 99.75%. The improved PCA-SVM algorithm can achieve fast and loss-less source tracing of wine.

Process monitoring of abnormal working conditions in the zinc roasting process with an ALD-based LOF-PCA method

Timely and accurate detection of abnormal working conditions can ensure stability, improve production efficiency and reduce pollution of an industrial process. However, the production data of an industrial process has non-Gaussian and time-varying characteristics due to the diverse feed composition and complex reaction mechanisms. To address the above issue, an improved online principal component analysis (PCA) algorithm based on the selective model update is proposed in this study. First, considering the non-Gaussian nature of the process data, a local outlier factor-based (LOF) abnormality detection logic is used to replace the T2 and squared prediction error (SPE) statistics in traditional PCA algorithms. Then, to adapt to the time-varying characteristics of the process data, an approximate linear dependence (ALD) algorithm is used to evaluate the independent degree between the new sample and training samples. Only those samples containing new information are used to update the monitoring model, which can improve model performance and reduce the frequency of online updates. The zinc roasting process (ZRP) is used as an example to illustrate the proposed approach. Industrial data collected from a ZRP is used to demonstrate the performance of the ALD-based LOF-PCA method in the early detection of two typical abnormal working conditions in the ZRP.

Research on Microvideo Character Perception and Recognition Based on Target Detection Technology

With the rapid development of the Internet short-video platform, microvideo is increasingly attracting users because of its small traffic and time period. However, the content of microvideo on the Internet is complex, with illegal information. This research is an improved algorithm based on target detection technology, aiming at the characteristics of difficult character recognition in microvideo. The algorithm combines Gabor wavelet transform algorithm and 2D principal component analysis (PCA) algorithm and makes full use of video perception technology to realize character recognition in microvideo. Through MATLAB simulation analysis, it can be seen that the accuracy of the optimization algorithm proposed in this paper is 86.34% and the recognition time is 34.28s. Compared with the traditional PCA algorithm and artificial neural network (ANN) algorithm, the optimization algorithm proposed in this paper has better recognition rate and recognition efficiency.

Research on parallel principal component analysis based on ternary optical computer

Principal component analysis (PCA) is a widely used algorithm in the field of artificial intelligence and statistical analysis. When dealing with a large number of high-dimensional data, the traditional PCA algorithm has some shortcomings, such as slow processing speed, high power consumption, low computational efficiency. By taking advantage of several characteristics which include large number of data bits, easy-expansion of data bits, and parallel computing in ternary optical computer (TOC), the implementation scheme of parallel PCA algorithm is studied. Through the analysis of the traditional PCA algorithm, the parallel parts are found out and parallel design is carried out. Then, the elaborate implementation flow of the algorithm is given, and the clock cycle and hardware resources of the algorithm are also analyzed. Finally, the validity of the parallel scheme is proved by experiments. Compared with the traditional PCA algorithm, the PCA algorithm based on TOC has better time performance, operational efficiency, and gives full play to the advantage of ternary optical computer in processing complex computation.

Probabilistic predictive principal component analysis for spatially misaligned and high-dimensional air pollution data with missing observations.

Accurate predictions of pollutant concentrations at new locations are often of interest in air pollution studies on fine particulate matters (PM2.5), in which data is usually not measured at all study locations. PM2.5 is also a mixture of many different chemical components. Principal component analysis (PCA) can be incorporated to obtain lower-dimensional representative scores of such multi-pollutant data. Spatial prediction can then be used to estimate these scores at new locations. Recently developed predictive PCA modifies the traditional PCA algorithm to obtain scores with spatial structures that can be well predicted at unmeasured locations. However, these approaches require complete data, whereas multi-pollutant data tends to have complex missing patterns in practice. We propose probabilistic versions of predictive PCA which allow for flexible model-based imputation that can account for spatial information and subsequently improve the overall predictive performance.

Research on Genetic Algorithm Fault Diagnosis in Chemical Process

Aiming at the high false positive rate of traditional PCA (principal component analysis) algorithm in chemical process fault diagnosis, an algorithm combining GA (genetic algorithm) and PCA algorithm is proposed. Optimize the training samples of the PCA algorithm. In this method, when selecting training samples, not only the specific data is selected, but the selected data is optimized by the GA. Select a set of best performing data as a new training sample. Experiments using continuous stirred tank reactor data show that the improved PCA algorithm can effectively reduce the false positive rate compared with the traditional PCA algorithm.

An Improved Algorithm for Extracting Subtle Features of Radiation Source Individual Signals

With the rapid development of communication and information technology, it is difficult for traditional signal detection and recognition methods to accurately acquire and identify the intelligence under complex environments. In order to solve this problem, this paper proposes a subtle feature extraction and recognition algorithm for radiation source individual signals based on multidimensional hybrid features. Firstly, Hilbert transform was performed on the radiation source signals from 10 identical radio devices, and the subtle features of different radiation sources’ signals were extracted. Then, traditional principal component analysis (PCA) algorithm was used to extract and reduce the principal components of the extracted feature data sets. Aiming at the insufficiency of traditional PCA algorithm, an improved principal component analysis algorithm was proposed. At last, a gray relation algorithm was used to classify and identify the radiation source individual signals, and the recognition rate was calculated. Experimental results show that Hilbert transform combined with the improved PCA algorithm can achieve a recognition rate of 99.67% for the "fingerprint" features of radiation source individual signals under the signal-to-noise ratio (SNR) of 20 dB. Compared with the traditional algorithms, the recognition rate increased by 5.67%. Therefore, it provides a powerful theoretical basis for extracting subtle features of radiation source devices under complex electromagnetic environments.

Bayesian variable selection for globally sparse probabilistic PCA

Sparse versions of principal component analysis (PCA) have imposed themselves as simple, yet powerful ways of selecting relevant features of high-dimensional data in an unsupervised manner. However, when several sparse principal components are computed, the interpretation of the selected variables is difficult since each axis has its own sparsity pattern and has to be interpreted separately. To overcome this drawback, we propose a Bayesian procedure called globally sparse probabilistic PCA (GSPPCA) that allows to obtain several sparse components with the same sparsity pattern. This allows the practitioner to identify the original variables which are relevant to describe the data. To this end, using Roweis' probabilistic interpretation of PCA and a Gaussian prior on the loading matrix, we provide the first exact computation of the marginal likelihood of a Bayesian PCA model. To avoid the drawbacks of discrete model selection, a simple relaxation of this framework is presented. It allows to find a path of models using a variational expectation-maximization algorithm. The exact marginal likelihood is then maximized over this path. This approach is illustrated on real and synthetic data sets. In particular, using unlabeled microarray data, GSPPCA infers much more relevant gene subsets than traditional sparse PCA algorithms.

Parallel and Distributed Dimensionality Reduction of Hyperspectral Data on Cloud Computing Architectures

Cloud computing offers the possibility to store and process massive amounts of remotely sensed hyperspectral data in a distributed way. Dimensionality reduction is an important task in hyperspectral imaging, as hyperspectral data often contains redundancy that can be removed prior to analysis of the data in repositories. In this regard, the development of dimensionality reduction techniques in cloud computing environments can provide both efficient storage and preprocessing of the data. In this paper, we develop a parallel and distributed implementation of a widely used technique for hyperspectral dimensionality reduction: principal component analysis (PCA), based on cloud computing architectures. Our implementation utilizes Hadoop's distributed file system (HDFS) to realize distributed storage, uses Apache Spark as the computing engine, and is developed based on the map-reduce parallel model, taking full advantage of the high throughput access and high performance distributed computing capabilities of cloud computing environments. We first optimized the traditional PCA algorithm to be well suited for parallel and distributed computing, and then we implemented it on a real cloud computing architecture. Our experimental results, conducted using several hyperspectral datasets, reveal very high performance for the proposed distributed parallel method.

Low-velocity impact damage monitoring of a sandwich composite wing

Impact damage has been identified as a critical form of defect that constantly threatens the reliability of composite structures, such as those used in aircrafts and naval vessels. Low-energy impacts can introduce barely visible damage and cause structural degradation. Therefore, efficient structural health monitoring methods, which can accurately detect, quantify, and localize impact damage in complex composite structures, are required. In this article, a novel damage detection methodology is demonstrated for monitoring and quantifying the impact damage propagation. Statistical feature matrices, composed of features extracted from the time and frequency domains, are developed. Kernel principal component analysis is used to compress and classify the statistical feature matrices. Compared with traditional principal component analysis algorithm, kernel principal component analysis method shows better feature clustering and damage quantification capabilities. A new damage index, formulated using the Mahalanobis distance, is defined to quantify impact damage. The developed methodology has been validated using low-velocity impact experiments with a sandwich composite wing.

IT Performance Key Factors Extraction Based on PCA Selective Ensemble

The IT performance evaluation is a complex analysis system. Some key factors that can reflect the essential characteristics and be easy to acquire, should be extracted to help us to clarify evaluation, analysis and diagnoses. In this paper for the uncertainty of IT performance data, an algorithm of IT performance key factors extraction based on selective ensemble of principal component analysis (PCA) is proposed. Unlike the traditional PCA algorithm, PCA selective ensemble increase the difference degree between the main component and the others. Moreover it expands the selective ensemble in the application of unsupervised algorithm. The validity of the algorithm is verified through the analysis of the actual data.

Algorithms for accelerated convergence of adaptive PCA

We derive and discuss new adaptive algorithms for principal component analysis (PCA) that are shown to converge faster than the traditional PCA algorithms due to Oja, Sanger, and Xu. It is well known that traditional PCA algorithms that are derived by using gradient descent on an objective function are slow to converge. Furthermore, the convergence of these algorithms depends on appropriate choices of the gain sequences. Since online applications demand faster convergence and an automatic selection of gains, we present new adaptive algorithms to solve these problems. We first present an unconstrained objective function, which can be minimized to obtain the principal components. We derive adaptive algorithms from this objective function by using: 1) gradient descent; 2) steepest descent; 3) conjugate direction; and 4) Newton-Raphson methods. Although gradient descent produces Xu's LMSER algorithm, the steepest descent, conjugate direction, and Newton-Raphson methods produce new adaptive algorithms for PCA. We also provide a discussion on the landscape of the objective function, and present a global convergence proof of the adaptive gradient descent PCA algorithm using stochastic approximation theory. Extensive experiments with stationary and nonstationary multidimensional Gaussian sequences show faster convergence of the new algorithms over the traditional gradient descent methods.We also compare the steepest descent adaptive algorithm with state-of-the-art methods on stationary and nonstationary sequences.