Transformation of measurement uncertainties into low-dimensional feature vector space

Advances in technology allow the acquisition of data with high spatial and temporal resolution. These datasets are usually accompanied by estimates of the measurement uncertainty, which may be spatially or temporally varying and should be taken into consideration when making decisions based on the data. At the same time, various transformations are commonly implemented to reduce the dimensionality of the datasets for postprocessing or to extract significant features. However, the corresponding uncertainty is not usually represented in the low-dimensional or feature vector space. A method is proposed that maps the measurement uncertainty into the equivalent low-dimensional space with the aid of approximate Bayesian computation, resulting in a distribution that can be used to make statistical inferences. The method involves no assumptions about the probability distribution of the measurement error and is independent of the feature extraction process as demonstrated in three examples. In the first two examples, Chebyshev polynomials were used to analyse structural displacements and soil moisture measurements; while in the third, principal component analysis was used to decompose the global ocean temperature data. The uses of the method range from supporting decision-making in model validation or confirmation, model updating or calibration and tracking changes in condition, such as the characterization of the El Niño Southern Oscillation.

Upper Limb Motion Recognition Based on LLE-ELM Method of sEMG

The purpose of this paper is to develop an effective method to identify upper limb motions based on EMG signal for community rehabilitation. The method will be applicable to the control system in the rehabilitation equipment and provide objective data for quantitative assessment. The recognition goal sets of upper limb motion are constructed by decomposing assessment activities of activity of daily living scale (ADL). The recognition feature vector space is established by Variance (VAR), Mean Absolute Value (MAV), the fourth-order Autoregressive (the 4thAR), Zero Crossings (ZC’s), integral EMG (IEMG), and Root Mean Square (RMS), and various feature sets are extracted to get the best classification. Locally linear embedding (LLE) algorithm is used to reduce the computational complexity, and upper limb motions about shoulder, elbow and wrist are quickly classified through extreme leaving machine (ELM), which obtained the average accuracy of 98.14%, 98.61% and 94.77%, respectively. Furthermore, when ELM is compared with Back-propagation (BP) and Support vector machine (SVM), it has performed relatively better than BP and SVM. The results show that the validity of the mixed model for recognition is verified. In addition, the method can also provide a basis for recognition and assessment of the angle of upper limb joint in the next study.

A Kind of Self-Constructed Category Dictionary in Chinese Text Classification

By means of word-segmentation technology in TRIP database and each word that appears in a database will be account in detail, a kind of self-constructed category dictionary (SCC-dictionary) in Chinese text classification is proposed. For solving high dimension and sparseness problem exit in vector space model, a four-dimensional feature vector space model (FFVSM) is presented in this paper. With Support Vector Machine (SVM) algorithm, the text classifier is designed. Experimental results show there are two achievements in this paper: first, SCC-dictionary can replace the artificial-written dictionary with the same effect; second, the FFVSM will not only reduce the computing load than high-dimensional feature vector space model, but also keep the precision of classification as 86.87%, recall rate as 95.12%, and F1 value as 90.81%.

Feasibility of polarized all-sky imaging for aerosol characterization

In this study, we investigate the method of polarized all-sky imaging with respect to aerosol characterization. As a technical frame work for image processing and analysis, we propose Zernike polynomials to decompose the relative Stokes parameter distributions. This defines a suitable and efficient feature vector which is also appealing because it is independent of calibration, circumvents overexposure problems and is robust against pixel noise. We model the polarized radiances of realistic aerosol scenarios and construct the feature vector space of the key aerosol types in terms of the first two principal components describing the maximal variances. We show that, using this representation, aerosol types can be clearly distinguished with respect to fine and coarse mode dominated size distribution and index of refraction. We further investigate the individual influences of varying aerosol properties and solar zenith angle. This suggests that polarized all-sky imaging may improve aerosol characterization in combination with sky scanning radiometers of the existing Aerosol Robotic Network (AERONET) especially at low aerosol optical depths and low solar zenith angles.

Chinese Semantic Word-Formation Analysis Using FKP-MCO Classifier Based on Layered and Weighted GED

For Chinese information processing, automatic classification based on a large-scale database for different patterns of semantic word-formation can remarkably improve the identification for the unregistered word, automatic lexicography, semantic analysis, and other applications. However, owing to noise, anomalies, nonlinear characteristics, class-imbalance, and other uncertainties in word-formation data, the predictive performance of multi-criteria optimization classifier (MCOC) and other traditional data mining approaches will rapidly degenerate. In this paper we put forward an novel MCOC with fuzzification, kernel, and penalty factors (FKP-MCOC) based on layered and weighted graph edit distance (GED): firstly the layered and weighted GEDs between each semantic word-formation graph and prototype graphs are calculated and used for the dissimilarity measure, then the normalized GEDs are embedded into a new feature vector space, and FKP-MCO classifier based on the feature vector space is built for predicting the patterns of semantic word-formation. Our experimental results of Chinese word-formation analysis and comparison with support vector machine (SVM) show that our proposed approach can increase the separation of different patterns, the predictive performance of semantic pattern of a new compound word.