基于向量模型的文本检索若干问题研究

针对向量空间模型存在的几个问题进行了研究和探讨。在文本切分上提出了一种基于方差的切词方法;在对TF-IDF因子改进方面提出了位置参数加权方式;对向量维数压缩、词语同现等问题方面总结了解决的方法。

矩阵方程AXB+CYD=I解存在的条件

本文利用矩阵的广义奇异值分解给出了 AXB+ CYD=I解存在的条件及解的表达式 。

Study on algorithms of low SNR inversion of T_2 spectrum in NMR

The method of regularization factor selection determines stability and accuracy of the regularization method. A formula of regularization factor was proposed by analyzing the relationship between the improved SVD and regularization method. The improved SVD algorithm and regularization method could adapt to low SNR. The regularization method is better than the improved SVD in the case that SNR is below 30 and the improved SVD is better than the regularization method when SNR is higher than 30. The regularization method with the regularization factor proposed in this paper can be better applied into low SNR （5〈SNR） NMR logging. The numerical simulations and real NMR data process results indicated that the improved SVD algorithm and regularization method could adapt to the low signal to noise ratio and reduce the amount of computation greatly. These algorithms can be applied in NMR logging.

The analysis on IP signals in TEM response based on SVD

During transient electromagnetic method （TEM） exploration of a copper mine, we detected the late-channel TEM signal reversal phenomenon （a voltage change from positive to negative） caused by the influence of the induced polarization （IP） effect, which affects the depth and precision of the TEM detection. The conventional inversion method is inefficient because it is difficult to process the data. In this paper, the Cole-Cole model is adopted to analyze the effect of Dc resistivity, chargeability, time constant, and frequency exponent on the TEM response in an homogeneous half space model. Singular Value Decomposition （SVD） is used to invert the measured TEM data, and the Dc resistivity, chargeability, time constant and frequency exponent were extracted from the measured TEM data in the mine area. The extracted parameters are used for interpreting the detection result as a supplement. This reveals why the TEM data acquired in the area has a low resolution. It was found that the DC resistivity and time constant do not significantly change the results, however, the chargeability and frequency exponent have a significant effect. Because of these influences, the SVD method is more accurate than the conventional method in the apparent resistivity profile. The area of the copper mine is confined accurately based on the SVD inverted data. The conclusion has been verified by drill and is identical to the practical geological situation.

Contourlet watermarking algorithm based on Arnold scrambling and singular value decomposition

A new digital watermarking algorithm based on the contourlet transform is proposed to improve the robustness and anti-attack performances of digital watermarking. The algorithm uses the Arnold scrambling technique and the singular value decomposition （SVD） scheme. The Arnold scrambling technique is used to preprocess the watermark, and the SVD scheme is used to find the best suitable hiding points. After the contourlet transform of the carrier image, intermediate frequency sub-bands are decomposed to obtain the singularity values. Then the watermark bits scrambled in the Arnold rules are dispersedly embedded into the selected SVD points. Finally, the inverse contourlet transform is applied to obtain the carrier image with the watermark. In the extraction part, the watermark can be extracted by the semi-blind watermark extracting algorithm. Simulation results show that the proposed algorithm has better hiding and robustness performances than the traditional contourlet watermarking algorithm and the contourlet watermarking algorithm with SVD. Meanwhile, it has good robustness performances when the embedded watermark is attacked by Gaussian noise, salt- and-pepper noise, multiplicative noise, image scaling and image cutting attacks, etc. while security is ensured.

STRUCTURE OPTIMIZATION STRATEGY OF NORMALIZED RBF NETWORKS

Aimed at studying normali zed radial basis function network (NRBFN), this paper introduces the subtractiv e clustering based on a mountain function to construct the initial structure of NR BFN, adopts singular value decomposition (SVD) to analyze the relationship betwe en neural nodes of the hidden layer and singular values, cumulative contribution ratio, index vector, and optimizes the structure of NRBFN. Finally, simulation and performance comparison show that the algorithm is feasible and effective.

信噪比盲估计算法性能比较

为了实现非协作环境下的通信信号信噪比估计,本文在功率谱分析的基础上,提出了一种基于功率谱差分的信噪比盲估计算法,并将其与传统的基于奇异值分解的信噪比盲估计算法进行了比较。理论与仿真结果表明,与传统方法相比,本文提出的算法不但复杂度低,而且在低信噪比情形下仍具有较高的估计精确度和稳健性。

OPTIMIZATION WITH QUADRATIC CONSTRAINT IN APPLICATION OF STRUCTURE DYNAMIC MODEL UPDATING

A model updating optimization algorithm under quadratic constraints is applied to structure dynamic model updating. The updating problems of structure models are turned into the optimization with a quadratic constraint. Numerical method is presented by using singular value decomposition and an example is given. Compared with the other method, the method is efficient and feasible.

The Necessary and Sufficient Conditions for Two Kinds of Matrix Equtions Having Semipositive Subdefinite Solutions

Finding solutions of matrix equations in given set SR n×n is an active research field. Lots of investigation have done for these cases, where S are the sets of general or symmetric matrices and symmetric positive definite or sysmmetric semiposite definite matrices respectively . Recently, however, attentions are been paying to the situation for S to be the set of general(semi) positive definite matrices(called as semipositive subdefinite matrices below) . In this paper the necessary and sufficient conditions for the following two kinds of matrix equations having semipositive, subdefinite solutions are obtained. General solutions and symmetric solutions of the equations (Ⅰ) and (Ⅱ) have been considered in in detail.

Some studies on finding the nearest volume-preserving matrix

Finding the nearest volume-preserving matrix for a given matrix is studied. Amatrix equation is first obtained, which is a necessary condition for the solution to the problem.Then the equation is solved by the singular value decomposition method. Some additional results arealso provided to further characterize the solution. Using these results, a numerical algorithm isintroduced and a numerical test is given to illustrate the effectiveness of the algorithm.

Latent semantic analysis for query interfaces of deep web sites

To further enhance the efficiencies of search engines,achieving capabilities of searching,indexing and locating the information in the deep web,latent semantic analysis is a simple and effective way.Through the latent semantic analysis of the attributes in the query interfaces and the unique entrances of the deep web sites,the hidden semantic structure information can be retrieved and dimension reduction can be achieved to a certain extent.Using this semantic structure information,the contents in the site can be inferred and the similarity measures among sites in deep web can be revised.Experimental results show that latent semantic analysis revises and improves the semantic understanding of the query form in the deep web,which overcomes the shortcomings of the keyword-based methods.This approach can be used to effectively search the most similar site for any given site and to obtain a site list which conforms to the restrictions one specifies.

Direct linear discriminant analysis based on column pivoting QR decomposition and economic SVD

A direct linear discriminant analysis algorithm based on economic singular value decomposition （DLDA/ESVD） is proposed to address the computationally complex problem of the conventional DLDA algorithm, which directly uses ESVD to reduce dimension and extract eigenvectors corresponding to nonzero eigenvalues. Then a DLDA algorithm based on column pivoting orthogonal triangular （QR） decomposition and ESVD （DLDA/QR-ESVD） is proposed to improve the performance of the DLDA/ESVD algorithm by processing a high-dimensional low rank matrix, which uses column pivoting QR decomposition to reduce dimension and ESVD to extract eigenvectors corresponding to nonzero eigenvalues. The experimental results on ORL, FERET and YALE face databases show that the proposed two algorithms can achieve almost the same performance and outperform the conventional DLDA algorithm in terms of computational complexity and training time. In addition, the experimental results on random data matrices show that the DLDA/QR-ESVD algorithm achieves better performance than the DLDA/ESVD algorithm by processing high-dimensional low rank matrices.

Ground roll attenuation based on an empirical curvelet transform

In the field of seismic exploration, ground roll seriously affects the deep effective reflections from subsurface deep structures. Traditional curvelet transform cannot provide an adaptive basis function to achieve a suboptimal denoised result. In this paper, we propose a method based on empirical curvelet transform （ECT） for ground roll attenuation. Unlike the traditional curvelet transform, this method not only decomposes seismic data into multiscale and multi-directional components, but also provides an adaptive filter bank according to frequency content of seismic data itself. So, ground roll can be separated by using this method. However, as the frequency of reflection and ground roll components are close, we apply singular value decomposition （SVD） in the curvelet domain to differentiate the ground roll and reflection better. Examples of synthetic and field seismic data reveal that the proposed method based ECT performs better than the traditional curvelet method in terms of the suppression of ground roll.

SVD-LSSVM and its application in chemical pattern classification

Pattern classification is an important field in machine learning; least squares support vector machine (LSSVM) is a powerful tool for pattern classification. A new version of LSSVM, SVD-LSSVM, to save time of selecting hyper parameters for LSSVM is proposed. SVD-LSSVM is trained through singular value decomposition (SVD) of kernel matrix. Cross validation time of selecting hyper parameters can be saved because a new hyper parameter, singular value contribution rate (SVCR), replaces the penalty factor of LSSVM. Several UCI benchmarking data and the Olive classification problem were used to test SVD-LSSVM. The result showed that SVD-LSSVM has good performance in classification and saves time for cross validation.

Impulsive component extraction using shift-invariant dictionary learning and its application to gear-box bearing early fault diagnosis

The impulsive components induced by bearing faults are key features for assessing gear-box bearing faults.However,because of heavy background noise and the interferences of other vibrations,it is difficult to extract these impulsive components caused by faults,particularly early faults,from the measured vibration signals.To capture the high-level structure of impulsive components embedded in measured vibration signals,a dictionary learning method called shift-invariant K-means singular value decomposition(SI-K-SVD)dictionary learning is used to detect the early faults of gear-box bearings.Although SI-K-SVD is more flexible and adaptable than existing methods,the improper selection of two SI-K-SVD-related parameters,namely,the number of iterations and the pattern lengths,has an adverse influence on fault detection performance.Therefore,the sparsity of the envelope spectrum(SES)and the kurtosis of the envelope spectrum(KES)are used to select these two key parameters,respectively.SI-K-SVD with the two selected optimal parameter values,referred to as optimal parameter SI-K-SVD(OP-SI-K-SVD),is proposed to detect gear-box bearing faults.The proposed method is verified by both simulations and an experiment.Compared to the state-of-the-art methods,namely,empirical model decomposition,wavelet transform and K-SVD,OP-SI-K-SVD has better performance in diagnosing the early faults of a gear-box bearing.

Super-resolution reconstruction of synthetic-aperture radar image using adaptive-threshold singular value decomposition technique

A super-resolution reconstruction approach of （SVD） technique was presented, and its performance was radar image using an adaptive-threshold singular value decomposition analyzed, compared and assessed detailedly. First, radar imaging model and super-resolution reconstruction mechanism were outlined. Then, the adaptive-threshold SVD super-resolution algorithm, and its two key aspects, namely the determination method of point spread function （PSF） matrix T and the selection scheme of singular value threshold, were presented. Finally, the super-resolution algorithm was demonstrated successfully using the measured synthetic-aperture radar （SAR） images, and a Monte Carlo assessment was carried out to evaluate the performance of the algorithm by using the input/output signal-to-noise ratio （SNR）. Five versions of SVD algorithms, namely 1 ） using all singular values, 2） using the top 80% singular values, 3） using the top 50% singular values, 4） using the top 20% singular values and 5） using singular values s such that S2≥/max（s2）/rinsNR were tested. The experimental results indicate that when the singular value threshold is set as Smax/（rinSNR）1/2, the super-resolution algorithm provides a good compromise between too much noise and too much bias and has good reconstruction results.

A New Statistical Downscaling Scheme for Predicting Winter Precipitation in China

An effective statistical downscaling scheme was developed on the basis of singular value decomposition to predict boreal winter(December-January-February)precipitation over China.The variable geopotential height at 500 hPa(GH5)over East Asia,which was obtained from National Centers for Environmental Prediction’s Coupled Forecast System(NCEP CFS),was used as one predictor for the scheme.The preceding sea ice concentration(SIC)signal obtained from observed data over high latitudes of the Northern Hemisphere was chosen as an additional predictor.This downscaling scheme showed significantly improvement in predictability over the original CFS general circulation model(GCM)output in cross validation.The multi-year average spatial anomaly correlation coefficient increased from–0.03 to 0.31,and the downscaling temporal root-mean-square-error(RMSE)decreased significantly over that of the original CFS GCM for most China stations.Furthermore,large precipitation anomaly centers were reproduced with greater accuracy in the downscaling scheme than those in the original CFS GCM,and the anomaly correlation coefficient between the observation and downscaling results reached～0.6 in the winter of 2008.

Wavelet matrix transform for time-series similarity measurement

A time-series similarity measurement method based on wavelet and matrix transform was proposed,and its anti-noise ability,sensitivity and accuracy were discussed. The time-series sequences were compressed into wavelet subspace,and sample feature vector and orthogonal basics of sample time-series sequences were obtained by K-L transform. Then the inner product transform was carried out to project analyzed time-series sequence into orthogonal basics to gain analyzed feature vectors. The similarity was calculated between sample feature vector and analyzed feature vector by the Euclid distance. Taking fault wave of power electronic devices for example,the experimental results show that the proposed method has low dimension of feature vector,the anti-noise ability of proposed method is 30 times as large as that of plain wavelet method,the sensitivity of proposed method is 1/3 as large as that of plain wavelet method,and the accuracy of proposed method is higher than that of the wavelet singular value decomposition method. The proposed method can be applied in similarity matching and indexing for lager time series databases.

Short-term photovoltaic power prediction using combined K-SVD-OMP and KELM method

For photovoltaic power prediction,a kind of sparse representation modeling method using feature extraction techniques is proposed.Firstly,all these factors affecting the photovoltaic power output are regarded as the input data of the model.Next,the dictionary learning techniques using the K-mean singular value decomposition(K-SVD)algorithm and the orthogonal matching pursuit(OMP)algorithm are used to obtain the corresponding sparse encoding based on all the input data,i.e.the initial dictionary.Then,to build the global prediction model,the sparse coding vectors are used as the input of the model of the kernel extreme learning machine(KELM).Finally,to verify the effectiveness of the combined K-SVD-OMP and KELM method,the proposed method is applied to a instance of the photovoltaic power prediction.Compared with KELM,SVM and ELM under the same conditions,experimental results show that different combined sparse representation methods achieve better prediction results,among which the combined K-SVD-OMP and KELM method shows better prediction results and modeling accuracy.

An Analysis of the Causes of Decadal Variations of Rainfall in Shandong in Summer

The precipitation in Shandong in July, August as well as the whole summer (JJA) and the corresponding 500 hPa geopotential height fields are analyzed by means of the SVD (singular value decomposition) methodology. It is found that the general circulations in East Asia and the Western Pacific underwent decadal changes around 1979. The geopotential height, in particular over key areas like the South China Sea and the Philippines, increased after 1979. Corresponding to the changes in the geopotential height, the rainfall in Shandong started to decrease around 1979. The synthesized analysis shows that when the geopotential height at 500hPa level decreases in the key areas, the Western Pacific subtropical high shifts northward and an anticyclonic anomalous cell enforces the southerly flow over Shandong-Korea-Japan, Shandong could experience a wet period. A dry period is likely to occur when the geopotential height increases in these key areas, the subtropical high moves southward or expands westward to a great distance, and a cyclonic anomalous cell controls Shandong. Respective conceptual models for the causative mechanism are obtained for the cases of July, August and the whole summer (JJA) .