A fuzzy method to learn text classifier from labeled and unlabeled examples

In text classification, labeling documents is a tedious and costly task, as it would consume a lot of expert time. On the other hand, it usually is easier to obtain a lot of unlabeled documents, with the help of some tools like Digital Library, Crawler Programs, and Searching Engine. To learn text classifier from labeled and unlabeled examples, a novel fuzzy method is proposed. Firstly, a Seeded Fuzzy c-means Clustering algorithm is proposed to learn fuzzy clusters from a set of labeled and unlabeled examples. Secondly, based on the resulting fuzzy clusters, some examples with high confidence are selected to construct training data set. Finally,the constructed training data set is used to train Fuzzy Support Vector Machine, and get text classifier. Empirical results on two benchmark datasets indicate that, by incorporating unlabeled examples into learning process,the method performs significantly better than FSVM trained with a small number of labeled examples only. Also, the method proposed performs at least as well as the related method-EM with Naieve Bayes. One advantage of the method proposed is that it does not rely on any parametric assumptions about the data as it is usually the case with generative methods widely used in semi-supervised learning.     

A Semi-Supervised WLAN Indoor Localization Method Based on l1-Graph Algorithm

For indoor location estimation based on received signal strength( RSS) in wireless local area networks( WLAN),in order to reduce the influence of noise on the positioning accuracy,a large number of RSS should be collected in offline phase. Therefore,collecting training data with positioning information is time consuming which becomes the bottleneck of WLAN indoor localization. In this paper,the traditional semisupervised learning method based on k-NN and ε-NN graph for reducing collection workload of offline phase are analyzed,and the result shows that the k-NN or ε-NN graph are sensitive to data noise,which limit the performance of semi-supervised learning WLAN indoor localization system. Aiming at the above problem,it proposes a l1-graph-algorithm-based semi-supervised learning( LG-SSL) indoor localization method in which the graph is built by l1-norm algorithm. In our system,it firstly labels the unlabeled data using LG-SSL and labeled data to build the Radio Map in offline training phase,and then uses LG-SSL to estimate user's location in online phase. Extensive experimental results show that,benefit from the robustness to noise and sparsity ofl1-graph,LG-SSL exhibits superior performance by effectively reducing the collection workload in offline phase and improving localization accuracy in online phase.     

Novel Apriori-Based Multi-Label Learning Algorithm by Exploiting Coupled Label Relationship

It is a key challenge to exploit the label coupling relationship in multi-label classification (MLC) problems. Most previous work focused on label pairwise relations, in which generally only global statistical information is used to analyze the coupled label relationship. In this work, firstly Bayesian and hypothesis testing methods are applied to predict the label set size of testing samples within their k nearest neighbor samples, which combines global and local statistical information, and then apriori algorithm is used to mine the label coupling relationship among multiple labels rather than pairwise labels, which can exploit the label coupling relations more accurately and comprehensively. The experimental results on text, biology and audio datasets shown that, compared with the state-of-the-art algorithm, the proposed algorithm can obtain better performance on 5 common criteria.     

Layer-Constrained Triangulated Irregular Network Algorithm Based on Ground Penetrating Radar Data and Its Application

In this paper,a layer-constrained triangulated irregular network (LC-TIN) algorithm is proposed for three-dimensional (3D) modelling,and applied to construct a 3D model for geological disease information based on ground penetrating radar (GPR) data.Compared with the traditional TIN algorithm,the LC-TIN algorithm introduced a layer constraint to the discrete data points during the 3D modelling process,and it can dynamically construct networks from layer to layer and implement 3D modelling for arbitrary shapes with high precision.The experimental results validated this method,the proposed algorithm not only can maintain the rationality of triangulation network,but also can obtain a good generation speed.In addition,the algorithm is also introduced to our self-developed 3D visualization platform,which utilized GPR data to model geological diseases.Therefore the feasibility of the algorithm is verified in the practical application.     

Layer-Constrained Triangulated Irregular Network Algorithm Based on Ground Penetrating Radar Data and Its Application

In this paper,a layer-constrained triangulated irregular network (LC-TIN) algorithm is proposed for three-dimensional (3D) modelling,and applied to construct a 3 D model for geological disease information based on ground penetrating radar (GPR) data.Compared with the traditional TIN algorithm,the LC-TIN algorithm introduced a layer constraint to the discrete data points during the 3D modelling process,and it can dynamically construct networks from layer to layer and implement 3D modelling for arbitrary shapes with high precision.The experimental results validated this method,the proposed algorithm not only can maintain the rationality of triangulation network,but also can obtain a good generation speed.In addition,the algorithm is also introduced to our self-developed 3D visualization platform,which utilized GPR data to model geological diseases.Therefore the feasibility of the algorithm is verified in the practical application.     

A Semi-Supervised WLAN Indoor Localization Method Based on 1-Graph Algorithm

For indoor location estimation based on received signal strength (RSS) in wireless local area networks (WLAN), in order to reduce the influence of noise on the positioning accuracy, a large number of RSS should be collected in offline phase. Therefore, collecting training data with positioning information is time consuming which becomes the bottleneck of WLAN indoor localization. In this paper, the traditional semi-supervised learning method based on k-NN and ε-NN graph for reducing collection workload of offline phase are analyzed, and the result shows that the k-NN or ε-NN graph are sensitive to data noise, which limit the performance of semi-supervised learning WLAN indoor localization system. Aiming at the above problem, it proposes a 1-graph-algorithm-based semi-supervised learning (LG-SSL) indoor localization method in which the graph is built by 1-norm algorithm. In our system, it firstly labels the unlabeled data using LG-SSL and labeled data to build the Radio Map in offline training phase, and then uses LG-SSL to estimate user’s location in online phase. Extensive experimental results show that, benefit from the robustness to noise and sparsity of 1-graph, LG-SSL exhibits superior performance by effectively reducing the collection workload in offline phase and improving localization accuracy in online phase.     

Adaptive Neighboring Selection Algorithm Based on Curvature Prediction in Manifold Learning

Recently manifold learning algorithm for dimensionality reduction attracts more and more interests, and various linear and nonlinear, global and local algorithms are proposed. The key step of manifold learning algorithm is the neighboring region selection. However, so far for the references we know, few of which propose a generally accepted algorithm to well select the neighboring region. So in this paper, we propose an adaptive neighboring selection algorithm, which successfully applies the LLE and ISOMAP algorithms in the test. It is an algorithm that can find the optimal K nearest neighbors of the data points on the manifold. And the theoretical basis of the algorithm is the approximated curvature of the data point on the manifold. Based on Riemann Geometry, Jacob matrix is a proper mathematical concept to predict the approximated curvature. By verifying the proposed algorithm on embedding Swiss roll from R3 to R2 based on LLE and ISOMAP algorithm, the simulation results show that the proposed adaptive neighboring selection algorithm is feasible and able to find the optimal value of K, making the residual variance relatively small and better visualization of the results. By quantitative analysis, the embedding quality measured by residual variance is increased 45.45％ after using the proposed algorithm in LLE.     

Audio-visual emotion recognition with multilayer boosted HMM

Emotion recognition has become an important task of modern human-computer interaction. A multilayer boosted HMM ( MBHMM) classifier for automatic audio-visual emotion recognition is presented in this paper. A modified Baum-Welch algorithm is proposed for component HMM learning and adaptive boosting ( AdaBoost) is used to train ensemble classifiers for different layers ( cues) . Except for the first layer,the initial weights of training samples in current layer are decided by recognition results of the ensemble classifier in the upper layer. Thus the training procedure using current cue can focus more on the difficult samples according to the previous cue. Our MBHMM classifier is combined by these ensemble classifiers and takes advantage of the complementary information from multiple cues and modalities. Experimental results on audio-visual emotion data collected in Wizard of Oz scenarios and labeled under two types of emotion category sets demonstrate that our approach is effective and promising.     

Coupled Attribute Similarity Learning on Categorical Data for Multi-Label Classification

In this paper a novel coupled attribute similarity learning method is proposed with the basis on the multi-label categorical data (CASonMLCD). The CASonMLCD method not only computes the correlations between different attributes and multi-label sets using information gain, which can be regarded as the important degree of each attribute in the attribute learning method, but also further analyzes the intra-coupled and inter-coupled interactions between an attribute value pair for different attributes and multiple labels. The paper compared the CASonMLCD method with the OF distance and Jaccard similarity, which is based on the MLKNN algorithm according to 5 common evaluation criteria. The experiment results demonstrated that the CASonMLCD method can mine the similarity relationship more accurately and comprehensively, it can obtain better performance than compared methods.     

A self region based real-valued negative selection algorithm

Point-wise negative selection algorithms, which generate their detector sets based on point of self data, have lower training efficiency and detection rate. To solve this problem, a self region based real-valued negative selection algorithm is presented. In this new approach, the continuous self region is defined by the collection of self data, the partial training takes place at the training stage according to both the radius of self region and the cosine distance between gravity of the self region and detector candidate, and variable detectors in the self region are deployed. The algorithm is tested using the triangle shape of self region in the 2-D complement space and KDD CUP 1999 data set. Results show that, more information can be provided when the training self points are used together as a whole, and compared with the point-wise negative selection algorithm, the new approach can improve the training efficiency of system and the detection rate significantly.     

Energy efficiency enhancement of two-way amplifier forward relaying with channel estimation error简

In order to save energy consumption of two-way amplifier forward （AF） relaying with channel estimation error, an energy efficiency enhancement scheme is proposed in this work. Firstly, through the analysis of two-way AF relaying mode with channel estimation error, the resultant instantaneous SNRs at end nodes is obtained. Then, by using a high SNR approximation, outage possibility is acquired and its simple closed-form expression is represented. Specially, for using the energy resource more efficiently, a low-complexity power allocation and transmission mode selection policy is proposed to enhance the energy efficiency of two-way AF relay system. Finally, relay priority region is identified in which cooperative diversity energy gain can be achieved. The computer simulations are presented to verify our analytical results, indicating that the proposed policy outperforms direct transmission by an energy gain of 3 dB at the relative channel estimation error less than 0.001. The results also show that the two-way AF relaying transmission loses the two-way AF relaying transmission loses its superiority to direct transmission in terms of energy efficiency when channel estimation error reaches 0.03.     

Corrosion behavior of ferritic stainless steel with 15wt% chromium for the automobile exhaust system

The effect of chloride ion concentration, pH value, and grain size on the pitting corrosion resistance of a new ferritic stainless steel with 15wt% Cr was investigated using the anodic polarization method. The semiconducting properties of passive films with different chloride ion concentrations were performed using capacitance measurement and Mott-Schottky analysis methods. The aging precipitation and intergranular corrosion behavior were evaluated at 400- 900℃. It is found that the pitting potential decreases when the grain size increases. With the increase in chloride ion concentration, the doping density and the flat-bland potential increase but the thickness of the space charge layer decreases. The pitting corrosion resistance increases rapidly with the decrease in pH value. Precipitants is identified as Nb（C,N） and NbC, rather than Cr-carbide. The intergranular corrosion is attributed to the synergistic effects of Nb（C,N） and NbC precipitates and Cr segregation adjacent to the precipitates.     

Recent developments on applications of sequential loop closing and diagonal dominance control schemes to industrial multivariable system

With a focus on an industrial multivariable system, two subsystems including the flow and the level outputs are analysed and controlled, which have applicability in both real and academic environments. In such a case, at first, each subsystem is distinctively represented by its model, since the outcomes point out that the chosen models have the same behavior as corresponding ones. Then, the industrial multivariable system and its presentation are achieved in line with the integration of these subsystems, since the interaction between them can not actually be ignored. To analyze the interaction presented, the Gershgorin bands need to be acquired, where the results are used to modify the system parameters to appropriate values. Subsequently, in the view of modeling results, the control concept in two different techniques including sequential loop closing control （SLCC） scheme and diagonal dominance control （DDC） schemes is proposed to implement on the system through the Profibus network, as long as the OPC （OLE for process control） server is utilized to communicate between the control schemes presented and the multivariable system. The real test scenarios are carried out and the corresponding outcomes in their present forms are acquired. In the same way, the proposed control schemes results are compared with each other, where the real consequences verify the validity of them in the field of the presented industrial multivariable system control.     

Dynamic response of cylindrical cavity to anti-plane impact load by using analytical approach简

The transient response of an unlimited cylindrical cavity buried in the infinite elastic soil subjected to an anti-plane impact load along the cavern axis direction was studied. Using Laplace transform combining with contour integral of the Laplace inverse transform specifically, the general analytical expressions of the soil displacement and stress are obtained in the time domain, respectively. And the numerical solutions of the problem computed by analytical expressions are presented. In the time domain, the dynamic responses of the infinite elastic soil are analyzed, and the calculation results are compared with those from numerical inversion proposed by Durbin and the static results. One observes good agreement between analytical and numerical inversion results, lending the further support to the method presented. Finally, some valuable shear wave propagation laws are gained： the displacement of the soil remains zero before the wave arrival, and after the shear wave arrival, the stress and the displacement at this point increase abruptly, then reduce and tend to the static value gradually at last. The wave attenuates along the radial, therefore the farther the wave is from the source, the smaller the stress and the displacement are, and the stress and the displacement are just functions of the radial distance from the axis.     

Biodegradable starch/poly （vinyl alcohol） film reinforced with tita- nium dioxide nanoparticles

Biodegradable starch/poly （vinyl alcohol）/nano-titanium dioxide （ST/PVA/nano-Ti02） nanocomposite films were prepared via a solution casting method. Their biodegradability, mechanical properties, and thermal properties were also studied in this paper. A general full factorial experimental approach was used to determine effective parameters on the mechanical properties of the prepared films. ST/PVA/TiO2 nanocomposites were characterized by scanning electron microscopy （SEM） and X-ray diffraction （XRD）. The results of mechanical analysis show that ST/PVA films with higher contents of PVA have much better mechanical properties. In thermal analysis, it is found that the addition of Ti02 nanoparticles improves the thermal stability of the films. SEM micrographs, taken from the fracture surface of samples, illustrate that the addition of PVA makes the film softer and more flexible. The results of soil burial biodegradation indicate that the biodegradability of ST/PVA/TiO2 films strongly depends on the starch proportion in the film matrix. The degradation rate is increased by the addition of starch in the films.     

Interfacial intermetallic compound growth and shear strength of low-silver SnAgCuBiNi//Cu lead-free solder joints

The growth rule of the interfacial intermetallic compound （IMC） and the degradation of shear strength of Sn-0.SAg-0.5Cu-2.0Bi-0.05Ni （SACBN）/Cu solder joints were investigated in comparison with Sn-3.0Ag-0.5Cu （SAC305）/ Cu solder joints aging at 373, 403, and 438 K. The results show that （Cul-x,Nix）6Sn5 phase forms between the SACBN solder and Cu substrate during soldering. The interracial IMC thickens constantly with the aging time increasing, and the higher the aging temperature, the faster the IMC layer grows. Compared with the SAC305/Cu couple, the SACBN/Cu couple exhibits a lower layer growth coefficient. The activation energies of IMC growth for SACBN/Cu and SAC305/Cu couples are 111.70 and 82.35 kJ/mol, respectively. In general, the shear strength of aged solder joints declines continuously. However, SACBN/Cu solder joints exhibit a better shear strength than SAC305/Cu solder joints.     

Effect analysis of friction and damping on anti-backlash gear based on dynamics model with time-varying mesh stiffness

A nonlinear model of anti-backlash gear with time-varying friction and mesh stiffness was proposed for the further study on dynamic characteristics of anti-backlash gear. In order to improve the model precision, applied force analysis was completed in detail, and single or double tooth meshing states of two gear pairs at any timing were determined according to the meshing characteristic of anti-backlash gear. The influences of friction and variations of damping ratio on dynamic transmission error were analyzed finally by numerical calculation and the results show that anti-backlash gear can increase the composite mesh stiffness comparing with the mesh stiffness of the normal gear pair. At the pitch points where the frictions change their signs, additional impulsive effects are observed. The width of impulsive in the same value of center frequency is wider than that without friction, and the amplitude is lower. When gear pairs mesh in and out, damping can reduce the vibration and impact.     

Nonlinear adaptive optimal control for vehicle handling improvement through steer-by-wire system简

A control algorithm for improving vehicle handling was proposed by applying right angle to the steering wheel, based on the nonlinear adaptive optimal control （NAOC）. A nonlinear 4-DOF model was initially developed, then it was simplified to a 2-DOF model with reasonable assumptions to design observer and optimal controllers. Then a simplified model was developed for steering system. The numerical simulations were carried out using vehicle parameters for standard maneuvers in dry and wet road conditions. Moreover, the hardware in the loop method was implemented to prove the controller ability in realistic conditions. Simulation results obviously show the effectiveness of NAOC on vehicle handling and reveal that the proposed controller can significantly improve vehicle handling during severe maneuvers.     

Numerical evaluation of buckling behavior in space structure considering geometrical parameters with joint rigidity简

The buckling behavior of single layer space structure is very sensitive. The joint rigidity, moreover, is one of the main factors of stability which may determine the entire failure behavior. Thus, the reasonable stiffness of joint system, which is neither total pin assumption nor perfect fix condition, is very important to apply to the real single layer space one. Therefore, the purpose of this work was to investigate the buckling behavior of single layer space structure, using the development of the upgraded stiffness matrix for the joint rigidity. To derive tangential stiffness matrix, a displacement function was assumed using translational and rotational displacement at the node. The geometrical nonlinear analysis was simulated not only with perfect model but also with imperfect one. As a result, the one and two free nodal numerical models were investigated using derived stiffness matrix. It was figured out that the buckling load increases in proportion to joint rigidity with rise-span ratio. The stability of numerical model is very sensitive with the initial imperfection, responding of bifurcation in the structure.     

Electrical and mechanical properties of vapour grown gallium monotelluride crystals

The physical vapour deposition （PVD） of gallium monotelluride （GaTe） in different crystalline habits was established in the growth ampoule, strongly depending on the temperature gradient. Proper control on the temperatures of source and growth zones in an indigenously fabricated dual zone furnace could yield the crystals in the form of whiskers and spherulites. Optical and electron microscopic images were examined to predict the growth mechanism of morphologies. The structural parameters of the grown spherulites were determined by X-ray powder diffraction （XRD）. The stoichiometric composition of these crystals was confirmed using energy dispersive analysis by X-rays （EDAX）. The type and nature of electrical conductivity were identified by the conventional hot probe and two probe methods, respectively. The mechanical parameters, such as Vickers microhardness, work hardening index, and yield strength, were deduced from microindentation measurements. The results show that the vapour grown p-GaTe crystals exhibit novel physical properties, which make them suitable for device applications.