Search results

Moreover, it is more suitable for the classification of low dimensional data dimensionally reduced by dimensionality reduction methods, especially those methods obtaining the low dimensional and neighborhood preserving embeddings of high dimensional data.
MNIST data set consists of handwritten digits 0-9 (60,000 training samples, 10,000 test samples).
Recognition rates (%) of several classifiers on MNIST data set (the dimensionality is reduced to 9 by LDA).
Although LPP is a linear transformation method, it doesn’t discover the globally linear structure of data set and preserve it in low dimensional space, but discovers the local neighborhood of data set and preserves it in low dimensional space.
Further, it seems that NN-LSRC performs better than SRC with the samples dimensionally reduced by several dimensionality reduction methods, especially those methods obtaining the low dimensional and neighborhood preserving embeddings of high dimensional data.

Analytical expressions for cathodic reaction were developed for the reduction of H+ and the reduction of oxygen.
According to testing data, the limiting current at -0.5V vs. the solution concentration change at P, then reaction mechanism changes at P, before point P, the corrosion proceeds via the reduction of H+, while after point P, the corrosion proceeds via the reduction of the oxygen, and the concentration near P can be estimated to be: 0.011～0.012mol/cm3 for APF=1.5, 0.01～0.011mol/cm3 for APF=2.875, 0.01～0.0106mol/cm3 for APF=3.3, 0.01～0.0102mol/cm3 for APF=3.8 and 0.009～0.0096mol/cm3 for APF=4.3.
The kinetic parameters for reduction of H+.
This relationships were also developed by the testing data and shown as fig.8 at -0.5V, then n can be calculated from the slope of lines with D≈9.34×10-5cm2/s and d ≈2.5×10-3cm.
The kinetic parameters for reduction of oxygen.

This method not only reduces the demand for data, but also ensures accuracy of the results using the sensitivity correction.
The key problem of condition assessment of distribution equipments is to obtain data of the state variables from the equipments.
The more complete and accurate the data is, the more exact the assessment results will be.
Therefore, the state data of the distribution equipments is very small.
Use these data to analysis the sensitivity of non-core status.

The basic data format for patient cases has listed in Table 4.
So dimensionality reduction is also required to develop a single dimension data mining model.
FP-tree algorithm is used to mine the new single dimension data, after which rule screening is carried out aiming to delete the inappropriate and take appropriate rules.[5] Example of Data Mining Course Table 5 lists the transaction data that need to be mined.
Patient data and mining record is shown in Table 6.
Data Mining: Concepts and Techniques[M].

Accordingly, the wavelength changes can be regarded as the initial data of alarm.
The deviation of the real-time value and this mean value can act as valid data.
Both groups include 12 sets of actual intrusion data and 18 sets of non-intrusion data.
Get waveform data: Process the sets of measured signals and get the waveform data, the dimension of which is 64 and the sum is 1.
Apply the feature of central moment of 4D to the 64D waveform data.

The biological selenate reduction generated alkalinity, increasing the wastewater pH from 6.0 to 8.6.
The IFBR was monitored and controlled using data acquisition and control hardware (CompactRio National Instruments, USA) and software (Labview, National Instrument, USA).
Data collection and read trimming/filtering was performed using TorrentSuite 5.0.
The formation of a red precipitate indicated reduction of selenate to elemental selenium.
The biological selenate reduction generated alkalinity, increasing the wastewater pH from 6.0 to 8.6.

Data Fusion Model of BP Neural Network The BP neural network sensor data fusion and can greatly reduce the feature dimension data transmission, data fusion to improve the efficiency of a wireless network.
BP neural network algorithm uses a single hidden layer of the data collected by wireless sensor networks for data fusion.
After the data fusion algorithm BP neural network applications for each cluster, the neural network models for each cluster, the nodes in the cluster to collect data using the input neurons function to preprocess the data collected and handled well data to the cluster head node.
Eigen value data fusion is much less than the amount of data collected, thereby reducing the amount of data transmitted, reducing power consumption and prolong the life of the wireless sensor network.
Improved model incorporates the processed data is much less than the amount of data collected by the sensor nodes before treatment, thereby reducing the energy consumed during data transmission achieving the purpose extend network lifetime.

By integration analysis of monitoring data of typical section, curves of horizontal and vertical displacement with time were drawn, shown in Figure 2.
Fig.3 Curves of external deformation rate with time According to the typical cross-section observation results and horizontal and vertical accumulative displacement variation, curves of horizontal and vertical displacement with time were drawn by integration analysis of data, shown in Figure 3.
By integration analysis of monitoring data of typical inclinometer hole, deep displacement curves with depth were drawn, shown in Figure 4.
Based on the observation results of typical inclinometer hole and variation of displacement deformation with depth, as well as integration analysis of the data, curves of displacement and its rate 1m away from the orifice with time are drawn, shown in Figure 5.
Slope Stability Analysis by Strength Reduction FEM [J].

Recrystallization texture also was examined during isothermal annealing at 500°C with reduction in area.
Stored energy associated with orientations in the deformed grains was evaluated by reconstructing of data measured using electron back-scattered diffraction (EBSD).
To obtain the electron backscatter diffraction (EBSD) data, high resolution EBSD (using a JEOL 6500F scanning electron microscope with an INCA/OXFORD EBSD system) was used and the data analysis was performed using the REDS [15].
Reduction in area is specified by 90, 92, 95, 98%.
Vinel: Physica Status Solidi (B) 112 (1982) p. 111 [15] REDS, Repressing of EBSD Data in SNU, User Manual, Texture Control Lab, 2002

The porous ceramic holds good potential as acoustic resistance and vibration reduction material during ultra-precision machining.
This is mainly dependent on two aspects, one is the manufacturing accuracy, the other is the vibration control and reduction on machine and its peripheral systems during the operation [2].
This paper probes into the acoustic properties of porous ceramic, which hold good potential as acoustic resistance and vibration reduction material on machine body, tool post, saddle, and so on.
The noise absorption and resistance properties are studied, and the calculation based on JCA model and the comparison between it and the reported data are presented.
The calculation results based on the above model are compared with the experimental data from literature [9], as shown in Fig. 3 and Fig. 4, which indicate the model conforms to the measurement well.