Search results

Collection S include s data sample, assuming Class label attribute has a different value m, making a kind of different definition.Set the number sample of , to classify a given sample, the information of the required expectation for: Whereis the probability of any sample belong to, with estimates of.
Then some scholars put forward some improved algorithm, such as C4.5 algorithm, CART algorithm, SLIQ algorithm, SPRINT algorithm, etc. 2 Multivariate Decision Tree Algorithm Based on Rough Sets 2.1 Multivariate Decision Tree Construction Algorithm Based on Rough Sets In the selection of conditional attributes set, firstly, to deal with the data previously, delete the data with noise, to the default value for neat, and a discrete a series of data value, its construct algorithm as follows.
Table 1 Decision table U A B C D E F G d U A B C D E F G d 1 1 2 2 1 1 2 1 1 9 1 2 2 2 1 2 1 1 2 1 2 3 2 2 2 1 1 10 1 1 3 2 1 2 1 1 3 1 2 2 3 1 2 1 1 11 2 1 2 2 1 2 1 2 4 2 2 2 1 1 2 1 1 12 1 1 2 3 1 2 1 1 5 2 3 2 2 1 1 1 2 13 1 2 3 2 1 2 2 1 6 1 3 2 1 1 2 1 1 14 2 3 2 2 1 2 2 2 7 1 2 3 1 1 2 1 2 15 1 2 3 2 3 2 2 2 8 2 3 1 2 1 2 1 1 16 2 3 2 2 1 3 2 1 We can get condition attribute set R = {A, C, D, E, F} after attribute reduction, and K (D, D) = 2/16, K (C, D) = K (E, D) = K (F, D) = 1/16 through calculation, so SIG = {{D}, {C,E,F}}, therefore =D，={C,E,F}.

With the use of image reduction and reconstruction, it can realize the optimization design of image.
Then the designed image transfers data to the computer receiving end through BeeCube platform, and the computer compares the received data with the software results.
In the process of calculation, the results of data processing are shown as follows.
Tab.1 Results of computer image processing Number of iterations Background rendering / frame Residuals 100 123 0.00122 200 156 0.00105 300 178 0.00086 400 205 0.00052 500 236 0.00011 Table 1 shows the statistical data obtained from the dance movements image processed by using computer multimedia technology.

When determining the grid number the analysis data type should be considered.
When calculating the positions with large gradient data changing or stress concentration we need utilize intensive grid.
While calculating the positions with small gradient data changing we should divide relatively sparse grids.
By comparison, the experimental data and calculated values are close, the average relative error 11.1% is within the allowable range, indicating the case finite element model and its results are correct.
Table 2 Effect of Rib Size on the Stress Model type Rib thickness (mm) Rib height (mm) The maximum von mises stress (MPa) The reduction of the maximum displacement Original model 6 30 587 - Option II A 6 32 325 81.2% Option II B 6 34 308 83.6% Option II C 8 30 291 84.7% Option II D 8 32 278 86.7% Option II E 8 34 262 91.7% Option II F 10 30 275 86.9% Changes in Size of Reinforcement Plates.

Therefore, research on highway hydrology Zoning can provides scientific basis for disaster prevention and reduction of highway.
In order to improve the accuracy of the division and provide basic data and theoretical basis for national highway hydrology Zoning, this paper established the index system of highway hydrology Zoning and proposed a framework of provincial highway research framework of hydrological Zoning based on the relationship of hydrological elements and highway engineering research.
Provincial Highway Hydrological Zoning Research Framework Provincial Highway hydrological Zoning research in general is divided into basic data collection, index analysis and zoning map compilation in three steps.
First, the data of geographical elements related to hydrological environment in the research area should be collected.
Then, the index system of highway hydrological Zoning should be established and the Zoning index should be analyzed based on basic data.

Content of study 1) Based on cloud computing collaboration and sharing of teaching resources, to promote the "cloud" interaction, establish the standard system of teaching resources, ensure the effectiveness and real-time of teaching resources. 2) Construct the cloud service platform system of integration of teaching data processing, analyzing and application.
Through user-friendly windows, cloud computing configuration and management of the client, based on multiple layers of IaaS, SaaS, PaaS, the user can use the teaching data (educational content, educational feedback, quality assessment, etc.), interoperability, mutual urge, to establish education quality assessment system of full participation. 3) Create a student-centered cloud service platform, based on short-board of students' knowledge, push the service of animation teaching and practice for wrong titles, using virtual rewards and integration mechanisms, inspire and mobilize the enthusiasm of students, to establish good learning habits. 4) Utilization blended learning mode, with the teachers' classroom teaching, online animation and video teaching, the students obtain knowledge from online, school and other channels. 5) Solve the unfair problem of education, based on the technical level.
Platform pilot, 30% reduction in the workload of teachers, and learning-targeted to be increased by 50%, a 50% increase in enthusiasm for learning.
Bigtable: A distributed storage system for structured data.
Data mining using high performance clouds: Experimental studies using sector and sphere.

Introduction OFDM is a kind of multicarrier parallel transmission scheme, divide the channel into several sub channels and put the high-speed serial data stream convert into a number of sub data stream whose rate is relatively slow and the data is parallel, also modulation the signal for each sub channels, at the same time make each signal to be orthogonally, so that spectrum overlap each other, thus improving the efficiency of the spectrum and reduce the mutual interference between channels.
And using IDFT for, converting the data in frequency domain to the time domain in order to get the time domain sample sequence of the modulated signal, after adding the guard interval, then to the time domain sequence for digital conversion, finally get the time domain waveform of the modulated OFDM signal.
The DFT of the receiver can be consider as the received signals convolution with the filter which the impulse response is , then a reduction of sample N.

Crystal Data [Mn2 (C8H7O3)4(H2O)2] Dx = 1.489 Mg m3 Mr = 375.23 Mo Kα radiation triclinic, P -1 Cell parameters from 8291 reflections a = 8.038(5) Å θ= 2.99 -27.43° b = 9.695(5) Å μ= 0.822 mm-1 c = 11.149(5) Å T = 293(2) K α= 86.158(5)° chunk, yellow β= 81.239(5)° 0.20 * 0.18 * 0.16 mm γ= 77.298(5)° V = 837.2(8) Å3 Z = 2 Data collection Rigaku R-AXIS RAPID 3795 independent reflections diffractometer 2816 reflections with I > 2σ(I) ω scans Rint = 0.0207 Absorption correction: multi-scan θmax = 27.43° (ABSCOR; Higashi, 1995) [11] h = -9→10 Tmin = 0.832, Tmax = 0.877 k = -12→12 6519measured reflections l = -14→14 Refinement Refinement on F2 H-atom parameters constrained R[
Hydrogen-bond geometry (Å , °) D—H…A D—H H…A D…A D—H…A O4—H4A…O13ii 0.73(4) 2.01(4) 2.733(5) 170(3) O4—H4B…O2iii 0.93(5) 1.69(5) 2.621(4) 177(6) [Symmetry codes: (ii) 2-x, 1-y, 1-z; (iii) 1-x, 1-y, 1-z] Data collection: PROCESS-AUTO [8]; cell refinement: PROCESS-AUTO; data reduction: CrystalStructure [9]; program(s) used to solve structure: SIR92 [1]; program(s) used to refine structure: SHELXL97 [2]; molecular graphics: ORTEP-3 for Windows [5]; software used to prepare material for publication: WinGX [6].

The study of possible mechanisms active during sintering was attempted by continuously recording the shrinkage vs. time and elaborating the data on the basis of Kingery' s model for liquid phase sintering.
Finally, magnetic ac susceptibility measurements were performed on the sintered samples (Quantum Design, Lakeshore), collecting data with a driving field amplitude varying from Hac≈10 A/m to Hac≈1200 A/m, oscillating at a frequency of 1000 Hz, on warming from 15 K after zero-field cooling of the samples.
The value of 1/6 is very near to the theoretical one (1/5) for a mechanism of diffusion of prismatic particles into the intergranular liquid phase as the rate-governing step; on the other hand 1/n=1/8 can be justified on the basis of a progressive reduction of the viscosity of the intergranular fluid with the increase of temperature.
To best characterize the samples as a function of treatment temperature (i.e. the actual density), we evaluated the critical current density (Jc) using the susceptibility data in the framework of the Bean model [5, 6]; the data seem to claim the existence of two different types of samples (Fig.4) and those with d > 90 % seem to produce the better results in term of critical current.

In the present paper the crack growth methodology is used to make predictions that are then compared to experimental crack growth data for the F/A-18 and F-111 load spectra.
The predictions for the F/A18 case are shown in Fig. 2, together with recent experimental data and FASTRAN [13] predictions which does not account for notch plasticity.
Discussion As shown by the solid lines in Fig. 2 and 3, the current crack growth predictions for both the F/A-18 and F-111 cases correlate closely with the experimental data.
Taking into account the elastic-plastic response of the material results in a reduction of the stress near the hole thereby reducing the stress intensity factors calculated.
Results of the solution technique, when compared to experimentally measured crack growth data under F/A-18 and F-111 load spectrums, showed improved correlation over the FASTRAN predictions.

The main advantages of using SHM for maintenance which are the driving factors for research and development are their ability to monitor large areas and blind spots without disassembling parts resulting in significant reduction in the aircrafts ground time.
To develop and assess the meta-models for impact detection and identification, a data base of 1500 various impact scenarios (changing location, mass, velocity) on a sensorised composite panel has been generated via SMART FE simulation running on HPC units, see [9] for details of the numerical analysis.
This effect was added to the sensor signals to represent sensor data acquired in service, see Figure 2(a).
(a) sensor data with & without operational load (b) PDF of the correlations between FE and filtered signals; f=30Hz Figure 2 Statistical analysis – Influence of operational load A band pass filter was developed to denoise the sensor signals and remove the influence of the load before inputting them to the developed meta-model for impact detection and force reconstruction.
(a) PDF (b) CDFs and errors Figure 3 Statistical analysis –PoFA; empirical data and best fitting distributions Results and Conclusion In this work a statistical analysis was carried to study the effect of a range of parameters on the performance and reliability of the proposed passive sensing algorithm based on ANN.