Search results

Preparation of data: includes the standardization of the data characteristics, and dimensionality reduction of high dimensional data.
Results and Analysis of Experiments Data Sets.
To investigate the effectiveness of the algorithm, it is applied to the following three UCI data sets [2], Iris data sets, and Wisconsin breast cancer: Original (BCW) data sets and Abalone data sets.
These data sets are widely used in performance tests of fields of pattern recognition, data mining and so on.
Table 2 shows the nature of these data sets.

The experiment is carried out based on KDD Cup 99 data set.
In this paper training data set is composed of 8 000 data of normal type and 8 000 data of attack type, selected randomly from KDD Cup99 of "10% KDD" dataset.
Testing data set is composed of 4 000 data of normal type and 4 000 data of attack type, selected randomly from KDD Cup99 of the "Corrected KDD" dataset.
The attack types are divided into normal data and attack data.
But for the detection of normal data, the detection of abnormal data and average detection, there is little difference between BP and New-GA-BP.

The data showed that L. ferriphilum was able to utilise ferrous iron at low-level concentrations of SCN- (0.25 mg/l to 1.75 mg/l), however exhibited a reduction in oxidation rate relative to the control (0 mg/l).
Historical process data suggests that the biooxidation community is highly sensitive to SCN- with the empirical limit supposedly 1 mg/l [2, 3].
To date, no kinetic data or information pertaining to the iron oxidation performance of L. ferriphilum in the presence of SCN- is available.
The data indicated a decrease in oxidation rate compared to that of the control.
Sensitivity of L. ferriphilum is evident from the data presented in Figure 1 and Table 1.

The HDH plate was rolled only to 75% reduction in thickness.
This is reasonable since the HDH product was formed using HIP forming methods and not by sintering. [20] Figure 4 shows a table of Charpy Impact data (A) and a histogram of the same data (B).
This was significantly lower than both the blended elemental sample rolled to 75% reduction, 34.03 J (25.06 ft lb), and the prealloyed HDH powder sample also rolled to 75% reduction in thickness, 33.19 J (24.44 ft lb).
Deviation Joules (ft.lb) 4.41 (3.25) 4.47 (3.29) 4.18 (3.08) (A) (B) Figure 4: Charpy Impact data BE 75% BE 87% HDH 1000x Magnification (D) (E) (F) Figure 5 SEM images of Charpy Impact Sample Fracture Surface in the short transverse direction BE 75% BE 87% HDH As Received (A) (B) (C) Mill Annealed (D) (E) (F) Figure 6 (0002) Pole Figures of selected samples Conclusions and Recommendations · HDH samples had equiaxed α grains while the Blended Elemental samples had elongated α grains
Pages 888-895. 2013 [6] ATI Ti-6Al-4V Data Sheet (Allegheny Technologies Incorporated, Philadelphia, USA, 2012) [7] Wang, H., Fang, Z.

The Study of Synchronous Reduction-carbonization of V2O3, Cr2O3 and W-Co Composite Oxides in Fluidization Nanyan GONG，Yafei OUYANG Diamond Road, Zhuzhou, Hunan, P.R.China GONGNY@126.COM , OUYANGYAFEI123@126.COM Key words: VC,Cr3C2, Fluidization, Reduction-carbonization Abstract.
Theory Analysis Thermodynamics calculation of synchronous reduction-carbonization of vanadium.
Thermodynamics computing of synchronous reduction-carbonization chromium.
The chemical compositions of composite powder are shown in Table.3 after reduction-carbonization of composite oxide in fluidized bed.
Practical manual with thermodynamics data on mineral.

Before analysis, arcsine transformation was applied on germination data.
Data are the means of nine replicates±se (P<0.05).
Data are the means of nine replicates±se (P<0.05).
Data are the means of nine replicates±se (P<0.05).
Therefore, previous data suggest the existence of plant eATP signaling.

H2-Temperature-programmed-reduction (TPR) profiles.
Genarally the reduction of iron oxides take place in two steps.
Table 1: H2-TPR data of the catalysts Catalyst 5wt%/Al2O3 Reduction temperature (◦C) Peak 1 Peak 2 Peak 3 Co/ Al2O3 507 650 731 70Co:30Fe/Al2O3 447 501 667 50Co:50Fe/Al2O3 328 412 614 30Co:70Fe/Al2O3 456 458 669 100Fe/ Al2O3 454 635 716 Bimetallic nanocatalysts showed different reduction patterns than monometallic nanocatalysts.
Total H2-consumption and degree of reduction of the catalysts.
Fig. 3: Representative spectrum of CO-Chemisorption at 250 ◦C for Co/Al2O3 Table 3: CO-Chemisorption data for 5wt%/Al2O3 catalysts Catalyst CO-Adsorbed (μmol/g.cat) Co/Al2O3 0.41 70Co:30Fe/Al2O3 0.50 50Co:50Fe/Al2O3 1.77 30Co:70Fe/Al2O3 0.57 Fe/Al2O3 0.27 X-ray diffraction analysis.

It was a normal probability plot that indicated the error of the data, and estimated standard deviation of the residual that measures the amount of standard deviations separating the observed and predicted values.
The data points lying on the straight line show that there was no problem with the normality of this curve.
As a histogram in Fig. 1b the results shows that the trend of the normal curve represented the accurate data.
The fit chart in Fig. 4c shows the random of the data.
The results of this experiment showed that the data were distributed as well.

In this paper, data of different microbial populations in the coal spoil substrate and geochemical background data are given, and they are compared with former data of microbial communities in sandy substrates and their impact to the environment.
The obtained data correspond well with a parallel and long-term monitoring of ground water in the coal spoil area which is characterized by high concentrations of SO42-, Fe and NH4+ and an enhanced acidity [3].
Therefore, a reduction of the high oxidation potential, acidity and salinity by sulfate reduction processes is too slow under the actual conditions and will show no substantial changes in the next future.
In the sandy Lusatian spoil material, only from a depth of 12 – 15 m a complete sulfate reduction was detected [1]; in the oxidation zones above no sulfate reduction could be shown.
Weber of the LMBV for the data support and cooperation in this project.

The two main problems in data-based SHM are therefore: (1) If supervised learning is necessary, how does one acquire data corresponding to damage states of the structure
Basically, a statistical model of the healthy system is created, based on normal data and then subsequent data are tested to see if they are statistically consistent or inconsistent with the normal data.
The first SVM (labelled SVM0) seeks to separate damage-state data from normal-state data.
Confusion matrix for Classifier 2 applied to testing data.
Also concerned with the problem of sourcing data, a classifier has been presented that has generalised from single-site damage data to multi-site data.