Search results

Dimensionality reduction is the mapping of high-dimensional data into some representation of lower dimensionality.
We apply GPLVM method for dimensionality reduction on the data set which is in high-dimensional form, and then we evaluate the quality of the resulting low-dimensional data representation by comparing the results attained to those we got when using intrinsic dimensionality estimation techniques, and lastly compare the performances of GPLVM and PCA on the same data set.
The data set consists of 53 data points with a dimensionality of 87 features.
Now, our data set has been successfully downsized using the above illustrated dimensionality reduction techniques, the next step is to make use of this outcome for effective and fast clustering.
Fig. 3 shows the resulting data clusters.

Introducing geodesic distance to replace the original European distance data will make twenty-four dimensional data, which using the traditional MFCC feature extraction down to ten dimensional data.
This paper put the Manifold learning used in speech recognition of the feature extraction, proposes a new MFCC - Manifold feature extraction method, to the low dimensionality reduction is effect, and the characteristic value of the data quantity have done a lot about minus.
Realize the data dimension reduction or about data visualization and obtain appropriate mapping relation.
For convenience of shows that the characteristic value data quantity about loss of direct effect to MFCC feature extraction of characteristic value data quantity as the standard, the type definition: Data ratio = (MFCC - Manifold characteristic value data quantity) / (Mel characteristic value data quantity) TABLE I.
Can be seen from table 1, compared with the traditional MFCC its final recognition rate has, and the data reduction about is above 70%.

Pawlak, is an extension of set theory for the study of the intelligent systems characterized by uncertain, imprecise, incomplete and inconsistent data.
Preliminary Information tables are firstly defined by Pawlak for rough data representation [1].
Pawlak, Rough Sets: Theoretical Aspects of Reasoning About Data, Kluwer Academic Publishers, Boston, 1991
Luo, Data analysis based on discernibility and indiscernibility, Information Sciences 177 (2007), p. 4959–4976
Xie, Information-preserving hybrid data reduction based on fuzzy-rough techniques, Pattern Recognition Letters 27 (2006), p. 414–423

Wide Area Distributed Data Acquisition Mode.
In each subsystem of a data acquisition area, data acquisition server is operated in the way of hot standby, operating data of other areas will not affect the operation and resource cost of this area [5].
Data Integration Technology.
FES server of region DAS is connected with data acquisition network of region DAS, which is used for data acquisition of region DAS and sent data to SCADA servers of region DAS, while FES server of county DAS is connected with data acquisition network of county DAS, which is used for data acquisition of county DAS and sent data to SCADA servers of county DAS.
Region DAS adopts subscribe method to get all county’s data by service bus, SCADA server of region DAS has the whole system’s data, thus realizing data share in the whole system.

Keywords- Data mining, Frequent pattern mining, Association rules Abstract-Frequent pattern mining has been a focused theme in data mining research.
Orlando et al. [13] proposed an algorithm that combines transaction reduction and direct data access.
Efficient Data Mining for Path Traversal Patterns.
Set-oriented data mining in relational databases.
Data Knowl.

The data acquisition system is composed of four modules: the transducers (the force transducer for the z direction, the force transducer for the x direction consisting of tensometric devices HBM 350XY11) placed on the two rings of the dynamometric table, the signal conditioning modules (MB-38 produced by Keithley Instruments Inc.), the analog – digital conversion device (KPCI 3108, Keithley Instruments Inc.) and a software package that controls the acquisition system and processes the collected data (TestPoint).
The system used to measure the strain and thickness reduction was an optical system Argus by GOM.
Fig. 5 and 6 presents the variation of thickness reduction th for cases Fem9 and Exp9.
Thickness reduction variation th for case Fem9 Fig. 6.
Thus, strain and thickness reduction rise with the decrease of vertical step.

According to the data presented in Table 1, in the near infrared region bismuth in some glass systems may have a broadband luminescence with a full width at half maximum (FWHM) of the order of 250-300 nm at 1.3 microns position.
Therefore, the reduction conditions are needed.
The absorption spectra of samples treated at reduction conditions were also measured.
Those data are in accordance to literature data and the measured Bi near infrared luminescence under 455 nm excitation is shown [2, 3].
Taking into account the data, we can conclude that the luminescence centers have the same nature.

In the dry category, the selective catalytic reduction (SCR) and selective non catalytic reduction (SNCR) were the most widely used [2-3].
This paper presented a new method of “absorption-oxidation-reduction” which used ferrous sulfate solution as absorbent, oxygen as oxidizer and urea as reducer to remove NOX from flue.
This method could provide important reference data for wet denitration, Moreover, it has good application prospects. 2 Experimental Sections 2.1 Experimental apparatus and instruments Experimental apparatus and flow diagram of “absorption–oxidation-reduction” coupling method in Fig.1.
Low-temperature selective catalytic reduction of NO with NH3 over Ce-Mn/TiO2 catalyst [ J].
NO removal by reducing agents and additives in the selective non-catalytic reduction(SNCR) process [J].

Compared to remote Sensing census data of 1988, the watershed erosion is reduced by 63 million tons and soil erosion area is reduced by 4.09%.
The above statistical analysis of sediment data measured by the main hydrological stations in the Three Gorges Reservoir (Yichang) shows that mainstreams of the Yangtze River above Zhutuo have no significant change and the runoff and sediment reduction of Cuntan and Yichang stations is mainly caused by the Jialing River runoff and sediment.
From the reductions of the Jialing River, the effect of sediment reduction is quitely prominent.
The data analysis shows that the river sediment discharge has decreased significantly since 1989, "Changzhi Project" carries out soil and water conservation.
According to the statistics data measured by soil and water conservation management of Beibei, Zhutuo, Cuntan and Yichang station, the sediment distribution of the Three Gorges changed significantly.

Dynamic predictive maintenance model based on data-driven machinery prognostics approach W.Z.
In this paper, a data-driven machinery prognostics approach is developed to describe machine degradation.
First, as the collected monitoring data usually contains multidimensional data sets, feature extraction is applied to obtain dominant information.
Here, and could be deduced from the history maintenance data of the system.
Comparison between the actual data and the estimated data of the 500 sets of samples In the predictive maintenance model, the related cost factors are: and .