Search results

The research of key data classification optimal mining methods for massive data Li Yinying, Wang Yunze HeBei institute of communictions, ShiJiazhuang Hebei 050071, China Keywords: inter-cell classification; mathematical model; fuzzy; convergence Abstract: The data classification is an important issue in massive data classification.
The data indexing is the procedure to extract and integrate the data features so that the data which are need to be indexed will maintain stable within a center range.
In the equation, denotes the single feature of the data, n denotes the amounts of the data features, t describes the final features of the data by calculation
（2）K samples are selected in the data set.
They are classified by K-means after dimension reduction with semantic matrix.

These additional data provide an idea on how the computational gain may evolve when the total time of the analysis (here tfinal=4∙10-5 s) is varied.
Time-invariant actuation, data related to: run times; overall error, and computational gain with respect to the staggered solution.  
Total time [sec] Snapshots/ Base generation Reduction system error w.r.t. stag. gain w.r.t.
Time-varying actuation, data related to: run times; overall error, and computational gain with respect to the staggered solution.  
Brand, Incremental Singular Value Decomposition of Uncertain Data with Missing Value, Lecture Notes in Computer Science, 2350 (2002) 707–720.

Although significant amount of information on conventional steel welds is available, lack of data still exists for the laser beam welded Al-alloys of aerospace grade.
Data were taken for sheet T4 at 15 points, separated by 3 mm, on a line perpendicular to the weld line.
The sheet was rotated from -2.5° to 2.5° around an axis perpendicular to the weld line during data acquisition to increase the number of diffracting grains.
The data acquisition time was 23 seconds to match it with the slow motion of the rotating table.
The program "Fit2d" was used for data reduction.

Zhang Yuzhuo thought the key of carbon emission reduction is the cleaning of coal in China.
Fig.1 The change of energy consumption structure in China Data from China Statistical Yearbook 2010 and 2010 National Economic and Social Development Statistical Bulletin.
Fig. 2 Proportion of output value of China's three industries in GDP Data from China Statistical Yearbook 2010.
After this, the regulations and measures of Energy saving and emission reduction and improving energy efficiency are drew up continuously.
Energy efficiency should be improved further China's energy efficiency improved from 1179 RMB per ton standard coal of 1985 to 10,818 RMB per ton standard coal of 2009 Data are calculated by energy efficiency=GDP/gross of energy consumption

Efficient Moving Target Tracking algorithm: Design and implementation of the Sand-table tracking algorithm Jiaqi Li Macao university of science and technology ,Taipa, Macau Mail:908934586@qq.com Keywords: dynamic recognition, target tracking, image processing, dynamic noise reduction Abstract.
It is important to develop robust real-time video understanding techniques which can process the large amounts of data attainable.
Learning classification based on pattern identification A.K-Nearest Neighbor[ 7] B.Bayes Classifier[ 8] C.Principle Component Analysis[ 9] 　　2.a Work well on the training sample and target recognition 　　2.b Need a large database of sample to training 　　2.c less recognition error need large data .large data need large cost in operation 3.Recognize and tracking by model Pre-summarize the key point of the target with different methods ,then tracking it with the model.
(hard to carry and expensive) 5.c Need a large database to match the data between the special equipment and the normal camera 5.d Need a algorithm with high complexity and low compatibility Algorithm overview The current target tracking algorithm has disadvantages: 1 .Slow because of the recognition / background rebuild/FFT operation 2.
Each pixel simulate the hight of the sand-table .We do Growth and Linear reduction at same time to imitate the the process of falling.

The best fit of eq (7) to the experimental data is shown, in figures (2-5), by a solid line.
The S-parameter as a function of thickness reduction (%) for 5005 Al-Mg alloy.
The S-parameter as a function of thickness reduction (%) for 5051 Al-Mg alloy.
The Sparameter was studied as a function of thickness reduction for the alloys.
Al-alloy (5083) S-paramter Thickness Reduction (%)

From these experimental data and taking into account the Arrhenius equation, the activation energy of the recrystallization process has been determined for each deformation rate.
The copper has been first industrially hot rolled and cold wire-drawn between ∆=51 and 94 % (reduction in the surface, diameter after reduction d=5.54 to 1.93 mm, true strain ε=0.73 to 2.84).
Fig. 3 Reaction advancement factor calculated from measurements of the diffracted intensity at the center of the {111} pole figure. a) Wire-drawn to 71 % reduction. b) Wire-drawn to 94 % reduction.
The fast decrease is rapidly slowed down at the highest level of reduction and value of Ea≈45 kJ/mol is reached for a reduction level equal and higher to 90 %.
Values around 45 kJ/mol are obtained at deformation level equal to and higher than 90 % reduction.

Critical potentials exist for the formation and reduction of oxide films in high temperature water.
The film reduction process is relatively slower than the film formation process.
Normalized CER (NCER=CER/CERlow DO) values calculated from the original CER data in figure 4a are shown in figure 5.
The time scale (in seconds) is also adjusted for easier data treatment.
The reduction process is more complicated.

These data show that the hospitality is both high energy consumption and high carbon emission.
Though years of data collecting and analyzing, we recommend that the Monitoring Reporting Verification (MRV) should consist the following index: (1) The comprehensive energy consumption.
Data source and its analysis As of December of 2013, monthly comprehensive energy consumption data form 14 high-star eastern China hotels are collected, the period of which various from 1 year to 6 years.
And the relevant data are also collected from 12 2-star and 3-star hotels, the period of which various from 2 to 4 years.
We have also collected the data from 15 economy hotels, the period of which ranges from 2 to 3 years.

The method includes the following steps, pretreatment, feature extraction by WPD and dimensionality reduction by LPP and classification of the test samples to a corresponding class according to the nearest neighbor classifier.
Wavelet packet decomposition of gait energy image Wavelet packet decomposition (WPD) is a wavelet transform where the discrete-time (sampled) data is passed through more filters than the discrete wavelet decomposition  (DWD) [11,12].
The direct use of the gait cycle gait recognition exists huge amount of data, increasing the difficulty of gait feature extraction and calculation of consumption, in order to guarantee not to abandon the gait characteristics while reducing pending gait amount of image data, gait energy image (GEI) is used to last cycle through a number of gait image synthesized by the method of weighted average image, this image contains a gait cycle each gait image contour, frequency, phase, etc. gait information.
Other image processing applications such as noise reduction, edge detection, and gait analysis is as in Ref. [5, 6, 7, 10].
Data Eng. 17 (12)( 2005)p.1624-1637.