Search results

Moreover, for training the sample data mingled with outlier data in the relatively class of sample, it often can not improve the classification capability.
Data preprocessing.We put a pedal on the stair of the bus.
Fig.1 The system frame diagram of data gathering Feature extraction.
Because the max value of the A/D is 1024, every data can be divided by 1100.
Of SIAM International Conference on Data Mining，Lake Buena Vista, FL, USA,2004

Substituting the above control experiment data into the above formula, the value of the specific reaction rate at any temperature can be obtained.
Results and Discussion Preliminary Results After many experiments and adjustments, it was found that in the case of 200 pellets, the most consistent with the original experimental data.
The results obtained are not much different from the experimental data in the literature, which proves that this method is feasible.
The degree of reduction and the reduction time of the pellets are obtained in continuous time.
Melt reduction [M].

Ferric iron reduction was observed in all incubation modes.
Members of genus Acidiphilium are known to couple the oxidation of glucose to the reduction of Fe(III) [4].
Values of redox potential were plotted against related Fe 3+/Fe 2+ data obtained from the chemical analysis of the solution for both iron species.
Bacterial growth and iron reduction is shown in Fig. 2.
Because the ferric iron reduction and Fe(III) coupled CO2 production were exponential, the corresponding rate constants were calculated from the respective slopes of lines on semilog plots (data not shown).

Research on Evaluation of Soil Fertilities of Farmland at County Level in Nongan Based on Data Mining Li Ma, Guifen Chen, Liying Cao, Yueling Zhao Jilin Agricultural University, Changchun, Jilin, China Mary19801976@sohu.com,752922110@qq.com,24426266@qq.com,14964585@qq.com Keywords：Evaluation of Soil Fertilities of Farmland; Rough set; Decision tree; Data mining Abstract：This research used the method of rough set and decision tree in data mining, building the evaluation model of soil fertility, to evaluate NongAn of Jilin province farmland productivity.
Select HAYATA 452 final, 11 paddy, vegetable 9, located in the county's 26 towns, farmland will generate sample data underlying database.
But in soil domain knowledge, a lot of data, such as soil humus layer thickness or floating-point values, are continuous values or the form of floating point Numbers, so the data must be processed before performing discrete attribute reduction, in order to accelerate the efficiency of knowledge acquisition.
With Soil data, the condition attribute is the 24 item properties as the evaluation factors above, the decision attribute is the soil fertility level, the rough set attribute reduction process is: Obtain the nuclear of condition attributes, and use it as the initial reduction.
In this paper, decision tree method is used to construct the decision tree model, using the rough data after intensive Jane，after reduction of the soil data input the decision tree algorithm, do decision tree structure and pruning, to get fertility level of the decision tree.

Quantitative data about crystal orientation within shear bands were taken from EBSD mapping data.
Crystal orientations within SBs were analyzed from EBSD data.
In each specimen, the EBSD mapping data in the vicinity of the largest angle of SBs were taken into account.
Reduction h1 q1 q2 h2 Fig. 4.
Change of SB angles during reduction.

Symbolic aggregate approximation is data dispersed dimension reduction method.
Time series data mining is an important research branch of data mining, has a wide application value.
SAX methods in the study of time series similarity is a transformation function with many advantages, such as high compression ratio, retain the local information of the data, effective implementation of the data dimension reduction, solve the problem of high dimension.Have higher tolerance to noise data, the segmentation process is realized to eliminate noise and implements the data smoothing processing;Visual intuitive concise;Multi-resolution characteristics, etc.Therefore, in the time series data mining in the many fields have a wide range of applications[9].
Generally, the dimensions of the new sequence are far less than the original time dimension data sequences, also achieved the purpose of dimension reduction.
There are 100 sample series ECG data set, is divided into two categories, each time sequence of length 96 （http://www.cs. ucr.edu/~eamonn/time_series_data/）.

At the same time, it is proved that the attribution reduction is equivalent to the attribution reduction based on the positive region.
However, in practical application due to measuring data error, restricting on access to knowledge, We usually face with the incomplete decision table[2].
At the same time, it is proved that the attribution reduction is equivalent to the attribution reduction based on the positive region.
Thus the attribute reduction is .
Approximation reduction in inconsistent incomplete decision tables[J].

This change can be analysis by the torque of wring, the data test under wind-force of 5 grades is shown in Fig.3.
Such result needs more time -load data to satisfy process of data transformation for obtaining the dynamic performance of wring.
For accurately analysing the working load of hydraulic system in typical working conditions, the root mean square value is to transform the data from testing and integrates the mean value and standard deviation.
Choosing a set of load data from Table.1, the strength test is imposed to reel and the result shown in Table 2.
The critical load of reel is computed by data in Table.2 due to Eq.3, its value is 520.52 kN, which shows that the component will not damage after one million repeatedly load test.

a15500092211@163.com, b15010203158@qq.com, cearly4932@163.com, d345319286@qq.com, echhdai@163.com Keywords: Multivariate Data Visualization, Clutter reduction, Parallel Sets, measurement ratio, corresponding degree for categorical measures Abstract.
Classic part-labeled data visualization method Parallel Sets is applied to represent visualization of multivariate data with measures.
Introduction Multivariate data with measures of metric attributes [9] has been explored in depth in data mining (DM).
Sales data from supermarket are used to prove the efficiency of the improved algorithm.
TVBPS：A Method Based On The Parallel Sets to Measure Properties of Multivariate Data Temporal Data Visualization [J].

The hydrogen reduction reaction kinetic parameters of different particle’ sizes Cu2O were calculated by DTA-TG-DTG data.
(4) Namely (5) Use TG data calculate α, rise temperature velocity β was 15℃·min-1, was obtained by TG and DTG data.
Reaction progression was defined as: n = 1.26I1/2 (6) By the DTA curve data, according to equation (6) the reaction progression n was can obtained.
Use α data, β data, data, n data and equation (5) the k was calculability obtained.
Fig. 3 DTA-TG-DTG curves of sample 1#~3# Using DTA-TG-DTG curves data and equation (2) calculated the different size the cuprous oxide hydrogen reduction reaction apparent activation energy E and frequency factors A.