Search results

· Extracting knowledge from data.
Roadmap on technologies of model order reduction, data-sciences and hybridation.
The first concerns data analysis and more particularly data-reduction.
By combining both, physics-based models, calibrated by using data assimilation, and operating in real time by using advanced model order reduction techniques, with a data-driven model for describing the gap between the measures and the physics-based model prediction.
Model order reduction, manifold learning, data-driven applications and nonlinear dimensionality reduction.

Another one for dimension reduction trains a subspace projection matrix to project original data space into some low-dimensional subspaces which have deep architecture, so that the low-dimensional codes would be learned.
Unfortunately, most effective visual features are at high-dimensional data space [2].
Thus, it is difficult for computers to process such a huge data.
It tries to represent the data by the linear combination of a small number of basic elements, and the combination coefficients will be used as low dimensional data.
Kim: Data Mining and Knowledge Discovery, Vol. 26 (2013) No.3, p. 512-32 [3] J.Y.

Finite Element Data Reduction Based Energy Release Rate for Delamination Tests S.
This is shown through two relevant aspects in delamination toughness measuring, say: data reduction and Iso-G delamination front design.
Many experimental data reductions are based on beam theories and thus assumes straight delamination front during propagation, which is not true when investigating laminates with general anisotropy.
Data Reduction Based Energy Release Rate Experimental Approaches.
This last expression (3) can be used as data reduction scheme more efficiently in delamination toughness measuring.

For the mixture, the experimental TG-DSC curves of mixture are compared with the calculated weighted sum, which is the sum of the TGA-DSC data of pure iron ores and biomass according to their weighting percentages, shown in Fig. 2 (b).
It is clear that the experimental data of the mixture is almost consistent with the calculated weight sum before 450 oC, mainly relating to the remove of the water combined in the iron ores.
Thereby, we could conclude that the other components in the iron ores have somewhat effect on the pyrolysis of the biomass, which leads to the difference between the experimental data and the calculated results.
The reduction of goethite ores was conducted in a lab-scale reactor.
Valix, Reduction roasting of limonite ores: effect of dehydroxylation, Int.

The structural effect value were 0.99% and -1.91% respectively, namely the breeding structure does not promote the COD discharge reduction but benefits NH4-N discharge reduction.
According to dynamically updated data from national census of pollution sources, the discharge of Chemical Oxygen Demand (COD) and ammonia nitrogen (NH4-N) of China’s livestock and poultry industry in 2010 were 11.48 million tons and 0.65 million tons respectively, which account for 45% and 25% of the total discharge of the country, and 95% and 79% of the total discharge of agricultural source.
Method and Data The pollutants discharge of large-scale livestock and poultry farms has been analyzed based on the hierarchical decomposition methods.
(8) (9) 1.2 Data source illustration and data processing The data used in the model for calculating breeding quantity and pollutant discharge is derived from 2010-2013 environmental statistics.
Structure effect presents an obvious affection on NH4-N discharge reduction of livestock and poultry breeding industry, but fails in COD reduction.

The evaluation from the data objects based on key attributes can reduce the data size and algorithm complexity.
After Clustering analysis of customers, then the evaluation analysis will process to the clustering data.
There are a lot of uncertain data of customer cluster, so the traditional method of classification and evaluation to the incomplete data is very difficult.
Tested by actual data analysis, cluster analysis can reduce the size of customer data and data noise, and the key class evaluation analysis can improve the evaluation efficiency of e-commerce customers.
Data Analysis Approaches Of Soft Sets Under Incomplete Information.

Introduction The theory of rough sets [1], proposed by Poland mathematician Pawlak in 1982, is a mathematical method to deal with insufficient and incomplete data.
So it is a set-theory-based technique to handle data, that is, through the known information to approximately describe the uncertainty concept [2, 3].
Reduction of a -consistent covering decision system.
Reduction of an inconsistent covering decision system.
Pawlak, Rough sets: Theoretical Aspects of Reasoning about Data, Kluwer Academic Publishers, boston,1991

These problems are caused by large-scale training sample set and outlier data immixed in the other class.
Moreover, for training the sample data mingled with outlier data in the relatively class of sample, it often can not improve the classification capability.
The ideal of SVM is to search for an optimal hyperplane to separate the data with maximal margin.
Of the 8st ACM SIGKDD international conference of knowledge discovery and data mining, Edmonton, Canada, 2002
Of SIAM International Conference on Data Mining，Lake Buena Vista, FL, USA,2004

But compare with these "massive" data, the ability of people to analyze the data and acquire knowledge exist a considerable gap, formed a "data glut" and "information poor" in a passive state.
Wisdom: Effective use of knowledge Meta-knowledge: Knowledge of the rules Knowledge: The rules of the use of information Information: Potentially useful of knowledge Data: Potentially useful information Noise: No obvious information Meta-knowledge Wisdom Knowledge Information Data Noise Fig.1 Hierarchy structure of knowledge In the hierarchy of knowledge, the lowest level (layer 6) is a noise which is almost made up of meaningless by the obscure nature of the data issues and the composition.
Level 5 is the data, some potentially significant issues.
Layer 4 is the information, is the result of a significant data processed.
The minimal attribute reduction sets is the serial code for {2,4,6,7,9,10,11,13,14}, TCF names are {Data communication, The importance of the system, Online data Processing, Multiple screens and multiple operations, Complex input and output, Complex internal processing, Code reusability, The perfect functionality and performance, Easy to maintain and modify}.

Without reactive power data, the system assumes that the power factor is fixed; without voltage data, the system makes use of the rated voltage.
In addition, with GPRS module installed in it, the meter can achieve wireless data transmission.
Therefore, decision support system for comprehensive power loss reduction of rural power network is designed with a common data interface.
In the process of data input and theoretical line loss calculation, the system can obtain existing data from other software systems avoiding manual input.
It ensures the accuracy of the data and improves efficiency of inputting data.