Search results

Under this kind of background, the author who has done the research and put forward some specific algorithms on the aspects of data reconstruction, multi-view data registration and data reduction.
Under the menu bar, a row of toolbar buttons' main functions are reading and storing data documents, changing observed view, dividing data space, data registration and data reduction.
Algorithm of Point Cloud Data Reduction Based on Curvature.
The three algorithms of topology reconstruction, data registration and data reduction are detailed in the literature [6, 7].
(in Chinese) [6] Xiao-lei Du and Yong Zhuo: A point cloud data reduction method based on curvature.

But people can't find relationships and rules that exist in the data and can not predict the future trend of development according to existing data.
And how to refine data mining attribute sets (called the attribute reduction) becomes a key link in data mining.
CART algorithm which not only can handle highly inclined or polymorphism numeric data, but also can handle sequence or unordered class attribute data; SPRING algorithm that through two data structure of attribute table and class histogram can be a real processing of super-large data from a database; PUBLIC algorithm which put forward the objective function of node cost and improved decision tree of efficiency.
[2] Ling Han, Xue-Gang Hu, etc.: An Algorithm of Attribute Reduction Based on Data Analysis Method.
Mehta: SPRINT: A Scalable Parallel Classifier for Data Mining.

However, the interaction between the cluster evolution and the dimensionality reduction is the most common scenario in the time decayed stream data.
Therefore, the dimensionality reduction should interact with cluster operation in the endless life cycle of stream data.
A data stream is defined as a sequence of unbounded, real time data items with a very high data rate that can be only read once.
From the dimensionality reduction point of view, to eliminate one attribute from the source data set means that the resulted data set become denser than the original data set, and vice versa, to append one attribute to the data set means that the resulted data set become sparser than the original data set.
Therefore, multi-fractal dimension of the data set will superior to single fractal dimension on the dimensionality reduction operation.

Vehicle networking data analysis based on Mapreduce Shu-li Huang College of Information Engineering, Jiangxi University of Technology, Nanchang 330098, China Keywords: car networking; data mining; Attribute Reduction Abstract: The number of vehicles using the rapid development of today, in order to better monitor the vehicle and road traffic, must find the data they need from the vast amounts of data, thus providing a good way to find the target from the mass of data is very important.
Reduction theory for these kind of little change in the amount of data and the data is not fruitful, but for those who handle incremental data or mass data often have significant limitations.
To accommodate this massive amount of data or data, LIU Zong-tian proposed incremental reduction algorithm based on rough set; Wang Jue put forward based on the difference between a distributed matrix reduction algorithm;.
Parallel reduction can be stable and strong generalization ability and adapt to the dynamic changes of the data reduction, it has all the advantages of a dynamic reduction, but to overcome the dynamic reduction defects.
Mapreduce The ultimate goal is to make data deduplication data appears more than once in the output file only once appeared in the original data.

A special kind of data is considered in this paper called multimodal data.
The purpose of dimensionality reduction is to find the low-dimensional embedding of high-dimensional data and keep the most of “intrinsic information”.
In many practical applications, one kind of data that we frequently deal with is the multimodal data.
So it is needed to explore dimensionality reduction methods on multimodal data, and several attentions have been attracted in recent years [1-5].
Dimensionality reduction of multimodal labeled data by local Fisher discriminant analysis.

The article chose data of environmental management investment and exhaust gas and waste water discharge amount of Chinese thirty major cities to analysis the regional differences in pollutants discharge reduction and environmental investment.
To this end, we chose the environmental investment and major pollutants discharge data of Chinese thirty major cities and analysed the regional differences in environmental investment and the effect of pollutants discharge reduction based on panel data models.
Time span is 2005-2010, the data comes from CEIC database.
In order to fully extract information from sample data, we chose panel data models.
Analysis on Environmental Pollution Decomposition under the Open Economy Based on Local Panel Data with a Time Span 1990-2003.

Keywords: energy saving and emission reduction, data reduction factor, super-efficiency DEA, efficiency evaluation Abstract: Data Reduction Factor (DRF) and the Super Efficiency DEA (SE-DEA) model are employed to study the energy saving and emission reduction efficiency of Shanxi Province.
Tab1 shows the input and output index data.
Data processing and the model operation Data processing.The data must be positive and dimensionless because they have different natures, units and dimensions.
There are many ways to convert the data into dimensionless data, and the paper adopts the threshold method.
China’s Commercial Banks Risk and Efficiency Study －Index Selection based on Data Reduction Factor [J].

Measured Scan Data Reduction Combination Algorithm Based on Scan Line Shigang Wang1, a, Yong Yan1, b, Fengjuan Wang1, c 1School of Mechatronics Engineering, Qiqihar University, Qiqihar 161006, China ahljwangsg@163.com, b185662045@qq.com, cwfj320110@163.com Keywords: Reverse Engineering; Laser Scanning; Data Reduction; Combination Algorithm Abstract.
Definite the data points to simplify as, and the data amount, is expected to simplify.
If have fetched, the state measurement data reduction is over.
Idea of the algorithm is: (1) the first the minimum distance method is used to data compression, streamlining smaller part of the point cloud data; (2) to eliminate the point cloud data reuse the angle-chord high combined code algorithm for secondary to streamline.
It is achieved that more data points are kept at high curvature; in the low curvature more point cloud data is cut, which conform to the principle of lean, as shown in Fig. 5.

Then, based on actual survey data and synthesis emission factor, the paper analyzes the energy-saving and emission-reduction effect of Jinan BRT.
Based on actual survey data, the paper analyzes the energy-saving and emission-reduction effect of BRT.
Some survey data for subsequent calculation is shown in Table 1.
Conclusions Based on actual survey data, the paper analyzes the energy-saving and emission-reduction effect of Jinan BRT.
However, the paper only surveys two conventional bus lines, and the data is limited.

Towards Efficient Dimensionality Reduction for Evolving Bayesian Network Classifier Wang LiMina , Li XiongFeib and Wang XueChengc College of Computer Science and Technology, JiLin University, ChangChun 130012, China awanglim@jlu.edu.cn, blxf@jlu.edu.cn, cwxc@jlu.edu.cn (corresponding author) Keywords: Bayesian Network; Dimensionality Reduction; Mixed Data Abstract.
Experimental Study We chose 12 data sets from the UCI machine learning repository for our experiments.
For almost every data set, HBC achieves slightly better classification performance than Naïve Bayes Since most of the data sets have just a few relevant attributes.
Experiments were performed on data sets from the UCI repository.
It has been observed that for certain data sets only few parameters are used. 6.