Search

When using international popular algorithms deal with RMB paper currency, there exist three main aspects. First, the size of RMB paper currency is different with other countries. Second, the color of RMB paper currency is different from other countries, and the different face values have different major color, pattern and texture. Third, the location of face value in paper currency of our country is different from other countries. For RMB paper currency image, using edge detection is to separate the paper currency from the image according to boundary. In other words, the background is removed and only paper currency is left. This paper firstly introduces the principles of removing background by gradient operator, LOG algorithm and Canny algorithm. Then it analyzes the edge image results obtained by these three kind algorithms. At last, these three algorithms are evaluated. Results show that the Canny algorithm can satisfy the requirement of removing background and can improve a good foundation for the subsequent of paper currency extraction.

This paper puts across the various approaches and methods that have been proposed in the context of Fuzzy Mathematical Morphology. The underlying principles of Dilation & Erosion, the structuring elements used in various techniques, the unique variations put forth by researchers, new applications in spatial relationships, decision making, segmentation of medical images have been discussed.

With the deepening of the degree of oil exploitation, investigation of geological structure is particularly important, especially those faults that have an important impact on the exploration and development of oil. However, seismic data is affected by various kinds of factors in the progress of data acquisition, which reduces SNR and interfere with the accuracy of geological structure interpretation. This paper based on image processing provides fault enhancement medthod. It can reduce random factors impact and depicte fault more clearly. This method combine anisotropy and orientation information of image, then use generalized Kuwahara filter to enhance fault. This technique has a most important value in seismic fault interpretation.

On the basic of multiple populations of immune algorithm and clonal selectionalgorithm, the purpose of this paper is to further improve the detectionefficiency and reducing the false alarm rate. This paper uses the kddcup99 dataset as the experimental data set, and chooses four types of attack data groupof experiment data set as initial population of multiple populations of clonalselection algorithm, through the algorithm to create the optimal model. Basedon the principle of normal data larger than the abnormal data, in turn,experimental data set matched with the normal data set and the optimal model bythe improved R matching algorithm. The results of this paper show that thedetection rate increased significantly

In view of the mine special environment, with passive millimeter wave imaging technology applied to mine body hidden contraband detection as the background, this paper has design of miner detection system. This paper gives the miner detection system composition block diagram and working process, and design the data acquisition module, give the data acquisition module internal logical structure.

This paper studies the power transformer fault quality diagnosis using rough sets theory and neural network. It is rough sets reduction as the pre-unit of neural network based on reduction algorithm with the attribute significance. The paper describes the reduction algorithm and implementation method detailed. Through the training and testing results with practical data, it is proved that the reduction algorithm with the attribute significance can make the number of input samples shorter, the training speed faster and the diagnostic accuracy higher. The algorithm is feasible and effective for applying to the fault diagnosis system of power transformer.

Teaching evaluation is to process the information of the huge amount of teaching data and to evaluate the teaching activity, process and outcome to provide the basis for the quality and decision of teaching. S-P Chart is a typical evaluating method. The paper introduces both traditional and fuzzy S-P chart. Then, the paper proposes three improved S-P methods, including single-knowledge-point analysis, multi-knowledge-point analysis and cognition analysis. All these three improved methods supplements the shortages of traditional S-P and fuzzy S-P, and are implied in an analysis system of the quality of teaching.

Warfare simulation scenarios will play a vital role in helping commanders to make decisions, nowadays. However, one warfare simulation scenario usually includes lots of various information with complicated relations and is not so easy to form. Thus, this paper focuses on the presentation of a basic framework of the storage of warfare simulation scenarios, and also the improvement of management capability of them based on the methods of ontology and CBR. Through researching the process of joint air defense warfare, the paper constructed the warfare scenario ontology in that domain based on the method of first-order logic and, in the following, has given the proof of its consistency and completeness. Meanwhile, the paper has also illustrated one case using one feature set and carry out case retrieval based on this expression, which has improved the effectiveness of scenario management.

To solve the problems such as randomness user behavior, effective privacy protection and massive data processing shortage, this paper has proposed a household load data processing method with privacy protection, combined data mining and cloud computing. First, it gave the platform architecture and individual protection model. Then, it proposed Mask-k_means database encryption method, algorithm parallelization was implemented by MapReduce. Finally, this paper varied the method was and implemented household load data processing based on statistics and real-measured data respectively. One conclusion is that feasibility of cluster analysis for multi-user was low, while single-user load analysis is high. The other is that this method is simple and practical. This can provide a new way to household load big data under smart electricity consuming.

Looking at the high rates of production and the steep competition in the world market, it becomes quite essential that the fault control is done in a very efficient way. This article presents a summary on the maintenance, the monitoring techniques, and the diagnosis methods for the condition based maintenance of machine tools. The paper initially gives a brief introduction on the condition based maintenance of machine tools. In the next part, the various methods for the monitoring are discussed followed by the models for data mining. The paper concludes that most of the techniques have their own advantages and drawbacks, so a careful selection of the techniques is needed to form a proper monitoring system.

In order to remedy the defects of traditional methods in the data mining, improving the data mining effect. The paper proposed the improved TFIDF algorithm and applied to the data mining based on the analysis of the flaw and the insufficiency in simple calculation method, minimum value calculation method and the traditional TFIDF algorithm. The paper proved that the improved TFIDF algorithm is better than other traditional methods more scientific, more advantages and practical value based on four data mining results are compared and evaluated.

In order to meet the current needs of the report dealing with large volumes of data, operation difficulties of data sources, fixed form and other issues, The paper raises that basing on principles of modular and hierarchical design thinking, preprocess various data sources in dimension, store in special data base, then dynamically generate report through a combination of different dimensions in the hierarchy. The report system in this paper is designed 4-layer architecture in hierarchical relationship, ensuring that coupling value of each part of the system is a minimum, and this improves system module’s universality and the efficiency of dynamic report system’s deployment and execution.

Non-stationary and nonlinear features in the dynamic trajectories of traffic parameters may stop rich and hidden multi-dimensional knowledge from being extracted, contributing a correlation result with false judgments or ignorable significance. As the fundamental improvement for a dynamic congestion analyzing method, the paper decomposes a time series of traffic data into 2 components based on a polynomial approximation method, called the “trend” and “detail” components. In order to simplify the selection of the fitting function to obtain the trend component, the paper proposes an adaptive partitioning method based on the sensitivity search technology. The given examples prove the effectiveness of the proposed methods for a spatiotemporal traffic congestion study.

The evaluation of surface water quality is a problem which relates to whether the environmental function and value do possess or not in the water. A new comprehensive evaluation method which combines Principal Component Analysis (PCA) and the weighted grey correlation method was presented in this paper. The reduction of indicators and the weakening of the multiple correlation among indicators were considered in Principal Component Analysis. According to different functional areas categories of every section, the weighted grey correlation method could determine the evaluation indicator’s weights, strengthen the effect of concerntrations of indicators and consider each indicator’s impact. The Mudanjiang River downstream sections were used as the research object. Based on the water quality monitoring data of typical monitoring section in 2013, the study on the environmental quality assessment of downstream section in mudanjiang was given in this paper. The efficiency of the evaluation process and the accuracy of the results can be improved.

Automatic dependent surveillance-broadcast (ADS-B) is a totally new surveillance method, so before practically applied and operated in China, abundant tests and evaluations are necessary to validate the performance of ADS-B and guarantee the operational security. To solve the asynchronous problem among radar data, ADS-B data and real-time kinematic (RTK) data caused by different update rates, this paper proposes the technique of synchronizing multi-surveillance data by extrapolating from the data of low update rate to high update rate according to velocity and heading. Meanwhile, because radar data, ADS-B data and RTK data are expressed in different coordinates and cannot be compared each other, this article provides a method to unifying the coordinates of multi-surveillance data. By the analysis and evaluation, we can conclude that the performance of ADS-B is better than radar. Addtionally, the paper makes focus study on the position quality indicator and the encoding rules, and puts forward the conversion between the different rules. The paper provides the technical guide for the manufacturers and the airlines.

This paper made a in-depth research on the application of data mining in academic management system based on overall understanding of data mining, and selected the data sample of students in grade 2007 of Logistics Management major of Logistics School of Beijing Wuzi University, this paper used Clementine software to establish decision tree model and clustering model to analyze students’ cluster characteristics by taking students’ academic achievement, score of college entrance examination, CET 4 and CET 6 level, students’ family income and psychological health, the research can help school notice development features of students in order to improve way of students’ cultivation.

The target of this paper is to introduce the improvement of the technique of image retrieval. At first, it comes up with the concept of image retrieval and shows the importance of this technique. As the techniques of multimedia and Internet are developing rapidly, the resources of images that users obtain are also extended. And then this paper gives the problem about the image retrieval, namely the information of images are disordering. As the result, it is significant to do the effective organization, management and retrieval based on the increasingly extensive image information storage. After that, this paper presents the concept of TBIR and CBIR and gives the definitions of them. It proposes an issue that CBIR is the improvement of TBIR. Based on CBIR, there are also some disadvantages that need to be improved. In terms of the main point of CBIR, the paper raises that the annotation is one of the most difficult techniques that need to be promoted. Then it describes some algorithms about the technique of automatic image annotation. After these algorithms, the paper shows the challenges and developing direction of the technique of image retrieval. At last, it presented the conclusion to emphasize the main points of this paper.

Materialized view is an important topic in data warehouse research, and also affects the query efficiency and maintenance cost. The disk-space view-selection problem is to select a set of materialized views for the purpose of minimizing the total query processing cost and the total maintenance cost. In this paper we introduce evolutionary algorithm using stochastic ranking algorithm, which can enable materialized view selection under disk-space constraint. The algorithm improve the stochastic ranking algorithm, which can find a near-optimal feasible solution. This paper use the algorithm in Police data warehouse.

Data mining is the general methodology for retrieving useful information from big data. Clustering analysis is a mathematical method of classification for unsupervised machine learning. It can be adopted for data classification in Data mining. This paper combines the clustering process by fuzzy way and then deduces a special clustering algorithm with fast fuzzy c-means (FFCM) method. In summary, the paper illustrates the adoption of a series of fuzzy clustering methods in Data Mining. These methods have improved the computational efficiency with learning as the convergence speed is fast. The methodology of this paper presents significantly meaningful for information retrieval of big data.

Analysis of user behavior management control has important significance for network users. This paper uses the clustering algorithm to cluster the user behavior for clicking certain software or Website in different times, through the time context of user behavior to analyze user behavior rules. Additionally, this paper analyzes the clustering result in detail, then, divides user behavior into different types. Finally, according to the clustering result, more targeted pages and applications are recommended to the network users to create huge business value.

An algorithm which named Boosting algorithm occurred in the last ten years. It can raise the learning algorithm accuracy with multiple learning and obviously improve the efficiency of learning algorithm by adopting the principle of “comprehensive optimizing”. In addition, it can effectively develop the “weak learning algorithm” with low efficiency into a “strong learning algorithm” with high efficiency. The boosting, a new learning method to integration machine, is based in learning theory and show its good qualities in many fields. The paper elaborates and summarizes the basic ideas of Boosting algorithm and applies it into Data Mining (DM). In recent years, the DM has been extensively adopted out of labs such as in commerce, technological research and engineering technology. In this background, the paper tries improving traditional DM algorithm to solve these problems in industrial application.

— This paper demonstrates to build a Fuzzy Inference System (FIS) for any model utilizing the Fuzzy Logic Toolbox graphical user interface (GUI) tools. A different conception for decision making process, based on the fuzzy approach, is propounded by authors of the paper.The paper is worked out in two sections. Description about the Fuzzy Logic Tool box is done in the first section.Illustration with an introductory example concludes the second section. Based on various assumptions the authors construct the rule statements which are then converted into fuzzy rules and the GUI tools of the Fuzzy Logic Toolbox built using MATLAB numeric computing environment is used to construct a fuzzy inference system for this process.The output membership functions are expected to be fuzzy sets in Mamdani-type inference.Defuzzification of fuzzy set for each output variable generated after the aggregation process has to be carried out. Application of information technology for Decisions in today's environment which is highly competitive are undeniable principles of organizations and helps managers in making useful decisions meaningfully.

The main idea of rough set theory is to extract decision rules by attribute reduction and value reduction in the premises of keeping the ability of classification. This paper presents the design of model for customer division based on rough set, and uses algorithms for attribute reduction and rule extraction in rough set to analyze the customer of supermarket. This paper also introduces how to achieve the minimum result of attribute reduction and decision-making via decision-making report, winkling redundant attribute and over-rule of decision.

Currently, rough set theory has been widely used in many fields. In rough set theory, how to measure the attributes significance of the data is a core content. In order to solve the problem that the existing attributes significance measure methods usually ignore the interaction among the attributes, the paper presents a measure method based on difference degree. When given a set, the proposed method first divides it into several subsets according to the value of condition attributes, and then computes the difference degree in the subsets. Secondly, the important attributes are selected based on the value of difference degree. Further the paper discussed some properties of the difference degree, and the experimental results shows the effectiveness of this method in the final.

According to the deficient situation of study in merging unit performance evaluation, this paper set up a comprehensive evaluation model for merging unit performance evaluation. Introduced triangular fuzzy number and determined the index weight through certain processing conversion, and set up a scientific system of performance indicators, and then selected normal distribution as the membership function. On this basis, this paper established a hierarchical and sub-aims comprehensive evaluation model. Finally, the model was applied for evaluating a merging unit performance. The evaluation result showed that the weight was effective and the fuzzy comprehensive evaluation model was simple and reliable.

This paper focuses on the correlation analysis method based on vector space model. In the case of dual classification, this paper made a Joint comparison to find the most appropriate method of selecting featured items for the focused crawler; and then made special effort on analysis and verification of LBTF-IDF algorithm in which the weight calculation method has been improved.

This paper mainly discusses the sites of complex mountainous area wireless communication stations which is based on the analysis of the top digital model in order to highlight the advantages and functions of GIS in the practical application. This paper is thought on the theory of visual analysis, which is based on the digital model. According to the characteristic value of terrain data extraction, the obtaining and conversion of interpolation parameters as well as the features of characteristics and hope vector coverage area, the verification of signal coverage appears to be consistent with the shape visibility analysis of the base station position. With the assistance of professional ARCGIS, it can be easily to track the data and complete the 3D analysis and surface analysis of the typical complex mountainous terrain. Finally, the paper presents the prediction, preliminary selection, adjustment, optimization and results of station construction.

Increasingly serious environmental pollution,trying to find a effective method to control NOx emission become more importance. Under this background, this paper adopts the naïve Bayesian classifier method which built on the basis of the probability density function to forecasting the NOx emission of diesel engine. This paper proposes a new approach to weight the super-parent one dependence estimators, and uses the UCI datasets to verify the validity of the proposed method. Finally, apply this diagnosis technology to the collected WD615 diesel engine data. The comparison experiments with other algorithms demonstrate the effectiveness of the proposed method.

The paper centers on the character of fault modes recognition for control system and introduces intelligent diagnosis based on signal computing that is called fault recognition system. The key point and direction in recent research about fault recognition is given out. Later classifier of the fault recognition system and its character occupy an important part in the paper. At last application prospect of pattern recognition in fault diagnosis is stated briefly.

Grey system theory has obvious advantage on handling “poor information and small sample data” problems. Some TV seeker’s faults performance of such characteristics that the date is less and not integrity, and failure phenomenon is not explicit. So it comes to be difficult to diagnose the faults at the Basic level by a quick way. This paper means to build a model using the general gray relational analysis in order to establish contact between the fault parameter names and failure phenomenon. Then by using gray relational analysis on the fault records in the database and checking the model with another data. Calculation result shows that the method can judge fault position simply and quickly. The paper provides a kind of relatively reliable quantitative analysis basis for the fault diagnosis and locating faults.

In this paper, we present an approach to characterize and model microblog traffic in cellular data network. In contrast to previous methods, our approach is based on the cloud computing platform and the cluster system, including the Hadoop Distributed File System (HDFS) and the parallel processing software framework MapReduce. Whats more, we focus on the contrast of Sina and Tencent microblogs. We analyze the features of microblog traffic in four aspects of increasing details, which are (i) traffic diurnal pattern, (ii) modeling the traffic distribution, (iii) user distribution, (iv) diversity usage of microblogs. This approach of analyzing microblog traffic comprehensively is probably the most important contribution of this paper. Furthermore, our approach has two important features. First, the massive mobile subscriber data we used in our experiments was collected from a commercial Internet Service Provider (ISP) covering an entire province in Southern China. Therefore, it ensures the results indicate the true characteristics of microblog traffic in network. Second, we investigate that the microblog traffic fits with the power law distribution. We demonstrate the electiveness of our approach on three real datasets. Our results are important for cellular network operators to learn user behavior and optimize the future microblog application designs.

Three-dimensional model is high degree of freedom information. In order to realize content-based 3D model retrieval, normalization preprocessing of 3D models is needed. This paper have proposed a normalization method with 3D modeling software based on center of mass, principal axis of inertia and volume parameters to carry out coordinate system normalization of three-dimensional mechanical entity models. In this paper, Specific method is listed. Application examples are given to show the feasibility of this method.

Fuzzy clustering is a kind of language approximation method. It is a multivariate analysis method to classify samples based on the attribute diversity by using fuzzy mathematic method. From the diversity and complexity of impact tendency indicators of coal mine and rock, paper started with extreme value standardization and combined the classification indicators of impact tendency of coal mine and rock, then determined the fuzzy rating, membership function and weight set. Afterwards, paper established the fuzzy similar matrix by using cosine method, then transformed it into equivalent matrix, and did the fuzzy clustering analysis at last. The result shows that the method overcomes the irrationalities which made by existence of impact tendency, achieves a more accurate classification and evaluation, and provides a new method to the comprehensive evaluation of coal rock impact tendency.

According to the internal mechanism of the formation of heat load, the formation of heat load consists of two parts, the systemic heat load, which is determined by the building envelope and outdoor environmental parameters and random load caused by the users randomness of events and solar radiation etc. Toward systemic heat load, this paper considered the influence of environmental parameters before the prediction time and used the method of stepwise trials and MSE to obtain the optimal solution. Toward random load, it is considered that the day of the same type have the same variation pattern. On this basis, this paper introduced a correction coefficient to obtain random load eventually. This paper selected DeST, the widely used energy simulation software in China, to analysis the case. The result shows that the prediction method is feasible and 50% of the predicted loads have the relative error of less than 5%.

In recent years, the increasing Debris flow has been drawing the geologists' attention for its massive destructive force. At present, the research on the debris flow are mainly based on comprehensive multi-factors evaluation or singlue factor qualitative anlalysis. However, all factors have the impact on the debris flow. At present, the research on the factors that foundamentally affecting the debris flow or its character is quite few. This paper use the method of stepwise discriminant analysis to minimize the correlation and repeatability between factors to ensure the selected factor can be fully represent the character of debris flow. In addition, classification is an important way to illustrate general picture of the debris flow. This paper classified the selected 6 key factors by method of FCM (Fussy C Means) to achieve the transition and evolution from general geological classification to the accurate quantitative analysis. The result of such classification will benefit the debris flow cognition and disaster prevention measures.

FTAe is software tool for Fault Tree Analysis in Microsoft Excel. This paper is focused on short description of created FTAe. A brief description of Fault Tree Analysis is in the first part of paper. The Structure, Logic and Minimal Cut Sets in Fault Tree analysis is described too. Next part of this paper is focused on the choice of algorithm for cut set calculation and brief description of structure of FTAe. Finally, advantages of FTAe software in Microsoft Excel are described in conclusions.

The present paper attempts to clarify which priority rule is appropriate for specific manufacturing environment variations, such as processing time variations, operation additions, and additional job releases. In such manufacturing environments, the performance of scheduling rules should be evaluated with a degree of robustness under dynamic situations and traditional measures, such as makespan, under static situations. The present paper evaluates the robustness of the priority rules FCFS, SPT, MOPNR, and MWKR from the viewpoint of both dynamic and static situations based on the similarity of schedules before and after variations and based on the rate of makespan deterioration. A new measure of the similarity of schedules is proposed. Guidelines for selecting priority rules, such as avoiding rules that directly use the uncertain information in calculating the priorities, are provided.

This paper describes the extending model of Multi-mode Failure Models by using the Weibull and Gamma distribution models presented in a conference [1,2]. Different than the models in the previous papers which are for variable test data, in this paper we will describe the use of attribute test data for our model. In reliability theory, the most important problem is to determine the reliability of a complex system from the reliability of its components. The weakness of most reliability theories is that the systems are described and explained as simply functioning or failed. In many real situations, the failures may be from many causes depending upon the age and the environment of the system and its components. Another problem in reliability theory is one of estimating the parameters of the assumed failure models. The estimation may be based on data collected over censored or uncensored life tests. In many reliability problems, the failure data are simply quantitatively inadequate. The Bayesian analyses are more beneficial than classical analyses in such cases. The Bayesian estimation analyses allow us to combine past knowledge or experience in the form of an apriori distribution with life test data to make inferences of the parameter of interest. In this paper, we have investigated the application of the Bayesian estimation analyses to multi-mode failure systems for attribute test data. The cases are limited to the models with independent causes of failure. We select our investigation by using the Multi-nomial distribution as our model. This distribution is widely used in reliability analysis for attribute test data. This model describes the likelihood function and follows with the description of the posterior function. A Beta prior is used in our analysis for each model and it is followed by the estimation of the point, interval, and reliability estimations.

The price of the wood according to the type of wood. Classification of the woods can be done by studying its texture. This paper introduces Fuzzy k Nearest Neighbor to classify 25 types of wood. The woods images have been taken from the Wood Database of the Centre for Artificial Intelligence & Robotics, Universiti Teknologi Malaysia. The features of wood images are extracted using Local Binary Pattern. The results of this paper shows improvement in wood classification compare to the previous literature.

Contaminated sites remediation is one of the hot points in environmental protection fields. Therefore, an important research project is how to expediently determine the bound and level of pollution of contaminated sites. This may help to decide on a timely and accurate rescue plan so as to minimize the costs and to improve the effect of restoration. This is also the basis of contaminated sites remediation acceptance. There are many influencing factors to confirm the remediation areas on contaminated sites. Studies show that three-dimensional interpolation method is effective tool to delimit the pollution scope. Different three-dimensional interpolation methods have different results in contaminate distribution prediction and directly affect the boundary of contamination. This paper selects one specific case with heavy metal lead contamination to study three-dimensional interpolation methods. The three-dimensional stratigraphic model is established for uncertainty analysis and error analysis in the grid through KrigingIDW(Shepard) and nearest neighbor interpolation method. Results show that 40m*40m grids have the highest accuracy of stratigraphic simulation. Nearest neighbor interpolation method is most suitable for stratigraphic model building. Kriging interpolation method is more suitable for constructing three-dimensional pollutant prediction model. On the basis of above, this text established one distribution prediction model of soil lead contamination, and calculated the contaminated earthwork quantity under different remediation goals. Finally, the paper achieved three-dimensional visualization of the site soil contamination. This study can be directly applied to the actual site remediation decision process, and is helpful for the contaminated sites evaluation and restoration in the future.

A short-term PV system power forecasting method is presented in the paper based on neural network considering fuzzy characteristics of weather factors. Weather factors that affect PV system power output mainly include temperature, radiation intensity, rain and relative humidity which are all of strong fuzziness. The paper firstly made use of membership functions to process their fuzziness. Then, the historical power data of a PV system was put into neural network together with fuzzy processed historical weather data to train the network, therefore, neural network that be able to forecast PV power was get. Finally, data of an actual PV system in Colorado was employed to methods with and without fuzzy processing of weather factors, results show that the method with fuzzy processing is more accurate than that without fuzzy processing.

This paper tries to use Data Envelopment Analysis (DEA) as the foundation, combining Fuzzy Comprehensive Assessment (FCA) method and Analytic Hierarchy Process (AHP), to make decisions for the project. First we use FCA method to analyze and quantize each index, then use AHP method to determine the weight of each index, use Evaluation Model of DEA Cross-efficiency to reduce the error of expert evaluation and weight in the FCA method. This paper will give the detailed steps. Finally, we use a case to test the reliability and feasibility of our method. We think, this paper will provide a now method for the Construction project bidders.

This paper introduced the ICA theory in the process of analyzing the entrepreneur mental factors which affect the value chain, and analyses the feasibility of applying ICA theory in the extraction of the entrepreneur mental factors. In the empirical study, we separated the main factors influencing value chain by the FastICA algorithm based on ICA theory, and the experimental results are in conformity with theory.It is feasible to sift Entrepreneur mental factors,the research results of this paper also provides a theoretical reference for the application of the ICA theory in value chain management and optimization.

This paper presents an improved GSA-GA algorithm to achieve bad-data detection, identification and correction in power grid. The algorithm combines BP neural network, K-means clustering algorithm, gap statistical algorithm (GSA) and genetic algorithm together. BP neural network preprocesses the data, K-means algorithm clusters the preprocessed data and GSA algorithm determines the optimal clustering number and identifies the presence of bad data. After identifying the bad data, GA-BP algorithm is used to correct the identified data. This paper takes simulation tests to verify the proposed algorithms correctness and effectiveness based on actual grid data considering multiple types of existed bad data.

Gene expression profiles of tumor have the limited amount of samples in comparison to the high dimensionality of the samples;this paper proposed a classification algorithm based on neighborhood rough set to improve classification accuracy.This paper first applied feature filtering method of kruskal-wallis rank sum test to select a set of top-ranked related genes, and then applied neighborhood rough set on these genes to generate a informative genes subset. Finally, SVM was used to classify the GEP data set. The result of the experiment indicates that this method can effectively improve classification accuracy, and it has higher generalization.

due to the time sequence has some unique characteristics, problems with the traditional point to point distance method are realized by more and more researchers when measuring the similarity of time series. Therefore, this paper uses a new similarity measure, i.e. global characteristics, extracting values of statistical distribution, seasonal, spectral and other characteristics from time series to construct eigenvector, based on the characteristic vector clustering analysis. By conducting time series clustering of communication base stations in a region, this paper confirms the method is able to deal with global characteristics that has a larger scale of time series, and to calculate effectively large time series.

This paper proposes a dimensionality reduction mathematical model based on feedback constraint for High-dimensional information. It uses feedback restriction technique to construct dimensionality reduction model for multidimensional product data. The data obtained is with high latitudes, where a large number of data are under components involved standardized restrictions. High-dimensional data participating in operation will increase the complexity of operation, and hence, we need to reduce its dimension. In this paper multi-constrained inverse regression model is adopted to reduce the dimension of cloud resource scheduling data in multi-constrained environments. Experimental results show that the proposed method increases the data coverage rate of high-dimensional data mining by 66%, and has great optimizing effect.

The paper introduced the Bayesian networks briefly and discussed the algorithm of transforming fault tree into Bayesian networks at first, then regarded the structures impaired caused by tunnel blasting construction as a example, introduced the built and calculated method of the Bayesian networks by matlab. Then assumed the probabilities of essential events, calculated the probability of top event and the posterior probability of each essential events by the Bayesian networks. After that the paper contrast the characteristics of fault tree analysis and the Bayesian networks, Identified that the Bayesian networks is better than fault tree analysis in safety evaluation in some case, and provided a valid way to assess risk in metro construction.

Dynamic traffic data quality evaluation can provide reliable data support for the traffic management system. This paper aim at three kinds of common failure data, put forward the traffic flow data quality control methods and process. First time evaluate the data quality in three stages, and each stage respectively for all kinds of ITS detector, and builds the six index system. Finally take Beijing as an example to analyze and verify the method of this paper.

This paper presents an overview of research works on the utilizing of soft computing in the optimization of process parameters and in the prediction of thin film properties in sputtering processes. The papers from this review were obtained from relevant databases and from various scientific journals. The papers collected were published from 2008 to 2012. The focus of the review is to provide an outlook on the utilization of soft computing techniques in sputtering processes. Based on the review, the soft computing techniques which have been applied so far are ANN, GA and Fuzzy Logic. The first finding of this review is that soft computing technique is a promising and more reliable approach to optimize and predict process parameters compared to the traditional methods. The second finding is that the utilizing of soft computing techniques in sputtering processes are still limited and still in exploratory phase as they have not yet been extensively and stably applied. The techniques applied are also limited to ANN, GA and Fuzzy, whereas the exploration into other techniques is also necessary to be conducted in order to seek the most reliable technique and so as to expand the application of soft computing approach. Future research could focus on the exploration of other soft computing techniques for optimization in order to find the best optimization techniques based on the specific processes.