Earticle

This paper has been explored information data prediction implementation access based on data mining combination model. With data mining technology as the entry point and in combination with the analysis on information data prediction characteristics. Research on variable substitution to non-linear regression forecast model precision's influence, and seek the modeling method that can improve the forecast precision. Based on the Data mining, the transform in space and the weighted processing combined method, make full use of information that the primary data provide. Given modeling method of combination forecast model based on the Data mining. Based on Data mining’s combination forecast model’s modeling method can reduce the serious influence that the variable substitution brings and has fully used useful information in the primary data. It obviously improved the accuracy of the prediction model.

Online learning began to spread of the Internet in the 1990s. Online education has now changed into a smart learning Smart learning has requested an extension to the mobile environment in a number of e-Learning content that has been fixed to the existing web environment. It was learning to enable anywhere, anytime, regardless of location, and the user has enabled the study of the interaction to the content. In this study, we studied the e-learning authoring tools that meet the requirements of the mobile environment. It is provided to the user in the form of SaaS, and it provides a method for creating the e-learning content made HTML5 technology. In addition, we analyze big data of various learners, the goal is to provide the personalized learning as integrated e-learning solutions.

Threats in Internet of things are ubiquitous such as counterfeiting, product piracy and product recall. China is no exception to this trend. The reader SoC (system on chip) chip of Ultra high frequency (UHF) Radio Frequency Identification is the key technology to solve these threats. Due to RF technology, tag data is read and written through wireless transmission directly in the air. In order to avoid tag theft and related backstage database attack, we provide UHF High Security System (UHS-HSS) to prevent the tag data monitoring in third party equipment. UHF-HSS regard UHF RFID reader SoC chip technology as the technology foundation provides chip level security solutions, system level information security service and industry level security applications for the IOT. This paper introduces a complete set of software platforms based on UHF RFID sensors including the underlying Linux operating system and related device driver, IOT platform technology, RFID middle-ware technology and software platform application. It solves the critical problem of security and reliability of UHF RFID applications for the national economy, which is of a great significance for the development of China's Internet of Things technology.

Short text similarity measure is the basis of classification and duplicate checking of the short texts. Allowing for the insufficient consideration of the sentence semantic and structure information in similarity calculation between two short texts, we propose a novel method of short text similarity calculation based on double vector space model on the basis of traditional vector space model. Creatively transforming traditional vector space model into double vector space model. We utilize the numeral data link relations of Wikipedia to calculate semantic similarity between words, and calculate text structure similarity by dependency trees. Finally, we get the synthetic similarity by combining the semantic similar vector and structure similar vector. Our experiment results demonstrate that the proposed method has higher accuracy than other methods.

Data provide the inputs to systems used to understand, explain, manage, regulate, and predict the world in which we live. A basic question in studying data is, What are data? What are and are not data, and how do data become information? Exploring a conception of data is fundamentally a philosophical problem and also an important issue in the area of database design. A firm understanding of the nature of the data being modeled enhances the process of modeling reality, and it helps in establishing a mental map of the computerized domain. This paper proposes a basic definition of data as interpreted things that flow. This definition is used in building structured data (e.g., tuples, tables) that form the foundation of database systems. The notion of things that flow is a concept based on a flow-based modeling language established on machines (extension of the input-process-output model) that create, process, release, transfer, and receive these things that flow. The study uses the proposed basic definition of data to build structured data, hence, applying the definition in constructing a data-based description of particular aspects of database systems.

In allusion to the shortcomings of easy falling into the local optimization and difficult obtaining Pareto optimal solutions for the original ant colony optimization algorithm in solving the complex optimization problems, multi-population, parallel mechanism, dynamic evaporation strategy and chaos theory are introduced into the original ant colony optimization algorithm in order to propose an improved multi-population ant colony optimization(MPPDCACO) algorithm in this paper. In the proposed MPPDCACO algorithm, the ant colony is divided into scout ants, search ants and worker ants in order to make the ACO algorithm as far as possible to avoid falling into local optimization and improve the local search ability of ant colony. The multi-population parallel mechanism is used to exchange the information and improve the computational effectiveness. The dynamic evaporation strategy is used to dynamically adjust the evaporation coefficient of pheromone in order to improve the global search capability of the ACO algorithm. The chaos theory is used to realize the optimization search in order to obtain the pheromone distributing in choosing path process. So the proposed MPPDCACO algorithm can prevent the local convergence caused by the misbalance of pheromone and can improve the searching ability. In order to test the optimization performance of the proposed MPPDCACO algorithm, 6 traveling salesman problems are selected from the TSPLIB in here. The experimental results show that the proposed MPPACACO algorithm takes on better global searching ability and higher convergence speed.

Under the conditions of different community formation, this paper proposed two different models of formation communities. Firstly, we put forward two kinds of similarity calculation models, and compare them with the traditional similarity model, Secondly, several similarity models are tested under different conditions of community formation. Finally it compares tow models of forming communities and finds that for non-strict division of community model has a higher accuracy and diversity of recommendation, compared with the strict division of community model. Thus, the experiments show that the non-strictly divided communities’ model is more suitable for recommendation system, especially for the personalized recommendation.

Traditional association rule mining method has the high redundancy of computation. Therefore, this paper proposes a kind of association rule algorithm of minimum single constraint based on post-processing closure operator. Firstly, it proposes association rule mining method of equivalence relation set based on closure operator constraint rule. It can meet the above minimum single constraint, maximum support and confidence coefficient threshold value effectively. In addition, it can divide constraint rule set into disjointed equivalence rule class. Secondly, it gives answers to questions and necessary and sufficient conditions of specific rule class existence. It can reduces redundant computation effectively and improve computation efficiency. At last, it verifies the validity of proposed algorithm through experimental contrast of standard test set.

There has been a growing practical need for querying XML streaming data efficiently. Stream requires to be read sequentially and only once into memory, the query must be processed on the fly. QXSList technique is proposed for massive data processing, which takes the SAX events sequence as input, buffer the incoming elements for further processing, remove unnecessary elements from the buffer in time, and give the results on the fly. Data model and algorithm integrated framework are defined, the integrate methods of how to process predicate and wildcard are discussed respectively. Level value is used for determining the relationship of two elements and relational pointers are constructed for linking multi lists in this method. The experimental results show that our approach is effective and efficient on this problem, and outperforms the state-of-the-art algorithms and query engines especially for data size is very large. At the same time, memory usage is nearly constant.

As a result of an empirical analysis of the fixed effect model, the following results have been drawn in this study: First, guarantee balance has a positive impact on the basic asset. As guarantee balance increases, the basic asset increases. Second, guarantee supply has a negative impact on the basic asset. As guarantee increases, the basic asset decreases. Third, the number of bankrupt companies and unemployment rate are estimated to have a positive impact on the basic asset. As the number of bankrupt companies and unemployment rate increase, demand for guarantee becomes higher, and therefore basic asset should be procured more. Lastly, this study did not identify the impacts of subrogation and dishonored ratio of checks and bills on the basic asset. Since basic asset has a character of the reserve fund of subrogation, the relation of the two variables is expected to have a negative relation. Although the sign is negative, no significance is found statistically. This study examined what variables affect the expansion of basic asset that becomes the basis of credit guarantee. For analysis, this study conducted a panel regression analysis with the panel data of three types of credit guarantee organizations: Credit guarantee fund, technology guarantee fund and regional credit guarantee foundations [1].

Intelligent call center is also known as the customer service center or telephone service center, it is a kind of integrated information service system. This paper analyzes the classification method based on cost sensitive variable precision rough set. And in this paper, we use the attribute weighted cost sensitive rough set classification method based on established and customer level, customer history records, agents business related dynamic queuing strategy. In addition, this paper improves the multi fractal BP neural network algorithm for the call center customer classification. Improved algorithm is able to use multifractal fluctuation and BP neural network to predict the call center traffic and seat allocation. The paper presents construction of intelligent call center system based on cost sensitive variable precision rough set and multi fractal BP neural network. Experimental results show that novel method proposed can classify customers, reduce the impact of missing data and noise data, and improve the efficiency of customer satisfaction and intelligent call.

In the terms of indoor positioning, the location of the fingerprint technology that based on receiving wireless LAN WIFI signal strength (RSS) has been widely used. In the process of location of fingerprint offline training, the traditional method has more manpower and time. In this paper, we propose a location of fingerprint database constructing method that based on Gaussian process regression (GPR), compared with the process of the traditional method for collecting a large number of fingerprint, now we based on the propagation law of space radio signals, the Gaussian process regression model is applied to the construction of the location fingerprint database, and forecast the location of fingerprint inside the located area through the study of collected samples, by doing this we can reduce the collecting density of fingerprint samples, improve the efficiency of position fingerprint positioning technology.

Deep neural networks (DNNs) and their learning algorithms have been widely used in large data analysis. With the practice of teaching network reform in higher education, many colleges began to use the network to change the traditional teaching methods, and made several achievements. In this paper, the authors first analyze the infinite depth neural network, and construct evaluation index system by using network data mining method. The result shows that 58.09% students think the evaluation from college physical education is formative assessment, 27.94% students think that the evaluation reflects the magnitude progress, and 30.15% students think that it reflects the students' self-evaluation. Physical teaching evaluation is an important part of physical education; however, the survey found that physical education students are lack of understanding of teaching evaluation. Therefore, colleges should make comprehensive evaluation of physical education, and formulate scientific and rational teaching evaluation concept.

Role based access control (RBAC) has been widely adopted in industrial and government. However RBAC is only suitable for closed enterprise environment. With modern Internet based application, collaboration and sharing among multiple organizations become essential and RBAC is no longer sufficient. Role mapping has been the solutions to deal with multiple domains, where the roles in the hierarchy of one organization are mapped to the roles in the hierarchy of another organization. But role mapping can be a tedious task for the security officers if it is done fully manually. Yet, performing role mapping automatically incur security risks. In this paper, we introduce a semi-automated role mapping process, where promising role mappings are generated automatically and recommended to the security officer(s). The security officers then approve or modify the recommended role mappings. We present a method for automatically generate role mappings based on the similarities of the roles in two role hierarchies. We use an example to illustrate our approach and show its feasibility.

Needs for the collection of data for text mining of the case with many protolanguages or emotional words distributed in SNS and following trends, and the treatment process method of purification of improved data in a previous step are raised. In data collection for mining, the online trend dictionary based on tag was referred and semi-structured data was effectively parsing processed based on tags of dictionaries according to domains of treating languages, and data for analysis was collected. Additionally, there were the cases to show inefficiency in the text processing of the general genre or the limitation of noun extraction, however, it can be suggested as an alternative on searching trend vocabularies which requires the timeliness or the class processing for corpus work of sentiment dictionary.

With the rise of cloud computing, internet of things, social networks, the type and scale of data in human society has increased at an unprecedented rate, making data from being a simple object to be process to being a basic resource. Fully mining the value of data resources that was hidden in SNS such Weibo Microblog, Wechat has become a common subject concerned by industrial circle, academic area and government departments. Although the distributed storage and analysis of cloud computing platform have been widely used in big data process of social networks, it has not been able to fully solve the problems of big data storage and process in social networks. In this paper, it proposed the big data process framework of social networks based on cloud computing. By adopting the mixing cloud model and coordinating the data storage framework and data computing framework, and regarding social networks features such as real-time, sharing, mobility, individuation, and interactivity, this big data process framework can be adopted to process large-scale massive amount of data, to research the unified management and sharing strategy of massive data, to propose data process strategy and the service application of big data such as Microblog and Wechat, and to discuss several urgent key problems in processing social networks big data.

Attribute reduction is one of the key issues for data preprocess in data mining. Many heuristic attribute reduction algorithms based on discernibility matrix have been proposed for inconsistent decision tables. However, these methods are usually computationally time-consuming. To address this issue, the derived consistent decision tables are defined for different definitions of relative reducts. The computations for different reducts of the original inconsistent decision tables are converted into the computations for their corresponding reducts of the derived consistent datasets. The relationships among different core sets and attribute reducts are further discussed. The relative discernibility object pair and the more optimal relative discernibility degree from view of the boundary region are designed to accelerate the attribute reduction process. An efficient attribute reduction framework using relative discernibility degree is proposed for large datasets. Experimental results show that our attribute reduction algorithms are effective and feasible for large inconsistent datasets.

The goal of his research was to analyze the meaning of creativity and character based on online text data analysis and identify the relationship between the two. To perform research, online text data was collected for analysis. The collected data was cleaned and then used to analyze the frequency of keyword text, network density, and centrality. The research concluded that both creativity and character have important socio-cultural significances on education. Also, the research showed that, while people tend to include character when they consider creativity, they rarely include creativity when they consider character. It is hoped that the results of this research will contribute to developing ways to integrate creativity and character in education.

In database outsourcing, the data owner delegates the tasks of data management to a third-party service provider. As the service provider may be untrusted or susceptible to attacks, query authentication is an essential part. Merkle R-tree (MR-tree) is one of the most efficient authenticated index that combines Merkle hash tree with R*-tree. MR-tree can provide an efficient range query authentication, however, as it uses the traditional R*-tree query structure in neighbor queries, a large number of unnecessary nodes may be accessed, and that can affect the efficiency of the query. In this paper, the neighbor relationship is introduced into the construction of MR-tree, and we propose a new index structure, called VMR-tree that incorporates the Voronoi diagram into MR-tree. In order to utilize VMR-tree index structure, we propose algorithms for spatial nearest neighbor queries and experiments to verify it has a better efficiency in spatial neighbor query authentication.

Affinity Propagation (AP) is a new clustering method to cluster data set efficiently. In this paper, a Distributed method of Local Affinity Propagation (DLAP) is proposed to solve hardware bottleneck and time-consuming problem. DLAP refines AP by reducing the calculating data scale in each iteration and keeps a high quality clustering result. The method is implemented on Apache Spark distributed computation framework. Depending on high iteration efficiency on Spark, the method has an impressive result in time complexity. Experiments are conducted on two-dimensional data to show that the time cost of LAP on single machine is better than the two methods, FSAP and FAP, meanwhile the result of DLAP on Spark is better than that on Hadoop.

As the starting age of English education becomes younger around the world where they teach English as a foreign language, the debate on early English education is an unsettling issue as not only academic research, but also educational policy. To resolve the unsettled issues, a new approach is used: the big data and its analytics. To explore the pros and cons of the early English education, the study uses the analysis of research abstracts collected from scholars.google.com and www.kci.go.kr. It also analyzes the data posted on the discussion sites such as agora, daum café and naver café, plus daily interactions of the early English education using SNS. The study uses opinion mining technique using tools such as Sisense and WEKA to lay out the data and analyze them as basic data analytics. The results show that pro early English education is commonly co-occur with critical period, lateralization, ultimate attainment, universal grammar, fossili-zation, inhibition, acquisition process, bilingualism and exposition. Essays against early English education are related to no critical period, no authentic input, not effective, no universal grammar, national identity loss, self-identity loss and L2 interference. Other extraneous factors are based on practical problems such as social pressure, outcome pressure, political pressure, test reform and statistics.

In the rapid development of big data, the platform-based trend of Micro-blog, Wechat, Fetion, Wing talk and other social media is obvious. The instant interaction between client and website is realized; the social attributes are becoming more and more stable; the commercialization attributes as a marketing platform and the social attributes as the media are also significantly enhanced. The influence of the whole field of media marketing and even the whole information industry is growing gradually. In the marketing process of social media, whether when it comes to the marketing with event or the marketing with content, a wave of using the large data is prominent. In the big trend of rapid development of big data marketing and social media marketing, the marketing concept, marketing methods and evaluation of the results of enterprises begin to change. Social media marketing based on big data has subverted the traditional way of transmission for Internet advertising and it has become the new trend of enterprise marketing.

Dynamic time warping algorithm (DTW) is a method of measuring the similarity of time series. Concerning the problem that DTW cannot keep high classification accuracy when the computation speed improved, a FG-DTW method based on the idea of naive granular computing is proposed. In this method, firstly, better temporal granularity is acquired by calculating temporal variance feature and it is used to replace original time series; Secondly, the elastic size of under comparing time series granularity allow dynamic adjustment through DTW algorithm and optimal time series corresponding granularity is obtained; Finally, DTW distance is calculated by optimal corresponding granularity model. At the same time, the early termination strategy of infimum function is introduced to improve the efficiency of FG-DTW algorithm. Experiments show that the proposed algorithm improves the running rate and accuracy effectively.

At the moment many of modern relational databases support set valued attributes. Despite such attributes don’t fit in classical relational theory, they expands the possibilities of data storage and manipulation. Search query on set valued attribute can be represented in specific search predicates which can be easily expressed in set-theoretic operations. Accurate enough selectivity estimation for search predicates on set valued attributes is essential for query optimizer in the same way as selectivity estimation for regular search predicates. This paper introduces a probabilistic model for estimating selectivity of search predicates on set valued attributes. This model uses frequencies of set elements occurrences as well as histogram of set values cardinality. Parameters of the model are estimated during preliminary analysis of database contents. The model was implemented for array types of DBMS PostgreSQL, which are implementation of ordered set valued attributes. Experimental verification of this implementation showed that highly accurate selectivity estimation is provided on the basis of the proposed model.

With the establishment of the power system in depth and marketing system, the power supply enterprise marketing has become a core business, develop marketing strategies to adapt to market is very important to develop efficient power marketing strategy needs to support comprehensive information. Multidimensional data analysis huge amounts of data and complex multi-angle, multi-level analysis and presentation, to obtain useful information hidden from the general to allow users to fully grasp the local business operation of multidimensional data analysis of historical data for the enterprise multi-angle, multi-level flexible as observed with high efficiency in the enterprise's existing historical data into useful information on. based on the multidimensional data analysis technology to build electric power marketing decision-oriented application of data analysis platform, through the companies have historical data analysis for the enterprise many policy-makers useful information.

Case-Based Reasoning (CBR) is widely used for problem solving in many fields, but there still exist limitations for the problems with dynamic characteristics. This work tries to introduce approaches in Sequence Pattern Mining (SPM) to extend the capability of CBR for solving problems described with sequential data. We propose a SPM algorithm named wGSP, which takes user’s different concerns on events into consideration by weight setting, to discover typical patterns in case base. Then the support information of cases to typical patterns is used to depict cases and facilitate efficient case retrieval. The contribution of this work lies in two aspects: firstly it is an improvement to traditional CBR method for coping with sequential data based cases with better interpretability and higher efficiency; secondly, it provides flexibility for parameters setting in SPM to satisfy the personalized preferences of users. Finally through a calculating instance, the advantages and effectiveness of the two-step CBR method based on sequential data is illustrated.

In LTER, curation activities play very essential role for data discovery and retrieval, quality control, and reuse over time. The K-ecohub system is a pilot repository to preserve long-term Korean national ecological data. The repository has been designed so as to manage and share data in an efficient manner. The paper presents the workflow-based data curation process in K-ecohub, which promotes collaboration and facilitates contribution from experts in the LTER field and also provides a way to automate and customize curation activities depending on the data types.

It remains both important and difficult for service-oriented computing to retrieve accurately and efficiently the service components fulfilling the client’s need. This paper intends to present a method for use in retrieving the personalized service components based on Hadoop architecture. It extends the semantic ontology of service components, and proposes the methods for building and inquiring the index files based on Lucene. By utilizing the HDFS and adopting the QoS computation and personalized recommendation techniques, the working principles of Map and Reduce of Hadoop are expanded. Experiment results show that the proposed method is applicable in realizing the retrieval of service components in distributed computing environment.

Distributed block storage system is one of the fundamental components of cloud computing, and many important services, including cloud database, cloud queue, are built upon it. It is common practice to build block storage system based on reliable and efficient Linux open source software, i.e. software RAID, to meet the I/O requirements of cloud database. Bitmap is a critical data structure of software RAID, and hence is important to reliability and performance of this kind of storage system. We describe several existing software RAID bitmap management solution, and propose REMB (Recoverable External Memory Bitmap), which is reliable and efficient. Experimental results show that REMB improve cloud database performance by 30%~60%.

Query optimization is a very complex task for commercial databases in case of performance issue that needs to be well known of entire structure of database. In desktop or web application at the back end, query processing aspires to be major factor for finding the better execution. We address the problem of SQL query optimization merely from the perspective of query response time in different databases invoked multiple queries are imparted in admission database management system, leveraging join and complex queries. Our proposed method adhering with respect to the underlying topics, to tune (Select, complex and join SQL) queries with optimized execution plan using PL/SQL features by incorporating database objects such as procedures, triggers and methods to improve query performance, instead amendment of query semantics which lessen time of developer or administrator to do manual tuning.