Applied Mechanics and Materials Vols. 543-547

Abstract: The purpose of association rules mining is to find rules which can meet the minimum support and minimum confidence from a large quantity of data. To find the valid association rules efficiently, we had a comprehensive analysis on some well-know parallel association rules mining algorithms and proposes a new parallel association rules mining algorithm (Array Based on Hadoop, short for ABH) based on the cloud computing platform. The ABH scans the database only once, uses the 0/1 array to represent one of the transactions and to record the frequency of the same transaction. Moreover, by utilizing the random access characteristics of the array and the special nature of the frequent itemset, the ABH can reduce the quantity of frequent candidate itemset effectively and find the frequent itemset quickly. We have compared the ABH with two classical algorithms CD and DD through experiment; we can find that ABH outperforms CD and DD.

Abstract: Cloud computing system preserves the security and privacy of a user's data by replicating data among several clouds and by using the secret sharing approach. It uses the database management system DBMS (data source) to manage and control the operations between the clients and them ulti-clouds inside a cloud service provider [1]. Generally speaking, The Shamir's threshold scheme is suitable for the shares in the cloud computing. The Shamir's threshold scheme as a secret sharing scheme will attract more attention in the cloud computing in future.

Abstract: In connected graph, edge centrality represents the importance of edge and loading degree in the process of information transmission. The second eigenvalue of Laplacian matrix decides connectivity of complex networks. We propose edge centrality model and cut model based on minimization model of the algebraic connectivity. Edge centrality function is derived in order to calculate edge centrality. Cut model deletes k edges at an iteration whose algebraic connectivity of complex networks decreased fastest. Choice of k is based on edge sparse degree of complex networks. By empirical analysis of real network, the implications of edge with higher centrality and its vertex coincide with the fact. The model can be used in medium-size networks with lower time complexity and higher efficiency.

Abstract: In this paper the new grid task attemper is discussed. It combined the grid computing with role layer resource organize, role working flow and role collaboration. Consequently, the tactical computing functions has been shifted from single node running into simultaneous multi-node operation. Thus, the problems caused by large amount of data and complicated operation, such as poor speed, high request for server equipment and task attemper intermission etc, in the traditional grid system can be solved effectively.

Abstract: To resolve conflicts between share and collaborative analysis requirements of security alarm and alert data holders worries about privacy, it firstly probes into the anonymized protection method Incognito. Based on that, it improves the algorithm to solve existing problems by extending common data like privacy protection targets to alert data. The generalized anonymous processing model for alert data is developed and the quantitative evaluation is realized between the level of alert datas secret protection and data quality. With authoritative data set of intrusion detection attack scenario as test data, the experiment validates efficiency and effectiveness of the proposed method on the part of performance and security.

Abstract: According to the existing deficiency of the process of network public opinion for the ideological and political work in colleges and universities, we put forward a new threshold network public opinion algorithm. Compared with the network public opinion algorithm which has the traditional soft and hard threshold and the existing threshold, the new algorithm has overcome the error about the estimation between network public opinion and the real public opinion, and has better regulation and continuity. Simulation results show that it can improve the quality of network public opinion by using the new threshold function, and reduce the intensity of interference on students.

Abstract: Graph algorithm has been widely used in modern computer network technology, and with the increase of data, graph algorithm is faced with the challenge of big data. Using MapReduce programming framework, this paper puts forward a multiple ring fast computing method in diagraph. Compared with the previous methods, this method is more efficient. And through the experiment the method are compared with the previous method.

Abstract: With the development of economy, the tourism industry has become a pillar industry for national development. To speed up the development of the tourism industry, more rational capital investment for tourism, the reasonable layout of the facilities, make scientific decisions will be very important to tourism. Data mining technology has had a profound impact in many industries and areas. Among data mining association rules mining, because it can find a lot of interesting connections between data items, can provide the basis for our decision-making. Therefore, we will introduce data mining into tourism industry. Find out the hidden link in the tourism industry, in order to provide the basis for scientific decision-making.

Abstract: E-commerce is an inevitable trend of the development of tourism industry according to the characteristics of the development of tourism electronic commerce, using Web data mining technology, the tourists, and tourism core product classification to develop personalized tourism electronic commerce services, identify potential customers, develop dynamic adjustment of tourism products, and improve the tourism electronic commerce web site design and the tourism information retrieval speed.[ It can provide the tourism electronic commerce website for tourism enterprise construction planning, strategic planning, website maintenance and web marketing to provide the reference for the construction of network brand. Analysis of the data mining technology and Web records the process of mining, mining and recommendation engine using Web records help the tourism electronic commerce personalized tourism network. Meanwhile, the usages of e-commence can recommend travel plans for travelers and improve the reservation rate of network of tourism. Tourism enterprises also can make more profit

Abstract: Along with the rapid development of image acquisition and image storage, a huge number of usable image data are obtained by people, such as satellite remote sensing image data, medical image data, etc. Data mining of images is to analyze these useful images and extract the usable information from them. How to effectively store rapidly make data mining for the increasing images has become the most challenging problem faced by people. This paper focuses on data mining of the massive images with the help of the Hadoop cloud platform.