Advanced Materials Research Vols. 989-994

Abstract: Blind source separation (BSS) is a technique for recovering a set of source signals without priori information on the transformation matrix or the probability distributions of the source signals. In the previous works of BSS, the choice of the learning rate would reflect a trade-off between the stability and the speed of convergence. In this paper, a particle swarm optimization (PSO)-based learning rate adjustment method is proposed for BSS. In the simulations, three source signals are mixed and separated and the results are compared with natural gradient algorithm. The proposed approach exhibits rapid convergence, and produces more efficient and more stable independent component analysis algorithms than other related approaches.

Abstract: Data mining technology based on the particle swarm optimization algorithm applied in earthquake prediction was presented. Making use of the characteristics of abnormally high-dimensional data of earthquake precursor, this paper studies an earthquake prediction model based on the Particle Swarm Optimization Clustering Algorithm. This model analyzes the relationship between earthquake precursor data and earthquake magnitude. Inputs are 14 abnormal indexes such as belt, seismic gap and short leveling, and output is earthquake magnitude classification. The cluster average-distance is set as the evaluation function of the Particle Swarm Optimization Algorithm. The experimental results indicate that, this model can effectively and validly predict the earthquake magnitude in accordance with the earthquake precursor data. Compared with the traditional clustering k-means Algorithm model, this stability is stronger, and the correctness of forecast is much higher. Through the research and analysis of the example of history source seismic data, the model of this paper is one of approaches to improve the efficiency of earthquake forecast.

Abstract: In wireless sensor networks, the precise positioning of unknown nodes is the precondition and security for all kinds of applications. In this paper, by analyzing the present research status of personnel positioning algorithm and on the basis of the study of traditional centroid algorithm, an improved based on RSSI (Received Signal Strength Indicator) of the weighted centroid localization algorithm is proposed. The algorithm takes the signal intensity meanwhile the distance between the unknown nodes and beacon nodes as a reference to correct RSSI value. Through the simulation of the localization algorithm it proves that the improved localization algorithm has higher precision than the traditional algorithm. Once the accidents can be timely and efficient rescue for the underground staff, better meet the actual demand of safety management.

Abstract: With the data explosion, data mining algorithms are required to deal with huge amounts of records. In the traditional way, the processing goes in one single control flow, the time spent in computing grows fast with the increasing of data scale. K-means is one of the widely used algorithms in cluster analysis. MapReduce is a programming model which has been widely used for processing data in a parallel environment. This paper gives an implementation of the K-means algorithm based on the MapReduce model, so that the clustering system could handle the massive data in a fast and scalable fashion. The brief structure of the algorithm and the analysis for the main improvement are also given. We demonstrated that the algorithm will be superior when the volume of data grows bigger or the number of nodes in the computer cluster grows much bigger.

Abstract: According to the question of the standard particle swarm optimization (PSO) algorithm is prone to premature and no convergence phenomenon, this paper proposed an algorithm of Inflection nonlinear global PSO. The algorithm introduces nonlinear trigonometric factor and the global average location information in the formula of velocity updating. It take advantage of the convex of the triangle function cause the particles early in the larger velocity search maintain long time and in the later searching with smaller speed maintain long time, use the global average position information make the population can use more information to update their position. The method are applied in optimizing in the parameters of the main steam temperature control system and furnace pressure control system for comparison, the results show that the method in the search speed and precision than standard PSO has significantly improved.

Abstract: First, we describe crossing-selling and association rules. And then with the study of a correlation clothing store sales data, it shows the Apriori algorithm applies in specific association analysis. We can propose a model which is suitable for crossing-selling. Through commercial test, the algorithm can significantly increase sales of the relevant product.

Abstract: In the past some of the social and economic system of the study, often simplified to a single person's behavior can be described using the Poisson process stationary random process. However, since 2005, and reply by email, mail Propagations, human behavior, the actual statistics of time intervals, it was found there with the above assumptions these acts very different characteristics: a long period of silence and short-term high-frequency The outbreak, while present in these human behavior, the time interval distribution also satisfy the inverse power function of the fat tail, that is, during the occurrence of these acts can not be described by Poisson process. In order to study public sentiment of the Tibetan language network generation, transmission and disappearance of the law, it is necessary in the public sentiment research networks in the spread of the main roles. In this paper, the network user behavior patterns in Tibetan, Tibetan text of automatic segmentation, user behavior of the subject property and the main information dissemination and other aspects of behavior modeling are discussed.

Abstract: With the widely promotion of education informationization, education workers all around the world are facing a challenge that the electronic homework plagiarism has become increasingly serious. Our research includes technology innovation and education innovation, the technology innovation will divide plagiarism into two types, by using different algorithms to analysis and identify the plagiarism under the computer room environment and network copying, and the digital watermarking technology is applied to electronic operations against plagiarism. In network condition, vector space algorithm and edit distance algorithm is used to judge the similarity between electronic assignments and related documents. In the two years of teaching practice, we also strengthened credit education and reformed teaching methods, the percentage of plagiarism was decreased obviously.

Abstract: Retracted paper: The implications of stochastic epistemologies have been far-reaching and pervasive. After years of natural research into massive multiplayer online role-playing games, we show the compelling unification of courseware and the Internet. Our focus in this work is not on whether courseware and hash tables can interact to overcome this issue, but rather on describing a novel application for the improvement of Internet QoS (WEY).

Abstract: Geomagnetic matching is a new developmental technology in recent years. For resolving the question of the searching space is too much, a geomagnetic matching algorithm based on AFSSS is proposed, which imitates the fish behaviors such as the preying, swarming, following and leaping etc to achieve the global optimum value, then achieve the emendation of inertial system. Firstly, the affine transformation model between inertial trajectory and real trajectory is displayed. Secondly, from the application of geomagnetic matching point of view, the state of artificial fish (AF), distance and food consistence are defined afresh, and the algorithm flow chart based on AFSSS is given. Lastly, the simulation analysis is performed in an actual magnetic reference map. The simulation results show that the algorithm actualizes the precision localization when the inertial system exist position error, velocity error and heading error and then validates the feasibility of the proposing method.