Under deregulated environment, accurate electricity price forecasting is a crucial issue concerned by all participants. Experience shows that single forecasting model is very difficult to improve the forecasting accuracy due to the complicated factors affecting electricity prices. A particle swarm optimization (PSO) based GM(1,2) method on day-ahead electricity price forecasting with predicted error improvement is proposed, in which the moving average method is used to process the raw series, the PSO based GM(1,2) model to the processed series and the time series analysis to further improve the predicted errors. The numerical example based on the historical data of the PJM market shows that the method can reflect the characteristics of electricity price better and the forecasting accuracy can be improved virtually compared with the conventional GM(1,2) model. The forecasted prices accurate enough to be used by market participants to prepare their bidding strategies.
Ihis paper use theories such as agency, management efficiency, and market power to explain the motive and performance of mergers and acquisitions by a cross-sectional logistic regression model, the feature indexes, and cumulative abnormal returns. The results show that the acquirers have significantly lower asset-liability ratio compared to the listed firms without takeover transactions, and firms with better profitability are more likely to execute a takeover transaction. The results also indicate that an increase in market value of acquirers, return on equity, and being overvalued could increase a firm's likelihood of acquiring because of high level of management or low agency costs. Increase in asset-liability ratio declines the possibility of acquire because of financial constraints and financial risk.
Due to uncertainties in target motion and limited sensing regions of sensors, collaborative target tracking in wireless sensor networks (WSNs) suffers from low tracking accuracy and lack of reliability when a target cannot be detected by a scheduled sensor. Generally, actuating multiple sensors can achieve better tracking performance but with high energy consumption. Tracking accuracy, reliability, and energy consumed are affected by the continuous sensing and transmission and coding method. In this paper, an optimized energy-efficient multisensor scheduling scheme is proposed for collaborative target tracking in WSNs. Simulation results show that, compared with existing scheduling mechanism the proposed scheme can achieve superior energy efficiency and tracking reliability while satisfying the tracking accuracy requirement. It is also robust to the uncertainty of the process noise.
MODIS has wide spectral range and spatial coverage, as well as the continuous coverage MODIS will provide over time, and observes the Earth as a unified nature, which is necessary for multidisciplinary studies of land, ocean, atmospheric processes and so on, at the same time, it provide us the global data for at least 15 years. All of these characters made the great use of MODIS data. Today the data of MODIS is usually used in the long term observation of the earth, the monitoring of earth environment and natural disasters. With its great important use and then there are many studies were worked on the data process of MODIS data. For the data process of MODIS, such as eliminate of bowtie and stripe noise, there are many method theories, but what is more easily is that we process these in mature RS software. In this paper, we introduced the reason of appearance of bowtie and stripe noise and the method to process these of MODIS 1B in ENVI and ERDAS, At the same time we use sample data with several corresponding processes to get verification.
The key to reduce coal pollution is the development of clean coal technology and the improvement of the backward coal-burning technology. The coal water slurry (CWS) is the first substitute of the oil. The particle size distribution of CWS plays an important role in the quality control of CWS. This article proposes an ultrasonic method with effective medium theory model which can be accurately reflected in the acoustic attenuation characteristics of coal-water slurry based on structural average. Experimental simulation proved that effective medium model is fully capable of achieving on-line detection of coal-water slurry particle size, for detection of fine-and coarse-sized particle size distribution. Non-linear relationship between attenuation and particle size, the three-frequency method can be used to inverse calculation of its. Which we can achieve CWS granularity on-line, and continuously control the quality of CWS.
To improve the reservoir long-term runoff forecasting accuracy. Adaptive regulation ant colony system algorithm (ARACS) is proposed. The forecast model is set up by using an adaptive regulation ant colony system algorithm and the radial basis function (RBF) neural network combined to form ARACS-RBF hybrid algorithm. Form the reservoir long-term runoff forecast model based on the hybrid algorithm. Then carry out the reservoir long-term runoff forecast by using the method and history runoff data. The result shows the convergence of method is faster and forecast accuracy is more accurate than that of the traditional ant colony system algorithm-RBF neural network and RBF neural network. The method improves forecast accuracy and improves the RBF neural network generalization capacity; it has a high computational precision, and in 98% of confidence level the average percentage error is not more than 6%. The hybrid algorithm can be used efficaciously in long-term runoff forecasting of the reservoir and river.
This paper introduces the SignalTap II basic content and features. Combined with concrete examples, described FPGA debugging methods and steps with SignalTap II in detail. Experimental results demonstrate that FPGA test with SignalTap II make debugging easier and speed up the system development, with very good results.
The iceberg join query is an important type of query which explores the relationships among the sensing data in wireless sensor network.It outputs the joined tuples whose aggregated results are above some threshold constraint. As the number of above-threshold results is often very small compared to the amount of sensing data,the iceberg join query is a challenging query in the resource-constraint sensor network. In this paper,we propose a new algorithm called SRJA for the iceberg join processing in the wireless sensor network. SRJA is output sensitive, and by"pushing"the iceberg constraint into the joining regions,it filters out large number of unsatisfied tuples, saving lots of tuple transmissions.Experiments based on simulations verify the performance efficiency of our algorithm.
The article studies the internal principle of the evolution of knowledge networks. Firstly, it analyses the evolution characteristics of the knowledge network. The knowledge network displays a series of self-organization characteristics during evolution. Then, it researches the impetus mechanism and points out that competition and cooperation among organizations are the fundamental impetus for the evolution of knowledge networks.
The competitive environment and changing business processes propose new requirements to ERP system's flexibility and its evaluation becomes the first and critical step to solve these problems. This paper begins with the analysis of flexibility's definition.Basing on the core idea of three-tier software architecture, the author refines the index system from the aspects of presentation flexibility, business flexibility and design flexibility. Then, treating the grey theory as a guide, the evaluation system is established by integrating AHP and Grey Evaluation Model. Finally, the corresponding example is given to illustrate the model.
An approach is proposed to discover closed frequent itemsets with a simple linear list structure called the Frequent Pattern List(FPL) in transaction database. The approach selects representation patterns from candidate itemsets to reduce combinational space of frequent patterns. By performing two operations, signature vertex conjunction and vertex counting, it simplify the process of closed itemsets generation.
Recently, social network privacy becomes a hot issue in the field of privacy. We are concerned about the path nodes in the social network. With the knowledge of the two endpoints of a path, the adversary can attack the privacy of the nodes on this path. In this paper, we define the adversary's background knowledge, propose the anonymity model, and propose PN- Anonymity algorithm to adjust paths. Experimental results show that our algorithms can achieve the path nodes anonymous, and information loss can be well controlled to ensure the availability of information.
This paper not only analyzes the principle of the periodic sampling,but also put forth a comprehensive visualization intuitive analysis method on Peer-to-Peer traffic detection with lower cost and higher performance.The method is based on detecting the in-degree and out-degree of the node,and it also combines periodic sampling with geo-information system. Finally the traffic characteristic of every node is displayed on the map in a visual way.
Previous studies have focused on serveral aspects of CRM (Customer Relationship Management). However, there is a lack of research that focuses on the customer segmentation of shipping enterprises using data mining. Data mining technology can be used to in modern CRM to greatly enhance it function and efficiency. Based on the technologies of clustering and classification in data mining, this paper discusses the method of segmentation of shipping enterprises' customers by mining the information in the mass data of documentation database. That is, we cluster history freight instances using cluster algorithm first, and then classify the new instance using Bayesian network classifier according to the results of former steps. The purpose is to support the marketing departments' decision-making, and improve the CRM level of shipping enterprises.
In this paper, the current forecast of storm surge based on BP is adapted to deal with the characteristics of storm surge. One main kind of fuzzy information in geology calamity predicting system is solved by information diffusion method. The whole process is as follows: Firstly, influential information is collected by single step predicting model and neural network predicting model separately to predict the extreme tide level. Secondly, information diffusion method is applied to dig up information as much as possible to improve the precision of risk recognition under the condition that the information is not enough.
In the competitive electronic commerce market, how to evaluate and improve the qualify of its service has become a focus problem. The quality evaluation model of electronic commerce service based on the AHP-FUZZY combines both qualitative analysis and quantitative analysis, forms the overall evaluation framework through the use of evaluation indexes, which draws the comparable and analyzable indexes and numerical value that represent the service quality. This model provides a common method of quantification, and is proved effective and feasible through case study.
The notion of 'classes of a path in an Entity-Relationship (ER) data schema' was put forward as a characterization of the capability of a schema of accommodating data instances whereby to provide information for various purposes including answering a query. Thus far though, criteria with which a process of classifying paths can be systematically carried out are yet to be found and formulated. To this end, one possible approach seems to first all accurately and formally describe paths so that we can reason about them, and then to examine the characteristics of the formal representations of the paths with a view to finding useful clues and indications. We propose to use Description Logics (DL) for this task. In this paper, we show how to use Description Logics to formulate paths in an ER schema whereby to find desirable criteria.
Knowledge acquisition is a dynamic process. Cognitive structure and cognitive process have great influence on knowledge acquisition and knowledge discovery. Previous research of knowledge acquisition technology focuses on local knowledge structure and static knowledge acquisition methods. Making full use of cognitive structure and cognitive process to complete knowledge acquisition, knowledge dynamic organization and unknown concept relation, and man-computer interaction of knowledge structure analysis, a kind of dynamic knowledge acquisition technology to drive the process of knowledge discovery based on cognitive is proposed in this paper. Using self-improvement of existing knowledge structure to drive knowledge discovery process in man-computer interaction process, thus the quality and efficiency of knowledge acquisition are improved. Knowledge acquisition theory based on cognitive and realization technology are studied in this paper, and the validity of the algorithm is verified by instances.
Information transmitting has been a very important research field of complex network and lots of researchers have been contributing to it for years to help improve the transmitting efficiency and the network's loading capacity, and build of the information transmitting mechanisms. But most of them are based on the global information of the topology architecture, which turns out to be quite difficult to obtain when the network becomes larger. In view of that situation, authors propose an information transmitting model based on the Local Selection Strategy. In this model, the source nodes where the information package locates are based on the topology information of its neighbor nodes, and will select a node as a target to transmit information until the packet reaches the target nodes. In this paper, authors study the efficiency of the information transmitting of the model in the network of BA. It is figured out in the final results, with the average degree of the network becomes bigger, the transmitting efficiency improves, while the speed of improvement slows down. In that case, elevation of the transmitting efficiency with increasing the scale of network is limited.
This paper studies the design of data fusion algorithm for asynchronous system with integer times sampling. Firstly, the multisensor asynchronous samplings is mapped to the basic axis, accordingly a sampling sequence of single sensor can be taken. Secondly, aiming at the sensor with the densest sampling points, the modified parallel filtering is given. Afterwards, the sequential filtering fusion method is introduced to deal with the case that there are multiple mapped measurements at some sampling point. Finally, a novel parallel filtering fusion algorithm for asynchronous system with integer times sampling is proposed. Besides, a judgment scheme to distinguish measurement number at every sampling point in the fusion period is also discussed.
