Showing papers in "Journal of Computer Applications in 2008"

Emotion recognition from surface EMG signal using wavelet transform and neural network: Emotion recognition from surface EMG signal using wavelet transform and neural network

Abstract: Emotion recognition is a pivotal question of affective computing. This paper adopts the wavelet transform to analyse the surface EMG signal instability feature. Surface EMG signal is decomposed by discrete wavelet transform (DWT) and selected maximum and minimum of the wavelet coefficients in every level. The extracted maximum and minimum of the wavelet coefficients is inputted to identify emotion by the BP neural network improved by Levenberg-Marquardt algorithm. Experimental result shows that identification purpose of four emotional signals (joy, anger, sadness and pleasure) is effective and have are a great potential in practical application of emotion recognition. Keywords-affective computing; emotion recognition; wavelet transform; BP Neural Network; EMG

Chaos particle swarm optimization algorithm

Abstract: By introducing chaos state into the original Particle Swarm Optimization(PSO),this paper proposed a new algorithm-Chaos Particle Swarm Optimization(CPSO).The new algorithm makes good use of the ergodicity,stochastic property,and regularity of chaos.By judging the local convergence,when PSO gets into the local convergence,CPSO can start the chaos researching.The experimental results demonstrate that the new algorithm has the ability to avoid being trapped in local minima,and improves computational precision,convergence speed and the ability of global optimization.

Named entity relation extraction based on SVM training by positive and negative cases

Abstract: Based on the characteristics of the Chinese named entity relation extraction,features were selected and feature vectors were constructed in terms of Chinese morphological,grammar and semantics.Then potential named entity pairs in accordance with the specific entity relation template were extracted and divided into positive and negative cases.Support Vector Machine(SVM) classifier was trained by the positive and negative cases and used to judge the relation of the potential named entity pairs.Experimental results prove that this new method can effectively improve the accuracy of Chinese named entity relation extraction.

Research and realization of RIA WebGIS based on Flex

Abstract: According to the characteristics of Flex and RIA, the architecture of RIA WebGIS based on Flex was analyzed. In order to improve the efficiency of the system, the tile and double cache technology was introduced with vector and raster integrated map service. At last, a solution was given and a certain company WebGIS system was developed. The results demonstrate that developing WebGIS by using Flex technology is an ideal method.

Analysis of risk evaluation techniques on information system security

Abstract: The risk evaluation on information system security is the basis and precondition of the security mechanism of information system This paper introduced the present evaluation techniques in detail and then summarized these techniques according to different characteristics After that, we analyzed the problems existing in risk evaluation on information system security and future prospect

Survey on research of sentiment analysis

Abstract: With the rapid growth of the Web text data,mining and analyzing these text data,especially the online review data posted by the users,can greatly help better understand the users' consuming habits and public opinions,and played an important role in decision-making for the enterprises and the governmentThis survey first introduced the motivation,research problems and goals of sentiment analysis,and presents some basic technologies used in sentiment analysisIt then described one of the major tasks in sentiment analysis,subjective sentence detection,by reviewing and analyzing some recent work in this areaNext,it focused on another important task in sentiment analysis,opinion classification,and discussed two leading feature extraction techniques for opinion classification,sentimental word based and frequent pattern based methodsFurthermore,it also introduced several other relevant sentiment analysis problemsFinally,the current status,remaining challenges,and future directions in the field of sentiment analysis was summarizedWith the rapid growth of the Web text data,mining and analyzing these text data,especially the online review data posted by the users,can greatly help better understand the users' consuming habits and public opinions,and plays an important role in decision-making for the enterprises and the governmentThis survey first introduces the motivation,research problems and goals of sentiment analysis,and presents some basic technologies used in sentiment analysisIt then describes one of the major tasks in sentiment analysis,subjective sentence detection,by reviewing and analyzing some recent work in this areaNext,it focuses on another important task in sentiment analysis,opinion classification,and discusses two leading feature extraction techniques for opinion classification,sentimental word based and frequent pattern based methodsFurthermore,it also introduces several other relevant sentiment analysis problemsFinally,the paper summarizes the current status,remaining challenges,and future directions in the field of sentiment analysis

Hybrid genetic ant colony algorithm for traveling salesman problem

Abstract: A new Heuristic Genetic Information based Ant Colony Genetic Algorithm (HGI-ACGA) to solve Traveling Salesman Problems(TSP) was proposed. HGI-ACGA's two sub-algorithms are Ant Colony Genetic Algorithm(ACGA) and Heuristic Genetic Information based Ant Colony Algorithm(HGI-ACA). ACGA enhances genetic algorithm's population diversity and reduces the search domain, while HGI-ACA eliminates the creation of invalid tours and also avoids depending on pheromone excessively. A combination of genetic information and pheromone leads to a significant improvement in performance. The strategy of HGI-ACGA algorithm can improve convergence rate and capacity of searching optimal solution. Experimental results show that the proposed algorithm generally exhibits a better solution and a higher rate of convergence for TSP than ACGA and ACA.

Adaptive image watermarking algorithm based on block DCT transform and Arnold shuffling

Abstract: In this paper, we proposed an adaptive image watermarking algorithm based on DCT and Arnold shuffling transform. With the features of Human Visual System (HVS) sufficiently taken into account during embedding, embedding different watermarking energy into different DCT blocks made the algorithm adaptive. The experimental results show that the algorithm is robust to common digital image processing operations, especially to image cut.

Tasks scheduling algorithm for parallel system with multi-core processor

Abstract: A proper algorithm for tasks scheduling based on features of parallel system on multi-core processor is proposed.The algorithm completed tasks allocating before tasks scheduling,which can improve the performance of tasks scheduling by decreasing the communication overhead among processors with the reasonable tasks allocating.The algorithm simulation was completed and the superiority of the algorithm with simulation results was demonstrated.

Improved approach to CHI in feature extraction

Abstract: Feature extraction technology is an essential part of text categorization,which directly affects the categorization precision.This paper comprehensively took frequency,distribution and concentration into account and proposed an improved Chi-square Statistic(CHI)approach.In order to verify the improved CHI approach,a contrastive experiment was carried out.The experimental results show that improved CHI approach is superior to traditional CHI approach in feature selection,which verifies the efficiency and probability of the improved CHI approach.

Demand-based spectrum allocation algorithm in multi-cells cognitive radio network

Abstract: A demand-based spectrum allocation algorithm between cognitive radio cells was proposed. Based on the combination of graph coloring model and the concept of spectrum quality grading, the proposed algorithm allocated the available spectrum to cognitive radio cells to meet demand maximally under QoS guarantee. Simulation results show that, compared with the original algorithm, the proposed algorithm can better satisfy demand in the network.

Education management information system based on SOA

Abstract: Through introducing the basic concepts of Service-Oriented Architecture(SOA) and analyzing the relationship between SOA and Web Service,a service-oriented design of education management information system that based on Web Service and a part of its implement was given.The aim is taking full advantage of education resource,reducing costs of education informatization,achieving the goal of information sharing and unified management.

Automated software test data generation based on immune genetic algorithm

Abstract: A kind of software test data automated generation method based on Immune Genetic Algorithms(IGA)was proposed.Immune operators including vaccine extraction,vaccination and immune selection were introduced into the traditional genetic algorithm.The experimental results show that this algorithm is more effective than the traditional one.

Implementation of heterogeneous system integration based on ESB

Abstract: Being the communication barrier in business,divergences of commercial software are hard to mutual integration.This article,to solve this problem,suggested that different systems can be integrated by Enterprise Service Bus(ESB)main line to achieve the goal of smooth communication.And it also enables the system integration to be more convenient and expandable while protecting the existing investment maximally.

Design of RFID middleware architecture

Abstract: Radio Frequency IDentification(RFID)middleware is the key software between tag readers and database and enterprise applications.It is the center of RFID application deployment operation.A Service-Oriented Architecture(SOA)real-time RFID middleware infrastructure was designed according to the requirement of related enterprises,and the corresponding UML class diagrams for the device management and integration components were given.The roles of RFID middleware components were described.The test of catering system shows that this middleware can improve the portability,maintainability and reliability of the system.

Improved Kalman filter method for speech enhancement

Abstract: An approach to speech enhancement based on spectral subtraction and masking properties of human auditory system was present.In this approach,spectral subtraction method was applied to estimatede the Autoregressive(AR) parameters,and the comutational complexity was reduced.A perceptual post-filter concatenated with a standard Kalman filter was proposed to make the estimation error obtained from the Kalman filter lower than the masking threshold and optimizes the trade-off between the speech distortion and noise reduction.Experimental results demonstrate that the proposed algorithm has better performance when compared with standard kalman methods.New method is more suitable for hunman auditory properties and this superiority is more significant at very low SNR.

ElGamal public-key cryptosystem based on integral quaternions

Abstract: This paper introduced the definition of congruence class and some basic property of integral quaternions,and proposed ElGamal Public-Key Cryptosystem(PKC) based on integral quaternions.Its security was based on the difficulty of large integer factorization and discrete logarithms problem.Simulation and realization were carried out in computer,furthermore the security of it was discussed and analysed.

Improved algorithm based on decision tree

Abstract: Decision tree is an important method in induction learning as well as in data mining,which can be used to form classification and predictive model.ID3 algorithm is the most widely used algorithm in the decision tree,based on the Decision Tree ideas in the data mining,for the shortcoming of inclining to chose attributions which has many values,the paper has improved ID3 by introducing irrelevant degrees.Experimental Analysis of the data shows that the new algorithm can be more reasonable and more effective rules.