Support vector machine (SVM) has been studied and applied extensively for its high accuracy, but it must construct SVM decision tree to classify sample sets with multiclass for it just be applicable for binary classif...
详细信息
ISBN:
(纸本)7900719229
Support vector machine (SVM) has been studied and applied extensively for its high accuracy, but it must construct SVM decision tree to classify sample sets with multiclass for it just be applicable for binary classification and solve a quadratic programming problem to gain optimal hyperplane either in sample spaces or in feature spaces;Alternative covering algorithm which designs neutral networks with spherical domains has the advantages of fast performance, some hard solving problems have been solved using this algorithm. In this paper, a new kind of structural learning algorithm which combining covering design algorithm, fuzzy set and SVM is put forward, instances show that this kind of networks has the virtue of both covering design algorithm and SVM.
The secondary structure prediction of protein plays an important role to obtain its tertiary structure and function. In the past thirty years, a huge amount of algorithms have been employed to this task. The better pr...
详细信息
The secondary structure prediction of protein plays an important role to obtain its tertiary structure and function. In the past thirty years, a huge amount of algorithms have been employed to this task. The better predicators are based on machine learning techniques, especially based on neural networks. But the architecture of neural network is hard to define, and the training process is time-consuming. In this paper,a constructive machine learning approach is used to predict protein secondary structure with five different encoding schemes,the results show that the constructive algorithm can achieve high predicting accuracies and the encoding schemes have influence on predicting result.
Two neural networks with the common input vector can finally synchronize their weight vectors by output- based mutual learning. It can be well utilized to negotiate secure information over a public channel. Designing ...
详细信息
Two neural networks with the common input vector can finally synchronize their weight vectors by output- based mutual learning. It can be well utilized to negotiate secure information over a public channel. Designing security protocols based on such synchronized neural network model is quite advantageous for its low-cost and high-performance. In this paper, we at first analyze and optimize the interacting network neurl, then present a cryptography-oriented secure parity model and implement the performance simulations. As an instance, a novelkey agreement protocol design scenario is finally proposed.
Most of the current trust models in peer-to-peer (P2P) systems are identity based, which means that in order for one peer to trust another, it needs to know the other peer's identity. Hence, there exists an inhere...
详细信息
Most of the current trust models in peer-to-peer (P2P) systems are identity based, which means that in order for one peer to trust another, it needs to know the other peer's identity. Hence, there exists an inherent tradeoff between trust and anonymity. To the best of our knowledge, there is currently no P2P protocol that provides complete mutual anonymity as well as authentication and trust management. We propose a zero-knowledge authentication scheme called pseudo trust (PT), where each peer, instead of using its real identity, generates an unforgeable and verifiable pseudonym using a one-way hash function. A novel authentication scheme based on zero-knowledge proof is designed so peers can be authenticated without leaking any sensitive information. With the help of PT, most existing identity-based trust management schemes become applicable in mutual anonymous P2P systems. We analyze the levels of security and anonymity in PT, and evaluate its performance using trace-driven simulations and a prototype implementation. The strengths of pseudo trust include the lack of need for a centralized trusted party or CA, high scalability and security, low traffic and cryptography processing overheads, and man-in-middle attack resistance. We aim for the pseudo trust design to be included in the P2P trust and anonymity context.
In this paper, an uncalibrated dynamic visual servoing algorithm is proposed and analyzed. No calibration or robot model is needed. After a brief introduction of the development of uncalibrated visual servoing, the th...
详细信息
In this paper, an uncalibrated dynamic visual servoing algorithm is proposed and analyzed. No calibration or robot model is needed. After a brief introduction of the development of uncalibrated visual servoing, the theoretical backgrounds and mathematical requirements of recursive least square (RLS) are stated respectively. Then the core uncalibrated visual servoing algorithm, in RLS form, or more technically, VS-RLS as well as its performance analysis, is investigated. After that, the experimental 6DOF Puma560 simulation of static and moving target tracking is demonstrated. Finally the weak and strength of the algorithm as well as the potential and promising improvements are discussed.
Partial least squares (PLS) is one of the widely used dimension reduction methods for analysis of gene expression microarray data, it represents the data in a low dimensional space through linear transformation, the s...
详细信息
Partial least squares (PLS) is one of the widely used dimension reduction methods for analysis of gene expression microarray data, it represents the data in a low dimensional space through linear transformation, the size of the reduced space by PLS is critical to generalization performance of classifiers. The previous works always determined the top fixed number of components or the top several components by cross-validation. Here we demonstrate the usage of feature selection for PLS based dimension reduction. As a case study, PLS is combined with two feature selection methods (genetic algorithm and sequential backward floating selection) to get more robust and efficient dimensional space, and then the constructed data from the selected components is used as input for the support vector machine (SVM) classifier. We use the method for tumor classification on gene microarray data, experimental results illustrate that our proposed framework is effective both to reduce classification error rates and get compact dimensional space.
Several artificial intelligent methods, including support vector regression (SVR), artificial neural networks (ANNs), and partial least square (PLS) are used for the multivariate calibration in the determination of th...
详细信息
Several artificial intelligent methods, including support vector regression (SVR), artificial neural networks (ANNs), and partial least square (PLS) are used for the multivariate calibration in the determination of the three aromatic amino acids (phenylalanine, tyrosine and tryptophan) in their mixtures by fluorescence spectroscopy. The results of the leave-one-out method show that SVR perform better than other methods, and appear to be good methods for this task. Furthermore, feature selection is performed for SVR to remove redundant features and a novel algorithm named PRIFER (prediction risk based feature selection for support vector regression) is proposed. Results on the above multivariate calibration data set show that PRIFER is a powerful tool for solving the multivariate calibration problems.
Most of the timed automata reachability analysis algorithms in the literature explore the state spaces by enumeration of symbolic states, which use time constraints to represent a set of concrete states. A time constr...
详细信息
Most of the timed automata reachability analysis algorithms in the literature explore the state spaces by enumeration of symbolic states, which use time constraints to represent a set of concrete states. A time constraint is a conjunction of atomic formulas which bound the differences of clock values. In this paper, it is shown that some atomic formulas of symbolic states generated by the algorithms can be removed to improve the model checking time- and spaceefficiency. Such atomic formulas are called as irrelevant atomic formulas. A method is also presented to detect irrelevant formulas based on the test-reset information about clock variables. An optimized model-checking algorithm is designed based on these techniques. The case studies show that the techniques presented in this paper significantly improve the space- and time-efficlency of reachability analysis.
暂无评论