Advanced Materials Research Vols. 532-533

Abstract: Function mining aims to discover valid functions from various data which seem not related. Nowadays, the results of most mining algorithms are not capable to characterize the things fully, because the algorithms aim at one simple function, meanwhile, another one can achieve this task, but its efficiency is very low because of the too much iteration. The paper presents one new algorithm named interest-based function set mining algorithm (IFSM), it applies the user's experience, expectations and prior knowledge to the mining process. IFSM succeeds in improving the description and enhancing the mining efficiency and quality.

Abstract: It urgently needs to solve the information security issues of RFID system at present. In order to increase the security of RFID Secure Mechanism without affecting the operation efficiency, a novel chaotic encryption scheme is proposed to improve the RFID information security in RFID. The Logistic map with variable initial value is used to generate chaotic cipher sequence, and then the sequence is encrypted. Related analysis and simulation experiment verified that this chaotic encryption RFID Secure Mechanism is easily to realize. It shows that the novel chaotic encryption can completely meet the security demand of the RFID systems effectively.

Abstract: Moving object detection and feature extraction algorithm in video sequences are discussed in this paper. There are several problems in moving object detection and extraction from outdoor video surveillance, that is, moving object detection algorithm is easily interfered by background of video monitor, the feature of moving object is difficult to extract from video source, and the vibration of picture frame in outdoor video surveillance causing by wind factors effects the incorrect extraction of the moving object. The vibration causing by wind factors was corrected and an enhancement inter-frame difference algorithm based on difference histogram threshold selection is presented in this paper. The experiment results prove that this method can detect and extract the moving object accurately and efficiently, and it can meet the needs of real-time detection.

Abstract: A new hybrid algorithm is presented in this paper, which solves the ill-posed inverse problem of magnetic induction tomography (MIT) and improves the quality of reconstructed image. The hybrid algorithm firstly produces the preliminary image region using Tikhonov regularization algorithm, and then it obtains the final reconstructed image using variation regularization algorithm. The hybrid algorithm, compared with the Tikhonov regularization algorithm and the variation regularization algorithm, overcomes the numerical instability of MIT image reconstruction and accelerates the convergence speed of image reconstruction, and it also improves the resolving power of targets conductor and the quality of the reconstructed image. Simulation results show that the quality of the reconstructed image obtained using the hybrid algorithm is enhanced, so an effective algorithm for magnetic induction tomography (MIT) is introduced.

Abstract: Fuzzy Cognitive Map (FCM) fails to represent the measures of uncertain causal relationships, proposes an evolutionary algorithm Based on Neural Network for FCM. This algorithm integrates the high non-linear mapping ability of neural network and the globally optimizing ability of evolutionary computation to improve the dynamic reasoning for fuzzy knowledge.

Abstract: To solve the problem of sparse keywords and similarity drift in short text segments, this paper proposes short text clustering algorithm with feature keyword expansion (STCAFKE). The method can realize short text clustering by expanding feature keyword based on HowNet and combining K-means algorithm and density algorithm. It may add the number of text keyword with feature keyword expansion and increase text semantic features to realize short text clustering. Experimental results show that this algorithm has increased the short text clustering quality on precision and recall.

Abstract: Most of planners can’t avoid state combinatorial explosion which is a main problem to planning failure. To solve it, the paper brings forward the arithmetic of goal decomposition and initial decompositioin to partition the fact file for planner, thereby to reduce number of combinated states. But then, for the planners in being can’t deal with the multiple fact files, to settle which, withal, the paper introduces the arithmetic of multi-fact file management. The application for three arithmetics is on IPP possessing preferable performance. The modified planner is called MF-IPP which is able to handle multiple fact files for avoiding the combinatorial explosion. At last, we compared the performance of the two planners, and the result showed that MF-IPP can avoid the combinatorial explosion well, and the test case covering rate is meseasued by two novel rules.

Abstract: An efficient algorithm for computing the k-error linear complexity spectrum of a q- ary sequence s with period 2 pⁿ is presented, where q is an odd prime and a primitive root modulo p². The algorithm generalizes both the Wei-Xiao-Chen and the Wei algorithms, The new algorithm can compute the k-error linear complexity spectrum of s using at most 4 ⁿ⁺¹ steps.

Abstract: In this paper, we proposed a novel filtering algorithm that using the Ricker wavelet kernel to reduce the noise. The algorithm based on Support vector machine (SVM) which is a machine learning method on the base of statistical learning theory. Those parameters of the new algorithm affect the rising edge, the band width and central frequency of passband. The experimental results of synthetic seismic data show that the filter with the Ricker wavelet kernel works better than other methods．

Abstract: Based on rough graph theory, this paper gives a new algorithm in rough network, which generalizes the classical algorithm of exploring maximum flow. This algorithm successfully deals with a kind of complex relationship mining problem between different relationship levels. Simulation shows the effectiveness of this algorithm.