Discovering frequent patterns in semi-structured and structured data is a new trend in the data mining community. The motivation for discovering such types of patterns can be understood when investigating the real wor...
详细信息
Discovering frequent patterns in semi-structured and structured data is a new trend in the data mining community. The motivation for discovering such types of patterns can be understood when investigating the real world problems. Several problems can be modeled only with structural data like graphs rather than with flat transactions. Thus the research directions must adjust to the requirements of these new kinds of problems. Frequent tree mining is a special part of frequent graph mining as trees are special graphs. Trees can model several real world problems such as handling XML documents or WEB navigation patterns, furthermore, trees can be handled easier than general graphs. Thus discovering frequent trees has justification. In this paper a novel approach is proposed for efficiently discovering frequent tree-like patterns in tree databases. The new algorithm presented in this paper is a level-wise, "candidate generate and test" algorithm. The main contribution of the new approach is to use pushdown automatons for subtree inclusion testing. In order to enhance the performance of counting the support of the candidate trees, the several automatons are joined such that only one stack is needed for handling the joined automatons. The advantage of the new method is that is has to read the items of the input exactly once while the support of all the candidates is determined.
Sequential pattern mining is a heavily researched area in the field of data mining with wide variety of applications. The task of discovering frequent sequences is challenging, because the algorithm needs to process a...
详细信息
Sequential pattern mining is a heavily researched area in the field of data mining with wide variety of applications. The task of discovering frequent sequences is challenging, because the algorithm needs to process a combinatorially explosive number of possible sequences. Most of the methods dealing with the sequential pattern mining problem are based on the approach of the traditional task of itemset mining, because the former can be interpreted as the generalization of the latter. Several algorithms use a level-wise "candidate generate and test" approach, while others use projected databases to discover the frequent sequences. In this paper a classification of the well-known sequence mining algorithm is presented. Because each algorithm has its own advantages and drawbacks regarding the execution time and the memory requirements, and the exact aim of the algorithms differs as well, thus an exact ranking of the methods is omitted. A basic level-wise algorithm, the GSP is described in detail. Because the level-wise algorithms need less memory in general than the projection-based ones, ah efficient implementation of the GSP algorithm is also suggested. Two novel methods, the Bitmap-based GSP (BGSP) and the SM-Tree (State Machine-Tree) algorithms are presented as an enhancement of the GSP-based sequential pattern mining approach.
The paper is devoted for a systematic discussion on the subspace based identification methods. The structure of the noise components acting additively at various system signals is considered as the guiding principle f...
详细信息
The paper is devoted for a systematic discussion on the subspace based identification methods. The structure of the noise components acting additively at various system signals is considered as the guiding principle for the analysis of the estimation algorithms. It is shown that replacing the state space approach typically applied in this field by a discrete time transfer function system realization the highly involved machinery of the subspace algorithms can be handled in a clearly ordered framework exhibited by a set of over-determined equations to be solved.
an adaptive gain, smooth sliding observer-controller is developed to control n -degree of freedom rigid robotic manipulators with uncertain parameters. Furthermore, an on-line closed loop identification scheme, for ti...
详细信息
An adaptive gain, smooth sliding observer-controller is developed to control n -degree of freedom rigid robotic manipulators with uncertain parameters. Furthermore, an on-line closed loop identification scheme, for ti...
详细信息
An adaptive gain, smooth sliding observer-controller is developed to control n -degree of freedom rigid robotic manipulators with uncertain parameters. Furthermore, an on-line closed loop identification scheme, for time-varying parameters is proposed in order to obtain useful information despite loads, external disturbances and faults detection. In order to reduce the chattering, a smooth switching function (parameterised tangent hyperbolic function) is used instead of pure relay one, in the observer and the controller. The gains of the switching functions are adaptively updated, depending on the estimation and tracking errors, respectively. Using adaptive gains, the transient and tracking responses are improved. Simulation results with a two degree of freedom (DOF) robot manipulator are presented to show the interest of the approach.
According to the multi-model approach a nonlinear dynamical process is approximated in different working points by local valid linear models. The global valid model output is calculated as the weighted sum of the sub-...
详细信息
The paper presents a control system design technique in delta domain for IMC (Internal Model control) structure. Also a generalised form for delta domain Dead-beat control algorithm is given, a hybrid implementation o...
详细信息
Long-range optimal prediction algorithms use the predicted output for several steps ahead. The prediction based on traditionally estimated model parameters does not result in an optimal prediction if the measurements ...
详细信息
In this paper, a direct learning control method for a class of switched systems is proposed. The objective of direct learning is to generate the desired control profile for a newly switched system without any feedback...
详细信息
For predictive control in industry often very long horizons for control error and manipulated signal are used because of the slow processes which take place in the petrochemical industry. In order to reduce the comput...
详细信息
暂无评论