Papers by Keyword: Rough Set (RS)

Paper TitlePage

Authors: Yun Hua Wang, Hui Yan Ke
Abstract: The demand for individualized teaching from E-learning websites is rapidly increasing due to the huge differences existed among E-learning learners. A method for clustering E-learningers based on rough set is proposed. The basic idea of the method is to reduce the learning attributes prior to clustering, and therefore the clustering of E-learningers is carried out in a relative low-dimensional space. Using this method, the E-learning websites can arrange corresponding teaching content for different clusters of learners so that the learners individual requirements can be more satisfied.
Authors: Xiao Kang Tang, Xue Zhi Zhang, Qiong Zou, You Guo Wei, Cheng Jun Cao
Abstract: when the rough set be used to deal with Knowledge representation system, the data in decision table should be expressed in discrete data, if some conditions or decision attribute is continuous value, which should be discrete Before process.Discretization is not specific data processing only by rough set theory , people have conducted extensive research on discretization problem before the rough set theory put forward , and Made a lot of progress ,but the discretization technique is can not be completely in common used in every subject, different areas have their own unique requirements and handling .This paper proposes a discretization algorithm based on regular conditional entropy.
Authors: Fang Yuan Wu, Feng Kong, Jiang Yun Yao
Abstract: This paper presents an intelligent fault diagnostic approach for a steer-by-wire (SBW) system. A rough set model is utilized to reduce the redundant information. On the base of the reduction, the classifying rules can be extracted. A radical basis function (RBF) neural network optimized by particle swarm optimization (PSO) algorithm is designed to learn the fault rules that are extracted from the reduction of the redundant information. The proposed approach is simulated in MATLAB. Simulation results show that the proposed intelligent fault diagnostic algorithm provides a higher level of diagnostic accuracy than the approach without any optimization.
Authors: De Wen Wang, Lin Xiao He
Abstract: With the development of on-line monitoring technology of electric power equipment, and the accumulation of both on-line monitoring data and off-line testing data, the data available to fault diagnosis of power transformer is bound to be massive. How to utilize those massive data reasonably is the issue that eagerly needs us to study. Since the on-line monitoring technology is not totally mature, which resulting in incomplete, noisy, wrong characters for monitoring data, so processing the initial data by using rough set is necessary. Furthermore, when the issue scale becomes larger, the computing amount of association rule mining grows dramatically, and its easy to cause data expansion. So it needs to use attribute reduction algorithm of rough set theory. Taking the above two points into account, this paper proposes a fault diagnosis model for power transformer using association rule mining-based on rough set.
Authors: Guo Qiang Sun, Hong Li Wang, Jing Hui Lu, Xing He
Abstract: Rough set theory is mainly used for analysing, processing fuzzy and uncertain information and knowledge, but most of data that we usually gain are continuous data, rough set theory can pretreat these data and can gain satisfied discretization results. So, discretization of continuous attributes is an important part of rough set theory. Field Programmable Gate Array(FPGA) has been became the mainly platforms that realized design of digital system. In order to improve processing speed of discretization, this paper proposed a FPGA-based discretization algorithm of continuous attributes in rough ret that make use of the speed advantage of FPGA and combined attributes dependency degree. This method could save much time of pretreatment in rough ret and improve operation efficiency.
Authors: Hong Xin Wan, Yun Peng
Abstract: The evaluation algorithm is based on the attributes of data objects. There is a certain correlation between attributes, and attributes are divided into key attributes and secondary attributes. This paper proposes an algorithm of attribute reduction based on rough set and the clustering algorithm based on fuzzy set. The algorithm of attributes reduction based on rough set is described in detail first. There are a lot of uncertain data of customer clustering, so traditional method of classification to the incomplete data will be very complex. Clustering algorithm based on fuzzy set can improve the reliability and accuracy of web customers.
Authors: Cheng Hua Wang, Lin Zhou, Feng Jiang, Hong Bo Zhao
Abstract: Decision tree algorithms have been widely used in intrusion detection. In this paper, within the framework of granular computing (GrC), we propose a new decision tree algorithm called DTGAE and apply it to intrusion detection. First, by virtue of the GrC model using information tables, we propose a new information entropy model, which contains two basic notions: approximation entropy of granule (AEG) and GrC-based approximation entropy (GAE), where the latter is defined based on the former. In algorithm DTGAE, GAE is adopted as the heuristic information for the selection of splitting attributes. When calculating AEG and GAE, we not only utilize the concept of conditional entropy in Shannon's information theory, but also use the concept of approximation accuracy in rough sets. Second, we present a method of decision tree pre-pruning based on Düntsch's knowledge entropy. Finally, the KDDCUP99 data set is used to verify the effectiveness of our algorithm in intrusion detection.
Authors: Na Su, Feng Feng Liao, Zhe Hui Wu
Abstract: The independency between two attribute subsets can be verified based on Chi square statistic to reduce candidate sets. Based on this measure, heuristic algorithm employing information entropy for reduction of decision systems is presented by combining rough sets and statistics. And the validity of this algorithm is analyzed.
Authors: Zhao Hui Ren, Yuan Hao, Bang Chun Wen
Abstract: Continuous attribute discretization based on rough set is to got possibly minimum number of cuts, and at the same time it should not weaken the indiscernibility ability of the original decision system. In order to obtain the optimal cut set of the continuous attribute system, based on research the choice of candidate cut set, this paper presents a heuristic genetic algorithm for continuous attribute discretization to decision tables. In this algorithm making the importance of the continuous cut as heuristic message, a new operator is constructed to not only maintain the discernibility of the cuts selected, but also improve local search ability of the algorithm. Compared the performance of this method with the others’, this method is proved effective and superiority.
Authors: Horng Lin Shieh
Abstract: In this paper, a hybrid method combining rough set and shared nearest neighbor algorithms is proposed for data clustering with non-globular shapes. The rough k-means algorithm is based on the distances between data and cluster centers. It partitions a data set with globular shapes well, but when the data are non-globular shapes, the results obtained by a rough k-means algorithm are not very satisfactory. In order to resolve this problem, a combined rough set and shared nearest neighbor algorithm is proposed. The proposed algorithm first adopts a shared nearest neighbor algorithm to evaluate the similarity among data, then the lower and upper approximations of a rough set algorithm are used to partition the data set into clusters.
Showing 1 to 10 of 249 Paper Titles