Since the early 1970s, decision support systems (DSS) have evolved significantly. In this paper, the design and implementation of MSMiner, a developing platform for DSS, is introduced. The system is constructed on a data warehouse and integrated with a number of data mining algorithms. It is well suited for on-line analytical processing (OLAP). The characteristics of MSMiner include the ability to support multiple data sources and data mining strategies, additional organizational flexibility in regard to data and mining strategies, and the powerful expansibility of data mining tasks.

Rough集理论是一种新型的处理含糊和不确定性知识的数学工具,将Rough集理论应用于知识发现中的聚类分析,给出了局部不可区分关系、个体之间的局部不可区分度和总不可区分度、类之间的不可区分度、聚类结果的综合近似精度等定义,在此基础上提出了一种基于Rough集的层次聚类算法,该算法能够自动调整参数,以寻求更优的聚类结果 .实验结果验证了该算法的可行性,特别是在符号属性聚类方面有较好的聚类性能 .

Efficiency of algorithms is always an important issue concerned by so many researchers. Rough set theory is a valid tool to deal with imprecise problems. However, some of its algorithms’ consuming time limits the applications of rough set. According to this, our paper analyzes the reasons of rough set algorithms’ inefficiency by focusing on two important factors: indiscernible relation and positive region, and analyzes an equivalent and efficient method for computing positive region. And according to our research on the efficiency of some basic issues of rough set, a complete algorithm for the reduction of attributes is designed and its completeness is proved. Theoretical analysis and experimental results illustrate that our reduction algorithm is more efficient than some other algorithms.

This paper presents a document clustering algorithm based on swarm intelligence and k-means: CSIM. First, a document clustering algorithm based on swarm intelligence is employed. It is derived from a basic model interpreting ant colony organization of cemeteries. Swarm intelligence for flexibility, self-organization and robustness has been applied in a variety of areas. Taking advantage of these traits, good initial clusters are obtained in the first phase in CSIM. We then combine it with the classical k-means clustering method by using the clusters as initial centers. CSIM inherits the prominent properties of both swarm intelligence and k-means. It also offsets the weakness of those two techniques. Experimental results show the good performance of the hybrid document clustering algorithm

The idea of performing client-server computing by transmission of executable programs between clients and servers has become highly popular among researchers and developers who are engaged in intelligent network services. Computing based on mobile agents is an important aspect of this idea. This paper focuses on researching the migration process of agents. A model based on modules is devised for constructing agents. A concurrent schedule method is presented, with which the agent migration can be easily implemented. Most of unnecessary transmission of codes and data can be avoided by module reuse. Consequently, the executing period of mobile agents is reduced and their efficiency is improved. Additionally, a Fault-tolerance mechanism is designed in the system to ensure that the agent can work even when some faults occur in the network or in the host.

Improves on the classical formula of calculating the term weight in the vector space model. Furthermore, an approach to multi-hierarchy text classification based on the vector space model is proposed. In this approach, all classes are organized as a tree according to some given hierarchical relations, and all the training documents in a class are combined into a class-document. In order to construct the class models, only the class-documents attached to the same node of the same layer are compared. When classifying the documents, one matching process is hierarchically performed from the root node to the leaf nodes until a corresponding subclass is found. The experiment and real systems indicate that the approach is of high classification precision and recall