Papers by Keyword: Knowledge Discovery

Abstract: Clinical data are inevitably incomplete, and most knowledge discovery algorithms lack the capability to contend with missing data. Flow-graph confers some distinct advantages in data mining and knowledge discovery. However, flow-graph methodology is not able to comprehensively solve the incomplete data problem. This paper proposes a flow-graph network approach for extracting knowledge from incomplete medical data. The concept of incomplete-medical-diagnosis-flow-graph (IMDFG) was defined. To evaluate the diagnosis rules within the IMDFG, the computing method for the certainty factor and coverage factor are presented. Moreover, the application of flow-graph network can be useful for extracting comprehensibility knowledge from the incomplete medical data. In an illustrative medical example, the clinical diagnosis rules are induced and interpreted in accordance to the generated flow graphs.

Abstract: Knowledge service is an important part of digital library. According to knowledge service mode of digital library, the paper builts a four-layer structure of knowledge service platform based on knowledge discovery. The platform consists of user layer, service layer, knowledge discovery layer and resources layer. Meanwhile the paper introduces the function modules of knowledge service platform.

Abstract: With the development of computer science and database system, millions of data will be generated every day. How to mining useful information and knowledge from large database is becoming a more and more popular research topic. In this paper, we introduced the origin of data mining, discussed some different data mining techniques in different fields, and made a conclusion of the application of data mining.

Abstract: With the explosive development of the Web2.0, Web education resources have been increasing dramatically. It is just so hard to mine some useful information among the Internet environment filled with a large number of Wed education resources. Due to the fact that human beings have strong ability to identify information quickly in visualization modes, we decide to change the Web resources into the visual form. Visual interface enables us to have more efficiency in observing, manipulating, researching, skimming, collating, comparing and comprehending large scale statistics of Web education resources^[1]. In this paper, we process multi-dimension data by some software, get visual results and analyze the Internet education texts.

Abstract: On the foundation of discovery of the traditional incomplete information system, this paper introduces ontology semantic similarity to improve the traditional similarity relationship, and eliminate the inaccuracy of discovery of knowledge brought by partial transmission. Based on the original model, this paper proposes a more obscure similarity relationship under the improvement of incomplete information system. Furthermore, this paper also advocates a knowledge discovery algorithm based on the combination of weak similarity relationship and attribute value similarity, which not only considers the VPRS of existing null semantics based on the weak similarity relationship and attribute value similarity, but also explores and studies the knowledge discovery method of non-existing null incomplete information system.

Abstract: Cloud computing is the latest trend in IT technical development, the importance of cloud databases has been widely acknowledged. There are numerous data in the cloud database and among these data, much potential and valuable knowledge are implicit. The key point is to discover and pick up the useful knowledge automatically. An association rule is one of the main models in mining out these data, and it mainly focuses on the relationship among different areas in the data. This paper puts forward the basic model of data mining based on association rules in cloud database and introduces corresponding mining algorithms.

Abstract: Fault prewarning is important to guarantee the safe and stable operation of Generator Sets (GS). In order to generate prewarnings quickly and accurately before the failures or faults occur in GS, a real-time fault prewarning approach to GS based on dynamic threshold was put forward. This approach was consisted of five steps, that is operating condition (or abnormal event) synchronous, dynamic threshold selection, threshold analysis, fault detection and fault prewarning. The dynamic threshold (closely related to operating condition or abnormal event of GS) was the key of this approach, which can be obtained by means of expertise knowledge discovery. This approach can effectively reduce false positives and false negatives for the faults of GS, whose effectiveness is validated by the applications and practices of Gezhouba power plant.

Abstract: Knowledge discovery model of making-decision table was established based on rough set theory and the wearing characteristics of the military launch vehicles hydraulic system. Compatibility relationship of the incomplete information system was used to achieve knowledge discovery from decision-making tables. Incomplete information systems knowledge representation which based on attribute reduction had been simplified, it had obtained simple and intuitive decision rules. The rules indicated that only monitor the normal wearing debris and non-ferrous debris in hydraulic oil could determine the wear state of the hydraulic system.

Abstract: Data mining is to discover knowledge which is unknown and hidden in huge database and would be helpful for people understand the data and make decision better. Some knowledge discovered from data mining is considered to be sensitive that the holder of the database will not share because it might cause serious privacy or security problems. Privacy preserving data mining is to hide sensitive knowledge and it is becoming more and more important and attractive. Association rule is one class of the most important knowledge to be mined, so as sensitive association rule hiding. The side-effects of the existing data mining technology are investigated and the representative strategies of association rule hiding are discussed.

Abstract: Massive data stored in a cloud database, in which lots of potential and valuable knowledge exist. In this paper, the data model of the cloud database is analyzed. Through analyzing, classifying, the common features of the data are extracted and form a feature data set, from which the new knowledge can be found. In the paper, the basic model based on the classification characteristic rules of the cloud database is defined, and the discovery algorithm of the classification characteristic rules is presented.