Combination of Cloud Model and Neural Network to Find Knowledge from Imperfect Data in IDSS

• Published in: 2007 International Conference on Machine Learning and Cybernetics Vol. 5; pp. 2749 - 2754
• Main Author: Hong-Li Wang
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2007
• Subjects: Artificial neural networks; Cloud model; Clouds; Cognition; Data mining; Data pretreatment; Decision support system; Decision support systems; Imperfectness of data; Intelligent networks; Intelligent systems; Mathematical model; Mathematics; Neural network; Neural networks
• ISBN: 1424409721; 9781424409723
• ISSN: 2160-133X
• DOI: 10.1109/ICMLC.2007.4370615

Decision support system using data mining to find decision knowledge is called intelligent decision support system (IDSS). Neural network as a data mining method commonly is used to find classification knowledge in IDSS. But the classic data mining based on neural network is short in dealing with the blank value data or data with the character of blurring and randomicity. Such data is called as imperfect data. In order to overcoming this shortcoming the method of combination of cloud model and neural network to find knowledge from imperfect data in IDSS is proposed. Firstly the cloud is used to depict the imperfect data by group decision. In the following, attribution generation based on cloud model or grey cloud model is used to generate the upper concept layer. In this step the cloud model depicting the imperfect data is classified into the concept layer that is proximal to itself according to distance between two cloud models. Then classic neural network method is used to gain knowledge. The data is input into the neural network to training and gaining the classification knowledge. Lastly an experiment is given to verify the validity of the method.