Interval Fuzzy C-means Approach for Incomplete Data Clustering Based on Neural Networks

Interval Fuzzy C-means Approach for Incomplete Data Clustering Based on Neural Networks

In the field of data mining and machine learning, the problem of recovering missing values from a dataset has become an important research issue. Recently, the numerical values may not be suitable for describing the uncertainty of missing attributes, and there is a certain degree of error. Hence, we propose an efficient interval approach which utilizes a missing-data back propagation to estimate the error value of the complete property of the missing samples and convert the value of the missing attribute to the form of an interval. Furthermore, fuzzy C means performs clustering analysis on the recovered data set. Therefore, the numerical data set is converted into an interval valued fuzzy C means clustering analysis, and the final clustering results are obtained. The experimental results demonstrate that our algorithm has good accuracy in data clustering performance.