This talk will cover the problems of clustering data with missing values. Clustering is regarded as one of the significant tasks in data mining and has been widely used in very large datasets. Real-world data usually contain some missing values as well as outliers. However, the well-known traditional approaches to restoring gaps could be applied only in cases when a source data table is set a priori and a number of its rows or columns cannot be changed during the whole processing. But it is more typical that data are fed for processing consistently in real time, while it is not known in advance which incoming vectors contain gaps. The fuzzy clustering approach could be applied in this case to establish the presence and the nature of hidden patterns in the data as well as for data restoration.