A FCA-based classification of uncertainty data using rough clustering

Although the amount of electronically stored data is continuously increasing on the Internet, there are no good solutions to easily deal with uncertainty contained in datasets. Formal concept analysis (FCA) classifies data based on the ordinary set into concept units which consists of objects and attributes that those objects have commonly. However, FCA is insufficient to process and analyze vague data, such as rough and fuzzy data. In this paper, we propose a new FCA-based approach for rough clustering in order to discovery implicit knowledge from given vague fuzzy datasets. Moreover, we show some experiments that demonstrate how our approach can be applied on Web mining. Our research results would be helpful for clustering and classifying the vague web data, in particular when dealing Web resources with the uncertainty.