Many real-world applications of edge-labeled graphs are now an active area of research, like social networks, protein-protein interactions, bibliographic data, etc. Lots of information can be extracted by recognizing the cohesive groups existing in these data networks. But this cannot be achieved through conventional clustering methods. Chromatic Correlation Clustering is a newly proposed formulation of the problem of discovering patterns in data networks. The proposed algorithm is named "Informed Chromatic Balls". Comparison with original CB is done over synthetic datasets of varying structures. ICB shows better performance than CB in all cases. Fuzzy clustering in a multi-view setting has been attempted for the first time in the proposed work. A global objective function for multi-view fuzzy clustering is proposed which includes the weights of attributes and views. A simple iterative algorithm for clustering is proposed. Experiments over data collected from real-life have been designed to test the performance of the proposal. Comparison with single view Fuzzy C-Means and a multi-view clustering algorithm has been done.