Analysis of Text Clusters Based on Fuzzy and Rough K-Means Strategies

There are several issues in the present data mining industry regarding the text classification process and finding the accuracy parameters while implementing fuzzy. Based on the fuzzy rule the results are purely produced on the probabilistic mean, so no one can guarantee regarding the classified output. Regularly the rough K-Means strategy achieves the performance, or speed, or accuracy. Additionally applying the Association Rule Mining, this gives feature subset from the transactional data in an efficient manner. The major goal of this approach is to apply association rules to generate quicker results that enhance algorithm process. By means of the Ofrecca (one by one Fuzzy Relational Eigenvector Centrality-based Clustering Algorithm) in normal text, the conditions are not connected to more than one. A rough k-means algorithm is also used for fixing the cluster heads in more efficient manner. A single constituent can fit in to manifold cluster, but using the k-means algorithm every constituent can precisely fit in to one cluster. Both HFRECCA and rough k-means clustering algorithms are used with association rule to achieve the fuzzy relation in successful manner. It generates the quicker results and to reduce the classifier computational weight and augment the information steadiness.