
In this modern era, a lot of data is available to all the organisations so the real task is to gather the useful information from the data. There is a need to develop a robust technique that can resolve this problem. Clustering is a technique in data mining that groups together the similar data items into cluster. The data in a cluster is more similar to each other than data in the other cluster. There are different types of clustering algorithms like K–means (partitioning based), greedy based, hierarchical based algorithms, density based. Clustering has a wide range of application such as text mining information retrieval, Business analytics, data analysis, machine learning etc. In this we are using K-means clustering, firstly we have discuss how K-means algorithm is implemented then we have discuss the advantages and disadvantages of K-means and then finally we have selected one of the shortcoming of K-means algorithm. Then we have proposed are own method to improve this shortcoming and then we applied this algorithm on textual dataset. Finally we get the result of our algorithm.