Document Clustering is the process of segmenting a particular collection of text into subgroups. Nowadays all documents are in electronic form, because of the issue to retrieve relevant document from the large database. The goal is to transform text composed of daily language in a structured, database format. In this way, different documents are summarized and presented in a uniform manner. The challenging problem of document clustering are big volume, high dimensionality and complex semantics. The objective of this paper is mainly focused on clustering multi-sense word embeddings using three different algorithms(K-means, DBSCAN, CURE). Among these three algorithm CURE gives better accuracy and it can handle large databases efficiently.</p>