The rapid expansion of digital content on the internet has significantly increased the volume of online documents, making the tasks of managing, searching, and retrieving information more complex. One of the most challenging aspects of this process is accurately assessing the similarities and differences between documents or their attributes. Various clustering algorithms have been proposed to address these challenges by determining the degree of similarity between elements in a dataset. Cluster analysis involves partitioning a set of N objects into smaller groups or clusters, such that the objects in the same cluster exhibit higher similarity to each other than to objects in other clusters. The similarity between objects can be explicitly or implicitly defined, depending on the context. This paper introduces a new approach to measuring similarity in document clustering based on multiple viewpoints (MVS). This method is compared with traditional K-means clustering algorithms. The fundamental difference between MVS and traditional approaches lies in the use of multiple viewpoints instead of a single viewpoint, which is common in traditional clustering methods. In this new approach, objects such as documents are measured not only with respect to their own cluster but also from viewpoints external to the clusters they belong to. The comparison between the K-means algorithm and the proposed incremental multiviewpoint-based clustering method is conducted, and simulation results demonstrate that the latter offers improved accuracy in clustering document data. The proposed method is implemented using Java programming, and the results highlight the advantages of the Incremental Multiviewpoint-Based Clustering approach in document similarity measurement.