Author(s): Vincent Marcov
The researchers argue that while performing document clustering, calculating the similarity measure between a pair of documents is a significant computational burden. Similarity measure involves determining the extent of similarity between a couple of reports then they are assigned a real number (0-1) . A value of one shows that the documents are essentially the same while a value of zero signifies that the documents are entirely disparate. Conventionally, the vector-based models are usually relied to compute similarity between a pair of documents.