Singular Value Decomposition is the process of breaking down a significant database into much smaller volumes to determine document relevance for numerous items via carrying out comparisons with other documents and items. The first step in the process is stemming which involves taking into account the various versions of a term. The next step is local weighting as per which the relevance of documents on the basis of the number of times a certain term appears in it is increased. The third step is called global weighting which involves increasing the relevance of search terms while the final step is normalization which is the process of penalizing or rewarding copies as applicable.