This document discusses web clustering engines. It begins by defining search engines and their role in returning ranked search results for user queries. It then explains that conventional search engines can be inefficient for ambiguous queries, as results may be irrelevant and mixed together. Web clustering engines aim to address this by grouping search results into meaningful labeled clusters. The document outlines the architecture and techniques used by web clustering engines, including acquiring search results from engines like Google, preprocessing them by identifying language, tokenizing, stemming and selecting features, and representing the results as vectors in vector space models.