The World Wide Web contains a huge amount of unclassified data and its continuous growth has made it a complex domain for information retrieval. Current Web information retrieval (IR) systems (i.e., search engines) very often overload the user with irrelevant search results. This has forced the user to perform a certain level of analysis on the results returned. Web IR systems are currently one of the most researched areas in the computer industry. So far there have been many attempts to incorporate soft computing techniques such as fuzzy logic, neural networks, genetic algorithms, etc. This work focuses on how fuzzy logic can be introduced to IR systems. The current applications of fuzzy techniques are analyzed and a concept called "macro-clustering" is introduced as a solution for optimizing results of generalized search queries.