Most of the queries given to search engines are composed of a few keywords and have a length ranging from two to eight words. The queries given to search engines are mostly verbose, written in natural language using more than four terms that might contain redundant and unnecessary keywords for the purposes of IR rather than selecting a small number of well-focused keywords and cause ambiguity and topic drifts. This paper proposes a methodology to automatically detect verbose queries and modify them conditionally. The methodology proposed in this paper combines concepts from a linguistic database, uses topic gisting algorithm that detects and processes verbose queries along with succinct queries and provides a modified query to IR system. Documents are extracted according to the modified query using proximity analysis algorithm and documents are ranked accordingly and presented to the user. The proximity analysis process would aim at increasing the efficiency of the retrieval process and also performance of the IR system depending on the input query be it verbose or succinct. © Research India Publications.