Urdu is a widely known/ spoken language used by approximately 200 million people worldwide and there is an abundance of raw Urdu text available. To refine and extract useful information, our product: Urdu Stemmer utilizes the data mining techniques such as Stemming. The stem of words (news) خبروں is (news) خبریں (newspapers) اخبارات (News-papers) اخبار is خبر (news). In this project, we have built a stemming system for Urdu text. This system performs stemming on Urdu words as well as words borrowed from other languages such as Arabic, Persian and Turkish. The state of the art stemmer has extensively improved search accuracy which is why it is being utilized for translation algorithms for systems at NADRA.