Design and Development of a Rule-Based Urdu Lemmatizer

Language is known to be one of the tools for communication in a translingual society. It is composed of many elements and the basic fundamental part of a language is the structure of words. Understanding the structure of word is not only necessary to gain the proper understanding about a language, but also an important factor for language translation. The words have numerous variant forms based on its usage; depend has variants as dependency, dependent, independent, etc., where depend is a root word. To drop the root from its variant form, some tools are required like Stemming or Lemmatizer. But to extract correct and meaningful root word, the mechanism of lemmatizer should be used because it is not always possible to use stemming to find the meaningful root word. Therefore, lemmatizer is an extended mechanism of stemming. In this paper, the rule-based Urdu Lemmatizer is created that works by eliminating suffix from the root word and adds some required and relevant information to extract the meaningful root.