Context Urdu is an indo-Aryan language with over 100 million speakers. It is mainly used in Pakistan. While often mutually intelligible with Hindi, the two languages use different scripts and as a result NLP tools developed for Hindi are often not extendible to Urdu. Some words, like “the” or “and” in English, are used a lot in speech and writing. For most Natural Language Processing applications, you will want to remove these very frequent words. This is usually done using a list of “stopwords” which has been complied by hand. Content This dataset is a list of 517 high-frequency Urdu words. Very high frequency words are not generally useful for most NLP tasks and are generally removed as part of pre-processing. The .json and .txt files have the same words in them. Acknowledgements: This dataset is Copyright (c) 2016 Gene Diaz and distributed here under an MIT license. See the attached license file for more information. Inspiration: This is more of a utility NLP dataset than one that is interesting in its own right. Try using it with these other Urdu datasets: * [Urdu-Nepali Parallel Corpus: A part of speech tagged corpus for Urdu & Nepali](https://www.kaggle.com/rtatman/urdunepali-parallel-corpus) * [The Holy Quran (in 21 languages, including Urdu)](https://www.kaggle.com/zusmani/the-holy-quran) * [Old Newspapers: A cleaned subset of HC Corpora newspapers](https://www.kaggle.com/alvations/old-newspapers)