Hindi Language Stop Words List

Stop words are frequent, evenly distributed, function words in any document corpus which does not add any meaning to the text content. Information retrieval from the corpus is not getting affected by the removal of these words. It has been proved that removing the stop words reduces the document size to a considerable extent and
saves time in text processing in Natural Language Processing.

There are two sources where Hindi stop words are available online. First is Kevin Bouge list of stop words in various languages including Hindi . Second is sarai.net list . Third source can be translation of English Stop words available in NLTK corpus into Hindi using
translator. In this work, the Stop words list is the extended list using all three resources and contains words as well as phrases. The combined list is verified by one native speaker of Hindi language and finalized after necessary corrections.

For the first time, the phrases are also kept in the list because a word in present continuous verb form changes to a phrase when written in Hindi. In this way, a phrase can be removed from the document corpus in one attempt and the document corpus can be reduced, without effecting the sentiment expressed in it.

The stop words list has total 264 words and phrases, where 1 phrase is of the size of four words, 3 phrases are
of the size of three words, 18 phrases are of the size of two words and 242 words are of the size of one word.

This list of stop words are useful for performing many tasks in text mining/Natural Language Processing in Hindi language like sentiment analysis, text summarization, question answering etc.