Abstract: A computer-implemented method to determine a part-of-speech (POS) category associated with a word in a text. The method includes determining a first set of candidate POS categories associated with the word based on a dictionary. The method further includes determining one or more contexts in which the word is used in the text based on a first set of rules. The method further includes determining a second set of candidate POS categories from the first set of POS categories based on the one or more contexts. The method furthermore includes determining the POS category from the second set of candidate POS categories based on a second set of rules.