Stemming is the process for finding original form of the derived or irregular words. Stemming is mainly used in modern web search engines (i.e. semantic search), and NLP (natural language processing) issues.
Old fashion search engines built on string searching algorithms which are capable to find only exact matches of the given text withing datasets. It works well till desired search results should contain derived or irregular forms of the given input (i.e. “run” -> “ran”). Easiest solution for the problem is stemming.
Stemming deals with morphological structure of the words which depends on the language itself. That’s why it is easy to find stemming rules and algorithms for English and for Turkish I only found Snowball and going to test it within unstructured sample data.