A .NET library for performing word stemming
EditByBSEditor: The Stemmer is an API for .NET applications that can be used to find the root of a word, for example, the root of programming is program. Stemmers are particularly usefull for search applications. When creating an index of words, you would typically stem each word before you added it to the index and Likewise when executing the search. Lets say that you had two pages about software development. O ne has the word programming the other has the word program. When someone performs a search for program you would really like both pages to be returned.
Most good search engines use a stemmer, and if you really need to implement full text search of a web site, you would probably be best of using an off the shelf search engine rather than trying to build your own. However, you may have need to implement some other type of indexing and searching that would not warrant the use of a search engine. If so, feel free to use this stemmer.
If you think this is cool, you may be wondering why I would give this away for free. Well, the library uses the Porter stemming Algorithm. Honestly, you can find code for the algorithm all over the place. We ported most of the code from the Java implementation written by Martin Porter (here ). You can also find source code implementations in a variety of languages, including C#.
So, you ask, why bother downloading this one when you can just get source code for another implementation? Not an unreasonable question. The Porter algorithm itself is an approximation. It works well for simple words like running, however, it cant handle irregular words, such as geese, and it can be a bit overzealous (it will trimm parents to par). So I addded capabilities to handle irregular nouns and verbs as well as in internal spell checker to make sure it spits out a legitimate word.If for whatever reason the input cannot be stemmed, it simply returns the word unchanged. you can free download .NET Stemmer 1.0 now.