Hi, The Rabin Karp algorithm performs better than KMP and Boyer Moore during multiple pattern searching. The interface that I have suggested is given below. Please let me know your opinions and if there is a better approach than the one I am suggesting. Any feedback/criticism/guidance will be appreciated and helpful. An overview of the Rabin Karp Algorithm: The Rabin Karp algorithm assigns hash values to the text. The pattern is made to 'slide' along the text to be searched. The hash value of the text is computed and is compared to the hash value of the pattern. A brute force search of each element in the pattern with the current window of text is performed if and only if the hash value of the text matches the hash value of the pattern (and this hash value can be calculated in O(1) time). There can be two interfaces that can be implemented for this a procedural and an object oriented interface (similar to the interfaces for the Boyer Moore Algorithm). There should be two iterator types. One to iterate through the text called 'textIter' and another to iterate through the set of patterns (multiple patterns can be stored as part of a set) called 'setIter'. In the object oriented interface, the hash value of the pattern can be precomputed in the constructor. The text can be searched to match the pattern using the () operator (similar to the implementation of Boyer Moore). In the procedural interface, the search as well as hash computation of the pattern string all happens in the single function 'rabin_karp_search'. template <typename patIter> class rabin_karp { public: rabin_karp ( patIter first, patIter last ); ~rabin_karp (); template <typename textIter> template <typename setIter> textIter operator () ( textIter text_first, textIter text_last, setIter set_first, setIter set_last); }; template <typename patIter, typename textIter> textIter rabin_karp_search ( textIter text_first, textIter text_last, setIter set_first, setIter set_last ); Thanks and Regards, Meghana