Phonetic Redundancy Avoidance Technique

Phonetic-based Algorithm is beneficial to avoid the data redundancy. There are numerous Phonetic Algorithm available such as Soundex, D-M Soundex, Metaphone, and Kolner Phonetic Algorithm. The objective of this research is to develop modified phonetic algorithm to minimize data duplication. Our research focused on the existing three phonetic-based algorithm Daitch–Mokotoff Soundex, Kolner Phonetik Metaphone. The purpose of our research is apply modified version of phonetic algorithm to sample dataset to English names, European name, Jarman names, and verify the results. It is observed that phonetic algorithm used for many decades but there is no any universal Phonetic-based Algorithm Existing Phonetic-based Algorithm is used for the particular work. Like Daitch, Mokotoff Soundex algorithm is developed for the European-based Surname, it will show better result for ‘Slavic’- and ‘Yiddish’-based surnames. The Metaphone algorithm is developed for the English language. The Kolner phonetic-based algorithm developed for German language names or surnames. This is suitable for German names. In all scenario the Soundex phonetic algorithm in inconsistent so these Phonetic Algorithm are to be enhanced. Our experimental results show 45.62% accuracy for Daitch–Mokotoff Soundex algorithm which is suitable for European Name, The Metaphone algorithm gives 93.25%. Useful for English language names and the Kolner Phonetik-based phonetic algorithm show 34.26% useful for the German Language, respectively. These phonetic-based algorithms are developed for the specific language; this is limitation of existing research so in current system proposed the universal algorithm which is more efficient than other phonetic algorithm. The developed phonetic-based algorithm gives the 98.00% accuracy for the all different kind of regional languages.