International Journal of Soft Computing and Engineering (IJSCE) ISSN: 2231-2307, Volume-5 Issue-1, March 2015 24 Retrieval Number: A2517035115/2015©BEIESP Published By: Blue Eyes Intelligence Engineering & Sciences Publication An Advanced Precision based Approach to String Transformation B. Sankara Babu, K. Rajasekhar Rao, P. Satheesh Abstract: Distinct obstacles occur in Natural language processing, Knowledge Engineering, Information Retrieval, Genetics Informatics, Computational molecular biology and Data Mining concerned to String Transformation. Consider an input string, the system automatically produces top k output strings referring to input string. Generally people perform various kinds of spelling errors such as misspell words accidentally while surfing the web. To circumvent such errors, this Paper propounds an advanced Precision based approach to string transformation which is very accurate. The proposed system comprises unique precision value allocated to each alphabet and these are aggregated to give the Total Precision of the particular word. Data sets are trained with the precision based approach by validating them to dictionary called the database. Misspell word precision is compared with the data sets precision and retrieves the top k nearest neighbour output strings relevant to input string. This is one of the best accurate Misspell word and sentence correction approach and experimentally proven on large data sets. Keywords: String Transformation, Precision based Approach, Misspell words, Total Precision. I. INTRODUCTION Required transformations that are necessary for the replacement of source string with different destination strings defined as String Transformation. In Natural language Processing, String Transformation is illustrated as Misspell word correction, word metamorphosis, word transition. In Information Retrieval, String transformation deals with retrieving the relevant records regarding the query. In Genetic Informatics, it is used to detect the disease in a person by using the standardized DNA patterns. In Data mining it is used as excavating the metonyms and opposite words. In String transformation, Input string is the string comprises the words, group of letters, tokens. Top k corrected output strings are produced by conducting various transformations on an input string which is misspell. Output strings are the strings which are corrected and generated by referring to the dictionary. Transformation is the several operations to replace the input string with different output strings. Correction of misspell word is done in two steps. In First step, it deals with the aspect of Keywords generation. In this step Misspelled words are corrected and distinct keywords are generated. Manuscript Received on February 16, 2015. B. Sankara Babu, Assoc. Prof., Department of CSE, Gokaraju Rangaraju Institute of Engineering and Technology Hyderabad, India. Dr. K. Rajasekhar Rao, Prof., Department of CSE, Koneru Laxmaiah University Guntur, India. Dr. P. Satheesh, Assoc. Prof., Department of CSE, Maharaj Vijayaram Gajapathiraj College of Engineering Vizianagaram, India. For example consider a misspelled word “Thught”, here the original word is “Thought “, here the letter’ o’ should be placed to make the misspelled word as correct word. Similarly, many other words are generated related to the misspelled word. In second step it deals with the aspect of Best key word selection. In this step Best key word is selected from distinct keywords based on the priority. The words which are wrongly typed by one character or group of characters to which certain modifications are needed to make them as correct words are defined as Misspell words.Table1 describes the misspell words correction. In the table we considered wrong words which are corrected by operations such as insertion, deletion, substitution to make them into original words. Table 1. Describes the misspell words correction. Misspell word Original Word Wrongly typed character Correction Action Performed Flowr Flower - e inserted Mengo Mango e - inserted Beauetiful Beautiful - - deleted Technlgy Technology e oo inserted Innovatiev e Innovative e - deleted Approech Approach e - substituted Word document contains misspell words are corrected syntactically and semantically by implicitly using ABC Spelling and Grammar. Figure1: Illustrates the correction of misspell word in word pad.