A Tool for String Transformation Enabling Flexibility in Real World Applications

Abstract: String is a sequence of characters whose manipulations have many real world utilities. Especially transforming a string from given input to various outputs has very important use in applications like bioinformatics, information retrieval, data mining and natural language processing. When an input string is given, the output has to be generated as multiple strings that can be used for solving problems such as query reformulation, correction of errors and so on. Recently Wang et al. proposed a probabilistic approach to string transformation. They proposed an algorithm for obtaining top-k candidates, a training model and log linear model for achieving string transformation which proved to be efficient and accurate. In this paper, we built an application, a prototype that can demonstrate the concept of string transformation. The experimental results reveal that the tool is useful in string transformation which can be used further in applications pertaining to error correction and query reformulation in online web search.

Keywords: String transformation, error correction, and query reformulation