LOCATION PREDICTION OF ANONYMOUS TEXT USING AUTHOR PROFILING TECHNIQUE

Author Profiling is used to predict the demographic characteristics of the authors like age, gender and location by analyzing their written text. Several researchers proposed different types of approaches to predict the gender and age of the authors from different types of corpuses. Few researchers concentrated on location prediction of the texts. In this work, we concentrated on prediction of the location of the authors. Most of the existing work in Author Profiling concentrates on extraction of various types of stylistic features to differentiate the writing style of the text. In this work a profile specific document weighted model is used to predict the location of the authors. In this model content based features were used to compute the weights of the documents specific to each profile group. These document weights were used to represent the vectors of documents for generating classification model. The obtained results were good when compared with exiting approaches for location prediction.