A New Term Representation Method for Gender and Age Prediction

Author Profiling is a kind of text classification method that used for detecting the personality profiles such as age, gender, educational background, place origin, traits, native language, etc., authors by processing their written texts. Several applications like forensic analysis, security and marking are techniques author profiling finding basic details authors. The main problem in domain preparation suitable dataset predicting characteristics PAN one organization conducting competitions on various types shared tasks. In 2013, organizers presented task series continued this further years. They arranged different kinds datasets varieties languages. From 2013 onwards several researchers proposed solutions to predict features utilizing provided competitions. Researchers character based, lexical or word structural features, syntactic, content style based distinguishing author’s writing styles Most observed words phrases those most useful work, experiment conducted with important terms age group gender from competition datasets. Two 2014 2016 experiment. documents converted vector representation which format giving training machine learning algorithms. term document plays crucial role improve performance prediction.The Term Weight Measures (TWMs) purpose represent significance value representation. we developed new TWM representing TWM’s efficiency compared other existing TWMs. Machine Learning (ML) algorithms SVM (Support Vector Machine) RF (Random Forest) considered estimating accuracy approach. We recognized accomplished best accuracies prediction two Datasets.