Development of a Technique for Gender Recognition on Twitter Users
Jatinder Kaur · International Journal for Research in Applied Science and Engineering Technology · 2018
Content is as utmost common Internet media sort. Instances of incorporate prominent social networking sites, for example, Twitter, Facebook, Craigslist and so on. Other web applications for example, email, chat rooms, blog and so forth are likewise generally message based. An inquiry we address in this paper manages content based Internet crime scene investigation is the accompanying: given a short content record, would we be able to distinguish the author of text as male or female? The analysis of author classification is influenced by late occasions where individuals try to counterfeit their sexual orientation on the Internet. In this paper, we propose feature selection with hybrid classification technique. Data set is collected and data preprocessing is performed to mine the data and feature selection is performed to reduce the dimensionality of dataset. Classification is done to improve the performance of individual classifiers. The proposed method is tested along various parameters like accuracy, precision, recall, f Measure.