Gender Prediction From Name Using Deep Learni...

Gender Prediction From Name Using Deep Learning

Author Name : Sarumathy N, Dr. S. Meera

ABSTRACT Gender information is no longer a mandatory input while registering for an account. However, the prediction of demographic information such as gender and age remains an important task. The challenge is that HR professionals and recruiters are interested to balance gender diversity among their employees. The majority of today's recruitment occurs on LinkedIn. The gender of a person is not available on LinkedIn. The gender inferred will be on the name of the profile. The dataset comprises over 30,000 full names with genders. The name and gender dataset is downloaded from Kaggle. By analyzing the first names, it is found that genders can be very effectively classified using the composition of the name strings. In this paper, recurrent deep neural network models such as RNN, GRU and LSTM are examined and implemented to classify gender through the first name. A few evaluation measures such as precision, recall, accuracy, and loss are used to select the optimal model.