SUCCESSIVE WORD PREDICTION

SUCCESSIVE WORD PREDICTION

Ch. Praneeth, Keerthi Veginati, Leela Sri Yarlagadda, Sirisha Pidamarthi, Supritha Murugu

Department of Information Technology

Prasad V Potluri Siddhartha Institute of Technology

Vijayawada, India

Abstract— Worldwide, people are using their mobile device, laptop, tablet, desktop, etc. for activities including messaging, emailing, banking, and social media connection anything else, including the media. The amount of time spent typing on these gadgets must be reduced. When the gadget can offer the user more possibilities for the potential next word for the currently written word, it can be accomplished. Also, it speeds up typing. In this paper, we propose and offer a comparative analysis of several models, including recurrent neural networks, stacked recurrent neural networks, long short-term memory networks (LSTM), and bi-directional LSTM, that provide a solution to the issue. Our main objective is to recommend which of the four models will forecast the following word for the supplied current word in the English language.

Keywords: Recurrent Neural Network (RNN), Long Short Term Memory (LSTM), Bi-Directional LST.