Next Word Prediction in Hindi Using Deep Learning Techniques

Radhika Sharma, Nishtha Goel, Nishita Aggarwal, Prajyot Kaur, Chandra Prakash

2019 · 35 citations · 9 references

Concepts

TL;DR

NLG aims to generate natural, human‑interpretable language and can be applied to tasks such as story and sentence auto‑completion. The study proposes a novel method for predicting the next word in Hindi sentences. The method employs deep learning models, specifically LSTM and Bi‑LSTM, to predict the next word. The approach reduces user keystrokes and achieves 59.46% accuracy with LSTM and 81.07% with Bi‑LSTM.

Abstract

Natural Language Generation (NLG) focuses on the generation of natural, human-interpretable language. This study proposes a novel methodology to predict the next word in a Hindi sentence. By predicting the next word in a sequence, the number of keystrokes of the user can be reduced. Two deep learning techniques namely Long Short Term Memory (LSTM) and Bi-LSTM have been explored for the task of predicting next word and accuracy of 59.46% and 81.07% was observed for LSTM and Bi-LSTM respectively. This approach may be used for various NLG tasks like story auto-completion, sentence autocompletion, etc.

References

9