E-mail autocomplete function using RNN Encoder-decoder sequence-to-sequence model

2021 5th International Conference on Electronics, Communication and Aerospace Technology (ICECA), pp. 710–714

Abstract

Text Auto-complete feature suggests a stream of words which complete a user's text as the user types each character. Such a feature is used in search engines, email programs, source code editors, database query tools etc. Earlier people have used traditional language models for this problem. But for better performance, a Neural network-based language model is needed. Here, an encoder-decoder based sequence-to-sequence language model has been used for performing text generation and the empirical results show that the model effectively suggests the incomplete sentences.

Authors 5

1. Rohan Modi

   Nirma University

   Affiliation as printed

   Nirma University

2. Kush Naik

   Nirma University

   Affiliation as printed

   Nirma University

3. Tarjni Vyas

   Nirma University

   Affiliation as printed

   Nirma University

4. Shivani Desai

   Nirma University

   Affiliation as printed

   Nirma University

5. Sheshang Degadwala Aachen Computer Engineering Sigma Institute of Engineering

   Sigma Engineering (Germany) · Sigma University

   Affiliation as printed

   Computer Engineering Sigma Institute of Engineering