Human Mimic Chatbot
Department of Computer Science and Engineering, ACE Engineering College, Hyderabad, Telangana, India.
Review Article
World Journal of Advanced Research and Reviews, 2023, 18(03), 1232–1239
Article DOI: 10.30574/wjarr.2023.18.3.1228
Publication history:
Received on 15 May 2023; revised on 24 June 2023; accepted on 26 June 2023
Abstract:
A Telegram bot that lets users enter a contact's name and a text file for a chat serves as the project's entry point. The bot turns the conversation data into a structured pandas dataframe after extracting it. The chatbot is then trained using the conversation data and the Microsoft Dialogue GPT model so that it can produce responses resembling those of the selected contact. The model is then deployed to the Hugging Face repository after training is finished, and users are given access to a run.py file. This Python programme interacts with WhatsApp using Selenium to keep track of new messages from the chosen contact. The chatbot concept is applied to construct an appropriate reply when a new message is received, automating the reply procedure. The project's benefits include improved productivity, tailored responses, and easy integration with well-liked messaging platforms. As the chatbot responds to incoming messages in line with the conversational style of the designated contact, users may now concentrate on their activities without the need for frequent manual engagement.
The project does, however, have several drawbacks, such as its reliance on the reliability and accessibility of the offered conversation history. Future improvements might use sentiment analysis, context awareness, and advanced natural language processing techniques to overcome these restrictions and improve the effectiveness of the chatbot.
Keywords:
Chatbot; Deep Learning; Transformers; Natural Language Processing.
Full text article in PDF:
Copyright information:
Copyright © 2023 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0