Skip to content

This publication, presented at an IEEE conference in 2020, unveils a chatbot designed through deep neural networks, advancing human-machine conversations with nuanced understanding and responses.

Notifications You must be signed in to change notification settings

Krishna-Vamsi-G/Chatbot-A-Deep-Neural-Network-Model

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 

Repository files navigation

Chatbot: A Deep Neural Network Based Human to Machine Conversation Model

Authors

  • G Krishna Vamsi, Computer Science and Engineering, Maulana Azad National Institute Of Technology, Bhopal, India
  • Akhtar Rasool, Computer Science and Engineering, Maulana Azad National Institute Of Technology, Bhopal, India
  • Gaurav Hajela, Computer Science and Engineering, Maulana Azad National Institute Of Technology, Bhopal, India

Abstract

A conversational agent (chatbot) is computer software capable of communicating with humans using natural language processing. Despite many developments in Natural Language Processing (NLP) and Artificial Intelligence (AI), creating a good chatbot model remains a significant challenge. This paper proposes a new method of creating a chatbot using a deep neural learning method, where a neural network with multiple layers is built to learn and process the data, enhancing chatbots' functionality and interaction quality.

Keywords

Machine Learning, Conversational Agent, Chatbot, Neural Networks, Deep Learning, Natural Language Processing.

Introduction

The development and effectiveness of chatbots have been a focal point of research in recent years, aiming to create models that can mimic human conversations accurately. Our work focuses on leveraging deep neural networks and advanced NLP techniques to build a chatbot that can perform various tasks, understand the user's intent, and deliver appropriate replies.

Contribution

Our research introduces a novel chatbot model trained on a diverse dataset encompassing 14 different tags with corresponding patterns and responses. The model demonstrates significant advancements in understanding and generating human-like responses, making it suitable for various applications, including customer service and healthcare.

Dataset

The dataset used in our study is derived from an open-sourced Kaggle healthcare services competition, comprising queries and keywords mapped to corresponding responses, showcasing the model's ability to handle real-world conversational scenarios.

Experimental Results

The study evaluates different neural network architectures, optimizers, and weight initializers, concluding that certain combinations yield optimal performance, particularly in accuracy and response generation.

Conclusion

Our findings suggest that deep learning models hold substantial promise in enhancing chatbot interactions, making them more natural and human-like. Future work could explore integrating voice and visual inputs to broaden chatbots' applicability.

Full Paper

Access the full paper here: Chatbot: A Deep Neural Network Based Human to Machine Conversation Model


This repository contains the source code and dataset used for the research paper titled "Chatbot: A Deep Neural Network Based Human to Machine Conversation Model" presented at the 11th ICCCNT 2020, IIT Kharagpur.

For more information, please refer to the publication.

About

This publication, presented at an IEEE conference in 2020, unveils a chatbot designed through deep neural networks, advancing human-machine conversations with nuanced understanding and responses.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages