Intent Classification using BERT for Chatbot application pertaining to Customer Oriented Services

N, Dr Geetha and G, Mr Vivek and T A, Vinetia (2021) Intent Classification using BERT for Chatbot application pertaining to Customer Oriented Services. In: ICCAP 2021, 7-8 December 2021, Chennai, India.

[thumbnail of PDF]
Preview
Text (PDF)
eai.7-12-2021.2314563.pdf - Published Version

Download (496kB) | Preview

Abstract

In current scenario, customer interactions with live chat interfaces are very popular and real time customer service are offered in a better way. Traditional human chat services are replaced by conversational agents. The conversational agents or chat bots had become an integral part of business for customer services. Chat bots are the software designed to perform textual or audio conversations with the customers/users. Chat bots provide services with less time but at times fail to meet the customer expectations. Initial days, chat bots are designed by using simple pattern matching and string processing methods using rule-based models. Recently AI based chat bots using machine learning algorithms are developed to handle conversation context. To be more interactive like humans, Conversational AI chat bots are developed with Natural Language Processing (NLP) and Natural Language Understanding (NLU). BERT Model and go Bot are used to identify intent by chat bot and tested using dataset pertaining to networking domain. The results show that the accuracy of conversation is better and can be improved for large dataset.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: bert intent chat bot neural network contextual go bot
Subjects: T Technology > T Technology (General)
Depositing User: EAI Editor IV
Date Deposited: 29 Mar 2022 08:13
Last Modified: 29 Mar 2022 08:13
URI: https://eprints.eudl.eu/id/eprint/9869

Actions (login required)

View Item
View Item