# Chatbot: A Conversational Agent employed with Named Entity Recognition Model using Artificial Neural Network

**Authors:** Nazakat Ali  
**arXiv:** [2007.04248](https://arxiv.org/abs/2007.04248)  
**Date:** 2020-06-19  
**Summary page:** https://www.summarizepaper.com/en/paper/2007.04248v1/

## Key points

- Chatbots are used in various business domains to enhance customer service and satisfaction
- Chatbots mimic human behavior using natural language
- Chatbots require a domain-specific knowledge base
- Natural Language Understanding (NLU) is the core component of a chatbot
- NLU has seen improvements through deep learning methods
- Building high-quality NLU modules from scratch can be time-consuming and challenging
- The research focuses on integrating Named Entity Recognition (NER) and Intent Classification models into the NLU service of a chatbot
- Named entities are manually inserted into the knowledge base and automatically detected within user input sentences
- The proposed architecture for the NER model is based on an artificial neural network trained on manually created entities
- The model's performance is evaluated using the CoNLL-2003 dataset
- This research contributes to improving chatbot capabilities by enhancing NLU through NER and Intent Classification models
- Automating entity detection and classification improves chatbot responses and overall user experience in various business domains.

## Summary

The research paper titled "Chatbot: A Conversational Agent employed with Named Entity Recognition Model using Artificial Neural Network" by Nazakat Ali explores the use of chatbots in various business domains to enhance customer service and satisfaction. Chatbots are designed to mimic human behavior using natural language, and they require a knowledge base specific to the domain they operate in. The core component of a chatbot is Natural Language Understanding (NLU), which has seen significant improvements through deep learning methods. However, building high-quality NLU modules from scratch can be time-consuming and challenging. To address this issue, the research focuses on integrating Named Entity Recognition (NER) and Intent Classification models into the NLU service of a chatbot. Named entities are manually inserted into the knowledge base and automatically detected within user input sentences. The proposed architecture for the NER model is based on an artificial neural network trained on manually created entities. The model's performance is evaluated using the CoNLL-2003 dataset. This research contributes to improving chatbot capabilities by enhancing NLU through NER and Intent Classification models. By automating entity detection and classification, chatbots can provide more accurate responses and improve overall user experience in various business domains.