Intelligent Chatbot System based on Entity Extraction using RASA NLU
👇Download Article👇
https://www.ijert.org/intelligent-cha...
IJERTV11IS020193
Intelligent Chatbot System based on Entity Extraction using RASA NLU
Raj Sinha , Pushpendra Kumar Singh
This paper is focusing on creating a chatbot from the RASA framework using custom components in the pipeline which is to be used by: 1) clients to get their queries responded easily from the telegram or website. The Enquiry Chatbot has the capacity to make friendly conversations; respond to the service and pricing details; give the link for the previous projects completed; answer the frequently asked questions; ask the clients for the budget and based on the clients input; and ask for the sample software which is to be made. Also a covid- 19 Tracker is made and integrated in Telegram which will answer the queries about Covid-19 and send emails about the covid-19, measures, precautions in a pdf to them after taking their phone numbers and email IDs. Maximum chatbots lack empathy and fail to understand anything which is out of the script. In order to address these problems, we used the RASA framework and implemented it in the telegram. Chatbot extends the implementation of the current chatbots by adding sentiment analysis and active learning. Upto some extent sentiment analysis can recognize the user's query as positive, negative and neutral, but the system was partially successful in adding empathy to the chatbot so that it can understand unscripted queries also. It is because the system requires more rigorous training data to handle all queries which are off script. However, for such queries, active learning helps to improve the chatbot performance since it correctly understands the user's questions, asks clarifying questions, and then retrains the system to give the response what the user intends to get. In this project, we will look into how, the recently released Rasa Core, which provides machine learning based dialogue management, helps in maintaining the context of conversations using machine learning in an efficient way.
