About this EdVid
These are solutions to build Chatbot, based on that towards developing a complete and sustainable Chatbot applied to business activities of the retail chain stores.
Objectives
A chatbot can automatically identify new customers or close customers, communicate with customers by voice, and allows customers to complain or order food, or drink at the store, the system switches from voice to text, and text can understand the meaning of a customer's statement, the system also identifies the emotion of customers to make appropriate communication and using data mining from users to give customers the right items.
The scope of the project application
Testing application chatbot to serve retail chain in Can Tho city.
• Business process:
- Chatbot consists of 2 basic components mounted:
• Component 1: Identify customers and identify emotions through a chatbot camera
- A customer standing at the chatbot, the chatbot will identify the face which is closest to the chatbot and gives greetings to the customer.
- At the same time, the data mining system will provide data recommender suitable items to the customer, then the system will switch to dialogue with the customer.
- The emotion identification will be integrated into the complain functional dialogue to provide the appropriate response to the customers' emotions.
• Component 2: Dialogue System
The chatbot will communicate with customers standing in front of the chatbot:
- Step 1: Chatbot welcomes the customer and receives requests from the customer
- Step 2: Customer says their request (Speech)
- Step 3: System to translate speech to text(Speech2Text)
- Step 4: The system translates text to semantic/context , it's Understanding (Text2Context)
- Step 5: Context comparison system with answers from database to find the appropriate answer (Context2Text)
- Step 6: Pronunciation system for answers (Text2Speech).