/CloudComputing_HW1_sj3828_vb2279

Serverless Dining Concierge chatbot with AWS technologies to provide tailored restaurant suggestions.

Primary LanguageJavaScript

Dining Concierge Chatbot Project

Cloud Computing - CSGY 9223 - NYU Tandon

  • Suriya Prakash Jambunathan (sj3828)
  • Vamsi Krishna Bunga (vb2279)

Welcome to the Dining Concierge Chatbot project

Link to Bot

This project seeks to revolutionize the customer service experience by leveraging the power of Natural Language Processing.
Our goal is to create a responsive chatbot that provides restaurant suggestions tailored to users’ preferences, which are collected through conversational interactions.

Features:

  • Web Interface: Frontend hosted on AWS S3.
  • API: Set up via API Gateway with CORS enabled.
  • Chatbot: Developed on Amazon Lex, it gathers user preferences such as location, cuisine, and more, pushing data to an SQS queue.
  • Lex Integration: Seamlessly integrated into our API using AWS SDK.
  • Data Collection: Over 5,000 Manhattan restaurants sourced from Yelp, stored in a DynamoDB table.
  • ElasticSearch: Efficient restaurant search capability indexed under “restaurants”.
  • Recommendation System: Fetches suggestions from ElasticSearch and DynamoDB, and sends them to users via SES.

This project is a comprehensive demonstration of serverless architecture, microservices, and integration of multiple AWS services to enhance the dining experience for users.

Chatbot Architecure

Conversation with Dining Suggestions Chatbot

User returns to chat session in a different browser

Scenario 1: User accepts previous recommendation

Scenario 2: User denies previous recommendation