Natural Language Integration into a React Based User Interface
Click the Poster to View Full Screen, Right click to save image
Grant: American Council on Education and Kean University
Cymantha Blackmon
CoPIs:
Habiba Morsy, Essence Toone, Mostafa Moamen
College:
The Dorothy and George Hennings College of Science, Mathematics, and Technology
Major:
Computational Science & Engineering
Faculty Research Advisor(s): Ensela Mema, George Avirappattu
Abstract:
The goal of this project is to develop a cohesive and immersive conversational experience for users, focusing on personalized AI interactions and conversation flow. We focus on the development of a question and answering application tailored to the Carnegie Classifications of Institutions of Higher Education (CCIHE), aiming to optimize the integration of natural language processing models with a React based user interface. React is a JavaScript library developed by Facebook that allows developers to build modular interfaces via a component-based architecture where reusable UI components can be updated and managed efficiently. Testing has been conducted using two advanced Large Language Models (LLM), GPT 3.5 (OpenAI) and Llama II (Facebook), both employing artificial intelligence-driven language comprehension to generate appropriate responses to user queries. During the model development, we employ Retrieval Augmented Generation, an AI framework that allows the LLMs to retrieve and generate responses tailored to specific categories or phrasing within user inquiries.