Please use this identifier to cite or link to this item:
Title: A Text Based Conversational Agent for a Higher Education Institute
Authors: Malalasena, C. N.
Keywords: Conversational Agent
Natural Language Processing
Natural Language Understanding
Issue Date: 2-Aug-2021
Abstract: As more and more businesses turn to doing business virtually through World Wide Web, providing best experience to their website visitors has become an important factor. Even though interacting with a website is a familiar task for many, it is a comfort to have a human like conversation with a customer care agent representing the company when a virtual company is visited. This requirement has given high importance to Conversational Agents also known as ChatBots which are the software that communicate with humans, in a natural language, like a real person. This document describes the study of available technologies to develop a Conversational Agent for an Educational Institution which provides Undergraduate Degree Programs. Then develop the desired product with suitable features and Evaluate its performance. As chatbots has become latest addition to the software business, there are many organizations providing their user friendly product to implement a chatbot for you, on your own, initially for free. Or you have the Open Source chatbot frameworks like RASA under GNU General Public License. This project implements a chatbot using RASA Framework, along with Python as the programming language and the database server MongoDB. The chatbot is capable of greeting the visitor upon arrival to the chat window, and offer assistance by providing information regarding the available degree programs and answering other general questions that a visitor might ask in the first visit or the phone conversation. It is also capable of extracting bits of information like contact numbers, names, from the text typed in by the visitor and store them in the Database.
Appears in Collections:2019

Files in This Item:
File Description SizeFormat 
2016MIT042.pdf1.51 MBAdobe PDFView/Open

Items in UCSC Digital Library are protected by copyright, with all rights reserved, unless otherwise indicated.