Named entity recognition ontology chatbot

• Main Author: Lim, Zhi Yang
• Other Authors: Chng Eng Siong
• Format: Final Year Project
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering::Computer science and engineering::Computer applications
• Online Access: http://hdl.handle.net/10356/76122
• Institution: Nanyang Technological University

The proliferation of conversation agents is becoming prevalent due to the myriad of benefits it brings about. This has led to a revolutionised way of consumers interacting with machines. Traditionally, chatbots are only able to handle simple queries that allow the users to order one food item at a single instance in the food and beverages industry. To address this limitation, this report will be on the development of a chatbot that allows the processing of complex queries; such queries contains multiple items, size, quantities, special symbols, punctuations and multiple trailing blank spaces. A food ordering Chatbot in the domain of Named Entity Recognition (NER), was developed in this project. This Chatbot is a proof-of-concept prototype for entity extraction with concept ontology using Dialogflow as the intent. The focus of this chatbot is on the parsing of complex queries using named entity recognition and returning the total cost and summary of the food order back to the users in NTU North Spine Canteen A. This Chatbot supports the ordering of the food that are in English as well as dialects such as Hokkien, for food like “Char Kway Teow”. Based on the food items queried by the users in the Facebook Messenger, the Chatbot can extract out the entities of the items that identifies the name, size, quantity and category of food the item falls into. With the prevalence of the assimilation of data in everything we do in the upcoming Global Technology Revolution 2020, the development of advanced ontologies would be promising.