dc.contributor.authorLeon, Shi Min
dc.date.accessioned2019-05-07T03:01:35Z
dc.date.available2019-05-07T03:01:35Z
dc.date.issued2019
dc.identifier.urihttp://hdl.handle.net/10356/77094
dc.description.abstractCustomer service is an important aspect of a business, regardless of industry and sizeof a corporation. An important area of improvement for a human operated customerservice is providing a professional, timely and reliable customer service. In this the-sis, we investigate the possibilities of having a customer service powered by ArtificialIntelligence, and what are the advantages that it brings. The purpose of this investiga-tion is to provide an efficient communication platform for Agoda Company, an onlinetravel reservation platform, to facilitate the communication between Agoda’s customerand its hotel. We propose using natural language processing to change text data, thatare meant for humans comprehension, to vectors, for the comprehension of machinelearning algorithms. This is followed by passing the data through an Artificial NeuralNetwork to predict the intentions of the customers who are trying to contact the hotels.This predictions would help the organization to determine the right actions requiredto address the inquires in an automated manner. Experiments have shown that artifi-cial neural network can perform reasonably well, with the use of various regularizationmethods, to prevent the model from over-fitting.en_US
dc.format.extent45 p.en_US
dc.language.isoenen_US
dc.subjectDRNTU::Engineering::Computer science and engineering::Computing methodologies::Artificial intelligenceen_US
dc.titleAutomated customer service powered by artificial intelligenceen_US
dc.typeFinal Year Project (FYP)en_US
dc.contributor.supervisorCheong Siew Annen_US
dc.contributor.schoolSchool of Physical and Mathematical Sciencesen_US
dc.description.degreeBachelor of Science in Applied Physicsen_US
dc.contributor.organizationAgoda.comen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record