UNISEL Bot: Designing Simple Chatbot System for University FAQs

Abstract

The UNISEL Bot system was developed for helping marketing department in order to help on giving information in interactive mode for marketing purpose. The current problem is the information are not served in interactive ways, manually serving information using portal and paper are complicated and there is no real-time customer support to help on question and answer. In the era of technology, the information should be served in an interactive platform. The interactive information tends to gather more user attention. Therefore, this project aims to develop marketing assistant Chatbot system for a private academic institution which known as UNISEL Bot. The Chatbot system development is expected to assist the marketing department to use smarter marketing and interactive ways, for instance; to receive FAQs from student and provide real-time feedback whilst encourage people to engage with latest technology. Agile methodology was used in the development of this Chatbot system. Qualitative data gathering using interview method with students and University staffs was implemented. Multiple diagram is presented in this paper to describe the process flow of UNISEL Bot system. UNISEL Bot system was made up of seven main modules including the Ask Question, Feedback, Registration, Event, Appointment, Survey and Map. In future, this Chatbot system can work effectively to replace the traditional method of manual customer service and can also helping people in capturing user data for building analytic data.

PixieGPT: Design and Implementation of a Generative Pre-Trained Transformer for Universities of Bangladesh

“…In a study by XINGGUANG et al (2022), a chatbot was designed for manufacturing, specifically for equipment like boilers, pressure vessels, elevators, and more. [16]Using Pressure Vessel Manufacturing (PVM), Quora Question Pair (QQP), LCQMC, and CCKS dataset for implantation. In manufacturing chatbot uses labeled corpus that is transformed into a word vector through the BeRT pretrained language model.…”

Perancangan Chatbot Hotel dengan Model Natural Language Processing Chatbot dan Button Based Chatbot

“…Terdapat beberapa model chatbot yang dapat diimplementasikan, dua di antaranya adalah button-based chatbot dan natural language processing chatbot. Chatbot yang memproses basis input dalam bentuk tombol adalah chatbot dengan model button-based [9]. Model chatbot yang mampu memproses input dalam bentuk phrases atau kalimat bahasa manusia adalah model natural language processing chatbot [18].…”

“…Perancangan chatbot yang dilakukan oleh Ahmad et al, menerapkan model button-based chatbot pada penelitiannya. Penggunaan model button-based memudahkan pengguna dalam mengakses informasi dan meminimlaisir terjadinya kesalahan input [9]. Penelitian oleh V. Naidoo menggunakan model button-based chatbot dengan studi kasus bidang perhotelan.…”