Building a mobile app for joke-generating with machine learning model

• Main Author: Toh, Rui Sheng
• Other Authors: Jong Ching Chuen
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Computer science and engineering::Software::Software engineering; Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/149713
• Institution: Nanyang Technological University

Several studies have utilised machine learning language models, such as GPT-2, to learn and generate new tweets; however, not many of them attempt to explore the usage of language learning in the entertainment field through a more accessible medium. Here we designed an Android mobile application which showcases the jokes obtained by training the GPT-2 model. The model is trained by feeding Tweets, from Twitter accounts of well-known comedians, into it. Outputs of the model are curated through both software and manual approaches, in order to ensure the end results possess proper grammatical structures and high qualities. The results are stored in MongoDB Atlas, a cloud database. Additionally, MongoDB Realm is used to synchronise data between the database and the Android application.