An extensible AI tool implementation to 2D game editor

Artificial intelligence has a more and more important role to play in our life nowadays. It has been applied in many fields, including virtual personal assistants, smart cars and fraud detection. It has an even wider scope of applications in the years to come. A team under Prof Ong Yew Soon in Nanya...

Full description

Saved in:
Bibliographic Details
Main Author: Youwei, Du
Other Authors: Ong Yew Soon
Format: Final Year Project
Language:English
Published: 2017
Subjects:
Online Access:http://hdl.handle.net/10356/70165
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Artificial intelligence has a more and more important role to play in our life nowadays. It has been applied in many fields, including virtual personal assistants, smart cars and fraud detection. It has an even wider scope of applications in the years to come. A team under Prof Ong Yew Soon in Nanyang Technological University is currently carrying on a project titled as “2DGameEditor with Nanyang Technological University”, which focuses on mainly key issues like AI advancement in Gaming Industry. In this project, an extensible AI tool based on decision tree learning algorithms is developed, with the objectives to dynamically generate game rules for the 2D game editor. As a result, users of the 2D game editor do not need to struggle with endless trials and errors to find the set of game rules that fit well with their intended storylines. To be more specifically, this project will conduct an in-depth study and a thorough analysis of different decision tree learning algorithms and their main strategies. With the consideration of the context of the 2D game editor and an understanding of the benefits and drawbacks of different algorithms, a suitable decision tree learning algorithm will be selected and implemented. After that, the AI tool developed will be tested using different databases from the UCI Machine Learning Repository. Tests which target at different features of the AI tool will also be performed using suitable databases. Moreover, some databases are used as benchmarks to measure the performance of the AI tool.