Yusuke Mori (AI Researcher, Square Enix Co, Ltd.)
Pass Type:
All Access Pass, Summits Pass
Topic:
Programming
Format:
Lecture
Vault Recording: TBD
Audience Level: All
In the 1980s, adventure games with free text input, such as Mystery House (On-Line Systems, 1980), was popular. However, due to the technical limitations of the time, the mainstream of adventure games was replaced by those where the user selects explicit commands, which aimed at such free text input by users.
Yusuke Mori, AI Researcher at Square Enix Co, Ltd., will go into how the team has developed a technology for a new generation of adventure games that can flexibly respond to arbitrary input from the user—by utilizing the recent advanced natural language processing (NLP) technology. Using this tech, they have revived one of Japan's most famous adventure games, The Portopia Serial Murder Case (Enix, 1983), in a new form.
Takeaway
Attendees will learn how natural language processing can be applied to adventure games. They implemented two systems using NLP: a text input acceptance system that allows for ambiguity and a chit-chat dialogue system. These allow for a new form of adventure games where users feel a high degree of freedom.
Intended Audience
This is for anyone curious about natural language processing (NLP), especially its application for adventure games. Anyone interested to know how to apply machine learning-based NLP to production and tweak it to the needs of the scenario.