The Tree of Babel

Project Idea


The project goal is understanding and generating interactive narrative structures using Machine Learning models.

This project is inspired by the idea of Standard Patterns in Choice-Based Games, that analyzing interactive narrative structures of games as graph and design pattern.

The dataset is customized for ML libraries and scraped from TransverseReadingGallery.

Basic concept:
    1. Convert graphs to feature matrix of vector embeddings*.
    2. Categorize graphs by clustering algorithms.
    3. Generate new graphs from categories using autoencoder.
 
Used libraries and ML models:

*An embedding is a relatively low-dimensional space into which you can translate high-dimensional vectors. Embeddings make it easier to do machine learning on large inputs like sparse vectors representing words. Ideally, an embedding captures some of the semantics of the input by placing semantically similar inputs close together in the embedding space. An embedding can be learned and reused across models. (Google Developers Machine Learning Course)
 
ITP Thesis 2019, Youjin Chung