Categories
Media Generation

Carleton Course Assistant

Date: Spring 2021

Partner(s): Theresa Chen, Alistair Pattison, and Logan Peters

View the code associated with this project

This project is a chatbot that prompts users to enter their favorite courses and then uses natural language processing techniques to compare course descriptions and suggest the most similar courses to the user’s favorites. It also uses NLP techniques to create an imaginary class to recommend. This project required my teammates and I to combine many NLP techniques into a cohesive project. It is the second chatbot we created as a term, and in it we attempted to create a different sort of interactive experience.

Categories
Media Generation

Murder at the Sherwood Forest Fair

Date: Spring 2021

Partner(s): Theresa Chen, Alistair Pattison, and Logan Peters

View the code associated with this project

For this project, my teammates and I created a dialogue system that prompts the user, who is acting as a detective, to solve a murder. It uses several natural language processing techniques to construct a unique mystery each time it is run and allow the user to conduct interviews as they search for the murderer. The bot was intended to bring users into a unique world and storyline each time they interacted with it, creating a novel experience every time. Throughout this project my teammates and I learned to write a complicated chatbot that integrated multiple NLP techniques and how to create a flexible narrative within the constraints of a computational setting.

Categories
Media Generation

Generating Poetry Pamphlets with Procedural Rules

Date: Fall 2020

View the code associated with this project – and a sample output

In this project, I used a grammar parser and rule system designed by Professor James Ryan to create a pamphlet of sonnets and free verse poems. I wrote rules for creating technically correct sonnets and less structured free verse poems from a corpus of words separated by part of speech and number of syllables. The PDF generator also uses a logic system to include an image with each poem. This is the only large project I have created that uses computational techniques to generate creative textual output. It combined words and phrases in unique ways that I would never have myself because of linguistic conventions that I have been taught since birth. However, once the program generated these unusual word combinations human creativity can find meaning in them.

Categories
Media Generation

Generating Dance Routines with a Procedural Grammar

Date: Fall 2020

View the code associated with this project – and a routine performed

For this project, I used a grammar system written by Professor James Ryan to procedurally generate International Rumba dance routines. The system ensures that the sequences of movements produced will be danceable and also determines what song the routine should be danced to and what the atmosphere of the performance should be. I performed an output of the system over Zoom with my dance partner, Stuart Kraabel. This program encouraged me to leave my comfort zone because it produced routines I would not have choreographed myself and dictated unusual styling choices that forced me to reconsider performance norms that were ingrained in me.

css.php