Physical Sciences and Mathematics Commons^™

Machine Learning And Natural Language Processing For Crossword Puzzles, Finn Brennan

Senior Projects Spring 2024

Senior Project submitted to The Division of Science, Mathematics and Computing of Bard College.

Dumonym: Crafting And Assessing Lexical Simplification, From Algorithms To Models, Jeremias Brea De Los Angeles

Senior Projects Spring 2024

Lexical Simplification is the process of replacing complex words with simpler alternatives in a given text. This project aims to use different approaches in the field of Natural Language Processing to create a series of lexical Simplification models. The framework of lexical simplifiers will also be explored and researched, to give more insight of the mechanisms and approaches used to achieve successful text simplification. I will develop a pipeline of steps, based on my research, with the aim to create a framework for a functional lexical simplification model. I will develop a series of distinct lexical simplification models based on …

Simulating The Spread Of Information Among Non-Playable Characters In Video Games, Summer Ashley Colon

Senior Projects Spring 2024

This project attempts to improve Non-Player Character (NPC) realism in video games by using an Agent-Based Model (ABM) to automate their communication and interactions. To evaluate how information spreads through interacting NPCs in video games, this simulation assumes that the main purpose of NPC societies is to provide a context in which NPCs interact with each other to spread information about player-generated events that change the game state. The goal is to successfully identify possible programming methods to automate NPC behavior and create a more realistic gaming experience for players. The simulation incorporates three archetypal types of characters that are …

¿Quién Soy Yo? [Who Am I?]: Exploring Identity Through Analyzing Afro-Cuban Poetry And Creative Coding In A Post-Secondary Spanish Literature Classroom, F. Megumi Kivuva

Senior Projects Spring 2022

With efforts to broaden participation in computing by integrating CS education into humanities and developing more critical pedagogy, this research focuses on teaching computing in a post-secondary Spanish literature class through analyzing Afro-Cuban poetry. Its goal was to evaluate how participants may use Twine to reflect on Afro-Cuban poetry and their own identities. A group of 5 participants, one professor, and five students, learned how to use Twine to create interactive narratives reflecting on “El apellido,” a poem by Afro-Cuban poet Nicolás Guillén. Through analyzing researcher notes, participants’ projects, post-workshop surveys, and interviews, the research revealed that students were able …

Predicting League Of Legends Ranked Games Outcome, Ngoc Linh Chi Nguyen

Senior Projects Spring 2022

League of Legends (LoL) is the one of most popular multiplayer online battle arena (MOBA) games in the world. For LoL, the most competitive way to evaluate a player’s skill level, below the professional Esports level, is competitive ranked games. These ranked games utilize a matchmaking system based on the player’s ranks to form a fair team for each game. However, a rank game's outcome cannot necessarily be predicted using just players’ ranks, there are a significant number of different variables impacting a rank game depending on how well each team plays. In this paper, I propose a method to …

Relevance-Tcav: Explaining Deep Neural Nets In Human Concepts, Henning Fischel

Senior Projects Spring 2021

Neural Networks, a form of machine learning, are used in increasingly important roles in the modern world. They are being used in self-driving cars and medical diagnoses. However, they are “Black Boxes”: they cannot be easily interpreted by humans. This project combines two methods of explaining a neural network’s decisions in an attempt to improve their accuracy. This new method, relevance-based testing with concept activation vectors (R-TCAV), yields promising results on two small experiments but is less precise than the previous TCAV method.

From Rural To Urban: Understanding America Through The Census, R. Owen Hartman

Senior Projects Spring 2020

The first ever Senior Project to combine Historical Studies and Computer Science. Examining United States Census data from 1870 to 1930, "From Rural to Urban" explores the changes in American life that occurred during the industrial revolution at the turn of the 20th century.

Relating Sentiment Expressed By Financial Twitter Accounts And Financial Index Price Movement, Jacob Edelstein Lester

Senior Projects Spring 2020

We apply sentiment analysis to correlate price movement for two financial indices with sentiment expressed on Twitter by a select group of 93 influential financial users. We gathered close prices for the VIX and SPX indices for one month from March through April 2020 during the early stage of the COVID-19 pandemic in the U.S. as schools and businesses shut down. Tweets were also gathered during this period, although there is a large gap in collected tweets of about two weeks. We examine correlations based on five temporal resolutions from 60 minutes to 1440, which is equivalently one day.. We …

An Study Of Sentiment Analysis Methods For Mandarin Chinese, Thalia Celeste Chai-Zhang

Senior Projects Spring 2020

Sentiment analysis is the study of automated methods of emotion detection in language and speech. It is an area of much active study, and has seen significant progress in improving accuracy over recent years. Many studies however are conducted on datasets of English and other Indo-European languages. We apply well studied methods for sentiment analysis to a Mandarin Chinese data set and compare the accuracies achieved against eachother and other studies. Our findings indicate that methods for improving sentiment analysis accuracy for English may not be as applicable to Mandarin Chinese.

The Spiral Model For Generative Harmony, Jackson Guy Spargur

Senior Projects Spring 2020

Generative music is a broad and well-explored field, in which researchers have attempted various approaches at creating algorithmic models for the creation of music. Researchers may attempt to model the composition of melody, or of musical phrase structure, or, as is the focus of this paper, the harmonization of multiple voices. I use as the core of my model Elaine Chew’s “Spiral Array”, outlined in her 2000 thesis “Towards a Mathematical Model Of Tonality”. Chew’s applications for this model were all analytical, gaining insights about human-composed pieces of music by running them through her model. My project is comprised of …

Testing And Improving An Optimization-Based Digital Colorblindness Corrective Filter, Zachary Kenneth Mcintyre

Senior Projects Fall 2020

Computers often communicate essential information via color which is lost to colorblind users. In order to address this information loss, designers and computer scientists have created a variety of different correction methods to improve computer accessibility. One such method was created by Luke Jefferson and Richard Harvey in their 2006 paper, “Accommodating Color Blind Computer Users” which consists of a difference histogram, differences of key colors, optimization and interpolation to adjust images for specific types of congenital colorblindness. I have recreated their algorithm as well as their original test images. I then conducted extensive tests on challenging images to examine …

A Computational Method For The Image Segmentation Of Pigmented Skin Lesions, Kaila M. Piscitelli

Senior Projects Spring 2020

Senior Project submitted to The Division of Science, Mathematics and Computing of Bard College.

Simulating An Immune Response With A Combined Agent-Based Model Of A Triple-Negative Breast Cancer Tumor And Vascular Network, Michael J. Ventoso

Senior Projects Fall 2019

Cytotoxic T-cells (CTLs) are one mechanism the immune system employs to eliminate cancer cells. In this study, I expand upon a previous 3-dimmensional agent-based model of triple-negative breast cancer to include a therapy simulating an immune response in the form of a CTL insertion into the tumor. The model consists of the tumor, comprised of progenitor cells, cancer stem cells, and tumor-associated macrophages, as well as an expanding vascular network. I investigate the effects of inserting different amounts of CTLs into the space, and their effect on tumor size in the short and longer terms. The results show that while …

Programming Proletarian Literature: Kobayashi Takiji’S "Kani Kôsen" And Gaming As Reading, Jacob Philip Fisher

Senior Projects Spring 2019

Abstract

This project translates a novel, Kobayashi Takiji’s, Kani Kôsen (The Crab Cannery Ship, 1929) into a video game. As a joint project between Computer Science and Japanese, its focus is to develop a game for the original Game Boy (1989) narratively based on a work of Japanese proletarian literature. Specific tools used in development were the Game Boy emulator: bgb, the Game Boy Developers Kit (gbdk), the Game Boy CPU manual, as well as a foundation in the C programming language, and some lower level systems experience. Being based on a novel, the play style utilizes text …

Is There Joy Beyond The Joystick?: Immersive Potential Of Brain-Computer Interfaces, Elias Solomon Posen

Senior Projects Spring 2018

Immersion, the state of being fully engaged in one's current operation, is a descriptor commonly used to appraise user experience in computer games and software applications. As the use of brain-computer interfaces (BCIs) begins to expand into the consumer sphere, questions arise concerning the ability of BCIs to modulate user immersion. This study employed a computer game to examine the effect of a consumer-grade BCI (the Emotiv EPOC) on immersion. In doing so, this study also explored the relationship between BCI usability and immersion levels. An experiment with twenty-seven participants showed that users were significantly more immersed when controlling the …

The Poem Will Resemble You: A Human-Computer Collaboration, Zoe M. Morgan-Weinman

Senior Projects Spring 2018

Senior Project submitted to The Division of Languages and Literature of Bard College.

Algorithmic Music Composition And Accompaniment Using Neural Networks, Daniel Wilton Risdon

Senior Projects Spring 2016

The goal of this project was to use neural networks as a tool for live music performance. Specifically, the intention was to adapt a preexisting neural network code library to work in Max, a visual programming language commonly used to create instruments and effects for electronic music and audio processing. This was done using ConvNetJS, a JavaScript library created by Andrej Karpathy.

Several neural network models were trained using a range of different training data, including music from various genres. The resulting neural network-based instruments were used to play brief pieces of music, which they used as input to create …

Evaluation Of Interfaces For Senior Citizens: Bigscreenlive, Eldy And Pointerware, Bella Manoim

Senior Projects Spring 2011

Senior Project submitted to The Division of Science, Mathematics, and Computing of Bard College.