Various Works and Essays

Risks and Remedies for Black Box Artificial Intelligence Aug. 2020

There is no denying that artificial intelligence (AI) and machine learning technologies will shape our future. Advanced AI systems can solve the most complex business problems and create significant economic benefits for organizations. But coupling such systems with poor interpretability - also referred to as explainability - can create a recipe for disaster.

In this article, we outline the issues that can be addressed through implementing interpretable AI, the challenges to AI interpretability, and how the C3 AI Suite supports AI interpretability.

Manifest Destiny: Are Progressive Web Application Destined to Expand Across Our Online Experiences? Jan. 2018

This project was completed during my Summer ‘17 internship at eBay, Inc. We explored the incorporation of Progressive Web Application features inside a design experiment enviroment and how this changes the strategies of designing a better user experience. Three angles were explored, the implication to web app design practices, the user experience, and design research.

Emotional Expressions in Online Settings: How I Spun Down a Data Mining Hole of Emotion Mar. 2017

Emotional expression through text is a complex process. It can be expressed very explicitly through descriptive wording, “I feel happy”, or be expressed through an implicit process “We are having cake!”. This work was a class exploration budding from my work on explicit emotional expressing into examining a possible measurement of implicit emotional expression.

Exploration Study of Retweet Propensity Dec. 2015

Causal explanations when working with pure observational data is a challange many Mathematical, Statiscal, and Computer Science researchers are focusing on improving. This work was a class project focusing on incorporating some known techniques of causal modeling toward an online media research question, what causes individuals to retweet on Twitter?

Are Emotions Detected Differently within First-Person Narratives? Mar. 2015

Much of human expression is done through narratives. While much more on emotion detection within text has been through the lens of simplistic natural language processing techniques, very few have explore the overall structure of the text effects on emotional expression. This work was a class project examining how taking into the account Narrative vs. Non-narrative hierarchical information may be more predictive for emotion classification.

Link predicition is a common problem for research on the Web. These links could be between two websites or two individuals. This work was a class project incorporating the unqiue relational feature set commonly used in collective classification frameworks, towards a task of predicting whether or not two online communities would hyperlink with each other.

Ranking Politicos. Mar. 2014

Ranking Politics is a Brailizian website going through the effort of information Brazilian voters about the quality of candidates running for office. A student group I was apart of for a class took the effort of designing a mobile interface for this website as the site itself was not mobile friendly (as of 2014). We incorporated many design research objectives such as Competitive Analysis, Personas, and Use Case research. We design low and high level prototypes and tested these with a small sample of Brazilian voters.

ELO Outside of the Competitive Gaming Realm. Mar. 2014

ELO was originally used as a chess ranking algorithm to quantatively measure skill. However the measure has found itself inside many more applications since. I used this algorithm to explore its ability in the YouTube slam competetion.

Fun Projects

ELO ability to predict baseball game outcomes. Jan. 2016

Further explored ELO inside the realm of baseball. Used ELO as a prediction model to judge its accurancy on predictive baseball game outcomes. To little surprise this wasn’t found to beat the odds in Vegas.

Procedural Generation of 2D Caves Sep. 2014

Game development is a hobby interest of mine and I have many friends who work in games using procedural generation. I wanted to explore some basic generation techniques so I found one that used a common mathematical simulation (Conway’s Game of Life) to generate caves and implemented this within a web based enviroment.