Uzodinma Ikwuakor

Intro

My name is Uzodinma Ikwuakor (I go by 'Uzi' or 'Uzo, for short), and I'm a Denver-based data scientist. I have a B.S. in Mathematics, with a double concentration in Applied Mathematics and Probability and Statistics, from the University of Colorado in Denver. I'm also a graduate of the Thinkful Data Science program, a 1-on-1 mentor- guided, data science curriculum covering machine learning concepts, statistical inference and industry best- practices. I've built various supervised and unsupervised machine learning models. Check out my work.

Work

Language Toxicity Classification

Many online content providers depend upon an engaged and active online community, oftentimes for the creation of content itself, and need ways to ensure their site is an attractive destination for repeat web traffic. One of the ways they often achieve this is through an engaging online forum. However, when abusive community members are allowed to make a forum feel toxic, many of the other members can be turned off by this behavior and begin to limit, if not completely eliminate, their level of engagement in the community. The following project uses Natural Language Processing (NLP) techniques to explore how to detect toxic language using text data from Wikipedia forums.
Language Toxicity Classification

Boston Crime

In the fallout of recent mass shootings and what seem like daily news reports of gun related tragedies, gun violence, and how best to stop or reduce it, has been a lightning rod of American political discourse. In this project, I explore the city of Boston, as a microcosm of a greater national concern, and the areas inside of the city where gun crime is most prevalent; emphasis on shootings and gun-related homicides in particular. I examine which parts of Boston, relative to the downtown Boston area, present the greatest risks of potential gun violence and towards which areas, or whether or not, gun activity is migrating within Boston. I also examine which time of the day, days of the week, and months of the year gun violence is most likely to occur.
Boston Crime

Predicting Medical Appointment No-Shows

Patients failing to make their appointments can be quite problematic for medical professionals. From wasted personnel on staff to other patients, who were otherwise available, being scheduled for a later date, missed appointments can be quite disruptive for medical offices. In fact, missed appointments might also be disruptive to the absent patient's own good health! The question is: how do medical professionals combat missed appointments? Is there a way to predict ahead of time who is most likely to miss their appointments? The following project explores a dataset from Brazil to determine to what extent we can predict whether patients will show up to their medical appointments as scheduled.
Predicting Medical Appointment No-Shows

Life Expectancy EDA

Finding the factors that affect the life expectancy, specifically, which factors increase the expected life in the countries and which factors decrease it. This data was collected from the websites of the World Health Organisation (WHO) and World Bank (WB) as well as coutry population data I pulled from photius.com for reference purposes.
Life Expectancy EDA

About

They say that a data scientist should be a good storyteller. Well, here’s mine.

Long before I ever understood the usefulness of math, I loved it. The same thrill and sense of accomplishment you might get from completing a crossword puzzle or solving a brain teaser, I would get from solving math problems. They were like puzzles to me, or maybe even art. When the other kids would ask, “Teacher, when will I ever use this stuff?”, I would think, “Can’t we just enjoy the ride?”

At some point, though, I probably should have asked the same question because once high school graduation was right around the corner, math was about to become something I used to enjoy. A pastime. It wasn’t until my second stint of college that a friend in my math department introduced me to machine learning and I finally saw a path to merging my technical brain and my creative brain. After graduation, and after I'd saved up enough money, I found myself in Thinkful's data science program, finally making sense of all the things my friend would talk about. I've spent the past several months deep- diving into various supervised and unsupervised machine learning algorithms, and now my coding, analytics and even my math skills, have taken off to a whole 'nother level! This journey of mine has been a long time coming, and once I land that dream job, the rest will be history.

In summary, aside from long walks on the beach and candlelit dinners, I enjoy understanding things to the point where they become trivial, making sense of things that don’t make sense at first glance and then testing what I 'know' by exposing it to scrutiny (in case you’ve ever wondered what type of person debates others in Youtube comment sections). I guess what I wish to convey, perhaps even more than general aptitude, is my rabbit-hole approach to, well, everything!

Contact

• fa-linkedin
• GitHub
• Gmail

Elements

i = 0;

while (!deck.isInOrder()) {
    print 'Iteration ' + i;
    deck.shuffle();
    i++;
}

print 'It took ' + i + ' iterations to sort the deck.';