About

I am Torin Stott, a recent Computer Science graduate at Georgia Southern University. Outside of schooling most of my time is spent coding and tinkering in various linux distributions. I am very interested in application security. I have participated in many CTF style programs such as tryhackme and hacker101.

TryHackMe Profile

Hacker101 Ctf

Capstone Project

Pictured below is my Capstone SQL Scaffolding project. This project was designed in conjunction with Tyler Porcher, and Charles Haislip. Currently the Github repo for this project remains private, but I wanted to document some screenshots displaying our progress.

Problem statement

Our application was designed to help make learning SQL easier for students. We worked closely with Dr.Otto Borchert to provide an interactive SQL learning experience.

Site Layout

Home

Base landing page showing Contributors.

Question

Page for students to enter SQL answers. This page is designed to be dropped into an IFrame on the TopHat website. TopHat is an interactive book website used by professors. When students press execute their code is run on SQLite and if they pass all test cases they will be given a code word they can enter into TopHat when they are finished.

Research

Page showing student progress. For use by Professors. Connects student data using Firebase from Amazon Web Services.

Database View

Place for professor to upload new Databases.

ML Project

View Code

Problem Statement

This project aimed to make predictions using a dataset derived from echocardiogram tests. It was designed to classify if patients will survive for at least one year after a heart attack.

Data Preparations

Initial dataset was loaded from UC Irvine which may be downloaded here.

Shown below is a snapshot of initial data

Some columns such as name and group were were stripped from the data. Note there are also some empty values "?" which also need to be removed to properly train models.

Column Descriptions

  • Age-at-heart-attack: age in years when patient experienced heart attack
  • Fractional-shortening: measures contractility around heart
  • Wall-motion-score: measures how parts of left ventricle move
  • LVDD: measure of the heart at end diastole

    • These features were fed into various training models to predict if the patient would be alive after one year

    KNN

    Perceptron

    SVM

    Conclusion

    In this project I learned the importance of utilizing model selection techniques. It is worth exploring using these models with a larger dataset to get more accurate classifications. SVM performed most effectively as a classifier because data was not linearly separable.

    Personal Project

    View Code Live Demo

    Problem Summary

    This small project uses the Github API to search for github users quickly. Information such as when the user created their profile, their profile picture and number of repositories are shown.

    *Note* Requests are limited when using this API. If search fails retry after a few seconds.
    Git Documentation

    Contact Form

     

    Elements

    i = 0;

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

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