91¿´Æ¬

Spring 2024

Fall 2023

 

Fall 2023 Projects

The students had to define their own objectives, obtain or collect the data, find patterns in the data through descriptive analytics techniques and create predictive models using machine learning approaches.

1. Understanding safety and performance in f1 racing - Formula One racing is a high-octane, high-intensity race tour that pits racers against each other with the most advanced and world’s highest-performing cars on some of the most complex tracks. In Formula One, one mistake could prove costly for a place on the podium and the driver's safety. With a select group of drivers and constructors dominating the sport and the associated risks of high-speed accidents, this project seeks to understand why accidents occur to help keep racers safe while balancing in-race strategies to help drivers reach the podium. 

2. Hanes Size Mix Optimization - Our project aims to identify the most prominent characteristics of the data involved in determining size mix. Our project also helps determine the best level to understand optimal size mixes for Hanes Brand products. We applied clustering techniques to gain insights into the profile of the optimal product mix.

3. Bonnie Plants Analytics - Our team elected to work with Bonnie Plants to improve the Sell-Through. Bonnie Plants is one of the country's largest distributors of grown plants. Our data is based on 6 stations that service 6 different regions across the country. Stations house massive greenhouse gas sites, and units are shipped from station to stores to be sold. Common customers are outdoor retail companies, like Home Depot, Lowes, and Tractor Supply. Our project analyzes the number of plants sold relative to delivered, based on location, item category, and weather. We are looking at weather patterns for fiscal months to observe how various factors affect the seasonality of sell-through.

4. Smart Grid Energy Savers - Our project centers around using smart grid data to develop a more intelligent and efficient use of our daily resources. We are taking incredibly thorough and comprehensive data that outlines all the different energy uses throughout the homes to highlight areas of inefficiency to address. Many factors can play into this, and we are working to address all environmental factors that play a significant role. Moving forward, we hope that our work throughout this project can improve the use of smart grids as they grow to other cities across the nation and the globe.

5. Bengals Defensive Analysis - The Cincinnati Bengals’ defense had not been able to obtain consistent success within their division and in postseason games. This, coupled with the rapid development of high-powered offenses, requires a tremendous effort to achieve the goals of the Bengals. Our group aimed to provide the Bengals’ defensive coordinator with trends and visualizations to improve the defense's performance. We intended to show the coach actionable insights within the data that wouldn’t otherwise be seen without in-depth analysis. Some of our ideas included modeling the Bengals’ win-down percentage, identifying where explosive plays occur, creating heatmaps to highlight defensive weaknesses, and evaluating how the defense fares against top QBs and top WRs in the league. We hope that uncovering certain tendencies about their previous games can improve play calling, defensive reads, and the defense's overall performance.

6. Bussin’ it Down: Tiger Transit Analytics - Our project is focused on descriptive analysis and anomaly detection within Auburn’s Tiger Transit system. Our analysis finds patterns, trends, and anomalies within Tiger Transit’s operations using engine time, speed, boardings, and more variables. Using this data, we pinpoint shortcomings in Tiger Transit that could be why patrons are not satisfied. Additionally, by locating anomalies within the data, the system will be better aware of these anomalies and better equipped against further issues.

 

Spring 2023

Spring 2023 Projects

The students had to define their own objectives, obtain or collect the data, find patterns in the data through descriptive analytics techniques and create predictive models using machine learning approaches.

1.  Bonnie Plants Sales Prediction – Bonnies Plants has shared their sales data from The Home Depot and Lowes. In addition, the total foot traffic and demographic information about the location where stores are located is available. The objective is to identify the factors affecting sales using descriptive and predictive analytics models.

2. Insomnia condition among college students – The objective is to understand the relationship between mental disorders and insomnia among college students. The students collected data from an online health forum.

3. Olympians performance prediction – The purpose is to understand the relationship between the performance of the players based on their country of origin. 

4. The Effect of Weather on Baseball Metrics and Pitch Success – Supported by AU Baseball, the students aim to identify the important sports-related and external factors that can predict pitch success.

5. Evaluating the performance of EBSCO customers – Sponsored by EBSCO, students are analyzing customer transaction data to find the differences between low and high-performing customers.

6. Predicting COVID-19 breakthrough infections in Alabama – Sponsored by the Alabama Department of Public Health, students are building models to compute the likelihood of a patient getting a breakthrough infection.

7. Will the bubble pop? - The housing market is in constant flux with houses being listed and sold every single day. This project examines the current housing market to determine if the conditions are similar with the housing market before the 2008 crash. Comparing these data points will allow us to forecast the potential of another crash in the current housing market.

Fall 2022

Fall 2022 Projects

There are seven teams in this class working on different data-driven projects. The students had to define their own objectives, obtain or collect the data, find patterns in the data through descriptive analytics techniques and create predictive models using machine learning approaches. A brief description of each dataset is given below.

1. Bonnie Plants Sales Prediction – Bonnies Plants has shared their two years of sales data from Home Depot and Lowes with the BASY department. In addition, the total foot traffic and demographics information about the location where stores are located is available. The objective is to identify the factors affecting the sales using descriptive and predictive analytics models.
2. Healthcare analytics – Focusing on the condition of epilepsy, students have obtained the medical conditions of epilepsy patients from an online health website. The objective is to find the health disparities among epilepsy patients by gender and age.
3. Sleep Study - We got access to the sleep-related data. The dataset contains sequential information about sleep stages recorded overnight from more than 2,600 patients with heart related disorders. The objective is to create classification models to diagnose the heart diseases using the sleep patterns. (Acknowledgements - Dr. Rupesh Agrawal, Northern Kentucky University)
4. Sports Analytics – We had access to NCAA basketball games data. In this unique dataset, the detail of each game is available including the different plays every second in the game. The objective is to understand the trends in the events over years across all teams. (Acknowledgements - Dr. David Paradice)
5. Game days and Crime rate – The objective is to find a relationship between the crime rate and the game days. The students have collected data about the crimes and game schedules from multiple sources.
6. American Hospital Association – The objective is to find the relationship between the use of technology by hospitals across all the regions in the US and the satisfaction of their employees. The BASY department has access to the unique AHA datasets for 13 years.

Spring 2022

View Spring 2022 Presentations