Capstone – Project
Machine Learning Technology Evaluation & Testing
A Collaboration Between
Students in the MSBA 790 will deliver the following:
- Evaluate the top five to seven Machine Learning platforms for our Industrial Internet of Things (IIOT) platform. Our IIOT solution is a big data cloud application that captures 31,536,000 rows of data containing over 1.5B data points every year for each device deployed. Students must present the pros and cons of using different ML tools and frameworks to process Big Data and extract key insights.
- An ML problem that is of interest to us is the impact of outliers on energy costs and long-term implications on energy consumption. We will provide students with a large data set to solve this problem using ML algorithms. Students must review and identify the appropriate ML algorithms for this problem. These algorithms must be executed (run) on the above ML platforms. An evaluation of the performance of the different ML platforms using this data set must be included in the presentation and report.
- Students must also divide the data set that we provide into testing and training data and provide a detailed explanation as to how this was done.
- The above information should be included both in your presentation and a detailed written report. This includes the following: a) Highlight the top choice ML platform given your understanding of Eco Enterprise's goals. b)Which of the ML algorithms presents the best framework for analyzing the data? c) What insights were you able to generate based on the sample data? d) How might these insights further improve as you add more data? What are the cost and benefit implications of ML, in general?
Project Topics
Data Management
Product Design & Development
Quality Control
Company Information
| Company | Eco-Enterprise |
| HQ | New Jersey |
| Revenue | Unlisted |
| Employees | 1-5 |
| Stage | Pre-Revenue Startup |
| Hiring Potential | Follow-on Projects, Formal Internship, Entry Level Full-Time, Upper Level Full-Time |
| Website | http://Eco-Enterprise.com |
Company Overview
eco-enterprise helps businesses to become more Efficient, Economic and Environmentally responsible via unique financial solutions and energy-saving technologies.
Experiential Learning Program Details
| School | University of South Carolina – Upstate |
| Engagement Format | - |
| Course | Summer Capstone in Business Analytics |
| Level |
|
| Students Enrolled | 12 - 4 students / project |
| Meeting Day & Time | Online |
| Student Time Commitment | 4-7 Hours Per Week |
| Company Time Commitment | 1 Hour |
| Duration | 8.29 Weeks |
Program Timeline
Key Project Milestones
-
June 18, 2021 - Define ML Problem to be Solved
What is the ML Problem to be Solved given the data set and product goals of the Eco Enterprise IIOT platform?
Suggested Deliverable:
Learn more about HVAC system and define the ML problem to be solved
-
July 2, 2021 - Initial Evaluation of ML Tools
- What are the top ML tools in the market?
- How are they used? By whom?
- How might Eco Enterprise be able to leverage different existing ML technology to glean relevant insights?
Suggested Deliverable:
Present top ML platform for consideration, how they’re used by similar companies for similar purposes, and initial pros/cons comparison of each
Ensure you have access to sample data for next phase
-
August 4, 2021 - Recommend Top Choice Machine Language Technology
Run the sample data through each of the machine learning technologies with the goal of identifying outliers and anomalies within the data set. Increase the quality of your pros and cons comparison analysis based on your experience running the data through the tool and developing insights.
Suggested Deliverable:
Produce a final report highlighting the top choice machine learning platform given your understanding of Eco Enterprise’s Machine Learning Problem to be solved, the sample data, and the outcomes of your pilot using the sample data.
Which of the ML algorithmes presents the best framework for analyzing the data?
What insights were you able to generate based on the sample data?
How might this further improve as you add further data?
What are the cost and benefit implications of choosing your selected tool? Why does that present a benefit over other choices?
Project Resources
There are no resources currently available
Industry Mentors
Academic Mentors
