Capstones

Capstone projects at Vanderbilt Data Science serve as the pinnacle of our students’ academic journey, offering them the chance to apply their skills in real-world settings. By partnering with industry leaders, we create opportunities for students to tackle complex, practical challenges. These collaborations not only bridge the gap between classroom learning and professional application but also help shape the future of data science, one project at a time.

--

What is a Capstone Project?

A Capstone Project is the culmination of the Data Science journey at Vanderbilt. It provides an opportunity to apply advanced skills in a real-world setting, addressing challenges presented by industry partners. This experience bridges the gap between academic learning and professional application, allowing participants to solve complex data problems and deliver impactful results.

Capstone Structure and Timeline

Overview: A Capstone Project follows a structured timeline, ensuring participants can fully apply their data science skills while meeting the expectations of industry partners. The process typically spans a semester and includes key phases that guide participants from project selection to final deliverables.


Capstones can happen either in the Fall Semester or Spring Semester.

Key Phases

  • Project Selection

     ~1 month prior to semester
    Participants begin by choosing a project that aligns with their interests and career goals. The primary task in this phase is clearly defining the problem statement and identifying and securing the necessary data and/or tools to solve the problem. 

  • Execution

    During the semester (Fall or Spring)
    The core of the Capstone Project involves executing the planned analysis or solution development. This phase requires regular check-ins with advisors and industry partners to ensure progress and address any challenges.

  • Final Deliverables

    End of semester/final exam week
    The project culminates with the submission of final deliverables, which include at a minimum a comprehensive report, a GitHub repository, and a presentation summarizing the project’s problem statement and outcomes

Where do Capstone projects come from?

Capstone Projects can arise from many different sources. Our primary goal is for you to work on something that is energizing and exciting, such that you look forward to the work and want to talk about it with others. 

  1. Personal passion. Some students have specific personal interests that they want to pursue on their own. As long as you can define a realistic business/organizational problem you are solving and have the necessary data to solve it, this can be a very fulfilling project. 
  2. Extension of current work. Many students maintain part-time jobs or find RA positions with faculty and want to leverage that work to become their Capstone. As long as there is a defined, discrete problem to be solved, this allows your capstone to enhance and support what you are already doing. 
  3. Something new. Many students want to do something different. Every semester, DS faculty pursue possible capstone projects with industry partners and researchers at Vanderbilt University or Vanderbilt Medical Center. Students apply for these and get selected by those partners running the project(s). 

Types of Capstone Projects

Capstone Projects come in various forms, allowing participants to engage with different challenges and methodologies. Each project is designed to apply data science skills to real-world problems, offering a unique learning experience. Here are some of the types of capstone projects students might encounter:

1. Individual Projects:

  • Overview: These projects are typically more focused, allowing a single participant to dive deep into a specific problem. Individual projects often involve developing a specialized tool, conducting in-depth data analysis, or creating a custom solution for an industry partner. 
  • Example: A student might work on developing a machine learning model to predict customer churn for a retail company, with the aim of improving retention strategies.

2. Group Projects:

  • Overview: Group projects bring together a team of students to tackle more complex, multifaceted problems. Collaboration is key, with students leveraging each other’s strengths to deliver a comprehensive solution.
  • Example: A team might work on designing a data pipeline for a healthcare provider, integrating multiple data sources to improve patient outcomes through predictive analytics.

3. Research-Oriented Projects:

  • Overview: These projects focus on advancing the field of data science itself, often involving exploratory research, developing new algorithms, or testing novel approaches to existing problems.
  • Example: Students might conduct research on improving the efficiency of deep learning models, publishing their findings in academic journals or presenting at conferences.

4. Tool and Infrastructure Development:

  • Overview: Some projects are centered around building or enhancing tools, platforms, or infrastructure that supports data-driven decision-making within an organization.
  • Example: A student might develop a dashboard for a financial firm, enabling real-time monitoring of key performance indicators using interactive visualizations.

5. Social Good Projects:

  • Overview: Focused on leveraging data science for societal impact, these projects partner with non-profits or public sector organizations to solve pressing social issues.
  • Example: A group could work on analyzing public health data to identify trends and suggest interventions that could help reduce the spread of diseases in underserved communities.

Roles and Responsibilities

Successful Capstone Projects rely on collaboration between students, faculty advisors, and industry partners. Each plays a critical role in ensuring the project meets its objectives and delivers meaningful results.

  • Students
    • Project Execution: Students are responsible for conducting the research, analysis, and development work required by the project. This includes adhering to the project plan, meeting deadlines, and producing high-quality deliverables.
    • Communication: Regular updates are key to keeping the project on track. Students are expected to maintain clear and professional communication with both their faculty advisor and industry partner.
    • Problem-Solving: Students should proactively address challenges that arise during the project, seeking guidance from their advisors when necessary and iterating on their work based on feedback.
  • Faculty Advisors
    • Guidance and Support: Faculty advisors provide mentorship throughout the project, helping students refine their research questions, methodologies, and overall approach.
    • Feedback: Advisors review the work at key milestones, offering constructive feedback to ensure the project stays on course and meets academic standards.
    • Assessment: At the end of the project, faculty advisors evaluate the student's performance, contributing to the final grade.
  • Industry Partners
    • Project Proposal: Industry partners propose relevant and impactful projects that align with their business needs and provide valuable learning experiences for students.
    • Data and Resources: Partners provide access to necessary data, tools, and other resources, ensuring students have what they need to succeed.
    • Regular Check-Ins: Industry partners engage in regular check-ins with students, providing feedback and direction to keep the project aligned with their expectations.
    • Final Evaluation: Industry partners contribute to the assessment of the project's success, offering insights on the practical value of the work completed.

Expected Outcomes

Completing a Capstone Project at Vanderbilt Data Science is a transformative experience, providing participants with the opportunity to apply their skills and knowledge in a real-world setting. Here’s what you can expect to achieve:

1. Deliverables:

  • Final Report: A detailed report summarizing the project’s objectives, methodologies, findings, and implications. This document will serve as a formal record of your work and can be used to demonstrate your expertise to potential employers or collaborators.
  • GitHub Repository: A well-organized repository containing all code, data, and documentation related to the project. This ensures your work is reproducible and can be easily shared with others in the field.
  • Presentation: A professional presentation that highlights the key aspects of your project, including identifying a business-oriented problem statement, not just tech for the sake of tech, challenges faced, solutions developed, and potential future directions. This is an opportunity to showcase your ability to communicate complex ideas clearly and effectively.

2. Skills Development:

  • Technical Proficiency: Gain hands-on experience with the latest tools and technologies in data science, from machine learning frameworks to data visualization platforms. This practical knowledge is crucial for advancing your career.
  • Project Management: Develop your ability to manage time, resources, and collaboration in a project setting. You’ll learn how to set realistic goals, meet deadlines, and adapt to challenges as they arise.
  • Communication Skills: Enhance your ability to communicate complex data science concepts to both technical and non-technical audiences. This is an essential skill for any data scientist working in a multidisciplinary environment.

3. Professional Growth:

  • Industry Connections: Build relationships with industry partners who may become future employers or collaborators. These connections are invaluable for expanding your professional network and exploring career opportunities.
  • Portfolio Piece: Your capstone project will become a centerpiece of your professional portfolio, demonstrating your ability to apply data science to real-world problems. This can be a significant asset when applying for jobs or further academic pursuits.
  • Personal Achievement: Completing a Capstone Project is a significant milestone in your data science journey, providing a sense of accomplishment and confidence in your abilities.

Capstone Case Studies

Nashville SC: Fan Sentiment Analysis

Students analyzed Reddit discussions to gauge fan sentiment for Nashville SC compared to other local teams. Using AI-driven tools, they identified key insights that helped NSC enhance fan engagement strategies.

AllianceBernstein: ESG Sentiment Analysis

Students collaborated with AllianceBernstein to analyze the impact of ESG-related news on stock prices. Using advanced NLP models, they identified how negative ESG sentiment could predict potential declines in stock returns, providing valuable insights for sustainable investment strategies.

HCA: AI for Medical Diagnostics

In collaboration with HCA Healthcare, a student fine-tuned an AI model to improve the identification of vascular conditions from radiology reports. By enhancing the accuracy of medical diagnoses, this project aims to streamline healthcare processes and improve patient outcomes.

Preverity: Predicting Pediatric Malpractice Risks

A predictive model was developed to assess malpractice risks in pediatric healthcare. By leveraging logistic regression with Lasso regularization, the project aimed to enhance patient safety by identifying potential risks in medical practices.

Explore All Our Case Studies