102: How to do a Data Science Project

Problems in Data Science

  • Understanding the problem

  • “the five-whys”

  • Different questions dramatically effect the tools and techniques used to solve the problem.

Data Science as a Process

  • More Science than Engineering
Research Problem Model
  • High risk
  • High reward
  • Difficult
  • Unpredictable
CRISP-DM Process

By Kenneth Jensen CC BY-SA 3.0, via Wikimedia Commons

Impacts of Data Science

  • What is the purpose of the project? Who is affected?

  • Which parts of the business are affected? Do we need help?

  • You must think about the human concerns.

  • You need buy-in from the business; the business will be affected.

Conclusions

Business goals: make money, save money or save time. Data Science generates profit.

Project justifications - you now know how they are judged:

  • Alignment with Business Goals
  • A well defined, testable requirement
  • A robust plan
    • Data Understanding
    • Data Preparation
    • Modelling
    • Evaluation
    • Deployment
    • Iteration of the above
  • Buy-in and integration with other parts of the business

However, there are more philanthropic, scientific reasons for undertaking a project too. So these arguments may not directly apply to charitable causes or academia.

More articles

LLM Architecture: RAG Implementation and Design Patterns

Discover the most common LLM architectures for retrieval augmented generation. Including pros and cons and how to choose the best RAG architecture.

Read more

Exploring Small Language Models

Learn about efficient Small Language Models (SLMs) as cost-effective, resource-saving alternative to Large Language Models (LLMs) for AI applications.

Read more
}