What uml diagrams can be used for a data science project

I am working on a data science project for my 3-2 mini project. My project analyzes the performance of a country in the Olympics based on some attributes. But I am confused about the UML diagrams I should be using in my project.

There are some 15 UML diagram types out there. A sensible sequence of diagrams to be created depends on your approach. If you'd like to create an analysis model that is a conceptual model of your problem domain then a sensible sequence of diagrams might be:

1. Usecase diagrams
2. Activity diagrams
3. Class diagrams

and if your project gets bigger you might need package diagrams.

If you'd like to create a design model that is a conceptual model of your solution domain then a sensible sequence of diagrams might be: 1. Component diagrams 2. Class diagrams 3. Sequence diagrams 4. Statecharts

In both cases a starting point is having a diagram for your system context. Some people like to mix component and usecase diagram features to denote a system context.

The aspects you might want to take into concideration of your diagram choices are:

• syntax - how strictly would you like to follow the UML standard and what use does adhering to the standard have for you
• semantics - what is your need - what do you want to document - and who needs to understand it
• pragmatics - what is the best way to achieve your projects goal e.g. being efficient and effective
• tool - what tools do you have at hand and are used and known to your peers - what can you afford to invest in keeping the tool infrastructure up