Good day!

There is a dataset of vehicle configurations:

• ID
• Brand
• Model
• Generation
• Engine Type
• Engine Capacity
• Number of cylinders
• Body type
• Gearbox type
• Engine code
• Year of manufacture

Task:
It is necessary to determine how one configuration is similar to another.

I assumed to represent each entry in the dataset as a vector, and calculate the cosine similarity between the vectors.

But there is a misunderstanding of how to represent values ​​in numerical form, for example, Body Type: sedan, crossover, coupe, etc.

Thanks for your help