Difference between Dimension, Attribute and Feature in Machine Learning [closed]

Question

This is two dimensional: [[2,2]] but it also has 2 features/attributes doesn't it. I am confused on what the difference between a dimension, attribute and feature is.

Answer 1

I have to dissagree with @Atilla answer

• Dimension usually refers to the number of attributes, although it can also be used in form of "second dimension of the data vector is person age", but it is rather rare - in most cases dimension is "number of attributes"
• Attribute is one particular "type of data" in your points, so each observation/datapoint (like personal record) contains many different attributes (like person weight, height, age, etc.)
• Feature may have multiple meanings depending on context:
  • It sometimes refers to attribute
  • It sometimes refers to the internal representation of the data generated by particular learning model, for example - neural networks extract features which are combinations of the attributes or other features
  • It sometimes refers to the hypothethical representation of the data induced by the kernel method (in Kernel PCA, Kernel k-means, SVM)

In general you have some objects X, which you describe using some attributes (which is the first step of feature extraction, and so these attributes are also sometimes refered as features), which creates a representation of given dimension (number of attributes, extracted features). Then you train some model, which often creates some kind of abstraction (sometimes even multi-level), and each of such abstractions generate new features (extracts features from features) which are more complex objects then the ones on the lower "level".

 X  --->   repr(X)   --->   f1(repr(X)) --->   ....  --->   fn(repr(X))
data      attributes         1st level                      nth level
        (0th features)       features                       features

      |repr(X)|=dimension

f's are often recurrent, so f2(repr(X)) is actually some f2'(f1(repr(X))

Answer 2

They are the same things. Attribute, dimension and feature. According to writer's background or domain they are used interchangeably.

For example if you are talking about mathematical aspects , you can say this is high dimension problem.