MCDA uses several criteria to rank "objects".

Multi-attribute analysis concerns "objects" that can be described through several attributes. I could use one criteria to rank object, but it should take into accounts all the relevant attributes.

Thank you very much for your answer.

I would like be sure that I good understand you. Let's conciderate example of buying a car .

For example, in MCDA we could take criteria of price (we take only attribute of buying price) and horsepower. On the other hand, in MAA - we could take criterion of acceleration with attributes: car weight. horsepower and shape of car.

It is the correct example.

In MAA you perceive that the objects to be evaluated have several dimensions (n) = attributes, like in the case of a car. In order to take into account all of these aspects, you can use some criteria (m). A criteriion will depend on some out of (may be all of) the dimensions. You're working in an (n,m) "ideal" matrix.

If m = 1 you accept trade-offs between each attribute.

If m > 1 you accept m (potentially) different trade-offs. Think of the case of a car and that you have to purchase it. Assume too you have a partner (it can happen), with potentially different tastes. Well, in this case you have to manage n properties of the car with m = 2 criteria. Good luck!

I agree with Mr. Elio Padoano.

Some researchers use to say that criteria and attribute are two words for the same concepts, some does not.

I have found in papers the both definitions, and particulary I think the concepts are more equal than different.

Thank you all, I was expecting that the naming in this area is a little vague. I saw that a some researchers are using of both approaches in their papers. I'm glad that i'm not alone with these insights. This state of affairs is quite strange for me. One more time, thank you all for your participation in the discussion.

Mcdm includes madm and modm.

MCDM is a general term and it is divided in two subsets as MADM (multi attribute decision making) and MODM (multi objective decision making). Hence there is no different between MCDM and MADM. all MADM methods like TOPSIS, AHP, PROMETHEE, etc are known as MCDM methods.

Hi Dear

MCDM consist of two subsets :

1-MADM   ,   2-MODM.

So to take making decision by MCDM techniques we face with  two types of problems: Qualitative and quantitative.

To making decision in  qualitative problems (like location to build a hospital in 3 Candidates places) use MADM.

To making decision in quantitative problems (like location to build a hospital in city ) use MODM.

multi attribute decision making is differs from multi objective decision making, because in data mining attribute selection is one of the objective. That is based on different objective we have to take final decision in data mining.

Hi 

Multi criteria decision making is divided by two main categories as Multi objective decision making (MODM) and multi attribute decision making methods (MADM).. These days usually authors use both these names for academic objectives, but MADM is more specific like methods TOPSIS, VIKOR, ...and MODM methods as MOORA.. 

I am also often confused. But I prefer to use multi-criteria.