machine learning - Difference between feature selection, clustering ,dimensionality reduction algorithm -

could indicate difference between feature selection , clustering , dimensionality reduction algorithms?

feature selection algorithms: allows find predominant variables either best represent data or best parameters indicate class eg : gbm / lasso

clustering helps indicate clusters of variables define output

isnt same dimensionality reduction algorithm? doesn't feature selection + clustering same dimensionality reduction algorithms?

feature selection:

in machine learning , statistics, feature selection, known variable selection, attribute selection or variable subset selection, process of selecting subset of relevant features (variables, predictors) use in model construction.

clustering:

cluster analysis or clustering task of grouping set of objects in such way objects in same group (called cluster) more similar (in sense or another) each other in other groups (clusters).

dimensionality reduction:

in machine learning , statistics, dimensionality reduction or dimension reduction process of reducing number of random variables under consideration, , can divided feature selection , feature extraction.

when have many features , want use of them can apply feature selection (i.e. mrmr). so, means have applied dimensionality reduction.

however, clustering assignment of set of observations subsets (called clusters) observations in same cluster similar in sense. clustering method of unsupervised learning, , common technique statistical data analysis used in many fields (check clustering in machine learning). when want group (cluster) different data points according features can apply clustering (i.e. k-means) with/without using dimensionality reduction.