Given
x1…xk∈R
x
1
…
x
k
,
we can define a linear space (vector space) XX as
X=spanx1…xk=
∑
i
=1k
α
i
xi
α
i
∈R
X
span
x
1
…
x
k
i
1
k
α
i
x
i
α
i
(1)
In this case,
x1…xk
x
1
…
x
k
form what is known as a
generating set for the
space
XX. That is to say that any vector in
XX
can be generated by a linear combination of the vectors
x1…xk
x
1
…
x
k
.
If
x1…xk
x
1
…
x
k
happen to be linearly independent, then they also form a
basis for the space XX. When
x1…xk
x
1
…
x
k
define a basis for XX, kk is the dimension of XX.
A basis is a special subset of a generating set. Every generating
set includes a set of basis vectors.
The following three vectors form a generating set for the linear space
R2
2
.
x1=11
x
1
1
1
,
x2=10
x
2
1
0
,
x3=21
x
3
2
1
It is obvious that these three vectors can be combined to form
any other two dimensional vector; in fact, we don't need this
many vectors to completely define the space. As these vectors
are not linearly independent, we can eliminate one of them.
Seeing that
x3
x
3
is equal to
x1+x2
x
1
x
2
,
we can get rid of it and say that our basis for
R2
2
is formed by
x1
x
1
and
x2
x
2
.