Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 63

Mathematics for Computing

DISCIPLINE SPECIFIC CORE COURSE – 1:


Course: B.Sc. (H) Computer Science
Unit-III
Content of Unit-III
I. EigenValue and EigenVector: Characteristic
Polynomial,
II.Cayley Hamilton Theorem (Only in numericals),
III.Eigen Value And Eigen Vector of a Matrix,
IV.Eigenspaces,
V. Diagonalization,
VI.Positive Definite Matrices,
VII.Applications to Markov Matrices
Firstly We Need to Revise
Unit-I’s Topic:-
DETERMINANTS
Ans.:-
Topic:- I, III & IV
EIGENVALUE AND EIGENVECTOR:
CHARACTERISTIC POLYNOMIAL,
EIGEN VALUE AND EIGEN VECTOR
OF A MATRIX, EIGENSPACES.
STEPS FOR FINDING EIGENVALUE AND EIGENVECTORS

• Step 1: Construct a polynomial


• Step 2: Find eigenvalues of by solving that will be .
• Step 3: Find the Eigen space for each of the eigenvalues of A.
• Step 4: Next, solve the homogeneous system for all the
Eigenvalue .
• Step 5: Further, with help of the augmented matrix and gauss
Jorden method find the solution set for the homogeneous
system which will give Eigen space .
• Step 6: Finally, find the eigenvectors for the Eigenvalue which
will be the nonzero scalar multiples of
Topic:-II
CAYLEY-HAMILTON THEOREM
(ONLY IN NUMERICALS)
Topic:-
DIAGONALIZATION
• One of the most important uses of eigenvalues and
eigenvectors is in the diagonalization of matrices.

• Because diagonal matrices have such a simple structure, it is


relatively easy to compute a matrix product when one of the
matrices is diagonal.
• As we will see later, other important matrix computations are
also easier when using diagonal matrices.
• Hence, if a given square matrix can be replaced by a
corresponding diagonal matrix, it could greatly simplify
computations involving the original matrix.
[ 𝟔 𝟔 𝟏𝟐
]
,− ,− ,𝟔 = [ 𝟏,−𝟐 ,−𝟒
𝟔 𝟑 𝟑
Topic:-VI
Positive Definite Matrices
• There are three method to test positive definite matrices,
which are as follows:
Topic:-VII
Applications to Markov Matrices
Check Matrix P is Markov Matrices or not?

You might also like