Elementary Linear Algebra

UVM/IIS
Thursday, July 8, 2010

EUCLIDEAN SPACE
Thursday, July 8, 2010

Euclidean Space is
The Euclidean plane and three-dimensional space of Euclidean geometry, as well as the generalizations of these notions to higher dimensions. The term “Euclidean” is used to distinguish these spaces from the curved spaces of non-Euclidean geometry and Einstein's general theory of relativity.

Thursday, July 8, 2010

Euclidean Space
Euclidean n-space, sometimes called Cartesian space, or simply n-space, is the space of all ntuples of real numbers (x1, x2, ..., xn). It is commonly denoted R , although older n literature uses the symbol E .
n

Thursday, July 8, 2010

Euclidean Space
R is a vector space and has Lebesgue covering dimension n. Elements of R are called n-vectors. R 1= R is the set of real numbers (i.e., the real line) R is called the Euclidean Space.
2 n n

Thursday, July 8, 2010

One Dimension
R = R is the set of real numbers (i.e., the real line)
1

-∞ √2

0

-∞

0

1 √2 (1.41)

Thursday, July 8, 2010

Two Dimensions
R is called the Euclidean Space.
∞ P(-2, 1) 0
2

-∞

-∞

Thursday, July 8, 2010

Three Dimensions
y
∞ P(2, 2, -2)

-∞

0

x ∞

z
Thursday, July 8, 2010

-∞

n Dimensions
R Space of One Dimension (x, y) R Space of Two Dimensions (x, y) R Space of Three Dimensions (x, y, z) R Space of Four Dimensions (x1, x2, x3, x4) R Space of n Dimensions (x1, x2, x3, ...., xn)
n 4 3 2 1

Thursday, July 8, 2010

SOLUTION OF EQUATIONS
Thursday, July 8, 2010

Solutions of Systems of Linear Equations
x1 + x 2 = 1 x1 - x 2 = 1
HAS ONLY ONE SOLUTION: ∞

x1 = 1 x2 = 0
-∞ 0 ∞

-∞
Thursday, July 8, 2010

Solutions of Systems of Linear Equations
x1 + x 2 = 1 x1 + x 2 = 2

HAS NO SOLUTIONS
0

-∞

-∞
Thursday, July 8, 2010

Solutions of Systems of Linear Equations
x1 + x 2 = 1 2x1 + 2x2 = 2 HAS INFINITELY MANY SOLUTIONS
-∞ ∞

0

-∞
Thursday, July 8, 2010

Solutions of Systems of Linear Equations
In general:

A SYSTEM OF LINEAR EQUATIONS CAN HAVE EITHER: No solutions Exactly one solution Infinitely many solutions
Definition: If a system of equations has no solutions it is called an inconsistent system. Otherwise the system is consistent.

Thursday, July 8, 2010

Matrix Notation
MATRIX = RECTANGULAR ARRAY OF NUMBERS
0 2 1 1 0 1 -2 0 3 4 1 9

( )( ) )
3 2 -1 0 1 2

EVERY SYSTEM OF LINEAR EQUATIONS CAN BE REPRESENTED BY A MATRIX
Thursday, July 8, 2010

Elementary Row Operations
1. INTERCHANGE OF TWO ROWS

) ( )( )
0 2 1 1 0 1 -2 0 3 4 1 9 1 2 0 1 0 1 3 0 9 1 4 -2
Thursday, July 8, 2010

Elementary Row Operations
2. MULTIPLICATION OF A ROW BY A NON-ZERO NUMBER

) ( ) ( )
1 2 5 0 1 5 3 2 1 4 3 0 1 6 5 0 3 5 3 6 1 4 9 0 *3
Thursday, July 8, 2010

Elementary Row Operations
3. ADDITION OF A MULTIPLE OF ONE ROW TO ANOTHER ROW

) ( ) ( )
1 2 5 0 1 5 3 2 1 4 3 0 1 2 7 0 1 5 3 2 7 4 3 8 *2
Thursday, July 8, 2010

How to Solve Systems of Linear Equations
-x1 + 2x2 + 3x3 = 4 2x1 + 6x3 = 9 4x1 - x2 - 3x3 = 0

x1 = ... x2 = ... x3 = ...
Thursday, July 8, 2010

( (

-1 2 4

2 0 -1

3 6 -3

4 9 0

NICE MATRIX

) )

)

Linear Algebra Application
Google PageRank
Thursday, July 8, 2010

Early Search Engines
SEARCH QUERY DATABASE OF WEB SITES
LIST OF MATCHING WEBSITES IN RANDOM ORDER

PROBLEM: HARD TO FIND USEFUL SEARCH RESULTS
Thursday, July 8, 2010

Google Search Engine
DATABASE OF WEB SITES
WITH RANKINGS!

SEARCH QUERY
MATCHING WEBSITES IMPORTANT SITES FIRST!

Thursday, July 8, 2010

How to Rank?
VERY SIMPLE RANKING: Ranking of a page = number of links pointing to that page

PROBLEM: VERY EASY TO MANIPULATE

Thursday, July 8, 2010

Google PageRank
IDEA: LINKS FROM HIGHLY RANKED PAGES SHOULD WORTH MORE
IF

Ranking of a page is x The page has links to n other pages
THEN

Each link from that page should be worth x/n

Thursday, July 8, 2010

Google PageRank
THIS GIVES EQUATIONS:

x1 = x3 + 1/2 x4 x2 = 1/3 x1 x3 = 1/3 x1 + 1/2 x2 + 1/2 x4 x4 = 1/3 x1 + 1/2 x2

Thursday, July 8, 2010

Google PageRank
MATRIX EQUATION:
x1 x2 x3 x4 0 0 0 1 0 0 0 1/2 0 x1 x2 x3 x4

( ) ( )( ) )
=
1/3 1/3 1/2 1/3 1/2 1/2 0

COINCIDENCE MATRIX OF THE NETWORK
Thursday, July 8, 2010

Google PageRank
x1 x2 x3 x4 0 0 0 1 0 0 0 1/2 0 x1 x2 x3 x4

( ) ( )( ) )
=
1/3 1/3 1/2 1/3 1/2 1/2 0

( x1, x2, x3, x4 ) is an eigenvector of the coincidence matrix corresponding to the eigenvalue 1.

Thursday, July 8, 2010