CS 2210 Discrete Math Graphs Fall 2019 Sukumar Ghosh Seven Bridges of Knigsbergnigsberg

Is it possible to walk along a route that cross each bridge exactly once? Seven Bridges of Knigsberg A Graph What is a Graph

A graph G = (V, E) consists of V, a nonempty set of vertices (or nodes) and E, a set of edges. Each edge connects a pair of nodes that are called its endpoints. Graphs are widely used to model various systems in the real world

Back to the Bridges of Knigsbergnigsberg Eulers solution Euler path Simple graph

Types of graph Definitions At most one edge between a pair of nodes

Multiple edges between some pair of nodes Simple graphs vs. multi-graphs

Undirected vs. directed graphs (digraphs) Each edge between a pair (u, v) of nodes is directed, and represents an ordered pair More examples of graphs

Hollywood graph Web graphs Each node denotes an actor or an actress, and each edge between P and Q denotes that P, Q worked

together in some movie. It is an undirected graph Each node denotes a web page, and each edge from page P to Q Q denotes a link on page P pointing to page Q.

It is a directed graph Application: Exam scheduling Why? Problems in a computer network

Vertex degree Degree sequence Handshaking theorem

Handshaking theorem A theorem THEOREM. An undirected graph has even number of vertices of odd degree. Can you prove this? It should follow from the

handshaking theorem. Review of basic definitions Review of basic definitions Types of graphs

A cycle of a graph is a subset of its edge set that forms a path such that the first node of the path corresponds to the last. Types of graphs The n-cube graph n=3

Complete graph: All vertices are adjacent Wheel graph Types of graphs

Bipartite graph A simple graph is called bipartite if its vertex set V can be partitioned into two disjoint subsets V1 and V2, such that every edge in the graph connects a vertex in V1 to a vertex in V2. Can always be colored using two colors.

Subgraphs Computer representation of graphs ADJACENCY MATRIX (taken from Wolfram Mathworld)

Computer representation of graphs ADJACENCY LIST 1 2 4

Vertex Adjacent to 1

3, 4 2 3, 4 3

1, 2 4 1, 2

3 Can be represented as a linked list Graph isomorphism Taken from MIT 6.042J/18.062J

Graph isomorphism Taken from MIT 6.042J/18.062J Graph isomorphism

Taken from MIT 6.042J/18.062J Connectivity An undirected graph is connected if there is a path between every pair of distinct vertices of the graph. A connected component is the maximal connected subgraph of the given graph.

Connectivity issues Erds number in academic collaboration graph Erds number = n means the person collaborated with someone whose Erds number is (n-1) Kevin Bacon number in Hollywood graph Actor Kevin Bacon once remarked that he worked with

everybody in Hollywood, or someone who worked with them. Kevin Bacon number is an adaptation of Erds number in Hollywood movie industry. Cut vertex, cut set, cut edge A cut vertex (or articulation point ) is a vertex, by removing which one can partition the graph. A cut edge is an edge by

removing which one can partition the graph. If multiple edges need to be remove to partition the graph, then the minimal set of such edges a cut set. Examples taken from Wikipedia Connectivity in directed graphs

A directed graph is strongly connected if there is a path from any vertex a to any other vertex b of the graph. A directed graph is weakly connected if there is a path between any two vertices of the underlying undirected graph. Strongly or weakly connected ?

More definitions Vertex cover is a famous problem in graph theory Vertex Cover A vertex-cover of an undirected graph G=(V,E) is a subset V of V such that if edge (u, v) is an edge of G, then u is in V, or v is in V, or both. The set

V is said to "cover" the edges of G Minimal or minimum vertex cover is a famous problem in graph theory Dominating Set A dominating set for a graph G = (V, E) is a subset D of V such that every vertex not in D is adjacent to at least one member of D.

(from Wikipedia) Computing a minimal or minimum dominating set is a famous problem in graph theory Independent Set Given a graph G = (V, E) an independent set is a subset of vertices

no two of which are adjacent to one another. (from Wikipedia) Computing the maximal independent set is a well-known problem in graph theory Euler path vs. Hamiltonian path

Hamiltonian path = A path that passes through every vertex exactly once. A closed path is a Hamiltonian circuit or cycle. Euler path = A path that includes every edge exactly once. A closed path is a Euler circuit or cycle. We have reviewed Euler path in the 7-bridges of Konigsberg Problem.

Hamiltonian path 1 2 5 4

3 Hamiltonian circuit/cycle colored red Does the above graph have

a Hamiltonian cycle? No! Shortest path Weighted graph. Compute the shortest path from a to z Shortest path: Dijkstras algorithm

Read the algorithm from page 712 of your text book Shortest path: Dijkstras algorithm Shortest path: Dijkstras algorithm

Shortest path: Dijkstras algorithm Computes the shortest path from a source node to each target node L (source) = 0, and for all other node u, L(u) := infinity, S:= null while z is not in S u := a vertex not in S with L(u) minimal; S := S {u}; {u}; for all vertices v not in S

if L(u) + w(u, v) < L(v) then L(v) := L(u) + w(u, v) {known as relaxation: this adds a vertex with minimal label to S and updates the labels of vertices not in S} return L(z) Traveling Salesman Problem (TSP) A traveling salesman wants to visit each of

n cities exactly once, and then return to the starting point. In which order should he visit the cities to travel the minimum total distance? TSP = Computing the minimum cost Hamiltonian circuit. TSP is an extremely

hard problem to solve (NP-complete) An optimal TSP tour through Germanys largest cities (Source: Wikipedia) Planar Graph

A planar graph is one that can be embedded in the plane, i.e., it can be drawn on the plane in such a way that its edges do not intersect except only at their endpoints. planar K4

Butterfly planar Non-planar Non-planar

K5 K3,3 Planar Graph How to verify that a graph is a planar graph? It should not depend upon how you draw the graph.

Any graph that contains a K5 or K3,3 as its sub-graph is not planar Graph Coloring Let G be a graph, and C be a set of colors. Graph coloring finds an assignment of colors to the different nodes of G, so that no two adjacent nodes have the same color.

The problem becomes challenging when the |C| is small. Chromatic number. The smallest number of colors needs to color a graph is called its chromatic number. Graph Coloring The chromatic number of a tree is 2.

Graph Coloring What are the chromatic numbers of these two graphs? Four color theorem Theorem. Any planar graph can be colored using

at most four colors. It all started with map coloring bordering states or counties must be colored with different colors. In 1852, an ex-student of De Morgan, Francis Guthrie, noticed that the counties in England could be colored using four colors so that no adjacent counties were assigned the same color. On this evidence, he conjectured the four-color theorem. It took nearly 124 years

to find a proof. It was presented by Andrew Appel and Wolfgang Haken. CS2210:0001 Discrete Structures Trees Fall 2019 Sukumar Ghosh

What is a tree? Rooted tree: recursive definition Rooted tree terminology

(If u = v then it is not a proper descendant) If v is a descendant of u, then u is an ancestor of v Rooted tree terminology A subtree

Rooted tree terminology Important properties of trees Important properties of trees Theorem. A tree with n nodes has (n-1) edges Proof. Try a proof by induction

Important properties of trees Theorem. A tree with n nodes has (n-1) edges Proof. Try a proof by induction Trees as models

Domain Name System Trees as models directory subdirectory

file file file subdirectory

file file file

subdirectory file file file

Computer File System This tree is a ternary (3-ary) tree, since each non-leaf node has three children Trees as models: game tree Binary and m-ary tree

Binary tree. Each non-leaf node has up to 2 children. If every non-leaf node has exactly two nodes, then it becomes a full binary tree. m-ary tree. Each non-leaf node has up to m children. If every non-leaf node has exactly m nodes, then it becomes a full m-ary tree Properties of trees

Theorem. A full m-ary tree with k internal vertices contains n = (m.k + 1) vertices. Proof. Try to prove it by induction. [Note. Every node except the leaves is an internal vertex] Properties of trees

Theorem. Every tree is a bipartite graph. Theorem. Every tree is a planar graph. Balanced trees The level of a vertex v in a rooted tree is the length of the unique path from the root to this vertex. The level of the root is zero. The height of a rooted tree is the maximum of the levels of vertices.

The height of a rooted tree is the length of the longest path from the root to any vertex. A rooted m-ary tree of height h is balanced if all leaves are at levels h or h 1. Balanced trees Theorem. There are at most mh leaves in an m-ary tree of height h.

Proof. Prove it by induction. Corollary. If an m-ary tree of height h has l leaves, then If the m-ary tree is full and balanced, then Binary search tree Ordered binary tree. For any non-leaf node The left subtree contains the lower keys.

The right subtree contains the higher keys. How can you search an item? How many steps A binary search tree of size 9 and depth 3, with root 8 and leaves 1, 4, 7 and 13

does each search take? Binary search tree Insertion in a binary search tree procedure insertion (T : binary search tree, x: item) v := root of T {a vertex not present in T has the value null }

while v null and label(v) x if x < label(v) then if left child of v null then v := left child of v else add new vertex as a left child of v and set v := null else if right child of v null then v := right child of v else add new vertex as a right child of v and set v := null

if root of T = null then add a vertex v to the tree and label it with x else if v = null or label(v) x then label new vertex with x and let v be the new vertex return v {v = location of x} Decision tree Decision trees generate solutions via a sequence of decisions.

Example 1. There are seven coins, all of which are of equal weight, and one counterfeit coin that is lighter than the rest. Given a weighing scale, in how many times do you need to weigh (each weighing determines the relative weights of the objects on the the two pans) to identify the counterfeit coin? {We will solve it in the class}.

Comparison based sorting algorithms A decision tree for sorting three elements Comparison based sorting algorithms Theorem. Given n items (no two of which are equal), a sorting algorithm based on binary comparisons requires at least

comparisons Proof. See page 761-762 of your textbook. We will discuss it in the class The complexity of such an algorithm is Why? Spanning tree

Consider a connected graph G. A spanning tree is a tree that contains every vertex of G Many other spanning trees of this graph exist Computing a spanning tree Given a connected graph G, remove the edges (in some

order) without disrupting the connectivity, i.e. not causing a partition of the graph. When no further edges can be removed, a spanning tree is generated. Graph G Computing a spanning tree

Spanning tree of G Depth First Search procedure DFS (G: connected graph with vertices v1vn) T := tree consisting only of the vertex v1 visit(v1) {Recursive procedure}

procedure visit (v: vertex of G) for each vertex w adjacent to v and not yet in T add vertex w and edge {v, w} to T visit (w) The visited nodes and the edges connecting them form a spanning tree. DFS can also be used as a search or traversal algorithm

Depth First Search: example Breadth First Search A different way of generating a spanning tree

Given graph G Spanning tree Breadth First Search ALGORITHM. Breadth-First Search. procedure BFS (G: connected graph with vertices v1, v2, vn)

T := tree consisting only of vertex v1 L := empty list put v1 in the list L of unprocessed vertices while L is not empty remove the first vertex v from L for each neighbor w of v if w is not in L and not in T then

add w to the end of the list L add w and edge {v, w} to T Minimum spanning tree A minimum spanning tree (MST) of a connected weighted graph is a spanning tree for which the sum of the edge weights is the minimum.

How can you compute the MST of a graph G? Huffman coding Consider the problem of coding the letters of the English alphabet using bit-strings. One easy solution is to use 5 bits for each letter (25 > 26). Another such example is The ASCII code. These are static codes, and do not make

use of the frequency of usage of the letters to reduce the size of the bit string. One method of reducing the size of the bit pattern is to use prefix codes. Prefix codes 0

e In typical English texts, e is most frequent, followed by, l, n, s, t The prefix tree assigns to each letter of the alphabet a code whose length depends on the

frequency: 1 0 1

a 0 1 l

0 1 n

0 s e = 0, a = 10, l= 110, n = 1110 etc 1 t

Such techniques are popular for data compression purposes. The resulting code is a variable-length code. Huffman codes Another data compression technique first developed

By David Huffman when he was a graduate student at MIT in 1951. (see pp. 763-764 of the textbook) Huffman coding is a fundamental algorithm in data compression, the subject devoted to reducing the number of bits required to represent information.

Huffman codes Example. Use Huffman coding to encode the following symbols with the frequencies listed: A: 0.08, B: 0.10, C: 0.12, D: 0.15, E: 0.20, F: 0.35. What is the average number of bits used to encode a character?

Huffman coding example Huffman coding example Huffman coding example Huffman coding example

So, in this example, what is the average number of bits needed to encode each letter? 3x 0.08 + 3 x 0.10 + 3 x 0.12 + 3 x 0.15 + 2 x 0.20 + 2 x 0.35 = 2.45 instead of 3-bits per letter.