Presentation is loading. Please wait.

Presentation is loading. Please wait.

HASH TABLE. HASH TABLE a group of people could be arranged in a database like this: Hashing is the transformation of a string of characters into a.

Similar presentations


Presentation on theme: "HASH TABLE. HASH TABLE a group of people could be arranged in a database like this: Hashing is the transformation of a string of characters into a."— Presentation transcript:

1

2 HASH TABLE

3 a group of people could be arranged in a database like this:
Hashing is the transformation of a string of characters into a usually shorter fixed-length value or key that represents the original string. Example a group of people could be arranged in a database like this: Allen, Jane Moore, Sarah Smith, Dan

4 H A S I N G Hash Table 7864 Allen, Jane Moore, Sarah 9802 1990
Smith, Dan HASH VALUES HASH KEYS HASH FUNCTION

5 Hash Table stores things and allows 3 operations: insert, search and delete. associated with a set of records

6 Bob Miller 34 John Smith Sally Wood 21 H John Smith 29 5

7 Each slot of a hash table is called a bucket and hash values are called bucket indices.
7864 Allen, Jane BUCKET BUCKET INDEX

8 HASH FUNCTION Mapping of the keys to indices of a hash table
2 compositions Hash code map: key integer Compression map: integer [0, N-1]

9 DIVISION Example: If table size m = 12 key k = 100
Map a key k into one of m slots by using this function: h(k) = k mod m Example:     If table size m = 12                 key k = 100     than         h(100) = 100 mod 12                    = 4

10 Ex. k=3121 then 31212=9740641 thus h(3121)= 406 MID-SQUARE FUNCTION
The key is squared and the mid part is used as the address. Ex. k=3121 then 31212= thus h(3121)= 406

11 Folding Key is divided into several parts 2 types 1. shift folding
2. boundary folding

12 Shift Folding Ex. (SSN) 123-45-6789 1. Divide into 3 parts:
123, 456 and 789. 2. Add them. =1368 3. h(k)=k mod M where M = 1000 h(1368) = 1368 mod 1000 = 368 1. Divide into five parts: 12, 34, 56, 78 and 9. 2. Add them. = 189 3. h(k)=k mod M where M = 1000 h(189) = 189 mod 1000 = 189

13 1st 4 digits = 1234 Last 4 digits = 6789
Extraction Only a part of the key is used to compute the address. Ex. (SSN) 1st 4 digits = 1234 Last 4 digits = 6789 1st 2 combined with the last 2 = 1289(address)

14 Hash Method : Folding Chopping the Key in Two Parts
Add the Two Parts to Generate the Hash Leading Digit will be Ignored Example Key Parts H(x) Option Rotate the Second Digit Parts H(x)

15 Radix Transformation K is transformed into another number base M = 100
21210=2559 M = 100 H(k) = k mod M H(255) = 255 mod 100 = 55

16 212= 9(9(2)+ 5)+ 5 = 2(92)+ 5(9)+ 5. divide 212 by 9.
9 divides into times with remainder = 9(23)+ 5 9 divides into 23 twice with remainder = 9(2)+5 212= 9(9(2)+ 5)+ 5 = 2(92)+ 5(9)+ 5.

17 different keys happen to have same hash value
Hash Collision different keys happen to have same hash value

18 Bob Miller 34 Jane Depp 18 Collision! Sally Wood 21 2 John Smith 29

19 Collision Resolution There are two kinds of collision resolution:
1 – Chaining makes each entry a linked list so that when a collision occurs the new entry is added to the end of the list. 2 – Open Addressing uses probing to discover an empty spot.

20 Collision Resolution – Open Addressing
the table is probed for an open slot when the first one already has an element. Linear probing in which the interval between probes is fixed — often at 1. Quadratic probing in which the interval between probes increases linearly (hence, the indices are described by a quadratic function). Double hashing in which the interval between probes is fixed for each record but is computed by another hash function.

21 H(x,i) = (H(x) + i)(mod M)
Linear Probing is a scheme in resolving hash collisions of values of hash functions by sequentially searching the hash table for a free location two values - one as a starting value and one as an interval between successive values  newLocation = (startingValue + stepSize) % arraySize H(x,i) = (H(x) + i)(mod M)

22 Linear Probing - Example
1 2 3 4 5 6 7 8 9 empty Insert 15, 17, 8 1 2 3 4 5 6 7 8 9 empty 15 17 8 H(8)=8 mod 10 = 8 H(15)=15 mod 10 = 5 H(17)=17 mod 10 = 7

23 1 2 3 4 5 6 7 8 9 empty 75 15 35 17 8 25 Insert 25 Insert 35 H(1,8)=(1 + 8) mod 10 = 9 H(35)=35 mod 10 = 5 H(1,6)=(1 + 6) mod 10 = 7 H(1,7)=(1 + 7) mod 10 = 8 H(1,5)=(1 + 5) mod 10 = 6 H(25)=25 mod 10 = 5 H(1,5)=(1 + 5) mod 10 = 6 Insert 75 H(1,9)=(1+9) mod 10 = 0 H(1,8)=(1+8) mod 10 = 9 H(1,5)=(1+5) mod 10 = 6 H(75)=75 mod 10 = 5 H(1,6)=(1+6) mod 10 = 7 H(1,7)=(1+7) mod 10 = 8

24 Has anyone spotted the flaw in the linear probing technique
Has anyone spotted the flaw in the linear probing technique? Think about this: what would happen if we now inserted 85, then 95, then 55?

25 Each one would probe exactly the same positions as its predecessors
Each one would probe exactly the same positions as its predecessors. This is known as clustering. It leads to inefficient operations, because it causes the number of collisions to be much greater than it need be.

26 eliminates primary clustering p(K, i) = c1 i2 + c2i + c3
Quadratic Probing eliminates primary clustering p(K, i) = c1 i2 + c2i + c3  p(K, i) = i2 (i.e., c1 = 1, c2 = 0, and c3 = 0)

27 Quadratic Probing - Example
Table Size is 11 (0..10) Hash Function: h(x) = x mod 11 Insert keys: 20 mod 11 = 9 30 mod 11 = 8 2 mod 11 = 2 13 mod 11 = 2  2+12=3 25 mod 11 = 3  3+12=4 24 mod 11 = 2  2+12, 2+22=6 10 mod 11 = 10 9 mod 11 = 9  9+12, 9+22 mod 11, 9+32 mod 11 =7 1 2 3 13 4 25 5 6 24 7 9 8 30 20 10

28 not all hash table slots will be on the probe sequence
Using p(K, i) = i2 gives particularly inconsistent results If all slots on that cycle happen to be full, this means that the record cannot be inserted at all!

29 Double Hashing P = (1 + P) mod TABLE_SIZE
 increment P, not by a constant but by an amount that depends on the Key.  P = (1 + P) mod TABLE_SIZE P = (P + INCREMENT(Key)) mod TABLE_SIZE

30 Double Hashing - Example
P = (P + INCR(Key)) mod TABLE_SIZE Suppose INCR(Key) = 1 + (Key mod 7) Adding 1 guarantees it is never 0! Insert 15, 17, 8:

31 Insert 35: P = H(35) = 5. P = (5 + ( mod 7)) mod 10 = 6. Insert 25:P = H(25) = 5. P = (5 + ( mod 7)) mod 10 = 0

32 10 3 2 1 4 9 5 8 6 7 Let’s try! Insert 75: P = (P + INCR(Key)) mod TABLE_SIZE Suppose INCR(Key) = 1 + (Key mod 7)

33 Chaining/Separate Chaining
uses an array as the primary hash table an array of lists of entries

34 Chaining nil nil nil : nil
One way to handle collision is to store the collided records in a linked list. The array now stores pointers to such lists. If no key maps to a certain hash value, that array entry points to nil. 1 nil 2 nil 3 4 nil 5 : Key: name: tom score: 73 HASHMAX nil

35

36 29 16 14 99 127 129 16 127 99 29 14 29 129

37 is a collision resolution method that
Coalesced Hashing is a collision resolution method that uses pointers to connect the elements of a synonym chain. A hybrid of separate chaining and open addressing. Linked lists within the hash table handle collisions. This strategy is effective, efficient and very easy to implement.

38 A5, A2, A3 B5, A9, B2 B9, C2 Insert: A2 A2 A2 A3 A3 A3 C2 A5 A5 A5 B9

39 Insert: A5, A2, A3 B5, A9,B2 B9,C2 A2 A2 A2 A3 A3 A3 A5 A5 A5 C2 A9 A9

40 using additional space
Bucket Addressing using additional space A bucket can be defined as a block of space that can be used to store multiple elements that hash to the same position.

41 Insert: A5, A2, A3, B5, A9, B2, B9 A2 B2 A3 A5 B5 A9 B9

42 TOMBSTONE DELETION Deleting a record must not hinder later searches.
The search process must still pass through the newly emptied slot to reach records whose probe sequence passed through this slot.  It should not mark the slot as empty. Freed slot should be available to a future insertion. TOMBSTONE

43 28 83 25 75 35 Insert: Collision Probing Sequence: 25 75 Delete: 35 83 Match Found! TOMBSTONE 75 28

44 A1, A4, A2,B4,B1 A2 A4 Delete: Insert: A1 B1 A2 B1 B4 A4 B4

45 Perfect Hash Functions
Quick to compute Distributes keys uniformly throughout the table Very rare(birthday paradox) No collisions Perfect hash functions are rare.

46 A Perfect Hash Function for Strings
R. J. Cichelli gave an algorithm for finding perfect hash functions for strings. He proposes the hash function: h(s)=size+g(s.charAt(0))+ g(s.charAt(size-1))%n where size = s.length(). The function g is to be constructed so that h(s) is unique for each string s.

47

48

49

50

51 Example 1: Illustrating Perfect Hashing
Use Cichelli's algorithm to build a minimal perfect hash function for the following nine strings: DO DOWNTO ELSE END IF IN TYPE VAR WITH

52 Example 1: Solution For Step 1 in the algorithm, we find the frequencies of the first and last letter of each word to find: D O E I F N T V R W H Next we find the sum of the first and last letter of each word: DO=5(D+0=3+2), DOWNTO=5, ELSE = 8, END=7, IF=3, IN=3, TYPE=5, VAR=2,WITH=2 Sorting the keywords in decreasing frequency yields: ELSE END DOWNTO DO TYPE IN IF VAR WITH We are now at step 5 of the algorithm, the heart of the algorithm. We try the words in frequency order:

53 Example 1: Cichelli's Method (cont'd)
s = ELSE g(E)=0 h(s)= s.length()+g(E)+g(E)=4 s = END g(D)=0 h(s)= s.length()+g(E)+g(D)=3 s = DOWNTO g(O)=0 h(s)= s.length()+g(D)+g(O)=6 s = DO h(s)= s.length()+g(D)+g(O)=2 s = TYPE g(T)=0 h(s)= s.length()+g(T)+g(E)=4* s = TYPE g(T)=1 h(s)= s.length()+g(T)+g(E)=5 s = IN g(I)=0,g(N)=0 h(s)=s.length()+g(I)+g(N)=2* s = IN g(I)=1,g(N)=0 h(s)=s.length()+g(I)+g(N)=3* s = IN g(I)=2,g(N)=0 h(s)=s.length()+g(I)+g(N)=4* s = IN g(I)=3,g(N)=0 h(s)=s.length()+g(I)+g(N)=5* s = IN g(I)=3,g(N)=1 h(s)=s.length()+g(I)+g(N)=6* s = IN g(I)=3,g(N)=2 h(s)=s.length()+g(I)+g(N)=7 s = IF g(F)=0 h(s)=s.length()+g(I)+g(F)=5* s = IF g(F)=1 h(s)=s.length()+g(I)+g(F)=6* s = IF g(F)=2 hash(s)=s.length()+g(I)+g(F)=7* s = IF g(F)=3 h(s)=s.length()+g(I)+g(F)=8

54 Example 1: Cichelli's Algorithm (cont'd)
VAR WITH DO END ELSE TYPR DOWNTO IN IF The hash table above is fully occupied with empty slots. Note that if there are empty slots or there is a collision, then the g-value assignments are in error.

55 Extendible hashing Hashing with buckets

56 Extendible Hashing - Class Example

57 Leaf Pages Directory d1 = local depth d = global depth rec 1 rec 4
splitting bucket rec 1 rec 2 d1=0 1 splitting bucket d = 0 record 3 = overflow!! d = 1 rec 2 rec 3 record 5 = overflow!! NEXT

58 rec 1 rec 4 00 01 rec 2 10 splitting bucket rec 3
d = 2 d1 = 2 d1 = 1 11 01 rec 1 rec 4 rec 2 splitting bucket rec 3 record 7 = overflow!! rec 5 rec 6 NEXT

59 000 110 d = 3 111 001 010 011 100 101 rec 1 rec 4 splitting bucket
record 8 = overflow!! d1 = 3 rec 2 rec 7 d1 = 3 rec 3 rec 5 rec 6 NEXT

60 rec 1 NEXT rec 4 rec 8 000 001 010 011 100 rec 2 101 rec 7 110 rec 3
d1 = 3 d1 = 2 rec 1 rec 4 rec 8 000 110 d = 3 111 001 010 011 100 101 d1 = 3 d1 = 2 rec 2 rec 3 rec 5 rec 6 rec 7 rec 9 splitting bucket record 10 = overflow!!

61 NEXT rec 7 rec 9 d1 = 3 rec 1 rec 4 000 110 d = 3 111 001 010 011 100 101 rec 2 rec 3 d1 = 2 rec 8 rec 11 rec 12 d1 = 3 rec 5 rec 6 splitting bucket rec 10 record 13 = overflow!!

62 d1 = 4 d1 = 3 d1 = 2 0000 1110 d = 4 1111 0001 0010 0011 1100 1101 0100 0101 0110 0111 1010 1011 1000 1001 rec 1 rec 4 rec 2 rec 3 rec 5 rec 7 rec 8 rec 11 rec 12 rec 14 rec 15 rec 6 rec 10 rec 13

63 Hash Table Uses driver's license record's Internet search engines telephone book databases electronic library catalogs implementing passwords for systems with multiple users. Hash Tables allow for a fast retrieval of the password which corresponds to a given username


Download ppt "HASH TABLE. HASH TABLE a group of people could be arranged in a database like this: Hashing is the transformation of a string of characters into a."

Similar presentations


Ads by Google