A.Y.

2021-2022

Dhruv Jain
60004190030
TE COMPS A
Experiment No. 9
Aim: -Implementation of HITS algorithm.

Theory: -
Hyperlink Induced Topic Search (HITS) Algorithm is a Link Analysis Algorithm that
rates webpages, developed by Jon Kleinberg. This algorithm is used to the web link-structures to
discover and rank the webpages relevant for a particular search.
HITS uses hubs and authorities to define a recursive relationship between webpages. Before
understanding the HITS Algorithm.

 Given a query to a Search Engine, the set of highly relevant web pages are called Roots.
They are potential Authorities.
 Pages that are not very relevant but point to pages in the Root are called Hubs. Thus, an
Authority is a page that many hubs link to whereas a Hub is a page that links to many
authorities.

Algorithm
-> Let number of iterations be k.
-> Each node is assigned a Hub score = 1 and an Authority score = 1.
-> Repeat k times:

 Hub update:Each node’s Hub score = \Sigma (Authority score of each node it pointsto).
 Authority update:Each node’s Authority score = \Sigma (Hub score of each node
pointing to it).
 Normalizethe scores by dividing each Hub, Authority score by sum of their individual
values

Code: -
importnumpyasnp

n = int(input("Enter number of pages:- "))

epochs = int(input("Enter number of iterations:- ")) graph =
np.array([[0foriinrange(n)]forjinrange(n)])

print("Enter 1 if link is present 0 otherwise:- \n")

foriinrange(n):
forjinrange(n):
graph[i][j] = int(input(f"{chr(65+i)}-->{chr(65+j)}: "))

auth_score = {} #in lines

hub_score = {} #out lines
#initializationfo
riinrange(n):
auth_score[chr(65+i)] = 1
hub_score[chr(65+i)] = 1

defoutgoing(page,old_auth):
count = 0
temp = 0
foriingraph[page]:
ifi==1:
count+=old_auth[chr(65+temp)]
temp+=1
returncount

defincoming(page,old_hub):
count = 0
temp = 0
foriingraph[:,page]:
ifi==1:
count+=old_hub[chr(65+temp)]
temp+=1
returncount

defnormalize(scores):
total = sum(scores.values())
a = {k: round(v / total,2)fork, vinscores.items()}
returna

foriinrange(epochs):
old_auth = auth_score.copy()
old_hub =
hub_score.copy()forjinrange(
n):
auth_score[chr(65+j)] = incoming(j,old_hub)
hub_score[chr(65+j)] = outgoing(j,old_auth)

print(i+1,")Auth score:- ",auth_score)

auth_score = normalize(auth_score)
print(i+1,")Hub score:- ",hub_score)
print(i+1,")normalized Auth:- ",auth_score)
hub_score = normalize(hub_score)
print(i+1,")normalized hub:-",hub_score,"\n")
 Input Graph: -

Output: -

PS F:\SEM 5\DMW> python .\hitsAlgorithm.py

Enter number of pages:- 8
Enter number of iterations:- 6
Enter 1 if link is present 0 otherwise:-

A-->A:0
A-->B:0
A-->C:0
A-->D:1
A-->E:0
A-->F:0
A-->G:0
A-->H:0
B-->A:0
B-->B:0
B-->C:1
B-->D:0
B-->E:1
B-->F:0
B-->G:0
B-->H:0
C-->A:1
C-->B:0
C-->C:0
C-->D:0
C-->E:0
C-->F:0
C-->G:0
C-->H:0
D-->A:0
D-->B:1
D-->C:1
D-->D:0
D-->E:0
D-->F:0
D-->G:0
D-->H:0
E-->A:0
E-->B:1
E-->C:1
E-->D:1
E-->E:0
E-->F:1
E-->G:0
E-->H:0
F-->A:0
F-->B:0
F-->C:1
F-->D:0
F-->E:0
F-->F:0
F-->G:0
F-->H:1
G-->A:1
G-->B:0
G-->C:1
G-->D:0
G-->E:0
G-->F:0
G-->G:0
G-->H:0
H-->A:1
H-->B:0
H-->C:0
H-->D:0
H-->E:0
H-->F:0
H-->G:0
H-->H:0
1 )Auth score:- {'A': 3, 'B': 2, 'C': 5, 'D': 2, 'E': 1, 'F': 1, 'G': 0, 'H': 1}
1 )Hub score:- {'A': 1, 'B': 2, 'C': 1, 'D': 2, 'E': 4, 'F': 2, 'G': 2, 'H': 1}
1 )normalized Auth:- {'A': 0.2, 'B': 0.13, 'C': 0.33, 'D': 0.13, 'E': 0.07, 'F': 0.07, 'G': 0.0, 'H':
0.07}
1 )normalized hub:- {'A': 0.07, 'B': 0.13, 'C': 0.07, 'D': 0.13, 'E': 0.27, 'F': 0.13, 'G': 0.13, 'H':
0.07}

2 )Auth score:- {'A': 0.27, 'B': 0.4, 'C': 0.79, 'D': 0.34, 'E': 0.13, 'F': 0.27, 'G': 0, 'H': 0.13}
2 )Hub score:- {'A': 0.13, 'B': 0.4, 'C': 0.2, 'D': 0.46, 'E': 0.6600000000000001, 'F': 0.4, 'G':
0.53, 'H': 0.2}
2 )normalized Auth:- {'A': 0.12, 'B': 0.17, 'C': 0.34, 'D': 0.15, 'E': 0.06, 'F': 0.12, 'G': 0.0, 'H':
0.06}
2 )normalized hub:- {'A': 0.04, 'B': 0.13, 'C': 0.07, 'D': 0.15, 'E': 0.22, 'F': 0.13, 'G': 0.18, 'H':
0.07}

3 )Auth score:- {'A': 0.32, 'B': 0.37, 'C': 0.81, 'D': 0.26, 'E': 0.13, 'F': 0.22, 'G': 0, 'H':0.13}
3 )Hub score:- {'A': 0.15, 'B': 0.4, 'C': 0.12, 'D': 0.51, 'E': 0.78, 'F': 0.4, 'G': 0.46, 'H':0.12}
3 )normalized Auth:- {'A': 0.14, 'B': 0.17, 'C': 0.36, 'D': 0.12, 'E': 0.06, 'F': 0.1, 'G': 0.0, 'H':
0.06}
3 )normalized hub:- {'A': 0.05, 'B': 0.14, 'C': 0.04, 'D': 0.17, 'E': 0.27, 'F': 0.14, 'G': 0.16, 'H':
0.04}

4 )Auth score:- {'A': 0.24000000000000002, 'B': 0.44000000000000006, 'C':

0.8800000000000001, 'D': 0.32, 'E': 0.14, 'F': 0.27, 'G':
0, 'H': 0.14}
4 )Hub score:- {'A': 0.12, 'B': 0.42, 'C': 0.14, 'D': 0.53, 'E': 0.75, 'F': 0.42, 'G': 0.5, 'H': 0.14}
4 )normalized Auth:- {'A': 0.1, 'B': 0.18, 'C': 0.36, 'D': 0.13, 'E': 0.06, 'F': 0.11, 'G': 0.0, 'H':
0.06}
4 )normalized hub:- {'A': 0.04, 'B': 0.14, 'C': 0.05, 'D': 0.18, 'E': 0.25, 'F': 0.14, 'G': 0.17, 'H':
0.05}

5 )Auth score:- {'A': 0.27, 'B': 0.43, 'C': 0.8800000000000001, 'D': 0.29, 'E': 0.14, 'F': 0.25,
'G': 0, 'H': 0.14}
5 )Hub score:- {'A': 0.13, 'B': 0.42, 'C': 0.1, 'D': 0.54, 'E': 0.78, 'F': 0.42, 'G':
0.45999999999999996, 'H': 0.1}
5 )normalized Auth:- {'A': 0.11, 'B': 0.18, 'C': 0.37, 'D': 0.12, 'E': 0.06, 'F': 0.1, 'G': 0.0, 'H':
0.06}
5 )normalized hub:- {'A': 0.04, 'B': 0.14, 'C': 0.03, 'D': 0.18, 'E': 0.26, 'F': 0.14, 'G': 0.16, 'H':
0.03}
 6 )Auth score:- {'A': 0.22, 'B': 0.44, 'C': 0.8800000000000001, 'D': 0.3, 'E': 0.14, 'F': 0.26,
'G': 0, 'H': 0.14}
6 )Hub score:- {'A': 0.12, 'B': 0.43, 'C': 0.11, 'D': 0.55, 'E': 0.77, 'F': 0.43, 'G': 0.48, 'H': 0.11}
6 )normalized Auth:- {'A': 0.09, 'B': 0.18, 'C': 0.37, 'D': 0.13, 'E': 0.06, 'F': 0.11, 'G': 0.0, 'H':
0.06}
6 )normalized hub:- {'A': 0.04, 'B': 0.14, 'C': 0.04, 'D': 0.18, 'E': 0.26, 'F': 0.14, 'G': 0.16, 'H':
0.04}

Conclusion: -Hyperlink Induced Topic Search (HITS) is an algorithm used in link analysis. It
could discover and rank the webpages relevant for a particular search. The idea of this algorithm
originated from the fact that an ideal website should link to other relevant sites and being linked
by other important sites.