METHOD AND SYSTEM FOR EVALUATING RELIABILITY BASED ON ANALYSIS OF USER ACTIVITIES ON SOCIAL MEDIUM
Disclosed are a method and system for evaluating reliability on the basis of an analysis on user activities on a social medium. The method of evaluating reliability of information in a social media service includes calculating an evaluation score of information provided by an information provider on the basis of a social activity of each of a plurality of information consumers relating to the information, and calculating a reputation score of the information provider in a category of the information on the basis of the category and the evaluation score of the information.
1 . A method of evaluating reliability of information in a social media service, the method comprising:
calculating an evaluation score of information provided by an information provider based on a social activity of each of a plurality of information consumers relating to the information; and
calculating a reputation score of the information provider in a category of the information based on the category and the evaluation score of the information.
2 . The method of claim 1 , wherein the evaluation score is determined based on a final implicit evaluation score of the information and a final explicit evaluation score of the information,
the final implicit evaluation score is determined based on a social activity of at least one implicit information consumer who has performed an implicit evaluation on the information as the social activity among the plurality of information consumers, and
the final explicit evaluation score is determined based on a social activity of at least one explicit information consumer who has performed an explicit evaluation on the information as the social activity among the plurality of information consumers.
3 . The method of claim 2 , wherein the social activity of the at least one implicit information consumer includes a positive implicit evaluation or a negative implicit evaluation of the information,
the positive implicit evaluation is classified as an active positive implicit evaluation or a passive positive implicit evaluation in consideration of whether the social activity is active, and
the negative implicit evaluation is classified as an active negative implicit evaluation or a passive negative implicit evaluation in consideration of whether the social activity is active.
4 . The method of claim 2 , wherein the final implicit evaluation score I ct n is calculated according to equations below:
I
ct
n
=
d
+
(
1
-
d
)
·
PI
ct
n
+
NI
ct
n
n
(
I
)
PI
ct
n
=
∑
i
=
1
n
PI
PI
i
ct
n
NI
ct
n
=
∑
i
=
1
n
NI
NI
i
ct
n
〈
Equation
〉
where PI ct n is a sum of all of n PI positive implicit evaluation scores of the information, NI ct n is a sum of n NI all of negative implicit evaluation scores of the information, n(I) is a number of implicit evaluations of the information, and d is a damping coefficient,
the final explicit evaluation score is calculated according to an equation below:
E
ct
n
=
1
n
E
·
∑
i
=
1
n
g
E
i
ct
n
〈
Equation
〉
where n E is a number of explicit evaluations, and E ct n is an average of n E explicit evaluations of the information and has a range E ct n ε[0,1], and
the evaluation score is calculated according to an equation below:
ct n =α·E ct n +β·I ct n <Equation>
where each of α and β is a weight, and a sum of α and β is 1.
5 . The method of claim 4 , wherein the reputation score UR C N is calculated according to an equation below:
UR
C
N
=
1
n
∑
i
=
1
n
ct
i
C
N
·
n
r
n
u
〈
Equation
〉
where n u is a number of users of the social media service, n r is a number of the plurality of information consumers, ct n C N is an evaluation score of each of the information and other information belonging to the category, and n is a number of pieces of the information and the other information.
6 . A system for evaluating reliability of information in a social media service, the system comprising:
a processor configured to calculate an evaluation score of information provided by an information provider based on a social activity of each of a plurality of information consumers relating to the information and calculate a reputation score of the information provider in a category of the information based on the category and the evaluation score of the information.
7 . The system of claim 6 , wherein the evaluation score is determined based on a final implicit evaluation score of the information and a final explicit evaluation score of the information,
the final implicit evaluation score is determined based on a social activity of at least one implicit information consumer who has performed an implicit evaluation on the information as the social activity among the plurality of information consumers, and
the final explicit evaluation score is determined based on a social activity of at least one explicit information consumer who has performed an explicit evaluation on the information as the social activity among the plurality of information consumers.
8 . The system of claim 7 , wherein the social activity of the at least one implicit information consumer includes a positive implicit evaluation or a negative implicit evaluation on the information,
the positive implicit evaluation is classified as an active positive implicit evaluation or a passive positive implicit evaluation in consideration of whether the social activity is active, and
the negative implicit evaluation is classified as an active negative implicit evaluation or a passive negative implicit evaluation in consideration of whether the social activity is active.
9 . The system of claim 7 , wherein the final implicit evaluation score I ct n is calculated according to equations below:
I
ct
n
=
d
+
(
1
-
d
)
·
PI
ct
n
+
NI
ct
n
n
(
I
)
PI
ct
n
=
∑
i
=
1
n
PI
PI
i
ct
n
NI
ct
n
=
∑
i
=
1
n
NI
NI
i
ct
n
〈
Equation
〉
where PI ct n is a sum of all of n PI positive implicit evaluation scores of the information, NI ct n is a sum of all of n NI negative implicit evaluation scores of the information, n(I) is a number of implicit evaluations of the information, and d is a damping coefficient,
the final explicit evaluation score is calculated according to an equation below:
E
ct
n
=
1
n
E
·
∑
i
=
1
n
g
E
i
ct
n
〈
Equation
〉
where n E is a number of explicit evaluations, and E ct n is an average of n E explicit evaluations of the information and has a range E ct n ε[0,1], and
the evaluation score is calculated according to an equation below:
ct n =α·E ct n +β·I ct n <Equation>
where each of α and β is a weight, and a sum of α and β is 1.
10 . The system of claim 9 , wherein the reputation score UR C N is calculated according to an equation below:
UR
C
N
=
1
n
∑
i
=
1
n
ct
i
C
N
·
n
r
n
u
〈
Equation
〉
where n u is a number of users of the social media service, n r is a number of the plurality of information consumers, ct n C N is an evaluation score of each of the information and other information belonging to the category, and n is a number of pieces of the information and the other information.