Available via license: CC BY 4.0
Content may be subject to copyright.
Applied Mathematics and Nonlinear Sciences (aop) (aop)
Applied Mathematics and Nonlinear Sciences
https://www.sciendo.com
†Corresponding author.
Email address: m15282101467@163.com
ISSN 2444-8656
https://doi.org/10.2478/amns.2023.2.00931
© 2023 Zheng Qiu, published by Sciendo.
This work is licensed under the Creative Commons Attribution alone 4.0 License.
Research on the ideologic and politic education of college students based on big data
analysis
Zheng Qiu1,†
1. School of Marxism, Neijiang Normal University, Neijiang, Sichuan, 641000, China.
Submission Info
Communicated by Z. Sabir
Received November 29, 2022
Accepted May 12, 2023
Available online November 1, 2023
Abstract
This paper firstly researches the ideological and political education in higher education institutions in the background of
the big data era and puts forward the innovation strategy of ideological and political education mode in higher education
institutions in the big data era. Secondly, it studies the application of data mining technology in colleges and universities
for ideological and political education. Based on data mining technology, the data on ideological and political education
is determined by the clustering algorithm, followed by a quantitative examination of the policy text of ideological and
political education of colleges and universities by calculating the quantitative statistical analysis of policy text, word
frequency, and word cloud analysis. Then, the research subjects were selected, and the initial data were obtained using
questionnaires, further pre-processing, and further empirical analysis of ideological and political education based on data
mining technology. The consequences show that the proportions contained in the three types of score bands, namely
17.99%, 59.38% and 9.96%, basically match with the sample proportions accounted for by the three types of grades
assessed by clustering, namely 30.51%, 62.02% and 8.47%, which confirms that such a model of data mining adopted
according to the quantitative table of ideologic and politic education assessment in higher education institution is a very
successful model. This study contributes to the innovation of online ideological and political education work in higher
education institutions in the context of big data.
Keywords: Data mining techniques; Clustering algorithms; Universities; Ideological and political education; Policy
texts.
AMS 2010 codes: 97Q70
Applied Mathematics and Nonlinear Sciences (aop) (aop)
1 Introduction
With the importance of big data technology in the world and the new requirements of the Party Central
Committee for ideological and political education in higher education institutions in the new era, big
data technology has been applied to the teaching management of ideological and political education
in higher education institution [1-2]. Traditional ideological and political education in higher
education institutions is mainly reflected in the classroom teaching of ideological and political
education, and many teachers of ideological and political education have old teaching contents, single
teaching methods, dull and lack of vitality in teaching form, and most of them teach students in “duck-
filling” and dogmatic way [3-4]. All these factors have led to the fact that most students in higher
education institutions do not attach much importance to the Civics course, and many college students
think that learning Civics course is useless, their political ideology is gradually fading, and some
students have a lack of ideals and beliefs [5-6]. Therefore, it is urgent to enhance the effectiveness of
Civic Education, and it becomes especially important to innovate the working methods of Civic
Education in higher education institutions [7]. Teachers can use multimedia teaching tools to make
theoretical knowledge in the form of animation or video to make ideological and political theoretical
knowledge more concrete and three-dimensional so that the ideological and political classroom
becomes more vivid and lively, thus strengthening students’ memory of theoretical knowledge and
effectively improving the overall efficiency of ideologic and politic education [8-9]. The necessity
and possibility of ideological and political education innovation in higher education institutions in
the era of big data have planned a clear path for ideological and political education innovation so that
educators can effectively comprehend the far-reaching significance of big data for ideological and
political education in a higher education institution and achieve high-quality development of
ideologic and politic education [10].
In this paper, we first study the ideology and politics of education data in higher education institutions,
then continuously innovate the ideology and politics of education model in higher education
institutions. The education model innovation is specifically reflected in optimizing the big data
system, improving the ability of big data applications, establishing the education system, and
highlighting students' development. Next, the ideologic and political education data types in the
clustering process are determined, and these data types are converted into a dissimilarity matrix,
followed by the use of simple matching correlation coefficients and dissimilarity matrix to describe
the dissimilarity between ideological and political education object
i
and ideologic and politic
education object
j
. Then, the policy texts, word frequencies, and word clouds were analyzed using
clustering algorithms, and the word frequencies and weights of keywords illustrated the subjects of
ideological and politic education policies in higher education institutions, which were used to confirm
the development of ideological and politic education policies in higher education institution. Finally,
the students’ ideological and political education performance is separated into five types of grades,
such as “excellent, good qualified, poor and poor”, and the correlation analysis between the
ideological and political quality of students and their growth environment and growth status confirms
the promotion effect of this research on ideologic and politic education through ideologic and politic
education clustering analysis.
2 Literature Review
The literature [11] argues that with the rapid development of big data technology, its application in
education is also receiving increasing attention. Although ideological and political education is a
significant component of higher education in China, it is urgently necessary to figure out how to use
big data technology to maximize these fields of study. The authors study and analyze the use of big
data technology in ideological and political education. The study's findings demonstrate that big data
Applied Mathematics and Nonlinear Sciences (aop) (aop)
technology may deliver more precise and individualized services for ideology and political education,
actualize all-encompassing and multi-faceted assessment and management, and foster students’
ability to think critically and innovatively on ideological and political issues. The literature [12]
argues that ideological and political education in higher education institutions is one of the significant
ways to cultivate students' ideological, moral character, and personality qualities. Showing a positive
and healthy personality in students is one of the key subjects in this process. The authors analyzed
and discussed the importance of personality shaping in ideological and political education in higher
education institutions. The research results show that ideological and political education in higher
education institutions should focus on cultivating students’ moral consciousness and social
responsibility, strengthening students’ self-awareness and psychological quality, and promoting
students’ self-improvement and development. The literature [13] points out that the representativeness
of political party members and supporters has become a source of concern after the realignment of
major political parties in Finland. The authors compared the differences in the social and ideological
representativeness of political party members and supporters through surveys and analyses. The
findings indicate some discrepancies between the representativeness of party members and supporters,
but most parties can maintain relative social and ideological representativeness.
The literature [14] states that ideological extremism has significantly impacted political participation
in Japan. The authors explored the connection between ideology extremism and political participation
by reviewing and analyzing literature related to the two in Japan. The findings suggest that ideological
extremism can negatively affect political participation, limiting people's free expression and
participation and undermining political stability and social harmony. Literature [15] argued that
ideological and political education in higher education should focus on the guidance and practice of
faith education to guide students to create the correct outlook on life, values and worldview. The
writers propose several reform initiatives and recommendations by reviewing and analyzing literature
related to the reform of ideological and political education in higher education institutions. The
research results show that ideological and political education in higher education institutions should
focus on cultivating students’ faith education, ideological and moral education and practical education
to encourage students’ comprehensive qualities of positivity, social accountability and innovation.
The literature [16] argues that cognitive neuroscience, an emerging interdisciplinary research field,
can deliver new perspectives and methods for studying ideological and political education in higher
education institutions. By reviewing and analyzing the literature related to the ideological and
political education of college students, the authors proposed a research method for ideological and
political education based on cognitive neuroscience. The findings of the research demonstrate that
the cognitive neuroscience-based research methodology can investigate the mental procedures and
neural system of students’ ideological and political education using scientific tools like functional
magnetic resonance imaging and brain waves to fully comprehend the development and creation of
students’ ideological and ethical behaviors, values, and beliefs.
3 Ideologic and politic education in higher education institution under the background of
big data era
3.1 Ideologic and politic education data in universities
3.1.1 Data sources of ideologic and politic education in universities
The data sources of ideologic and politic education in higher education institution are shown in Figure
1. Ideologic and politic education in higher education institution widely uses the Internet in
curriculum teaching, daily management, campus cultural activities, social practice and
communication, which has formed a large quantity of data. The amount of macro, micro, structural
Applied Mathematics and Nonlinear Sciences (aop) (aop)
and non-structural information in these data makes data collection challenging. To evaluate the data,
it is necessary to filter it, keep usable data, and create appropriate correlations and aggregations.
Ideology Political
Education Linked Data
Content of
ideological and
political education
The field of
ideological and
political education
Communication
Information
Communication carrier
of ideological and
political education
Figure 1. Ideologic and politic education data sources in higher education institution
3.1.2 Display of ideologic and politic education data in higher education institution
Many universities form a huge amount of ideologic and politic education data, scattered in different
places and departments, becoming a treasure trove to be discovered for students’ ideological and
political education. However, in the face of a large amount of data, how to utilize it to demonstrate
its value becomes a new problem. The data associated with ideological and political education mainly
includes the following parts.
1) Ideological and political education course data, including attendance, participation,
assignments, grades, and other information. These data reflect the students' course learning
status.
2) Information about the participation of students in party and caucus training. This includes the
number of students participating in caucus training, the number of training sessions, training
results, etc., thus forming the data of students’ party and caucus education.
3) Data on students' participation in practical activities outside ideological and political courses.
For example, students’ participation in extra-curricular practical activities or volunteer
activities is converted into certain quality education points to form quality education data,
which is convenient for understanding students’ participation in extra-curricular practical
activities.
4) Information about the participation of students in online communities, social platform
interactions, and game activities. These online activities can generate data on the number of
occurrences, frequency of occurrences, and activity, which is convenient for understanding
the status of students’ participation in online social activities and analyzing students’
preferences and participation.
Applied Mathematics and Nonlinear Sciences (aop) (aop)
5) Information regarding the involvement of students in online communities, social platforms,
and games. These online activities can generate data such as the number of appearances,
frequency of appearances, and activity, which make it simple to determine the level of pupils’
engagement in online social events and examine their preferences.
The preceding data must be screened, combined, and rebuilt to be shown; this creates the constantly
evolving data model of ideologic and political education depicted in Figure 2.
Data usage
Data source
The data shows
Data prediction and intervention Knowing, feeling, thinking and acting
Student ideological and political education data
composition
Learning Data Activity Data Community Data Life Data
Academic
Performance Number of
participations
Frequency of
participation
Number of
participation
Duration
Dormitory Star
Rating
Bedtime
Figure 2. Ideologic and politic education data and the path to analysis
3.2 Innovation of ideologic and politic education mode in higher education institution in the
era of big data
3.2.1 Concepts and characteristics of the big data era
About people’s ability to access information and data, big data also introduces new changes in many
domains, such as people’s lives and learning, and the alterations of big data bring beneficial and
positive benefits to people. The big data model is an innovative product created using the new process
model of the new era. The use of information assets in people’s production lives presents various
challenges and opportunities in many sectors, and the growth of data sources also plays a vital role in
various industries. Information and data are characterized by their variety, richness, and accuracy.
Colleges are now playing a crucial role in delivering education in the big data era.
3.2.2 Innovation of ideologic and politic education mode in higher education institution
The step into the era of big data has gradually changed the system of social existence. While
innovating the height of ideological and political education in higher education institutions, it has also
created a precedent for the educational theory system in the new era. Based on this phenomenon,
many network managers have started combining big data management with ideological and political
education to explore new educational models. Through this change, colleges and universities'
ideologic and political education modes are constantly innovated, and good results are achieved. The
innovative ideology and political education modes of colleges and universities are depicted in Figure
3.
1) Optimize big data system.
Applied Mathematics and Nonlinear Sciences (aop) (aop)
There are many ways to address problems with big data, starting with altering the traditional modes
of instruction and learning and expanding fundamentally. Given that big data is a kind of
comprehensive evaluation index, it has two characteristics: dynamic and a process, mainly combined
with the feedback situation of students’ comprehensive ability to assume to judge. In this way, the
main goal of ideological and political education in higher education institutions can be more
systematically highlighted. Thus, creating an ideological and political education system in college
that addresses the above points is especially significant.
2) Advance the ability of big data application
In the big data era, developing multidimensional abilities that combine theoretical understanding with
fundamental professional skills is imperative. Activities related to education and training must be
carried out with great enthusiasm. The application abilities of big data technologies can be enhanced,
and the useful information gathered by students can be arranged and documented comprehensively,
according to training in statistics, Internet mechanisms, and other areas. Furthermore, the use of
network platforms is prevalent in the field of big data, and when applied skillfully, it may produce
twice as much output with a third of the effort.
3) Establishing education system
Establishing education schemes and reflecting ideology and politicalal thoughts is one of the effective
strategies for innovating ideological and political education modes in higher education institutions in
the big data era. First of all, building an information data gathering system is required to guarantee
that students’ privacy rights are not violated, and educating students on how to use cutting -edge
technologies like the Internet to gather data and information by students’ developmental needs is also
vital, to facilitate the process of promoting ideologic and politic education in a higher education
institution. Strictly by the standards required by the school, accurate records of student attendance
and extracurricular activities, continuous modification and improvement of student participation in
data collection information, to ensure that the school improves the construction of education, but also
the innovation of the ideologic and politic education in higher education institution will be more
smoothly carried out, and ultimately achieve a comprehensive realization of the situation of political
education invention.
4) Highlighting students’ development
College students have to go through ten years of hard study, and they successfully enter the gate of
colleges and universities with the vigor of more and more battles and the tenacity of insisting on not
giving up. They have a certain understanding of China's scientific and technological development,
political culture, and human geography and are the inheritors and transmitters of the new era. A few
days ago, college students facing social experiences still faced certain challenges, and it is more likely
that they will choose the wrong decompression behavior. Therefore, how to self-understand, improve
themselves, and solve unfair matters in society with optimistic thinking is the root of the problem.
Applied Mathematics and Nonlinear Sciences (aop) (aop)
Optimize big data
system Dynamicity
Improve the ability to
apply big data Possess theoretical
knowledge
Enhance the
application skills of
big data technology
Establishing an
education system
Building an
information data
collection system
Improve student
engagement data
Highlighting
students' own
development
Highlighting
students' own
development
The tenacious spirit
of perseverance and
not giving up
Process
Figure 3. Innovation of ideologic and politic education model in higher education institution
4 Application of data mining technology in the ideologic and politic education of colleges
and universities
4.1 Data mining techniques
4.1.1 Data types in the clustering process
For all the
n
objects,
p
variables were designated to define the “ideology and politic education”
object, like teaching methods, teaching efficiency, teaching quality, teaching tools, teaching principles,
and other attributes. These data types were presented in the form of relational tables after being
collected slowly on an interval scale, i.e.
np
matrix:
11 12 1
21 22 2
12
p
p
n n np
x x x
x x x
x x x
(1)
The similarity matrix is applied to place the similarity in two of
n
ideologic and politic education
objects, and its particular manifestation can be characterized as a
nn
matrix:
Applied Mathematics and Nonlinear Sciences (aop) (aop)
0
(2,1) 0
(3,1) (3,2) 0
( ,1) ( ,2) 0
d
dd
d n d n
(2)
Where
( , )d i j
is the exact quantitative procedure of the dissimilarity between two ideologic and
politic education items
i
and
j
, it is typically a number that is higher than or equal to zero, when
two ideologic and politic education
i
and
j
are tremendously similar,
( , )d i j
tends to 0: conversely,
when
i
and
j
are different, then the worth of
( , )d i j
tends to be large, you can get
( , ) ( , )d i j d j i=
,
and also get
( , ) 0d i i =
.
The difference in distance among each group of ideologic and politic schooling objects, which is the
most traditional method of calculating the distance, is used to measure the dissimilarity (or similarity)
among ideologic and politic education objects defined by the interval scale variable. The researcher
defines the Euclidean distance as follows:
2 2 2
1 1 2 2
( , ) i j i j in jh
d i j x x x x x x= − + − + + −
(3)
( )
12
, , ,
i i in
i x x x=
as well as
( )
12
,,
j j jn
j x x x=
are
n
- dimensional data ideologic and politic
education objects.
Another very classical distance metric is the Manhattan (or city block) distance, which is defined by
the researcher as summarized as
1 1 2 2
( , ) i j i j in jn
d i j x x x x x x= − + − + + −
(4)
The Manhattan distance and the Euclidean distance both meet the requirements of:
1)
( , ) 0d i j
: Both distances are non-negative values:
2)
( , ) 0d i j =
: The detachment between itself is 0:
3)
( , ) ( , )d i j d j i=
: Symmetry exists in the function connecting the two distances:
4)
( , ) ( , ) ( , )d i j d i h d h j+
: The distance from the ideologic and politic education object
i
to the
ideologic and politic education object
j
directly is less than or equal to the distance from the
other ideologic and politic education object
k
to the ideologic and politic education object
j
.
The generalization of the first two distances is the Minkowski distance. The researcher offers the
following succinct definition of it:
1
1
( , ) np
p
i j ik jk
pk
d i j x x x x
=
= − = −
(5)
Applied Mathematics and Nonlinear Sciences (aop) (aop)
This distance is a generalization of an infinite distance measure, if
p
is occupied as
[1, )p
, then
( , )d i j
is the Manhattan distance, and if
p
is occupied as 2, then
( , )d i j
is the Euclidean distance.
4.1.2 Clustering algorithm
For example, the attribute “quality” is a symmetric binary variable with good and bad states. The
degree of dissimilarity measured by a symmetric binary variable for the study of political education
objects is called symmetric binary dissimilarity. The straightforward matching correlation coefficient
is frequently used by researchers to characterize the disparity between ideologic and politic education
object
i
and ideologic and politic education object
j
, which can be distinct as the following formula:
( , ) ( ) / ( )d i j r s q r s t= + + + +
(6)
The number of negative matches is often ignored as a non-significant coefficient, and the Jaccard
coefficient is the most traditional evaluation coefficient to gauge the disparity of asymmetric binary
variables.
( , ) ( ) / ( )d i j r s q r s= + + +
(7)
For example, the record table of college student attributes includes attributes name, gender, over age,
over weight, project-1, project-2, project-3, project-4, where name is the ideologic and politic
education object identification mark, the other qualities are all asymmetrical binary variables, while
gender is a symmetrical binary variable. The relationship of the binary attributes to describe the
attribute records of college students is exposed in Table 1.
Table 1. Ideological and political education
name
gender
over age
over weight
project-1
project-2
project-3
project-4
Green
M
Y
N
P
N
N
N
Mick
F
Y
N
P
N
P
N
Robert
M
Y
Y
N
N
N
N
…
…
…
…
…
…
…
…
The values
Y
and
P
of the asymmetric attributes in the table are set to 1 and the value
N
is set to 0.
If only the asymmetric variables are used to measure the disparity between the ideologic and politic
education targets (personnel situations). Then the Jaccard coefficient formula (7) can be used to
measure the dissimilarity between the three company personnel Green, Mick and Robert two by two
as shown in formula (8), (9) and (10):
01
( , ) 0.33
2 0 1
d Green Mick +
==
++
(8)
11
( , ) 0.67
111
d Green Rober +
==
++
(9)
12
( , ) 0.75
1 1 2
d Robert Mick +
==
++
(10)
Applied Mathematics and Nonlinear Sciences (aop) (aop)
The values
i
and
j
of the asymmetric attributes in the table are set to 1 and the value 3 is set to 0. If
only the asymmetric variables are used to measure the disparity between the ideologic and politic
education targets (personnel situations). Then the Jaccard coefficient formula (7) can be used to
measure the dissimilarity between the three company personnel Green, Mick and Robert two by two
as shown in formula (8), (9) and (10):
( , ) pm
d i j p
−
=
(11)
Where
m
is the amount of matches and
p
is the quantity of all variables. It is often necessary to
radiate the values of all variables into the
[0.0,1.0]
region space, in order to make the weights of all
variables of the same size, which can be implemented using
kf
z
instead of
kf
r
, where
1
1
if
if
f
r
zM
−
=−
(12)
If there is
p
varied type variable, we put the dissimilarity between ideologic and politic education
objects
i
and
j
can be calculated using the formula below:
()
1
1
()
( , ) ()
pf
jf jf
f
p
ij
f
fd
d i j f
=
=
=
(13)
Where the pointer term
()
if f
takes the value of 0 if there is no measure of
f
between ideologic and
politic education objects
i
and
j
, then
0
if jf
xx==
, and when variable
f
is not a binary symmetric
variable; conversely, the term
()
if f
takes the value of 1.
The researcher used the following four categories of inter-cluster distance calculations to measure
this clustering:
Minimum distance:
( )
min ,
, min ij
i j p c p c
d c c p p
= −
(14)
Maximum distance:
( )
max ,
, max ij
i j p c p c
d c c p p
=−
(15)
Mean distance:
( )
mean ,
i j i j
d c c m m=−
(16)
Average distance:
( )
1
,
ij
avg i j p c p c
ij
d c c p p
nn
=−
(17)
Where
PP−
represents the size of the detachment between two ideological education
P
and
P
,
where
i
m
represents the number of mean values of cluster
i
c
, and
i
n
represents the number of data
ideological education objects in cluster
i
c
. The probability case represents the conceptual probability
as well as the conditional probability in the form of
( )
/
ij k
P Ai V C=
, Where
i ij
AV=
refers to a one-
attribute one-value pair and
k
C
refers to the conceptual class. On a particular layer of such a tree, a
Applied Mathematics and Nonlinear Sciences (aop) (aop)
new split of ideology training nodes is produced. The classification efficiency is calculated by the
algorithm using heuristic estimation to direct the construction of the classification tree. The efficiency
of classification can be summed up as follows:
( )
( )
22
1
1nij
k i i ij
k i j i j
k
v
p c p A p A v
nc
=
= − =
(18)
Where
n
indicates the amount of nodes, ideas or “kinds” in a new division
12
, , , n
C C C
generated
at some level of the classification tree. Similarity within classes and dissimilarity between classes are
given as feedback for classification effectiveness:
4.2 Quantitative analysis of ideologic and politic education policy texts in universities
4.2.1 Statistical analysis of the number of policy texts
Universities' ideological and political education was employed to examine how political and
ideological education programs in colleges and universities developed. The annual number of
ideologic and political education policies promulgated in higher education institutions is shown in
Figure 4, and the trend of annual changes can be seen. The quantitative changes in ideological and
political education policies in higher education institutions show a consistency in the fluctuation of
the number of ideological and political education policy studies. Before 2010, the number of literature
on ideologic and political education policy research was small but generally showed a trend of
fluctuating growth, with the highest number of literature in 2017 and fluctuations after that, reflecting
the value of cutting-edge research in the field being explored continuously. As a subdivision of
ideological and political education policy, the change in the number of policies is consistent with the
number of literature in the whole research field, which is determined by the systematic nature of
ideological and political education policy and also shows from the side that ideologic and politic
education policy is closely related to ideologic and politic education practice. Since the ideologic and
political education policy in higher education institutions is an emerging subdivision with less
literature, it is expected that in the future, more and more individuals will be interested in this
subdivision.
Figure 4. Number of ideologic and politic education policies in higher education institution
Applied Mathematics and Nonlinear Sciences (aop) (aop)
4.2.2 Word frequency and word cloud analysis
After the text of online ideological and politic education policy was word-separated, and the word
density analysis of the text that has been divided into words, the word frequency and weight of online
ideological and political education policy are exposed in Table 2. The higher the word frequency, the
more it appears in the network ideologic and political education policy and the weight's size reflects
the keyword's importance in the network ideologic and political education policy. It can be gotten
that “network” is the keyword with the highest word frequency in the text, reaching 831 with a weight
of 4.6606, “ideology” with a word frequency of 570 and a weight of 4.3997, and “university “The
word frequency and weight of keywords such as “Ministry of Education” and “Communist Youth
League” indicate the main body of online ideologic and politic education policies, and the results of
word frequency analysis can show that The consequences of the word frequency examination can
show that the main body of policy formulation and policy implementation is closely related to politics,
propaganda and culture. Word cloud can be used to analyze further online ideologic and political
education policy keywords.
Table 2. Online ideologic and politic education policy word frequencies and weights
No.
Keywords
Word
frequency
Weights
No.
Keywords
Word
frequency
Weights
1
Network
832
4.6605
15
Advocacy
152
3.4822
2
Ideas
569
4.3996
16
Government
145
3.4592
3
Colleges and
Universities
568
4.3996
17
Ideological and Political
Work
141
3.4144
4
Education
527
4.3442
18
Online
132
3.3846
5
Politics
427
4.2014
19
Development
128
3.3738
6
Construction
379
4.1157
20
Ministry of Education
129
3.3738
7
Strengthening
358
4.0845
21
Communist Youth League
117
3.3127
8
Services
359
4.0844
22
Guidance
114
3.2828
9
Students
339
4.0382
23
China
114
3.2828
10
Management
258
3.8359
24
Internet
111
3.3185
11
Campus
182
3.8245
25
Team
105
3.3948
12
Department
177
3.7353
26
Leadership
105
3.2258
13
Culture
168
3.6245
27
School
96
3.2057
14
Organization
161
3.5304
28
Advancing
97
3.1852
5 Empirical analysis of ideologic and politic education based on data mining technology
5.1 Ideologic and politic education Cluster Analysis
5.1.1 Data pre-processing
“The quantitative table of ideologic and political education in higher education institutions requires
the teachers to “adhere to the standard, reward and punish clearly, and treat each student objectively
and fairly. In the items such as “accurately grasp the situation of poor students, earnestly do the work
of diligent work and student loans”, according to the students’ ideologic and political education
performance, they are divided into five types of grades, such as “excellent, good, qualified, poor,
poor”. We have classified students’ ideologic and political education performance according to four
Applied Mathematics and Nonlinear Sciences (aop) (aop)
major aspects, namely, “ideologic and political education attitude”, “ideologic and political education
ability”, “ideologic and politic education method,” and “ideologic and political education effect,”.
We reorganize and merge the data in the quantitative evaluation table of ideologic and political
education in higher education institutions according to four attributes, namely “ideologic and political
education attitude”, “ideologic and political education ability”, “ideologic and politic education
method,” and “ideologic and politic education effect”. The five levels of assessment ratings,
“excellent, good, qualified, poor, poor,” are arranged in a special order. These ratings are mapped to
a range of regions to assign the same weight to all variables. According to equation (12), the five
values of the appraisal ratings were calculated as “1, 0.75, 0.5, 0.25, 0”. Table 3 lists the data samples
that will be used in the analysis process using the pre-processed data sample details procedure.
Table 3. Sample data for clustering
Ideologic and politic
education attitude
Ideologic and politic
education ability
Ideologic and politic
education methods
Efficiency of ideological
and political education
0.64
0.61
0.55
0.57
0.64
0.61
0.55
0.57
0.34
0.34
0.32
0.34
0.77
0.75
0.69
0.76
0.81
0.81
0.81
0.82
We divided the data samples in the top three in the starting position as the vital case of clustering by
the parameters of the clustering algorithm, and we then took further steps to enhance the clustering
algorithm following arriving at the samples that can serve each class to replace the first three samples
of data that have been identified as the center of grouping at the beginning, to prevent the skew of the
data and the number of errors.
Table 4. Sample data representing 3 levels
Type
Ideologic and
politic education
attitude
Ideologic and
politic education
ability
Ideologic and
politic education
methods
Efficiency of
ideological and
political education
Indicates a better situation
0.751
0.751
0.751
0.751
Indicates a medium situation
0.52
0.52
0.52
0.52
Indicates poor condition
0.253
0.253
0.253
0.253
5.1.2 Analysis of clustering results
This study, 122 sample data were analyzed using a clustering method, including 118 sample data from
the quantitative job assessment form and 3 standard samples that represented better, moderate, and
poorer data, respectively. The sample data was divided into four categories of attributes: attitude
toward political and ideological education, aptitude for ideological and political education, the
delivery method for ideological and political education, and the impact of political and ideological
education. These characteristics were clustered by data mining, with the initial K value set at 3. Figure
5 displays the final cluster evaluation findings from the mining process. The data items included in
each cluster’s final proportionate distribution fall within the following range: the first cluster (better),
with a total of 37 data samples, deleting a pre-defined standard sample, leaving 36 data samples,
accounting for 36/118=30.51%. Cluster 2 (medium), with a total of 73 samples, deleting a standard
sample, leaving 72, or 72/118 = 62.02%. The third cluster (poor), with a total of 11 data samples,
deleted a pre-defined standard sample data, leaving 10 data samples, accounting for 10/118 = 8.47%.
Applied Mathematics and Nonlinear Sciences (aop) (aop)
Figure 5. Results of cluster analysis of ideological and political education
We once more requested via the Office of Student Work the ideological and political education of the
247 students in the class of 2022 at the School of Marxism, as well as the pertinent comprehensive
quantitative scores obtained in the eight activities sponsored by the School to validate the data
mining’s final findings further. The total score was calculated as 100 points, and we then classified
these data samples according to the criteria of 0 to 100 points, which can be classified into three rating
levels: 81 points or more, 61 to 80 points (including 61 and 80 points) and less than 60 points. The
fallouts of the comprehensive quantitative score situation examination are compared, as shown in
Table 5. The percentage of branches among each category of levels has increased when comparing
the two groups of levels (above 81, 61 to 80 (which includes 61 and 80), and below 60) with the
ranges used for clustering ratings (0.751, 0.52, 0.253). According to the combined analysis above, the
proportion of branches in the three score bands is 17.99%, 59.38%, and 9.96%. It is confirmed that
the data mining model used to manage and educate about ideologies and politics in higher education
institutions is very effective. It brings reference and guiding significance to the management and
schooling of these activities. The model was adopted by the quantitative table for evaluating
ideological and political education in higher education institutions.
Table 5. Comprehensive quantitative score situation analysis results
Score band
Sample size
Corresponding percentage
Above 81 points
511
17.99%
61 to 80 points
1687
59.38%
Below 60
283
9.96%
5.2 Analysis of the correlation between students’ ideological and political quality and their
growth environment and growth status
5.2.1 Analysis of the correlation between the ideological and political quality of college
students and their growth environment
College students' political and ideological views and the quality of their learning environment are
significantly positively correlated. The correlation study between college students' political and
ideological views and their learning environment is displayed in Table 6. The values of the product
Applied Mathematics and Nonlinear Sciences (aop) (aop)
distinction correlation coefficient (Pearson correlation coefficient, also known as the product
difference connection coefficient, which can be used to measure the linear connection between fixed
distance variables) for the ideological and partisan quality and growth environments of college
students, the absolute value of the correlation coefficient represents the degree of correlation
r
, the
value of correlation coefficient ranges between -1 and +1, that is,
11r−
, the coefficient of
correlation is stronger when the absolute value of the correlation value is higher. The correlation
coefficient (the stronger the relationship, the higher the value in absolute terms, the closer the
correlation value is to 1 or -1, which indicates the correlation, and the weaker the connection, the
closer the relationship coefficient is to 0), which is 0.697, shows an important positive correlation.
The companion probability’s p-value, however, is 0.00, much less than the significance limit of 0.01,
further indicating that the two are highly positively and linearly correlated. There are significant
correlations between college students’ growth environment and college students’ political outlook
(r=0.611, sig=0.00), ideological awareness (r=0.635, sig=0.00), moral outlook (r=0.587, sig=0.00),
and legal outlook (r=0.628, sig=0.00). The most important thing for ideological and political
education in higher education institutions is that we must protect the main channel and main position
taken by ideological and political theory courses, enhance the educational effect, optimize the campus
developing environment, maximize the network environment, and foster a welcoming group
environment, and assist students in understanding and analyzing social issues objectively, thoroughly,
rationally, and correctly.
Table 6. Correlation between growth environment and ideological and political quality
Projects
Political
View
Thoughtfulness
Morality
Legal
View
Mental
Health
Ideological and political
qualities
Mental Health
Pearson
.332**
-.463**
-.352**
-.322**
1
-.413**
Salience
.000
.000
.000
.000
.000
N
162
162
162
162
162
162
Home
Environment
Pearson
.399**
.437**
.461**
.401**
-.315**
481**
Salience
.000
.000
.000
.000
.000
.000
N
162
162
162
162
162
162
School
Environment
Pearson
.585**
.602**
565**
651**
-.327**
682**
Salience
.000
.000
.000
.000
.000
.000
N
162
162
162
162
162
162
Social
Environment
Pearson
.337**
.312**
.171*
361**
-.145.
336**
Salience
.000
.000
.032
.000
0.63
.000
N
162
162
162
162
162
162
Media
Environment
Pearson
.447**
.527**
534**
.435**
-.386**
548**
Salience
.000
.000
.000
.000
.000
.000
N
162
162
162
162
162
162
Youkun
Environment
Pearson
.455**
.455**
.466**
.431**
-.416**
.512**
Salience
.000
.000
000
.000
.000
.000
N
162
162
162
162
162
162
Growth
Environment
Pearson
.611**
.635**
.587**
.627**
-.425**
.697**
Salience
.000
.000
.000
.000
.000
.000
N
162
162
162
162
162
165
Applied Mathematics and Nonlinear Sciences (aop) (aop)
5.2.2 Analysis of the correlation between the ideological and political quality of college
students and their growth status
College students' present-day ideological and political climate is significantly positively correlated
with their development. Table 7 displays the results of the correlation study between the ideological
and political levels of college students and the current state of growth. There is a substantial positive
correlation between the ideological and partisan quality of college students and the current state of
growth (r=0.692, sig=0.00), as well as among the quality of ideological and political excellence of
college students, the degree of group-self harmony (r=0.537, sig=0.00), and the degree of harmony
between heaven and man (r=0.584, sig=0.00) all have significant positive correlations. The
relationship between college students' ideological and political quality and the degree of healthy and
harmonious growth of college students is strong. That is, improving the ideological and political
quality of college students and achieving harmonious growth of college students are inherently
unified.
Table 7. Correlation between ideological and political quality and growth status
Projects
Man and Self
White Self
Harmony
People and Society
Harmony of the group
and self
Man and Nature
Harmony between
Heaven and Man
Growth
Status
Political View
Pearson
.508**
.402**
.483**
.538**
Salience
.000
.000
.000
000
N
162
162
162
162
Thought Awareness
Pearson
.602**
598**
.611**
.702**
Salience
.000
.000
.000
.000
N
162
162
162
162
Moral values
Pearson
.554**
.505**
.595**
.642**
Salience
.000
.000
.000
.000
N
162
162
162
162
Legal View
Pearson
.482**
.404**
.584**
.566**
Salience
.000
.000
.000
.000
N
162
162
162
162
Ideological and
political quality
Pearson
.605**
.537**
642**
.692**
Salience
.000
.000
.000
.000
N
162
162
162
162
6 Conclusion
The use of big data technology for ideological and political education in higher education institutions
is currently a research hotspot. This paper proposes the application of data mining technology to
ideological and political education in higher education institution to promote ideological and politic
education in higher education institution to keep pace with the times, break the traditional ideological
and political education in a single way of communication, and also greatly improve the timeliness of
ideologic and politic education in higher education institution. The following conclusions can be
inferred:
1) The proportions contained in the three types of score bands, namely 17.99%, 59.38% and
9.96%, basically coincide with the sample proportions accounted for by the three types of
Applied Mathematics and Nonlinear Sciences (aop) (aop)
grades assessed by clustering, namely 30.51%, 62.02% and 8.47%, which confirms that such
a model of data mining adopted according to the quantitative table of ideologic and politic
education assessment in higher education institution is a very successful model. It helps to
give better play to the efficacy of ideological and political education, meet the needs of “real
people”, and improve the nurturing and guiding role of ideological and political education in
higher education institutions.
2) There are significant correlations between college students’ growth environment and college
students’ political outlook (r=0.611, sig=0.00), ideological awareness (r=0.635, sig=0.00),
moral outlook (r=0.587, sig=0.00) and legal outlook (r=0.628, sig=0.00). There are significant
positive correlations between the quality of college students’ ideological and political quality
and the degree of self-harmony (r=0.605, sig=0.00), group-self harmony (r=0.537, sig=0.00)
and celestial-human harmony (r=0.584, sig=0.00) of college students. Ideologic and political
education in higher education institutions should update the education concept, enhance the
cultural connotation of ideologic and political education in higher education institutions,
promote the optimization of the content system of ideological and political education in higher
education institutions, innovate the ways and methods of ideologic and politic education in
higher education institution, build a harmonious environment of ideologic and politic
education in higher education institution, and strengthen the construction of soft power of
ideological and political educators.
References
[1] Wang, N. (2021). Ideologic and politic education recommendation system based on ahp and improved
collaborative filtering algorithm. Scientific programming(Pt.13), 2021.
[2] Li, C. Y., & Zheng, L. (2021). Analysis of tai chi ideological and political course in university based on
big data and graph neural networks. Scientific Programming, 2021(1), 1-9.
[3] Lewis, L. J. (2015). Education and political independence in africa, and other essays. Frontiers in Aging
Neuroscience, 7(3), 183.
[4] Zhang, R. (2021). Research on the security of web-based ideologic and politic education resource
information system based on amp. Journal of Intelligent and Fuzzy Systems(2), 1-12.
[5] Sun, X., & Zhang, Y. (2021). Research on the framework of university ideologic and politic education
management system based on artificial intelligence. Journal of Intelligent and Fuzzy Systems(5), 1-10.
[6] Li, X. (2021). Web remote ideologic and politic education system constructed by using agent technology.
Journal of Intelligent and Fuzzy Systems(2), 1-10.
[7] Zhu, L. (2021). Research on the design and application of ideologic and politic education platform in
higher education institution based on moodle. Journal of Intelligent and Fuzzy Systems(3), 1-8.
[8] He, T. (2020). Research on the effectiveness of the ideologic and politic education in higher education
institution based on the computer data analysis. Basic & clinical pharmacology & toxicology.(S1), 127.
[9] Li, K., Jing, M., Tao, X., & Duan, Y. (2021). Research on online management system of network ideologic
and politic education of college students. International Journal of Electrical Engineering Education,
002072092098370.
[10] Xia, Y. (2020). Big data based research on the management system framework of ideologic and politic
education in higher education institution. Journal of Intelligent and Fuzzy Systems(6), 1-10.
[11] Bin, S., & Shuqin, L. (2017). Integration of big data into ideological and political education. Journal of
Higher Education.
[12] Lei, X. (2015). Discussion on shaping the sound personality of college students in the ideologic and politic
education in higher education institution. International Journal of Technology Management(005), 000.
Applied Mathematics and Nonlinear Sciences (aop) (aop)
[13] Koivula, A., Koiranen, I., Saarinen, A., & Keipi, T. (2019). Social and ideological representativeness: a
comparison of political party members and supporters in finland after the realignment of major parties.
Party Politics.
[14] Asano, T. A. (2022). Ideological extremism and political participation in japan. Social Science Japan
Journal.
[15] Liang, L. (2016). Research on ideologic and politic education reform in higher education institution based
on the perspective of faith education. International Journal of Technology, Management.
[16] Li, F. (2018). Research method innovation of college students’ ideologic and politic education based on
cognitive neuroscience. Neuroquantology, 16(5).
