Classification Method of Japanese Teaching Resources Based on

Density Clustering Algorithm

Qinlang Li

Chengdu Neusoft University, Chengdu City, Sichuan Province, 611844, China

Keywords: Computer, Classification Method Based on Density Clustering Algorithm, Japanese Language Teaching,

Teaching Resources.

Abstract: In Japanese language teaching, teaching resources are very important and make the process of learning

Japanese easier. From this, we can know the classification of resources, you can better find the required

information, and the general classification method cannot solve the problem of inaccurate classification of

Japanese teaching resources. Therefore, this paper proposes a classification method based on density

clustering algorithm for resource classification analysis. First, computers are used to classify Japanese

language teaching resources, and indicators are divided according to resource classification requirements to

reduce resource classification in the interfering factor. Then, the computer classifies the results of Japanese

teaching resources, forms a resource classification scheme, and classifies the resources Conduct a

comprehensive analysis. MATLAB simulation shows that the classification method based on density

clustering algorithm can classify resources in Japanese language teaching accurately under certain evaluation

criteria The efficiency of resource classification is better than that of ordinary classification methods.

1 INTRODUCTION

With the continuous development of artificial

intelligence and big data technology, these

technologies have gradually been introduced in the

field of education to optimize and improve the

classification and management of educational

resources (Adebayo, and Quadri, et al. 2023). The

density clustering algorithm is a commonly used

clustering algorithm that does not require a pre-

defined number of clusters and can automatically

discover clusters in a dataset (Arabit-Garcia, J and

Prendes-Espinosa, et al. 2023). This article will

explore the application of density clustering

algorithm in the classification of Japanese teaching

resources.

1.1 Description of the classification of

Japanese language teaching

resources

In Japanese language education, there are a large

number of teaching resources, such as teaching

materials, vocabulary, grammar, listening and

reading materials, etc. These resources are large in

number and complex in classification, and need to be

automatically classified using clustering algorithms

to better manage and utilize these resources

(Arblaster, and Mackenzie, et al. 2023). The goal of

the Japanese Language Teaching Resource

Classification is to put similar resources in the same

category and different resources in different

categories for better management and utilization.

1.2 Fundamentals of density clustering

algorithms

The density clustering algorithm is a density-based

non-hierarchical clustering algorithm, which does not

need to pre-set the number of clusters, but calculates

the local density of data points to mine out the clusters

in the dataset (Chan, 2023). The basic principles of

the density clustering algorithm can be summarized

in the following steps:

Density reachability: According to the specified

distance threshold, the density reachability judgment

is made for each data point, and if one data point can

be reached from another data point (Chen, 2023), the

two data points are considered to be density

reachable.

250

Li, Q.

Classiﬁcation Method of Japanese Teaching Resources Based on Density Clustering Algorithm.

DOI: 10.5220/0013539100004664

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 3rd International Conference on Futuristic Technology (INCOFT 2025) - Volume 1, pages 250-256

ISBN: 978-989-758-763-4

Core object: A data point is considered core if it

contains enough data points within a radius centered

on it and within a specified distance threshold (Chen,

2023).

Direct density reachability: For a core object, if a

data point is within the radius of the specified distance

threshold, the data point is considered to be directly

density reachable (Cohen, and Lefstein, et al. 2023).

Density connected: If a data point is both a core

object and another core object has a direct density

reachability, the two data points are considered to be

density connected (Gosztonyi, and Varga, 2023).

Cluster formation: By continuously adding

densely connected data points to the same cluster,

several clusters are eventually formed (Javier Robles-

Moral, and Fernandez-Diaz, et al. 2023).

Noise points: Data points that are not included in

any cluster are considered noise points (Jin, and Lu,

2023).

The advantage of the density clustering algorithm

is that it does not need to pre-set the number of

clusters, can automatically discover clusters in the

dataset, and has good fault tolerance for noise points.

1.3 Application of Density Clustering

Algorithm in the Classification of

Japanese Teaching Resources

In the classification of Japanese teaching resources,

the density clustering algorithm can be applied to two

aspects: grammar and vocabulary (Karol, and

Shaylor, et al. 2023).

1.3.1 Grammatical Classification

In grammatical classification, each sentence can be

regarded as a data point, and the similarity between

sentences can be judged by calculating the distance

between sentences, and similar sentences can be

placed in the same cluster. In the process of

grammatical classification, the subject, predicate,

object and other information of the sentence can be

added to the distance calculation, so as to make the

classification result more accurate (Kazima, and

Jakobsen, et al. 2023).

1.3.2 Vocabulary Classification

In lexical classification, you can think of each word

as a data point, calculate the distance between words,

judge the similarity between words, and put similar

words in the same cluster. In the process of

vocabulary classification, the part of speech,

meaning, frequency and other information of words

can be added to the distance calculation, so as to make

the classification results more accurate (Krieg, 2023).

1.4 Practical Application of Density

Clustering Algorithm in the

Classification of Japanese Teaching

Resources

1.4.1 Dataset Selection and Preprocessing

Choosing the right dataset is one of the keys to

classification. In Japanese language education,

datasets can be constructed by collecting various

teaching resources such as teaching materials,

listening materials, reading materials, etc. In the

preprocessing of a dataset, each data point can be

converted into a vector, adding different features to

the vector (Kwee, and Santos, 2023).

1.4.2 Practical Case Application

In practical case applications, the density clustering

algorithm can be applied to the classification of

Japanese teaching resources, such as classifying

Japanese textbooks and Japanese phonetic

vocabulary. In the process of classification, the

appropriate distance calculation method and density

threshold can be selected according to specific needs,

so as to obtain reasonable classification results.

1.5 Challenges of Density Clustering

Algorithm In The Classification Of

Japanese Teaching Resources

1.5.1 The Quality of the Dataset

The quality of the dataset directly affects the accuracy

of the classification results. How to ensure the quality

of data sets and avoid duplicate data and erroneous

data is one of the issues that need to be studied and

explored.

1.5.2 Selection of Distance Calculation

Method

The distance calculation method directly affects the

effectiveness and speed of classification. How to

choose the appropriate distance calculation method to

improve the accuracy and speed of classification is

also one of the problems that need to be studied and

discussed.

Classiﬁcation Method of Japanese Teaching Resources Based on Density Clustering Algorithm

251

1.5.3 Parameter selection

There are many parameters to choose in the density

clustering algorithm, such as density threshold,

neighborhood radius, and so on. How to choose the

appropriate parameters is one of the keys to affect the

effect of the algorithm.

The classification of Japanese teaching resources

is a complex problem, and the density clustering

algorithm is an effective classification algorithm,

which can automatically discover the clusters in the

dataset, without presetting the number of clusters, and

has good fault tolerance for noise points (Li, and Qu,

et al. 2023). However, there are challenges such as the

quality of the dataset, the choice of distance

calculation method, and the selection of parameters.

In future research, it is necessary to further optimize

the parameters of the algorithm, improve the effect

and speed of the algorithm, and improve the

robustness and adaptability of the algorithm.

Teaching resources are one of the important

components of Japanese language teaching and are of

great significance for learning Japanese. However, in

the process of resource classification, the resource

classification scheme has the problem of poor

accuracy, which has a certain impact on Japanese

language learning. Some scholars believe that

applying the classification method based on density

clustering algorithm to Japanese teaching analysis

can effectively analyze the resource classification

scheme and provide corresponding support for

resource classification. On this basis, this paper

proposes a classification method based on density

clustering algorithm to optimize the resource

classification scheme and verify the effectiveness of

the model.

2 RELATED CONCEPTS

2.1 Mathematical Description of a

Classification Method Based on

Density Clustering Algorithm

The classification method based on the density

clustering algorithm uses computer operations to

optimize the resource classification scheme, and finds

the unqualified values in Japanese language teaching

according is

, and the indicators in the resource

classification

, and classifies the resources the

scheme is

(

iij

tol y h⋅ ）

, integrated to determine the

feasibility of Japanese language teaching, and the

calculation is shown in Equation (1).

() max( )

iij ij ij ii h

tol y h y h hY

−

⋅=≥ ⋅



)

Among them, the judgment of outliers is shown in

Equation (2).

max( ) ( 3) ( )

ij ij ij i

hh meanh h

−

=÷ ⋅





)

The classification method based on density

clustering algorithm combines the advantages of

computer computing and uses Japanese language

teaching for quantification, which can improve the

accuracy of resource classification.

Hypothesis I. The resource classification

requirements is

, the resource classification

scheme is

, the satisfaction of the resource

classification scheme is

, and the resource

classification scheme judgment function is

(0)

Dh≈

, As shown in Equation (3).

() lim

ii i

Dp h y

→

=→











)

2.2 Selection of Teaching Resource

Programs

Hypothesis II. The Japanese teaching function is

()

, the weight coefficient is

, then, the

resource classification requires unqualified Japanese

teaching as shown in Equation (4).

()= ( ) lim

ii i i

jh z D p w

→

⋅−→

∏



)

Based on assumptions I and II, a comprehensive

function for Japanese language teaching can be

obtained, and the result is shown in Equation (5).

() ( ) max( )

ii ij

hDp h+≤

)

INCOFT 2025 - International Conference on Futuristic Technology

252

In order to improve the accuracy of resource

classification effectively, all data needs to be

normalized, and the results are shown in Equation (6).



() ( ) (

ii ij i

hDp meanh h

−

+↔ ⋅



(6)

2.3 Analysis of Resource Classification

Schemes

Before the classification method based on density

clustering algorithm, multi-dimensional analysis

should be performed on the resource classification

scheme, and the resource classification requirements

should be mapped to the Japanese teaching library to

eliminate the unqualified resource classification

scheme is

()

No h

, According to Equation (6), the

anomaly evaluation scheme can be proposed, and the

results are shown in Equation (7).



() ( )

()

ij i

jh Dp

No h

mean h h

−

⋅



(7)

Among them,



() ( )

()

ij i

jh D p

mean h h

≤

−

⋅



is stated that the scheme needs to be proposed,

otherwise the scheme integration required is

()

, and the result is shown in Equation (8).



() min[ () ( )]

iii

hh jh D p=+



(8)

Japanese language teaching conducts

comprehensive analysis, and sets the threshold and

indicator weights of the resource classification

scheme to ensure the accuracy of the classification

method based on the density clustering algorithm.

Japanese language teaching is a systematic test

resource classification scheme that needs to be

analyzed. If Japanese language teaching is in a

nonnormal distribution is

()

unno h

, its resource

classification scheme will be affected, reducing the

accuracy of the overall resource classification, and

the calculation result is

()

accur h

, shown in

Equation (9).



min[ ( ) ( )]

() 100

() ( )

jh Dp

accur h

jh Dp

=×



)

The survey resource classification scheme shows

that the teaching resource scheme shows a multi-

dimensional distribution, which is in line with the

objective facts. Japanese language teaching is not

directional, indicating that the teaching resource

scheme has a strong randomness, so it is regarded as

a high analytical study. If the random function of

Japanese teaching is

()

randon h

, then the

calculation of formula (9) can be expressed as

formula (10).



min[ ( ) ( )]

( ) 100% (

)

() ( )

jh Dp

accur h randon h

jh Dp

=×+



(10

)

Among them, Japanese language teaching meets

the normal requirements, mainly computer operations

to adjust Japanese language teaching, eliminate

duplicate and irrelevant schemes, and supplement the

default scheme, so that the dynamic correlation of the

entire resource classification scheme is strong.

3 OPTIMIZATION STRATEGIES

FOR JAPANESE LANGUAGE

TEACHING

The classification method based on density clustering

algorithm adopts the random optimization strategy

for Japanese language teaching, and adjusts the

parameters of Japanese teaching resources to realize

the optimization of Japanese language teaching. The

classification method based on the density clustering

algorithm divides Japanese language teaching into

different resource classification levels, and randomly

selects different schemes. In the iterative process, the

resource classification schemes of different resource

classification levels are optimized and analyzed.

After the optimization analysis is completed, compare

the resource classification levels of different schemes

and record the best Japanese language teaching.

Classiﬁcation Method of Japanese Teaching Resources Based on Density Clustering Algorithm

253

4 PRACTICAL EXAMPLES OF

JAPANESE LANGUAGE

TEACHING

4.1 Introduction to Resource

Classification

In order to facilitate resource classification, this paper

takes Japanese language teaching in complex

situations as the research object, with 12 paths and a

test time of 12h, and the specific resource

classification of Japanese language teaching The

scheme is shown in Table 1.

Table 1: Resource classification requirements

Scope of

application

grade accuracy Teaching

resources

Basic

equivalents

routine 85.47 87.54

Highe

85.13 85.68

Advanced

equivalents

routine 87.82 86.31

Highe

86.35 87.52

Audition routine 83.50 85.48

Highe

86.15 86.00

The resource classification process in Table 1 is

shown in Figure 1.

Density

clustering

Teaching

resources

Classification

accuracy

Japanese

teaching

Classification

efficiency

Resource

classification

Japanese

learning

Figure 1: The analytical process of Japanese language

teaching

Compared with the general classification method,

the resource classification scheme of the

classification method based on the density clustering

algorithm is closer to the actual resource

classification requirements. In terms of the rationality

of Japanese language teaching, the classification

method based on density clustering algorithm is better

than the ordinary classification method. It can be seen

from the change of resource classification scheme in

Figure II that the classification method based on

density clustering algorithm has higher accuracy and

faster classification speed. Therefore, the resource

classification scheme speed, teaching resource

scheme resource classification scheme and

summation stability of the classification method

based on density clustering algorithm are better.

4.2 Japanese Language Teaching

The resource classification scheme for Japanese

language teaching includes unstructured information,

semi-structured information, and structured

information. After the pre-selection of the

classification method based on the density clustering

algorithm, the preliminary resource classification

scheme for Japanese language teaching is obtained,

and the Japanese language teaching is obtained

Analyze the feasibility of resource classification

schemes. In order to verify the results of Japanese

language teaching more accurately, select Japanese

language teaching with different resource

classification levels, and the resource classification

scheme is shown in Table 2.

Table 2: The overall picture of the teaching resource

program

Categor

Rationalit

Analysis rate

Basic

uivalents

91.72 87.68

Advanced

Equivalents

87.48 92.07

Audition 88.77 90.41

Mean 90.12 90.03

89.39 90.71

P=1.696

4.3 Teaching Resources and Stability of

Resource Classification

In order to verify the accuracy of the classification

method based on the density clustering algorithm, the

resource classification scheme is compared with the

general classification method, and the resource

classification scheme is shown in Figure 2.

Figure 2: Teaching resources for different algorithms

INCOFT 2025 - International Conference on Futuristic Technology

254

It can be seen from Figure 2 that the teaching

resources of the classification method based on

density clustering algorithm are higher than those of

ordinary classification methods, but the error rate is

lower, indicating that the resource classification of

the classification method based on density clustering

algorithm is relatively stable The classification of

resources in the general classification method is

uneven. The average resource classification scheme

of the above three algorithms is shown in Table 3.

Table 3: Comparison of resource classification accuracy of

different methods

Algorithm Teaching

resources

Magnitude

of change

Error

classification

method based

on density

clustering

orith

93.24 95.46 94.44

Common

classification

metho

89.94 92.54 91.42

P 91.78 85.96 90.97

Table 3 shows that the general classification

method has shortcomings in the accuracy of the

classification of teaching resources in Japanese

language teaching, and the Japanese language

teaching has changed significantly, and the error rate

is high. The general results of the classification

method based on the density clustering algorithm

have higher teaching resources and are better than the

general classification method. At the same time, the

teaching resources of the classification method based

on density clustering algorithm are greater than 93%,

and the accuracy does not change significantly. In

order to further verify the superiority of the

classification method based on density clustering

algorithm. In order to further verify the effectiveness

of the proposed method, the classification method

based on density clustering algorithm is analyzed by

different methods, as shown in Figure 3.

It can be seen from Figure 3 that the teaching

resources of the classification method based on the

density clustering algorithm are significantly better

than the ordinary classification method, and the

reason is that the classification method based on the

density clustering algorithm increases the adjustment

coefficient of Japanese teaching and sets it

Thresholds for Japanese teaching resources that

exclude resource classification schemes that do not

meet the requirements.

Figure 3: Classification method based on density clustering

algorithm: teaching resource for resource classification

5 CONCLUSIONS

Aiming at the problem that the classification of

Japanese teaching resources is not ideal, this paper

proposes a classification method based on density

clustering algorithm, and combines computer

operation to optimize Japanese language teaching. At

the same time, the accuracy of resource classification

is analyzed in depth, and a collection of Japanese

teaching resources is constructed. This study shows

that the classification method based on the density

clustering algorithm can improve the rationality of

Japanese language teaching, and can classify general

resources for Japanese language teaching 。

However, in the process of classification method

based on density clustering algorithm, too much

attention is paid to the analysis of resource

classification, resulting in unreasonable selection of

resource classification indicators.

REFERENCES

Adebayo, S. B., Quadri, G., Igah, S., & Azubuike, O.

B.(2023) Teaching in a lockdown: The impact of

COVID-19 on teachers? capacity to teach across

different school types in Nigeria. Heliyon, 9(3).

Arabit-Garcia, J., Prendes-Espinosa, M. P., & Serrano, J.

L.(2023) Open Educational Resources and active

methodologies for STEM teaching in Primary

Education. Revista Latinoamericana De Tecnologia

Educativa-Relatec, 22(1): 89-106.

Arblaster, K., Mackenzie, L., Buus, N., Chen, T., Gill, K.,

Gomez, L., Hamilton, D., Hancock, N., McCloughen,

A., Nicholson, M., Quinn, Y., River, J., Scanlan, J. N.,

Schneider, C., Schweizer, R., & Wells, K.(2023) Co-

design and evaluation of a multidisciplinary teaching

resource on mental health recovery involving people

Classiﬁcation Method of Japanese Teaching Resources Based on Density Clustering Algorithm

255

with lived experience. Australian Occupational

Therapy Journal, 70(3): 354-365.

Chan, H.(2023) 3E Model: How to Use OERs to Enhance

Teaching/Learning. Computer, 56(4): 139-142.

Chen, J.(2023a) Reform of English Writing Teaching

Method Under the Background of Big Data and

Artificial Intelligence. International Journal of E-

Collaboration, 19(4): 20-20.

Chen, L.(2023b) Empirical analysis of integrated teaching

mode of international trade based on practical training

data detection and neural network. Soft Computing.

Cohen, E., Lefstein, A., & Dishon, G.(2023) The end of the

textbook and the beginning of teaching? Tradeoffs in

designing on-line support for K-12 teachers. Education

and Information Technologies.

Gosztonyi, K., & Varga, E.(2023) Teachers' practices and

resources in the Hungarian "Guided Discovery"

approach to teaching mathematics: presenting and

representing "series of problems". Zdm-Mathematics

Education.

Javier Robles-Moral, F., Fernandez-Diaz, M., & Enrique

Ayuso-Fernandez, G.(2023) A Study of the Usefulness

of Physical Models and Digital Models for Teaching

Science to Prospective Primary School Teachers.

Education Sciences, 13(4).

Jin, H., & Lu, Q.(2023) Exploring the application of

inquiry-based defensive teaching based on the internet

and clinical resources during probationary teaching of

dermatovenereology. Chinese Journa of Dermatology,

56(4): 353-356.

Karol, D., Shaylor, R., Fiszer, E., & Weiniger, C. F.(2023)

Serious games as an innovative learning tool at a

medical conference to teach medical knowledge and

crisis resource management A narrative report.

European Journal of Anaesthesiology, 40(2): 143-146.

Kazima, M., Jakobsen, A., Mwadzaangati, L., & Gobede,

F.(2023) Teaching the concept of zero in a Malawi

primary school: illuminating the language and resource

challenge. Zdm-Mathematics Education.

Krieg, A.(2023) Teaching the Abstract: An Evaluation of

"Social Structure" in Introductory Textbooks. Teaching

Sociology.

Kwee, C. T. T., & Dos Santos, L. M.(2023) An international

study of high school teachers' experience of

incorporating water resources in their teaching.

Frontiers in Education, 7.

Li, D., Qu, P., Jin, T., Chen, C., & Bai, Y.(2023) Automatic

classification of multi-source and multi-granularity

teaching resources based on random forest algorithm.

International Journal of Continuing Engineering

Education and Life-Long Learning, 33(2-3): 177-191.

INCOFT 2025 - International Conference on Futuristic Technology

256