Application of Apriori Algorithm in the Mining of Common

Technical Actions in Basketball Games

Wei Li and Wenjing Chen

Yunnan College of Business Management, KunMing, 650000, China

Keywords: Mining Algorithm, Apriori Algorithm, Technical Actions, Basketball.

Abstract: The role of basketball technical actions in basketball games is very important, but there is the problem of

inaccurate reference information. Ordinary technical algorithms cannot solve the problem of basketball

technical actions in basketball games, and the information is inaccurate. Therefore, this paper proposes an

Apriori algorithm for professional basketball technical action analysis. First, the mining algorithm is used to

collect the information of basketball technical actions, and divide the indicators according to the requirements

of basketball technical actions to reduce them Disturbing factors in basketball technical actions. Then, the

mining algorithm forms a basketball technical action plan for the professional basketball technical actions of

college students, and synthesizes the basketball technical action results Analyse. MATLAB simulation shows

that under certain evaluation criteria, Apriori algorithm has a professional basketball technical action for

college students The scoring rate and basketball technical action foul rate are better than ordinary technical

algorithms.

1 INTRODUCTION

With the continuous development and change of

basketball games, the identification and analysis of

actions in basketball games is becoming more and

more important. Apriori's algorithm is a classic

association rule mining algorithm (Bowman, and

Harmon, et al. 2023), which can be used to analyze

frequent itemsets and association rules in large-scale

data (Chun, and Lee, et al. 2023). In this paper, the

Apriori algorithm is applied to the recognition and

analysis of actions in basketball games, and its

application effect and technical ideas are explored.

1.1 Apriori Algorithm Principle

Apriori's algorithm is an association rule mining

algorithm based on frequent itemsets. The core idea

is based on the "a priori principle", which states that

if a set of terms is frequent (El-Saleh, 2023), all

subsets of it are also frequent. The algorithm is

mainly divided into two steps: the first step is to

generate the candidate set, and the second step is to

filter out the frequent itemset based on the minimum

support threshold (Ferioli, and Conte, et al. 2023).

The specific process is as follows:

1.1.1 Build a Candidate Set

First, all terms are used as candidate 1 itemsets to

calculate their support; then, according to the rule

satisfying the "a priori principle" (Ferioli, and

Rampinini, et al. 2023), the candidate k-1 itemsets are

connected to obtain the candidate k itemset; finally,

the support and join operations are repeated until all

frequent itemsets are obtained (Goldschmied, and

Raphaeli, et al. 2023).

1.1.2 Filter for Frequent Itemsets

Based on the minimum support threshold, filter out

frequent itemsets that meet the criteria. At the same

time, according to the frequent itemset, the

association rule is generated, and its confidence (Han,

2023), support degree and improvement degree are

calculated to obtain a high-quality rule set.

1.2 Application of Apriori Algorithm in

Action Recognition Analysis in

Basketball Games

The wide variety of actions in basketball games is

complex and diverse, which puts forward high

requirements for action recognition and analysis

326

Li, W. and Chen, W.

Application of Apriori Algorithm in the Mining of Common Technical Actions in Basketball Games.

DOI: 10.5220/0013540900004664

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 326-331

ISBN: 978-989-758-763-4

(Hassan, and Alibrahim, et al. 2023). As a classic

association rule mining algorithm, Apriori algorithm

can be used for the recognition and analysis of actions

in basketball games. The specific applications are as

follows:

1.2.1 Data Preprocessing

Firstly, the action data in basketball games is

collected, sorted and cleaned to obtain a dataset

(Hoelzemann, and Romero, et al. 2023). Then,

according to the specific needs, the actions are

classified and coded for subsequent analysis and

mining.

1.2.2 Build a Candidate Set

According to the principle of Apriori's algorithm, the

action data is generated by the candidate set. Various

actions in basketball can be regarded as different

items (Horvat, and Job, et al. 2023), and different

items can be combined, connected and screened to

obtain various frequent item sets.

1.2.3 Filter for Frequent Itemsets

Based on the minimum support threshold, all frequent

itemsets are found by filtering, that is, itemsets with

occurrence frequencies higher than the specified

threshold. At the same time, according to the frequent

itemset, the association rules between various actions

are excavated, and the indicators such as confidence,

support and improvement are calculated to obtain

high-quality rule sets (Hulka, and Strniste, et al.

2023).

1.2.4 Action Recognition Analysis

Through the application of the above algorithm, the

relationship model and rule set of various actions in

basketball games can be obtained, and then the

actions can be identified and analysed (Ibanez, and

Lopez-Sierra, et al. 2023). According to the

association relationship in the rule set, various actions

can be classified and combined, and different action

combination modes and action sequences can be

obtained (Ibanez, and Pinar, et al. 2023).

1.3 Manage Policies

Data collection and collation: Action data collection

and collation in basketball games is the premise of

action recognition and analysis, and it is necessary to

strengthen the technological innovation and

application of data collection and collation (Mengi,

and Alemdaroglu, et al. 2023).

Algorithm improvement and optimization:

Apriori algorithm is a classic association rule mining

algorithm (Merino-Campos, and Leon-Quismondo, et

al. 2023), but it has problems such as high time

complexity when processing a large amount of data,

and it is necessary to strengthen the algorithm

improvement and optimization (Morales-Belando,

and Canovas-Lopez, et al. 2023).

Action coding and classification: Action coding

and classification are essential for action recognition

and analysis, and it is necessary to strengthen the

standardization and standardization of action coding

and classification.

Rule generation and filtering: Rule generation and

filtering is the core step of action identification and

analysis, and the accuracy and efficiency of rule

generation and filtering need to be further

strengthened.

Application expansion and innovation: Action

recognition analysis has a wide range of application

prospects and market demand in basketball games,

and it is necessary to strengthen application

expansion and innovation to provide strong support

for the development of motion recognition analysis.

Based on Apriori algorithm, this paper explores its

application in motion recognition analysis in

basketball games. Through the application of data

preprocessing, candidate set generation, frequent

itemset screening, and action recognition and

analysis, the recognition and analysis of actions in

basketball games can be effectively realized. In the

future application process, it is necessary to further

strengthen the research and application of algorithm

improvement, data collection and sorting, action

coding and classification, etc., so as to provide more

comprehensive and in-depth support for the

development and application of action recognition

and analysis in basketball games.

2 RELATED CONCEPTS

2.1 Mathematical Description of

Apriori's Algorithm

Apriori's algorithm optimizes the basketball technical

action scheme by using association rule mining, Ask

for characteristic actions during basketball games.

The scheme is integrated to finally determine the

winning rate of the basketball game. Apriori's

algorithm combines the advantages of association

rule mining and uses basketball games for

Application of Apriori Algorithm in the Mining of Common Technical Actions in Basketball Games

327

quantification, which can improve basketball

technical action scoring technical actions.

Hypothesis 1: The basketball technical action

requirements is

, the basketball technical action

plan is

, the satisfaction of the basketball

technical action plan is

, and the basketball technical

action plan judgment function is

(0)

Ja≈

,As

shown in Equation (1).

11 12

21 22

()

Jeu e j



=⋅









(1)

2.2 Selection of Scoring Technical

Action Schemes

Hypothesis 2: The basketball game function is

()

, and the weight coefficient is

, then, the

basketball technical action requires an unqualified

basketball game as shown in equation (2).

()= (, )

ii ii

ee Jey x

⋅−

∏



(2)

2.3 Analysis of Basketball Technical

Action Schemes

Basketball, as an important technology, involves

more contents, including physical fitness and

psychological factors, so it is necessary to

comprehensively analyze basketball technology and

data. First of all, we should analyze the technical

movements of basketball and the physical fitness of

basketball, and then determine the relationship

between various technologies and data, as well as the

analysis conditions and methods. Secondly, we

should obtain the corresponding feedback

information, and make a comprehensive judgment on

basketball data and basketball methods, so as to

realize the comprehensiveness of data and analyze

and verify the effectiveness of basketball methods. As

shown in the Figure 1.

Analyze the relevant data in basketball and test the

relationship between basketball skills and

movements, form the optimization scheme of

basketball, and then analyze the key indicators and

key contents in basketball skills, which has improved

the shooting rate of basketball and the overall score

of basketball. In addition, we should analyze and

judge the comprehensive tension of players, realize

the comprehensive judgment and analysis of data, so

as to improve the overall skills of basketball.

Foul rate

Scoring rate

Technical

action

Reference

information

Basketball match

Figure 1: Score the results of the selection of technical

action schemes

3 OPTIMIZATION STRATEGIES

FOR BASKETBALL GAMES

Apriori algorithm adopts a random optimization

strategy for basketball games, and adjusts technical

action parameters to realize the scheme optimization

of basketball games. Apriori's algorithm divides

basketball games into different basketball technical

action levels, and randomly selects different schemes.

In the process of comprehensive analysis and internal

analysis of data, it is necessary to judge and optimize

basketball data accordingly, and identify the key

indicators and key contents in basketball data, so as

to integrate them with scores and realize the

corresponding relationship and corresponding

processing between data scores.

4 PRACTICAL EXAMPLES OF

BASKETBALL GAMES

4.1 Introduction to Basketball

Technical Movements

The actual basketball game and 6 basketball scores as

the research object, the comprehensive judgment of

basketball score data, and the key scores and key

actions in basketball are summarized to form key

INCOFT 2025 - International Conference on Futuristic Technology

328

identification, information and characteristics. As

shown in the Table 1.

Table 1. College basketball technical action requirements

Scope of

application

Grade Scoring

rate

Score

technical

moves

Emergency

sto

Standar

86.46 88.10

86.15 89.27

Sliding step Standar

83.82 88.41

89.98 87.41

Stride Standar

89.32 86.31

86.89 85.20

The basketball technique action process in Table

1. is shown in Figure 2.

Basketball technical action

Scoring technical action

Apriori

Scoring

rate

Basketball

match

Figure 2: The analytical process of a basketball game

Compared with ordinary technical algorithms, the

basketball technical action scheme of Apriori

algorithm is closer to the actual basketball technical

action requirements. In terms of scoring rate, foul

rate, etc. Basketball in the whole game process this

article proposed the algorithm, can carry on the

recognition to the basketball movement, especially to

the foul movement, the score movement as well as the

athlete false movement carries on the recognition, and

provides the technique for the later stage plan and the

implementation to carry on the basketball training

better, therefore said I directly the algorithm sound

quite effective.

4.2 Basketball Matches

In the whole data analysis process of basketball, we

should analyze the key indicators and key contents of

basketball, realize the comprehensive judgment of

data, and realize the matching between games and

data. In order to more accurately verify the scoring

rate of basketball games, select basketball games with

different basketball technical action levels, and the

basketball technical action scheme is shown in Table

Table 2: Score the overall picture of the technical action

plan

Cate

Satisfaction Anal

sis rate

Emer

enc

Sto

89.00 85.85

Sliding Step 87.29 87.78

Stride 81.65 90.38

Mean 87.73 87.55

84.47 87.57

P=2.17

4.3 Scoring Technical Action and

Stability of Basketball Technical

Action

How to move needs to be analyzed by intelligent

methods, and the relationship between action,

psychology and physical fitness is used to find out

certain laws, and form joint analysis and results

between them, so as to realize comprehensive

judgment of relevance. As shown in the Figure 3.

Figure 3: Scoring technical actions of different algorithms

From the analysis just now, we can see that

although both methods change within the constraints,

the original technical algorithm is more closely

related to the middle line, and there are relatively few

particles and data close to both sides, which shows

that the algorithm in this paper is relatively

concentrated. One-to-one basketball game in the key

indicators and data need to be judged, especially the

error rate and other aspects of the content needs to be

analyzed. As shown in the Table 3.

Three-week data analysis shows that the

algorithm in this paper has a great advantage in the

score of the whole action and the correction rate of

wrong data, and the improvement range is about 3%

~ 5%. Moreover, the whole data change and data

analysis process show a relatively stable process, so

Application of Apriori Algorithm in the Mining of Common Technical Actions in Basketball Games

329

Table 3: Comparison of the accuracy of basketball

techniques and movements of different methods

Algorithm Score

technical

moves

Magnitude

of change

Action is

victory.

Apriori

orith

92.76 93.87 92.24

Ordinary

technical

algorithms

87.11 90.05 86.55

P 88.77 85.93 86.95

the whole analysis score and effect are better. In

order to verify the effective needs of this analysis, the

continuous graphic analysis is shown in Figure 4.

Figure 4: Apriori algorithm basketball technical action

scoring technical action

From the data analysis results in the atlas, we can

know that the algorithm recorded in this paper has

relatively large fluctuations, mainly because my

algorithm is a step-by-step process, which requires

large fluctuations at the beginning, and then carries

out data fusion and data concentration, and optimizes

the whole analysis process and the whole change

process of data, so as to realize the overall analysis of

data.

5 CONCLUSIONS

If you can be widely used in various fields of society

as a common economic movement, Lanzhou New

Energy Power is a common training method, but there

is a lack of effective guidance and targeted guidance

in previous algorithms, so it is possible to use

intelligent methods to optimize and discover this

method, and the key indicators can be optimized and

guided.At the same time, the scoring rate and foul rate

of basketball technical actions are analyzed in depth,

and a collection of technical actions is constructed.

Research shows that the Apriori algorithm can

improve the scoring rate and win rate of basketball

games Basketball games perform general basketball

technical moves. However, in the process of Apriori

algorithm, too much attention is paid to the analysis

of basketball technical actions, resulting in

irrationality in the selection of basketball technical

action indicators.

REFERENCES

Bowman, R. A., Harmon, O., & Ashman, T.(2023)

Schedule inequity in the National Basketball

Association. Journal of Sports Analytics, 9(1): 61-76.

Chun, D., Lee, M. Y., Kim, S. W., Cho, E. Y., & Lee, B.

H.(2023) The Mediated Effect of Sports Confidence on

Competitive State Anxiety and Perceived Performance

of Basketball Game. International Journal of

Environmental Research and Public Health, 20(1).

El-Saleh, M. S.(2023) The Effect of an Educational

Program for Mental Visualization to Teaching Some

Shooting Skills for Basketball Beginners. Annals of

Applied Sport Science, 11(2).

Ferioli, D., Conte, D., Scanlan, A. T., & Vaquera,

A.(2023a) Technical-Tactical Demands of 3 x 3

International Basketball Games According to Game

Outcome, Player Sex, and Competition Phase. Journal

of Strength and Conditioning Research, 37(2): 403-412.

Ferioli, D., Rampinini, E., Conte, D., Rucco, D.,

Romagnoli, M., & Scanlan, A.(2023b) Physical

demands during 3? 3 international male and female

basketball games are partially impacted by competition

phase but not game outcome. Biology of Sport, 40(2):

377-387.

Goldschmied, N., Raphaeli, M., & Morgulev, E.(2023)

"Icing the shooter" in basketball: The unintended

consequences of time-out management when the game

is on the line. Psychology of Sport and Exercise, 68.

Han, F. L.(2023) STRENGTH TRAINING INFLUENCES

ON BASKETBALL PLAYERS. Revista Brasileira De

Medicina Do Esporte, 29.

Hassan, A. K., Alibrahim, M. S., & Ahmed, Y.(2023) The

effect of small-sided games using the FIT LIGHT

training system on some harmonic abilities and some

basic skills of basketball players. Frontiers in Sports

and Active Living, 5.

Hoelzemann, A., Romero, J. L., Bock, M., Van Laerhoven,

K., & Lv, Q.(2023) Hang-Time HAR: A Benchmark

Dataset for Basketball Activity Recognition Using

Wrist-Worn Inertial Sensors. Sensors, 23(13).

Horvat, T., Job, J., Logozar, R., & Livada, C.(2023) A Data-

Driven Machine Learning Algorithm for Predicting the

Outcomes of NBA Games. Symmetry-Basel, 15(4).

Hulka, K., Strniste, M., Hruby, M., & Belka, J.(2023)

Validity and reliability of fatigue manifestation during

basketball game-based drill. Journal of Human Sport

and Exercise, 18(3): 555-562.

INCOFT 2025 - International Conference on Futuristic Technology

330

Ibanez, S. J., Lopez-Sierra, P., Hernandez-Beltran, V., &

Feu, S.(2023) Is Basketball a Symmetrical Sport?

Symmetry-Basel, 15(7).

Ibanez, S. J., Pinar, M. I., Garcia, D., & Mancha-Triguero,

D.(2023) Physical Fitness as a Predictor of

Performance during Competition in Professional

Women's Basketball Players. International Journal of

Environmental Research and Public Health, 20(2).

Mengi, E., Alemdaroglu, B. U., & Erturan, A. G.(2023)

Technical and Internal Load Responses in 3-A-Side

Full-Court Basketball Games: The Effects of Coaches'

Verbal Feedback. European Journal of Human

Movement, 50: 92-102.

Merino-Campos, C., Leon-Quismondo, J., Perez, J. G., &

Fernandez, H. D.(2023) Use of video games in Physical

Education and self-concept development in

adolescence: sex-based differences. Retos-Nuevas

Tendencias En Educacion Fisica Deporte Y

Recreacion(47): 110-118.

Morales-Belando, M. T., Canovas-Lopez, M., & Arias-

Estero, J. L.(2023) Differences in the perception of

mini-basketball players' motivational climate according

to the games' results. A pilot study in youth sport.

Retos-Nuevas Tendencias En Educacion Fisica Deporte

Y Recreacion(48).

Application of Apriori Algorithm in the Mining of Common Technical Actions in Basketball Games

331