International Journal of Engineering Intelligent Systems

Algorithm of Scheduling Game Management Based on Big Data Integrated Energy Management and Control

2025-03-29T21:50:59-07:00

Currently, the information society is being transformed into an intelligent society. Various emerging technologies are being applied in various fields, and the application of big data (BD) in the energy field is also driving the transformation of the energy field to an intelligent industry. BD has five main characteristics: a huge amount of data, the general process of parallel processing when importing data into the database, diverse data types, little high-value data and real data sources. BD technology is used in the energy field to comprehensively consider the data of various types of energy, geographical environment, meteorological and other factors, for the purpose of energy collection, data analysis, energy processing and use. BD energy management and control not only make the energy production and management and control mode in the energy field enter a new era, but also further promote the development of the energy industry and optimize the energy scheduling algorithm. Comprehensive energy management and control involves collecting the data of the existing energy using equipment, monitoring the energy consumption data of each equipment in real time, and conducting a statistical analysis of this data, so as to obtain a more energy-saving air conditioning and equipment management strategy, ultimately achieving the goal of conserving energy and reducing operating and production costs. Game theory is derived from applied mathematics, and is used for better decision-making. The current scheduling algorithm is similar to the process scheduling in the computer field. In this paper, the traditional scheduling game management algorithm through the BD integrated energy management and control mode was analyzed. The traditional
scheduling game management algorithm was optimized using several BD technologies, and the main steps of energy production, storage and use were improved through BD technology. Finally, model experiments were conducted on the new scheduling game management algorithm based on BD energy integrated management and control and on the traditional scheduling game management algorithm; then the differences in their performance indicators were compared. Compared with the traditional scheduling game management algorithm, the new scheduling game management algorithm had an average improvement of 38.5% in multiple performance indicators. At present, however, the research and building of BD technology infrastructure has not been well popularized, so the new algorithm proposed in this paper has limited application.

Keywords: scheduling game management; big data; energy control; energy report

Cite As

W. Zhao, Y. Ai, " Algorithm of Scheduling Game Management Based on Big Data Integrated Energy Management and Control"
Engineering Intelligent Systems, vol. 33 no. 2, pp. 111-120, 2025.

Application and Evaluation of Intelligent Management Accounting Platform Based on Association Rule Algorithm and PS-DR-DP Model

2025-03-29T21:57:05-07:00

Management accounting plays a key role in unit-level planning, decision-making, control and evaluation and is an important subfield of accounting. It is mainly intended to meet the internal management requirements of organizational units by utilizing relevant data and integrating financial and operational activities in a coherent manner. The management accounting platform is oneofthe centers of business management. Therefore, thestability and status of its platform requires extra attention. In this study, the researchers propose an intelligent management accounting platform assessment method based on big data association rules algorithm and a “pressure-support”, “destructiveness-resilience”, “degradation-promotion” model. The results revealed that the expert scores of the pressure and support indicator system ranged from 75 to 88, the destructive and resilient indicator system
ranged from 69 to 93, and the degradation and enhancement indicator system ranged from 78 to 88. The credibility values of the three sub-indicator systems were 0.908, 0.989, and 0.955 respectively. Most of the carrying contribution values were below 0.9, and only the carrying state of support and degradation power in 2023 exceeded 1.0, at 1.6894 and 1.0832 respectively. Hence, the system was better equipped to withstand outside pressure as support, resilience, and deterioration all increased yearly along with the support capacity. The system damage or pressure resilience was improved, although the system may have some long-term degradation trends. The carrying state value also increased to 1.3956, and the mean value of carrying contribution increased significantly from 0.6025 to 0.9527 in 2023, indicating that the overall carrying capacity and contribution of the system is increasing year by year. It can be concluded that the proposed model can effectively assess the state of an intelligent management accounting platform. This provides a technical basis for promoting the management accounting platform to assess decision-making efficiency, resource allocation optimization, continuous improvement and strengthening of competitiveness.

Keywords: Association rule algorithm; management accounting; PS-DR-DP model; platform evaluation

H. Li, X. Yu, "Application and Evaluation of Intelligent Management Accounting Platform Based on Association Rule Algorithm
and PS-DR-DP Model", Engineering Intelligent Systems, vol. 33 no. 2, pp. 121-130, 2025.

Association Rule Mining for English Digital Archive System Based on Improved Apriori Algorithm

2025-03-29T22:04:28-07:00

With the widespread use and implementation of digital technology, the English digital archive system has accumulated a significant amount of data encompassingvarious dimensionssuchaslearning behavior, teachingprocesses, andstudentfeedback. Extracting valuable information andknowledge from this vast data has now become crucial for educational research and management. To this end, it is suggested that an enhanced convolutional neural network be combined with an Apriori algorithm to design and optimize a digital archive management system and association rule mining for English language. The improved Apriori algorithm takes into account the data’s peculiarities and mining demands, thereby yielding comprehensible and high-quality results. The study’s outcomes revealed that when the system reached a maximum iteration of 114 times, the proposed method attained the highest fitness value of 98.25 on the training set, in comparison with other fitness values. Similarly, when the proposed method was tested on the validation set, it achieved a fitness value of 98.74 after reaching a maximum iteration of 61 times, compared to fitness values obtained by other methods. The change in the overfitting curves shows that the model’s performance in managing the data was stabilized after 30 iterations. With the optimized training model, the accuracy of data manipulation progressively increased to a consistent 90% and eventually converged to 99.99%. In practical applications, the test scores obtained by the proposed system for data transmission, preservation, retrieval, and management, all surpassed
84 points–a significant improvement. Notably, the system’s data management score exceeded 92 points. The research findings demonstrate the clear advantages of the new methodology over the traditional approach in regards to accuracy and operational efficiency. It also indicates the proposed method’s capacity to effectively manage the vast amount of data in the English digital archives of colleges and universities, thus yielding robust data support for research and management purposes in English education.

Keywords: Improved Apriori algorithm; English language; digital archiving system; association rules; educational management

Cite As

X. Fang, " Association Rule Mining for English Digital Archive System Based on Improved Apriori Algorithm",
Engineering Intelligent Systems, vol. 33 no. 2, pp. 131-140, 2025.

Coupling of Image Authentication and Identification of Logistics Blockchain Based on Cloud Computing

2025-03-29T22:10:02-07:00

There is a one-to-one correspondence between logistics information and express order numbers. If the logistics information is directly disclosed in the blockchain, it is easy to infer the addresses of the recipient and the sender, and the personal privacy of the user cannot be guaranteed. Once information leakage or information theft occurs, serious consequences can ensue. With regard to the protection of digital image copyright, transaction deposit and digital image copyright authentication in the management of existing logistics blockchain image authentication and identification, user privacy protection, personalized recommendation, etc., in this study, digital watermarking, blockchain technology and recommendation technology is applied to digital image transaction management. We propose an intelligent recommendation algorithm for personalized digital images based on weighted TextRank and SOM. A logistics blockchain image transaction management system is designed that can solve the current problems in the
cloud computing environment. It benefits both buyers and sellers. The median filter attack experiment found that after the image to be detected is attacked by the median filter with the filter window size ranging from 2 to 99, all of the NC values of the extracted robust watermark are 1. The NC values of the extracted semi-fragile watermarks range from 0.9696 to 0.8909, and theTAF values of the semi-fragile watermarks range from 0.0676 to 0.1808. The trusted framework based on blockchain for outsourcing data entry (TFBO) in this paper is not only credible but also practical. This study helps to solve the privacy protection problem in the logistics process and improve the user’s trust in the logistics company.

Keywords: logistics blockchain, image authentication recognition, cloud computing, user privacy protection

Cite As

S. Liu, " Coupling of Image Authentication and Identification of Logistics Blockchain Based on Cloud Computing",
Engineering Intelligent Systems, vol. 33 no. 2, pp. 141-153, 2025.

Design of Function Approximation Algorithm Based on RBF Neural Network

2025-03-29T22:14:56-07:00

Function approximation is an important part of function theory, and its role in numerical computation is crucial. The application of a neural network in function approximation is an innovative means of developing function approximation. However, most of the current researches compare the function approximation of various networks in depth. The purpose of this study is to investigate ways to examine and analyze function approximation based on a neural network, and describe a radial neural network. This study addresses the problem of function approximation algorithm design. Because this is based on an artificial neural network, a typical neural network is taken as an example, and the related concepts and algorithms are described in detail. The radial basis function neural network (RBFNN) for function approximation is designed and analyzed using experimental simulation. According to
the results of RBFNN and BPNN simulation experiments, the errors of RBFNN are close to 0, while the errors of BPNN have large oscillations around 0. From the results and analysis, it can be concluded that the approximation effect of RBFNN is better than that of BPNN, providing an important reference value for research on function approximation.

Keywords: function approximation, artificial neural network, radial neural network, BP neural network

Cite As

L. Hou, " Design of Function Approximation Algorithm Based on RBF Neural Network", Engineering Intelligent Systems, vol. 33 no. 2, pp. 155-167, 2025.

Effectiveness of Improved Personalized Intelligent Intelligent Systems Recommendation Model for Travel Information by Collaborative Filtering Algorithm Incorporating User Interests

2025-03-29T22:21:20-07:00

A critical issue for customized intelligent travel information recommendation is the exponential growth of travel information and the wide range of travel needs. In order to address the issues of cold start and poor recommendation results yielded by the recommendation system, in this study, a collaborative filtering method is proposed that integrates user interests. This algorithm is an item-oriented recommendation algorithm called iExpand algorithm. The algorithm views user conduct as a collection of relatively implicit interests that need to be examined, with the specific preferences of each individual user being considered. As a result, the algorithm is better able to examine the user’s interests and grasp how those interests change over time. The iExpand algorithm has a 0.962 accuracy, a 0.038 loss rate, a 0.033 suggestion error, and a runtime of 1.04s. The findings for the customized intelligent recommendation model for trip information showed that the proposed algorithm has a greater recommendation accuracy and a quicker reaction time, resolving the issues of cold start.

Keywords: iExpand; user interests; collaborative filtering algorithm; personalisation; travel information

Cite As

Y. Li, X. Gou, "Effectiveness of Improved Personalized Intelligent Intelligent Systems Recommendation Model for Travel Information by Collaborative Filtering Algorithm Incorporating User Interests", Engineering Intelligent Systems, vol. 33 no. 2,
pp. 169-178, 2025.

Real-time Integration Technology of Large-scale Heterogeneous Data Based on Big Data and Artificial Intelligence

2025-03-29T22:26:59-07:00

In today’s era of increasingly sophisticated network communication technology, large-scale heterogeneous data are emerging, and people’s requirements for data integration technology are increasing. To address this challenge, many scholars have begun to combine big data with artificial intelligence to integrate large-scale heterogeneous data in real time. Since the neural network can fully take into account the characteristics of the data, it has strong upgrade ability and fault resistance, and does not need to know the real-time data integration for specific instance training in advance. Hence, it is able to deal with the real-time integration problem of large-scale heterogeneous data. In order to increase the speed and effectiveness of data integration, this study presents a neural network based on big data and artificial intelligence to assess large-scale heterogeneous data real-time
integration technology. According to this study’s experimental findings, the particle swarm-BP neural network has an error of about 0.020 and the BP neural network has an error of about 0.021 in terms of training results for the three techniques. The enhanced particle swarm-BP neural network technique has an inaccuracy of 0.15 to 0.2; hence, with the enhanced particle swarm-BP neural network technique, there is significantly less error. Also, the improved algorithm requires much less training time than the other two algorithms in the range of 114ms-121ms, indicating that the algorithm has better integration efficiency. Therefore, this method is very effective for the real-time integration of large-scale heterogeneous data.

Keywords: heterogeneous data, neural networks, artificial intelligence, big data

Cite As

J. Zhang, "Real-time Integration Technology of Large-scale Heterogeneous Data Based on Big Data and Artificial Intelligence", Engineering Intelligent Systems, vol. 33 no. 2, pp. 179-188, 2025.

Research on Knowledge Tracking in Foreign Language Teaching Based on Neural Network Computing

2025-03-29T22:31:39-07:00

Knowledge tracking offers new opportunities for integrating intelligent assistance in foreign language teaching, greatly promoting foreign language teaching in terms of research generally to accurate tracking of student progress, and suggestions. When teaching a foreign language, it is essential that teachers track students’ knowledge acquisition scientifically and accurately. In order to establish and optimize the foreign language information assisted teaching environment, this paper first proposes the integration of big data in foreign language teaching, and improves the mutual adaptation of the teaching environment. Next, it strengthens the construction of teaching informatization, then conducts an in-depth exploration of knowledge measurement in foreign language teaching and the tracking of students’ knowledge level. Finally, the paper demonstrates that the proposed knowledge tracking model can improve students’ overall learning outcomes. By means of effective, multi-dimensional and multi-index analysis methods, the paper explores the main problems that currently challenge the teaching of a foreign language. Through the data collection, knowledge point coding, testing and analysis in the process, the neural network computing knowledge tracking model is designed to indicate the students’ recall of information, effectively track students’ level of knowledge mastery, and optimize and improve teaching practices. The knowledge tracking study of 352 students shows that it is feasible to divide knowledge points according to 15 semantics, 46 phrases, 63 grammars, 95 structures and 126 logical relationships. Data preprocessing ofknowledgetracking is necessary, asit determines the effectiveness andstability of knowledge tracking, and indicates that students’ knowledge level and forgetting level follow certain rules. The reasoning ability of the model is improved after integrating knowledge tracking and forgetting information. The effective application of the knowledge tracking model can improve teachers’ teaching efficacy and students’ autonomous learning ability. The proposed knowledge tracking model and path for foreign language teaching is a new method that strengthens the driving force of informatization and provides a new means of improving and fostering foreign language learning.

Keywords: image dehazing; neural network computing; information management; knowledge tracking; foreign language teaching; model design

Cite As

H. Wu, " Research on Knowledge Tracking in Foreign Language Teaching Based on Neural Network Computing",
Engineering Intelligent Systems, vol. 33 no. 2, pp. 189-200, 2025.

Student Concentration Recognition Model in English Classroom Based on ResNet50 andVGG16

2025-03-29T22:38:11-07:00

Students’ classroom behavior can indicate their listening status in real time, and teachers judge students’ listening status based on their own experience and students’ behavior. However, this method greatly tests teachers’ teaching experience and energy, and it is difficult to achieve good results on this basis. Therefore, this study proposes an English classroom student focus recognition model based on deep residual networks and visual geometry group networks. The model selects the “You Only Look OnceV3” algorithm as the target position detector, and then uses deep residual networks and visual geometry group networks to recognize and classify student classroom behavior, thereby determining each student’s class state. The experimental
results show that the accuracy of the cropped model is significantly higher than that before cropping. When the number of iterations reaches 500, the accuracy of the model before and after image cropping in the visual geometry group network model is 0.88 and 0.97, respectively. In regard to the deep residual network model, the accuracy of the model before and after image cropping is 0.86 and 0.98, respectively. For the dual network hybrid model, the model accuracy before and after image cropping was 0.90 and 0.99, respectively. The research results indicate that the proposed dual network hybrid algorithm model has excellent performance in recognizing student state.

Keywords: Behavior recognition, transfer learning, ResNet50,VGG16, concentration

Cite As

D. Tang, Q. Jiang, "Student Concentration Recognition Model in English Classroom Based on ResNet50 andVGG16",
Engineering Intelligent Systems, vol. 33 no. 2, pp. 201-211, 2025.

The Evaluation of Enterprise Financing Structure Capability Based on RF-CART Integrated Algorithm

2025-03-29T22:44:12-07:00

The continuous advancement of information technology and the expansion of application scenarios have highlighted the advantages of data fusion and personalized evaluation of computer technology, which has great potential for application in the financing field of small and medium-sized enterprises. Based on the current information asymmetry and limitations of subjective evaluation, an evaluation model to determine the capability of a financing structure is proposed, based on the RF-CART integrated algorithm. Firstly, the impact indicators of financing structure for small and medium-sized enterprises were collected, a principal component analysis was conducted, and the characteristic variables for the indicators were constructed. Secondly, the CART decision tree was used as a weak learner to construct a credit evaluation model under the RF model, and the classification results processed by multiple decision trees were integrated to obtain the optimal classification result. By using indicator testing and
cross validation to analyze the effectiveness of the model algorithm, it was found that there is a negative correlation between the tax effect and financing capacity of small and medium-sized enterprises. The testing and training time of ensemble learning algorithms are both less than 21 seconds, with an average recognition accuracy of over 95%. The accuracy difference of other comparative calculation methods is 4%, and their AUC area for recognition performance is 0.915, indicating good sample discrimination ability. The proposed method can effectively determine the operational status of enterprises and provide indicators and warnings related to financing risks and any decisions pertaining to growth.

Keywords: integrated learning; small and medium-sized enterprises; financing structure; KS; random forest algorithm

Cite As

Z. Meng, M. Liu, B. Li, " The Evaluation of Enterprise Financing Structure Capability Based on RF-CART Integrated Algorithm", Engineering Intelligent Systems, vol. 33 no. 2, pp. 213-221, 2025.

Transport Vehicle Route Decision Based on Particle Swarm Optimization Algorithm in Logistics Distribution Management

2025-03-29T22:49:42-07:00

In recent years, the economic boom has prompted various industries to usher in better development prospects, including the continuous development of the logistics industry. In all aspects of logistics, the most important issue is the issue of distribution, which corresponds to the issue of transportation timeliness. The transportation timeliness is closely related to the route of the transportation vehicle to a large extent, so the optimal selection of the transportation route plays an important role in solving the timeliness problem of distribution. In order to better optimize the transportation routeof transportation vehicles, this paper will use particle swarm optimization algorithm to solve such decision-making problems. And through the logistics distribution and transportation vehicle route optimization problem, the logistics transportation optimization method based on the traditional particle swarm optimization algorithm and the logistics distribution method of the improved particle swarm optimization algorithm are used to construct the decision-making of the transportation vehicle transportation route. Through the simulation experiment of transportation route optimization based on particle swarm optimization, the particle optimization algorithm was run 99 times, of which the optimal solution is 57 times, and the probability of achieving the final efficient delivery success reaches 57%. The corresponding optimal solution results of 57.1 and 25.1 show that the transportation vehicle route based on the particle swarm optimization algorithm can successfully obtain the optimization results for the route, achieving the timeliness of distribution in logistics transportation.

Keywords: Logistics Distribution Management, Particle Swarm Optimization Algorithm, Transportation Vehicle Route, Decision Optimization

Cite As

L. Yu, L. Tu, "Transport Vehicle Route Decision Based on Particle Swarm Optimization Algorithm in Logistics Distribution Management", Engineering Intelligent Systems, vol. 33 no. 2, pp. 223-233, 2025.