International Journal of Engineering Intelligent Systems

A Study on Computer Simulation and User Experience Evaluation for Product Interface Design from the Perspective of Human-Computer Interaction

Hui Wang, Zhenfeng Qu — Wed, 01 Jul 2026 00:00:00 -0700

The current user interface design processes do not include early quantitative estimates of how users interact with and experience aninterface. Therefore, when poor usability and unsatisfactory user experience are discovered, the issues often emerge only after the product has been released. This paper proposes using an interface design analysis method that incorporates insight into user interface interaction modeling, computer simulation, and user experience evaluation within a human-computer interaction context. By formally modeling the interface structure, user tasks, and operational behaviors, computer simulations may be used to predict the user’s path, the success of their task completion, and the distribution of errors.The simulation results are then validated against multi-dimensional user experience evaluation indicators. This enables the systematic evaluation and
optimization of interface design during the early stages of development. Experimental results show that the simulated user operation paths maintain 88% consistency with real user behavior under interface scheme S1, with the lowest average completion time for core tasks at 32 seconds; in S5, the highest error frequency is 8.1%. After simulation-driven optimization, task completion time is reduced to 32–44 seconds, and the overall user experience score improves to above 4.0/5. The research results demonstrate that the proposed method provides a model- and data-driven approach for early-stage validation and experience evaluation during the design of an interface.

Keywords: human-computer interaction modeling; interface interaction simulation; user experience prediction; operational path simulation; task performance evaluation

Cite As

H. Wang, Z. Qu, "A Study on Computer Simulation and User Experience Evaluation for Product Interface Design from the Perspective of Human-Computer Interaction", Engineering Intelligent Systems, vol. 34 no. 4, pp. 441-453, 2026.

Adaptive Modeling and Anti-error Correction Algorithm for Abnormal Behavior of Power Load Based on Reinforcement Learning

Liang Xiao, Guanghu Xu, Shijin Xu, Quan Yuan, Xiaobo Zhang — Wed, 01 Jul 2026 00:00:00 -0700

In regard to the high uncertainty and error correction risk problems in the identification of abnormal behavior of power load, this paper proposes an adaptive modeling and anti-error correction algorithm based on reinforcement learning. In the adaptive modeling stage, a Markov decision process with random perturbations is constructed; multi-scale state representation is integrated; and the load fluctuation characteristics are dynamically captured. In the anti-error correction stage, a hierarchical safety constraint action space and an interpretable reward function are designed, and the attention-driven strategy network is combined to focus on key features. A dynamic safety barrier module based on load fluctuation entropy is applied
to achieve risk perception and real-time interception of high-risk correction. Experiments show that the retention rate of the proposed model reaches 0.87, and the correction success rate is 92% under a noise intensity of 20db, which is up to 22% higher than the baseline model.

Keywords: reinforcement learning; power load anomaly identification; adaptive modeling; anti-error correction; multi-scale state representation

Cite As

L. Xiao, G. Xu, S. Xu, Q. Yuan, X. Zhang, "Adaptive Modeling and Anti-error Correction Algorithm for Abnormal Behavior of Power Load Based on Reinforcement Learning", Engineering Intelligent Systems, vol. 34 no. 4, pp. 455-468, 2026.

Coupled Optimization Algorithm of Intelligent Recommendation Model Based on Computer Natural Language Processing

Dianqing Bao, Wen Su — Wed, 01 Jul 2026 00:00:00 -0700

To address the issues of fragmented semantic representation of items and the difficulty of capturing dynamic user interests in traditional recommendation models, this paper proposes a coupling optimization algorithm for intelligent recommendation models based on natural language processing (NLP). A heterogeneous multi-source, information-deep, semantic-unified representation module is constructed, integrating knowledge graph embedding and a Bi-LSTM (Bidirectional Long Short-Term Memory) model to combine structured and unstructured item information. A time-aware user dynamic interest coupling modeling module is designed, using GRU (Gated Recurrent Unit) sequence encoding to capture behavioral temporal dependencies and combining an interest-semantic coupling attention mechanism to dynamically track interest evolution. A multi-objective collaborative adversarial coupling optimization module is introduced toachieve multi-objective collaborative optimization. Experiments show that the proposed model achieves 0.937 on Recall@20 and 0.959 on NDCG@20 (Normalized Discounted Cumulative Gain); the intra-list similarity is as low as 0.25, and the coverage
reaches 0.934.

Keywords: computer natural language processing, intelligent recommendation models, coupling optimization, dynamic interest coupling, deep semantic unification

Cite As

D. Bao, W. Su, "Coupled Optimization Algorithm of Intelligent Recommendation Model Based on Computer Natural Language Processing", Engineering Intelligent Systems, vol. 34 no. 4, pp. 469-481, 2026.

Creation of Augmented Reality and Virtual Reality Technology in VR Animation Design

Tao Zhang — Wed, 01 Jul 2026 00:00:00 -0700

VR refers to virtual reality. In the animation industry, VR technology offers people a new type of animation experience. Nowadays, VR animation is being used in a wide range of industries, bringing huge economic benefits. However, there are several problems that cannot be ignored in the current design and production of VR animation, mainly due to the use of a single technology. Relying on a single virtual reality technology for design and production will lead to unsatisfactory results and poor quality VR animations. Therefore, this present study integrated augmented reality technology and virtual reality technology into VR animation design and creation, and developed a model based on the UNITY 3D platform. After simulating 50 samples, the experimental results showed that the application of the improved development mode could increase the production effect of VR
animation design by 7.04%, which enables VR animation to benefit the social economy and cultural construction.

Keywords:VR animation design, augmented reality technology, virtual reality technology, UNITY 3D

Cite As

T. Zhang, "Creation of Augmented Reality and Virtual Reality Technology in VR Animation Design",
Engineering Intelligent Systems, vol. 34 no. 4, pp. 483-491, 2026.

Cross Modal Image and Text Retrieval Algorithm Based on Deep Learning

Jiana Bi — Wed, 01 Jul 2026 00:00:00 -0700

Multi-modal data contains rich and complementary information that provides a more comprehensive means of understanding and analyzing things deeply. Cross-modal image retrieval is related to the effective association between two different modal data, which can accurately retrievea semantically matched image through the input text, or accurately find the corresponding text through the input image. In this study, the algorithm model is built based on deep learning, and the dual tower structure and multi-modal fusion structure are combined to give full play to the advantages of different structures. Image construction based on CNN is responsible for feature extraction of input images, and text construction based on Transformer is used to accurately capture semantic dependencies in text. The addition of a fusion layer to the twin tower structure can better mine the potential
semantic associations between images and text. Two open data sets, MSCOCO and Flickr30k, were used to carry out experiments that demonstrated the superiority of the proposed model in cross-modal image retrieval task.

Keywords: deep learning; cross modal; image and text retrieval; twin tower structure; CNN; Transformer

Cite As

J. Bi, "Cross Modal Image and Text Retrieval Algorithm Based on Deep Learning", Engineering Intelligent
Systems, vol. 34 no. 4, pp. 493-499, 2026.

Exploration of Closed-loop Supply Chain Network Model Based on Mobile Wireless Communication

Jinjin Guo — Wed, 01 Jul 2026 00:00:00 -0700

With the continuous development of global resources and the increasing awareness of the need to protect the environment, remanufacturing, reverse logistics, especially the closed-loop supply chain (CLSC), has attracted extensive attention from the business community and academia. On the one hand, the CLSC can reduce energy consumption and environmental pollution, and achieve resource recycling. At the same time, it can also strengthen the competitiveness of the company to achieve the comprehensive benefits of “economy and environment”. By establishing a CLSC network model based on mobile communication technology, this paper introduces CLSC management into the mobile Internet era, and provides new ideas and methods for solving problems such as the inability to directly connect consumers and production enterprises under the traditional market transaction mode, and the inability to timely understand production progress and quality. The experimental results showed that the material value changed from 11180 to 9800 in the whole CLSC process, and the change was small after removing the normal loss. Therefore, the stability of the model could be confirmed, whicheffectively promoted the informatization and datalization of logistics management, and improved the enterprise operation efficiency.

Keywords: closed-loop supply chain; wireless communication; sustainable development; network model

Cite As

J. Guo, "Exploration of Closed-loop Supply Chain Network Model Based on Mobile Wireless Communication",
Engineering Intelligent Systems, vol. 34 no. 4, pp. 501-510, 2026.

Intelligent Investigation of Manufacturing Supply Chain Management System Based on Industrial Big Data

He Yu — Wed, 01 Jul 2026 00:00:00 -0700

In today’s society, supply chain management systems are becoming more widely used, and their complexity is increasing. Suppliers and customers are distributed all over the world, leading to the traditional manufacturing supply chain management system being unable to operate optimally in the global context. Therefore, this current study uses industrial big data technology to examine the intellectualization of the manufacturing supply chain management system. Taking the supply chain management system of B manufacturing enterprise as an example, the study first determined the procurement process of the supply chain. Based on the comparison between common procurement quantities and actual demand results, it showed that the intelligent model of the system had stronger capture ability than traditional supply chain management systems, could timely understand market changes, and reduced procurement losses. The procurement volume was close to the actual demand. Secondly, the production process of the supply chain was examined. Based on the comparison of the total production quantity and qualified quantity results of the workshop within the specified time, it was found that the intelligent model of the system had higher production efficiency and quality every time compared to traditional supply chain management systems, especially with a production qualification rate exceeding 0.95 every time. Next, the warehousing process of the supply chain was examined. Based on the comparison of each count duration and the number of invalid counts, it was found that the total count duration
of the system’s intelligent model ranged from 30 minutes to 34 minutes, and the number of invalid counts for each experiment ranged from 1 to 5. The efficiency of each inventory was higher than that of traditional supply chain management systems. Finally, the logistics link of the supply chain was determined. Based on the comparison of logistics accuracy and logistics cost results, it was found that the logistics accuracy of traditional supply chain management systems was lower than 80%, while the logistics accuracy of intelligent system models was higher than 80%. The logistics costs generated by the intelligent system model were less than those generated by traditional supply chain management systems.

Keywords: supply chain management system; industrial big data; system intelligence; data acquisition and pre-processing; modeling analysis; decision support

Cite As

H. Yu, "Intelligent Investigation of Manufacturing Supply Chain Management System Based on Industrial Big Data",
Engineering Intelligent Systems, vol. 34 no. 4, pp. 511-521, 2026.

Potential Threats and Prevention Strategies of IoT Robots to Network Information Security

Wanrong Bai, Zuohu Chen, Lei Zhang, Di Wang, Rangle Wu — Fri, 16 Oct 2026 00:00:00 -0700

This paper presents a range of holistic prevention strategies comprising hardware attack mitigations, firmware vulnerabilities management, vulnerability patches of the operating system, malware protections, encryption solutions for communicating information, strong authentications, access permission management, automatic systems updating mechanisms, vulnerability patches management, and the monitoring of networking and detection of anomalies. Through practical application, the effectiveness of the protection mechanisms for IoT robot security and performance has been gradually integrated and evaluated. In addition, performance monitoring tools such as PerfMon and top have been used to log system response time and resource
consumption of an IoT robot under varying system loads after security measures had been deployed. The system response time did increase slightly, but it was still within an acceptable range, and did not have a significant impact on the performance of the system itself. Overall, the proposed holistic prevention strategies demonstrated its ability to provide a level of security protection to the IoT robot.

Keywords: Internet of Things robots; network information security; potential threats; prevention strategies; vulnerability detection

Cite As

W. Bai, Z. Chen, L. Zhang, D. Wang, R. Wu, "Potential Threats and Prevention Strategies of IoT Robots to Network Information Security", Engineering Intelligent Systems, vol. 34 no. 4, pp. 523-536, 2026.

Prediction of College Student Employment Characteristics Based on Deep Manifold Learning

Wenzhi Su — Wed, 01 Jul 2026 00:00:00 -0700

The employment of college students is of great significance to the development of the national economy. Due to the information gap between enterprisesand universities, the use of advanced artificial intelligence technology to analyzeand predict student information can provide reliable employment positioningreferences and effectively alleviate the current tight employment situation for college students. To improve the accuracy of predicting college students’ employment trends, a deep learning network based on the geometric structure of datamanifolds is proposed for forecasting employment characteristics. First, employment data from a university were collected and normalized, and covariance featurematrices were computed to
form a Lie group manifold. Second, by integrating differential geometry theory with deeplearning principles, a deep dataset recognition network based on Riemannian manifolds was developed. During model training, feature weight coefficients and learning rates were dynamically adjusted to obtain the optimal network model. Experimental results demonstrate that the deepmanifold learning algorithm achieves a 5.52%improvement in prediction accuracy compared to conventional convolutional neural network models, while also exhibiting higher efficiency and robustness. The predictive outcomes of this model can provide valuable insights for university employment departments, improving the foresight, targeting, and precision of
employment-related initiatives.

Keywords: college student employment; Lie group manifold; deep manifold learning;employment prediction

Cite As

W. Su, "Prediction of College Student Employment Characteristics Based on Deep Manifold Learning",
Engineering Intelligent Systems, vol. 34 no. 4, pp. 537-543, 2026.

Research on Task Allocation of Crowdsensing User Activity Based on the ResNet Neural Network

Li Zhang, Shukui Zhang, Ye Tao, Pingquan Wang, Hao Long — Wed, 01 Jul 2026 00:00:00 -0700

Currently, there is a lack of multi-dimensional research on sensing task allocation in mobile crowdsensing networks. This paper proposes a task allocation algorithm based on the positive association between sensing entities. First, sensing activity is measured by the number of successful completions of sensing tasks by participating users, divided into direct and indirect activities. Meanwhile, a correlation model is constructed by combining direct and indirect activities. Furthermore, to improve the success rate of sensing task allocation, the ResNet neural network is trained to eliminate the gradient vanishing problem. Moreover, the AM Softmax cross-entropy loss function is adopted to narrow the gap between peers and expand the gap between different classes, thereby improving the accuracy of sensing task allocation. Finally, the constructed association model and the classification neural network are integrated for sensing task allocation. Experimental verification indicates that the proposed algorithm has significantly enhanced the accuracy of task allocation and can better provide accurate task allocation services.

Keywords: mobile crowdsensing; mobile users; activate; task allocation

L. Zhang, S. Zhang, Y. Tao, P. Wang, H. Long, "Research on Task Allocation of Crowdsensing User Activity Based on the ResNet Neural Network", Engineering Intelligent Systems, vol. 34 no. 4, pp. 545-557, 2026.

Role of Artificial Intelligence Robots in Data Processing and Privacy Protection

Pei Sun, Gang Wang, Zuohu Chen, Zhenguo Peng, Yuntao Dou — Wed, 01 Jul 2026 00:00:00 -0700

Data processing and data privacy are two pressing social, ethical and technological issues that have emerged as a result of the transformational influence of the digital era. The focus of this paper is to provide an overview of the role of AI robots with a focus on the use of LSTM to predict Internet traffic and how to implement the LSTM/Differential Privacy combination to harness AI for creating Internet traffic prediction models. As indicated by the tests that have been performed to evaluate the implementation of Differential Privacy, while it may impact the overall quality of the model’s output, a reduction in the model’s prediction quality, the model greatly increases the model’s level of data security and protection of users’ privacy. Additionally, by means of multiple layers of neuronal configurations, the model is able to generalize better when an additional Dropout layer is added and the Adam optimizer is utilized to avoid overfitting. Results of the experiments illustrated that the model is able to make high-quality predictions even with Differential Privacy included in its design, specifically for both datasets with a RMSE of 87.93 (DS1), 88.46 (DS2) and MAE of 64.25 (DS1), 65.73 (DS2), thus demonstrating the ability of AI to strike a balance between data and privacy value. The research results indicate that a stacked LSTM model can accurately and effectively model temporal characteristics in Internet traffic allowing for enhanced network resource optimization and security alerting.

Keywords: data processing; privacy protection; artificial intelligence; long short-term memory; differential privacy

Cite As

P. Sun, G. Wang, Z. Chen, Z. Peng, Y. Dou, "Role of Artificial Intelligence Robots in Data Processing and Privacy Protection", Engineering Intelligent Systems, vol. 34 no. 4, pp. 559-567, 2026.

The Impact of English Movies on Students’ Learning of English

Hongwei Zhen — Wed, 01 Jul 2026 00:00:00 -0700

This study was conducted to determine whether the use of English movies in the classroom has a positive impact on students’ learning of English. Two second-year undergraduate classes at the Langfang Open University in Hebei Province, China, were taken as the subjects, denoted as Class A and Class B. Class A students were taught by traditional means in addition to experiencing English movie-assisted teaching. At the end of the semester, the English learning outcomes of the two classes were compared, and a questionnaire survey was used to gather the students’views on English movie assisted teaching. The results showed that the final examination score for Class A (92.36 ± 11.05) was significantly higher than the score achieved for the start-of-term examination (87.64 ± 12.39) (p = 0.039 < 0.05), and was also significantly higher than that of Class B (87.02 ± 11.03). There
was no significant improvement in the start-of-term and final examination scores of Class B. The survey found that more than 90% of the students thought English movie-assisted teaching was useful for strengthening learning motivation, more than 70% thought it was useful for improving language outcomes, and more than 80% thought it was useful for enhancing cross-cultural awareness. Among the designed teaching activities, students thought movie dubbing and role-playing were the most useful. The results verify the positive impact of English movie assistance on students’ English learning, and the study demonstrates that the method has practical value when applied in the language-learning classroom.

Keywords: English movie, English learning, final examination score, questionnaire survey

Cite As

H. Zhen, "The Impact of English Movies on Students’ Learning of English", Engineering Intelligent Systems, vol. 34 no. 4,
pp. 569-573, 2026.