Research on entity recognition and alignment of APT attack based on Bert and BiLSTM-CRF by Xiuzhang YANG, Guojun PENG, Zichuan LI, Yangqi LYU, Side LIU, Chenguang LI

“…Compared with the CRF model, LSTM-CRF model,GRU-CRF model, BiLSTM-CRF model, CNN-CRF model, and Bert-CRF model, the F1-score values of the proposed model are improved by 27.42%, 18.78%, 23.62%, 13.25%, 14.88%, and 14.46%. …”
Get full text

Research on entity recognition and alignment of APT attack based on Bert and BiLSTM-CRF by Xiuzhang YANG, Guojun PENG, Zichuan LI, Yangqi LYU, Side LIU, Chenguang LI

“…Compared with the CRF model, LSTM-CRF model,GRU-CRF model, BiLSTM-CRF model, CNN-CRF model, and Bert-CRF model, the F1-score values of the proposed model are improved by 27.42%, 18.78%, 23.62%, 13.25%, 14.88%, and 14.46%. …”
Get full text

Design and interactive performance of human resource management system based on artificial intelligence. by Yangda Gong, Min Zhao, Qin Wang, Zhihan Lv

“…The experimental results demonstrate that the algorithm optimized by the Nadm has shown improved convergence speed and forecast effect, with 187 iterations. …”
Get full text

Privacy-preserving federated learning framework with dynamic weight aggregation by Zuobin YING, Yichen FANG, Yiwen ZHANG

“…There are two problems with the privacy-preserving federal learning framework under an unreliable central server.① A fixed weight, typically the size of each participant’s dataset, is used when aggregating distributed learning models on the central server.However, different participants have non-independent and homogeneously distributed data, then setting fixed aggregation weights would prevent the global model from achieving optimal utility.② Existing frameworks are built on the assumption that the central server is honest, and do not consider the problem of data privacy leakage of participants due to the untrustworthiness of the central server.To address the above issues, based on the popular DP-FedAvg algorithm, a privacy-preserving federated learning DP-DFL algorithm for dynamic weight aggregation under a non-trusted central server was proposed which set a dynamic model aggregation weight.The proposed algorithm learned the model aggregation weight in federated learning directly from the data of different participants, and thus it is applicable to non-independent homogeneously distributed data environment.In addition, the privacy of model parameters was protected using noise in the local model privacy protection phase, which satisfied the untrustworthy central server setting and thus reduced the risk of privacy leakage in the upload of model parameters from local participants.Experiments on dataset CIFAR-10 demonstrate that the DP-DFL algorithm not only provides local privacy guarantees, but also achieves higher accuracy rates with an average accuracy improvement of 2.09% compared to the DP-FedAvg algorithm models.…”
Get full text

Enhancing the e-commerce shopping experience with IoT-enabled smart carts in smart stores by Jian Wang

“…The findings demonstrate that the proposed IoT-enabled model efficiently balances classification tasks, reduces congestion during algorithm execution, and minimizes energy waste. …”
Get full text

Leader-Follower Game Mechanism and Strategy of Industrial Park Demand Response with User Aggregator by Zhangyi LI, Xin MA, Wei PEI, Liang ZHANG, Mingshuang LIU, Qiuhong LIANG

“…Besides, two stage-optimization algorithms are used to solve this model. …”
Get full text

Design and Implementation of Local Threshold Segmentation Based on FPGA by Shangshang Gao, Yuanyuan Wang, Zhaofeng Chen, Feng Zhou, Rugang Wang, Naihong Guo

“…Finally, the design algorithm is verified by ModelSim simulation software and QT5 software. …”
Get full text

Secure beamforming design for IRS-assisted SWIPT Internet of things system by Zhengyu ZHU, Jinlei XU, Gangcan SUN, Ning WANG, Wanming HAO

“…In order to meet the new requirements of intelligent signal processing deployment and physical layer security for green interconnection of things, a design method of secure beamforming was proposed to solve the problem of the shortage of sustainable energy supply in the information and energy transmission at the same time Internet of things (IoT) system assisted by intelligent reflecting surface (IRS).Considering the constraints of secrecy rate, transmit power and IRS reflection phase shift, the optimization problem was modeled as a non-convex quadratic programming problem with quadratic constraints, aiming at maximizing the acquisition power of energy collector, and jointly optimizing the base station transmit beamforming matrix, jammer covariance matrix and IRS phase shift.The non-convex quadratic problem was transformed into an equivalent convex problem by using the relaxation variable, semidefinite relaxation method, auxiliary variable and sequence parameter convex approximation method, and an alternative iterative optimization algorithm was proposed to obtain the feasible solution of the original problem.Simulation results show that the proposed algorithm can converge quickly and improve performance effectively compared with the benchmark scheme.…”
Get full text

A Deep Learning Method for Photovoltaic Power Generation Forecasting Based on a Time-Series Dense Encoder by Xingfa Zi, Feiyi Liu, Mingyang Liu, Yang Wang

“…Deep learning has become a widely used approach in photovoltaic (PV) power generation forecasting due to its strong self-learning and parameter optimization capabilities. In this study, we apply a deep learning algorithm, known as the time-series dense encoder (TiDE), which is an MLP-based encoder–decoder model, to forecast PV power generation. …”
Get full text

An Evolutionary Deep Reinforcement Learning-Based Framework for Efficient Anomaly Detection in Smart Power Distribution Grids by Mohammad Mehdi Sharifi Nevisi, Mehrdad Shoeibi, Francisco Hernando-Gallego, Diego Martín, Sarvenaz Sadat Khatami

“…This optimization ensures improved accuracy, faster convergence, and better generalization to unseen data. …”
Get full text

Quantitative evaluation and obstacle factor diagnosis of drug regulatory capacity in China. by Mingming Zhai, Liwen Huang, Shijie Sun, Liying Cao, Xueqiong Yue, Yuanxia Hu

“…<h4>Methods</h4>Using the methods of literature research, expert interviews, investigation and analysis, the quantitative evaluation indicator system of supervision ability was established in all directions; the indicator data were collected and quantified; the indicator weight setting algorithm of the evaluation system was improved and the indicator weight was set by combining AHP and entropy method; the differences among eastern, central, and western provincial-level regions were analyzed by variance analysis; panel data were constructed for spatio-temporal evolution analysis; obstacle factor diagnosis model was used to analyze the obstacle factors.…”
Get full text

A Minimal Path-Based Method for Computing Multistate Network Reliability by Xiu-Zhen Xu, Yi-Feng Niu, Can He

“…To advance the solution efficiency of d-MPs, an improved model is developed by redefining capacity constraints of network components and minimal paths (MPs). …”
Get full text

A localization strategy combined with transfer learning for image annotation. by Zhiqiang Chen, Leelavathi Rajamanickam, Jianfang Cao, Aidi Zhao, Xiaohui Hu

“…Experiments on the Corel5k multilabel image dataset verify that CNN-2L improves the labeling precision by 18% and 15% compared with the traditional multiple-Bernoulli relevance model (MBRM) and joint equal contribution (JEC) algorithms, respectively, and it improves the recall by 6% compared with JEC. …”
Get full text

High-Resolution Direction of Arrival Estimation of Underwater Multitargets Using Swarming Intelligence of Flower Pollination Heuristics by Nauman Ahmed, Huigang Wang, Shanshan Tu, Norah A.M. Alsaif, Muhammad Asif Zahoor Raja, Muhammad Kashif, Ammar Armghan, Yasser S. Abdalla, Wasiq Ali, Farman Ali

“…For this purpose, particle swarm optimization (PSO), minimum variance distortion-less response (MVDR), multiple signal classification (MUSIC), and estimation of signal parameter via rotational invariance technique (ESPRIT) standard counterparts are employed along with Crammer–Rao bound (CRB) to improve the worth of the proposed setup further. …”
Get full text

Real-Time Recognition of Loading Cycles’ Process Based on Electric Mining Shovel Monitoring by Bonan Wang, Yun Duan, Wei Xu

“…Moreover, the dynamic time warping (DTW) algorithm was used to detect and classify the preliminary recognition results by optimizing its distance threshold parameters, reducing the error rate of the model. …”
Get full text

Apple Pest and Disease Detection Network with Partial Multi-Scale Feature Extraction and Efficient Hierarchical Feature Fusion by Weihao Bao, Fuquan Zhang

“…To address this issue, this study proposes an improved pest and disease detection algorithm, YOLO-PEL, based on YOLOv11, which integrates multiple innovative modules, including PMFEM, EHFPN, and LKAP, combined with data augmentation strategies, significantly improving detection accuracy and efficiency in complex environments. …”
Get full text

Application of Markov processes for analysis and control of aircraft maintainability by Yu. M. Chinyuchin, A. S. Solov'ev

“…In order to reduce the number of the mathematical operations for the analysis of aeronautical engineering maintainability by using non-stationary Markov processes an algorithm for their optimization is presented. The suggested methods of the analysis by means of Markov chains allow to execute comparative assessments of expected maintenance and repair costs for one or several one-type objects taking into account their original conditions and operation time. …”
Get full text

Berth Allocation and Quay Crane Assignment Considering the Uncertain Maintenance Requirements by Siwei Li, Liying Song

“…To address this novel problem, we propose a proactive-reactive method that incorporates a reliability-based model into the Swarm Optimization with Differential Evolution (SWO-DE) algorithm. …”
Get full text

Machine learning based assessment of hoarseness severity: a multi-sensor approach centered on high-speed videoendoscopy by Tobias Schraut, Anne Schützenberger, Tomás Arias-Vergara, Melda Kunduk, Matthias Echternach, Stephan Dürr, Julia Werz, Michael Döllinger

“…Subjects were classified into two hoarseness groups based on auditory-perceptual ratings, with predicted scores serving as continuous hoarseness severity ratings. A videoendoscopic model was developed by selecting a suitable classification algorithm and a minimal-optimal subset of glottal parameters. …”
Get full text

Detection of Critical Parts of River Crab Based on Lightweight YOLOv7-SPSD by Guoai Fang, Yu Zhao

“…These additions help achieve an initial reduction in model size while preserving detection accuracy. Furthermore, we optimize the model by removing redundant parameters using the DepGraph pruning algorithm, which facilitates its application on edge devices. …”
Get full text