Computer and Modernization

Construction of Single-condition Triadic Concept and Its Fusion Recommendation Application

LIU Yuxuan1, 2, LIAO Yuchen1, LIU Zhonghui1

2024, 0(07): 1-6. doi:10.3969/j.issn.1006-2475.2024.07.001

Asbtract ( 106 )

PDF (988KB) ( 79 )

References | Related Articles | Metrics

Abstract: Triadic concept analysis has been introduced into the field of recommendation systems. However， the fusion step of concepts increases the complexity of constructing triadic concepts， and the concept information is not fully utilized in recommendation. This paper directly uses single-condition triadic concepts for recommendation， and designs a construction method and fusion recommendation algorithm for single-condition triadic concepts. Firstly， the triadic context is decomposed into multiple single-condition triadic contexts， and the concept proportion is designed as heuristic information to generate single-condition triadic concept. Then， the popularity of the recommended items on the single-condition triadic concepts is calculated， and the fusion recommendation confidence is designed by combining the item condition weight of the triadic context. Finally， the target user is recommended by combining the fusion recommendation confidence and the recommendation threshold. This paper conducts experiments on six public datasets. The results show that on datasets with low sparsity， the algorithm proposed in this paper is slightly better than the recommendation effects of GRHC and GreConD-kNN， and comparable to the effects of IBCF and kNN.

A Task Scheduling Method for Biological Gene Multi Sequence Alignment Algorithm

YANG Bo, WANG Hongjie, XU Shengchao, MAO Mingyang, JIANG Jinling, JIANG Darui

2024, 0(07): 7-12. doi:10.3969/j.issn.1006-2475.2024.07.002

Asbtract ( 112 )

PDF (1405KB) ( 70 )

References | Related Articles | Metrics

Abstract： Aiming at the problem of slow alignment efficiency in current biological gene multi sequence alignment algorithms when facing large-scale data， a task scheduling method for biological gene multi sequence alignment algorithms is proposed to improve the efficiency of biological gene multi sequence alignment. Firstly， the Trie tree method is used to segment biological gene multi sequence data， thereby optimizing the efficiency of data search and matching in the subsequent gene multi sequence alignment process; Secondly a gene multi sequence BWT index is constructed and the BWT index method is used to complete biological gene multi sequence alignment; Finally， based on the multi sequence alignment method， a heterogeneous parallel system of CPU and GPU is used to complete the task scheduling of multi sequence alignment. The experimental results show that the proposed task scheduling method for biological gene multi sequence alignment algorithm is more efficient， performs better， and is more suitable for practical applications.

Wind Power Prediction Method Based on STAGCN-Informer Spatiotemporal Fusion Model

YANG Shaozu1, 2, WANG Haicheng1, 2, WU Jinya1, 2, MA Jiying1, 2

2024, 0(07): 13-20. doi:10.3969/j.issn.1006-2475.2024.07.003

Asbtract ( 139 )

PDF (3491KB) ( 112 )

References | Related Articles | Metrics

Aiming at the problem that the spatial information cannot be effectively extracted due to the influence of spatiotemporal fluctuation and randomness in wind power forecasting， resulting in insufficient prediction accuracy， a model named STAGCN-Informer-DCP is proposed based on Variational Mode Decomposition （VMD），fusion of Spatiotemporal Attention Graph Convolutional Network （STAGCN） and improved Informer combination model. Firstly， VMD is used to perform modal decomposition on the original features， and the feature information on different time scales is extracted. At the same time， the selection of core parameters （penalty factor and K value） of VMD is optimized by using northern goshawk optimization （NGO）. Secondly， the STAGCN module that integrates spatio-temporal attention is used to dynamically capture the spatio-temporal features of the target wind turbine and its neighbors， and fuses them with the original signal components to obtain a feature vector carrying spatial scale information. Finally， the improved Informer model is used to extract the long-term dependencies of temporal context and realizes multi-step output prediction. The experimental results show that the combination model can better capture the dynamic space-time dependence， and effectively improve the accuracy of medium and long-term wind power forecasting.

Stock Price Prediction Based on Business Content to Construct Stock Association Relationships

YANG Jiang1, SUN Xiaomei1, XU Tao2

2024, 0(07): 21-25. doi:10.3969/j.issn.1006-2475.2024.07.004

Asbtract ( 119 )

PDF (1254KB) ( 95 )

References | Related Articles | Metrics

Traditional stock price prediction methods are mostly based on the time series of a single stock， ignoring the complex interrelationships between stocks. In response to this issue， the article proposes a stock price prediction method based on business content to construct stock correlation relationships from the perspective of building a more effective stock portfolio. The model consists of three components: the association relationship construction component， the temporal feature extraction component and the association capture component. The association relationship construction component uses improved TF-IDF to extract the similarity of business content keywords in the annual reports of listed companies to construct stock correlation relationships. The temporal feature extraction component uses LSTM to extract temporal features of stock trading data. The association capture component utilizes GCN to capture high-dimensional features of stock interactions， and finally outputs the predicted stock price through the fully connected layer. The experimental results in the Chinese A-share market indicate that this model has the smallest error， the better fit， and can more effectively predict stock prices compared to single stocks and industry relationship based prediction methods. It is a stock price prediction model that captures the mutual influence between stocks more fully.

Circular Convolutional Neural Network-based Defect Detection Method for#br# Drainage Pipe Networks

LIU Cunli1, LEI Zhanzhan2, ZHENG Ao2

2024, 0(07): 26-35. doi:10.3969/j.issn.1006-2475.2024.07.005

Asbtract ( 108 )

PDF (4477KB) ( 109 )

References | Related Articles | Metrics

Municipal drainage systems are critical to the safety of urban road traffic， so it is important to assess their condition. In developed countries， closed-circuit television （CCTV） is the main detection tool for sewer assessment and maintenance， but it brings new challenges for its data processing. This paper proposes a drainage network defect detection method based on recurrent convolutional neural network （RCNN）. The RCNN uses a residual network （ResNet） as feature extraction module to extract visual features of drainage network image sequences， and a bidirectional LSTM is used to learn to identify temporal features to accomplish drainage network defect classification task. The method recognizes image sequences as a whole， and the training set， validation set and test set contain a total of 8800 image sequences， and 211200 images. The data set are trained and tested by the RCNN model， and the highest accuracy rate of the test set is 90.3%. Six sets of control experiments are carried out with four different fusion methods introduced to the proposed method， the SVM-based method and the method based on single frames， as well as three fusion methods based on visual attention mechanism are introduced into the proposed method and control tests are carried out. The experimental results show that the highest accuracy （90.3%） of the fusion experiments is achieved by RCNN taking the average value， and the feasibility analysis of engineering applications is realized， and the recall rate of RCNN reaches 0.977， which confirms the feasibility of the proposed method in engineering applications.

Knowledge Prompt Fine-tuning for Event Extraction

LI Lu, ZHU Yan

2024, 0(07): 36-40. doi:10.3969/j.issn.1006-2475.2024.07.006

Asbtract ( 118 )

PDF (1020KB) ( 108 )

References | Related Articles | Metrics

Event extraction is an important research focus in information extraction， which aims to extract event structured information from text by identifying and classifying event triggers and arguments. Traditional methods rely on complex downstream networks， require sufficient training data， and perform poorly in situations where data is scarce. Existing research has achieved certain results in event extraction using prompt learning， but it relies on manually constructed prompts and only relies on the existing knowledge of pre-trained language models， lacking event specific knowledge. Therefore， a knowledge based fine-tuning event extraction method is proposed. This method adopts a conditional generation approach， injecting event information to provide argument relationship constraints based on existing pre-trained language model knowledge， and optimizing prompts using a fine-tuning strategy. Numerous experiment results show that compared to traditional baseline methods， this method outperforms the baseline method in terms of trigger word extraction and achieves the best results in small samples.

Hardware Monitoring and Warning System of Provincial Meteorological Data Center Based on Distributed

LIU Yang1, HUANG Zhi2, XU Juan1, GAO Peng1, CHEN Xuhui1

2024, 0(07): 41-46. doi:10.3969/j.issn.1006-2475.2024.07.007

Asbtract ( 99 )

PDF (1687KB) ( 57 )

References | Related Articles | Metrics

With the rapid development of the meteorological industry， the support of various meteorological departments for information technology is increasing. The development of meteorological business needs to rely on various information systems and infrastructure. At present， the provincial meteorological data center carries a variety of meteorological services. However， due to the wide variety of services， the existence of a variety of heterogeneous devices， and the lack of an effective centralized hardware monitoring and warning platform， it has brought great difficulties to the system operation and maintenance personnel. Aiming at these problems， this paper analyzes the differences and monitoring status of heterogeneous devices in multi-source complex meteorological data environment， and designs a comprehensive hardware monitoring platform with “whole process”， “centralization” and “visualization”. The technical architecture of Mycat+MySQL+Haproxy+Keepalived is adopted to build a highly available， high-performance and scalable distributed database cluster architecture， which realizes the collection， processing， storage， user-defined warning and display of hardware monitoring data. At present， the system has been used in Gansu Meteorological Data Center， which is stable operation， safe and reliable， and can help the operation and maintenance personnel to locate the system fault in time， thus improving the efficiency of monitoring and operation and maintenance and providing efficient meteorological data services for social users.

Survey on Multimodal Information Processing and Fusion Based on Modal Categories

HUANG Wendong, WANG Yifan

2024, 0(07): 47-62. doi:10.3969/j.issn.1006-2475.2024.07.008

Asbtract ( 146 )

PDF (1939KB) ( 133 )

References | Related Articles | Metrics

With the continuous advancement of artificial intelligence and deep learning technologies， research in the field of multimodal information processing and fusion has garnered widespread attention from researchers. This paper provides a comprehensive overview of the development history and milestone works of multimodal information processing， along with strategies and models for multimodal fusion. Based on different modalities，mainstream datasets for multimodal information processing and fusion are systematically classified and summarized. Using modality type as the classification criterion， this paper systematically reviews the research progress in multimodal information processing and fusion， emphasizing the distinctions between different modalities. Multimodal information processing and fusion are categorized into four types： audio-visual processing and fusion， audio-text processing and fusion， visual-text processing and fusion， and visual-audio-text processing and fusion. Detailed investigations are conducted on methods and models for processing and fusing different input modalities. Finally， a summary and outlook on the development of multimodal processing and fusion are provided.

Commuting Traffic Analysis Zone Recognition Using Improved K-means Algorithm

QIN Yang, ZHAN Yong, MING Luyao, YANG Shuqi, LAN Zhenyi

2024, 0(07): 63-68. doi:10.3969/j.issn.1006-2475.2024.07.009

Asbtract ( 101 )

PDF (3217KB) ( 59 )

References | Related Articles | Metrics

Commuting is a periodical and stable travel behavior of urban residents， which is an important research content of urban development planning and public transportation management. Taxi GPS trajectory data reflects urban traffic conditions and citizens’ travel patterns to a certain extent. Aiming at the problem of taxi regional commuting pattern recognition， a commuting traffic analysis zone recognition method based on improved K-means algorithm is proposed. This method mainly includes three steps： dividing traffic analysis zones， generating flow transfer matrix between traffic analysis zones， and identifying commuting traffic analysis zone pairs. Referring to the existing traffic analysis zones division methods， a bottom-up division method based on fine-grained elements is proposed. In the recognition model of commuting traffic analysis zone pairs， the traffic flow and its dispersion coefficient during peak hours are taken as input features， and the commuting traffic analysis zone pairs are identified based on the improved K-means algorithm. Finally， an experimental verification is carried out based on the Chongqing taxi GPS data set， and the experimental results show that the method is effective.

Blockchain-enhanced Vehicle Edge Computing Networks Security Data Storage and Share

LI Junxiao1, ZHANG Xiaolin1, SHI Jing2

2024, 0(07): 69-75. doi:10.3969/j.issn.1006-2475.2024.07.010

Asbtract ( 109 )

PDF (3063KB) ( 68 )

References | Related Articles | Metrics

The traditional centralized Vehicular Ad Hoc Networks （VANET） architecture is difficult to overcome the increasingly complex problems of Intelligent Transportation System （ITS） applications， as well as the challenges of large amounts of data storage， trust management and information security. Therefore， Vehicular Edge Computing Networks （VECNets） emerged as the times require， providing powerful network edge computing capabilities for massive storage resources. However， due to potential data leakage and security risks， the centralized server in VECNets has a Single Point of Failure （SPoF）. In response to the above problems， a distributed network framework that integrates blockchain and smart contracts is proposed to ensure a secure environment for data storage and share in the system. We utilize the decentralization， anti-tampering characteristics and controllability balance of the alliance chain to ensure the security of data storage in the Internet of Vehicles. Additionally we combine edge computing and Byzantine Fault Tolerance consensus model （Practical Byzantine Fault Tolerance， PBFT） to distribute the data storage to road edges， thereby reducing data transmission latency. Experimental results show that the protocol proposed in this article has a good effect in improving the system throughput， reduces consensus delay and communication cost performance parameters of the vehicle network.

Application of Data Synchronization Technology in External Services of Meteorological Big Data Cloud Platform

QIU Ling1, 2, SONG Zhi1, 2, LYU Shuang1, 2, YANG Xue1, 2

2024, 0(07): 76-81. doi:10.3969/j.issn.1006-2475.2024.07.011

Asbtract ( 61 )

PDF (1198KB) ( 49 )

References | Related Articles | Metrics

Currently， the meteorological big data cloud platform exclusively serves meteorological Intranet users. In order to further expand the platform’s capability for external services and meet the extensive demands of meteorological industry users for the storage and sharing of massive meteorological data applications， a secure and efficient meteorological big data storage and service system is designed based on key technologies such as distributed storage and data service interfaces. This system is implemented in the Unidirectional Network Security Isolation Zone （Demilitarized Zone， DMZ）. For structured meteorological data， an optimized data synchronization method based on the Database Binlog in a distributed database is researched and designed. For unstructured meteorological data， an optimized data synchronization method combining the Database Binlog and ETL mechanisms is researched and designed to achieve real-time second-level unidirectional synchronization of massive meteorological data between the meteorological intranet big data cloud platform and the DMZ area. The method proposed in this paper addresses issues such as low data volume， poor sharing service capabilities， and low security for meteorological data in industry user services. Application practices demonstrate that the data synchronization technology achieves synchronization delays within seconds，maintains good data consistency， and ensures the real-time and accurate external service of the meteorological big data cloud platform.

PCB Board Welding Path Optimization Based on Improved Cuckoo Algorithm

WANG Qian, HUANG Miao, Tao Lili

2024, 0(07): 82-86. doi:10.3969/j.issn.1006-2475.2024.07.012

Asbtract ( 46 )

PDF (1219KB) ( 51 )

References | Related Articles | Metrics

An improved cuckoo algorithm is proposed to solve the path planning problem in the process of PCB welding. This paper adds 2opt neighborhood search strategy to the classic cuckoo algorithm. After the original algorithm produces a new solution， 2opt neighborhood search operation is performed on the new solution， and whether the better solution nearby to be replaced is judged by search， so as to improve the local search ability and solution accuracy of the algorithm. In order to verify the effectiveness of the improved algorithm， this paper carries out simulation experiments on three kinds of PCB boards with different hole numbers and hole distribution complexity， and compares the optimization effects among the classic cuckoo algorithm， the algorithm in Reference［28］ and the improved algorithm proposed in this paper. The experimental results show that， among the three PCB boards， the improved algorithm not only has the shortest optimization path distance， but also has the highest solution accuracy， and also the convergence speed is improved. When the number of holes increases and the complexity of hole distribution increases， the improved algorithm shows greater advantages in path optimization effect than the other two algorithms. To sum up， the improved cuckoo algorithm has certain effect in optimizing the welding path of PCB， and has the advantages of strong search ability， high solution accuracy and fast convergence speed.

Lane Line Detection Algorithm Based on Improved SCNN Network

WU Li1, ZHANG Zhenghao2, GE Caicheng2, YU Jun2

2024, 0(07): 87-92. doi:10.3969/j.issn.1006-2475.2024.07.013

Asbtract ( 98 )

PDF (2200KB) ( 101 )

References | Related Articles | Metrics

Lane line detection is the prerequisite for the realization of lane keeping system and lane departure warning system. In order to further improve the accuracy of detection， combining deep learning and lane line detection， this paper proposes an improved SCNN lane line recognition algorithm. Based on the improved SCNN network， this method introduces PSA attention module， and combines it with VGG （Visual Geometry Group） network to propose a lane line recognition network called VGG-K. This network fuses context information， helps the message transmission between row and column pixels in each layer， enhances its recognition ability for continuous transformation targets， and then uses quadric model fitting to obtain the final lane line detection result. The improved model is tested on the dataset CULane. Training results show that the comprehensive evaluation index F1 value of the method reaches 92.1 in normal scenarios and 75.3 in harsh scenarios. Compared with other models， the detection ability of the proposed algorithm is significantly improved， and the proposed algorithm has a better recognition for lane lines under various complex conditions.

Security Game Analysis Model of RFID System Based on Bayesian Attack Graph

MA Huiping1, LI Peng1, 2, HU Sujun1

2024, 0(07): 93-99. doi:10.3969/j.issn.1006-2475.2024.07.014

Asbtract ( 85 )

PDF (1497KB) ( 56 )

References | Related Articles | Metrics

In view of the lack of comprehensive and effective risk management and security assessment of RFID systems， in order to achieve effective analysis of the security risks of RFID systems and the assessment of the overall risk status of target RFID systems， this paper proposes a Bayesian attack graph-based RFID system security game analysis method. On the basis of Bayesian attack graph model， combined with game thought， the risk situation of RFID system is analyzed， and the process of the attacker invading the system is abstracted into the game model of the attack and defense. This paper firstly determines the offensive and defensive strategy based on the relevant information of the target system， and constructs the corresponding offensive and defensive game matrix by calculating the strategic income of the attacker and the defender， then obtains the Nash equilibrium state， determines the optimal strategy of each participant， and finally calculates the expected income of both parties to determine the security state of the target RFID system: If the expected return of the attacker is greater than the expected return of the defender， the system is in the risk state; otherwise， the system is in the security state. The experiment results show that the game model proposed in this paper can well realize the security analysis of target RFID system.

Laser Constant Current Source Controller Based on CDKF-RBFPID

WU Junkai, MAO Zhengchong

2024, 0(07): 100-105. doi:10.3969/j.issn.1006-2475.2024.07.015

Asbtract ( 91 )

PDF (1573KB) ( 50 )

References | Related Articles | Metrics

Stable operation of lasers requires a constant current source control system with high precision and stable output current. An algorithm based on the combination of central differential Kalman filtering （CDKF） and improved radial basis function （RBF） neural network adaptive PID control， named CDKF-RBFPID， is proposed to address the problem that the output accuracy of laser constant current source system is poor in noisy environment and the parameters are difficult to be adjusted by PID algorithm. By using CDKF， we update the state and covariance of the constant current source system， so as to filter out the state noise and measurement noise in the system. To accomplish adaptive parameter tuning， the RBF-PID parameters are modified using the reinforcement learning Actor-Critic architecture. Comparing the output current of the constant current source system and the output power of a laser， the experimental results demonstrate that the CDKF-RBFPID method can effectively lessen the impact of noise on the system， further enhance the accuracy of the constant current source output current and the stability of the laser output power， with the response time improved by 58.3%， the steady-state error reduced by 71.4%， and the output current control accuracy reaching 1%.

Prediction Model of Hydraulic Engineering Slope Mechanical Parameters Based on DEFA-LSSAR

CAO Ning1, YAN Xin’e1, XU Genqi2, XU Youwen1, ZHANG Zhengbo2, DU Qianyun2

2024, 0(07): 106-111. doi:10.3969/j.issn.1006-2475.2024.07.016

Asbtract ( 65 )

PDF (1453KB) ( 46 )

References | Related Articles | Metrics

In order to solve the problem of low accuracy of existing hydraulic engineering slope mechanical parameter prediction models， the least squares support vector machine LSSAR is used to predict the hydraulic engineering slope mechanical parameters （elastic modulus E）， and the improved firefly algorithm is used to optimize the model. A hydraulic engineering slope mechanical parameter prediction model based on DEFA-LSSAR is proposed. We compare the model proposed in this article with the LSSAR model optimized by the Salp Swarm Algorithm， Drosophila Algorithm， and Harris Eagle Optimization Algorithm， respectively. The analysis results show that the proposed model has the highest prediction accuracy， reaching over 94%， and has the smallest fitness value， verifying the effectiveness and correctness of the proposed model in this article.

MPA Algorithm Based on Adaptive Rotation Learning and Crisis Awareness Strategy

HONG Guangjie, CAI Maoguo, ZHAN Kaijie, OU Jifa

2024, 0(07): 112-119. doi:10.3969/j.issn.1006-2475.2024.07.017

Asbtract ( 67 )

PDF (2411KB) ( 60 )

References | Related Articles | Metrics

Aiming at the shortcomings of the Marine Predators Algorithm （MPA） such as slow convergence speed， low solution accuracy and easy to fall into local optimum， a Marine Predators Algorithm based on adaptive rotation learning and crisis awareness strategy （ARCMPA） is proposed. First of all， in view of the slow convergence speed and low convergence precision of the MPA， a crisis awareness strategy is introduced to improve the ability of the algorithm to explore the solution space， strengthen the lack of early development capabilities of the algorithm， speed up the convergence speed of the early algorithm， and improve the quality of the algorithm solution. Secondly， an adaptive rotation learning mechanism is introduced to make the position distribution of the entire population more uniform， effectively enhancing the diversity of the population during the iteration of the algorithm， and preventing the algorithm from falling into local optimum after accelerating the convergence speed in the early stage. Through the introduction of two strategies， the overall performance of the algorithm is effectively enhanced. In this paper， 10 benchmark functions are selected and compared with other metaheuristic algorithms. Experimental results show that the above improvements help to improve the overall performance of the algorithm.

Underwater Trash Detection Method Based on Improved YOLOv5

PANG Mei, WANG Gong, ZHAN Yong, HUANG Zhefa

2024, 0(07): 120-126. doi:10.3969/j.issn.1006-2475.2024.07.018

Asbtract ( 163 )

PDF (3845KB) ( 145 )

References | Related Articles | Metrics

To address the limitations of underwater image acquisition such as insufficient light， high noise and unclear object recognition， which lead to the ineffectiveness of existing object detection algorithms， an underwater garbage object detection algorithm based on improved YOLOv5 is proposed. The purpose of the improved object detection algorithm is to achieve more accurate detection and removal of underwater plastic trash from the ocean. The improved algorithm containes some improvements：using the Contrast Limited Adaptive Histogram Equalization（CLAHE） algorithm to enhance data features， which reduces the difficulty of feature extraction and enables the network to be detected more flexibly and more accurately； introducing a parameter-free attention module SimAM， using the lightweight convolution method GSConv to enhance network extraction capability while reducing model computation; At the same time， multi-scale feature fusion detection is added to solve the problem of small target location of underwater debris. Numbers of experiments are conducted based on MarineTrash which is a self-built real underwater environmental litter dataset， the results show that the improved method has good performance， in which the accuracy is increased by 4.3 percentage points， the mAP is increased by 3.5 percentage points， the GFLOPs is reduced by 0.3， and the model weight is only 13.9 MB， which is 0.6 MB lower than the baseline. The research on the underwater trash detection algorithm based on the improved YOLOv5 provides sufficient technology for deploying and installing detectors in Autonomous Underwater Vehicles （AUVs） to achieve detection and automatic removal of marine underwater trash and maintain the marine ecosystem.

Table of Content