International Journal of Computer Applications in Technology (21 papers in press)

Regular Issues

• FPGA implementation and Multisim simulation of a new four-dimensional two-scroll hyperchaotic system with coexisting attractors  
  by Sundarapandian Vaidyanathan, Esteban Tlelo-Cuautle, Khaled Benkouider, Aceng Sambas, Ciro Fabian Bermudez-Marquez, Samy Abdelwahab Safaan 
  Abstract: Field-programmable gate array (FPGA) design of a new four-dimensional two-scroll hyperchaotic system is investigated in this work. A detailed system modelling of the new system with a hyperchaotic attractor begins this work with phase plots, which is followed by a bifurcation study of the new system. Special dynamic properties such as multistability and symmetry are also investigated for the new system. Using Multisim software, a circuit model is designed and simulated for the new hyperchaotic system. FPGA design and Multisim simulation of the new system enable practical applications in science and engineering. The implementation of the FPGA design in this work is carried out by applying two numerical schemes, viz. Forward Euler and Trapezoidal methods. Experimental attractors observed in the oscilloscope show good match with the Matlab signal plots.The FPGA hardware resources are detailed for both numerical methods.
  Keywords: hyperchaos; bifurcation; symmetry; phase plots; hyperchaotic system;rnparameters; stability; multistability; circuit model; FPGA implementation.
  
  
• Improving hybrid-layer convolutional neural network system for lung cancer nodule classification using enhanced weight optimisation algorithm  
  by Vikul Pawar, P. Premchand 
  Abstract: In recent times, lung cancer is evolving as a highly life-threatening disease for human beings. According to the WHO, lung cancer disease is the second largest cause of deaths as compared to all other types of cancer. The prevailing available technology is striving to get more exposure in the field of medical science using Computer Assisted Diagnosis (CAD), where image processing is playing a crucial role for detecting the cancerous nodules in computer tomographic images. Augmenting the machine learning techniques with image processing algorithms is becoming a more comprehensive examination of cancer disease in proposed CAD systems. This paper is describes a heuristic approach for lung cancer nodule detection, and the proposed model predominantly consists of the following tasks, which are image enhancement, segmenting ROI (Region of Interest), features extraction, and nodule classification. In pre-processing, primarily the Adaptive Median Filter (AMF) filtering method is applied to eliminate the speckle noise from input CT images of Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI): in the LIDC-IDRI dataset, the quality of input image is improved by applying Histogram Equalization (HE) technique with Contrast-Limited Adaptive (CLA) approach. Secondly, in the successive stage the Improved Level-Set (ILS) algorithm is used to segment the ROI. Furthermore, the third step of the projected work is applied to extract the definite learnable texture features and statistical features from the segmented ROI. The extracted features in the subsequent stage of classification are applied to Hybrid-Layer Convolutional Neural Network (HL-CNN) architecture to classify the lung cancer nodule as either benign or malignant. Principally this research is carried out by contributing to each stage of it, where the novel concept of the improved Hybrid-Layer Convolutional Neural Network (HL-CNN) is employed by optimising and selecting the optimal weight using the Enhanced Cat Swarm Optimisation (ECSO) algorithm. The experimental result of the proposed HL-CNN using the weight optimisation algorithm ECSO is achieved an accuracy of 93%, which is comparatively efficient with respect to existing models such as DBN, SVM, CNN, WOA, MFO, and CSO. Moreover, the proposed model conclusively gives a decision on the detected nodule as either benign or malignant.
  Keywords: Computer Assisted Diagnosis (CAD); Computer Vision; Cancer Diagnosis; Image Classification; Image Enhancement; Image Segmentation; Feature Extraction.
  
  
• Prediction model for total amount of coke oven gas generation based on FCM-RBF  
  by Lili Feng, Jun Peng, Zhaojun Huang 
  Abstract: The rational use of Coke Oven Gas (COG) is of great significance to improve the economic efficiency of enterprises. In this paper, a COG generation prediction model based on fuzzy C-mean clustering (FCM) and radial basis function (RBF) neural network is proposed to address the problems such as the difficulty of accurate modelling of COG generation process and the difficulty of real-time flow prediction. Firstly, the coke oven production process is analysed and correlation analysis is used to select the influencing factors. Secondly, the FCM is used to classify the working conditions of the coke oven, and the appropriate number of working conditions is selected through experiments. Finally, the prediction models under different working conditions are established separately by using RBF. The experiments were carried out using actual industrial production data, and the experimental results showed that the model could provide guidance reference for the dispatchers.
  Keywords: coking oven process; fuzzy C-means clustering; prediction model; radial basis function neural network.
  
  
• Research on machine reading comprehension for BERT and its variant neural network models  
  by Yanfeng Wang, Ning Ma, Wenrong Lv 
  Abstract: Currently, machine reading comprehension model is mainly models primarily rely on the basis of LSTM network networks with the gate mechanism. In the present this study, we employ BERT and its variant pre training-trained language model are used models to conduct research and experiment experimentation on the DuReader dataset. It is found We find that improved mask methods enhanced masking techniques, such as full -word mask masking and dynamic mask masking, can notably significantly enhance the model's performance of the model in machine reading comprehension tasks. Therefore, the ROUGE-L and BLEU-4 values of Consequently, the best RoBERTRoBERTa-wwm-ext model on the test set are achieves ROUGE-L and BLEU-4 scores of 51.02% and 48.14% separately, on the test set, respectively, which are 19.12% and 8.94% higher than the benchmark model. In addition, in view of the problem that the, Moreover, addressing the issue of suboptimal model performance is not optimal when the data dealing with large-scale is large data and the relatively dispersed effective information is relatively dispersed, this paper adopts employs a three-step preprocessing of approach for the dataset.
  Keywords: machine reading comprehension; BERT pre-training language model; masking mode.
  DOI: 10.1504/IJCAT.2024.10063844
   
  
• Fast recognition method of pedestrian signs based on colour features and SVM  
  by Lu Gan 
  Abstract: In order to overcome the problems of low recognition efficiency and poor recognition accuracy in traditional recognition methods of pedestrian signs, a fast recognition method of pedestrian signs based on color features and SVM is proposed. Firstly, the median filter is used to filter the salt and pepper noise in the image, and the RGB image of the sidewalk indicator is converted into HSV image. Secondly, two morphological processing methods, expansion and corrosion, are used to obtain the color features of the sidewalk indicator image. Finally, support vector machine is used to classify the image of sidewalk indicator signs, construct the loss function of sidewalk indicator sign image recognition, and solve it to get the final image recognition result. The experimental results show that the recognition algorithm in this paper can achieve 98.2% recognition accuracy, and the recognition time is always less than 5 s.
  Keywords: colour characteristics; SVM; sidewalk sign; HSV colour model; median filtering.
  DOI: 10.1504/IJCAT.2024.10066590
   
  
• Multi-objective optimisation of traffic data transmission load based on improved machine learning in the edge computing framework  
  by Zhichang Huang 
  Abstract: To solve the problems of high packet loss rate, high delay and low load balance in the traditional method, a multi-objective optimization method of traffic data transmission load based on improved machine learning in the edge computing framework is proposed.In the edge computing framework, traffic data is collected and traffic data transmission path is selected. The multi-objective optimization function of traffic data transmission load is built, and the traffic data transmission load scheduling model is built by combining the multi-objective optimization function and the improved support vector machine. The traffic data transmission path and traffic data are input into the model, and the relevant optimization results are obtained. The experimental results show that the maximum packet loss rate of traffic data transmission is 6.28%, the maximum delay of traffic data transmission is 57.91ms, and the load balance of traffic data transmission is between 0.96 and 0.98.
  Keywords: edge computing framework; improved machine learning; traffic data; transmission load; multi-objective optimisation; improved support vector machine.
  DOI: 10.1504/IJCAT.2024.10066591
   
  
• A collaborative filtering-based network multimedia English teaching resource recommendation  
  by Huanxia Deng 
  Abstract: Owing to the problems of high recommendation error and low recommendation satisfaction in traditional network multimedia English teaching resources recommendation methods, a collaborative filtering-based network multimedia English teaching resource recommendation method is proposed in this paper. Firstly, the entities contained in the network multimedia English teaching resources are extracted, and a hierarchical allocation model for resource recommendation is established. Then the semantic information of the network multimedia English teaching resources is mined. The hidden items of different semantic features are analyzed, and different levels of concern are set. Finally, the distribution of the filtering model parameter under different difference levels is obtained by the collaborative filtering method, and the resources are recommended based on the knowledge graph fusion inference technology. The test results indicate that this method has a high level of satisfaction and low recommendation error in recommending resources.
  Keywords: collaborative filtering; network multimedia; English teaching resources; feature clustering; knowledge graph.
  DOI: 10.1504/IJCAT.2024.10066592
   
  
• Malicious communication node location method of smart campus network considering multi-dimensional mapping relationship  
  by Wenjing Ma, Ketong Liu, Lixia Hou 
  Abstract: In order to improve the security of smart campus network, a malicious communication node location method considering multidimensional mapping is designed. Firstly, the attributes of smart campus network communication nodes and the attack mode of malicious communication nodes are analyzed. Then, the temporal behavior association between the attacker and the attacked party is established by using the reverse backtracking method, and abnormal variables in the network are obtained based on the multidimensional mapping relationship, so as to detect malicious nodes. Finally, the modified RSSI model and trilateral prime-center method are used to locate the malicious communication nodes accurately. According to the experimental results, the positioning error rate of the proposed method is always below 0.05, and the minimum positioning time during the test is only 1841ms, indicating that the proposed method is effective.
  Keywords: smart campus network; malicious communication node; node detection; node location; multidimensional mapping relationship; RSSI model; trilateral prime centre method.
  DOI: 10.1504/IJCAT.2024.10066593
   
  
• Virtual data generation method for simulation scenes based on scene flow prediction  
  by Lizhen Jiang, Xu Lan 
  Abstract: In this paper, a virtual data generation method for simulation scenes based on scene flow prediction is proposed. Firstly, a variational autoencoder is introduced to determine the relationship between virtual data in different dimensions through backpropagation performance; Then, with the help of deep semantic information, automatic annotation of simulation scene data is completed; Finally, determine the different displacement changes in the three-dimensional space of the data object, encode the entire virtual data, and fuse it. Combined with the matching of encoded features, ensure the consistency of the encoded virtual data features. Finally, the objective function is introduced, and a data generator is set up to achieve virtual data generation in simulation scenarios. The results show that the proposed method reduces the false positive rate of virtual data generation in simulation scenarios and has good performance.
  Keywords: scene flow; prediction; simulation scenario; virtual data; generation method; deep semantic information; back-propagation.
  DOI: 10.1504/IJCAT.2024.10066594
   
  
• Research on storage and encryption method for private electronic communication information based on fuzzy rules  
  by Rui Sun, Zhenzhen Wang, Huanhuan Liu, Yu Ji 
  Abstract: To improve the security of private electronic communication information, the paper proposes an information storage encryption method based on fuzzy rules. Firstly, after establishing a private electronic communication information set, fill in the missing values. Secondly, using the fuzzy C-means algorithm, electronic communication information is classified according to fuzzy rules and stored in the underlying information database according to the index of communication information. Finally, for the data in the information database, the AES algorithm is applied to encrypt lightweight information, and a hybrid algorithm of DES and ECC is applied to encrypt high-dimensional sparse information. The experimental results show that when the number of plaintext in private electronic communication information increases from 30 bits to 150 bits, the number of ciphertext bits in this method increases from 65 bits to 309 bits, indicating that this method can effectively ensure the security of private electronic communication information
  Keywords: private electronic communication information; information classification; fuzzy C-means algorithm; fuzzy rules; information encryption.
  DOI: 10.1504/IJCAT.2024.10066595
   
  
• An outlier detection method for open-source software running data based on Bi-LSTM network  
  by Jiehai Deng, Weihong Li 
  Abstract: In order to solve the problem of large errors in existing data outlier detection methods, an open-source software based on Bi-LSTM network is proposed to run data outlier detection method. By introducing both parametric and non parametric methods, a threshold for outlier features in open source software running data is set, and features are extracted using the Manhattan distance and anomaly factor in the K-nearest neighbour algorithm. Introduce XGBoost algorithm to define decision tree, use loss function to reduce classification residual, train leaf nodes, and obtain classification features. The features obtained through standardization and normalization are detected using the Bi-LSTM network algorithm and attention mechanism is applied. The results show that this method has better detection performance.
  Keywords: Bi-LSTM network; open-source software; operating data; abnormal value detection; Manhattan distance; forward detection; reverse detection.
  DOI: 10.1504/IJCAT.2024.10066596
   
  
• Multi-source network attack tracing method based on traffic characteristics  
  by Hui Hong, Wei Yang, Bocheng Sun, Ling Zhang 
  Abstract: In order to overcome the shortcomings of low accuracy in traditional methods for extracting traffic characteristics, low traceback accuracy, and long duration, a Multi-source network attack tracing method based on traffic characteristics is proposed. A network traffic data collection window is set up and traffic characteristics are extracted. Mutual information is used to achieve traffic feature selection. Hidden Markov model is utilized to determine the status of the multi-source network and hosts, and combined with anomaly traffic detection for multi-source network attacks. Based on the detection results, the IP addresses of the originating hosts are traced and the results of multi-source network attack traceback is obtained. The experimental results show that the mean accuracy of traffic feature extraction in this method is 95.68%, the mean traceback accuracy is 97.25%, and the time fluctuates between 0.22 s and 0.35 s.
  Keywords: traffic characteristics; multi-source network attacks; traceback; feature selection; hidden Markov model.
  DOI: 10.1504/IJCAT.2024.10066597
   
  
• Traceability method for multisource heterogeneous data of power grid business based on multidimensional feature clustering  
  by Zhiguo Zhou, Hao Chen, Mengxiao Ni 
  Abstract: To shorten the data traceability delay and improve the traceability effect, a multi-source heterogeneous data traceability method for power grid business based on multi-dimensional feature clustering is proposed. Firstly, through unstructured data transformation and cleaning, preprocess multi-source heterogeneous data of power grid business. Then, multi-dimensional features of the data are extracted through sliding clustering, and the fusion results of data features are obtained through fuzzy decision-making. Finally, combining the K value in RFID encoding with the hash value in the index structure, perform forward tracing, backward tracing, and process tracing on the data. According to the experiment, it is known that after applying this method, the maximum value of traceability delay is only 2861 ms, the traceability error rate is basically maintained at around 0.85%, and the data integrity coefficient can reach up to 97.96%, indicating that this method has high feasibility.
  Keywords: power grid business; multisource heterogeneous data; data traceability; data conversion; data cleaning; feature extraction; feature fusion.
  DOI: 10.1504/IJCAT.2024.10066598
   
  
• A deep fusion method for asynchronous data in the Internet of Things based on data feature mining  
  by Yongyi Huang 
  Abstract: In order to reduce the packet loss rate and fusion time during the data fusion process, a deep fusion method for asynchronous data in the Internet of Things based on data feature mining is proposed. Firstly, analyze the structure of the Internet of Things and the characteristics of asynchronous data. Secondly, sparse filtering is applied to asynchronous data in the Internet of Things to improve the quality of asynchronous data. Finally, association rules are used to mine the association features of asynchronous data in the Internet of Things. Calculate the similarity feature of IoT asynchronous data and achieve deep fusion of IoT asynchronous data. The test results show that compared with existing data fusion methods, the packet loss rate of our method is significantly reduced, with the highest packet loss rate not exceeding 0.1, and the data fusion time is significantly shortened.
  Keywords: data feature mining; Internet of Things; asynchronous data; deep fusion.
  DOI: 10.1504/IJCAT.2024.10066599
   
  
• Extracting Abnormal Frequency Signals from Power Grid Based on Spectrum Analysis  
  by Chuning Peng, Jun Zhang, Xiaodong Yin, Yi Cao, Quan Wang, Haibin Chen, Jiachuan Long 
  Abstract: In order to improve the accuracy and efficiency of extracting abnormal frequency signals, a power grid abnormal frequency signal extraction method based on spectrum analysis is proposed. Firstly, singular value decomposition is used to denoise the power grid signal. Secondly, for the denoised power grid signal, a bispectral interpolation correction algorithm is used to correct the power grid signal. Finally, based on the observed signal matrix, mixed matrix, and source matrix of the power grid signal, the frequency ratio of the power frequency signal to the abnormal frequency signal in the power grid signal is calculated, and the abnormal frequency signal is extracted from the mixed signal of the power grid signal and abnormal frequency signal. The experimental results show that the method proposed in this paper can effectively remove noise from power grid signals, and the average accuracy of heterofrequency signal extraction reaches 98.23%.
  Keywords: spectrum analysis; power grid abnormal frequency signal; signal extraction; singular value decomposition.
  DOI: 10.1504/IJCAT.2024.10066600
   
  
• A distributed network data security and confidentiality system based on 5G networking  
  by Han Ge, Shumin Ren, Jian Xie 
  Abstract: In order to overcome the problems of weak data attack resistance, low data integrity, and poor system performance in distributed network data security and confidentiality systems, a distributed network data security and confidentiality system based on 5G networking is designed. Firstly, design the overall structure of the 5G networking gateway for the distributed network data security and confidentiality system; Then, analyze the 5G networking application scenarios and determine the 5G intelligent networking mode; Finally, generate encrypted ciphertext, receive the secret key through a delegated algorithm, obtain the public key by accessing the Ethereum block, and use the CP-ABE algorithm for distributed network data key conversion to achieve distributed network data security and confidentiality. The experimental results show that the data integrity of the designed system is 98.2%, and the energy consumption for data security and confidentiality is 4.5KJ, which verifies the data security and confidentiality performance of the system.
  Keywords: 5G networking; delegated algorithm; access tree; CP-ABE algorithm; key conversion.
  DOI: 10.1504/IJCAT.2024.10066601
   
  
• A fast retrieval method for legal judgment documents based on multi-granularity semantic interaction understanding  
  by Jianhua Guo 
  Abstract: In order to improve the normalized loss gain and recall rate of information retrieval results, the paper proposes a fast retrieval method based on multi granularity semantic interaction understanding for legal judgment documents. After constructing a multi granularity encoder, establish a local semantic interaction understanding module and a global semantic interaction understanding module to extract the associated information of each statement in the document from the perspective of statements and paragraphs; Then, the attention signal of the target statement is obtained, the semantic matching score is calculated and weighted to obtain the final search result. Experiment shows that after applying this method, the semantic interaction understanding recall rate of document information remains around 98.0%, the F1 statistical value of the search result is higher than 90.0, and the normalized loss gain of document is close to 1, indicating the effectiveness of this method.
  Keywords: multi-granularity semantic interaction understanding; legal judgment documents; information retrieval; semantic matching.
  DOI: 10.1504/IJCAT.2024.10066786
   
  
• YOLO-based gripping method for industrial robots  
  by Wei Gao 
  Abstract: With the current development of industrial intelligence in society, new challenges have been created for traditional industrial robots, and grasping is a significant capability of robots. The problem of robot grasping has been a famous research problem at home and abroad. With the rise of deep learning technology in recent years, it has been applied to various fields because it can extract better features. In this paper, two target detection models are optimized, and the Faster R-CNN target detection model is optimized to adjust the network structure, the scale size of anchors, the target classification, and the position regression structure. The YOLO-v2 target detection model is optimized, and the Darknet-19 feature extraction network structure and the loss function are adjusted. The experimental results demonstrate that the target detection network learns useful image features, and the grasping system can complete the autonomous grasping task.
  Keywords: convolutional neural network; target detection; automatic grasping.
  DOI: 10.1504/IJCAT.2023.10067187
   
  
• An unsupervised video summarisation method based on temporal convolutional networks  
  by Ke Jin, Haoran Li, Hui Li, Qichuang Liu, Rong Chen, Shikai Guo 
  Abstract: Video summarisation automatically select a sparse subset of video frames that best represent the semantic content of the input video. Previous work mainly used Long Short-Term Memory (LSTM) networks to learn how to assess the importance of each video frame and then select appropriate frames to compose a video summary. However, these models still represent shortcomings, such as limited memory capacity in handling long-term dependencies, and extended training time. Therefore, we present a deep video summarisation model, named TCN-SUM, which centres around Bi-TCN and models the frame sequence. TCN-SUM consists of three modules: frame selection module incorporates both Bidirectional Temporal Convolutional Network (Bi-TCN) and self-attention to model inter-frame dependencies, video reconstruction module reconstructs the original video based on the summary, and discriminator module measures the similarity between the original and reconstructed videos. Experimental studies on two benchmark datasets demonstrates that TCN-SUM outperforms state-of the- art techniques, achieving superior performance in unsupervised approaches and showing competitiveness compared to supervised methods.
  Keywords: Bi-TCN; video summarisation; self-attention; LSTM.
  DOI: 10.1504/IJCAT.2024.10067536
   
  
• An access control of enterprise financial privacy information based on RBAC model  
  by Xuejiao Shi 
  Abstract: It is essential to improve the role-based access control (RBAC) method for controlling access to enterprise financial privacy information. Firstly, this study proposes to address the information leakage and distortion issues concerning access to enterprise financial privacy information. The statistical characteristics of privacy information resources are analyzed using entropy analysis. Following the analysis of abnormal access to privacy information, an RBAC model is constructed, and multi-agent technology is introduced to ensure user privacy access. This approach aims to enhance the security and management effectiveness of enterprise financial privacy information and achieve access control to enterprise financial privacy information. Experimental results indicate that the proposed method can effectively reduce information leakage and distortion issues. However, it should be noted that due to legal and ethical considerations, it is necessary to adhere to data protection laws and regulations in implementing these access control measures.
  Keywords: RBAC model; corporate finance; privacy information; access control; anomaly analysis.
  DOI: 10.1504/IJCAT.2024.10067841
   
  
• A positioning method of intelligent networked vehicle based on laser radar  
  by Zhiting Liu, Fanxiu Shi 
  Abstract: To effectively solve the problem of large positioning deviation and high positioning delay, an intelligent networked vehicle positioning method based on LiDAR is proposed. Based on a planar model, ground point cloud extraction is carried out, and non ground point clouds are clustered using point cloud clustering combining density and Euclidean distance to filter out noisy points in the point cloud. Motion distortion compensation is applied to the point cloud data. Extracting feature point clouds based on roughness, utilizing a LiDAR odometer for localization, and eliminating accumulated errors during this process through loop detection and backend pose map optimization, to achieve intelligent networked vehicle localization. The results show that the proposed method always maintains a positioning delay of less than 2.3ms, and the positioning deviation can be controlled within the range of 20cm to 40cm, with high positioning accuracy, which can provide key technical support for intelligent transportation systems.
  Keywords: Lidar; intelligent connected vehicle; Lidar odometer; positioning method.
  DOI: 10.1504/IJCAT.2024.10067842