Most Read Articles

Published in last 1 year |  In last 2 years |  In last 3 years |  All
Please wait a minute...
Compound fault decoupling diagnosis method based on improved Transformer
Yu-xiang WANG,Zhi-wei ZHONG,Peng-cheng XIA,Yi-xiang HUANG,Cheng-liang LIU
Journal of ZheJiang University (Engineering Science)    2023, 57 (5): 855-864.   DOI: 10.3785/j.issn.1008-973X.2023.05.001
Abstract   HTML PDF (2584KB) ( 634 )  

Most of the compound fault diagnosis methods regard the compound fault as a new single fault type, ignoring the interaction of internal single faults, and the fault analysis is vague in granularity and poor in interpretation. An improved Transformer-based compound fault decoupling diagnosis method was proposed for industrial environments with very little compound fault data. The diagnosis process included pre-processing, feature extraction and fault decoupling. With introducing the decoder of the Transformer, the cross-attention mechanism enables each single fault label to adaptively in the extracted feature layer focus on the discriminative feature region corresponding to the fault feature and predicts the output probability to achieve compound fault decoupling. Compound fault tests were designed to verify the effectiveness of the method compared with the advanced algorithms. The results showed that the proposed method had high diagnostic accuracy with a small number of single fault training samples and a very small number of compound fault training samples. The compound fault diagnosis accuracy reached 88.29% when the training set contained only 5 compound fault samples. Thus the new method has a significant advantage over other methods.

Table and Figures | Reference | Related Articles | Metrics
Multimodal image retrieval model based on semantic-enhanced feature fusion
Fan YANG,Bo NING,Huai-qing LI,Xin ZHOU,Guan-yu LI
Journal of ZheJiang University (Engineering Science)    2023, 57 (2): 252-258.   DOI: 10.3785/j.issn.1008-973X.2023.02.005
Abstract   HTML PDF (928KB) ( 311 )  

A multimodal image retrieval model based on semantic-enhanced feature fusion (SEFM) was proposed to establish the correlation between text features and image features in multimodal image retrieval tasks. Semantic enhancement was conducted on the combined features during feature fusion by two proposed modules including the text semantic enhancement module and the image semantic enhancement module. Firstly, to enhance the text semantics, a multimodal dual attention mechanism was established in the text semantic enhancement module, which associated the multimodal correlation between text and image. Secondly, to enhance the image semantics, the retain intensity and update intensity were introduced in the image semantic enhancement module, which controlled the retaining and updating degrees of the query image features in combined features. Based on the above two modules, the combined features can be optimized, and be closer to the target image features. In the experiment part, the SEFM model was evaluated on MIT-States and Fashion IQ datasets, and experimental results show that the proposed model performs better than the existing works on recall and precision metrics.

Table and Figures | Reference | Related Articles | Metrics
Surface defect detection algorithm of electronic components based on improved YOLOv5
Yao ZENG,Fa-qin GAO
Journal of ZheJiang University (Engineering Science)    2023, 57 (3): 455-465.   DOI: 10.3785/j.issn.1008-973X.2023.03.003
Abstract   HTML PDF (1697KB) ( 479 )  

For the poor real-time detection capability of the current object detection model in the production environment of electronic components, GhostNet was used to replace the backbone network of YOLOv5. And for the existence of small objects and objects with large scale changes on the surface defects of electronic components, a coordinate attention module was added to the YOLOv5 backbone network, which enhanced the sensory field while avoiding the consumption of large computational resources. The coordinate information was embedded into the channel attention to improve the object localization of the model. The feature pyramid networks (FPN) structure in the YOLOv5 feature fusion module was replaced with a weighted bi-directional feature pyramid network structure, to enhance the fusion capability of multi-scale weighted features. Experimental results on the self-made defective electronic component dataset showed that the improved GCB-YOLOv5 model achieved an average accuracy of 93% and an average detection time of 33.2 ms, which improved the average accuracy by 15.0% and the average time by 7 ms compared with the original YOLOv5 model. And the improved model can meet the requirements of both accuracy and speed of electronic component surface defect detection.

Table and Figures | Reference | Related Articles | Metrics
Code development and verification for weak coupling of seepage-stress based on TOUGH2 and FLAC3D
Xia-lin LIU,Sheng-bin ZHANG,Quan CHEN,Heng SHU,Shang-ge LIU
Journal of ZheJiang University (Engineering Science)    2022, 56 (8): 1485-1494.   DOI: 10.3785/j.issn.1008-973X.2022.08.002
Abstract   HTML PDF (1589KB) ( 713 )  

Traditional and new geotechnical engineering problems such as compressed air energy storage, intercepting water with compressed air, carbon dioxide sequestration and oil and gas underground reserve project are all involving air-water two-phase flow and stress coupling problems. For this engineering reality, based on the weak coupling theory of gas-water two-phase seepage and stress in unsaturated soil, a air-water two-phase percolation-stress coupling calculation program based on coupled TOUGH2 and FLAC3D was developed. The calculation program can simulate real air-water two phase flow, and can investigate the gas-water interaction of seepage process. The calculation program considers the direct interaction between gas-water two-phase seepage and soil skeleton deformation, reflects the process of porosity, permeability, capillary pressure and the change of soil physical and mechanical parameters, and achieve a more perfect gas-water two-phase seepage-stress coupling analysis. Furthermore, by comparing with classical drainage test and model test, it is verified that the program can accurately simulate the gas-water two-phase flow-stress interaction.

Table and Figures | Reference | Related Articles | Metrics
Steel surface defect detection based on deep learning 3D reconstruction
Huan LAN,Jian-bo YU
Journal of ZheJiang University (Engineering Science)    2023, 57 (3): 466-476.   DOI: 10.3785/j.issn.1008-973X.2023.03.004
Abstract   HTML PDF (5141KB) ( 297 )  

A new 3D reconstruction network was proposed in order to resolve the difficulty of 2D detection method to detect defects with depth information. CasMVSNet with multiscale feature enhancement (MFE-CasMVSNet) was combined with the technology of point cloud processing for steel plate surface defect detection. In order to improve the accuracy of 3D reconstruction, a position-oriented feature enhancement module (PFEM) and a multiscale feature adaptive fusion module (MFAFM) were proposed to effectively extract features and reduce information loss. A density clustering method, curvature-sparse-guided density-based spatial clustering of applications with noise (CS-DBSCAN), was proposed for accurately extracting defects in different parts, and the 3D detection box was introduced to locate and visualize defects. Experimental results show that compared with the reconstruction method based on images, MFE-CasMVSNet can realize the 3D reconstruction of steel plate surface more accurately and quickly. Compared with 2D detection, 3D visual defect detection can accurately obtain the 3D shape information of defects and realize the multi-dimensional detection of steel plate surface defects.

Table and Figures | Reference | Related Articles | Metrics
Ship detection algorithm in complex backgrounds via multi-head self-attention
Nan-jing YU,Xiao-biao FAN,Tian-min DENG,Guo-tao MAO
Journal of ZheJiang University (Engineering Science)    2022, 56 (12): 2392-2402.   DOI: 10.3785/j.issn.1008-973X.2022.12.008
Abstract   HTML PDF (1335KB) ( 353 )  

A ship object detection algorithm was proposed based on a multi-head self-attention (MHSA) mechanism and YOLO network (MHSA-YOLO), aiming at the characteristics of complex backgrounds, large differences in scale between classes and many small objects in inland rivers and ports. In the feature extraction process, a parallel self-attention residual module (PARM) based on MHSA was designed to weaken the interference of complex background information and strengthen the feature information of the ship objects. In the feature fusion process, a simplified two-way feature pyramid was developed so as to strengthen the feature fusion and representation ability. Experimental results on the Seaships dataset showed that the MHSA-YOLO method had a better learning ability, achieved 97.59% mean average precision in the aspect of object detection and was more effective compared with the state-of-the-art object detection methods. Experimental results based on a self-made dataset showed that MHSA-YOLO had strong generalization.

Table and Figures | Reference | Related Articles | Metrics
Bearing life prediction based on multi-scale features and attention mechanism
Ren-peng MO,Xiao-sheng SI,Tian-mei LI,Xu ZHU
Journal of ZheJiang University (Engineering Science)    2022, 56 (7): 1447-1456.   DOI: 10.3785/j.issn.1008-973X.2022.07.020
Abstract   HTML PDF (1709KB) ( 353 )  

A bearing RUL prediction method based on multi-scale features and attention mechanism was proposed aiming at the problem that the previous remaining useful life (RUL) prediction methods were insufficient in mining bearing degradation information and ignored the difference in the contribution of different features, which affected the prediction accuracy. Several time-domain and frequency-domain features of the original bearing vibration signal at multiple scales were calculated as the input feature set. The multi-scale feature set was input into the network, and the attention module was used to adaptively assign the best weights to different features. Then the convolutional neural network (CNN) module was used for deep feature extraction and multi-scale feature fusion. The RUL prediction value was obtained through the feedforward neural network (FNN) module mapping. The proposed method was applied to the public bearing datasets for comparative studies. Results showed the superior prediction performance of the proposed method.

Table and Figures | Reference | Related Articles | Metrics
Solving combustion chemical differential equations via physics-informed neural network
Yi-cun WANG,Jiang-kuan XING,Kun LUO,Hai-ou WANG,Jian-ren FAN
Journal of ZheJiang University (Engineering Science)    2022, 56 (10): 2084-2092.   DOI: 10.3785/j.issn.1008-973X.2022.10.020
Abstract   HTML PDF (2369KB) ( 302 )  

Two typical cases including the stiff system of ordinary differential equations ROBER problem and the steady-state mixture fraction equation in jet flame were selected in order to efficiently embed the complex physicochemical information of turbulent combustion into physics-informed neural networks (PINNs). The potential of PINNs in solving combustion chemical differential equations was explored. Results show that the PINNs model can correctly capture the evolution of the zero-dimensional stiff reaction system. PINNs solution accorded well with the conventional numerical solution for steady jet flame. The selection of residual points was particularly important for solving complex differential equations in the field of combustion and chemistry, which should be considered based on the specific configuration in detail.

Table and Figures | Reference | Related Articles | Metrics
Graph neural network based node embedding enhancement model for node classification
Ju-xiang ZENG,Ping-hui WANG,Yi-dong DING,Lin LAN,Lin-xi CAI,Xiao-hong GUAN
Journal of ZheJiang University (Engineering Science)    2023, 57 (2): 219-225.   DOI: 10.3785/j.issn.1008-973X.2023.02.001
Abstract   HTML PDF (1473KB) ( 263 )  

In reality, the structure of most graphs could be noisy, i.e., including some noisy edges or ignoring some edges that exist between nodes in practice. To solve these challenges, a novel differentiable similarity module (DSM), which boosted node representations by digging implict association between nodes to improve the accuracy of node classification, was presented. Basic representation of each target node was learnt by DSM using an ordinary graph neural network (GNN), similar node sets were selected in terms of node representation similarity and the basic representation of the similar nodes was integrated to boost the target node’s representation. Mathematically, DSM is differentiable, so it is possible to combine DSM as plug-in with arbitrary GNNs and train them in an end-to-end fashion. DSM enables to exploit the implicit edges between nodes and make the learned representations more robust and discriminative. Experiments were conducted on several public node classification datasets. Results demonstrated that with GNNs equipped with DSM, the classification accuracy can be significantly improved, for example, GAT-DSM outperformed GAT by significant margins of 2.9% on Cora and 3.5% on Citeseer.

Table and Figures | Reference | Related Articles | Metrics
Research progress on interlayer damage identification technology of slab track structures
Wei DU,Juan-juan REN,Shu-yi ZHANG,Jun-hong DU,Shi-jie DENG
Journal of ZheJiang University (Engineering Science)    2023, 57 (2): 353-366.   DOI: 10.3785/j.issn.1008-973X.2023.02.015
Abstract   HTML PDF (1642KB) ( 209 )  

Slab track suffers material performance decline and structural damage accumulation in the long-term service process under the coupling effect of train load and complex environment, resulting in a gradual deterioration of its service performance. The forms and causes of common interlayer damages on prefabricated slab track and double-block slab track in China were comprehensively discussed. The application of ground penetrating radar method, impact echo method and other local damage identification methods used in slab tracks were summarized. And it was proposed that combining multiple local damage identification techniques was the key to achieve accurate local damage identification of track structures. In addition, the overall damage identification technologies based on modal parameters, slab bed vibration signals and vehicle vibration signals were outlined. The need to expand the detection sample of field damages to improve the generalization of the overall identification method was pointed out. The advantages and limitations of various identification methods were analyzed in detail to provide guidance for improving the identification technology system of slab track structures in China and making scientific and reasonable maintenance strategies.

Table and Figures | Reference | Related Articles | Metrics
Multi-scale spatiotemporal influencing factors of bike-sharing parking demand
Biao XU,Qing-chang LU
Journal of ZheJiang University (Engineering Science)    2023, 57 (2): 380-391.   DOI: 10.3785/j.issn.1008-973X.2023.02.017
Abstract   HTML PDF (2519KB) ( 197 )  

In order to reveal the spatiotemporal relationship between urban multi-dimensional features and bike-sharing parking demand and their associated scales, combined with multi-source data in Shanghai, a multiscale geographically and temporally weighted regression model constrained by riding distance (RD-MGTWR) was constructed to explore the spatiotemporal heterogeneity patterns of the impact of built environment and regional economic attributes on parking demand. The model comparison analysis shows that the MGTWR model exhibits better explanatory power and reliability than the geographically and temporally weighted regression model (GTWR), and the introduction of riding distance further improves the robustness of the MGTWR model. Results show that the scale of the positive impact of socioeconomic attributes on parking demand is global, while the negative impact of location conditions presents local heterogeneity, and is most significant in the inner ring central area during the commuter morning peak. In addition, bus station density, metro station density and shopping service facility density with micro-spatial or temporal scales have positive and negative effects on parking demand. The findings of the scale effect of influencing factors can help guide parking facility zoning development and bike sharing time-sharing scheduling.

Table and Figures | Reference | Related Articles | Metrics
Traffic signal control method based on deep reinforcement learning
Zhi-min LIU,Bao-Lin YE,Yao-dong ZHU,Qing YAO,Wei-min WU
Journal of ZheJiang University (Engineering Science)    2022, 56 (6): 1249-1256.   DOI: 10.3785/j.issn.1008-973X.2022.06.024
Abstract   HTML PDF (1377KB) ( 238 )  

A traffic signal control method based on an improved deep reinforcement learning was proposed for an isolated intersection, aiming at a problem that the traffic signal control methods based on deep reinforcement learning were difficult to update the traffic signal control strategy in time. A new reward function of the proposed method was built by utilizing the real-time change of vehicle numbers at an intersection between two adjacent sampling time steps, whereby the dynamic change process of traffic status at the intersection was tracked and utilized in time. In addition, double network structure and experience playback were respectively used to improve the learning efficiency and convergence rate of the proposed method. SUMO simulation test results show that the proposed method can significantly shorten the average waiting time and average queue length of vehicles at the intersection, and improve the traffic efficiency at the intersection.

Table and Figures | Reference | Related Articles | Metrics
Task allocation method for Internet of vehicles spatial crowdsourcing with privacy protection
Xue-jiao LIU,Hui-min WANG,Ying-jie XIA,Si-wei ZHAO
Journal of ZheJiang University (Engineering Science)    2022, 56 (7): 1267-1275.   DOI: 10.3785/j.issn.1008-973X.2022.07.001
Abstract   HTML PDF (1158KB) ( 317 )  

A task allocation method for Internet of vehicles spatial crowdsourcing with privacy protection was proposed under the blockchain architecture in order to solve the problem that centralized spatial crowdsourcing server in the traditional spatial crowdsourcing of Internet of vehicles was untrusted and vulnerable to malicious attacks, which posed a great threat to users’ privacy. A distributed and trusted spatial crowdsourcing system of Internet of vehicles was designed based on the blockchain technology. The multi-key homomorphic encryption algorithm was adopted to distribute tasks, which supported task allocation of location ciphertext data of different vehicle users (keys). Then the possibility of privacy disclosure of vehicle users was reduced. The experimental results show that the proposed method can effectively protect users’ privacy information, reduce the computing overhead of task allocation by 34.3% compared with the existing methods, and improve the efficiency of task allocation.

Table and Figures | Reference | Related Articles | Metrics
Automatic hot spot detection method for photovoltaic aerial infrared image
Jie-feng XIA,Wu-qin TANG,Qiang YANG
Journal of ZheJiang University (Engineering Science)    2022, 56 (8): 1640-1647.   DOI: 10.3785/j.issn.1008-973X.2022.08.018
Abstract   HTML PDF (2340KB) ( 168 )  

A two-stage hot spot detection method of aerial infrared image was proposed to realize component level positioning and fine classification diagnosis of hot spot defects in infrared image, aiming at the problems of high cost, low efficiency and low accuracy of traditional inspection technology of photovoltaic power station. This method combined the traditional image processing technology with the deep learning method to further improve the accuracy and efficiency of defect diagnosis. Specifically, firstly, based on the difference between the gray values of the front and back scenes of aerial infrared images, a component segmentation method based on edge detection was proposed to extract the contour of photovoltaic components to achieve component level positioning. This method achieved the effective detection rate of photovoltaic components up to 99.3% with relatively small hardware requirements. Secondly, considering the differences in the causes, hazards and corresponding treatment methods of hot spots, an infrared defect classification model based on EfficientNet was proposed to finely classify the hot spots, so as to provide more accurate decision support for the operation and maintenance personnel of the power station. The model obtained hot spot classification accuracy of 97.0% when it occupied 20.17 MB. Through experimental comparison and analysis, it is demonstrated that the proposed method has greatly improved the efficiency and accuracy of defect diagnosis.

Table and Figures | Reference | Related Articles | Metrics
High sensitivity flexible tactile sensor with hierarchical tilted micro-pillar structure
Xiao-hui GUO,Wei-qiang HONG,Guo-qing ZHENG,Jing-yi WANG,Guo-peng TANG,Jin-yang YANG,Chao-qiang ZHUO,Yao-hua XU,Yu-nong ZHAO,Hong-wei ZHANG
Journal of ZheJiang University (Engineering Science)    2022, 56 (6): 1079-1087, 1126.   DOI: 10.3785/j.issn.1008-973X.2022.06.004
Abstract   HTML PDF (3894KB) ( 189 )  

A flexible tactile sensor was proposed based on the hierarchical titled micro-pillar structure, in order to meet the precise tactile sensing application requirements of capacitive flexible tactile sensors in wearable electronic devices, intelligent robots and other fields. When the sensor was subjected to external pressure, the titled micro-pillar was deformed in stages, resulting in an increase in the capacitance of the sensor. Combined with simulation and experiment, the influence of sensor structure characteristics on its sensitivity were studied, the relationship between sensor signal output and loading pressure designed with different structural parameters was revealed, and the structure of tactile sensor was optimized. The experimental data showed that the sensor had good sensitivity (0.44 kPa?1) and low detection limit (2.6 Pa), the response time was 40 ms, the maximum hysteresis error was 6.7% The sensor exhibits excellent stability and repeatability during 2 400 cycles of loading/unloading, and it can be extended to "skin" arrays of different sizes and shapes, the precise perception of the manipulator and the monitoring of motion postures of the human body were realized.

Table and Figures | Reference | Related Articles | Metrics
Finite element implementation and application of strength theory based cohesive zone model
Tian-xiang SHI,Xin ZHANG,Yang-yang WANG,Ke-hong ZHENG,Yong-qiang ZHANG
Journal of ZheJiang University (Engineering Science)    2023, 57 (3): 573-582.   DOI: 10.3785/j.issn.1008-973X.2023.03.015
Abstract   HTML PDF (2243KB) ( 278 )  

The two-dimensional strength theory based cohesive zone model (ST-CZM) was extended to the three-dimensional case for a wider application. Furthermore, the finite element implementation of the ST-CZM was carried out using the Abaqus user element subroutine (UEL). The validity and accuracy of the ST-CZM were validated by several typical numerical benchmarks. On this basis, the ST-CZM finite element model was used to simulate the bond-slip behavior of the interface between fiber reinforced polymer (FRP) and concrete, which extended the application of the ST-CZM in complex working conditions. Compared to the traditional “traction laws” based cohesive zone model (CZM), the ST-CZM provides improved flexibility in mode mixity and allows independent selection of strength models in normal and tangent directions. In addition, the ST-CZM exhibits better convergence performance and more accurate strength predictions compared to “traction laws” based CZMs. All examples show that compared to the traditional “traction laws” based CZM, the ST-CZM finite element model can better predict the peak stress of the bonding interface and simulate the mixed-mode damage process, showing more realistic cracking process.

Table and Figures | Reference | Related Articles | Metrics
Experimental research on wet-dry cycle of MICP cemented calcareous sand in seawater environment
Yi-long LI,Zhen GUO,Qiang XU,Yu-jie LI
Journal of ZheJiang University (Engineering Science)    2022, 56 (9): 1740-1749.   DOI: 10.3785/j.issn.1008-973X.2022.09.007
Abstract   HTML PDF (2670KB) ( 189 )  

In order to explore the applicability of microbial induced calcium carbonate precipitation (MICP) to cement calcareous sand in seawater environment and the wet-dry cycle resistance of MICP cemented bodies, the calcareous sands were cemented in seawater and freshwater environment respectively, and the wet-dry cycles were carried out in seawater environment. The element and mineral composition of the cemented bodies were analyzed based on the energy dispersive spectroscopy(EDS), X-ray diffraction(XRD). In addition, the relationships between mechanical properties, mass loss and wet-dry cycle were established through unconfined compressive strength test, weighing; and the weakening mechanism of the wet-dry cycle on samples were analyzed by scanning electron microscope (SEM). Results show that, in seawater environment, the cementation effect of MICP on calcareous sand is better than that in freshwater environment. The resistance to wet-dry cycle of calcareous sand cemented in seawater environment is larger than that cemented in freshwater environment. After 21 wet-dry cycles, the strength of cemented calcareous sand in seawater and freshwater environment decreased to 30% and 7.53% of the original samples, respectively. The wet-dry cycle reduces the particle surface roughness and intergranular cementation strength, which is manifested in the reduction of strength and stiffness of cemented calcareous sand in macro-characteristics.

Table and Figures | Reference | Related Articles | Metrics
Medical image segmentation method combining multi-scale and multi-head attention
Wan-liang WANG,Tie-jun WANG,Jia-cheng CHEN,Wen-bo YOU
Journal of ZheJiang University (Engineering Science)    2022, 56 (9): 1796-1805.   DOI: 10.3785/j.issn.1008-973X.2022.09.013
Abstract   HTML PDF (1159KB) ( 306 )  

A neural network based segmentation model MS2Net was proposed to automatically and accurately extract regions of interest from medical images. In order to better extract context information, a network architecture combining convolution and Transformer was proposed, which solved the problem that traditional convolution operations lacked the ability to acquire long-range dependencies. In the Transformer-based context extraction module, multi-head self-attention was used to obtain the similarity relationship between pixels. Based on the similarity relationship, the features of each pixel were fused, so that the network had a global view, while the relative positional encoding enabled Transformer to retain the structural information of an input feature map. Aiming at making the network adapt to different sizes of regions of interest, the multi-scale features of decoders were used by MS2Net and a multi-scale attention mechanism was proposed. The group channel attention and the group spatial attention were applied to a multi-scale feature map in turns, so that the reasonable multi-scale semantic information was selected adaptively by the network. MS2Net had achieved better intersection-over-union than advanced methods such as U-Net, CE-Net, DeepLab v3+, UTNet on both ISBI 2017 and CVC-ColonDB datasets, which reflected its excellent generalization ability.

Table and Figures | Reference | Related Articles | Metrics
Medical image segmentation method based on multi-source information fusion
Chang-chun YANG,Zan-ting YE,Ban-teng LIU,Ke WANG,Hai-dong CUI
Journal of ZheJiang University (Engineering Science)    2023, 57 (2): 226-234.   DOI: 10.3785/j.issn.1008-973X.2023.02.002
Abstract   HTML PDF (1105KB) ( 195 )  

The segmentation model construction and training based on single source data may lead to insufficient segmentation accuracy due to the defects of various imaging methods in medical images. Aiming at this problem, a medical image segmentation method based on multi-source information fusion was proposed. The FFDM and DBT data sources in the breast tumour microcalcification cluster lesion were used as examples to verify the effectiveness of the proposed method. The Yolov4 region candidate network was used to screen the suspicious regions of the FFDM data. DBT image was preprocessed by using the suspicious region information. The preprocessed DBT image was used as the input of the improved U-Net model to achieve lesion segmentation. Finally, through the fusion strategy of fault segmentation results based on sequential similarity discrimination, the multi-slice results in DBT were combined to complete the final lesion segmentation. True positive rate of 98.52%, false positive rate of 10.45% and accuracy of 94.07% were obtained from the FFDM and DBT data of 20 patients by using this method. Results show that the medical image segmentation method based on multi-source information fusion can effectively utilize the advantages of multi-source data, and achieve the rapid and accurate segmentation of lesions. The method can provide a novel solution for intelligent medical image diagnosis and treatment.

Table and Figures | Reference | Related Articles | Metrics
New method for news recommendation based on Transformer and knowledge graph
Li-zhou FENG,Yang YANG,You-wei WANG,Gui-jun YANG
Journal of ZheJiang University (Engineering Science)    2023, 57 (1): 133-143.   DOI: 10.3785/j.issn.1008-973X.2023.01.014
Abstract   HTML PDF (1590KB) ( 464 )  

A news recommendation method based on Transformer and knowledge graph was proposed to increase the auxiliary information and improve the prediction accuracy. The self-attention mechanism was used to obtain the connection between news words and news entities in order to combine news semantic information and entity information. The additive attention mechanism was employed to capture the influence of words and entities on news representation. Transformer was introduced to pick up the correlation information between clicked news of user and capture the change of user interest over time by considering the time-series characteristics of user preference for news. High-order structural information in knowledge graphs was used to fuse adjacent entities of the candidate news and enhance the integrity of the information contained in the candidate news embedding vector. The comparison experiments with five typical recommendation methods on two versions of the MIND news dataset show that the introduction of attention mechanism, Transformer and knowledge graph can improve the performance of the algorithm on news recommendation.

Table and Figures | Reference | Related Articles | Metrics