Pub Date : 2022-12-09DOI: 10.1109/ACAIT56212.2022.10137944
Lei Yang, H. Huang, Suli Bai, Yanhong Liu
Medical image segmentation is a basal and essential task for computer-aided diagnosis and quantification of diseases. However, robust and precise medical image segmentation is still a challenging task on account of much factors, such as complex backgrounds, overlapping structures, high variation of appearances and low contrast. Recently, with the strong support of deep convolutional neural networks (DCNNs), the encoder-decoder based segmentation networks have been the popular detection schemes for medical image analysis, yet image segmentation based on DCNNs still faces some limitations, such as restricted receptive field, limited information flow, etc. To address such challenges, a novel dual-branch deep residual U-Net network is proposed in this paper for medical image detection which provides more avenues for information flow to gather both high-level and low-level feature maps and a greater depth of contextual data.A residual U-Net network is constructed for efficient feature expression using residual learning, attention block, and feature expression. Meanwhile, fused with atrous spatial pyramid pooling (ASPP) block and squeeze-and-excitation (SE) block, The residual U-Net network is suggested to embed an attention fusion block to gather multi-scale contextual data. On the basis, To fully utilize local contextual data and increase segmentation precision, a dual-branch deep residual U-Net network is built by stacking two residual U-Net networks. Combined with multiple public benchmark data sets on medical images, including the CVC-ClinicDB, the GIAS set and LUNA16 set, experimental results indicate the superior ability of proposed segmentation network on medical image segmentation compared with other advanced segmentation models.
{"title":"An Automatic Medical Image Segmentation Approach via Dual-Branch Network","authors":"Lei Yang, H. Huang, Suli Bai, Yanhong Liu","doi":"10.1109/ACAIT56212.2022.10137944","DOIUrl":"https://doi.org/10.1109/ACAIT56212.2022.10137944","url":null,"abstract":"Medical image segmentation is a basal and essential task for computer-aided diagnosis and quantification of diseases. However, robust and precise medical image segmentation is still a challenging task on account of much factors, such as complex backgrounds, overlapping structures, high variation of appearances and low contrast. Recently, with the strong support of deep convolutional neural networks (DCNNs), the encoder-decoder based segmentation networks have been the popular detection schemes for medical image analysis, yet image segmentation based on DCNNs still faces some limitations, such as restricted receptive field, limited information flow, etc. To address such challenges, a novel dual-branch deep residual U-Net network is proposed in this paper for medical image detection which provides more avenues for information flow to gather both high-level and low-level feature maps and a greater depth of contextual data.A residual U-Net network is constructed for efficient feature expression using residual learning, attention block, and feature expression. Meanwhile, fused with atrous spatial pyramid pooling (ASPP) block and squeeze-and-excitation (SE) block, The residual U-Net network is suggested to embed an attention fusion block to gather multi-scale contextual data. On the basis, To fully utilize local contextual data and increase segmentation precision, a dual-branch deep residual U-Net network is built by stacking two residual U-Net networks. Combined with multiple public benchmark data sets on medical images, including the CVC-ClinicDB, the GIAS set and LUNA16 set, experimental results indicate the superior ability of proposed segmentation network on medical image segmentation compared with other advanced segmentation models.","PeriodicalId":398228,"journal":{"name":"2022 6th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131022596","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the gradual expansion of the scale of”coal to electricity” users, the number of complaints about the heating effect, electricity safety and heating equipment safety guarantee in the clean heating operation process continues increasing. It is not possible to warn of possible emergencies in advance, and can be only remedied afterwards, completely in a passive response state. Therefore, rapid and accurate positioning of the key link is an urgent problem to be solved, but also the key to improve user satisfaction. Aimed at above problems, this paper established an automatic, information and intelligent electric heating risk warning mechanism. Based on the embedded feature selection algorithm and the DBSCAN adaptive clustering algorithm, a standardized vocabulary of customer appeals and complaint topics were constructed, combined with user historical electricity consumption data, and through the monitoring and matching of risk topics, an early warning model of customer electric heating abnormal risks was established. The model proposed in the article has strong practicability and provides strong support for lean management on the grid side, precise positioning of problems on the operation and maintenance side, government-side management decisionmaking and user satisfaction, and can promote safe, reliable and economical operation of the grid.
{"title":"Analysis and Research on Electric Heating Risk Early Warning Based on Embedded Feature Selection and DBSCAN Adaptive Clustering","authors":"Hui Xu, Lu Zhang, Longfei Ma, Xianglong Li, Siyue Lu, Shaokun Chen, Yifeng Ding, Wenbin Zhou","doi":"10.1109/ACAIT56212.2022.10137835","DOIUrl":"https://doi.org/10.1109/ACAIT56212.2022.10137835","url":null,"abstract":"With the gradual expansion of the scale of”coal to electricity” users, the number of complaints about the heating effect, electricity safety and heating equipment safety guarantee in the clean heating operation process continues increasing. It is not possible to warn of possible emergencies in advance, and can be only remedied afterwards, completely in a passive response state. Therefore, rapid and accurate positioning of the key link is an urgent problem to be solved, but also the key to improve user satisfaction. Aimed at above problems, this paper established an automatic, information and intelligent electric heating risk warning mechanism. Based on the embedded feature selection algorithm and the DBSCAN adaptive clustering algorithm, a standardized vocabulary of customer appeals and complaint topics were constructed, combined with user historical electricity consumption data, and through the monitoring and matching of risk topics, an early warning model of customer electric heating abnormal risks was established. The model proposed in the article has strong practicability and provides strong support for lean management on the grid side, precise positioning of problems on the operation and maintenance side, government-side management decisionmaking and user satisfaction, and can promote safe, reliable and economical operation of the grid.","PeriodicalId":398228,"journal":{"name":"2022 6th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132522586","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sarcasm is a meaningful and effective form of expression which people often use to express sentiments that are contrary to their literal meaning. It is fairly common to encounter such expressions on social media platforms. Comparing with the traditional approach of text sarcasm detection, multi-modal sarcasm detection is proved to be more effective when dealing with information on social networks with various forms of communication. In this work, a prompt-tuning method is proposed for multi-modal sarcasm detection (Pmt-MmSD). Specifically, to model the incongruity of text modalities, we first build a prompt-PLM network. Second, to model the text-image incongruity, an inter-modality attention network (ImAN) is designed based on self-attention mechanism. In addition, we utilize the pre-trained Vision Transformer (ViT) network to process the image modality. Extensive experiments demonstrated the effectiveness of the proposed Pmt-MmSD model for multi-modal sarcasm detection, which significantly outperforms the state-of-the-art results.
{"title":"Multi-Modal Sarcasm Detection with Prompt-Tuning","authors":"Daijun Ding, Hutchin Huang, Bowen Zhang, Cheng Peng, Yangyang Li, Xianghua Fu, Liwen Jing","doi":"10.1109/ACAIT56212.2022.10137937","DOIUrl":"https://doi.org/10.1109/ACAIT56212.2022.10137937","url":null,"abstract":"Sarcasm is a meaningful and effective form of expression which people often use to express sentiments that are contrary to their literal meaning. It is fairly common to encounter such expressions on social media platforms. Comparing with the traditional approach of text sarcasm detection, multi-modal sarcasm detection is proved to be more effective when dealing with information on social networks with various forms of communication. In this work, a prompt-tuning method is proposed for multi-modal sarcasm detection (Pmt-MmSD). Specifically, to model the incongruity of text modalities, we first build a prompt-PLM network. Second, to model the text-image incongruity, an inter-modality attention network (ImAN) is designed based on self-attention mechanism. In addition, we utilize the pre-trained Vision Transformer (ViT) network to process the image modality. Extensive experiments demonstrated the effectiveness of the proposed Pmt-MmSD model for multi-modal sarcasm detection, which significantly outperforms the state-of-the-art results.","PeriodicalId":398228,"journal":{"name":"2022 6th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132979905","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ACAIT56212.2022.10137854
Yaqi Chen, Dan Qu, Wenlin Zhang, Fen Yu, Haotong Zhang, Xukui Yang
Low-resource automatic speech recognition is a chal- lenging task. To solve this issue, multilingual meta-learning learns a better model initialization from many source language tasks, allowing for rapid adaption to the target language. However, due to the lack of limitations on multilingual pre-training, the shared semantic space of different languages is difficult to learn. In this work, we propose an adversarial meta-learning training approach to solve this problem. By using the adversarial auxiliary aim of language identification in the meta-learning algorithm, it will guide the model encoder to generate language-independent embedding features, which can improve model generalization. And we use Wasserstein distance and temporal normalization to optimize our adversarial training, making the training more stable and easier. The approach is evaluated on the IARPA BABEL. The results reveal that our approach only requires half as many meta learning training epochs to attain comparable multilingual pre-training performance. It also outperforms the meta learning in all target languages fine-tuning and achieves comparable performance in small data scales. Specially, it can reduce CER from 71% to 62% with fine-tuning 25% of Vietnamese data. Finally, we show why our approach is superior than others by using t-SNE.
{"title":"Adversarial Meta Learning Improves Low-Resource Speech Recognition","authors":"Yaqi Chen, Dan Qu, Wenlin Zhang, Fen Yu, Haotong Zhang, Xukui Yang","doi":"10.1109/ACAIT56212.2022.10137854","DOIUrl":"https://doi.org/10.1109/ACAIT56212.2022.10137854","url":null,"abstract":"Low-resource automatic speech recognition is a chal- lenging task. To solve this issue, multilingual meta-learning learns a better model initialization from many source language tasks, allowing for rapid adaption to the target language. However, due to the lack of limitations on multilingual pre-training, the shared semantic space of different languages is difficult to learn. In this work, we propose an adversarial meta-learning training approach to solve this problem. By using the adversarial auxiliary aim of language identification in the meta-learning algorithm, it will guide the model encoder to generate language-independent embedding features, which can improve model generalization. And we use Wasserstein distance and temporal normalization to optimize our adversarial training, making the training more stable and easier. The approach is evaluated on the IARPA BABEL. The results reveal that our approach only requires half as many meta learning training epochs to attain comparable multilingual pre-training performance. It also outperforms the meta learning in all target languages fine-tuning and achieves comparable performance in small data scales. Specially, it can reduce CER from 71% to 62% with fine-tuning 25% of Vietnamese data. Finally, we show why our approach is superior than others by using t-SNE.","PeriodicalId":398228,"journal":{"name":"2022 6th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133881718","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ACAIT56212.2022.10138004
Qun Jia, Jing Li, Yongze Li, Yilin Liu, Xiaoqing Gao, Bin Li
In the process of using sewage detecting for monitoring and assessment of drug situation, the authenticity and objectivity of the sewage sampling data is directly related to the true grasp of the drug situation. In this paper, an unattended and traceable of full process automatic sewage sampling and traceability system is developed, which enhances the automation of the sampling process and improves the safety and reliability of the data. The paper firstly describes the overall design of the automatic sewage sampling and traceability system, which mainly consists of a sewage sampler, a wireless transmission module, a mobile processing centre and a backstage management system. The structural design of the sampler is then described in detail, including the distribution of sensors on the sampler box, the implementation of the various sampling methods and sampling modes, as well as the traceability of the sampling process and the function realization of abnormal state alarm. Finally, the system software is introduced, including the mobile application and the backstage management system.
{"title":"Sewage Automatic Sampling and Traceability System for Monitoring and Assessment of Drug Situation","authors":"Qun Jia, Jing Li, Yongze Li, Yilin Liu, Xiaoqing Gao, Bin Li","doi":"10.1109/ACAIT56212.2022.10138004","DOIUrl":"https://doi.org/10.1109/ACAIT56212.2022.10138004","url":null,"abstract":"In the process of using sewage detecting for monitoring and assessment of drug situation, the authenticity and objectivity of the sewage sampling data is directly related to the true grasp of the drug situation. In this paper, an unattended and traceable of full process automatic sewage sampling and traceability system is developed, which enhances the automation of the sampling process and improves the safety and reliability of the data. The paper firstly describes the overall design of the automatic sewage sampling and traceability system, which mainly consists of a sewage sampler, a wireless transmission module, a mobile processing centre and a backstage management system. The structural design of the sampler is then described in detail, including the distribution of sensors on the sampler box, the implementation of the various sampling methods and sampling modes, as well as the traceability of the sampling process and the function realization of abnormal state alarm. Finally, the system software is introduced, including the mobile application and the backstage management system.","PeriodicalId":398228,"journal":{"name":"2022 6th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127708864","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ACAIT56212.2022.10137949
Chuanfu Yang, Xiujiang Fu, K. Zhu, Jian Zhang, Chunyang Xiong
In order to improve the anti-seismic performance and detection ability of highway tunnel, a performance-based anti-seismic design method of highway tunnel based on simulated annealing algorithm is proposed. Taking model material, design parameters of highway tunnel structure and counterweight of frame structure under earthquake excitation as constraint index parameter set, the seismic structural performance test model of highway tunnel is constructed. The parameter adaptive estimation method is used to estimate the influence coefficient of seismic strength and heavy load parameters of highway tunnel structure. The characteristic equation of seismic tensile stress is established by simulated annealing algorithm, and the seismic design of highway tunnel structure is realized by parameter optimization. The test results show that the seismic performance detection ability of this method is good, and the tension prediction accuracy can reach 99.3%, which improves the stability of highway tunnel structure.
{"title":"Performance Seismic Design Method of Highway Tunnel Structure Based on Simulated Annealing Algorithm","authors":"Chuanfu Yang, Xiujiang Fu, K. Zhu, Jian Zhang, Chunyang Xiong","doi":"10.1109/ACAIT56212.2022.10137949","DOIUrl":"https://doi.org/10.1109/ACAIT56212.2022.10137949","url":null,"abstract":"In order to improve the anti-seismic performance and detection ability of highway tunnel, a performance-based anti-seismic design method of highway tunnel based on simulated annealing algorithm is proposed. Taking model material, design parameters of highway tunnel structure and counterweight of frame structure under earthquake excitation as constraint index parameter set, the seismic structural performance test model of highway tunnel is constructed. The parameter adaptive estimation method is used to estimate the influence coefficient of seismic strength and heavy load parameters of highway tunnel structure. The characteristic equation of seismic tensile stress is established by simulated annealing algorithm, and the seismic design of highway tunnel structure is realized by parameter optimization. The test results show that the seismic performance detection ability of this method is good, and the tension prediction accuracy can reach 99.3%, which improves the stability of highway tunnel structure.","PeriodicalId":398228,"journal":{"name":"2022 6th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129342815","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ACAIT56212.2022.10138006
Xin Wang, Jiale Ren, Wei Shi, Tao Wang, Xuhui Guo, Yiyuan Han
Silver coin is an important circulating currency in modern China, and the edge teeth of silver coins are the key factor to identify its authenticity. However it is difficult for some hobbyists to distinguish the authenticity. So we propose an improved yoloV5 neural network algorithm, which can distinguish the authenticity of silver coin through its edge tooth images, and the value of mAP is more than 0.8. The algorithm in this paper adopts the Self-Attention mechanism, which can make full use of the correlation between image pixels and fully focus on the key details in the image, so that the network model can capture the global features of the image when learning a few parameters. Compared with yoloV5, the improved network model in this paper performs better on the public data set. No matter the value of mAP, FLOPs or average processing speed all have improved significantly. In addition, this paper also constructs a set of silver coin edge tooth images data set to facilitate relevant research in the future.
{"title":"Improved YoloV5 for the Authenticity Identification of Silver Coins in Modern China","authors":"Xin Wang, Jiale Ren, Wei Shi, Tao Wang, Xuhui Guo, Yiyuan Han","doi":"10.1109/ACAIT56212.2022.10138006","DOIUrl":"https://doi.org/10.1109/ACAIT56212.2022.10138006","url":null,"abstract":"Silver coin is an important circulating currency in modern China, and the edge teeth of silver coins are the key factor to identify its authenticity. However it is difficult for some hobbyists to distinguish the authenticity. So we propose an improved yoloV5 neural network algorithm, which can distinguish the authenticity of silver coin through its edge tooth images, and the value of mAP is more than 0.8. The algorithm in this paper adopts the Self-Attention mechanism, which can make full use of the correlation between image pixels and fully focus on the key details in the image, so that the network model can capture the global features of the image when learning a few parameters. Compared with yoloV5, the improved network model in this paper performs better on the public data set. No matter the value of mAP, FLOPs or average processing speed all have improved significantly. In addition, this paper also constructs a set of silver coin edge tooth images data set to facilitate relevant research in the future.","PeriodicalId":398228,"journal":{"name":"2022 6th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"2590 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128803647","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ACAIT56212.2022.10137975
Rui Qu
The identification and procedural orientation of juvenile criminal responsibility is an important part of maintaining social stability, standardizing social order and maintaining legal fairness and justice. Aiming at the low accuracy of the traditional method for locating juvenile criminal responsibility procedure, a method for locating juvenile criminal responsibility procedure based on multi-source data fusion is proposed. According to the distribution of explanatory variables of the orientation of juvenile criminal responsibility procedure, and taking the age of juvenile criminal responsibility, the protection of legal interests, social interests and other factors as reference variables, this paper makes a dynamic analysis of the statistical orientation of juvenile criminal responsibility procedure using multi-source information scheduling method. Based on the analysis results, according to the matching filter detection of the statistical information of the juvenile criminal responsibility procedure, a guidance model for juvenile criminal proceedings is constructed to obtain the statistical information of the juvenile criminal responsibility procedure. Under the parameters of the constraint index model, the statistical information characteristics of the transfer of the constraint index of juvenile criminal procedure law are extracted. The multi-source data fusion method is used for fuzzy clustering of the extracted feature quantities. According to the result of clustering, the criminal responsibility procedure for minors is determined. The simulation results show that the method has a high accuracy in determining the juvenile criminal responsibility procedure, and improves the scientificity, rationality and progressiveness of the age division of juvenile criminal responsibility.
{"title":"A Method for Locating Juvenile Criminal Responsibility Procedure Based on Multi-Source Data Fusion","authors":"Rui Qu","doi":"10.1109/ACAIT56212.2022.10137975","DOIUrl":"https://doi.org/10.1109/ACAIT56212.2022.10137975","url":null,"abstract":"The identification and procedural orientation of juvenile criminal responsibility is an important part of maintaining social stability, standardizing social order and maintaining legal fairness and justice. Aiming at the low accuracy of the traditional method for locating juvenile criminal responsibility procedure, a method for locating juvenile criminal responsibility procedure based on multi-source data fusion is proposed. According to the distribution of explanatory variables of the orientation of juvenile criminal responsibility procedure, and taking the age of juvenile criminal responsibility, the protection of legal interests, social interests and other factors as reference variables, this paper makes a dynamic analysis of the statistical orientation of juvenile criminal responsibility procedure using multi-source information scheduling method. Based on the analysis results, according to the matching filter detection of the statistical information of the juvenile criminal responsibility procedure, a guidance model for juvenile criminal proceedings is constructed to obtain the statistical information of the juvenile criminal responsibility procedure. Under the parameters of the constraint index model, the statistical information characteristics of the transfer of the constraint index of juvenile criminal procedure law are extracted. The multi-source data fusion method is used for fuzzy clustering of the extracted feature quantities. According to the result of clustering, the criminal responsibility procedure for minors is determined. The simulation results show that the method has a high accuracy in determining the juvenile criminal responsibility procedure, and improves the scientificity, rationality and progressiveness of the age division of juvenile criminal responsibility.","PeriodicalId":398228,"journal":{"name":"2022 6th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115818189","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ACAIT56212.2022.10137855
Jia Liu
In order to improve the accuracy of English long sentence machine translation, the fuzzy semantic optimal selection model of English long sentence machine translation is constructed by combining fuzzy semantic optimal selection and feature extraction methods. The fuzzy semantic optimal selection model of English long sentence machine translation based on adaptive learning of machine neural network is proposed, and the constraint object model of fuzzy semantic selection of English long sentence machine translation is constructed. The method of context correlation mapping is used to analyze the fuzzy semantic features and construct the ontology structure model in the process of English long sentence machine translation. The linear mapping and statistical information analysis of English long sentence machine translation are realized by using the linear semantic ontology structure mapping mechanism and the corresponding text sequence parameter mapping in the dictionary, and the language semantic correlation calculation model of the optimal selection of fuzzy semantics in English long sentence machine translation is established. The machine neural network adaptive learning method is adopted to realize the segmented learning control of the non-sentence backbone in the process of fuzzy semantic selection of English long sentence machine translation. Weighted learning and adaptive weight analysis are realized according to the machine neural network adaptive learning result of the optimal selection of fuzzy semantic of English long sentence machine translation, and the optimal design of fuzzy semantic optimal selection model of English long sentence machine translation is realized. The simulation results show that this method is robust and the evaluation result is accurate, which improves the accuracy and anti-interference of English long sentence machine translation.
{"title":"Research on the Optimal Selection Method of Fuzzy Semantics in English Long Sentence Machine Translation","authors":"Jia Liu","doi":"10.1109/ACAIT56212.2022.10137855","DOIUrl":"https://doi.org/10.1109/ACAIT56212.2022.10137855","url":null,"abstract":"In order to improve the accuracy of English long sentence machine translation, the fuzzy semantic optimal selection model of English long sentence machine translation is constructed by combining fuzzy semantic optimal selection and feature extraction methods. The fuzzy semantic optimal selection model of English long sentence machine translation based on adaptive learning of machine neural network is proposed, and the constraint object model of fuzzy semantic selection of English long sentence machine translation is constructed. The method of context correlation mapping is used to analyze the fuzzy semantic features and construct the ontology structure model in the process of English long sentence machine translation. The linear mapping and statistical information analysis of English long sentence machine translation are realized by using the linear semantic ontology structure mapping mechanism and the corresponding text sequence parameter mapping in the dictionary, and the language semantic correlation calculation model of the optimal selection of fuzzy semantics in English long sentence machine translation is established. The machine neural network adaptive learning method is adopted to realize the segmented learning control of the non-sentence backbone in the process of fuzzy semantic selection of English long sentence machine translation. Weighted learning and adaptive weight analysis are realized according to the machine neural network adaptive learning result of the optimal selection of fuzzy semantic of English long sentence machine translation, and the optimal design of fuzzy semantic optimal selection model of English long sentence machine translation is realized. The simulation results show that this method is robust and the evaluation result is accurate, which improves the accuracy and anti-interference of English long sentence machine translation.","PeriodicalId":398228,"journal":{"name":"2022 6th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127385063","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ACAIT56212.2022.10137920
Qinglong Ge
There is a local sparsity of user interest data in current e-commerce, resulting in low accuracy of personalized product recommendation. An item personalized recommendation model based on improved local similarity prediction of CNN (LSPCNN) is constructed. Firstly, the convolutional neural network CNN is used to extract local features. Then, a regulating layer is added on the basis of CNN network, and the item scoring matrix is constructed for the initial users to make their interest locally characterized. Finally, CNN is used to predict the missing score, thus realizing personalized recommendation. Experimental results show that compared with the improved CNN network model and the collaborative filtering recommendation model based on hybrid neural network, the data sparsity of the proposed LSPCNN model is significantly reduced, and the mean absolute error (MAE) is smaller. Therefore, the proposed algorithm can accurately extract the local feature data that users are interested in, which improves the accuracy of e-commerce personalized recommendation, and has certain feasibility.
{"title":"E-Commerce Personalized Recommendation Based on Convolutional Neural Network","authors":"Qinglong Ge","doi":"10.1109/ACAIT56212.2022.10137920","DOIUrl":"https://doi.org/10.1109/ACAIT56212.2022.10137920","url":null,"abstract":"There is a local sparsity of user interest data in current e-commerce, resulting in low accuracy of personalized product recommendation. An item personalized recommendation model based on improved local similarity prediction of CNN (LSPCNN) is constructed. Firstly, the convolutional neural network CNN is used to extract local features. Then, a regulating layer is added on the basis of CNN network, and the item scoring matrix is constructed for the initial users to make their interest locally characterized. Finally, CNN is used to predict the missing score, thus realizing personalized recommendation. Experimental results show that compared with the improved CNN network model and the collaborative filtering recommendation model based on hybrid neural network, the data sparsity of the proposed LSPCNN model is significantly reduced, and the mean absolute error (MAE) is smaller. Therefore, the proposed algorithm can accurately extract the local feature data that users are interested in, which improves the accuracy of e-commerce personalized recommendation, and has certain feasibility.","PeriodicalId":398228,"journal":{"name":"2022 6th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122633052","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：