Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering最新文献_第5页

An Innovative Approach To Software Modularization Based On The Artificial Fish Swarm Algorithm 基于人工鱼群算法的软件模块化创新方法

Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering Pub Date : 2022-10-21 DOI: 10.1145/3573428.3573602

Jianqiang Pan, Cheng Zhang, Huihui Jia

引用次数: 0

Hybrid Genetic Algorithms for Large Scale Optimization Problem 大规模优化问题的混合遗传算法

Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering Pub Date : 2022-10-21 DOI: 10.1145/3573428.3573739

Qingteng Guo, Qingshun Li, Xueshi Dong

引用次数: 1

Research on Recognition Method of Floating Objects on Water Surface Based on improved ResNet 基于改进ResNet的水面漂浮物识别方法研究

Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering Pub Date : 2022-10-21 DOI: 10.1145/3573428.3573710

Haonan Guo, Yangyi Ma, Lei Hua, D. Wan

引用次数: 0

Application of the perspective based on virtual reality technology to relieve anxiety in practice 基于视角的虚拟现实技术在缓解焦虑实践中的应用

Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering Pub Date : 2022-10-21 DOI: 10.1145/3573428.3573585

Qian Li, Huajie Sui

引用次数: 0

Self-attention Pyramidal Convolutional Network for Weakly-supervised Video Anomaly Detection 弱监督视频异常检测的自关注金字塔卷积网络

Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering Pub Date : 2022-10-21 DOI: 10.1145/3573428.3573698

Tianhao Liu, Yiheng Cai, Panjian Jun

引用次数: 0

Global Optimization Based on Hybrid Adaptive Differential Evolution Algorithm and Sooty Tern Optimization Algorithm 基于混合自适应差分进化算法和灰燕鸥优化算法的全局优化

Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering Pub Date : 2022-10-21 DOI: 10.1145/3573428.3573789

Hui Yang, Luo Jin, Mingyue Yang, Li Jin, Jiabo Jian

引用次数: 0

Rapid detecting of n-3 PUFAs content of enriched eggs during household storage based on near-infrared reflectance spectroscopy 基于近红外反射光谱法快速检测家庭储藏中富集鸡蛋中n-3 PUFAs含量

Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering Pub Date : 2022-10-21 DOI: 10.1145/3573428.3573773

Chang-Tao Shen, Xinhui Guan, Xueze Lü, Yuanzhi Zhang, Xue Liu

{"title":"Rapid detecting of n-3 PUFAs content of enriched eggs during household storage based on near-infrared reflectance spectroscopy","authors":"Chang-Tao Shen, Xinhui Guan, Xueze Lü, Yuanzhi Zhang, Xue Liu","doi":"10.1145/3573428.3573773","DOIUrl":"https://doi.org/10.1145/3573428.3573773","url":null,"abstract":"Eggs have been proved to be an ideal delivery for the human being to enhance the intake of Omega-3 polyunsaturated fatty acids (n-3 PUFAs). The variation of n-3 PUFAs concentration of the enriched eggs in household storage is a common concern, so, how to measure the n-3 PUFAs concentration in enriched eggs avoiding complicated and time-consuming procedures is a meaningful study. Near infrared spectroscopy (NIRS) technique was applied to achieve this aim in this research. 200 n-3 PUFAs enriched eggs were selected and stored under different household storage conditions (refrigeration and room temperature). Whole egg and liquid yolk for spectral scanning and measuring, Gas chromatography and NIRS were used to detect the content of n-3 PUFAs. The n-3 PUFAs content prediction models were developed based on NIRS. In comparison, the liquid yolk spectrum provided the best prediction using PLS combined with second derivative (2nd D) and Wavelet transform (WT), that Rc2=0.95, RMSECV=5.71, Rp2=0.83, RMSEP=16.9. This research demonstrated that NIRS technique coupled with opportune chemometric algorithms could be applied as a rapid detection technique for the content of n-3 PUFAs in eggs.","PeriodicalId":314698,"journal":{"name":"Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123523179","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}

引用次数: 0

Building Semantic Segmentation Through an Ensemble Architecture of UNet and Graph Convolution Networks 通过UNet和图卷积网络的集成体系结构构建语义分割

Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering Pub Date : 2022-10-21 DOI: 10.1145/3573428.3573445

Yuxin Shang, Hao Cuib, Haitao Guo

{"title":"Building Semantic Segmentation Through an Ensemble Architecture of UNet and Graph Convolution Networks","authors":"Yuxin Shang, Hao Cuib, Haitao Guo","doi":"10.1145/3573428.3573445","DOIUrl":"https://doi.org/10.1145/3573428.3573445","url":null,"abstract":"Extracting building from remote sensing images remains a challenge because some obstacles, such as trees, cars, farmland, and shadows, disturb the identification accuracy. In recnet years, deep convolution neural networks have shown significant improvements in computer vision, however it is still difficult to extract irregular and small buildings. To develop the extraction accuracy of building, we propose a deep learning model, which combined UNet and graph convolution network, to capture the long-range information. Seven deep convolution neural networks were used on the Massachusetts Buildings Dataset for comparison. And the proposed model, which performed best among these compared models, achieved a building Intersection over Union (IOU) of 71.10%, mean IOU of 81.88%, recall of 81.90%, precision 84.36% and F1 score of 83.11%. This result indicated that the proposed model can accurately extract building objects. The visualization results showed that the proposed model performs better in extracting irregular and small buildings and caneffectively extracts buildings from high-resolution remote-sensing images.","PeriodicalId":314698,"journal":{"name":"Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering","volume":"126 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122022261","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}

引用次数: 0

Machine learning performance comparison of CSI 300 stock price movement prediction after disclosure of corporate annual reports of year 2020 2020年企业年报披露后沪深300股价走势预测的机器学习性能比较

Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering Pub Date : 2022-10-21 DOI: 10.1145/3573428.3573725

Fengyu Han, Yue Wang

{"title":"Machine learning performance comparison of CSI 300 stock price movement prediction after disclosure of corporate annual reports of year 2020","authors":"Fengyu Han, Yue Wang","doi":"10.1145/3573428.3573725","DOIUrl":"https://doi.org/10.1145/3573428.3573725","url":null,"abstract":"In the current stock market, machine learning based investment robots are widely applied to predict stock price movement. This work studies predicting the stock price movement on the next day just after the release of the annual reports of enterprises, which is different from the scenarios of related work. We use five kinds of machine learning methods, including neural network, decision tree, random forest, logistic regression, prototypical network (a few-shot learning method). Different financial indicators (core and extended) are used in the experiments of this work. We obtain these financial data from the EastMoney website, and we get the results indicating that the machine learning models we use do not behave well to predict companies' price trend. Besides, we also filtrate those stocks which have ROE above 0.15 and net profit cash ratio above 0.9 in hope of improving prediction under this good financial criterion. We conclude that the stock price tendency's predictability on the next day just after the release is weak, and we get the highest accuracy about only 59.6% and the highest precision about only 56% using the random forest classifier which is the best, but the corresponding recall is low (only 13%), and the filtering does not help to enhance the performance very much. Among the models we use, random forest is the best classifier and prototypical network performs better than MLP.","PeriodicalId":314698,"journal":{"name":"Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117220863","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}

引用次数: 1

Adaptive network selection strategy for vehicle velocity in multiple lanes 多车道车辆速度自适应网络选择策略

Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering Pub Date : 2022-10-21 DOI: 10.1145/3573428.3573510

Xinyi Liu, Chenqi Di, Cheng Ke, Xizheng Ke, Haotong Zhang, Xin-xing Zhao

引用次数: 0