2021 2nd International Conference on Computing and Data Science (CDS)最新文献_第9页

Performance comparison and optimization of mainstream NIDS systems in offline mode based on parallel processing technology 基于并行处理技术的主流网络入侵检测系统离线模式性能比较与优化

2021 2nd International Conference on Computing and Data Science (CDS) Pub Date : 2021-01-01 DOI: 10.1109/CDS52072.2021.00030

Tianyang Zhou

引用次数: 0

Multi-language Handwritten Numeral Recognition with Convolutional Neural Network 基于卷积神经网络的多语言手写数字识别

2021 2nd International Conference on Computing and Data Science (CDS) Pub Date : 2021-01-01 DOI: 10.1109/CDS52072.2021.00082

Yihan Wang

引用次数: 0

Predictive Modeling of Wildfires in the United States 美国野火的预测模型

2021 2nd International Conference on Computing and Data Science (CDS) Pub Date : 2021-01-01 DOI: 10.1109/CDS52072.2021.00102

Lang Qin, W. Shao, Guofei Du, Junlin Mou, R. Bi

引用次数: 0

High Precision Autocollimation Measurement Technology Based on Image Recognition 基于图像识别的高精度自准直测量技术

2021 2nd International Conference on Computing and Data Science (CDS) Pub Date : 2021-01-01 DOI: 10.1109/CDS52072.2021.00028

Li Zhongtang, Liu Mingyao, Luan Jianfeng

引用次数: 1

Comparison of stock price prediction ability based on GARCH and BP_ANN 基于GARCH和BP_ANN的股价预测能力比较

2021 2nd International Conference on Computing and Data Science (CDS) Pub Date : 2021-01-01 DOI: 10.1109/CDS52072.2021.00021

Ercheng Liu

引用次数: 0

Hardware-friendly model compression technique of DNN for edge computing 面向边缘计算的深度神经网络硬件友好模型压缩技术

2021 2nd International Conference on Computing and Data Science (CDS) Pub Date : 2021-01-01 DOI: 10.1109/CDS52072.2021.00066

Xinyun Liu

{"title":"Hardware-friendly model compression technique of DNN for edge computing","authors":"Xinyun Liu","doi":"10.1109/CDS52072.2021.00066","DOIUrl":"https://doi.org/10.1109/CDS52072.2021.00066","url":null,"abstract":"The research proposes a design methodology to compress the existing DNN models for low-cost edge devices. To reduce the computation complexity and memory cost, several novel model compression techniques are proposed. (1) A DNN model used to conduct image classification tasks is quantized into integer-based model for both the inference and training. 8-bit quantization is chosen in this work to balance the model training accuracy and cost. (2) A stochastic rounding scheme is implemented during the gradient backpropagation process to relieve the gradient diminishing risk. (3) To further reduce the training error caused by the gradient diminishing problem, a dynamic backpropagation algorithm is implemented. By dynamically scaling the magnitudes of gradient during the backpropagation, e.g. enlarging the magnitude of the gradient when it's too small to be quantized, it can effectively overcome the information loss due to the quantization error. As a result, such a DNN model for image classification is quantized into 8-bit model including training, which reduces the computation complexity by 8X and decreases the memory size by 6X Owing to the proposed dynamic backpropagation and stochastic training algorithms, the gradient diminishing issue during backpropagation is relieved. The training speed is reduced by 3X while classification error rates of state-of-art databases, e.g. ImageNet and CIFAR-10, are maintained similarly compared to the original model without quantization.","PeriodicalId":380426,"journal":{"name":"2021 2nd International Conference on Computing and Data Science (CDS)","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134151767","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

Applications of BP, Convolutional and RBF Networks BP、卷积和RBF网络的应用

2021 2nd International Conference on Computing and Data Science (CDS) Pub Date : 2021-01-01 DOI: 10.1109/CDS52072.2021.00099

Zebu Lan

引用次数: 0

Flexible Job Shop Scheduling Rules Mining Based on Random Forest 基于随机森林的柔性作业车间调度规则挖掘

2021 2nd International Conference on Computing and Data Science (CDS) Pub Date : 2021-01-01 DOI: 10.1109/CDS52072.2021.00045

Yizhong Wang

引用次数: 1

A new perspective of credit scoring for small and medium-sized enterprises based on invoice data 基于发票数据的中小企业信用评分新视角

2021 2nd International Conference on Computing and Data Science (CDS) Pub Date : 2021-01-01 DOI: 10.1109/CDS52072.2021.00088

Yuan Sun, Weifeng Jian, Yufeng Fu, Huiping Sun, Yuesheng Zhu, Zhiqiang Bai

{"title":"A new perspective of credit scoring for small and medium-sized enterprises based on invoice data","authors":"Yuan Sun, Weifeng Jian, Yufeng Fu, Huiping Sun, Yuesheng Zhu, Zhiqiang Bai","doi":"10.1109/CDS52072.2021.00088","DOIUrl":"https://doi.org/10.1109/CDS52072.2021.00088","url":null,"abstract":"Credit Scoring takes a prominent part in the finance of small and medium-sized enterprises (SMEs), and it is also an invaluable tool to predict credit default. However, due to the variety of market size, capital scale, and the limitations of Credit Scoring Model, it is difficult for SMEs to refer to large enterprises for Credit Scoring. And as an important reference, financial statements are insufficient and time window is inflexible, which would fail to make data reflect the enterprise's operating conditions correctly and timely, inaccurate credit prediction arising. Therefore, we offer an inspiring perspective to search elastic and time-independent evidence. Served as an indispensable basis of accounting in China, invoices take full notes on taxes of economic business, with more details about financial statements and more flexibility over periods, which can develop a sustainable approach to master the operation information of SMEs in time. To deal with invoice data of SMEs, we study influential variables under the first digit law inspired by Benford's law, apply machine learning techniques, and guide experiment by the construction of score card. It shows that our method formed by easy-to-accomplish steps is of applicability and effectiveness, to support powerfully the existing Credit Scoring system.","PeriodicalId":380426,"journal":{"name":"2021 2nd International Conference on Computing and Data Science (CDS)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125334298","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

Node Augmentation Methods for Graph Neural Network based Object Classification 基于图神经网络的对象分类的节点增强方法

2021 2nd International Conference on Computing and Data Science (CDS) Pub Date : 2021-01-01 DOI: 10.1109/CDS52072.2021.00101

Yifan Xue, Yixuan Liao, Xiaoxin Chen, Jingwei Zhao

{"title":"Node Augmentation Methods for Graph Neural Network based Object Classification","authors":"Yifan Xue, Yixuan Liao, Xiaoxin Chen, Jingwei Zhao","doi":"10.1109/CDS52072.2021.00101","DOIUrl":"https://doi.org/10.1109/CDS52072.2021.00101","url":null,"abstract":"Graph neural networks (GNNs) are powerful models for learning representations of relational data, and have achieved impressive results both academically and industrially. To further enhance the performance of GNNs on the most studied node classification problem, we present NodeAug, a novel augmentation method that operates on graph-structured data, yielding virtual nodes by mixing pairs of nodes and corresponding graph structures. We first generate isolated virtual nodes with features and labels being convex combinations of existing nodes, without considering edges, which is termed as NodeAug-I and can enhance many GNN variants, demonstrating the effectiveness of node augmentation. Still, it does not exploit the graph structure, which is essential in GNNs while incorporating the structure in augmentation is a key challenge. Regarding this, we further propose two novel algorithms that can mix information of neighbors as well, taking graph structures into account. We conduct experiments on a wide range of benchmarks, including Cora, CiteSeer, Pubmed, CoraFull, Amazon and CLUSTER, and obtain considerable enhancement for well-known models, including GCN, GraphSAGE, GIN and GAT.","PeriodicalId":380426,"journal":{"name":"2021 2nd International Conference on Computing and Data Science (CDS)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116743727","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: 8