2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC)最新文献_第4页

LwF4IEE: An Incremental Learning Method for Interactive Event Extraction LwF4IEE:一种交互式事件抽取的增量学习方法

2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC) Pub Date : 2022-10-01 DOI: 10.1109/CyberC55534.2022.00026

Jiashun Duan, Xin Zhang, Chi Xu

{"title":"LwF4IEE: An Incremental Learning Method for Interactive Event Extraction","authors":"Jiashun Duan, Xin Zhang, Chi Xu","doi":"10.1109/CyberC55534.2022.00026","DOIUrl":"https://doi.org/10.1109/CyberC55534.2022.00026","url":null,"abstract":"Albeit great progress has been witnessed in event extraction, the accuracies achieved up to now by various automatic models still can not meet the performance requirements of some special applications such as disaster monitoring and rescue. It motivates us to introduce a new human-in-loop extraction mode called interactive event extraction (IEE), which works iteratively. Each iteration consists of three main steps: \"model recommending candidate results → manual selecting and correcting → model re-training and updating\". For candidate recommendation, we build an MRC (Machine Reading Comprehension)-based model that can output several most likely candidate elements, i.e., candidate triggers and arguments, by confidence evaluation. For model re-training and updating, we proposed an incremental learning method named LwF4IEE (Learning without Forgetting for IEE), which employs manual selected and corrected results as hard label and prediction of original model as soft label to avoid catastrophic forgetting. We conduct extensive experiments on datasets constructed from real-world Chinese texts. The results show that when setting the number of candidates to be 5, recalls of triggers and arguments reach 93.80% and 90.58% respectively, which is 11.51% and 11.33% higher compared with the basic MRC-based automatic extraction model. Moreover, LwF4IEE increases the recall of triggers by 2.71% on specific event types and only decreases by 0.24% on other types, achieving the purpose of learning without forgetting.","PeriodicalId":234632,"journal":{"name":"2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126807479","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

A Personalized Recommendation Fusing Tag Feature and Temporal Context 一种融合标签特征和时间上下文的个性化推荐

2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC) Pub Date : 2022-10-01 DOI: 10.1109/cyberc55534.2022.00058

Ling Li

引用次数: 0

Research on Multiple Targets Pedestrian Reidentification with Night Scene Image Enhancement 基于夜景图像增强的多目标行人再识别研究

2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC) Pub Date : 2022-10-01 DOI: 10.1109/CyberC55534.2022.00045

Minkang Zhang, Ding Chen, Yongxin Huang

{"title":"Research on Multiple Targets Pedestrian Reidentification with Night Scene Image Enhancement","authors":"Minkang Zhang, Ding Chen, Yongxin Huang","doi":"10.1109/CyberC55534.2022.00045","DOIUrl":"https://doi.org/10.1109/CyberC55534.2022.00045","url":null,"abstract":"Pedestrian reidentification is a popular research topic in the field of computer vision in recent years, and is a technique that uses computer vision techniques to determine whether a specific pedestrian is present in an image or video. After research and experiment, we found that YOLOv3-based pedestrian reidentification in practice has the problem of low accuracy rate of recognizing pedestrian pictures taken at night and cannot recognize multiple pedestrians at one time. In this paper, we improve the above problems by introducing a picture enhancement module to improve the brightness and defogging of night pictures before recognition, and improve the practice of averaging the distance values of multiple results for the same pedestrian to enable multiple targets pedestrian recognition. The experimental results demonstrate that the average accuracy rate of recognizing pedestrian pictures taken at night has improved from 6.85% to 80%, while the average accuracy rate of multiple targets pedestrian recognition has reached 85.9%, which is competent for multiple targets pedestrian recognition tasks at night.","PeriodicalId":234632,"journal":{"name":"2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121159862","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

A CVD Critical Level-aware Scheduling Model Based on Reinforcement Learning for ECG Service Request 基于强化学习的心电服务请求CVD临界电平感知调度模型

2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC) Pub Date : 2022-10-01 DOI: 10.1109/CyberC55534.2022.00039

Jian Gao, Jie Yu, Ning Wang, Panpan Feng, Huiqing Cheng, Bing Zhou, Zongmin Wang

引用次数: 0

Comprehensive defense scheme against container escape related to container management procedure 涉及集装箱管理程序的集装箱逃逸综合防范方案

2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC) Pub Date : 2022-10-01 DOI: 10.1109/cyberc55534.2022.00051

Zhimin Guo, Zhuo Lv, Nuannuan Li, Tao Yuan, Xue Gao, Zekun Yuan

引用次数: 0

Study on the automatic modeling method of 3D information model for substations 变电站三维信息模型自动建模方法研究

2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC) Pub Date : 2022-10-01 DOI: 10.1109/CyberC55534.2022.00053

Junfeng Ding, Haonan Zong, Jian Zhou, Deyong Wu, Xuan Chen, Lei Wang

{"title":"Study on the automatic modeling method of 3D information model for substations","authors":"Junfeng Ding, Haonan Zong, Jian Zhou, Deyong Wu, Xuan Chen, Lei Wang","doi":"10.1109/CyberC55534.2022.00053","DOIUrl":"https://doi.org/10.1109/CyberC55534.2022.00053","url":null,"abstract":"Reverse modeling is a kind of technology that transforms natural scenes into three-dimensional models. The models are generated manually by referring to the point cloud obtained by a laser scanner, which is time-consuming, laborious, and complicated. To address this problem, we propose an automatic modeling method for the point cloud data of substations. First, an algorithm is designed to automatically generate a component model library by referring to the standard structure of the substation equipment. Then, we improve the Euclidean clustering to segment disconnected point cloud data. Finally, corresponding points are found according to the SHOT feature descriptor, and each component is identified with Hough voting. After the location information for each element in the scene is obtained, the models can be transferred to the scene to replace the corresponding part in the point cloud, thus completing the process of automatic modeling. The experiment compares the results of improved Euclidean clustering and traditional Euclidean clustering. The clustering method in this paper has a significant improvement in execution efficiency. In addition, we also give the final modeling result of the method in this paper. Compared with the Delaunay triangulation and Poisson surface reconstruction methods, the model built by this method is more complete.","PeriodicalId":234632,"journal":{"name":"2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121422109","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

Big data encrypting transmission framework for a multi-AGV system 多agv系统大数据加密传输框架

2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC) Pub Date : 2022-10-01 DOI: 10.1109/CyberC55534.2022.00019

Tongpo Zhang, Yuxin Wan, M. López-Benítez, Enggee Lim, Fei Ma, Limin Yu

引用次数: 0

Deep Neural Networks Enhanced Phase Asynchronous Physical-Layer Network Coding 深度神经网络增强相位异步物理层网络编码

2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC) Pub Date : 2022-10-01 DOI: 10.1109/cyberc55534.2022.00037

Xuesong Wang, Lu Lu

引用次数: 1

Conditional Metadata Embedding Data Preprocessing Method for Semantic Segmentation 语义分割的条件元数据嵌入数据预处理方法

2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC) Pub Date : 2022-10-01 DOI: 10.1109/CyberC55534.2022.00057

Juntuo Wang, Qiaochu Zhao, Dongheng Lin, Erick Purwanto, K. Man

{"title":"Conditional Metadata Embedding Data Preprocessing Method for Semantic Segmentation","authors":"Juntuo Wang, Qiaochu Zhao, Dongheng Lin, Erick Purwanto, K. Man","doi":"10.1109/CyberC55534.2022.00057","DOIUrl":"https://doi.org/10.1109/CyberC55534.2022.00057","url":null,"abstract":"Semantic segmentation is one of the key research areas in computer vision, which has very important applications in areas such as autonomous driving and medical image diagnosis. In recent years, the technology has advanced rapidly, where current models have been able to achieve high accuracy and efficient speed on some widely used datasets. However, the semantic segmentation task still suffers from the inability to generate accurate boundaries in the case of insufficient feature information. Especially in the field of medical image segmentation, most of the medical image datasets usually have class imbalance issues and there are always variations in factors such as shape and color between different datasets and cell types. Therefore, it is difficult to establish general algorithms across different classes and robust algorithms that differ across different datasets. In this paper, we propose a conditional data preprocessing strategy, i.e., Conditional Metadata Embedding (CME) data preprocessing strategy. The CME data preprocessing method will embed conditional information to the training data, which can assist the model to better overcome the differences in the datasets and extract useful feature information in the images. The experimental results show that the CME data preprocessing method can help different models achieve higher segmentation performance on different datasets, which shows the high practicality and robustness of this method.","PeriodicalId":234632,"journal":{"name":"2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131125495","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

RTOSExtracter: Extracting user-defined functions in stripped RTOS-based firmware RTOSExtracter:在剥离的基于rtos的固件中提取用户定义的函数

2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC) Pub Date : 2022-10-01 DOI: 10.1109/CyberC55534.2022.00024

Xinguang Xie, Junjian Ye, Lifa Wu, Rong Li

{"title":"RTOSExtracter: Extracting user-defined functions in stripped RTOS-based firmware","authors":"Xinguang Xie, Junjian Ye, Lifa Wu, Rong Li","doi":"10.1109/CyberC55534.2022.00024","DOIUrl":"https://doi.org/10.1109/CyberC55534.2022.00024","url":null,"abstract":"In recent years, Real-Time Operating System (RTOS) has been widely used in the Internet of Things (IoT) devices in many fields. Meanwhile, IoT devices running RTOS are facing an increasing number of security vulnerabilities, which are caused mainly by user-defined functions. Therefore, researchers usually need to manually identify and analyze user-defined functions in the firmware to detect vulnerabilities. However, stripped RTOS-based firmware does not contain the debug symbols such as function names. There is no clear boundary between the system and user-defined functions, making it laborious and inefficient to identify user-defined functions from the thousands of functions.In this paper, we design and implement RTOSExtracter, an automated static analysis tool for identifying user-defined functions and their names in stripped RTOS-based firmware, which can be extended to support multiple RTOS types. This tool can disassemble the target firmware, recover the names of the task creation APIs, identify the parameter structure, and generate the parameter values that contain user-defined function addresses and function name addresses. To evaluate RTOSExtracter, we implemented a prototype of RTOSExtracter on IDA Pro with support for five common types of RTOS including FreeRTOS, LiteOS, RT-Thread, μC/OS-II, and μC/OS-III. We compiled 30 open-source projects covering these five RTOS types with 12 different compilers and optimizations and generated 275 firmware without the debug symbols to test RTOSExtracter. The experimental results show that RTOSExtracter identifies user-defined function addresses and function name addresses with high accuracy and low time cost. Furthermore, the case study shows that RTOSExtracter can effectively identify user-defined functions and their names in actual firmware.","PeriodicalId":234632,"journal":{"name":"2022 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128649230","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