2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC)最新文献

Exploring Generalizability of Fine-Tuned Models for Fake News Detection 探索假新闻检测微调模型的泛化性

2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC) Pub Date : 2022-12-01 DOI: 10.1109/CIC56439.2022.00022

Abhijit Suprem, Sanjyot Vaidya, C. Pu

{"title":"Exploring Generalizability of Fine-Tuned Models for Fake News Detection","authors":"Abhijit Suprem, Sanjyot Vaidya, C. Pu","doi":"10.1109/CIC56439.2022.00022","DOIUrl":"https://doi.org/10.1109/CIC56439.2022.00022","url":null,"abstract":"The Covid-19 pandemic has caused a dramatic and parallel rise in dangerous misinformation, denoted an ‘infodemic’ by the CDC and WHO. Misinformation tied to the Covid-19 infodemic changes continuously; this can lead to performance degradation of fine-tuned models due to concept drift. Degredation can be mitigated if models generalize well-enough to capture some cyclical aspects of drifted data. In this paper, we explore generalizability of pre-trained and fine-tuned fake news detectors across 9 fake news datasets. We show that existing models often overfit on their training dataset and have poor performance on unseen data. However, on some subsets of unseen data that overlap with training data, models have higher accuracy. Based on this observation, we also present KMeans-Proxy, a fast and effective method based on K-Means clustering for quickly identifying these overlapping subsets of unseen data. KMeans-Proxy improves generalizability on unseen fake news datasets by 0.1-0.2 f1-points across datasets. We present both our generalizability experiments as well as KMeans-Proxy to further research in tackling the fake news problem.","PeriodicalId":170721,"journal":{"name":"2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114889878","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: 3

An Imperative to Assess Socio-Technical Impact of Algorithms in Online Spaces and Wresting Responsibility from Technology Companies 评估在线空间中算法的社会技术影响和从技术公司争取责任的必要性

2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC) Pub Date : 2022-12-01 DOI: 10.1109/CIC56439.2022.00017

C. Buntain

{"title":"An Imperative to Assess Socio-Technical Impact of Algorithms in Online Spaces and Wresting Responsibility from Technology Companies","authors":"C. Buntain","doi":"10.1109/CIC56439.2022.00017","DOIUrl":"https://doi.org/10.1109/CIC56439.2022.00017","url":null,"abstract":"Today’s information ecosystem has a plethora of tools to support decision-making and cognitive offloading. As this ecosystem has evolved and grown, however, we increasingly rely on recommendation systems, algorithmic curation, and other technologies to make decisions on our behalf and help manage an otherwise-overwhelming information environment. This cognitive offloading comes at a price, as these technologies make decisions that govern the information sources to which we are exposed, the celebrities we are likely follow, the content likely to become popular, the order of information shown to us, and even the state of our emotions. Despite the potential effects of this tradeoff, the true socio-technical impact of this reliance and imbuing these technologies with so much influence over the information space remains both an open and a controversial question. This paper outlines different forms of this question, wherein I summarize the controversies surrounding these technologies and the mixed evidence from academic research on them. I then describe the barriers impeding resolution of these concerns and how the current trajectories of online social platforms and the technology companies that own them are unlikely to provide answers to these issues without external intervention. I close by describing a \"socio-technical safety triad\" for online social spaces, where the responsibilities and incentives for reducing societal harm are devolved from corporations and spread across academic, governmental, and corporate stakeholders.","PeriodicalId":170721,"journal":{"name":"2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC)","volume":"104 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131629381","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

Next Generation Federated Learning for Edge Devices: An Overview 边缘设备的下一代联邦学习:概述

2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC) Pub Date : 2022-12-01 DOI: 10.1109/CIC56439.2022.00012

Jianyi Zhang, Zhixu Du, Jingwei Sun, Ang Li, Minxue Tang, Yuhao Wu, Zhihui Gao, Martin Kuo, Hai Helen Li, Yiran Chen

引用次数: 2

Leveraging Synergies Between AI and Networking to Build Next Generation Edge Networks 利用人工智能和网络之间的协同作用，构建下一代边缘网络

2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC) Pub Date : 2022-12-01 DOI: 10.1109/CIC56439.2022.00013

Sen Lin, Ming Shi, A. Arora, Raef Bassily, E. Bertino, C. Caramanis, K. Chowdhury, E. Ekici, A. Eryilmaz, Stratis Ioannidis, Nan Jiang, Gauri Joshi, J. Kurose, Yitao Liang, Zhiqiang Lin, Jia Liu, M. Liu, T. Melodia, Aryan Mokhtari, Rob Nowak, Sewoong Oh, S. Parthasarathy, Chunyi Peng, H. Seferoglu, N. Shroff, S. Shakkottai, K. Srinivasan, Ameet Talwalkar, A. Yener, Lei Ying

{"title":"Leveraging Synergies Between AI and Networking to Build Next Generation Edge Networks","authors":"Sen Lin, Ming Shi, A. Arora, Raef Bassily, E. Bertino, C. Caramanis, K. Chowdhury, E. Ekici, A. Eryilmaz, Stratis Ioannidis, Nan Jiang, Gauri Joshi, J. Kurose, Yitao Liang, Zhiqiang Lin, Jia Liu, M. Liu, T. Melodia, Aryan Mokhtari, Rob Nowak, Sewoong Oh, S. Parthasarathy, Chunyi Peng, H. Seferoglu, N. Shroff, S. Shakkottai, K. Srinivasan, Ameet Talwalkar, A. Yener, Lei Ying","doi":"10.1109/CIC56439.2022.00013","DOIUrl":"https://doi.org/10.1109/CIC56439.2022.00013","url":null,"abstract":"Networking and Artificial Intelligence (AI) are two of the most transformative information technologies over the last few decades. Building upon the synergies of these two powerful technologies, we envision designing next generation of edge networks to be highly efficient, reliable, robust and secure. To this end, in this paper, we delve into interesting and fundamental research challenges and opportunities that span two major broad and symbiotic areas: AI for Networks and Networks for AI. The former deals with the development of new AI tools and techniques that can enable the next generation AI-assisted networks; while the latter focuses on developing networking techniques and tools that will facilitate the vision of distributed intelligence, resulting in a virtuous research cycle where advances in one will help accelerate advances in the other. A wide range of applications will be further discussed to illustrate the importance of the foundational advances developed in these two areas.","PeriodicalId":170721,"journal":{"name":"2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC)","volume":"153 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116507249","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 Comparative Approach to Threshold Optimization for Classifying Imbalanced Data 不平衡数据分类阈值优化的比较方法

2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC) Pub Date : 2022-12-01 DOI: 10.1109/CIC56439.2022.00028

John T. Hancock, Justin M. Johnson, T. Khoshgoftaar

{"title":"A Comparative Approach to Threshold Optimization for Classifying Imbalanced Data","authors":"John T. Hancock, Justin M. Johnson, T. Khoshgoftaar","doi":"10.1109/CIC56439.2022.00028","DOIUrl":"https://doi.org/10.1109/CIC56439.2022.00028","url":null,"abstract":"For the practical application of a classifier, it is necessary to select an optimal output probability threshold to obtain the best classification results. There are many criteria one may employ to select a threshold. However, selecting a threshold will often involve trading off performance in terms of one metric for performance in terms of another metric. In our literature review of studies involving selecting thresholds to optimize classification of imbalanced data, we find there is an opportunity to expand on previous work for an in-depth study of threshold selection. Our contribution is to present a systematic method for selecting the best threshold value for a given classification task and its desired performance constraints. Just as a machine learning algorithm is optimized on some training data set, we demonstrate how a user-defined set of performance metrics can be utilized to optimize the classification threshold. In this study we use four popular metrics to optimize thresholds: precision, Matthews’ Correlation Coefficient, f-measure and geometric mean of true positive rate, and true negative rate. Moreover, we compare classification results for thresholds optimized for these metrics with the commonly used default threshold of 0.5, and the prior probability of the positive class (also known as the minority to majority class ratio). Our results show that other thresholds handily outperform the default threshold of 0.5. Moreover, we show that the positive class prior probability is a good benchmark for finding classification thresholds that perform well in terms of multiple metrics.","PeriodicalId":170721,"journal":{"name":"2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130221345","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: 2

Reproducible Workflows for Exploring and Modeling EMA Data 用于探索和建模EMA数据的可重复工作流程

2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC) Pub Date : 2022-12-01 DOI: 10.1109/CIC56439.2022.00026

Ching-Yun Yu, Yingzi Shang, T. Trull

引用次数: 0

The Social Emotional Web 社交情感网

2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC) Pub Date : 2022-12-01 DOI: 10.1109/CIC56439.2022.00014

Kristina Lerman

引用次数: 0

Network SheafModels for Social Information Dynamics 社会信息动态的网络模型

2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC) Pub Date : 2022-12-01 DOI: 10.1109/CIC56439.2022.00015

R. Ghrist

引用次数: 2

Semantic Communications: A Paradigm Whose Time Has Come 语义交际:一个时代已经到来的范式

2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC) Pub Date : 2022-12-01 DOI: 10.1109/CIC56439.2022.00020

Emrecan Kutay, Aylin Yener

{"title":"Semantic Communications: A Paradigm Whose Time Has Come","authors":"Emrecan Kutay, Aylin Yener","doi":"10.1109/CIC56439.2022.00020","DOIUrl":"https://doi.org/10.1109/CIC56439.2022.00020","url":null,"abstract":"Communication system design to date is predicated on principles that abstract information as digital sequences irrespective of their meanings. Such a semantic-agnostic approach leads to fundamental limits that are application and technology independent, and offers efficient engineering of communication systems. Emerging applications that involve communications however, call for going beyond the engineering problem of reliably reconstructing a digital sequence. Most current communication devices are computing devices that execute tasks. Increasingly communication is needed for a purpose and is integrated with decision making, machine learning and sensing. Further, human networks operate over machine networks, which necessitate more sophisticated human-machine communication. The goal of this vision paper is to advocate for Semantic Communications, i.e., communication system design that, at the outset, pays attention to the content, its meaning, context, and purpose. We will argue that taking into account the meanings and context of information can lead to better communication designs. In particular, we argue in favor of semantic distortion, a novel metric introduced nearly a decade ago, based upon which communications systems design aiming to convey the meaning and purpose can be designed. We review the current efforts of Semantic Communications which has recently become a popular area of 6G, and potential directions of Semantic Communications, which can explore various different directions with novel metrics including using Knowledge Graphs (KL), information theoretic approaches, and machine/Deep Learning (DL).","PeriodicalId":170721,"journal":{"name":"2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC)","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114617187","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

Attributes Aware Relationship-based Access Control for Smart IoT Systems 智能物联网系统基于关系的属性感知访问控制

2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC) Pub Date : 2022-12-01 DOI: 10.1109/CIC56439.2022.00021

Lopamudra Praharaj, Safwa Ameer, Maanak Gupta, R. Sandhu

{"title":"Attributes Aware Relationship-based Access Control for Smart IoT Systems","authors":"Lopamudra Praharaj, Safwa Ameer, Maanak Gupta, R. Sandhu","doi":"10.1109/CIC56439.2022.00021","DOIUrl":"https://doi.org/10.1109/CIC56439.2022.00021","url":null,"abstract":"The pervasive nature of smart connected devices has intruded on our daily lives and has become an intrinsic part of our world. However, the wide use of the Internet of Things (IoT) in critical application domains has raised concerns for user privacy and security against growing cyber threats. In particular, the implications of cyber exploitation for IoT devices are beyond financial losses and could constitute risks to human life. Most deployed access control solutions for smart IoT systems do not offer policy individualization, the ability to specify or change the policy according to the individual user’s preference. As a result, currently deployed systems are not well suited to specify access control policies in a multi-user environment, where users access the same devices to perform different operations. The system’s security gets tricky when the smart ecosystem involves complicated social relationships, much like in a smart home. Relationship-based access control (ReBAC), widely used in online social networks, offers the ability to consider user relationships in defining access control decisions and supports policy individualization. However, to the best of our knowledge, no such attempt has been made to develop a formal ReBAC model for smart IoT systems. This paper proposes a ReBACIoT dynamic and fine-grained access control model which considers the social relationships among users along with the attributes to support an attributes-aware relationship-based access control model for smart IoT systems. ReBACIoT is formally defined, illustrated through different use cases, implemented, and tested.","PeriodicalId":170721,"journal":{"name":"2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC)","volume":"266 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122720308","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