Proceedings. IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology最新文献

{"title":"Session details: Chapter 6: IDP: 10th International Workshop on Intelligent Data Processing","authors":"","doi":"10.1145/3530279","DOIUrl":"https://doi.org/10.1145/3530279","url":null,"abstract":"","PeriodicalId":89230,"journal":{"name":"Proceedings. IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology","volume":"22 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90067793","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}

{"title":"On the Impact of Dataset Size:A Twitter Classification Case Study","authors":"Thi-Huyen Nguyen, Hoang H. Nguyen, Zahra Ahmadi, Tuan-Anh Hoang, Thanh-Nam Doan","doi":"10.1145/3486622.3493960","DOIUrl":"https://doi.org/10.1145/3486622.3493960","url":null,"abstract":"The recent advent and evolution of deep learning models and pre-trained embedding techniques have created a breakthrough in supervised learning. Typically, we expect that adding more labeled data improves the predictive performance of supervised models. On the other hand, collecting more labeled data is not an easy task due to several difficulties, such as manual labor costs, data privacy, and computational constraint. Hence, a comprehensive study on the relation between training set size and the classification performance of different methods could be essentially useful in the selection of a learning model for a specific task. However, the literature lacks such a thorough and systematic study. In this paper, we concentrate on this relationship in the context of short, noisy texts from Twitter. We design a systematic mechanism to comprehensively observe the performance improvement of supervised learning models with the increase of data sizes on three well-known Twitter tasks: sentiment analysis, informativeness detection, and information relevance. Besides, we study how significantly better the recent deep learning models are compared to traditional machine learning approaches in the case of various data sizes. Our extensive experiments show (a) recent pre-trained models have overcome big data requirements, (b) a good choice of text representation has more impact than adding more data, and (c) adding more data is not always beneficial in supervised learning.","PeriodicalId":89230,"journal":{"name":"Proceedings. IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology","volume":"30 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86590371","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}

{"title":"Bias detection in Wikipedia articles. A study on Polish and English Datasets.","authors":"Weronika Stankiewicz, Katarzyna Baraniak, M. Sydow","doi":"10.1145/3486622.3494007","DOIUrl":"https://doi.org/10.1145/3486622.3494007","url":null,"abstract":"Nowadays, an almost unlimited number of information sources and polarized media hamper our ability to distinguish between biased and neutral speech. It creates a need for an automatic tool that could comprehend human language and assess whether presented information is conveyed without any editorial bias. In this work, we introduce models able to detect bias on a sentence level. As other authors before, we continue to utilize Wikipedia revision history to collect examples of biased and unbiased speech. For the first time, this work introduces a corpus of labeled subjective and neutral sentences in the Polish language. Furthermore, we compare the performance of LSTM and BERT models trained on English and Polish sentences. As part of the findings, we present cues for subjectivity that were detected during the analysis. We also present a new dataset WNC-pl a Polish corpus of biased and unbiased sentences.","PeriodicalId":89230,"journal":{"name":"Proceedings. IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology","volume":"25 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79118738","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}

{"title":"Event Causal Relationship Retrieval","authors":"Yasunobu Sumikawa","doi":"10.1145/3486622.3493936","DOIUrl":"https://doi.org/10.1145/3486622.3493936","url":null,"abstract":"Analyzing history has numerous benefits, including understanding what the people in the past did for events and what results they obtained and using historical knowledge to the present. Several past studies have analyzed historical events based on the assumption that each event is described in texts. Most of them analyze how similar the words and their categories used in the descriptions are instead of taking care of event-causal relationships. In this study, we propose an algorithm named the Event Causality relationship similarity Measurement (ECM) to measure the similarity between event-causal relationships. The ECM solves a maximum weight matching problem on a bipartite graph, where the weights are the similarities between the event-causal relationships. We evaluated ECM with previous related works and confirmed that the ECM is the best.","PeriodicalId":89230,"journal":{"name":"Proceedings. IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology","volume":"41 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79156058","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}

{"title":"Why is it so hard to make self-driving cars? (Trustworthy autonomous systems)","authors":"J. Sifakis","doi":"10.1145/3486622.0000002","DOIUrl":"https://doi.org/10.1145/3486622.0000002","url":null,"abstract":"Why is self-driving so hard? Despite the enthusiastic involvement of big technological companies and the massive investment of many billions of dollars, all the optimistic predictions about self-driving cars ''being around the corner'' went utterly wrong. I argue that these difficulties emblematically illustrate the challenges raised by the vision for trustworthy autonomous systems. These are critical systems intended to replace human operators in complex organizations, very different from other intelligent systems such as game-playing robots or intelligent personal assistants. I discuss complexity limitations inherent to autonomic behavior but also to integration in complex cyber-physical and human environments. I argue that existing critical systems engineering techniques fall short of meeting the complexity challenge. I also argue that emerging end-to-end AI-enabled solutions currently developed by industry, fail to provide the required strong trustworthiness guarantees. I advocate a hybrid design approach combining model-based and data-based techniques and seeking tradeoffs between performance and trustworthiness. I also discuss the validation problem emphasizing the need for rigorous simulation and testing techniques allowing technically sound safety evaluation. I conclude that building trustworthy autonomous systems goes far beyond the current AI vision. To reach this vision, we need a new scientific foundation enriching and extending traditional systems engineering with data-based techniques.","PeriodicalId":89230,"journal":{"name":"Proceedings. IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology","volume":"47 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79244742","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}

{"title":"X-ray Image Blind Denoising in Hybrid Noise Based on Convolutional Neural Networks","authors":"Jie Wang, Huaiwei Cong, Xin Wei, Baolian Qi, Jinpeng Li, Ting Cai","doi":"10.1145/3498851.3498952","DOIUrl":"https://doi.org/10.1145/3498851.3498952","url":null,"abstract":"Low-dose X-ray imaging is a medical imaging method used for disease screening and diagnosis. However, the interpretation of such images is a challenging task because of machine noise. Although some deep learning-based denoising algorithms have made considerable progress, they do not perform well on real X-ray images. Because the actual noise of the X-ray image is more complicated. In this paper, we design a noise model according to the physical principle of X-ray imaging, which is used to simulate the real X-ray image. On this basis, we propose a blind denoising convolutional neural network (X-BDCNN) for low-dose X-ray image enhancement. X-BDCNN consists of two networks. One is used to estimate the noise level of the input noise X-ray image. The other is used to obtain the residual noise image by taking the noisy X-ray image and the estimated noise level as input. The final denoised X-ray image is obtained by subtracting the residual noise image from the input noise X-ray image. In addition, we add a structural similarity (SSIM) loss function to X-BDCNN to maintain the structural information. The experimental results show that the denoising performance of X-BDCNN is better than the existing denoising methods. Code is available online.","PeriodicalId":89230,"journal":{"name":"Proceedings. IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology","volume":"36 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91307003","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}

{"title":"Proof of Importance: A Consensus Algorithm for Importance Based on Dynamic Authorization","authors":"Bingbing Xiao, Chenguang Jin, Zheng Li, Bingnan Zhu, Xiaoruo Li, D. Wang","doi":"10.1145/3498851.3499007","DOIUrl":"https://doi.org/10.1145/3498851.3499007","url":null,"abstract":"Proof of stake consensus algorithm (PoS) has the advantage of not requiring arithmetic power, however, the higher the stake, the more likely the node will get the right to account. As a result, the accounting nodes are highly deterministic and the rich get richer, once the node with the highest equity fails to account for the block properly. The rest of the nodes have to compete again for accounting rights, and the probability of system stagnation will increase sharply. To address these two shortcomings, a consensus algorithm based on dynamic delegation of proof of importance (DPoI) is proposed.The algorithm introduces an importance assessment scheme that calculates the importance ivlaue (iValue) of nodes in each round based on nodes aValue, iTrade, Ltime and Credit.Finally, the Fibonacci series is used to divide nodes with similar iValue into a group, and the group draws on the DPoS voting strategy ranking to act as alternative nodes, thus forming a disaster recovery scheme to effectively avoid system stagnation, which effectively enhances the security and stability of the blockchain system. The experimental results show that the speed of DPoI block out is about 6 times faster than PoI, which significantly improves the block out speed.","PeriodicalId":89230,"journal":{"name":"Proceedings. IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology","volume":"54 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80751419","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}

{"title":"Fake News Detection via Biased User Profiles in Social Networking Sites","authors":"Ryoya Furukawa, Daiki Ito, Yuta Takata, Hiroshi Kumagai, Masaki Kamizono, Yoshiaki Shiraishi, M. Morii","doi":"10.1145/3486622.3493939","DOIUrl":"https://doi.org/10.1145/3486622.3493939","url":null,"abstract":"The spread of fake news on social networking sites has become a problem. Users who share fake news have strong human needs (such as the desire for approval, belonging, and self-expression) and are likely to have characteristic words in their self-descriptions. In this paper, we propose a method for detecting fake news based on the biased words in those self-descriptions. In the proposed method, feature vectors are first created from words in the self-descriptions of multiple users who post the same news URL on Twitter. Subsequently, they are classified into fake or not fake using machine learning. In experiments conducted using multiple datasets, including real and fake news from Japan and the U.S., the proposed method achieved an average classification accuracy of 90.2%. Furthermore, we show that the proposed method is effective for multi-domain fake news detection and analysis of users targeted by fake news in case studies.","PeriodicalId":89230,"journal":{"name":"Proceedings. IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology","volume":"27 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79721480","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}

{"title":"Redistribution in Public Project Problems via Neural Networks","authors":"Guanhua Wang, Wuli Zuo, M. Guo","doi":"10.1145/3486622.3493922","DOIUrl":"https://doi.org/10.1145/3486622.3493922","url":null,"abstract":"Many important problems in multiagent systems involve resource allocations. Self-interested agents may lie about their valuations if doing so increases their own utilities. Therefore, it is necessary to design mechanisms (collective decision-making rules) with desired properties and objectives. The VCG redistribution mechanisms are efficient (the agents who value the resources the most will be allocated), strategy-proof (the agents have no incentives to lie about their valuations), and weakly budget-balanced (no deficits). We focus on the VCG redistribution mechanisms for the classic public project problem, where a group of agents needs to decide whether or not to build a non-excludable public project. We design mechanisms via neural networks with two welfare-maximizing objectives: optimal in the worst case and optimal in expectation. Previous studies showed two worst-case optimal mechanisms for 3 agents, but worst-case optimal mechanisms have not been identified for more than 3 agents. For maximizing expected welfare, there are no existing results. We use neural networks to design VCG redistribution mechanisms. Neural networks have been used to design the redistribution mechanisms for multi-unit auctions with unit demand. We show that for the public project problem, the previously proposed neural networks, which led to optimal/near-optimal mechanisms for multi-unit auctions with unit demand, perform abysmally for the public project problem. We significantly improve the existing networks on multiple fronts: We conduct a GAN network to generate worst-case type profiles and feed prior distribution into loss function to provide quality gradients for the optimal-in-expectation objective. We adopt dimension reduction to handle a larger number of agents and we adopt supervised learning into the best manual mechanism as initialization, then leave it into unsupervised learning. For the worst case, we get better results than the existing manual mechanisms, and for the optimal-in-expectation objective, our mechanisms’ performances are close to the theoretical optimal performance.","PeriodicalId":89230,"journal":{"name":"Proceedings. IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology","volume":"30 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85026272","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}

{"title":"A Subgraph Isomorphism-based Attack Towards Social Networks","authors":"Mengjiao Guo, Chi-Hung Chi, Hui Zheng, Jing He, Xiaoting Zhang","doi":"10.1145/3498851.3499024","DOIUrl":"https://doi.org/10.1145/3498851.3499024","url":null,"abstract":"It has been widely recognized that social network analysis of group relationships and behaviors come to thrive, so publicity social networks have gained growing attention from third-party individuals for academic researchers and advertisers. Anonymous versions are generally obtained from the naive anonymization mechanism through identity transformation to fend off attacks on socially sensitive information. The adversaries intend to implement person re-identification in anonymous data, and they generally possess a subset of social interaction information of the target user. In this way, a privacy breach could be achieved by exploiting the neighbourhood of the object's known structural information. Say, if one node's information is breached, other nodes’ private information will be compromised according to the detected structural information. Therefore, all the mentioned above are equivalent to the subgraph isomorphism problem to identify who is who in the social networks. Existing enumeration and indexing-related subgraph isomorphism methods cannot process matching problems with both large target and query graphs. Therefore, subgraph querying is a knotty problem pressing for a solution. In this work, we elaborate on the subgraph of structural attack. Our subgraph isomorphism-based method adopts a 3-stage framework for learning and refining structural correspondences over a large graph. First, we generate a set of candidate matches and compare the query graph with these candidate graphs over the corresponding number of vertex and edge, which can noticeably reduce the number of candidate graphs. Secondly, we employ the permutation theorem to evaluate the row sum of vertex and edge adjacency matrix of query graph and candidate graph. Lastly, our proposed scheme deploys the well-found equinumerosity theorem to verify if the query graph and candidate graph satisfy the isomorphic relationship. Solid evaluation criteria on time complexity verify the proposed attack strategy.","PeriodicalId":89230,"journal":{"name":"Proceedings. IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology","volume":"19 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87701349","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}