With the widespread use of the Internet of Things, data-driven services take the lead of both online and off-line businesses. Especially, personal data draw heavy attention of service providers because of the usefulness in value-added services. With the emerging big-data technology, a data broker appears, which exploits and sells personal data about individuals to other third parties. Due to little transparency between providers and brokers/consumers, people think that the current ecosystem is not trustworthy, and new regulations with strengthening the rights of individuals were introduced. Therefore, people have an interest in their privacy valuation. In this sense, the willingness-to-sell (WTS) of providers becomes one of the important aspects for data brokers; however, conventional studies have mainly focused on the willingnessto-buy (WTB) of consumers. Therefore, this paper proposes an optimized trading model for data brokers who buy personal data with proper incentives based on the WTS, and they sell valuable information from the refined dataset by considering the WTB and the dataset quality. This paper shows that the proposed model has a global optimal point by the convex optimization technique and proposes a gradient ascentbased algorithm. Consequently, it shows that the proposed model is feasible even if the data brokers spend costs to gather personal data.
With the widespread of Internet of Things (IoT) environment, a big data concept has emerged to handle a large number of data generated by IoT devices. Moreover, since datadriven approaches now become important for business, IoT data markets have emerged, and IoT big data are exploited by major stakeholders such as data brokers and data service providers. Since many services and applications utilize data analytic methods with collected data from IoT devices, the conflict issues between privacy and data exploitation are raised, and the markets are mainly categorized as privacy protection markets and privacy valuation markets, respectively. Since these kinds of data value chains (which are mainly considered by business stakeholders) are revealed, data providers are interested in proper incentives in exchange for their privacy (i.e., privacy valuation) under their agreement. Therefore, this paper proposes a competitive data trading model that consists of data providers who weigh the value between privacy protection and valuation as well as other business stakeholders. Each data broker considers the willingness-to-sell of data providers, and a single data service provider considers the willingness-to-pay of service consumers. At the same time, multiple data brokers compete to sell their dataset to the data service provider as a non-cooperative game model. Based on the Nash Equilibrium analysis (NE) of the game, the feasibility is shown that the proposed model has the unique NE that maximizes the profits of business stakeholders while satisfying all market participants.
The key challenge of unsupervised vehicle reidentification (Re-ID) is learning discriminative features from unlabelled vehicle images. Numerous methods using domain adaptation have achieved outstanding performance, but those methods still need a labelled dataset as a source domain. This paper addresses an unsupervised vehicle Re-ID method, which no need any types of a labelled dataset, through a Self-supervised Metric Learning (SSML) based on a feature dictionary. Our method initially extracts features from vehicle images and stores them in a dictionary. Thereafter, based on the dictionary, the proposed method conducts dictionary-based positive label mining (DPLM) to search for positive labels. Pairwise similarity, relative-rank consistency, and adjacent feature distribution similarity are jointly considered to find images that may belong to the same vehicle of a given probe image. The results of DPLM are applied to dictionary-based triplet loss (DTL) to improve the discriminativeness of learnt features and to refine the quality of the results of DPLM progressively. The iterative process with DPLM and DTL boosts the performance of unsupervised vehicle Re-ID. Experimental results demonstrate the effectiveness of the proposed method by producing promising vehicle Re-ID performance without a pre-labelled dataset. The source code for this paper is publicly available on https://github.com/andreYoo/VeRI_SSML_FD.git.
In this paper, we propose Normality-Calibrated Autoencoder (NCAE), which can boost anomaly detection performance on the contaminated datasets without any prior information or explicit abnormal samples in the training phase. The NCAE adversarially generates high confident normal samples from a latent space having low entropy and leverages them to predict abnormal samples in a training dataset. NCAE is trained to minimise reconstruction errors in uncontaminated samples and maximise reconstruction errors in contaminated samples. The experimental results demonstrate that our method outperforms shallow, hybrid, and deep methods for unsupervised anomaly detection and achieves comparable performance compared with semi-supervised methods using labelled anomaly samples in the training phase. The source code is publicly available on https://github.com/andreYoo/ NCAE_UAD.git.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.