In this article, we aim to design an architecture for privacy-preserved credit data and model sharing to guarantee the secure storage and sharing of credit information in a distributed environment. The proposed architecture optimizes the data privacy by sharing the data model instead of revealing the actual data. This article also proposes an efficient credit data storage mechanism combined with a deletable Bloom filter to guarantee a uniform consensus for the training and computation process. In addition, we propose authority control contract and credit verification contract for the secure certification of credit sharing model results under federated learning. Extensive experimental results and security analysis demonstrate that our proposed credit model sharing system based on federated learning and blockchain is of high accuracy, efficiency, as well as stability. In particular, the findings of this article could alleviate the potential credit crisis under financial pressure that assist to economic recovery after the global COVID-19 pandemic. Our approach has further boosted up the demand for efficient, secure credit models for Industry 4.0.
3 October 2024 in IEEE Transactions on Engineering Management
Digital platforms are experiencing a growing presence of generative artificial intelligence (AI) content, raising concerns due to the prevalence of misinformation that disrupts market integrity. Consequently, the development of effective regulatory measures for overseeing generative AI content becomes imperative. This necessitates the establishment of mechanisms to detect and filter out inaccuracies, ensuring compliance with regulatory requirements. In addition, collaboration among experts, regulators, and AI developers is essential to encourage responsible AI deployment on digital platforms. Successful governance hinges on principles of transparency, accountability, and proactive risk management to navigate the evolving generative AI on digital platforms. Therefore, in order to address the security issues currently faced by Artificial Intelligence Generated Content (AIGC), this article first proposes a method of efficient cache mechanism for AIGC content. The secure method of determining the identity of AIGC content owners is proposed based on blockchain technology. Subsequently, it suggests mechanisms for access control and data encryption for generated content within a blockchain environment. Finally, it presents an efficient data supervision mechanism tailored to the AIGC environment. The methods outlined in this article aim to enhance security from three perspectives: protection of content creators' identities, safeguarding data security, and ensuring effective data supervision within the AIGC framework. The experimental results further confirm that our proposed method not only ensures the security of the AIGC framework but also provides an efficient data analysis and supervision solution for digital platforms.
Carbon emissions trading has become an increasingly hot topic nowadays, due to the fact that how to reduce carbon emissions has been a common effort of different countries. However, traditional methods are plagued by issues, such as inadequate privacy protection mechanisms and the challenge of representing data assets in a comprehensive form using blockchain data models. In this article, we propose carbon emissions trading scheme (CETS), a secure carbon emissions trading system using blockchain combined with digital assets transactions. The proposed CETS scheme enhances the performance of models for carbon emissions trading by prioritizing the efficiency, privacy, and traceability of carbon emissions trading. Simultaneously, it improves the consistency of digital asset trading throughout the chain. First, we propose a dual-blockchain-based method for storing and tracing carbon emission data, which ensures the privacy of the data. Next, we propose algorithms for transaction of digital assets in carbon emission trading scheme, which include digital asset uniqueness algorithm, serializable mechanism, and cross-chain algorithm of digital assets. Finally, we propose an automated machine learning pipeline approach based on the carbon trading price forecasting model construction method, which can provide efficient, automatic price forecasting model construction and training. The experimental results prove that our proposed carbon emission trading system can provide an efficient and stable carbon emission trading solution.
4 March 2025 in IEEE Transactions on Engineering Management
In response to the pressing imperative of combating climate change on a global scale, a new era of carbon neutrality is steadily emerging. Achieving carbon neutrality is critical, and in the digital economy, technology-driven business models are essential for reducing carbon emissions through effective carbon emission trading systems. However, current research on carbon emission trading suffers from inadequate privacy protection, low efficiency in data sharing and model construction, as well as insufficient capabilities in automated and autonomous model building. Therefore, this study focuses on utilizing blockchain and automated machine learning for data sharing and modeling to enhance carbon neutrality. First, we design the architecture of the system and the mechanism for storing data on the blockchain. We then devise methods for storing and trading carbon emission transactions on the blockchain and construct the process for issuing carbon credits. Additionally, our proposed method incorporates Neural Architecture Search (NAS) to develop a carbon trading price forecasting model. By leveraging data augmentation for carbon emission price time series and utilizing triplet loss for model training, we enhance the reliability and security of carbon trading investment through accurate price forecasting. The experimental results further demonstrate the robust performance and precision of our carbon emission price forecasting module. Consequently, our approach provides efficient carbon emission trading services to businesses and individuals, offering a robust solution for global carbon emission reduction and the achievement of carbon neutrality.
Timely and accurate depth estimation of a shallow waterway can improve shipping efficiency and reduce the danger of waterway transport accidents. However, waterway depth data measured during actual maritime navigation is limited, and the depth values can have large variability. Big data collected in real time by automatic identification systems (AIS) might provide a way to estimate accurate waterway depths, although these data include no direct channel depth information. We suggest a deep neural network (DNN) based model, called DDTree, for using the real-time AIS data and the data from Global Mapper to predict waterway depth for ships in an accurate and timely way. The model combines a decision tree and DNN, which is trained and tested on the AIS and Global Mapper data from the Nantong and Fangcheng ports on the southeastern and southwestern coast of China. The actual waterway depth data were used together with the AIS data as the input to DDTree. The latest data on waterway depths from the Chinese maritime agency were used to verify the results. The experiments show that the DDTree model has a prediction accuracy of 91.15%. Therefore, the DDTree model can provide an accurate prediction of waterway depth and compensate for the shortage of waterway depth monitoring means. The proposed hybrid DDTree model could improve marine situational awareness, navigation safety, and shipping efficiency, and contribute to smart navigation.