Author name cluster

Zuyuan Yang

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

2 papers

1 author row

EAAI Journal 2026 Journal Article

Hierarchical metering data imputation with multi-view learning for accurate electricity consumption prediction

Zitan Xie
Zuyuan Yang
Weifeng Zhong
Shengli Xie

Accurate electricity consumption prediction enables energy providers to optimize resource allocation, enhance grid stability, and meet fluctuating demands efficiently, with applications spanning across smart grids, energy management systems, and electricity markets. However, this process relies on large amounts of complete meter data, which may be missing or corrupted for various reasons during data collection. This study proposes a novel multi-view learning-based imputation model to deal with missing data in a hierarchical metering system. In this study, the hierarchical system is analyzed using two views to constitute a multi-view dataset, where the master-node view reflects the overall electricity consumption and the sub-node view provides detailed electricity consumption information. Unlike traditional single-view methods, the master-node view is developed to generate complementarity information for effective imputation of sub-nodes. Moreover, a feature alignment-based cross-view mapping is proposed to exploit the alignment relationship between views, and the adaptive multi-view graph learning is introduced to capture the distribution structure of the electricity consumption at different time points. Experimental results on real datasets from a ceramic factory show that our proposed method outperforms the other compared methods. In the prediction after imputation, when the missing rate is 50%, 70%, and 90%, the average performance improvement of our proposed method is about 10. 08% in Root Mean Square Error (RMSE) and 11. 9% in Mean Absolute Error (MAE) across all datasets.

Details DOI

AAAI Conference 2025 Conference Paper

FedMSGL: A Self-Expressive Hypergraph Based Federated Multi-View Learning

Daoyuan Li
Zuyuan Yang
Shengli Xie

Federated learning is essential for enabling collaborative model training across decentralized data sources while preserving data privacy and security. This approach mitigates the risks associated with centralized data collection and addresses concerns related to data ownership and compliance. Despite significant advancements in federated learning algorithms that address communication bottlenecks and enhance privacy protection, existing works overlook the impact of differences in data feature dimensions, resulting in global models that disproportionately depend on participants with large feature dimensions. Additionally, current single-view federated learning methods fail to account for the unique characteristics of multi-view data, leading to suboptimal performance in processing such data. To address these issues, we propose a Self-expressive Hypergraph Based Federated Multi-view Learning method (FedMSGL). The proposed method leverages self-expressive character in the local training to learn uniform dimension subspace with latent sample relation. At the central side, an adaptive fusion technique is employed to generate the global model, while constructing a hypergraph from the learned global and view-specific subspace to capture intricate interconnections across views. Experiments on multi-view datasets with different feature dimensions validated the effectiveness of the proposed method.

PDF Details DOI