Research Article
Pattern-Preserved Normalization Enabled User Profiling
@INPROCEEDINGS{10.1007/978-3-031-31733-0_28, author={Fengchao Chen and Lide Zhou and Junni Su and Xin Zhang}, title={Pattern-Preserved Normalization Enabled User Profiling}, proceedings={Smart Grid and Innovative Frontiers in Telecommunications. 7th EAI International Conference, SmartGIFT 2022, Changsha, China, December 10-12, 2022, Proceedings}, proceedings_a={SMARTGIFT}, year={2023}, month={5}, keywords={Clustering data-driven Consumer Analysis}, doi={10.1007/978-3-031-31733-0_28} }- Fengchao Chen
Lide Zhou
Junni Su
Xin Zhang
Year: 2023
Pattern-Preserved Normalization Enabled User Profiling
SMARTGIFT
Springer
DOI: 10.1007/978-3-031-31733-0_28
Abstract
The legacy power grid is evolving into a more intelligent grid, and the classical preventive control paradigm is also evolving into a more modern data-driven control paradigm. However, the massive data also poses challenges on the data-driven techniques. In this paper, we focus on the clustering problem in the residential energy sector based on long-term energy consumption data. We employ the classical k-means clustering algorithm and analyze the drawbacks of Min-Max normalization and the disadvantages of utilizing Euclidean distance. We further provide a potential solution, PP-normalization, to solve these issues to achieve a better performance in residential consumption data clustering.
Copyright © 2022–2025 ICST
