Loading...
Send to Friend
 Bookmark
 Review List
  1. Data-driven facility location balancing using machine learning and shannon entropy
Data-driven facility location balancing using machine learning and shannon entropy
195 viewed

Data-driven facility location balancing using machine learning and shannon entropy

Dastgheib, M. A

Data-driven facility location balancing using machine learning and shannon entropy

Dastgheib, M. A ; Sharif University of Technology | 2024

195 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/MECOM61498.2024.10881283
  3. Publisher: IEEE , 2024
  4. Abstract:
  5. This paper presents a novel approach to data-driven facility location balancing that leverages the power of machine learning (ML) and Shannon entropy. We introduce a ML algorithm that, for the first time, employs nearest neighbor methods for stochastic gradient estimation to learn continuous-space facility locations from data. This innovative approach achieves a speedup of two orders of magnitude over traditional methods. Evaluated through a proposed adjusted entropy ratio metric, our method improves load balancing by 62% compared to k -means, particularly in densely populated areas where k- means falls short. Shannon Entropy ensures balanced coverage in facility location tasks, offering significant benefits in scenarios requiring more equitable subdivision of crowded regions. © 2024 IEEE
  6. Keywords:
  7. Facility location ; Machine learning ; Nearest neighbor methods ; Shannon entropy ; Shape optimization ; Stochastic gradients ; Stochastic optimization ; Federated learning
  8. Source: 2024 IEEE Middle East Conference on Communications and Networking, MECOM 2024 ; 2024 , Pages 41-46 ; 979-835037671-5 (ISBN)
  9. URL: https://ieeexplore.ieee.org/document/10881283