Authors: Qian Xiao, Dan Liu, Kevin Credit

Published on: April 14, 2024

Impact Score: 8.2

Arxiv code: Arxiv:2404.12399

Summary

• What is new: Introduction of CLEAR, a self-supervised contrastive learning approach to identify inconsistencies in BER assessments.
• Why this is important: The BER assessment process is prone to errors due to missing or inaccurate measurements.
• What the research proposes: CLEAR leverages data-driven methods to pinpoint and assess the inconsistencies in BER assessments, offering a more reliable evaluation.
• Results: CLEAR successfully identified evidence of inconsistent BER assessments within the real-world dataset of Irish building stocks, indicating measurement data corruption.

Technical Details

Technological frameworks used: Self-supervised contrastive learning

Models used: nan

Data used: Dataset representing Irish building stocks

Potential Impact

Real estate, construction, urban planning, and policy-making sectors could benefit from more accurate BER assessments, leading to enhanced energy efficiency and climate improvement initiatives.

Want to implement this idea in a business?

We have generated a startup concept here: EcoScore Analytics.