Domain-Informed Label Fusion Surpasses LLMs in Free-Living Activity Classification (Student Abstract)

Shovito Barua Soumma, Abdullah Mamun, Hassan Ghasemzadeh

AAAI 2025 pp. 29495-29497

doi:10.1609/AAAI.V39I28.35301 /aaai/2025/soumma2025aaai-domain/

Abstract

FuSE-MET addresses critical challenges in deploying human activity recognition (HAR) systems in uncontrolled environments by effectively managing noisy labels, sparse data, and undefined activity vocabularies. By integrating BERT-based word embeddings with domain-specific knowledge (i.e., MET values), FuSE-MET optimizes label merging, reducing label complexity and improving classification accuracy. Our approach outperforms the state-of-the-art techniques, including ChatGPT-4, by balancing semantic meaning and physical intensity.

AAAI Semantic Scholar

Cite

Text

Soumma et al. "Domain-Informed Label Fusion Surpasses LLMs in Free-Living Activity Classification (Student Abstract)." AAAI Conference on Artificial Intelligence, 2025. doi:10.1609/AAAI.V39I28.35301

Markdown

[Soumma et al. "Domain-Informed Label Fusion Surpasses LLMs in Free-Living Activity Classification (Student Abstract)." AAAI Conference on Artificial Intelligence, 2025.](https://mlanthology.org/aaai/2025/soumma2025aaai-domain/) doi:10.1609/AAAI.V39I28.35301

BibTeX

@inproceedings{soumma2025aaai-domain,
  title     = {{Domain-Informed Label Fusion Surpasses LLMs in Free-Living Activity Classification (Student Abstract)}},
  author    = {Soumma, Shovito Barua and Mamun, Abdullah and Ghasemzadeh, Hassan},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2025},
  pages     = {29495-29497},
  doi       = {10.1609/AAAI.V39I28.35301},
  url       = {https://mlanthology.org/aaai/2025/soumma2025aaai-domain/}
}