Transfer Learning for Power Outage Detection Task with Limited Training Data
This work addresses power outage detection for utility management, but it is incremental as it applies existing transfer learning methods to a specific domain with limited data.
This research tackled power outage detection using limited labeled data by applying transfer learning and language models, finding that few-shot fine-tuning significantly improves performance, with BERT achieving 81.3% accuracy (+15.3%) and GPT achieving 74.5% accuracy (+8.5%).
Early detection of power outages is crucial for maintaining a reliable power distribution system. This research investigates the use of transfer learning and language models in detecting outages with limited labeled data. By leveraging pretraining and transfer learning, models can generalize to unseen classes. Using a curated balanced dataset of social media tweets related to power outages, we conducted experiments using zero-shot and few-shot learning. Our hypothesis is that Language Models pretrained with limited data could achieve high performance in outage detection tasks over baseline models. Results show that while classical models outperform zero-shot Language Models, few-shot fine-tuning significantly improves their performance. For example, with 10% fine-tuning, BERT achieves 81.3% accuracy (+15.3%), and GPT achieves 74.5% accuracy (+8.5%). This has practical implications for analyzing and localizing outages in scenarios with limited data availability. Our evaluation provides insights into the potential of few-shot fine-tuning with Language Models for power outage detection, highlighting their strengths and limitations. This research contributes to the knowledge base of leveraging advanced natural language processing techniques for managing critical infrastructure.