We introduce W EATHER S EARCH , an integrated search system deployed at the Korea Meteorological Administration (KMA) 1 . W EATH - ER S EARCH enables users to retrieve all the relevant data for weather forecasting from a massive weather database with simple natu-ral language queries. We carefully design and conduct multiple expert surveys and interviews for template creation and apply data augmentation techniques including template filling to collect 4 million data points with minimal human labors. We then finetune mT5 (Xue et al., 2021) on the collected dataset and achieve an average MRR of 0.66 and an average Recall of 0.82. We also discuss weather-data-specific characteristics that should be taken into account for creating such a system. We hope our paper serves as a simple and effective guideline for those designing similar systems in other regions of the world.
Paper
Full text
An Integrated Search System for Korea Weather Data
Semantic Scholar · Environmental Science · 2023
Abstract
We introduce W EATHER S EARCH , an integrated search system deployed at the Korea Meteorological Administration (KMA) 1 . W EATH - ER S EARCH enables users to retrieve all the relevant data for weather forecasting from a massive weather database with simple natu-ral language queries. We carefully design and conduct multiple expert surveys and interviews for template creation and apply data augmentation techniques including template filling to collect 4 million data points with minimal human labors. We then finetune mT5 (Xue et al., 2021) on the collected dataset and achieve an average MRR of 0.66 and an average Recall of 0.82. We also discuss weather-data-specific characteristics that should be taken into account for creating such a system. We hope our paper serves as a simple and effective guideline for those designing similar systems in other regions of the world.