This page was machine-translated and may differ from the original. View original

Creating Customized Weather Observation Stations with AI

Google 우선 소스Published2020.02.26 08:38
KIMM Hosts Competition for Temperature Estimation in Specific Regions
Enhancing Value of Public Data Through AI Algorithm Application


The Korea Institute of Machinery and Materials is launching an AI algorithm development competition starting March 1st in collaboration with the Korea Atomic Energy Research Institute, Dacon—a data science platform company, and AI Friends, an artificial intelligence researcher community.

▲ A competition to estimate temperatures in specific regions using public data is being held

This competition to estimate temperatures in specific regions using public data is being held as part of the next-generation public technology discovery network project supported by the Research & Development Special Zone Promotion Foundation.

Participants develop AI analysis algorithms by utilizing public data provided by the Korea Meteorological Administration and temperature data from specific regions provided by KIMM. Dacon analyzes the performance of these algorithms to determine rankings.

The competition is conducted online and is open to participants regardless of location. Results will be announced through the Dacon website on April 27th, with total prize money of 2 million won awarded: 1 million won for first place, 700,000 won for second place, and 300,000 won for third place.

Baek Dong-cheon, senior researcher at KIMM, stated, "Individual facility temperatures often differ significantly from those measured at meteorological observation stations, so more accurate information is necessary to provide practical benefit in daily life. We hope that through this competition, customized temperature estimation technology utilizing Korea Meteorological Administration public data will advance and contribute to providing customized temperature history records."

Meanwhile, KIMM's Reliability Assessment Research Laboratory is providing indoor and outdoor temperature data from 19 locations measured over three months in the Daejeon region, and the research team has been developing outdoor structure temperature estimation methods utilizing Korea Meteorological Administration public data as part of its own research projects since May of last year.
To request a correction, reply or follow-up report on this article, see how to file a request. Previously published statements are collected in corrections & replies.
최인영 Reporter