Jiji Auction Introduces AI to Auction Market "Selected for National Project on Winning Bid Price Prediction"

[Asia Economy Reporter Onyu Lim] Technology that can provide benefits not only to investors but also to stakeholders such as financial institutions and tenants through artificial intelligence (AI)-based auctions is being actively developed.

Gigi Auction, a specialized auction company, announced on the 1st that it was recently selected as the lead organization for the Ministry of Science and ICT’s national project, the ‘2020 Non-face-to-face Business Digital Innovation Technology Development Project.’ Following this selection, Gigi Auction plans to develop a ‘Deep Learning-based Auction Fair Price Prediction System’ over the next year in collaboration with Yonsei University’s Financial Engineering Laboratory.

Yonsei University’s Financial Engineering Laboratory has been researching AI in finance for over 20 years and is currently focusing on deep learning. They have published over 70 papers related to finance and artificial intelligence, including more than 40 in SCI-level international journals, and have filed two domestic patents. The total funding amounts to 350 million KRW.

Currently, although there are many proptech companies both in Korea and abroad, most focus on brokerage or rental platforms, making prediction services rare. Among these, real estate price prediction is a field that requires vast data and highly accurate analytical technology.

Gigi Auction has already built big data with statistical consistency using over 4 million auction and public sale real estate records accumulated over 37 years. With this project, applying deep learning to 200,000 cases of in-depth analysis data on special rights will allow further qualitative advancement in building prediction models. A Gigi Auction official stated, "The in-depth analysis data on special rights such as retention rights reflect the most important factors individually affecting the winning bid price of each property, which can significantly enhance prediction accuracy."

By utilizing the winning bid price prediction system that Gigi Auction plans to develop, it will be easy to predict the timing of winning bids and auction yields. Financial institutions can use the predictions as a basis to determine loan amounts and interest rates for individual real estate properties. In particular, it can replace the appraisal process, which is mandatory during loans and auction proceedings, acting as a catalyst to improve the current high-cost appraisal structure. It also enables safe and reasonable deposit negotiations in lease contracts.

Professor Kyungjoo Oh of Yonsei University’s Financial Engineering Laboratory said, “Statistical methods alone have limitations in accurately predicting real estate values,” and added, “By integrating various AI technologies developed through this project, such as auction index calculation systems, natural language-based news analysis, and big data visualization, accuracy can be improved.”

Kang Eun, director and project leader at Gigi Auction, said, “In the future, providing simple information about individual properties will no longer suffice; future outlooks on the auction market will be essential, making AI indispensable in the auction industry.” He added, “Winning bid price prediction clarifies the basis for loan decisions and can contribute to the government’s efforts to stabilize national housing life, making it socially and economically significant.”

This content was produced with the assistance of AI translation services.