딥러닝 예측 모델을 통한 중고자동차 진단평가 방법 연구

Abstract

In the Korea Consumer Agency's 2021 consumer market evaluation index, which measures the level of consumer orientation that consumers feel in the product and service market, the used car market scored the lowest point in the evaluation score of all items except one item. In order to avoid the problem of this significantly low consumer orientation in the used car market, a market price prediction model based on a deep neural network architecture that infer the price of used cars based on the standard price table that complies with the articles of association and regulations of the Korea Diagnostic Guarantee Association approved by the Ministry of Land, Infrastructure and Transport was designed. From the perspective of consumers, it is possible to recognize prices with levels of accuracy and objectivity in the used car market in advance, similar to how a vehicle is evaluated by a professional person such as a state-certified automobile diagnostic evaluator through an estimator inferred from the predictive model. In this way, it was intended to suggest a plan to create a consumer environment in the used car market that is reliable by providing benefits tailored to the consumer's perspective and demand.