Word2Vec vehicle price prediction method using Word2Vec

• Main Authors: JEON MIN SOO, IM JOON, CHOI JOSEPH SUNGPIL, LEE HYUN BOK, CHOI YUN SIK
• Format: Patent
• Language: English; Korean
• Published: 30.09.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The present invention relates to a vehicle price prediction method using Word2Vec. The vehicle price prediction method comprises: a data collection step of extracting and collecting one or more pieces of element information from input vehicle image information for prediction of the price of a vehicle; a data processing step of generating pieces of word to vectors (Word2Vec) corresponding to the respective elements based on the collected pieces of element information; and a step of predicting the price of the vehicle at the point of purchase desired by a user by inputting the generated Word2Vec to an artificial intelligence model by using the generated Word2Vec. Therefore, the vehicle price prediction method can numerically handle the pieces of element information, thereby predicting the price of the vehicle through vector operation on the pieces of element information. 본 발명은 워드투벡터(Word2Vec)를 이용한 차량의 가격예측 방법에 관한 것이다. 본 발명의 방법은 차량의 가격예측을 위한, 입력된 차량 이미지정보로부터 적어도 하나 이상의 엘리먼트 정보들을 추출하여 수집하는 데이터 수집 단계, 수집된 엘리먼트 정보들에 기초하여, 각 엘리먼트들에 대응되는 워드투벡터(Word2Vec)들을 생성하는 데이터 가공 단계 및 생성된 워드투벡터(Word2Vec)들을 이용하여, 생성된 워드투벡터(Word2Vec)들을 인공지능 모델에 입력함으로써 사용자가 원하는 구매시점의 해당 차량의 가격을 예측할 수 있다.