Model AI - Car Recognition & Price Suggestion
Technology stacks
Development Team
Development of the Recognition AI Model: 1 AI engineer, 1 Data Crawler
Development of the Price Suggestion AI Model: 2 AI engineer, 1 Data Crawler
Duration
Development of the Recognition AI Model: 2 months
Development of the Price Suggestion AI Model: 3 months
CHALLENGE
The customer's website suggests the price of used cars based on the information provided by the users.
However, the huge source of information takes such a long time to find out and choose. An inefficient on-web experience leads to increasing bounce rates.
HBLAB's Solutions
Optimize users experience with Solutions that automatically import information to the web using just a photo. Users only need to upload photos to the web, most of the information will be automatically analyzed and filled in the system.
Model AI automatically fills in information of 250 types of Cars including color, model, year of manufacture with high accuracy.
The AI system automatically analyzes and suggests the car price based on experience parameters from more than 300,000 similar products on the market.
The system is deployed on the AWS Cloud, flexible in scaling or adapting functions.
RESULTS
Model AI automatically fills in information of 250 types of Cars including color, model, year of manufacture with 95% accuracy.
The suggested value of the car has a maximum price difference of 20,000 Yen (~$160), depending on the accuracy of the input information and vehicle price classification.