In conventional EC sites, the data provided by suppliers is a wide variety of product data, so it was not possible to use it as a company's data in a unified manner.

Since the classification and registration of products into their own product categories were performed manually, efficiency and accuracy were uneven depending on the level of skill of the workers. However, the speed at which new products were introduced to the market was a major issue.

In terms of cost, the cost rises in proportion to the number of registered products. The company was faced with management issues such as being unable to adopt a strategy to increase the number of products handled.

In response to these issues, regardless of product supplier or product, AI automatically identifies product categories and product attributes from product images and text data, and builds its own database. As a result, the time required for product registration has been greatly reduced, the number of products handled has increased significantly, and it has become possible to speedily launch new products into the market. In addition, we were able to eliminate unevenness in product classification and attribute extraction work, create a system that can continuously improve accuracy, and realize an environment that allows us to propose products that meet the needs of each customer in a timely manner.