AI Recommends Products I Want... Naver Strengthens Hyper-Personalized Commerce

The 'Personalized Block' (left and center) provides reasons for recommendations based on the user's shopping history and search queries, while the 'Recommended Block' (right) suggests shopping themes and products that the user might be interested in. (Photo by Naver)

Naver is advancing its AI-based product recommendation technology. Users will be able to enjoy highly personalized product recommendation services.

On the 20th, Naver applied a more advanced AI technology to shopping search by combining its self-developed AI product recommendation technology 'AiTEMS' with the large-scale AI 'HyperCLOVA.' When users enter product-related keywords in Naver Search, a 'personalized block' that analyzes users' shopping activity history and shows the reasons for product recommendations together, and a 'recommendation block' that suggests users' shopping interests based on queries will be introduced.

As of June, the transaction amount of AI-recommended products increased by 30% compared to the same period last year, reflecting growing user demand and satisfaction with AI personalized recommendations. Naver plans to further enhance AI curation usability through this 'personalized product recommendation block.'

To provide a highly personalized AI product recommendation experience, Naver strengthened the personalized recommendation model and engine structure and built a large-scale data recommendation system to enable more diverse and accurate recommendations from the 1.5 billion Naver Shopping product database (DB).

In particular, the 'personalized block' applied with 'recommendation reason modeling technology' analyzes users' shopping history in real time and reflects the reasons for recommendations to immediately show product recommendation results linked to user preferences and search queries.

For example, if a user who previously clicked on a moisturizing cream or has a history of adding it to the cart, wishlist, or purchasing searches for 'moisturizing cream' again, the 'Personalized Shopping for Hong Gil-dong' block will show reasons for recommending the product, such as ▲the product clicked 3 days ago ▲a store purchased twice in 3 months ▲a brand visited 8 times in 3 months.

Even with the same keyword, personalized recommendations are more effective because they reflect different search queries and purchase histories for each user. Internal pre-tests showed that the product click-through rate in the personalized block was up to 27% higher compared to the existing shopping recommendation model.

The 'recommendation block' integrates Naver's large-scale AI 'HyperCLOVA' technology to recommend shopping themes and sub-products that users might be interested in according to the type of search query. HyperCLOVA automatically generates shopping interests and keyword lists, and suggests interest keywords by analyzing the correlation between users' search queries and shopping histories.

For example, when searching for 'basil,' for users who have previously browsed flowerpots or other plants, a 'Frequently Bought Together #BalconyDecoration Products' block is created to show air-purifying plants or seedlings. For users who have looked for tableware or salad sauces, the 'Frequently Bought Together #Brunch Products' interest keyword is recommended, suggesting bagels or other salad vegetables.

As Naver's AI personalized recommendation technology advances, users will continue to experience more refined and expanded product exploration, and unique SME (small and medium-sized enterprises) products are expected to have a higher chance of matching with users.

Currently, AiTEMS technology recommends 35% of less popular SME products and exposes 52% of products from new stores, recommending a variety of SME products. This helps alleviate the concentration on popular products. The personalized product recommendation block is expected to increase sales opportunities for SMEs.

Lee Jung-tae, the lead responsible for commerce AI technology at Naver, said, "Based on Team Naver's technological synergy, we will connect user satisfaction with SME growth and establish differentiated AI technology competitiveness within the industry."

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