What is Knowledge Graph

I will share quick introduction to Knowledge Graph which will make it easy for you to understand it. This I cam across while going through the Generative AI course Knowledge Graphs for RAG on deeplearning.ai

A knowledge Graph is a database that stores information in Nodes and Relationships

Both nodes and relationships can be properties.

Nodes can be given labels to group them together. Nodes are data records, so they have properties as well.

Relationships have direction and a type.

If you see above image, Person Andreas knows Andrew since 2024. Person Andreas teaches course since 2024. Andrew introduces course as per given relationship.

Example

One prominent example of a knowledge graph in e-commerce is Google's own product, Google Shopping. Here's how it functions:

Data Integration: Google Shopping gathers data from various sources:

• Product Listings: Retailers upload information about their products, including details like brand, model, price, and specifications.
• User Reviews: Customer reviews provide valuable insights into product quality and performance.
• Customer Search Data: Analyzing what users search for and how they interact with product listings helps understand customer preferences and trends.

Knowledge Graph Construction:

• Entities: Google creates individual entities for products, brands, categories, and other relevant concepts.
• Relationships: Relationships between entities are established, connecting products with their brands, categories, and related items.
• Attributes: Entities are enriched with additional attributes like price range, color, size, and other relevant characteristics.

Enhancing User Experience:

• Improved Search: Knowledge graphs enable Google to understand the context of user searches, leading to more accurate and relevant product recommendations.
• Personalized Recommendations: By analyzing user preferences and purchase history, Google can provide personalized product suggestions and recommendations.
• Product Comparison: Users can easily compare different products based on their attributes and features.
• Cross-selling and Upselling: Knowledge graphs enable the identification of complementary or related products, boosting sales opportunities.

Additional Benefits:

• Enhanced SEO: Optimized product data and relevant keywords within the knowledge graph improve search engine optimization.
• Dynamic Pricing: Prices can be dynamically adjusted based on real-time market data and competitor analysis.
• Inventory Management: Knowledge graphs facilitate better inventory management by providing insights into product demand and stock levels.

Overall, Google Shopping's knowledge graph illustrates how this technology can revolutionize the e-commerce industry by offering a more intelligent and personalized shopping experience for users.

Ref -4 ways Google’s Shopping Graph helps you find what you want (blog.google)