📢 Do you remember Google’s blog post?

“Introducing the Knowledge Graph: things, not strings” ❔

This post indicates that Google now understands things and is no longer just matching keywords (strings).

❓ The question is, “How Google Understands Things?”

The answer includes many techniques, and one of them is Context Vectors.

👉 Google’s Patent on Context Vectors

In September 2016, a Google patent, titled “User-context-based search engine”, introduced the concept of context vectors as a “data extraction tool” that combines the benefits of both search engines and directories while eliminating their drawbacks.

✅ What are Context Vectors?

Context vectors are the vectors that are used for determining the context of the content or content publisher (Website) in a given search intent. Google simply tries to profile the words that are being used in a certain domain to see what the unique sides of a context are.

Simply: Context Vectors are mathematical representations of words and phrases that capture their meanings based on the context in which they are used.

📌 Example

Consider the word ‘apple’. With context vectors, a search engine can understand the different contexts in which the word ‘apple’ is used, thereby improving the accuracy of search results.

• 🍏 Fruit: When searching for healthy snacks.
• 💻 Technology: When looking for Apple Inc. products.