What is embeddings?
Embeddings are lists of numbers that represent the meaning of a piece of text, so that a computer can tell how similar two pieces of text are even when they use different words.
Why it matters
Customers rarely use the same words as a help article. One person writes about a lost parcel and the article is titled missing delivery. Keyword search can miss the match. Embeddings let a system compare meaning, not spelling, which improves search, helps an AI agent find the right source and makes it possible to group similar tickets together.
How it works
A model reads a sentence or passage and turns it into a long list of numbers. Texts with similar meanings end up with similar lists. To compare two texts, the system measures how close their lists are. Help articles are usually converted in advance, and each customer question is converted when it arrives so that the closest articles can be found.
Example
A customer types, “My card got swallowed by the cash machine.” The bank’s help centre has no article containing the word swallowed. The question is turned into an embedding and compared with the embeddings of every article. The closest match is an article called “What to do if a cash machine keeps your card”, and the chat agent answers from it.
