Introduction

Welcome to this lesson on text embeddings, the central concept of semantic search. Before we delve into state-of-the-art text retrieval and semantic search methods in later lessons, it is necessary to first understand how text embeddings work.

In this lesson, we'll start by exploring a friendly introduction to text embeddings, followed by a more technical definition and tools for visualizing embeddings. Then, we'll move on to similarity scoring, a crucial component of text retrieval that allows us to measure the semantic similarity between two embeddings. And finally, we'll meet two of the first-ever text embedding models, word2vec for word embeddings and doc2vec for document embeddings.