エンベディングは、テキスト、画像、動画の数値表現であり、入力間の関係をとらえます。ML モデル（特に生成 AI モデル）は、大規模なデータセット内のパターンを識別してエンベディングを作成することに適しています。アプリケーションでは、言語を処理および生成するためにエンベディングを使用して、コンテンツに固有の複雑な意味とセマンティックな関係を認識できます。
Google Distributed Cloud（GDC）エアギャップ上の Vertex AI は、英語と多言語のテキスト入力用の Text Embedding API をサポートしています。テキスト エンベディングは、テキストをベクトルと呼ばれる浮動小数点数の配列に変換することで機能します。こうしたベクトルは、テキストの意味を捉えるように設計されています。エンベディング配列の長さは、ベクトルの次元数と呼ばれます。たとえば、テキストの一節は、数百の次元を含むベクトルで表せます。次に、2 つのテキストのベクトル表現間の数値的な距離を計算することで、アプリケーションでオブジェクト間の類似性を判断できます。
[[["わかりやすい","easyToUnderstand","thumb-up"],["問題の解決に役立った","solvedMyProblem","thumb-up"],["その他","otherUp","thumb-up"]],[["わかりにくい","hardToUnderstand","thumb-down"],["情報またはサンプルコードが不正確","incorrectInformationOrSampleCode","thumb-down"],["必要な情報 / サンプルがない","missingTheInformationSamplesINeed","thumb-down"],["翻訳に関する問題","translationIssue","thumb-down"],["その他","otherDown","thumb-down"]],["最終更新日 2025-09-04 UTC。"],[],[],null,["# Text embeddings overview\n\n| **Important:** This content applies to version 1.14.4 and later.\n\nEmbeddings are numerical representations of text, images, or videos that capture\nrelationships between inputs. Machine learning models, especially\nGenerative AI models, are suited for creating embeddings by identifying\npatterns within large datasets. Applications can use embeddings to process and\nproduce language, recognizing complex meanings and semantic relationships\nspecific to your content.\n\nVertex AI on Google Distributed Cloud (GDC) air-gapped supports\nText Embedding APIs for English and multilingual textual input.\nText Embedding works by converting text into arrays of floating-point\nnumbers called vectors. These vectors are designed to capture the meaning of the\ntext. The length of the embedding array is called the vector's dimensionality.\nFor example, one passage of text might be represented by a vector containing\nhundreds of dimensions. Then, by calculating the numerical distance between the\nvector representations of two pieces of text, an application can determine the\nsimilarity between the objects.\n| **Note:** Only text is supported as an input type for embeddings on GDC.\n\nFor a list of supported models, see\n[Embeddings models](/distributed-cloud/hosted/docs/latest/gdch/application/ao-user/genai/genai-models#embeddings-models).\n\nFor a list of supported multilingual languages, see\n[Supported text embedding languages](/distributed-cloud/hosted/docs/latest/gdch/application/ao-user/genai/supported-text-languages).\n\nText embeddings use cases\n-------------------------\n\nSome common use cases for text embeddings include:\n\n- **Semantic search**: Search text ranked by semantic similarity.\n- **Classification**: Return the class of items with attributes similar to the given text.\n- **Clustering**: Cluster items whose text attributes are similar to the given text.\n- **Outlier detection**: Return items where text attributes are least related to the given text.\n- **Conversational interface**: Cluster groups of sentences that can lead to similar responses, like in a conversation-level embedding space.\n\n### Example use case: Develop a book recommendation chatbot\n\nIf you want to develop a book recommendation chatbot, the first thing to do is\nto use a deep neural network (DNN) to convert each book into an embedding\nvector, where one embedding vector represents one book. You can feed the book\ntitle or text content as input to the DNN. Alternatively, you can use both of\nthese inputs together, along with any other metadata describing the book, such\nas the genre.\n\nThe embeddings in this example could include thousands of book titles with\nsummaries and their genre. It might have representations for books like\n*Wuthering Heights* by Emily Brontë and *Persuasion* by Jane Austen that are\nsimilar to each other (small distance between numerical representation). In\ncontrast, the numerical representation of the book *The Great Gatsby* by F.\nScott Fitzgerald would be further, as the time period, genre, and summary are\nless similar.\n\nThe inputs are the main influence on the orientation of the embedding space. For\nexample, if we only had book title inputs, then two books with similar titles\nbut very different summaries could be close together. However, if we include the\ntitle and summary, then these same books are less similar (further away) in the\nembedding space.\n\nWorking with Generative AI, this book-suggestion chatbot could summarize,\nsuggest, and show you books that you might like (or dislike) based on your\nquery.\n\nWhat's next\n-----------\n\n- Learn how to [get text embeddings](/distributed-cloud/hosted/docs/latest/gdch/application/ao-user/genai/get-text-embeddings)."]]
