Introduction to generic search

This page introduces and lists the capabilities of Vertex AI Search for generic apps. The page also provides links to the available features, tutorials, and checklists, to get you started with Vertex AI Search for generic apps.

What is Vertex AI Search for generic apps?

Vertex AI Search for generic apps is a powerful, Google-quality search and content discovery engine that you can integrate into your applications that contain website data and other structured or unstructured data. The search capability is beyond basic keyword matching and uses AI to deliver highly relevant results, provide personalized browse and search experiences, and generate AI answers grounded in your data.

You can use generic search app for vertical-agnostic data that's on public websites or is in structured or unstructured format. Additionally, Vertex AI Search offers other vertical-specific search and recommendations apps.

Key capabilities

The key capabilities of Vertex AI Search are as follows:

  • High-quality search: Leverages Google's search expertise to understand user intent, even with complex queries and natural language queries. It combines keyword and semantic search to serve the best results.
  • Personalized browse: Provides personalized results without a specific search query and personalized feed based on a user's context and navigation patterns. It is ideal for discovery experiences to view personalized category pages and home feeds.
  • Data sources: Works with the following variety of data sources:
    • Website: Index your public websites and use advanced features, such as index enrichment with the structured data in your websites.
    • Structured Data: Search over data organized in a defined format, such as databases, JSON files in Cloud Storage, or BigQuery tables—for example, hotel catalogs, real estate listings, and restaurant directories.
    • Unstructured Data: Search over documents like PDFs, HTML files, and TXT files or image files like JPEG and PNG files that are stored in Cloud Storage or BigQuery.
    • Blended Search: Search over multiple data stores that blend data from the data sources mentioned above. For example, you can create a search app and connect it to a website data store and a document data store. This lets your users search over all of your content at once.
  • Grounded AI answer generation: Generates AI answers grounded in your data, with citations to the source documents. You can also ask follow up questions and related queries.
  • Personalization: Improves results and ranking over time by learning from user interactions captured in user events, such as clicks and conversions.
  • Customization: Offers several ways to tune and configure the search and browse experience fit for your business needs.

Overview

The following diagram shows the key components of generic search and how they work together:

key components of generic custom search
Figure 1. Different components of generic search

The components of Vertex AI Search for generic search can be explained as follows:

  • Data store: Your content from different data sources is stored in a Vertex AI Search data store. The source data can be public website data or structured and unstructured data.
  • Data processing and indexing: Vertex AI Search understands and indexes your data, creating a searchable and retrievable representation. This includes the following:
    • Keyword extraction: Identifies and generates important terms necessary to retrieve the correct information.
    • Semantic understanding using embeddings: Creates vector embeddings to capture the meaning of the content.
    • Metadata processing: Processes your documents using the document's structured data or metadata. For example, location in a hotel catalog, modification or creation dates in a web page's metadata.
    • Advanced document parsing: Understands document structure and annotates advanced information, such as tables, images, and graphs, using OCR or layout parsing.
  • Search app: At the heart of the generic search is a search app, which connects to one or more data stores that bring data from different sources. For blended search, the data is ingested through connectors. You configure the search and browse behavior at the app level.
  • User query: The input from a user intended to retrieve information from your app, which can be of two types:
    • Search query: The user enters a targeted search query using text or images. Textual search is powered by autocomplete.
    • Navigational query or browse: An exploratory search to deliver personalized relevant content with no specific query. It is powered by the user's past activity and other signals, such as current category page and location.
  • Retrieval and ranking: There are several sub-components to retrieval and ranking of results:
    • Query understanding for search: Vertex AI Search analyzes a search query using the following:
      • Natural language processing: To understand the intent.
      • Filters with natural language understanding: Translates locations from natural language queries into geo-coordinates and the conditions in natural language queries into filters.
      • Knowledge graph: To disambiguate terms and expand the search.
      • Optional features: Includes spelling correction, synonyms, and query rephrasing.
    • Retrieval: Vertex AI Search finds the most relevant documents or chunks based on the following methods:
      • Keyword matching for search: Conventional search based on terms.
      • Semantic search: Using embeddings to find conceptually similar content.
      • Filtering: Applying any filters you've configured—for example, date, category, or relevance score.
    • Ranking: Vertex AI Search ranks the results based on the following factors:
      • Relevance: A combination of keyword and semantic matching during search.
      • Web signals for website search: Factors like page quality and popularity.
      • Boosting and burying: Your custom rules to promote or demote certain results.
      • Personalization: Learning from user interactions. This is optional but highly recommended.
      • Ordering: Applying ordering instructions, for example, by date.
  • Results and answer generation:
    • Search results: A ranked list of relevant documents or chunks is returned with optional features, such as snippets, extractive answers, and extractive segments. The results that are served can be configured with the help of serving controls. You can also tune the search results.
    • Answer generation: A concise, synthesized answer is generated based on the top and relevant results, with citations. This uses advanced LLM features.
    • Personalized browse: A personalized set of documents with the highest predicted likelihood of engagement or conversion is returned. This prediction uses an advanced model that learns from user interactions.
  • User events: A tracker for user interactions, such as clicks and views, that helps Vertex AI Search learn and improve search and personalization. User events aid in optimizing your business KPIs including engagement, conversion, and revenue.

Key features and configurations

The following features and configurations are available for your generic search apps. At each stage you can customize these settings to serve the best results to your users.

key components of generic custom search
Figure 2. Key features and configurations in generic search

To elaborate, here are the available configurations:

What's next