Share via

Using the Bing Text Search (Preview)

  • Article

Warning

The Semantic Kernel Text Search functionality is in preview, and improvements that require breaking changes may still occur in limited circumstances before release.

Overview

The Bing Text Search implementation uses the Bing Web Search API to retrieve search results. You must provide your own Bing Search Api Key to use this component.

Limitations

Feature Area Support
Search API Bing Web Search API only.
Supported filter clauses Only "equal to" filter clauses are supported.
Supported filter keys The responseFilter query parameter and advanced search keywords are supported.

Tip

Follow this link for more information on how to filter the answers that Bing returns. Follow this link for more information on using advanced search keywords

Getting started

The sample below shows how to create a BingTextSearch and use it to perform a text search.

using Microsoft.SemanticKernel.Data;
using Microsoft.SemanticKernel.Plugins.Web.Bing;

// Create an ITextSearch instance using Bing search
var textSearch = new BingTextSearch(apiKey: "<Your Bing API Key>");

var query = "What is the Semantic Kernel?";

// Search and return results as a string items
KernelSearchResults<string> stringResults = await textSearch.SearchAsync(query, new() { Top = 4, Skip = 0 });
Console.WriteLine("--- String Results ---\n");
await foreach (string result in stringResults.Results)
{
    Console.WriteLine(result);
}

// Search and return results as TextSearchResult items
KernelSearchResults<TextSearchResult> textResults = await textSearch.GetTextSearchResultsAsync(query, new() { Top = 4, Skip = 4 });
Console.WriteLine("\n--- Text Search Results ---\n");
await foreach (TextSearchResult result in textResults.Results)
{
    Console.WriteLine($"Name:  {result.Name}");
    Console.WriteLine($"Value: {result.Value}");
    Console.WriteLine($"Link:  {result.Link}");
}

// Search and return s results as BingWebPage items
KernelSearchResults<object> fullResults = await textSearch.GetSearchResultsAsync(query, new() { Top = 4, Skip = 8 });
Console.WriteLine("\n--- Bing Web Page Results ---\n");
await foreach (BingWebPage result in fullResults.Results)
{
    Console.WriteLine($"Name:            {result.Name}");
    Console.WriteLine($"Snippet:         {result.Snippet}");
    Console.WriteLine($"Url:             {result.Url}");
    Console.WriteLine($"DisplayUrl:      {result.DisplayUrl}");
    Console.WriteLine($"DateLastCrawled: {result.DateLastCrawled}");
}

Coming soon

More coming soon.

Coming soon

More coming soon.

Next steps

The following sections of the documentation show you how to:

  1. Create a plugin and use it for Retrieval Augmented Generation (RAG).
  2. Use text search together with function calling.
  3. Learn more about using vector stores for text search.

Text Search Abstractions Text Search Plugins Text Search Function Calling Text Search with Vector Stores