The Brave search engine has a query generalization feature based

The Brave search engine has a query generalization feature based on AI algorithms

Functions based on neural networks are increasingly appearing in popular search engines. The developers of the Brave search engine decided to keep up with this trend and introduced the Summarizer tool, which was based on several large language models (Open AI’s ChatGPT tool was not used). The name of the function suggests that its purpose is to generalize search queries for which the algorithm processes data from different sources.

    Image Source: janeb13 / Pixabay

Image Source: janeb13 / Pixabay

According to reports, the Summarizer feature has become available to all users of the Brave search engine on computers and mobile devices. In the examples given by the developers, this tool summarizes information about queries such as Are paracetamol and ibuprofen the same thing?Use of medical resource data or “What Happened in East Palestine, Ohio” based on data from news sites.

“With 22 million daily searches, Brave is the fastest growing search engine since Bing. Unlike an AI chatbot, which can provide inaccurate answers, Summarizer creates a simple summary at the top of search results by aggregating data from various online sources and providing source attribution for transparency and accountability. Starting today, this open system is available to all Brave users to help them navigate search results.”— said the representative of Brave.

    Image source: Brave

Image source: Brave

It should be noted that the AI ​​function of the Brave search engine does not generate any generalization for queries at this time. Currently, this approach is only used for 17% of searches, but the company plans to increase that number over time. At the heart of the Summarizer feature is Brave’s proprietary language model, which combines three algorithms. The first question and answer model allows you to collect textual data from different sources. The second model filters out spam and hate speech. The third model processes the information collected and generates a short summary that the search engine user sees.

About the author

Robbie Elmers

Robbie Elmers is a staff writer for Tech News Space, covering software, applications and services.

Add Comment

Click here to post a comment