Algolia Launches NeuralSearch, A Vector And Keyword Search API
Algolia NeuralSearch understands natural language and delivers results in milliseconds. Algolia launched Algolia NeuralSearch, providing vector and keyword search in a single API with end-to-end artificial intelligence processing every query. Algolia NeuralSearch understands natural language and delivers results in milliseconds. It uses advanced large language models (LLMs), the same technology underpinning ChatGPT and generative AI, […]
Topics
Algolia NeuralSearch understands natural language and delivers results in milliseconds.
Algolia launched Algolia NeuralSearch, providing vector and keyword search in a single API with end-to-end artificial intelligence processing every query.
Algolia NeuralSearch understands natural language and delivers results in milliseconds. It uses advanced large language models (LLMs), the same technology underpinning ChatGPT and generative AI, and goes a step further with Algolia’s Neural Hashing for hyper-scale and constantly learns from user interactions.
Algolia NeuralSearch analyses the relationships between words and concepts, generating vector representations that capture their meaning in an abstract and contextual manner. Because vector-based understanding and retrieval is combined with Algolia’s full-text keyword engine, it works for exact matching too. Neural Hashing compresses search vectors from 2,000-decimal-long numbers into static-length expressions.
Additionally, as the index changes, new products are added, new content is uploaded, or as terms take on new meaning, the AI-powered Algolia NeuralSearch product will learn and adjust automatically.
“Algolia is committed to advancing AI-powered search, and we believe Algolia NeuralSearch does just that. Algolia NeuralSearch, a first-of-its-kind hybrid search product, provides users with a smarter and more intuitive way to discover the most relevant content they want, when they need it, irrespective of the type of query presented,” said Bernadette Nixon, CEO of Algolia, in a statement.
“Importantly, we make it easy to achieve live production quickly. Specifically, we provide the setup, scaling, and management of all search capabilities and services, all of which helps accelerate and power discovery. Moreover, Algolia NeuralSearch is backward-compatible, which means there is zero engineering required for customers to become AI-enabled.”
Frasers Group, a fashion apparel provider, was among the first Algolia customers to use Algolia NeuralSearch in a real world environment. “We tested Algolia NeuralSearch with two of our brands (Missguided and Isawitfirst), and were thrilled to see above a 65 percent drop in zero search results and up to 17 percent uplift in conversion rates. These results, despite only sending a portion of our query traffic to Algolia NeuralSearch over a four-week period, exceeded our expectations,” said Kyle Sanders, head of digital optimisation at Frasers Group.
“Notably, our existing search implementation seamlessly evolved to further improve our customers’ discovery journey and improve their experience on our website, all without having to make any changes to a single line of code. We are excited to see what the future holds with Algolia NeuralSearch.”
