Pinecone Introduces the Most Powerful Hybrid Search Functionality in the Globe

For the first time, a new keyword-aware semantic search will provide tech giant-level technologies to businesses of all sizes.

Fremont, CA: Pinecone created the first vector database to enable developers to include vector search into commercial applications easily. It combines cutting-edge vector search libraries with additional features like filtering and distribution infrastructure to deliver great speed and reliability at any scale.

"Our research shows that keyword-aware semantic search is superior to either semantic search or keyword search separately. This release finally allows businesses to provide their users with the most relevant possible results no matter how specific the query or how unique the topic," said Edo Liberty, Founder and CEO of Pinecone.

Pinecone Systems Inc. launched a revolutionary keyword-aware semantic search solution that provides the world's most accessible and sophisticated mix of semantic and keyword search results.

In contrast to basic keyword-based searches, "vector search" enables businesses to give relevant results based on semantic or related meanings. Conversely, keywords continue to be important in searches containing rare words such as names or industry-specific phrases. Therefore, with rare exceptions, businesses must choose between semantic and keyword searches or operates both systems concurrently.

Both of these alternatives could be better. When businesses select one or the other, the results could be more complete; when they run both systems concurrently and try to aggregate the data, the cost and complexity skyrocket.

"Most companies do not have the resources to provide their users with the types of advanced search solutions that tech giants do, and this release from Pinecone changes all of that," concluded Liberty, who headed Amazon AI Labs and Yahoo's Research Lab in New York before founding Pinecone.

Pinecone created a first-of-its-kind hybrid search system that allows for keyword-aware semantic search. This technique can search across two data kinds — "dense vectors" created by ML models to express meaning and "sparse vectors" generated by classic keyword-ranking algorithms like BM25 — before merging everything into a single ranked list of the best relevant results.

Pinecone's new keyword-aware semantic search solution means that businesses of all sizes and sorts can now give the same highly relevant searches that only tech titans capable of investing considerably in data science and engineering effort could previously offer. In addition, their programs can now comprehend what users say and what they mean.