Real estate online marketplace Zillow has launched natural language-based home searches. Using AI-based natural language processing (NLP), the feature allows users to skip filters and use natural language phrases instead. The AI understands phrases relating to layout, location, style, condition, price, and lifestyle.

Users can enter phrases like “$700K homes in Charlotte with a backyard” or “open house near me with four bedrooms” into the Zillow search bar. This replaces step-wise filtering. They can also save searches and Zillow will notify them when new qualifying listings come online.

Zillow’s natural language search feature takes users’ queries and scans millions of listing details to bring relevant results to the surface. At the same time, the feature is training machine learning models to better respond to search queries that use natural, human-like sentences.

According to the company, it is the first major residential real estate marketplace to implement AI-powered search.

Many Zillow features are powered by AI and machine learning, including Neural Zestimate valuation, personalized home recommendations, and generating floor plans from panoramic photos.

“Beyond easy-to-filter criteria like bedrooms and bathrooms, buyers are considering many other specific features that match their unique lifestyle,” said Jenny Arden, Zillow’s chief design officer. “This new tool is a game changer for home shopping, because it helps shorten the sometimes long and stressful house-hunting process by creating an easy, more modern way to search, and it delivers relevant search results in a simple, uncluttered way.”

“The future of real estate will be powered through AI,” Arden said. “We are proud to be the first in the industry to offer this smarter way to search, and excited to see how it learns and evolves to help each Zillow shopper find their perfect home.”