Facebook Dating AI system

UPDATED MAR 7, 2025
The content you see on your Facebook Dating is selected, ranked and delivered to you by an artificial intelligence (AI) system. Within one AI system, multiple machine learning models work together to deliver your experience. These models and their input signals are dynamic and they change frequently as the system learns and improves over time.

Overview of Facebook Dating

When you view profiles on Dating, one of the underlying AI systems determines which profiles are suggested to you as a potential match and what order they’re shown in. These are profiles we think you might be interested in based on information you’ve added to or shared with Dating.
How Facebook Dating works
The AI system behind Dating automatically orders the profiles you see by predicting who you might like the most. Here’s how it works:
  1. Gather profiles
    First, the system gathers profiles of people that might interest you. This does not include anybody who you’ve blocked on Facebook or is currently your Facebook friend—but you can still match and see somebody’s Dating profile if you unblock or become friends with them.
  2. Leverage signals
    Next, the system considers a variety of input signals about the profiles. These signals might include the preferences you set for who you want to see, such as gender(s), location(s), age, or information people have added to their profile, such as hometown or college.
  3. Rank profiles
    The system calculates scores for each profile based on how likely you are to have a conversation with the person or like the profile. Then the system puts these profiles in order by their scores.
  4. Apply filters
    The system applies filters to remove profiles that you may not want to see based on things like your preferences or if a profile is incomplete. It also doesn't show you profiles of current Facebook friends or people you’ve blocked on Facebook.
    Finally, certain integrity processes are applied to all results. These determine which integrity measures, if any, need to be applied to the profiles selected for ranking.
How to customize what you see
Your experience on Facebook Dating is personalized based on your activity, and you have options to control or customize what you see. Below, we describe how to do this with different in-product features. Options shown here may not be available to everyone.
Show more or less people like a profile
This feature lets you customize which kinds of profiles are shown to you in Dating home. Choosing "Show more” or “Show less” will increase or decrease the ranking score for a profile and other profiles like it.
Update your Dating preferences
You can change and update your Dating preferences at any time. For example, you can set preferences for age, height and distance from you.
Swipe through profiles in “Not sorted”
To see profiles in a randomized order, with no ranking applied, go to Dating and select “Not sorted.”
Report or block a profile
If a conversation or profile in Dating makes you uncomfortable, you can report it or block the profile.
Like or pass on a profile
You can send a like to the profiles you’re interested in. You can pass on profiles you’re not interested in and the system won’t recommend certain profiles to you.
How the AI delivers content to you
We want you to see content you enjoy and find interesting. To achieve this, the AI system has models that help it make predictions about content you'll find most relevant and valuable. These prediction models use underlying input signals to help select content you're most likely to engage with.
Below are some of the significant predictions–and input signals that inform them–that we use in this AI system.
How likely you are to exchange messages with a specific profile
Signals influencing this prediction include:
  • How often other people like, pass on or exchange messages with the same profile
  • Attributes of profiles that you’ve interacted with or liked in the past, such as photos, age or colleges
  • How profile attributes match your preference settings, such as age, height and education level
  • How many likes, passes and messages you send to other profiles
  • Your Facebook activity, such as friend requests and accepts, the Pages and profiles you follow and what content you like and comment on in your Feed
How likely you are to spend time looking at aprofile
Signals influencing this prediction include:
  • Attributes of profiles that you’ve interacted with or liked in the past, such as photos, age or colleges
  • How many photos as well as the quality and contents of the photos on the profile
  • How many likes or passes you sent to other profiles in the past day
  • How much time you spend looking at profiles, both on average and compared to others
  • How much information has been added to the profile, including the length of the bio, quantity of photos and completed attributes
How likely you are to like a profile
Signals influencing this prediction include:
  • Your Facebook activity, such as friend requests and accepts, the Pages and profiles you follow and what content you like and comment on in your Feed
  • How many likes and passes you send on other profiles
  • How often other people like or pass on the same profile
  • Attributes of profiles that you’ve interacted with or liked in the past, such as photos, age or colleges
  • How profile attributes match your preference settings, such as age, height and education level