Abstract: An online system using attributes of users to select content for presentation to the users predicts one or more attributes of users whose attributes are unavailable to the online system. For a user with one or more attributes unavailable to the online system, the online system applies a model to attributes of additional users to predict one or more attributes of the user. Attributes of the additional user use in the prediction may include demographic information and interactions with content by the additional users. The online system may determine an accuracy of the model by using the model to predict attributes for users whose attributes are known to the online system and comparing the predicted attributes to the known attributes. If the model's accuracy is less than a threshold value, the online system discontinues using the model to predict attributes of users.

Abstract: For a visit of a user to a web page where the user's identity on an online system is not presently known to the online system, the online system uses a machine learning model to make a prediction of the user's identity. The online system obtains visit data about the visit of the user to the web page. The online system identifies candidate user IDs that may represent the user, based on the visit data and data known about previous visits of the candidate user IDs. The online system derives visit features for each candidate user ID based on a relationship between the current visit data and previous visit data for the candidate user ID. The online system provides the visit features for each candidate user ID to a prediction model that determines whether, or how likely, the candidate user ID accurately identifies the visiting user, and based on the determinations selects one of the candidate user IDs as the most likely user ID for the visiting user.

Abstract: An online system using attributes of users to select content for presentation to the users predicts one or more attributes of users whose attributes are unavailable to the online system. For a user with one or more attributes unavailable to the online system, the online system applies a model to attributes of additional users to predict one or more attributes of the user. Attributes of the additional user use in the prediction may include demographic information and interactions with content by the additional users. The online system may determine an accuracy of the model by using the model to predict attributes for users whose attributes are known to the online system and comparing the predicted attributes to the known attributes. If the model's accuracy is less than a threshold value, the online system discontinues using the model to predict attributes of users.

Abstract: For a visit of a user to a web page where the user's identity on an online system is not presently known to the online system, the online system uses a machine learning model to make a prediction of the user's identity. The online system obtains visit data about the visit of the user to the web page. The online system identifies candidate user IDs that may represent the user, based on the visit data and data known about previous visits of the candidate user IDs. The online system derives visit features for each candidate user ID based on a relationship between the current visit data and previous visit data for the candidate user ID. The online system provides the visit features for each candidate user ID to a prediction model that determines whether, or how likely, the candidate user ID accurately identifies the visiting user, and based on the determinations selects one of the candidate user IDs as the most likely user ID for the visiting user.

Abstract: An online system using attributes of users to select content for presentation to the users predicts one or more attributes of users whose attributes are unavailable to the online system. For a user with one or more attributes unavailable to the online system, the online system applies a model to attributes of additional users to predict one or more attributes of the user. Attributes of the additional user use in the prediction may include demographic information and interactions with content by the additional users. The online system may determine an accuracy of the model by using the model to predict attributes for users whose attributes are known to the online system and comparing the predicted attributes to the known attributes. If the model's accuracy is less than a threshold value, the online system discontinues using the model to predict attributes of users.