Abstract: An online system may identify content with which a user has an interest. For example, the online system may determine that a user has an interest in the content based on interaction information indicating that the user interacted with the content. In a particular example, the online system may identify image concepts included in the content based on computer vision techniques that recognize the image concepts. The online system may model probabilities that image concepts will appeal to users. Based on the modeled probabilities, the online system may automatically recommend image concepts for inclusion in candidate images, automatically generate candidate images, or assess candidate images to determine a probability of user interaction with the assessed candidate images.

Abstract: An online system may identify content with which a user has an interest. For example, the online system may determine that a user has an interest in the content based on interaction information indicating that the user interacted with the content. In a particular example, the online system may identify image concepts included in the content based on computer vision techniques that recognize the image concepts. The online system may model probabilities that image concepts will appeal to users. Based on the modeled probabilities, the online system may automatically recommend image concepts for inclusion in candidate images, automatically generate candidate images, or assess candidate images to determine a probability of user interaction with the assessed candidate images.

Abstract: A server system receives, from client devices, a first plurality of messages associated with user actions in a first version of an application. The first plurality of messages includes performance indicators relating to the first version. The method further includes receiving, from client devices, a second plurality of messages associated with user actions in a second version of an application. The second plurality of messages includes performance indicators relating to the second version. The second version of the application is more recent than the first version of the application. Subsequently, compare the performance indicators relating to the first version of the application with the performance indicators relating to the second version of the application. In response to a determination that a difference between a respective performance indicator in the first version and the second version satisfies a threshold value, provide a report of the respective performance indicator.