Abstract: A machine learning model classifies points of interest in a parallel reality game hosted by a server. The server generates training data sets that include verified properties for points of interest. The machine learning model may predict unverified properties for points of interest. Players in the parallel reality game may input properties for the points of interest. The machine learning model use the received properties from players as inputs to the machine learning model to verify unverified properties or generate new properties for the points of interest. The server may classify the points of interest as suitable for particular activities, and the server may use the classifications for future activities within the parallel reality game.

Abstract: A game server generates a parallel-reality game that users may interact with in the real-world. The game server receives user information as users interact with game content. The game server includes a content marketplace that manages the exchange value of additional content items. The content marketplace matches additional game content with users playing the game based on the exchange value of the additional content items and user information. To do so, the content marketplace determines a propensity score for each additional content item quantifying a likelihood the user will interact with the additional content item while interacting with content in the parallel-reality game. The content marketplace provides the additional content item to the user's client device for display in the parallel-reality game based on the propensity score.

Abstract: A location-based game may include language learning features to help a player learn a foreign language. To generate these features, the game may generate language information by allowing players to play language gathering games. For example, a virtual character can use a native language of a player to instruct the player to identify an object in an environment. The player may then identify the object in the environment. The player may also speak the name of the identified object. The name of the object, a recording of the player speaking the name, and pixel values representing the object may be associated together. By gathering this information across different languages, a database can be formed that includes language information for many objects. This language information may then be used to generate language learning features.

Abstract: A location-based game helps players interact with their environment without interrupting the game. For example, while a player is playing the location-based game on a client device, the client device displays a virtual content item adjacent to a location (or building) associated with the content item. The content item may provide information about the location. The content item may be displayed without obscuring virtual objects of the game, thus allowing the player to continue playing the game and simultaneously learn about and interact with their environment.

Abstract: A graphical text entry system for an electronic device is provided. The electronic device may be, for example, a portable or hand-held electronic device such as a cellular telephone, PDA, or the like. The system includes a graphical text entry screen and a graphical text entry ring displayed on the graphical text entry screen. A plurality of characters such as letters or numbers is positioned on the graphical text entry ring. The system also includes a touch sensor ring located on the graphical text entry screen for selecting individual characters from among the plurality of characters located on the text entry ring.