Abstract: Embodiments utilize interpolation to smooth animation frames by generating synthetic Shapley values utilized to generate the animation. Additionally, some embodiments provide for generation of an embedding space for improved understandability of the model training based on corresponding Shapley values. The embedding space is mapped to using activations derived from a specially configured LSTM for each model, where an embedded representation of the LSTM activations is generated by a specially configured autoencoder such that the embedded representation may be mapped to the embedding space. The embedding space and animations are renderable to an interface for user investigation and may be further customizable to narrow investigation on particular data thereof.

Abstract: The APPARATUSES, METHODS AND SYSTEMS FOR AN ONLINE GAME MANAGER (“OGM”) transform user social graph and online behavioral pattern data inputs via OGM components into optimized server computational load reports. In one embodiment, the OGM may process a user server migration request for a user by obtaining user social graph data identifying co-users, and social activity pattern data associated with the user and the co-users. The OGM may also obtain server load indices of servers associated with the co-users, and server load analysis rules. By analyzing the server load indices using the server load analysis rules, the OGM may identify a subset of the servers as non-overloaded transfer candidate servers. The OGM may calculate transfer candidate server scores for the non-overloaded transfer candidate servers using the user social graph data and the social activity pattern data.

Abstract: Methods, systems, and computer programs are presented for expanding gaming social networks with players having no existing social relationships with other players. One method includes an operation for identifying first filters selected by a first user, the first filters identifying a first criteria for desired new friends. Another operation identifies second filters selected by a plurality of users in the network, where the second filters are entered by each user to identify second criteria for being added as a friend in the network. A plurality of candidates to be added as friends for the first user are determined, where each candidate conforms to the first criteria and where the first user conforms to the second criteria set by each candidate. Additionally, friendship requests are sent to one or more of the candidates or candidates are added as friends when the candidates have chosen automatic addition of friends.

Abstract: In a method and system of sending an out-of-band instant message from a game networking system, an instant messaging system is accessed from the game networking system to send the out-of-band instant message to a friend of the user on behalf of the user, with the out-of-band instant message pertaining to the game networking system.

Abstract: The APPARATUSES, METHODS AND SYSTEMS FOR AN ONLINE GAME MANAGER (“OGM”) transform user social graph and online behavioral pattern data inputs via OGM components into optimized server computational load reports. In one embodiment, the OGM may process a user server migration request for a user by obtaining user social graph data identifying co-users, and social activity pattern data associated with the user and the co-users. The OGM may also obtain server load indices of servers associated with the co-users, and server load analysis rules. By analyzing the server load indices using the server load analysis rules, the OGM may identify a subset of the servers as non-overloaded transfer candidate servers. The OGM may calculate transfer candidate server scores for the non-overloaded transfer candidate servers using the user social graph data and the social activity pattern data.

Abstract: The APPARATUSES, METHODS AND SYSTEMS FOR A ONLINE GAME MANAGER (“OGM”) transform user social graph and online behavioral pattern data inputs via OGM components into optimized server computational load reports. In one embodiment, the OGM may process a user server migration request for a user by obtaining user social graph data identifying co-users, and social activity pattern data associated with the user and the co-users. The OGM may also obtain server load indices of servers associated with the co-users, and server load analysis rules. By analyzing the server load indices using the server load analysis rules, the OGM may identify a subset of the servers as non-overloaded transfer candidate servers. The OGM may calculate transfer candidate server scores for the non-overloaded transfer candidate servers using the user social graph data and the social activity pattern data.

Abstract: Methods, systems, and computer programs are presented for expanding gaming social networks with players having no existing social relationships with other players. One method includes an operation for identifying first filters selected by a first user, the first filters identifying a first criteria for desired new friends. Another operation identifies second filters selected by a plurality of users in the network, where the second filters are entered by each user to identify second criteria for being added as a friend in the network. A plurality of candidates to be added as friends for the first user are determined, where each candidate conforms to the first criteria and where the first user conforms to the second criteria set by each candidate. Additionally, friendship requests are sent to one or more of the candidates or candidates are added as friends when the candidates have chosen automatic addition of friends.

Abstract: Methods, systems, and computer programs are presented for expanding gaming social networks with players that do not have an existing social relationship with other players. One method includes an operation for determining one or more selection rules for a player of an online game, where each selection rule defines a criterion for selecting users from a plurality of users. The determined one or more selection rules are applied to the plurality of users resulting in a plurality of candidate game friends from the plurality of users. At least one of the candidate game friends does not have a social link with the player. In addition, the method includes an operation for presenting the plurality of candidate game friends. An operation to link, in the online game, the player with at least one of the plurality of candidate game friends is initiated, in response to an input from the player.

Abstract: Methods, systems, and computer programs for expanding gaming social networks with unrelated players are presented. For the player of an online game, one or more selection rules are determined defining a criterion for selecting users. The determined selection rules are applied resulting in a set of candidate game friends, where at least one of the candidate game friends does not have a social link with the player. Additionally, the method includes an operation for presenting the plurality of candidate game friends. An operation to link, in the online game, the player with at least one of the plurality of candidate game friends is then initiated in response to an input from the player. The operation to link may send a request to befriend the candidate, or to link the candidate without the need for permission when an automatic acceptance is in place.

Abstract: The APPARATUSES, METHODS AND SYSTEMS FOR A ONLINE GAME MANAGER (“OGM”) transform user social graph and online behavioral pattern data inputs via OGM components into optimized server computational load reports. In one embodiment, the OGM may process a user server migration request for a user by obtaining user social graph data identifying co-users, and social activity pattern data associated with the user and the co-users. The OGM may also obtain server load indices of servers associated with the co-users, and server load analysis rules. By analyzing the server load indices using the server load analysis rules, the OGM may identify a subset of the servers as non-overloaded transfer candidate servers. The OGM may calculate transfer candidate server scores for the non-overloaded transfer candidate servers using the user social graph data and the social activity pattern data.