U.S. Patent No. 9,616,345: Systems and methods for
crowd-sourced game strategy
Issued April 11, 2017, to Electronic Arts Inc.
Priority Date July 1, 2014
Priority Date July 1, 2014
Playing EA’s Madden can be daunting because of the number of
systems and strategies concurrently happening, especially for a person who does
not understand American Football. The playcall selection menu can very be
confusing to a novice who may not understand the differences between a running
play and a passing play. To help the player, Madden will provide suggested
playcalls based on the situation, but early incarnations of the suggestion
system were all based on pre-programmed suggestions. The game had no way of
learning a players habits or adjusting the suggestion based on the opponent. U.S. Patent No.9,616,345 addresses this problem by creating a system and method to collect
game play strategy data from online matches and use the data to adjust
suggestions accordingly. Now, Madden will provide you suggestions not only
based off your previous calls but also what you did in your previous matches as
well as how other players handled similar situations. The game will even show
you the number of times you have called a play and your success rate. The
patent also suggests that this data could be used to alter AI player strategy,
but the primary focus of the patent is on playcall suggestions.
Abstract:
Various embodiments provide systems and methods that collect
data regarding game strategy decisions by human players during video game, and
utilize the collected data to either adjust or replace behaviors of computer
players and/or suggest game strategies to human players during video game
sessions. The game strategy decisions may be harvested from human-vs-human and
human-vs-computer video game sessions. The data may be harvested from
online-connected video game sessions, which may be hosted over an online video
game network. Depending on the embodiment, the harvested data can include
information regarding game strategies used by players during the video game
sessions, the game contexts in which the game strategies were respectively
used, and the results achieved by the respective use of the game strategies.
Systems and methods described herein may facilitate a computer player having
behavior that is (at least partially) "crowd-sourced" based on game
strategies used by online-connected video game sessions.
Illustrative Claim:
1. A method comprising: as implemented by a processor
configured with computer-executable instructions, receiving first data from a
first client device, the first data regarding a first game strategy used, by a
first game player, in a first game context of a first video game session
between the first game player and at least a second game player, the first
video game session executing within a video game, the first game strategy being
a playcall; updating a game strategy dataset based on the first data, the first
data at least including an identity of the first game strategy and an
association between the first game context and the first game strategy, the
game strategy dataset including a set of one or more previous playcalls used in
a set of one or more previous game contexts; receiving gameplay data from a
second client device during a second gameplay session, the gameplay data
identifying a second game context of the second video game session, the second
video game session being between a third player and a fourth player; identifying
in the game strategy dataset a set of one or more relevant game strategies
based, at least in part, on the second game context during the second video
game session, wherein individual game strategies are playcalls available for
use within the second video game session; and transmitting to the second client
device second data regarding the set of one or more relevant game strategies,
the second client device being configured to use the second data to identify at
least one suggested game strategy of the set of one or more relevant game
strategies for use in the second game context of the second video game session,
and generate instructions to output the at least one suggested game strategy
within a video game user interface during the second video game session.
Researched By: Andrew F. Thomas
