A method and device for pushing game packages

Abstract

An embodiment of the present invention provides a method and device for pushing game packages. The method includes: pulling player information in a game running on the terminals of a player group; the player information includes social behaviors among player groups used to construct a social network; performing a clustering operation in the social network constructed based on the social behaviors to determine the game circle to which the player group belongs; determining the social potential information of the player group according to the belonging game circle, and locating target players with social potential from the player group according to the social potential information; obtaining discount information for game packages according to the social potential information of the target players, and pushing game packages to the corresponding terminals of the target players based on the discount information. By locating players with social potential and giving more emphasis on low prices to these players with social potential, while promoting the conversion of this part of players into socially mature players, the game stickiness of their surrounding players is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of games, and particularly to a game gift package pushing method and a game gift package pushing device. Background Art

[0002] The store in a game can provide goods that can meet the needs of players. The goods provided are different based on different game attributes. For example, it can provide weapon - type goods that can enhance the combat force value of virtual game characters, appearance clothing - type goods that meet the aesthetics of players, social - type goods that enhance the relationship between players, and so on.

[0003] In games with extremely strong social attributes such as language reasoning games, there are rich social behaviors among players. For the purpose of enhancing the social relationship among players and providing a large number of social - type goods at the same time, some of these goods may have the characteristic that they can only be given to others and cannot be used by the players themselves. This characteristic determines that most of such goods are consumables, that is, different from other goods (such as appearance clothing goods, etc.), players can purchase them multiple times. Such goods have potential mining value. However, there is currently no reasonable recommendation method that can meet the user's needs for the recommendation of social - type goods. Summary of the Invention

[0004] In view of the above problems, embodiments of the present invention are proposed to provide a game gift package pushing method and a corresponding game gift package pushing device that can overcome or at least partially solve the above problems.

[0005] Embodiments of the present invention disclose a game gift package pushing method, and the method includes:

[0006] Pulling player information in the game running on the terminals of the player group; the player information includes the social behaviors among the player groups used to construct a social network;

[0007] Performing a clustering operation in the social network constructed based on the social behaviors to determine the game circle to which the player group belongs;

[0008] Determining the social potential information of the player group according to the game circle to which it belongs, and positioning target players with social potential from the player group according to the social potential information;

[0009] Obtaining discount information for the game gift package according to the social potential information of the target players, and pushing the game gift package to the corresponding terminals of the target players based on the discount information.

[0010] Optionally, the social behaviors among the player groups include the social behaviors between different groups of players, and different social behaviors are set with different weights;

[0011] The construction method of the social network among the player groups is as follows:

[0012] Using the social behaviors among different groups of players and the corresponding weights of the social behaviors, calculate the social relationship weights among different groups of players in the player group;

[0013] Based on the social relationship weights among different groups of players, construct a social network for the player groups.

[0014] Optionally, performing a clustering operation in the social network constructed based on the social behaviors to determine the game circles to which the player groups belong, including:

[0015] Obtain a large-scale connected subgraph from the social network;

[0016] Perform a decomposition operation on the large-scale connected subgraph, and cluster the decomposed connected subgraphs to obtain the game circles to which the player groups belong.

[0017] Optionally, the social network has edge weights constructed based on the social relationship weights among player groups. The obtaining of the large-scale connected subgraph from the social network includes:

[0018] Filter out the edges with edge weights lower than the preset weight threshold in the social network, and filter the selected edges to obtain a connected subgraph;

[0019] Obtain the number of players in each connected subgraph, and determine the connected subgraph with the number of players greater than the preset quantity threshold as the large-scale connected subgraph.

[0020] Optionally, the large-scale connected subgraph contains multiple complete subgraphs. The performing of the decomposition operation on the large-scale connected subgraph and clustering the decomposed connected subgraphs to obtain the game circles to which the player groups belong includes:

[0021] Based on the common nodes, decompose each complete subgraph in the large-scale connected subgraph to obtain complete subgraphs with adjacent relationships;

[0022] Cluster the complete subgraphs with adjacent relationships into a subgraph set, and determine the subgraph set as the respective game circles to which the player groups belong; wherein, the same player belongs to at least one game circle.

[0023] Optionally, the social potential information includes a social potential score. The determining of the social potential information of the player groups according to the game circles to which they belong includes:

[0024] Obtain social metrics for a player group within their respective game circle, where the social metrics include betweenness centrality metrics, eigenvector centrality metrics, degree centrality metrics, and closeness centrality metrics for players within their respective game circle; among them, different social metrics are set with different weights;

[0025] Use the social metrics of the player within their respective game circle and the corresponding weights of the social metrics to calculate the social score of the player within their respective game circle;

[0026] Summarize the social scores to obtain the social potential score for the player.

[0027] Optionally, the social potential information includes the social potential score. Locating target players with social potential from the player group according to the social potential information includes:

[0028] Sort the players in the player group according to the social potential score in a preset order, and determine the players located at the head position of the social potential ranking as the target players with social potential.

[0029] Optionally, the discount information for the game package includes the discount value for the game package. Obtaining the discount information for the game package according to the social potential information of the target player, and pushing the game package to the corresponding terminal of the target player based on the discount information includes:

[0030] Obtain the dynamic discount range for the game package, and obtain the Gaussian distribution graph corresponding to the dynamic discount range;

[0031] Use the social potential score of the target player and the Gaussian distribution graph to determine the discount range interval for the game package and the discount value within the discount range interval;

[0032] Push the game package discounted based on the discount value to the corresponding terminal of the target player.

[0033] An embodiment of the present invention also discloses a game package pushing device, and the device includes:

[0034] A player information pulling module, configured to pull player information in the game running on the terminals of the player group; the player information includes the social behaviors among the player groups used to construct the social network;

[0035] A game circle determination module, configured to perform clustering operations in the social network constructed based on the social behaviors to determine the game circle to which the player group belongs;

[0036] A social potential information determination module, configured to determine the social potential information of the player group according to the game circle to which it belongs;

[0037] A target player positioning module, configured to locate target players with social potential from the player group according to the social potential information;

[0038] A game package push module, configured to obtain discount information for the game package according to the social potential information of the target player, and push the game package to the corresponding terminal of the target player based on the discount information.

[0039] Optionally, the device further includes:

[0040] A social network construction module, configured to construct a social network among the player groups;

[0041] The social behaviors among the player groups include social behaviors among different groups of players, where different social behaviors are set with different weights;

[0042] The social network construction module includes:

[0043] A social relationship weight calculation sub-module, configured to calculate the social relationship weights among different groups of players in the player group by using the social behaviors among different groups of players and the corresponding weights of the social behaviors;

[0044] A social network construction sub-module, configured to construct a social network among the player groups based on the social relationship weights among different groups of players.

[0045] Optionally, the game circle determination module includes:

[0046] A large-scale connected subgraph acquisition sub-module, configured to acquire a large-scale connected subgraph from the social network;

[0047] A game circle determination sub-module, configured to perform a decomposition operation on the large-scale connected subgraph, and cluster the decomposed connected subgraphs to obtain the game circles to which the player group belongs.

[0048] Optionally, the social network has edge weights constructed based on the social relationship weights among the player groups, and the large-scale connected subgraph acquisition sub-module includes:

[0049] A connected subgraph acquisition unit, configured to filter out the edges with edge weights lower than a preset weight threshold in the social network, and filter the filtered edges to obtain a connected subgraph;

[0050] A large-scale connected subgraph acquisition unit, configured to obtain the number of players in each connected subgraph, and determine the connected subgraph with the number of players greater than a preset quantity threshold as the large-scale connected subgraph.

[0051] Optionally, the large-scale connected subgraph includes a plurality of complete subgraphs, and the game circle determination sub-module includes:

[0052] A complete subgraph decomposition unit, configured to decompose each complete subgraph in the large-scale connected subgraph based on common nodes to obtain complete subgraphs with an adjacent relationship;

[0053] A subgraph set clustering unit, configured to cluster the complete subgraphs with an adjacent relationship into a subgraph set, and determine the subgraph set as each game circle to which the player group belongs; wherein, the same player belongs to at least one game circle.

[0054] Optionally, the social potential information includes a social potential score, and the social potential information determination module includes:

[0055] A social index determination sub-module, configured to obtain social indexes of the player group in the game circles to which they belong, where the social indexes include betweenness centrality indexes, eigenvector centrality indexes, point centrality indexes, and closeness centrality indexes of the players in the game circles to which they belong; wherein, different weights are set for different social indexes;

[0056] A social score calculation sub-module, configured to calculate the social score of the player in the game circle to which they belong by using the social indexes of the player in the game circle to which they belong and the corresponding weights of the social indexes;

[0057] A social potential score summarization sub-module, configured to summarize the social scores to obtain the social potential score for the player.

[0058] Optionally, the social potential information includes a social potential score, and the target player positioning module includes:

[0059] A target player positioning sub-module, configured to perform social potential ranking on the players in the player group in a preset order according to the social potential score, and determine the player located at the head position of the social potential ranking as the target player with social potential.

[0060] Optionally, the discount information of the game package includes a discount value for the game package, and the game package push module includes:

[0061] A Gaussian distribution sub-module, configured to obtain a dynamic discount interval for the game package, and obtain a Gaussian distribution diagram corresponding to the dynamic discount interval;

[0062] A discount value determination sub-module, configured to use the social potential score of the target player and the Gaussian distribution diagram to determine the discount range interval for the game package and the discount value within the discount range interval;

[0063] A game package push sub-module for pushing a game package discounted based on the discount value to the corresponding terminal of the target player.

[0064] An embodiment of the present invention also discloses an electronic device, including: a processor, a memory, and a computer program stored on the memory and capable of running on the processor. When the computer program is executed by the processor, the steps of any one of the game package push methods are implemented.

[0065] An embodiment of the present invention also discloses a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the game package push methods are implemented.

[0066] The embodiments of the present invention have the following advantages:

[0067] In the embodiments of the present invention, by pulling the player information in the game running on the terminals of the player group, clustering operations are performed in the social network constructed based on the existing social behaviors to determine the game circle to which the player group belongs and the social potential information of the game group determined based on the belonging game circle. Target players with social potential are located from the player group, and a game package discounted based on the discount information is pushed to the corresponding terminals of the target players, so as to improve the stickiness of the target players with social potential to the game, and at the same time be able to drive the enthusiasm of the surrounding players with social interactions with them for the game. That is, by positioning players with social potential through the social network constructed based on player social behaviors, and giving more emphasis on low prices to players with social potential, encouraging players to purchase social packages. While promoting this part of players to become socially mature players, it can improve the enthusiasm of their surrounding players for the game based on the social behaviors between players and other players, so as to improve the game stickiness of the surrounding players, and achieve a double improvement in social interaction among player groups and player retention in the application scenario of social packages. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 is a flowchart of the steps of a game package push method provided by an embodiment of the present invention;

[0069] Figure 2A-2B is a schematic diagram of the construction of a social network among player groups provided by an embodiment of the present invention;

[0070] Figure 3 is a flowchart of the steps of another game package push method provided by an embodiment of the present invention;

[0071] Figure 4 is a schematic diagram of a large-scale connected subgraph provided by an embodiment of the present invention;

[0072] Figure 5 It is a schematic diagram of the game circle to which the player group belongs provided by an embodiment of the present invention;

[0073] Figure 6 It is a schematic diagram of a complete subgraph provided by an embodiment of the present invention;

[0074] Figure 7 It is a schematic diagram of the process of clustering a set of subgraphs provided by an embodiment of the present invention;

[0075] Figure 8 It is an example diagram for calculating the social metrics of a player group in the game circle to which it belongs provided by an embodiment of the present invention;

[0076] Figure 9 It is a schematic diagram of a Gaussian distribution graph corresponding to a dynamic discount interval provided by an embodiment of the present invention;

[0077] Figure 10 It is an application scenario diagram of a game package push method provided by an embodiment of the present invention;

[0078] Figure 11 It is an implementation process diagram of positioning socially potential players provided by an embodiment of the present invention;

[0079] Figure 12 It is a structural block diagram of a game package push device provided by an embodiment of the present invention. Detailed implementation manners

[0080] To make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific implementation manners.

[0081] Generally, in games with extremely strong social attributes such as language reasoning games, there are rich social behaviors among players. For the purpose of enhancing the social relationship among players and providing a large number of social goods for players at the same time (which may have the characteristic that they can only be given to others and cannot be used by players themselves), most of these social goods are consumables and have potential mining value. If the demand of users / players for social goods can be reasonably grasped, this will be a sustainable recommendation project, such as recommending game packages to users / players. Among them, the relevant recommendation methods for game packages can include the form of daily special packages, which can improve ARPU (Average Revenue Per User, that is, the average revenue per user, an indicator that can be used to measure the operator's revenue. ARPU focuses on the revenue obtained by the operator from each user within a certain period). If it is assumed that social-related item combinations are customized into social packages and players with social potential are discovered, and exclusive customized packages with extremely strong social attributes are pushed to this group of players, then it is possible to achieve a double improvement in social interaction and player retention among player groups in the application scenario of social packages.

[0082] One of the core ideas of the embodiments of the present invention is to propose a customized package recommendation method based on a social potential player positioning system. It mainly locates players with social potential through a social network constructed based on players' social behaviors, and gives more emphasis on low prices for players with social potential, encouraging players to purchase social packages. While promoting this part of players to become socially mature players, it can improve the enthusiasm of their surrounding players for the game based on the social behaviors between players and other players, so as to improve the game stickiness of surrounding players and achieve a double improvement in social interaction and player retention among player groups in the application scenario of social packages.

[0083] Refer to Figure 1 , which shows the step flowchart of a game package push method provided by the embodiments of the present invention, mainly focusing on the construction process of the social network, and can specifically include the following steps:

[0084] Step 101, pull the player information in the game running on the player group terminal;

[0085] In the embodiments of the present invention, the proposed social potential player positioning system can pull the player information in the game running on the player group terminal, so that the social potential player positioning system can construct the social network among the player groups according to the pulled player information.

[0086] Among them, players with social potential can mainly be determined through the social behaviors carried out between the player and other players. Then, the player information pulled from the player group terminal can include the social behavior information between player groups, which can be used to construct a social network for player groups.

[0087] It should be noted that the social potential player positioning system can be embodied as a server that communicates with the player group terminal. This server can be a game-related server and has the function of positioning players with social potential. In this regard, the embodiments of the present invention are not limited.

[0088] Step 102, construct a social network for player groups.

[0089] In an embodiment of the present invention, in order to realize the push of game packages based on the social potential player positioning system, first, a social network for player groups can be constructed, so as to be able to locate target players with social potential from the player group and push discounted game packages to the corresponding terminals of the target players.

[0090] Specifically, a social network can be constructed based on the pulled social behavior information between player groups.

[0091] When initially constructing the social network, the social relationship weight (weight) between players can be calculated to construct an exclusive social network. Among them, the social behaviors between the pulled player groups can include the social behaviors between different groups of players, and different weights can be set for different social behaviors.

[0092] In practical applications, the social behaviors between different groups of players and the corresponding weights of the social behaviors can be used to calculate the social relationship weights between different groups of players in the player group, and then a social network for player groups can be constructed based on the social relationship weights between different groups of players.

[0093] In a specific implementation, when calculating the overall social relationship weight based on the social behaviors between different groups of players, the number of social behaviors carried out between different groups of players within the historical time period can be obtained respectively, and calculated with the corresponding weights.

[0094] In an alternative embodiment, the social behaviors carried out can be divided into private interactions such as private chatting and following, sect interactions such as sect activities and sect chatting, game interactions such as liking, following, and blocking, and gift-giving interactions, etc.; among them, the gift-giving interactions can also be divided into one-to-one gift-giving modes such as gift-giving between acquaintances, gift-giving to players in the same game session, and gift-giving to strangers, and one-to-many gift-giving modes such as in-game red envelopes and world red envelopes. When calculating the social relationship weight, weighted processing can also be carried out for the social behaviors related to gift-giving.

[0095] As an example, assume that a game has five types of social behaviors among players, namely private chat, follow, focus, block, and gift giving. The weights of these five social behaviors are 1, 10, 20, -20, and 100 respectively. At this time, by pulling the historical data in the game running on the player group terminal, it is known that player A and player B had 20 private chats, 1 follow, 1 focus, 1 block, and 1 gift giving in a certain historical period. Then, the overall social relationship weight calculated between player A and player B can be 20*1 + 10*1 + 20*1 - 20*1 + 100*1. Again, assume that there are five players A, B, C, D, and E in the game. Then, the social relationship weights between different groups of players, that is, 6 groups of players, can be calculated respectively, and a dedicated social network for the five players A, B, C, D, and E can be constructed based on the calculated social relationship weights.

[0096] It should be noted that the setting of the corresponding weights of social behaviors is related to the game business. That is, according to the different game businesses, there will be a certain tendency for the weights of specific social behaviors. When the game business and gift giving are highly relevant, the weight of the gift giving social behavior can be adaptively increased. Taking the above example, the weight set for the gift giving behavior is 100, which is much higher than the weight of the follow behavior, which is 10. The setting of the weights can be flexibly adjusted as needed. In this regard, the embodiments of the present invention do not impose any restrictions.

[0097] In the embodiments of the present invention, for a social network, the main thing is the construction of edges. Then, when constructing based on the social relationship weights between different groups of players, each player in the player group can be used as a node of the social network, and two nodes are connected based on the calculated social relationship weights (that is, constructing the edges in the social network). Finally, a network with social weights is obtained.

[0098] Specifically, referring to Figure 2A-2B , a schematic diagram of the construction of the social network for the player group provided by the embodiments of the present invention is shown. Assume that the social relationship weight between player A and player B is 2, the social relationship weight between player B and player C is 3, and the social relationship weight between player A and player C is 4. Then, the social network constructed based on the social relationship weights of the foregoing three groups of players can be as shown in Figure 2A and Figure 2B . The constructed social network is only a visual display of an abstract concept, and it only needs to be divided and connected according to the more prominent social behaviors.

[0099] It should be noted that there is no absolute position for players in the social network. For example, Figure 2A and Figure 2BThey represent the same social network (both are social networks for player group ABC), but the positions of the corresponding nodes of players ABC will not affect the construction of the social network, and weight(2, 3, 4) will not affect the distances between nodes ABC either. That is, as long as the weights between the relative positions of the nodes are ensured to be the same during construction, it can be recognized as the same social network, and there is no need to consider how to connect the nodes. In this regard, the embodiments of the present invention do not impose any restrictions.

[0100] In the embodiments of the present invention, social networks constructed based on players' social behaviors are used to locate players with social potential, and more emphasis is placed on these players with social potential at a lower price to encourage players to purchase social packages. While promoting the transformation of this part of players into socially mature players, it is possible to improve the enthusiasm of surrounding players for the game based on the social behaviors of players and other players, so as to increase the game stickiness of surrounding players, and achieve a double improvement in social interaction among player groups and player retention in the application scenario of social packages.

[0101] Referring to Figure 3 , a flowchart of steps of another game package pushing method provided by the embodiments of the present invention is shown, which mainly focuses on the usage process of the constructed social network, and specifically may include the following steps:

[0102] Step 301, pull the player information in the game running on the terminals of the player group and construct a social network;

[0103] In an embodiment of the present invention, in order to push game packages based on the social potential player positioning system, first, the social network among the player group can be constructed so as to locate target players with social potential from the player group and push discounted game packages to the corresponding terminals of the target players.

[0104] Among them, a dedicated social network can be constructed by inducing the social behaviors among players, that is, a social network can be constructed based on the social behavior information among the pulled player group. Specifically, the edges can be constructed based on the calculated social relationship weights between different groups of players, and finally a network with social weights is obtained.

[0105] Step 302, perform a clustering operation in the social network constructed based on social behaviors to determine the game circles to which the player group belongs;

[0106] After constructing the dedicated social network by inducing the edges through the social behaviors among players, a clustering operation can be performed on the social network to achieve the clustering of the game circles among complex player groups and determine the game circles to which each player in the player group belongs.

[0107] Specifically, in the process of clustering the game circles among complex player groups, first, the largest connected subgraph can be obtained from the constructed social network, and then the large-scale connected subgraph is decomposed. The decomposed connected subgraphs are clustered to obtain the game circles to which the player groups belong.

[0108] Among them, in the process of obtaining the large-scale connected subgraph, the constructed social network mainly constructs edges. In the process of edge construction, the social relationship weights (i.e., edge weights) among player groups can be included. At this time, invalid social interactions can be excluded to retain the valid social relationship weights, that is, the invalid edges included in the social network can be filtered based on the edge weights.

[0109] In an embodiment of the present invention, since the social network will include edges with low social weights, at this time, the edges with edge weights lower than the preset weight threshold are screened out in the social network, and the screened edges are filtered to obtain all connected subgraphs. At this time, the relatively complex large-scale connected subgraphs can also be screened out according to the number of players in each connected subgraph, that is, the number of players in each connected subgraph can be obtained, and then the connected subgraphs with the number of players greater than the preset quantity threshold (for example, the number of players > 100 / 500 / 1000) are determined as large-scale connected subgraphs, so as to improve the system performance and calculation efficiency when positioning players with social potential based on the social network.

[0110] In practical applications, referring to Figure 4 , a schematic diagram of the large-scale connected subgraph provided by the embodiment of the present invention is shown. The social network after retaining the valid social relationship weights can include large-scale connected subgraphs and small-scale connected subgraphs. When determining the game circles to which the player groups belong, the k_clique percolation algorithm can be used to decompose the obtained large-scale connected subgraphs to cluster the game circles to which each player belongs. For small-scale connected subgraphs, that is, the connected subgraphs with the number of players not reaching the preset quantity threshold, there is no need to further process them, and they can be regarded as a game circle by themselves.

[0111] In specific implementation, in the process of decomposing and clustering the large-scale connected subgraph, the k_clique percolation algorithm can be used to decompose the obtained large-scale connected subgraph. The k_clique percolation algorithm believes that a community is a set of fully connected subgraphs with shared nodes, and a clique filtering algorithm can be used to identify the community structure in the constructed social network. The identified community structure is the game circle among complex player groups as shown in Figure 5 Figure.

[0112] Specifically, the k_clique percolation algorithm can be used to decompose each complete subgraph in a large-scale connected subgraph based on common nodes, obtaining complete subgraphs with adjacent relationships. Then, the complete subgraphs with adjacent relationships are clustered into a set of subgraphs, and the set of subgraphs is determined as the respective game circles to which the player groups belong.

[0113] Among them, the large-scale connected subgraph contains multiple complete subgraphs. For example, Figure 6 as shown, for a graph, if there exists a subgraph where there is an edge between any two nodes, it can be called a complete subgraph. If the number of its nodes is k, then this complete subgraph can be called a k-clique. For example, Figure 6 the 3-clique subgraph, 4-clique subgraph, and 5-clique subgraph in

[0114] Furthermore, after searching for all complete subgraphs with k nodes, a new graph with k-cliques as nodes can be established. Referring to Figure 7 , it shows a schematic diagram of the process of clustering the set of subgraphs provided by an embodiment of the present invention. When using the k_clique percolation algorithm to decompose the complete subgraph with k nodes, if there are k - 1 common nodes between two k-cliques, then an edge can be established between the nodes representing them in the new graph, and these two cliques can be said to be "adjacent", that is, these two complete subgraphs have an adjacent relationship. And the game circle can refer to the set of subgraphs composed of complete subgraphs with adjacent relationships, that is, a string of adjacent cliques forms the largest set. Then, finally, each connected subgraph in the new graph is a community (i.e., a game circle).

[0115] As an example, as Figure 7 shown, assuming that the weights of all edges in the constructed social network are 1, at this time, the 3-cliques algorithm can be used to process the player group. Among them, the subgraph within the first circle can be a 3-cliques, and there are (k - 1), that is, 2 nodes that coincide with the 3-cliques within the range of the second circle, indicating that these two 3-cliques have an adjacent relationship with each other. By deducing other complete subgraphs in this way, the 3-cliques within the range of the second circle and the third circle are also two subgraphs with an adjacent relationship. Then, at this time, a string of cliques with adjacent relationships to each other can form the largest set, that is, the game circle is obtained. Among them, the game circles to which the same player belongs can be at least one game circle, that is, a player can appear in multiple game circles at the same time. For example, Figure 7 the solid node in

[0116] It should be noted that when clustering game circles, factors such as the weights of different edges and the parameter selection of k-cliques need to be considered according to the actual situation. Some players may also exist in more than two game circles. In this regard, the embodiments of the present invention do not impose any restrictions.

[0117] Step 303: Determine the social potential information of the player group according to the affiliated game circle, and locate the target players with social potential from the player group according to the social potential information.

[0118] In the embodiments of the present invention, after clustering the game circles among complex player groups is achieved, the social potential information of the player group can be determined according to the affiliated game circle, so as to locate the target players with social potential.

[0119] Specifically, multiple social metrics such as betweenness centrality can be used to measure the social potential of players. Among them, the social potential information may include social potential scores. At this time, social metrics can be used to determine the social potential scores of the player group.

[0120] In an embodiment of the present invention, the social metrics of a player group in its affiliated game circle may include the betweenness centrality index, eigenvector centrality index, degree centrality index, and closeness centrality index of the player in the affiliated game circle. Specifically, the betweenness centrality can be used as the core index, and the eigenvector centrality, degree centrality, and closeness centrality can be used as auxiliary indexes to comprehensively calculate the social potential score of the player.

[0121] In practical applications, when a player appears in multiple game circles at the same time, the superposition of their social potential will be relatively high, indicating that this player is very likely to be a player with social potential and has the potential to drive social behaviors in multiple circles. This player should be determined as the key push object of the social package. For example, if a player has two circles: the primary school classmate circle and the university classmate circle, this player generally does not play with primary school classmates and university classmates at the same time, that is, their primary school classmates and university classmates usually do not have social behaviors. Then, when calculating the social potential score of the player, it should be calculated separately in the primary school and university circles and then superimposed.

[0122] In a specific implementation, when calculating the social potential scores of different game circles to which a player belongs, different weights can be set for different social metrics. The weights corresponding to the social metrics and social indicators of the player in the game circle to which the player belongs are used to calculate the social score of the player in the game circle to which the player belongs, and then the social scores are aggregated to obtain the social potential score for the player. Among them, as a core metric, betweenness centrality can be set with more social potential weights, that is, its weight is higher than that of the auxiliary metrics.

[0123] Exemplarily, assume that the social metrics of a player group in the game circle to which the player belongs have four metrics A, B, C, and D, which can be betweenness centrality as the core metric, eigenvector centrality, degree centrality, and closeness centrality respectively. Among them, based on betweenness centrality as the core metric and eigenvector centrality, degree centrality, and closeness centrality as auxiliary metrics, the corresponding weights can be set to 70%, 10%, 10%, and 10% respectively. Then, at this time, the social score of a certain player in a certain game circle can be 70% * A + 10% * (B + C + D); after calculating the social scores of the player in each game circle to which the player belongs, the social scores can be aggregated to obtain the social potential score for the player.

[0124] Betweenness centrality, as a core metric, can describe the position of a certain node in the social network, determine whether it can control two nodes that do not have direct connectivity, and can be used to reflect the degree of correlation between players, that is, the tightness or connection strength between players can be known through this metric. Mainly for all the shortest paths between any two nodes in the social network, if many of these shortest paths pass through a certain node, then it is considered that the betweenness centrality of this node is high, and specifically, it can be manifested as a high degree value of this node.

[0125] The degree value of each node n can be determined based on the ratio of the shortest paths passing through a certain point (n) and connecting two points (s, t) to the total number of shortest path lines between these two points. The specific calculation formula can be as follows:

[0126] C b (n) = ∑ s≠n≠t (σ st (n) / σ st )

[0127] Among them, s and t are nodes different from n in the social network, σst is used to represent the number of shortest paths from s to t, and σst(n) is the number of times the node n is located in the shortest path from s to t.

[0128] In a general explanation, if a player (i.e., a node) often appears in the shortest distance paths among other players (i.e., the shortest distance paths often include this player), it can be shown that this player has more potential to facilitate communication among other players. At the same time, such players are the most sensitive to social behaviors and are also easily influenced by social behaviors (such as giving or receiving social packages).

[0129] In a specific implementation, the calculation result of the above formula can also be normalized. Specifically, the calculated betweenness centrality index can be divided by the pairwise permutations and combinations of the nodes other than n, so that the value of betweenness centrality is between 0 and 1.

[0130] The degree value of each node n is normalized by dividing it by the pairwise permutations and combinations of the nodes other than n. The specific formula is as follows: (N - 1)(N - 2) / 2, where N can be the total number of connected nodes in the game circle to which n belongs, so that the betweenness of each node is a value between 0 and 1.

[0131] As an example, as Figure 8 shown, the formula for normalizing the betweenness centrality index of point b can be:

[0132] C b (b) = (∑ a≠b≠c≠d≠e (σ acde (b) / σ acde )) / ((N - 1)(N - 2) / 2) = ((σ ac (b) / σ ac ) + (σ ad (b) / σ ad ) + (σ ae (b) / σ ae ) + (σ cd (b) / σ cd ) + (σ ce (b) / σ ce ) + (σ de (b) / σ de )) / ((5 - 1)(5 - 2) / 2) = ((1 / 1) + (1 / 1) + (2 / 2) + (1 / 2) + 0 + 0) / 6 ≈ 0.583.

[0133] In addition, for the eigenvector centrality as an auxiliary metric, its basic idea is that the centrality of a certain node is a function of the centralities of adjacent nodes, which can emphasize the value of a player within their game circle. It can be determined by the social capabilities of the friends they make, that is, it indicates that the more important the social capabilities of the players directly connected to this player in the social network, the more important this player is. It should be noted that even if a certain node has a high degree centrality, that is, there are many nodes with connection relationships, when the eigenvector centrality of the connected nodes may be very low, the eigenvector centrality of this player may also be very low; the degree centrality as an auxiliary metric can be understood as the number of in-game friends of this player, which can be manifested in the constructed social network as the number of nodes directly connected to a certain node. For example, Figure 8 the degree centrality metric of node b in

[0134] Figure 8

[0135] In a preferred embodiment, when determining the target players with social potential, the players in the player group can be sorted according to their social potential scores in a preset order, and the players located at the head position of the social potential ranking are determined as the target players with social potential.

[0136] It should be noted that the head position can refer to the range in the ranking, that is, the determined target players can include multiple players, and its range can be determined according to the actual situation. In this regard, the embodiments of the present invention do not impose any restrictions.

[0137] In an embodiment of the present invention, after identifying the core users affected by the circle algorithm, that is, determining the target players with social potential, discounted game packages can be pushed to the corresponding terminals of the target players, achieving a double improvement in social interaction and player retention among the player group in the application scenario of social packages.

[0138] Specifically, to push the discounted game package to the corresponding terminal of the target player, it is first necessary to determine the discount information of the game package, that is, the discount value for the game package can be determined. At this time, the dynamic discount range for the game package can be obtained, and the Gaussian distribution graph corresponding to the dynamic discount range can be obtained. Then, the social potential score of the target player and the Gaussian distribution graph are used to determine the discount range interval for the game package and the discount value within the discount range interval, so as to push the game package discounted based on the discount value to the corresponding terminal of the target player, implementing a push strategy that focuses on giving more low-price game packages to the top players with social potential.

[0139] In practical applications, referring to Figure 9 , a schematic diagram of the Gaussian distribution graph corresponding to the dynamic discount range provided by the embodiment of the present invention is shown. Specifically, the Mean and SD (standard deviation) of the dynamic discount Gaussian distribution can be determined through the dynamic discount range given by the planner, and then, according to the ranking of the social potential scores of the players, the discount range interval and the final discount value that can be given to this target player can be determined, and the object selection algorithm model is used for targeted object selection.

[0140] In the specific implementation, the method for determining the discount range interval and the final discount value can be through the normal distribution function, with the input parameters being Mean and SD, and the result output each time being the result that conforms to this distribution. By analogy with throwing a dice, assuming that its input parameters stipulate an equal probability distribution of 1-6, and the result output each time is 1-6, then the results of throwing the dice in the long term conform to the equal probability distribution of 1-6. The same is true for the normal distribution. If the result after multiple trials based on the result output each time can meet the preset result, then the discount value given by each independent implementation should be unique.

[0141] It should be noted that when using the object selection algorithm model for targeted object selection, it can include general features (such as active features (determined based on the activity of total online duration, total login days, online duration in the recent week / month, login days in the recent week / month, and the game level dimension based on character level, total number of games, number of wins, MVP, and honor value), payment features (determined based on the payment dimension of cumulative payment, payment in the recent week / month, and the token dimension based on the consumption of diamonds or vouchers or gold in the recent week / month, and the remaining amount of diamonds or vouchers or gold), and social features (determined based on the social capital dimension of charm value, number of fans, and lit titles, and the social relationship dimension based on the social circle, friends / intimate friends / CP / master-apprentice, and sect positions, as well as the social index dimension), etc.) and purchase motivation features (such as consumption preferences (based on consumption types, holdings, gift price preferences, etc.), pursuit of fun (such as pursuit of titles, pursuit of sets, social initiative, sect activity, etc.), recent behaviors (such as level improvement, honor value improvement, new social relationships, improvement of intimacy, progress of gift titles, progress of set titles, participation in popular rooms, consumption of related items, etc.)), but the item selection and discount concessions are independent and irrelevant. Regardless of the content of the game package, for any gift, players with high social potential can be given a lower discount. For social packages, additional relevant parameters of social indicators can be added to ensure more accurate delivery of social packages.

[0142] In the embodiment of the present invention, by pulling the player information in the game running on the player group's terminal, clustering operations are performed in the social network constructed based on the existing social behaviors to determine the game circle to which the player group belongs and the social potential information of the game group determined based on the belonging game circle. Target players with social potential are located from the player group, and game packages discounted based on discount information are pushed to the corresponding terminals of the target players to improve the stickiness of the target players with social potential to the game, and at the same time, it can drive the enthusiasm of the surrounding players with social interactions with them for the game. That is, by positioning players with social potential through the social network constructed based on player social behaviors, and giving more emphasis on low prices to players with social potential, encouraging players to purchase social packages, while promoting this part of players to become socially mature players, it can improve the enthusiasm of their surrounding players for the game based on the social behaviors between players and other players, so as to improve the stickiness of the surrounding players to the game, and achieve a double improvement in social interaction and player retention among player groups in the application scenario of social packages.

[0143] To facilitate further understanding by those skilled in the art of the game package pushing method proposed in the embodiment of the present invention, the following description is made in combination with the application scenario of the game package pushing method and the implementation process of positioning socially potential players:

[0144] In addition to how to locate socially potential players, another key aspect of the push game package proposed in the embodiments of the present invention lies in how to promote the conversion of such socially potential players into socially mature players for this part of players, and it is not limited to using betweenness centrality to measure the social potential of players.

[0145] As Figure 10 shown, the player group in the game can be divided into three categories based on social activity: (1) Socially mature players, who have high consumption ability, fixed social circles, rich overall social behaviors, stable retention, and strong game stickiness; (2) Socially potential players, who have consumption ability, a high proportion of social consumption in total consumption, a high frequency of social behaviors per unit time, and a certain degree of game stickiness; (3) Socially insensitive players: referring to a part of the player group with low social desire, low consumption desire, and low game stickiness.

[0146] Specifically, a social potential player positioning system based on betweenness centrality and k_clique percolation algorithm constructs an exclusive social network by summarizing the social behaviors among players, uses the k_clique percolation algorithm to decompose large-scale connected subgraphs to achieve game circle clustering among complex player groups, then uses multiple social indicators such as betweenness centrality to measure the social potential of players, demarcates the core users affected by the algorithm, locates the key groups that the social package focuses on, and accurately delivers social products to socially sensitive potential players with dynamic discount concessions through scientific and effective social algorithms, encourages players to purchase social packages, promotes the conversion of the socially potential player population into socially mature players, radiates the game stickiness of surrounding players to drive the game stickiness of the entire social circle, and realizes the double improvement of social interaction and player retention among player groups in the application scenario of the social package.

[0147] Among them, referring to Figure 11, which shows the implementation process diagram of the social potential player positioning provided by the embodiments of the present invention. Assuming that the social relationships of the player group in the game are simulated as a traffic network, when positioning social potential players, it is first necessary to preliminarily construct a global social network, mainly by inducing the social behaviors among players to calculate the social relationship weights between players for constructing the edges in the social network; then it is necessary to exclude the invalid social interactions in the constructed edges to retain the valid social relationship weights and obtain all connected subgraphs, such as the three connected subgraphs of Guangzhou, Jiangmen, and Zhongshan; then use the k_clique percolation algorithm to decompose the large-scale connected subgraphs to determine the game circles to which each player belongs. For example, decompose the large-scale connected subgraph of Guangzhou into three circles of Yuexiu, Tianhe, and Huangpu, while Jiangmen and Zhongshan, as small-scale connected subgraphs, can be directly regarded as one circle; finally, the potential of players can be evaluated by calculating the social potential of the player group, mainly by traversing all the game circles of the players and calculating the social potential indicators (such as betweenness centrality, etc.) of the players in each circle respectively, and comprehensively obtaining the social potential value of the players.

[0148] When giving dynamic discount concessions to players with social potential, sort the players according to their social potential. Give more emphasis on low prices to the top players in social potential, encourage players to purchase social packages, so as to reach more players, improve payment and activity, and enhance the game stickiness of surrounding players. Among them, through the Gaussian distribution of the dynamic discount range, determine the discount range and the final value according to the social potential score of the player. Exemplarily, the scope of action of the low-price package (that is, the game package pushed) can also be restricted. For example, a certain package contains 3 low-price title gifts. While giving concessions to players, it can be set that players can only give them to 3 different players first. The embodiments of the present invention do not limit this.

[0149] It should be noted that using this social algorithm system can achieve the common improvement of retention and revenue in multiple verification cycles, and the difference value of the improvement has passed the T-test verification of the hypothesis test, with a reliable confidence level.

[0150] In the embodiments of the present invention, social potential players are positioned through the social network constructed based on the social behaviors of players, and more emphasis is placed on low prices for players with social potential, encouraging players to purchase social packages. While promoting the transformation of this part of players into socially mature players, it can improve the enthusiasm of surrounding players for the game based on the social behaviors between players and other players, so as to improve the game stickiness of surrounding players, and achieve the dual improvement of social interaction and player retention among player groups in the application scenario of social packages.

[0151] It should be noted that for the method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should be aware that the embodiments of the present invention are not limited by the described action sequences, because according to the embodiments of the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential for the embodiments of the present invention.

[0152] Referring to Figure 12 , a structural block diagram of a game gift package pushing device provided by an embodiment of the present invention is shown, which may specifically include the following modules:

[0153] A player information pulling module 1201, configured to pull player information in the game running on the terminals of the player group; the player information includes the social behaviors among the player groups for constructing a social network;

[0154] A game circle determining module 1202, configured to perform a clustering operation in the social network constructed based on the social behaviors to determine the game circle to which the player group belongs;

[0155] A social potential information determining module 1203, configured to determine the social potential information of the player group according to the game circle to which it belongs;

[0156] A target player positioning module 1204, configured to locate target players with social potential from the player group according to the social potential information;

[0157] A game gift package pushing module 1205, configured to obtain discount information for the game gift package according to the social potential information of the target player, and push the game gift package to the corresponding terminals of the target players based on the discount information.

[0158] In an embodiment of the present invention, the device may further include the following modules:

[0159] A social network construction module, configured to construct a social network for the player groups;

[0160] In an embodiment of the present invention, the social behaviors among the player groups include the social behaviors among different groups of players, and different weights are set for different social behaviors; the social network construction module may include the following sub-modules:

[0161] A social relationship weight calculation sub-module, configured to calculate the social relationship weights among different groups of players in the player group by using the social behaviors among different groups of players and the corresponding weights of the social behaviors;

[0162] A social network construction sub-module, configured to construct a social network for the player groups based on the social relationship weights among different groups of players.

[0163] In an embodiment of the present invention, the game circle determination module 1202 may include the following sub-modules:

[0164] A large-scale connected subgraph acquisition sub-module, configured to acquire a large-scale connected subgraph from the social network;

[0165] A game circle determination sub-module, configured to perform a decomposition operation on the large-scale connected subgraph, and cluster the decomposed connected subgraphs to obtain the game circles to which the player groups belong.

[0166] In an embodiment of the present invention, the social network has edge weights constructed based on the social relationship weights among player groups. The large-scale connected subgraph acquisition sub-module may include the following units:

[0167] A connected subgraph acquisition unit, configured to screen out the edges with edge weights lower than a preset weight threshold in the social network, and filter the screened-out edges to obtain a connected subgraph;

[0168] A large-scale connected subgraph acquisition unit, configured to obtain the number of players in each connected subgraph, and determine the connected subgraph with the number of players greater than a preset quantity threshold as a large-scale connected subgraph.

[0169] In an embodiment of the present invention, the large-scale connected subgraph contains multiple complete subgraphs. The game circle determination sub-module may include the following units:

[0170] A complete subgraph decomposition unit, configured to decompose each complete subgraph in the large-scale connected subgraph based on common nodes to obtain complete subgraphs with an adjacent relationship;

[0171] A subgraph set clustering unit, configured to cluster the complete subgraphs with the adjacent relationship into a subgraph set, and determine the subgraph set as the respective game circles to which the player groups belong; wherein, the same player belongs to at least one game circle.

[0172] In an embodiment of the present invention, the social potential information includes a social potential score. The social potential information determination module 1203 may include the following sub-modules:

[0173] A social index determination sub-module, configured to acquire social indexes of a player group in the game circles to which it belongs. The social indexes include betweenness centrality indexes, eigenvector centrality indexes, degree centrality indexes, and closeness centrality indexes of a player in the game circles to which it belongs; wherein, different social indexes are set with different weights;

[0174] A social score calculation sub-module, which is used to calculate the social score of the player in the game circle to which the player belongs by using the weights corresponding to the social indicators and social indicators of the player in the game circle to which the player belongs;

[0175] A social potential score summarization sub-module, which is used to summarize the social scores to obtain the social potential score for the player.

[0176] In an embodiment of the present invention, the social potential information includes a social potential score, and the target player positioning module 1204 may include the following sub-modules:

[0177] A target player positioning sub-module, which is used to sort the players in the player group according to the social potential in a preset order according to the social potential score, and determine the player located at the head position of the social potential ranking as the target player with social potential.

[0178] In an embodiment of the present invention, the discount information of the game package includes the discount value for the game package, and the game package push module 1205 may include the following sub-modules:

[0179] A Gaussian distribution sub-module, which is used to obtain the dynamic discount interval for the game package and obtain the Gaussian distribution diagram corresponding to the dynamic discount interval;

[0180] A discount value determination sub-module, which is used to use the social potential score of the target player and the Gaussian distribution diagram to determine the discount range interval for the game package and the discount value within the discount range interval;

[0181] A game package push sub-module, which is used to push the game package discounted based on the discount value to the corresponding terminal of the target player.

[0182] For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For the relevant parts, please refer to the partial description of the method embodiment.

[0183] The embodiment of the present invention also provides an electronic device, including:

[0184] It includes a processor, a memory, and a computer program stored on the memory and capable of running on the processor. When the computer program is executed by the processor, it realizes each process of the above-mentioned game package push method embodiment and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0185] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements each process of the above embodiment of the game gift package pushing method and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0186] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other.

[0187] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a device, or a computer program product. Therefore, the embodiments of the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0188] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the method, terminal device (system), and computer program product according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal devices generate a device for implementing the specified function in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0189] These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing terminal devices to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device implements the specified function in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0190] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal devices, so that a series of operation steps are executed on the computer or other programmable terminal devices to generate a computer-implemented process. Therefore, the instructions executed on the computer or other programmable terminal devices provide for implementing the specified function in Figure 1Steps of the functions specified in one or more processes and / or boxes Figure 1 or in one or more boxes.

[0191] Although the preferred embodiments of the embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present invention.

[0192] Finally, it should also be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or terminal device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or terminal device comprising the element.

[0193] The above provides a detailed introduction to a game gift package pushing method and a game gift package pushing device provided by the present invention. Specific examples are used in this article to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A method for pushing game gift packages, characterized in that, The method includes: Pulling player information in the game running on the terminals of the player group; the player information includes the social behaviors among the player groups for constructing a social network; Performing a clustering operation in the social network constructed based on the social behaviors to determine the game circles to which the player groups belong; Determining the social potential information of the player group according to the game circles to which it belongs, and positioning target players with social potential from the player group according to the social potential information; Obtaining the dynamic discount range for the game package and obtaining the Gaussian distribution graph corresponding to the dynamic discount range; Using the social potential score of the target player and the Gaussian distribution graph to determine the discount range for the game package and the discount value within the discount range; Pushing the game package discounted based on the discount value to the corresponding terminals of the target players.

2. The method according to claim 1, characterized in that The social behaviors among the player groups include the social behaviors among different groups of players, and different weights are set for different social behaviors; The construction method for the social network among the player groups is as follows: Calculating the social relationship weights among different groups of players in the player group using the social behaviors among different groups of players and the corresponding weights of the social behaviors; Constructing the social network among the player groups based on the social relationship weights among different groups of players.

3. The method according to claim 1, characterized in that The performing a clustering operation in the social network constructed based on the social behaviors to determine the game circles to which the player groups belong includes: Obtaining a large-scale connected subgraph from the social network; Performing a decomposition operation on the large-scale connected subgraph, and clustering the decomposed connected subgraphs to obtain the game circles to which the player groups belong.

4. The method according to claim 3, characterized in that The social network has edge weights constructed based on the social relationship weights among the player groups. The obtaining a large-scale connected subgraph from the social network includes: Filtering out the edges with edge weights lower than the preset weight threshold in the social network, and filtering the filtered edges to obtain a connected subgraph; Obtaining the number of players in each connected subgraph, and determining the connected subgraph with the number of players greater than the preset quantity threshold as the large-scale connected subgraph.

5. The method according to claim 3, characterized in that: The large-scale connected subgraph contains multiple complete subgraphs. The performing a decomposition operation on the large-scale connected subgraph and clustering the decomposed connected subgraphs to obtain the game circles to which the player groups belong includes: Decomposing each complete subgraph in the large-scale connected subgraph based on common nodes to obtain complete subgraphs with adjacent relationships; Clustering into a set of subgraphs based on the complete subgraphs with adjacent relationships, and determining the set of subgraphs as the respective game circles to which the player groups belong; where the same player belongs to at least one game circle.

6. The method according to claim 1, 2 or 3, characterized in that: The social potential information includes a social potential score. The determining the social potential information of the player group according to the game circles to which it belongs includes: Obtaining the social metrics of the player group in the game circles to which it belongs, where the social metrics include the betweenness centrality metric, eigenvector centrality metric, degree centrality metric, and closeness centrality metric of the player in the game circles to which it belongs; different weights are set for different social metrics; Calculate the social score of the player in the game circle by using the player's social indicators and the corresponding weights of the social indicators in the game circle; The social scores are aggregated to obtain a social potential score for the player.

7. The method according to claim 1 or 2 or 3, characterized in that, The social potential information includes a social potential score, and locating a target player with social potential from the player group according to the social potential information includes: The players in the player group are ranked according to their social potential scores in a preset order, and the players at the head of the social potential ranking are determined as target players with social potential.

8. A game gift package push device, characterized in that, The device comprises: A player information extraction module is used to extract player information from games running on player group terminals; the player information includes social behaviors between player groups used to build a social network; A game circle determination module is used to perform a clustering operation in the social network constructed based on the social behavior to determine the game circle to which the player group belongs; A social potential information determination module, configured to determine the social potential information of the player group based on the game circles to which they belong; a target player positioning module, configured to locate target players with social potential from the player group based on the social potential information; A Gaussian distribution submodule, configured to obtain a dynamic discount range for the game gift package and obtain a Gaussian distribution graph corresponding to the dynamic discount range; a discount value determination submodule, configured to determine a discount range for the game gift package and a discount value within the discount range using the target player's social potential score and the Gaussian distribution graph; The game gift package pushing submodule is used to push the game gift package discounted based on the discount value to the corresponding terminal of the target player.

9. An electronic device, characterized in that, include: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein when the computer program is executed by the processor, the steps of the game gift package pushing method as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the game gift package pushing method as described in any one of claims 1 to 7 are implemented.

Images