Personalized information push method, device, storage medium and electronic device

Abstract

This specification discloses a personalized information push method, device, storage medium and electronic device. For each user to whom information is to be pushed, the game character information and historical event information of the user are obtained, and the historical operation data of the user on the game operation object in the game is determined based on the historical event information. The game style of the user is determined based on the game character information and the historical operation data. The user portrait of the user is determined based on the user information, game character information and game style of the user. According to the user portrait, user groups are divided, and the information pushed to at least one user group is determined. In this method, when determining the user portrait of the user, the game style and game character information are taken into account, which can reflect the user's personal game behavior, increase the type of data used for user portraits, and the user groups divided based on user portraits are more accurate, so that accurate personalized push can be performed according to the characteristics of the user groups.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a personalized information push method, device, storage medium and electronic device. Background Art

[0002] With the development of the Internet, the types of information are becoming more and more complicated. Users need to spend a lot of time and energy to find the information they need, and it is difficult to find the most suitable products, services or content for themselves. Personalized information push, through artificial intelligence and big data, data analysis based on user characteristics, can accurately determine user needs and improve user experience.

[0003] In the field of games, due to the complexity of the game's own content and gameplay, more diverse types of data will be generated in the game. However, the existing information push method does not fully utilize the diversity of data types generated by the game, and the types of data used for data analysis are not rich enough, resulting in inaccurate information push. Therefore, this specification provides a personalized information push method. Summary of the invention

[0004] The present specification provides a personalized information push method, device, storage medium and electronic device to at least partially solve the above-mentioned problems existing in the prior art.

[0005] This manual adopts the following technical solutions:

[0006] This specification provides a personalized information push method, including:

[0007] Determine the users to whom information is to be pushed;

[0008] For each user to whom information is to be pushed, obtain the user's game character information and the user's historical event information;

[0009] Determine historical operation data of the user on the game operation object in the game according to the historical event information;

[0010] Determining the game style of the user according to the game character information and the historical operation data;

[0011] Determine a user profile of the user based on the user information, game character information, and the game style of the user;

[0012] According to the user portraits of the users to whom the information is to be pushed, the users to whom the information is to be pushed are divided into groups, and for at least one of the divided user groups, according to the gaming styles of the users in the user group, the information to be pushed to the users in the user group is determined to push the information.

[0013] Optionally, determining the historical operation data of the user on the game operation object in the game according to the historical event information specifically includes:

[0014] Determine each game operation object included in the historical event information;

[0015] For each game operation object, determine each sub-event corresponding to the game operation object;

[0016] Sorting the sub-events according to the timestamps of the sub-events to determine the event sequence of the game operation object;

[0017] According to the event sequence of each game operation object, the historical operation data of the user on the game operation object in the game is determined.

[0018] Optionally, determining the game style of the user according to the game character information and the historical operation data specifically includes:

[0019] According to the game character information, determining the user's frequently used character among the game characters used by the user;

[0020] Determine the user's frequently used type according to the type of each game character used by the user;

[0021] Determining the role style of the user according to the common roles and the common types;

[0022] The game style of the user is determined according to the character style and the historical operation data.

[0023] Optionally, determining the game style of the user according to the character style and the historical operation data specifically includes:

[0024] Determining a scan length according to the number of sub-events included in the event sequence of each game operation object;

[0025] Perform multiple rounds of scanning on each event sequence in descending order of the scanning length;

[0026] For each round of scanning process, according to the scanning length of the round of scanning process, a sequence segment corresponding to the scanning length is determined in each event sequence;

[0027] Determining a sequence segment set consisting of the same sequence segments according to the names of the sub-events included in the sequence segments and the order of the sub-events included in the sequence segments;

[0028] If there is a sequence segment set including a number of sequence segments greater than a preset threshold value among the sequence segment sets, determining the behavior style of the user according to the sequence segment corresponding to the sequence segment set including a number of sequence segments greater than the threshold value, and determining the game style of the user according to the game character information and the behavior style;

[0029] If there is no sequence segment set in each sequence segment set that contains a number of sequence segments greater than the threshold value, and the scanning length of the scanning process is equal to the preset stop value, the game style of the user is determined according to the game character information.

[0030] Optionally, dividing the users of the information to be pushed into groups according to the user portraits of the users to whom the information is to be pushed, and determining, for at least one of the divided user groups, the information to be pushed to each user in the user group according to the game style of each user in the user group, specifically includes:

[0031] Determine user characteristics of each user of the information to be pushed according to the user portrait of each user of the information to be pushed, and cluster each user of the information to be pushed according to the user characteristics to determine each user group;

[0032] For each user group, determine the offline interval of each user in the user group;

[0033] For each user in the user group, when the offline interval of the user is greater than a preset interval, the user is determined to be a lost user; when the offline interval of the user is not greater than the preset interval, the user is determined not to be a lost user;

[0034] Determine whether the ratio of the number of lost users in the user group to the total number of users in the user group is greater than a preset ratio threshold; if so, determine that the user group is a lost group; if not, determine that the user group is not a lost group;

[0035] According to the game style of each user in the churn group, information to be pushed to each user in the churn group is determined.

[0036] Optionally, before determining that the user is a lost user when the offline interval is greater than a preset interval and determining that the user is not a lost user when the offline interval is not greater than the preset interval, the method further includes:

[0037] For each user in the user group, determine the average single online time of the user;

[0038] Determine whether the user's average single online time is greater than the preset time;

[0039] If yes, determine the additional duration according to the difference between the average single online time and the preset duration, and update the preset interval according to the sum of the additional duration and the preset interval, wherein the additional duration is proportional to the average single online time;

[0040] If not, the preset interval is not updated.

[0041] Optionally, determining the information to be pushed to each user in the user group according to the game style of each user in the user group specifically includes:

[0042] For each user in the at least one user group, determining a game resource corresponding to the game style of the user;

[0043] According to the game resources, information to be pushed to the user is determined.

[0044] This specification provides a personalized information push device, the device comprising:

[0045] User preliminary selection module, to determine the users to whom information is to be pushed;

[0046] The information determination module obtains the game character information and the historical event information of each user to whom the information is to be pushed;

[0047] An operation data determination module, which determines the historical operation data of the user on the game operation object in the game according to the historical event information;

[0048] A game style determination module, which determines the game style of the user according to the game character information and the historical operation data;

[0049] A user portrait module, which determines the user portrait of the user based on the user information, game character information and the game style of the user;

[0050] The information push module divides the users of the information to be pushed into groups according to the user portraits of the users, and determines, for at least one of the divided user groups, the information to be pushed to each user in the user group according to the game style of each user in the user group, so as to push the information.

[0051] This specification provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned personalized information push method is implemented.

[0052] This specification provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned personalized information push method when executing the program.

[0053] At least one of the above technical solutions adopted in this specification can achieve the following beneficial effects:

[0054] In the personalized information push method provided in this specification, for each user to whom information is to be pushed, the game character information and historical event information of the user are obtained, and the historical operation data of the user on the game operation object in the game is determined based on the historical event information, and the game style of the user is determined based on the game character information and the historical operation data, and the user portrait of the user is determined based on the user information, game character information and game style of the user. According to the user portrait of each user, the user group is divided, and the information to be pushed to at least one user group is determined to push information. In the method provided in this specification, when determining the user portrait of the user, the game style and game character information are taken into account, which can reflect the user's personal game behavior, increase the type of data used for user portraits, and the user group divided based on the user portrait is more accurate, so that accurate personalized push can be performed according to the characteristics of the user group. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The drawings described herein are used to provide a further understanding of this specification and constitute a part of this specification. The illustrative embodiments and descriptions of this specification are used to explain this specification and do not constitute an improper limitation on this specification. In the drawings:

[0056] Figure 1 A flowchart of a personalized information push method in this manual;

[0057] Figure 2 A sequence segment scanning process diagram provided in this specification;

[0058] Figure 3 A schematic diagram of determining a lost population provided in an embodiment of this specification;

[0059] Figure 4 A schematic diagram of a personalized information push device provided in this manual;

[0060] Figure 5 The corresponding Figure 1 Schematic diagram of electronic equipment. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solutions and advantages of this specification more clear, the technical solutions of this specification will be clearly and completely described below in combination with the specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0062] The technical solutions provided by the embodiments of this specification are described in detail below in conjunction with the accompanying drawings.

[0063] Figure 1 The following is a flowchart of a personalized information push method in this specification, which specifically includes the following steps:

[0064] S100: Determine the users to whom information is to be pushed.

[0065] All steps in the personalized information push method provided in this specification can be implemented by any electronic device with computing functions, such as a terminal, a server, etc. For ease of description, the personalized information push method provided in this specification is described below with the server as the execution subject.

[0066] The personalized information push method provided in this specification is used to push information to game users. For a game, the server can use the registration data stored in the user database of the game to determine all users who have registered the game as users to be pushed information.

[0067] Alternatively, the server may also perform a preliminary screening of all users registered for the game according to preset screening conditions, and use the remaining users after the preliminary screening as users to be pushed information, so as to reduce the user base for determining user portraits and group division in subsequent steps, thereby speeding up execution.

[0068] The screening criteria here can be determined according to the target audience of a specific game marketing campaign. For example, in a game marketing campaign on Children's Day, the target audience is users under the age of 16. The server can first screen each registered user according to age and exclude users who are obviously not in the target audience.

[0069] S102: For each user to whom information is to be pushed, obtain the game character information of the user and the historical event information of the user.

[0070] For each user to whom information is to be pushed, the server obtains the game character information of the user from the user database, and obtains the historical event information of the user from the game log.

[0071] The game characters used by users during the game reflect the user's habitual preferences. The user's game character information records the game characters used by the user in various games in history, as well as the character attributes such as the survivability, damage ability, skills, and skill effects of each game character. The user's historical event information records various operation data generated by the user's manipulation of the user's own game character, system characters, environmental elements and other game operation objects during the game.

[0072] By analyzing game character information and historical event information, we can determine the user's operating habits in the game and make a more accurate portrait of the user in terms of gaming style.

[0073] S104: Determine historical operation data of the user on the game operation object in the game according to the historical event information.

[0074] In the game, users can perform different types of operations on different game operation objects. The operations performed on different operation objects can represent the user's behavioral habits when facing different operation objects. Operations related to game characters, such as walking, skill casting, attack target selection, etc., operations related to system characters, such as dialogue, trading, attacking, etc., operations related to environmental elements, such as entering bushes, passing through walls, etc.

[0075] Therefore, the server classifies the historical event information according to the game operation object and determines the historical operation data.

[0076] Specifically, the server determines each game operation object included in the historical event information. For each game operation object, the server sequentially extracts each sub-event corresponding to the game operation object in the historical event information as the event sequence of the game operation object. According to the event sequence of each game operation object, the server determines the historical operation data of the user on the game operation object in the game.

[0077] A sub-event records an operation performed by a user on a game operation object in the game, such as a skill release, an attack target selection, etc. The historical event information is composed of sub-events for each game operation object.

[0078] Generally, historical event information is recorded in chronological order. However, in an offline state, the operation data generated by users in the game cannot be recorded in real time, resulting in some historical operation data not being recorded in time in the historical event information, affecting the subsequent determination of user portraits.

[0079] Therefore, preferably, the server can sort each sub-event according to the timestamp of each sub-event corresponding to the game operation object to obtain the event sequence of the game operation object.

[0080] S106: Determine the game style of the user according to the game character information and the historical operation data.

[0081] In one or more embodiments of the present specification, the game style may be composed according to game character information and behavior style.

[0082] The server determines the behavior style according to the following method:

[0083] First, the server determines the scan length according to the number of sub-events contained in the event sequence of each game operation object, and performs multiple rounds of scans on each event sequence in descending order of the scan length.

[0084] The sub-events in the event sequence are the operations performed by the user in the game. If multiple consecutive sub-events frequently appear in different event sequences or at different positions in the same event sequence, it can be considered that the operations corresponding to the multiple consecutive sub-events occur frequently, which represents the user's behavioral habits.

[0085] In an event sequence, the more continuous sub-events there are, the more accurate the description of user behavior habits is. Therefore, in this embodiment, the scanning length is gradually reduced so that the server can scan the sequence segments composed of continuous sub-events in order from long to short.

[0086] The scan length determined by the number of sub-events included in the event sequence of each game operation object is the initial scan length of the first round of scanning process. There are multiple ways to determine the initial scan length. The server can use the minimum or maximum number of sub-events included in the event sequence of each game operation object as the initial scan length. Alternatively, a number is randomly selected from the number of sub-events included in each game operation object as the initial scan length.

[0087] Secondly, for each round of scanning process, the server determines a sequence segment of a corresponding scanning length in each event sequence according to the scanning length of the scanning process.

[0088] Starting from the initial scan length, the server will intercept a sequence segment consisting of all consecutive sub-events that match the scan length in the event sequence during each round of scanning.

[0089] Specifically, the server determines, for each event sequence in the scanning process, according to the scanning length of the scanning process, the number of sub-events covered by the scanning frame of the scanning process.

[0090] Align the first position of the scan frame with the first sub-event of the event sequence, and move the scan frame backwards one sub-event position each time until the tail position of the scan frame is aligned with the last sub-event of the event sequence. The sub-events covered by the scan frame after each movement are regarded as sequence segments.

[0091] The following will be combined Figure 2 The determination of the sequence segment in this embodiment is described. Figure 2 A sequence segment scanning process diagram provided in this specification. Figure 2 In the example, different numbers represent different sub-events, and different sub-events are connected by straight lines. Figure 2 The figure shows an event sequence containing 5 sub-events, and the scanning process when the scanning length is 3. The scanning length is 3, and the scanning frame of this round of scanning process covers 3 sub-events. The scanning frame is moved from the first sub-event of the event sequence, and the sequence segments determined by the server scanning the event sequence are 1→2→3, 2→3→4, and 3→4→5.

[0092] Then, the server determines a sequence segment set consisting of the same sequence segments according to the names of the sub-events included in the sequence segments and the order of the sub-events included in the sequence segments.

[0093] If there is a sequence segment set including sequence segments whose number is greater than a preset threshold value among the sequence segment sets, the behavior style of the user is determined according to the sequence segment corresponding to the sequence segment set including sequence segments whose number is greater than the threshold value.

[0094] If there is no sequence segment set in each sequence segment set that contains a number of sequence segments greater than the threshold value, and the scanning length of this round of scanning process is equal to the preset stop value, then the user's game style is determined to be none based on the character style.

[0095] Specifically, the server determines, for each sequence segment set, whether the number of sequence segments contained in the sequence segment set is greater than a preset threshold value. If so, the sequence segments corresponding to the sequence segment set are used as the behavior style of the user. If not, when the scanning length of the current scanning process is not equal to the preset stop value, the next scanning process is continued, and when the scanning length of the current scanning process is equal to the stop value, the scanning process is stopped.

[0096] This specification does not limit the specific value of the stop value, which is used to determine the stop condition of the scanning process. Generally, the stop value can be set to 1, so that in the last round of scanning, the server can determine a sequence segment consisting of one sub-event. Considering that a sequence segment containing too few sub-events does not have a clear representation effect on user behavior habits, the stop value can also be set to other values ​​greater than 1.

[0097] Finally, after determining the user's behavior style, the server determines the user's game style based on the game character information and the behavior style.

[0098] In particular, when the scanning process stops, if the server still has not determined a sequence segment set including a number of sequence segments greater than a preset threshold value, the server cannot determine the user's behavior style, indicating that the user's behavior style is none. In this case, the server can determine the user's game style only based on the game character information.

[0099] S108: Determine a user portrait of the user based on the user information, game character information, and game style of the user.

[0100] In the personalized recommendation method of this specification, the server first determines the user's gaming style to enrich the data type of the data used to determine the user's portrait, and finally makes a more accurate user portrait of the user from multiple dimensions based on the user information, game character information and gaming style.

[0101] S110: Divide the users of the information to be pushed into groups according to the user portraits of the users, and for at least one of the divided user groups, determine the information to be pushed to each user in the user group according to the game style of each user in the user group, so as to push the information.

[0102] After obtaining the user portrait of each user, the server can divide the user groups based on the user portrait and determine users with similar characteristics. In game marketing activities, the server can determine the target group based on the characteristics of each user group, and determine the information to be pushed to the target group based on the game style of each user in the target group. The target group can be determined based on the actual operation of the business, and the target group can be a churn group, a novice group, a high-activity group, etc.

[0103] In one or more embodiments of the present specification, the target group is a lost group. The server pushes information in the following manner.

[0104] First, the server determines the user characteristics of each user based on the user portrait of each user to whom information is to be pushed, and clusters the users based on the user characteristics to determine the user groups.

[0105] The user features in this embodiment can be obtained by encoding the user portrait through any embedding model, such as the bidirectional encoder model based on transformer (Bidirectional Encoder Representations from Transformers, BERT), the bag of words model (Bag of Words, BOW), the global word frequency statistics-based word representation model (Global Vectors for Word Representation, GloVe), etc. This specification does not limit the specific selection of the embedding model.

[0106] Secondly, the server determines the lost users in each user group according to the offline interval of the users in each user group.

[0107] Specifically, for each user group, the server determines the most recent login time of each user in the user group. For each user in the user group, the server determines the user's offline interval based on the time interval between the user's most recent login time and the current time, and determines users whose offline interval is greater than a preset interval as lost users.

[0108] Due to the diversity of user behaviors, determining lost users based only on the most recent login time may be too one-sided. Therefore, in another embodiment, the server may also determine lost users based on the user's offline interval and average single online time.

[0109] Specifically, for each user group, the server determines the offline interval and average single online time of each user in the user group. For each user in the user group, the server determines whether the average single online time of the user is greater than the preset time. If so, the additional time of the user is determined according to the difference between the average single online time and the preset time, and the preset interval is re-determined according to the sum of the additional time and the preset interval. The additional time is proportional to the average single online time. If not, the additional time of the user is not determined, and the preset interval is not re-determined.

[0110] The server determines the user's offline interval based on the time interval between the user's most recent login time and the current time, and determines users whose offline interval is greater than a preset interval as lost users.

[0111] In this embodiment, the server updates the preset interval according to the average single online time of the user, that is, it gives additional accommodating offline time to users with a relatively long average single online time. If the average single online time of the user is greater than the preset time, it means that the average single online time of the user is long. Generally, users with a long average single online time will participate in the game less frequently, and the time interval between two adjacent games will usually be longer.

[0112] If the same preset interval is used to measure whether a user with a long average single online time and a user with a short average single online time is a lost user, the judgment will be inaccurate. This embodiment dynamically adjusts the preset interval for judging lost users according to the average single online time of the user to determine a more accurate measurement standard and accurately determine lost users.

[0113] Then, the server determines the proportion of lost users in the user group to the total number of users in the user group, and determines whether the proportion is greater than a preset proportion threshold. If so, the user group is determined to be a lost group. If not, the server determines that the user group is not a lost group.

[0114] Figure 3 This is a schematic diagram of a loss group determination principle provided in an embodiment of this specification, such as Figure 3 As shown in the figure, letters a to e represent the five user groups obtained after clustering, the shaded part represents the proportion of lost users in each lost group, and the dotted line represents the proportion threshold of 0.5. Figure 3 In the example, user groups a, c, and d can be determined as churn groups.

[0115] The server determines the information to be pushed to the lost group, and pushes the information to attract the users to return to the game. The specific content of the information pushed to the lost group can be determined according to the content of the specific game marketing activities.

[0116] Because user groups are divided by clustering, users in the same user group have similar characteristics. When most users in a user group are churned users, it means that all users in this user group are likely to churn. Therefore, the server pushes information to user groups where the number of churned users is greater than the proportion threshold, which can more effectively prevent user churn.

[0117] Furthermore, considering that the gaming styles of users in the same user group may also be different, the server can determine the information to be pushed to each user based on the gaming style of each user in the target group, generate personalized push information for each user, and enhance the user experience.

[0118] Specifically, the server determines, for each user in at least one user group, information to be pushed to the user according to the game resources corresponding to the game style of the user.

[0119] The information pushed by the server to the user may include basic information of the game resources, channels for obtaining the game resources, etc. The game resources are virtual currency and equipment that match the game style of the user.

[0120] For example, in a certain game, users can choose to walk or drive a vehicle when fighting in the game. In determining the churn group, by analyzing a user's game style, it is found that 90% of the user's total game time is spent driving a vehicle, but because there are very few game tasks that support driving a vehicle, the user leaves the game after trying all the ways to drive a vehicle. By analyzing the user's game style, the server can determine that the user is a vehicle enthusiast. When generating a push message to the user, the server also provides the user with map information containing the vehicle base, as well as a prop card that can be used to complete more game tasks by driving a vehicle. With the help of the prop card, the vehicle can be driven through areas that could only be passed on foot before. In this way, when the user receives the push message, there is a greater possibility that he or she will return to the game.

[0121] In the personalized information push method provided in this specification, for each user to whom information is to be pushed, the game character information and historical event information of the user are obtained, and the historical operation data of the user on the game operation object in the game is determined based on the historical event information, and the game style of the user is determined based on the game character information and the historical operation data, and the user portrait of the user is determined based on the user information, game character information and game style of the user. According to the user portrait of each user, the user group is divided, and the information to be pushed to at least one user group is determined to push information. In the method provided in this specification, when determining the user portrait of the user, the game style and game character information are taken into account, which can reflect the user's personal game behavior, increase the type of data used for user portraits, and the user group divided based on the user portrait is more accurate, so that accurate personalized push can be performed according to the characteristics of the user group.

[0122] In the above step S106, the game style can also be determined based on the character style and historical operation data.

[0123] In one or more embodiments of the present specification, the server determines the character style according to the following method:

[0124] First, the server determines the user's frequently used character among various game characters used by the user according to the game character information.

[0125] Because the user's game style generally changes dynamically with the degree of the user's participation in the game, in order to obtain the user's current game style, the server can determine the number of times the user uses each game character within a specified time period, and use the game character with a number of uses greater than a first value as a frequently used character. Alternatively, the server can also use the game character with the most uses as a frequently used character.

[0126] Secondly, the server determines the role type of each game role used, and determines the user's common role type based on the role type of each game role used by the user.

[0127] Specifically, the server determines the number of game characters corresponding to each character type among the game characters used by the user, and uses the character type whose number of corresponding game characters is greater than the second value as the commonly used type. Alternatively, the server may use the character type with the largest number of corresponding game characters as the commonly used type.

[0128] In this embodiment, the first numerical value determines the number of commonly used characters determined, and the second numerical value determines the number of commonly used types determined. When the number of commonly used characters and the number of commonly used types are both one, the role type of the commonly used characters is the same as the commonly used types. When the number of commonly used characters and the number of commonly used types are different, the role type of the commonly used characters is not necessarily the same as the commonly used types.

[0129] Finally, the server determines the user's character style based on the determined common characters and common types, and determines the user's game style based on the character style and historical operation data.

[0130] The game characters in the game can be divided into different character types according to their character attributes. The character attributes of the game characters of the same character type are similar. Common characters reflect the user's preference for specific game characters, and common types reflect the user's preference for a certain type of character attributes.

[0131] The designated time, the first value and the second value in this embodiment can be set according to the requirements. It should be noted that the designated time is a period of time with the current time as the end time, and the specific duration of the designated time period can be set according to the actual requirements.

[0132] Generally, the longer the designated time is, the more the number of used game characters determined by the server is, and the shorter the designated time is, the fewer the number of used game characters determined by the server is.

[0133] Therefore, in another embodiment, when the number of game characters used within a specified time is less than a preset lower limit, the server can directly use the game characters used within the specified time as common characters, and determine the common types based on the character types of the common characters.

[0134] The behavior style represents the user's behavior habits in the game, and the character style represents the user's character preferences and a certain type of character attribute preferences in the game. In order to determine a more accurate game style, the server can also determine the game style based on the character style and behavior style.

[0135] In the above step S106, the game style is generated according to the game character information and historical operation data of the user. In another embodiment, the type of game style can also be preset, and the server determines which preset game style the user belongs to through the style recognition model.

[0136] The training process of the style recognition model is as follows:

[0137] The server obtains the game character information and historical operation data of each historical user, as well as the annotation style of each historical user. For each historical user, the server inputs the game character information and historical operation data of the historical user into the style recognition model to be trained, obtains the predicted style of the historical user, and trains the style recognition model according to the difference between the predicted style and the annotation style.

[0138] After the style recognition model training is completed, the server inputs each user's game character information and historical operation data into the style recognition model to determine the user's game style.

[0139] In another embodiment of the above step S110, the server may also determine the users who are about to churn through a churn prediction model, and send push information to these users to prevent these users from churn.

[0140] The training process of the churn prediction model is as follows:

[0141] First, the server obtains the user records of two consecutive time periods in history, and for each user in the user record of the first period of the two time periods, determines whether the user exists in the user record of the second period of the two time periods. If so, the user is determined to be marked as a retained user. If not, the user is determined to be marked as a lost user.

[0142] Execute the above steps S102 to S108, determine the user profile of the user, input the user profile of the user into the encoding layer of the churn prediction model to be trained, determine the user features of the user, input the user features into the prediction layer of the churn prediction model, and output the predicted churn result of the user, which is divided into two types: predicted retention and predicted churn. According to the difference between the predicted churn result of the user and the annotation, train the churn prediction model.

[0143] After the prediction churn model is trained, for each user to whom information is to be pushed, the user profile of the user is input into the prediction churn model to obtain the predicted churn result of the user. The server determines the churn group for each user predicted to churn based on the predicted churn result, determines the information to be pushed to the churn group, and pushes the information.

[0144] In the above step S110, the target group may also be a high-activity group. Then the server may push information in the following manner.

[0145] The server determines the number of times each user has logged into the game and the cumulative online time within the specified time period. The specified time period in this embodiment has the same meaning as the specified time period in S106.

[0146] For each user, the server determines whether the number of times the user has logged into the game is greater than a preset third value, and determines whether the cumulative online time of the user is greater than a preset fourth value.

[0147] If the number of times the user logs into the game is greater than the preset third value, and the cumulative online time of the user is greater than the preset fourth value, the user is determined to be a high-activity user. If the number of times the user logs into the game is greater than the preset third value, but the cumulative online time of the user is not greater than the preset fourth value, or if the cumulative online time of the user is greater than the preset fourth value, but the number of times the user logs into the game is not greater than the preset third value, the user is determined to be a normal-activity user. If the number of times the user logs into the game is not greater than the preset third value, and the cumulative online time of the user is not greater than the preset fourth value, the user is determined to be a low-activity user.

[0148] The server determines that the user group composed of highly active users is a high-activity group, the group composed of ordinary active users is an ordinary active group, and the group composed of low-activity users is a low-activity group. When generating push information, the information pushed to the high-activity group contains the claiming channel of exclusive props and limited skins, the information pushed to the ordinary active group prompts that after completing specific game tasks, additional gold coins or double experience rewards will be obtained, and the information pushed to the low-activity group contains a reminder of the gift of a specified amount of gold coins or experience. In this way, different message content is pushed to user groups with different levels of activity, enhancing the user experience of user groups with different levels of activity.

[0149] The above is the personalized information push method provided in this specification. Based on the same idea, this specification also provides a corresponding personalized information push device, such as Figure 4 shown.

[0150] Figure 4 A schematic diagram of a personalized information push device provided in this manual specifically includes:

[0151] A user preliminary selection module 200 is used to determine the users to whom information is to be pushed;

[0152] The information determination module 202 is used to obtain the game character information of each user to which the information is to be pushed, as well as the historical event information of the user;

[0153] An operation data determination module 204, used to determine the historical operation data of the user on the game operation object in the game according to the historical event information;

[0154] A game style determination module 206, configured to determine the game style of the user according to the game character information and the historical operation data;

[0155] A user portrait module 208, for determining a user portrait of the user based on the user information, game character information and the game style of the user;

[0156] The information push module 210 is used to divide the users of the information to be pushed into groups according to the user portraits of the users, and for at least one of the divided user groups, determine the information to be pushed to each user in the user group according to the game style of each user in the user group, so as to push the information.

[0157] Optionally, the operation data determination module 204 is specifically used to determine the game operation objects contained in the historical event information, determine, for each game operation object, the sub-events corresponding to the game operation object, sort the sub-events according to the timestamps of the sub-events, determine the event sequence of the game operation object, and determine the historical operation data of the user on the game operation object in the game according to the event sequence of each game operation object.

[0158] Optionally, the game style determination module 206 is specifically used to determine the user's commonly used characters among the game characters used by the user based on the game character information, determine the user's commonly used type based on the character types of the game characters used by the user, determine the user's character style based on the commonly used characters and the commonly used types, and determine the user's game style based on the character style and the historical operation data.

[0159] Optionally, the game style determination module 206 is specifically used to determine the scanning length according to the number of sub-events included in the event sequence of each game operation object, and perform multiple rounds of scanning on each event sequence in descending order of the scanning length. For each scanning process, according to the scanning length of the scanning process, a sequence segment of the corresponding scanning length is determined in each event sequence, and a sequence segment set composed of the same sequence segments is determined according to the name of each sub-event included in each sequence segment and the order of each sub-event included in each sequence segment. If there is a sequence segment set in each sequence segment set containing a number of sequence segments greater than a preset threshold value, the user's behavior style is determined according to the sequence segment corresponding to the sequence segment set containing a number of sequence segments greater than the threshold value, and the game style of the user is determined according to the game character information and the behavior style. If there is no sequence segment set in each sequence segment set containing a number of sequence segments greater than the threshold value, and the scanning length of the scanning process is equal to the preset stop value, the game style of the user is determined according to the game character information.

[0160] Optionally, the information push module 210 is specifically used to determine user characteristics of each user of the information to be pushed based on the user portrait of each user of the information to be pushed, cluster each user of the information to be pushed based on the user characteristics, determine each user group, and for each user group, determine the offline interval of each user in the user group, for each user in the user group, when the offline interval of the user is greater than a preset interval, determine that the user is a lost user, and when the offline interval of the user is not greater than the preset interval, determine that the user is not a lost user, and judge whether the ratio of the number of lost users in the user group to the total number of users in the user group is greater than a preset ratio threshold; if so, determine that the user group is a lost group; if not, determine that the user group is not a lost group, and determine the information to be pushed to each user in the lost group based on the game style of each user in the lost group.

[0161] Optionally, the information push module 210 is specifically used to determine the average single online time of each user in the user group, and judge whether the average single online time of the user is greater than a preset duration; if so, determine an additional duration according to the difference between the average single online time and the preset duration; update the preset interval according to the sum of the additional duration and the preset interval, and the additional duration is proportional to the average single online time; if not, the preset interval is not updated.

[0162] Optionally, the information push module 210 is specifically configured to determine, for each user in the at least one user group, a game resource corresponding to the game style of the user, and determine information to be pushed to the user based on the game resource.

[0163] This specification also provides a computer-readable storage medium, which stores a computer program, which can be used to execute the above Figure 5 Provides personalized information push method.

[0164] This manual also provides Figure 5 The schematic structure diagram of the electronic device shown in FIG. Figure 5 As mentioned above, at the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include other hardware required for the business. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to achieve the above Figure 1 Of course, in addition to the software implementation, this specification does not exclude other implementations, such as logic devices or a combination of software and hardware, etc., that is, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0165] For the improvement of a technology, it can be clearly distinguished whether it is a hardware improvement (for example, improvement of the circuit structure of diodes, transistors, switches, etc.) or a software improvement (improvement of the method flow). However, with the development of technology, many improvements of the method flow today can be regarded as direct improvements of the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that the improvement of a method flow cannot be implemented with a hardware entity module. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming themselves, without having to ask chip manufacturers to design and make dedicated integrated circuit chips. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.

[0166] The controller can be implemented in any appropriate manner, for example, the controller can take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) that can be executed by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in a purely computer-readable program code manner, the controller can be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, this controller can be considered as a hardware component, and the devices included therein for implementing various functions can also be regarded as structures within the hardware component. Or even, the devices for implementing various functions can be regarded as both software modules for implementing the method and structures within the hardware component.

[0167] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0168] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0169] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0170] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0171] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0172] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0173] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0174] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0175] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0176] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0177] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems or computer program products. Therefore, this specification may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, this specification may 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.

[0178] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0179] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0180] The above description is only an embodiment of this specification and is not intended to limit this specification. For those skilled in the art, this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this specification should be included in the scope of the claims of this application.

Claims

1. A personalized information push method, characterized in that: include: Determine the users to whom information is to be pushed; For each user to whom information is to be pushed, obtain the user's game character information and the user's historical event information; Determine the historical operation data of the user on each game operation object in the game according to the historical event information; Determining the game style of the user according to the game character information and the historical operation data; Determine a user profile of the user based on the user information, game character information, and the game style of the user; Determine the user characteristics of each user of the information to be pushed based on the user portrait of each user of the information to be pushed, cluster the users of the information to be pushed based on the user characteristics, and determine each user group; for each user group, determine the offline interval of each user in the user group; for each user in the user group, when the offline interval of the user is greater than a preset interval, determine that the user is a lost user, and when the offline interval of the user is not greater than the preset interval, determine that the user is not a lost user; determine whether the ratio of the number of lost users in the user group to the total number of users in the user group is greater than a preset ratio threshold; if so, determine that the user group is a lost group; if not, determine that the user group is not a lost group; determine the information to be pushed to each user in the lost group based on the game style of each user in the lost group, so as to push the information; The step of determining the game style of the user according to the game character information and the historical operation data specifically includes: Determine the scanning length according to the number of sub-events included in the event sequence of each game operation object included in the historical event information, wherein for each game operation object, the event sequence of the game operation object is determined by sequentially extracting sub-events corresponding to the game operation object in the historical event information, and the sub-event records an operation of the user on the game operation object in the game; Perform multiple rounds of scanning on each event sequence in descending order of the scanning length; For each scanning process, according to the scanning length of the scanning process, determining a sequence segment corresponding to the scanning length in each event sequence; Determining a sequence segment set consisting of the same sequence segments according to the names of the sub-events included in the sequence segments and the order of the sub-events included in the sequence segments; If there is a sequence segment set including a number of sequence segments greater than a preset threshold value among the sequence segment sets, determining the behavior style of the user according to the sequence segment corresponding to the sequence segment set including a number of sequence segments greater than the threshold value, and determining the game style of the user according to the game character information and the behavior style; If there is no sequence segment set in each sequence segment set that contains a number of sequence segments greater than the threshold value, and the scanning length of the scanning process is equal to the preset stop value, the game style of the user is determined according to the game character information.

2. The method according to claim 1, characterized in that Determine the historical operation data of the user on the game operation object in the game according to the historical event information, specifically including: Determine each game operation object included in the historical event information; For each game operation object, determine each sub-event corresponding to the game operation object; Sorting the sub-events according to the timestamps of the sub-events to determine the event sequence of the game operation object; According to the event sequence of each game operation object, the historical operation data of the user on the game operation object in the game is determined.

3. The method according to claim 1, characterized in that Determining the game style of the user according to the game character information and the historical operation data specifically includes: According to the game character information, determining the user's frequently used character among the game characters used by the user; Determine the user's frequently used type according to the type of each game character used by the user; Determining the role style of the user according to the common roles and the common types; The game style of the user is determined according to the character style and the historical operation data.

4. The method according to claim 1, characterized in that Before determining that the user is a lost user when the offline interval is greater than a preset interval and determining that the user is not a lost user when the offline interval is not greater than the preset interval, the method further includes: For each user in the user group, determine the average single online time of the user; Determine whether the user's average single online time is greater than the preset time; If yes, determine the additional duration according to the difference between the average single online time and the preset duration, and update the preset interval according to the sum of the additional duration and the preset interval, wherein the additional duration is proportional to the average single online time; If not, the preset interval is not updated.

5. The method according to claim 1, characterized in that According to the game style of each user in the user group, determine the information to be pushed to each user in the user group, specifically including: For each user in the at least one user group, determining a game resource corresponding to the game style of the user; According to the game resources, information to be pushed to the user is determined.

6. A personalized information push device, characterized in that: include: User preliminary selection module, to determine the users to whom information is to be pushed; The information determination module obtains the game character information and the historical event information of each user to whom the information is to be pushed; An operation data determination module, which determines the historical operation data of the user on each game operation object in the game according to the historical event information; A game style determination module, which determines the game style of the user according to the game character information and the historical operation data; A user portrait module, which determines the user portrait of the user based on the user information, game character information and the game style of the user; The information push module determines the user characteristics of each user of the information to be pushed according to the user portrait of each user of the information to be pushed, clusters the users of the information to be pushed according to the user characteristics, and determines each user group; for each user group, determines the offline interval of each user in the user group; for each user in the user group, when the offline interval of the user is greater than a preset interval, determines that the user is a lost user, and when the offline interval of the user is not greater than the preset interval, determines that the user is not a lost user; determines whether the ratio of the number of lost users in the user group to the total number of users in the user group is greater than a preset ratio threshold; if so, determines that the user group is a lost group; if not, determines that the user group is not a lost group; determines the information to be pushed to each user in the lost group according to the game style of each user in the lost group, so as to push the information; The game style determination module is specifically used to: determine the scanning length according to the number of sub-events contained in the event sequence of each game operation object contained in the historical event information, for each game operation object, the event sequence of the game operation object is determined by sequentially extracting the sub-events corresponding to the game operation object in the historical event information, and the sub-event records an operation of the user on the game operation object in the game; perform multiple rounds of scanning on each event sequence in descending order of the scanning length; for each scanning process, determine the sequence segment of the corresponding scanning length in each event sequence according to the scanning length of the scanning process; and determine the number of sub-events contained in each sequence segment according to the number of sub-events contained in each sequence segment. The names of the sub-events and the orders of the sub-events included in the sequence segments are used to determine a sequence segment set composed of the same sequence segments; if there is a sequence segment set in each sequence segment set whose number of sequence segments is greater than a preset threshold value, the behavior style of the user is determined according to the sequence segments corresponding to the sequence segment set in which the number of sequence segments is greater than the threshold value, and the game style of the user is determined according to the game character information and the behavior style; if there is no sequence segment set in each sequence segment set whose number of sequence segments is greater than the threshold value, and the scanning length of this round of scanning process is equal to the preset stop value, the game style of the user is determined according to the game character information.

7. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 5 is implemented.

Images