A game platform user accurate portrait construction method and system

Abstract

This invention relates to a method and system for constructing accurate user profiles on a game platform, belonging to the field of game data processing technology. It includes: collecting user login information from the game platform; retrieving login time and login address from the login information; obtaining login duration based on the login information and login time; determining whether adjacent login addresses are consistent; if consistent, retrieving the game type from the game platform using the login duration and login information, and constructing a user profile based on the game type; if inconsistent, defining adjacent and inconsistent login addresses as abnormal login addresses; obtaining the geographical route based on the abnormal login address; and constructing a user profile based on the geographical route, login duration, and login information. This application has the effect of improving the accuracy of subsequent game recommendations from the game platform.

Description

Technical Field

[0001] This invention relates to the field of game data processing technology, and in particular to a method and system for constructing accurate user profiles for game platforms. Background Technology

[0002] Game data processing is a method that uses formatted processing techniques to extract accurate and valid data from a large amount of messy game data, and is used by technicians to build user profiles.

[0003] In the process of game data processing and user profile building, game platforms typically extract statistical values ​​of a user's independent behavioral dimension (such as total spending or online time) according to a preset long period. These values ​​are then directly matched with a preset configuration table to generate a profile tag corresponding to that dimension. The platform then collects user login logs, calculates their average online time, distribution of active periods, and long-term offline tasks (such as full recalculation by week or month), and calculates basic activity indicators using formulas such as SUM, COUNT, MAX / MIN, and AVG. The basic activity indicators are then compared with a fully deterministic configuration table to output the user's tag to overwrite the old tag.

[0004] In the process of game data processing to build user profiles, user profile updates rely on fixed, long-term offline data collection, which cannot adapt to changes in users' preferences for game types in real time, thereby reducing the accuracy of subsequent game recommendations from the game platform. Summary of the Invention

[0005] To improve the accuracy of subsequent game recommendations on game platforms, this invention provides a method and system for constructing precise user profiles for game platforms.

[0006] Firstly, the present invention provides a method for constructing accurate user profiles on a game platform, employing the following technical solution: A method for constructing precise user profiles for a game platform includes: Collect user login information from the game platform; Retrieve the login time and login address from the login information; The login duration is obtained based on the login information and login time. Determine if adjacent login addresses are identical; If they match, the game type is retrieved from the game platform using the login duration and login information, and a user profile is built based on the game type. If they are inconsistent, adjacent but inconsistent login addresses will be defined as abnormal login addresses; Obtain the geographical route based on the abnormal login address; User profiles are built based on geographical routes, login duration, and login information.

[0007] By adopting the above technical solution, login information is analyzed to obtain login addresses. By comparing adjacent login addresses and combining them with login information, user profiles can be constructed. This enables real-time data collection to adapt to changes in user preferences for game types. Furthermore, by observing changes in login addresses, it is possible to determine whether the same game account is being used by the user, reducing interference from behavioral data of accounts used by unauthorized individuals on user profiles and improving the accuracy of user profile construction.

[0008] Optionally, methods for building user profiles based on geographic routes, login duration, and login information include: Traffic type is determined based on geographical routes; The login date is obtained based on the login time point of the abnormal login address, and the login duration of the abnormal login address is defined as the abnormal login duration; The traffic type is determined by combining the duration of abnormal logins with the traffic type. Average spending information is obtained based on the marked traffic type and login date; Retrieve the consumption information corresponding to the login information from the game platform; Login information is updated based on consumption information and average consumption information, and the game type is retrieved again based on the updated login information and login duration to build a user profile.

[0009] By adopting the above technical solution, average consumption information is obtained by analyzing abnormal login addresses, login times, and geographical routes. Then, user consumption information is retrieved from the game platform, and the average consumption information is compared with the actual consumption information to determine whether the user has the ability to switch addresses. This helps determine whether someone else is logging into the user's account at another address, thereby eliminating login information from other abnormal login addresses and improving the accuracy of login information, which in turn improves the accuracy of user profile construction.

[0010] Optionally, methods for updating login information based on consumption information and average consumption information include: The consumption coefficient is obtained based on the login duration and login information; Consumption information is derived from the product of the consumption coefficient and consumption information. Determine whether the detected consumption information is greater than the average consumption information; If the value is greater than the original value, continue outputting the original login information. If it is not greater than, then consumption trends can be obtained through consumption information; Login information will be updated based on consumption trends and abnormal login addresses.

[0011] By adopting the above technical solution, the login duration and login information are mapped to a consumption coefficient, which is then multiplied with the consumption information to obtain the detected consumption information. This coefficient is then compared with the average consumption information to determine the user's ability to afford transportation costs. This helps identify whether the user is traveling on their own or if someone else is traveling on their behalf at a different address. This process can filter out login information that actually belongs to the user, thus improving the accuracy of login information.

[0012] Optionally, methods for updating login information based on consumption trends and abnormal login addresses include: The difference between the detected consumption information and the average consumption information is calculated as the consumption deviation value; The duration of trend decline was determined based on consumption trends, abnormal login addresses, and login times. The amount of consumption decrease is obtained based on consumption trends and the duration of the trend decline. The product of the decrease in consumption and the duration of the downward trend is used as the accumulated consumption amount. The login information is updated based on the difference between the accumulated spending and the spending deviation value.

[0013] By employing the aforementioned technical solution, login information is updated through analysis of detected consumption information, average consumption information, consumption trends, and abnormal login information. This allows for the determination of whether users have accumulated consumption in advance to meet subsequent gaming expenses, and further verification that abnormal login addresses belong to the users themselves, thereby improving the accuracy of login information.

[0014] Optionally, methods after updating login information include: The arrival time is obtained by combining the marked traffic type with the abnormal login address; The login time period is determined based on the landing time. The landing and login information is obtained based on the landing time and login information; Baseline login information is derived based on login information and login time period; Update the login information by comparing the actual login information with the baseline login information.

[0015] By adopting the above technical solution, the login information is updated by analyzing the marked traffic type, abnormal login address, and login information. This allows for analysis of the user's actual game login situation to determine whether someone else is playing on their behalf, thereby improving the accuracy of the login information.

[0016] Optionally, the methods after updating login information also include: When the login time period is a preset baseline type, the average duration of the marker is obtained based on the login time period and the login duration. Retrieve the marked login time points that exceed the average marked duration from the landing login information; Retrieve the initial login time from the login information; The baseline time point is obtained based on the marked traffic type and the initial login time point; The duration between the baseline time point and the marked login time point is calculated as the target interval duration. The baseline interval range is obtained by combining the login time period and the baseline type. Update login information by comparing the target interval duration with the baseline interval range.

[0017] By adopting the above technical solution, the login information is updated by analyzing the login time period, login duration, landing login information, and marked traffic type. This further identifies the login situation when the user actually travels to play the game under the benchmark type, and then determines whether it is a case of someone else playing on their behalf, thereby improving the accuracy of the login information.

[0018] Optional methods for obtaining user profiles include: In response to the activity output signal, retrieve the activity item type based on the activity output signal and the game type; The baseline duration range is determined based on the type of event item. Use the login time points within the benchmark duration range as the calibration time points; Update user profiles by combining activity output signals, login information, and baseline duration range.

[0019] By adopting the above technical solution, the user profile is updated by analyzing the activity output signal, game type, login time point and login information, thereby determining the types of activity items that users like in the game, further increasing the basis for building the user profile and improving the accuracy of the user profile.

[0020] Optional methods for updating user profiles include: The average activity value is obtained by combining the activity output signal with login information. The activity duration value is obtained based on the baseline duration range and login information; Determine if the activity duration is less than the activity average. If it is less than that, then remove the event item type from the game type to rebuild a new user profile; If the value is not less than the specified value, the user profile will be updated based on the game type and the type of event items.

[0021] By adopting the above technical solution, user profiles are updated by analyzing activity output signals, login information, and baseline duration ranges. This allows users to reflect their preferences for game items based on their actual activity duration in the game. The types of activity items are used to assist in building user profiles, making them more closely aligned with user preferences and thus improving the accuracy of user profile construction.

[0022] Optionally, methods after obtaining user profiles also include: Retrieve a base single type from the game genres; The number of items that overlap with the baseline single type is obtained by comparing the activity item type with the standard single type. When the number of overlaps exceeds 1, the type of activity item that overlaps with the baseline single type is used as the activity type to be detected. The activity output signal is updated based on the activity type and re-output. The activity duration value is then obtained again based on the updated activity output signal and the reference duration range. Update user profiles based on activity duration and activity type.

[0023] By adopting the above technical solution, user profiles are updated through analysis of game type, event item type, event output signals, and event duration. This allows for more precise segmentation and matching of multiple baseline single types within game event items, and proactive determination of user preferences by updating the event output signals of individual game items in subsequent events. This enables more accurate user preference identification for a specific type, thereby improving the matching between event recommendations and user preferences, and further enhancing the accuracy of user profile construction.

[0024] Secondly, this application provides a system for constructing precise user profiles for a game platform, employing the following technical solution: A system for constructing precise user profiles for a game platform, comprising: The acquisition module is used to retrieve login information; The memory is used to store a program for constructing a precise user profile for a game platform; The processor is used to load and execute programs stored in memory.

[0025] In summary, this application includes at least one of the following beneficial technical effects: 1. By analyzing login information to obtain login addresses, comparing adjacent login addresses and combining them with login information to build user profiles, real-time data collection can be performed to adapt to changes in user game types. Furthermore, by observing changes in login addresses, it can be determined whether the same game account is being used by the user, reducing interference from behavioral data of accounts not being used by the user and improving the accuracy of user profile construction. 2. By analyzing abnormal login addresses, login times, and geographical routes, average spending information is obtained. Then, user spending information is retrieved from the game platform. The average spending information is compared with the spending information to determine whether the user has the ability to switch addresses. This helps to determine whether someone else is logging into the user's account at another address, thereby eliminating login information from other abnormal login addresses and improving the accuracy of login information, which in turn improves the accuracy of user profile construction. 3. By analyzing activity output signals, game types, login times, and login information, user profiles can be updated to determine the types of in-game items that users prefer, thereby increasing the basis for user profile construction and improving the accuracy of user profile construction. Attached Figure Description

[0026] Figure 1 This is a flowchart of a method for constructing a precise user profile for a game platform according to an embodiment of the present invention; Figure 2 This is a system flowchart for constructing precise user profiles on a game platform according to an embodiment of the present invention. Detailed Implementation

[0027] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0028] Reference Figure 1 and Figure 2 This application discloses a method for constructing accurate user profiles on a game platform, including the following steps: S10: Collect user login information from the game platform.

[0029] A game platform is a software platform designed by technicians to store different types of games.

[0030] Login information refers to information such as the time, number of times, spending, and address of a user's login when playing games on a game platform. This information is obtained by directly retrieving the time, number of times, and address of the user's game play from the game platform's system.

[0031] S11: Retrieve the login time and login address from the login information.

[0032] The login time point refers to the point in time when a user logs into the game on the game platform.

[0033] The login address refers to the IP address that a user uses to log in when playing on a game platform.

[0034] By retrieving the login time and login address from the login information.

[0035] S12: Calculate the login duration based on the login information and login time.

[0036] Login duration refers to the length of time a user spends logging into and playing a game on a gaming platform. It is calculated by taking the time difference between the user's login time and the time they log out of the game.

[0037] S13: Determine whether adjacent login addresses are the same.

[0038] By checking whether adjacent login addresses are consistent, it can be determined whether the game account belongs to the user.

[0039] S14: If they match, retrieve the game type from the game platform using the login duration and login information, and build a user profile based on the game type.

[0040] Game type refers to the category of games played by users on the game platform. The game type is determined by retrieving the type of games played by the user (e.g., card games, competitive games, casual games, etc.) from the game platform based on the login duration and login information.

[0041] User profiles refer to user tag information (e.g., high spending, PVP preference, active on weekends and late at night). When adjacent login addresses are the same, it means that the game account belongs to the user. The game type is then input into the tree model (XGBoost / LightGBM) algorithm to obtain the accurate user profile as the user profile.

[0042] S15: If they are inconsistent, define adjacent but inconsistent login addresses as abnormal login addresses.

[0043] An abnormal login address refers to the current login address of the user's game when the previous login address is different.

[0044] By retrieving the user's current login address and its previous login address, if the two login addresses are inconsistent, the user's current login address is defined as an abnormal login address.

[0045] S16: Obtain the geographical route based on the abnormal login address.

[0046] Geographic route refers to the travel route between a user's abnormal login address and its previous login address.

[0047] The abnormal login address and its adjacent login addresses (refer to the login addresses compared with each other in S15) are entered into the route generated in the preset traffic query software as the regional route.

[0048] Traffic query software is navigation software designed by technicians to query relevant routes and traffic information. (See reference...) Figure 2 The traffic query software is the same as the query software shown in the picture.

[0049] S17: Build user profiles based on geographical routes, login duration, and login information.

[0050] User profiles are constructed by analyzing geographical routes, login duration, and login information. The specific construction method is described in S20 to S67.

[0051] Methods for building user profiles based on geographic routes, login duration, and login information include: S20: Traffic type is determined based on geographical routes.

[0052] Transportation type refers to the mode of transportation a user uses between an abnormal login address and their previous login address.

[0053] Enter the geographical route into the traffic query software to obtain the traffic type.

[0054] S21: Obtain the login date based on the login time point of the abnormal login address, and define the login duration of the abnormal login address as the abnormal login duration.

[0055] The login date refers to the date when a user logged into the game from an abnormal login address. The login date is retrieved from the login time of the abnormal login address.

[0056] Abnormal login duration refers to the length of time a user logs into the game at an abnormal login address. It is calculated as the difference between the login time and the time the user logs out of the game.

[0057] S22: Combine the abnormal login duration with the traffic type to obtain the marked traffic type.

[0058] The marked traffic type refers to the type of traffic that can satisfy regional route movement within the abnormal login duration.

[0059] In the traffic query software, query the total time required for a traffic type to travel within a region, and then mark the traffic type whose total time is less than the abnormal login time as the traffic type.

[0060] S23: Obtain average consumption information based on the marked traffic type and login date.

[0061] Average cost information refers to the average cost required to identify a particular transportation type on the login date.

[0062] Retrieve the costs for each traffic type marked on the login date from the traffic query software and calculate their average as the average cost information.

[0063] S24: Retrieve the consumption information corresponding to the login information from the game platform.

[0064] Consumption information refers to information about a user's spending on games on a gaming platform. It is obtained by retrieving the spending amount corresponding to the user's login information on the gaming platform.

[0065] S25: Update login information based on consumption information and average consumption information, and re-retrieve game type using the updated login information and login duration to build user profile.

[0066] The login information is updated based on the user's consumption information and average consumption information. The specific login update method is shown in S30 to S35. The updated login information, login duration, and game type are used as shown in S14 to obtain the user profile.

[0067] Methods for updating login information based on consumption information and average consumption information include: S30: The consumption coefficient is obtained based on the login duration and login information.

[0068] The consumption coefficient is a coefficient used to reflect a user's actual consumption level. The consumption coefficient is obtained by matching the login duration with the frequency in the login information from a preset consumption mapping table.

[0069] The consumption mapping table stores the consumption coefficients corresponding to different login durations and frequencies in the login information. The parameters in the consumption mapping table are set in advance by those skilled in the art based on actual conditions.

[0070] S31: Detect consumption information based on the product of consumption coefficient and consumption information.

[0071] Monitoring consumption information refers to a user's daily spending level.

[0072] The value calculated using a product algorithm based on the consumption coefficient and consumption information is used to detect consumption information.

[0073] S32: Determine whether the detected consumption information is greater than the average consumption information.

[0074] The system determines whether a user's detected consumption information exceeds their average consumption information, thereby assessing whether the user can afford the transportation costs associated with an abnormal login address.

[0075] S33: If it is greater than, continue to output the original login information.

[0076] If the detected consumption information is greater than the average consumption information, it proves that the user can afford the transportation costs required for this abnormal login address, and the original login information continues to be output.

[0077] S34: If it is not greater than, then consumption trends are obtained through consumption information.

[0078] Consumer trends refer to the changes in the amount of money users spend on game recharges on gaming platforms.

[0079] If the detected consumption information is not greater than the average consumption information, it is necessary to check whether the user's consumption has decreased, and then construct a curve of amount change by combining the consumption information with the corresponding time series to represent the consumption trend.

[0080] S35: Update login information based on consumption trends and abnormal login addresses.

[0081] New login information is obtained based on abnormal login addresses and consumption trends.

[0082] The login information is updated based on the user's abnormal login address and consumption trend. The specific login information update method is described in S40 to S67.

[0083] Methods for updating login information based on consumption trends and unusual login addresses include: S40: Calculate the difference between the detected consumption information and the average consumption information as the consumption deviation value.

[0084] The consumption deviation value refers to the deviation between the detected consumption information and the average consumption information.

[0085] The difference between the detected consumption information and the average consumption information is used as the consumption deviation value.

[0086] S41: The duration of the trend decrease is obtained based on consumption trends, abnormal login addresses, and login times.

[0087] The duration of a downward trend refers to the length of time during which users' consumption levels decrease.

[0088] Find the curve where consumption decreases in the consumption trend. When the number of decreases exceeds the preset threshold, the excess part of the curve is taken as the consumption decrease curve. Extract the duration of the consumption decrease curve before the login time of the abnormal login address from the consumption trend as the trend decrease duration.

[0089] The frequency threshold is a threshold set by technical personnel to determine whether a user's consumption is continuously declining. For example, the frequency threshold can be 3 times or more consecutively.

[0090] S42: Obtain the amount of consumption reduction based on consumption trends and the duration of the trend decline.

[0091] The decrease in consumption refers to the difference between the consumption value and the average consumption value during a time series in which the consumption trend is declining.

[0092] Based on the consumption trend, retrieve the consumption volume before the consumption decline curve and calculate the average consumption volume. Then, retrieve the marked consumption volume from the consumption decline curve and calculate the difference between the average consumption volume and the marked consumption volume as the consumption decline amount.

[0093] S43: Calculate the product of the decrease in consumption and the duration of the downward trend as the accumulated consumption amount.

[0094] Accumulated spending refers to the cumulative amount of spending by a user within the gaming platform over the duration of a downgrade.

[0095] The amount of money a user accumulates in the game is calculated by multiplying the amount of money the user spends in the game center by the duration of the downward trend.

[0096] S44: Compare accumulated spending with spending deviation values ​​to update login information.

[0097] The system updates a user's in-game spending information based on whether their accumulated spending in the game exceeds the spending deviation value. If the accumulated spending exceeds the spending deviation value, it means the user can afford transportation costs, and the original login information can be used. If the accumulated spending does not exceed the spending deviation value, it means the user cannot afford transportation costs, and the login information from this abnormal login address needs to be removed to update the login information.

[0098] The methods after updating login information include: S50: The landing time is obtained by combining the marked traffic type and the abnormal login address.

[0099] Landing time refers to the time it takes for a user to arrive at an abnormal login address using the marked traffic type.

[0100] The traffic type is entered into the traffic query software to obtain the arrival time of each traffic type at the abnormal login address. Then, the time point before the login time of the abnormal login address is retrieved from each time point as the landing time.

[0101] S51: Obtain the login time period based on the landing time.

[0102] The login time period refers to the time interval in which a user is logged in.

[0103] The login time period is matched from the preset login mapping table based on the landing time.

[0104] The login mapping table stores the login time periods corresponding to different landing times. The parameters in the login mapping table are set in advance by those skilled in the art based on actual conditions.

[0105] For example: Morning: 5:00 to 8:00, Late Morning: 8:00 to 12:00, Afternoon: 12:00 to 7:00, Evening: 7:00 to 12:00, Midnight: 12:00 to 5:00. Based on the landing time falling into different time ranges, the corresponding time period is used as the login time period.

[0106] S52: Obtain landing login information based on landing time and login information.

[0107] Landing login information refers to the data of a user logging in and playing the game after landing.

[0108] The login information after the landing time is retrieved from the login information and used as the landing login information.

[0109] S53: Derive baseline login information based on login information and login time period.

[0110] Baseline login information refers to the login information of a user during normal gameplay on the game platform after the login period.

[0111] Baseline login information is matched from the login mapping table based on login information and login time period.

[0112] The baseline login mapping table stores baseline login information corresponding to different login information and login time periods. The parameters in the baseline login mapping table are pre-set experimentally by those skilled in the art based on actual conditions. For example, in the morning, users log in for shorter periods and less frequently after landing, while users log in for longer periods and more frequently in the morning.

[0113] S54: Update the login information by comparing the actual login information with the baseline login information.

[0114] The system determines whether the login frequency of the logged-in information matches the login frequency in the baseline login information.

[0115] If they match, it means it's the account holder logging in, so continue outputting the original login information. If they don't match, discard this login information.

[0116] Methods after updating login information also include: S60: When the login time period is a preset baseline type, the average duration of the marker is obtained based on the login time period and the login duration.

[0117] The baseline type is a time period set by technical personnel where similar subsequent login information exists, such as the time periods of noon and evening.

[0118] The average duration of the marker is the average time a user spends playing the game normally after the login period. It is used to determine whether a user is playing the game normally or is not just logging in briefly.

[0119] By extracting the login durations after the login time period from the login durations at various time points and marking them, the login durations that are actually spent playing the game are retrieved from the game platform as the target durations, and the average of each target duration is calculated as the average marked duration.

[0120] S61: Retrieve the marked login time points that exceed the average marked duration from the landing login information.

[0121] The marked login time point refers to the time when a user logs in and plays the game normally after landing. This time point is used to exclude situations where a user logs in immediately after landing to claim event items.

[0122] Retrieve the user's login duration in the game from the login information, and use the login time point corresponding to the login duration exceeding the average duration as the marked login time point.

[0123] S62: Retrieve the initial login time from the landing login information.

[0124] The initial login time point refers to the time when a user logs into the game for the first time after landing.

[0125] The initial login time is obtained by retrieving the user's first login time from the login information.

[0126] S63: Obtain the baseline time point based on the marked traffic type and the initial login time point.

[0127] The baseline time point refers to the landing time point corresponding to the marked traffic type that can meet the initial login time point for game login.

[0128] The system retrieves the landing time of the marked traffic type using traffic query software, marks the landing time that is earlier than the initial login time, calculates the duration between the marked time and the initial login time, and uses the landing time corresponding to the minimum duration as the base time.

[0129] S64: Calculate the duration between the baseline time point and the marked login time point as the target interval duration.

[0130] The target interval duration refers to the time difference between the baseline time point and the marked login time point.

[0131] The target interval duration is calculated by taking the difference between the baseline time point and the marked login time point.

[0132] S65: Obtain the baseline interval range by logging in during the time period and using the baseline type.

[0133] The baseline interval range refers to the range of time a user can play the game normally after traveling.

[0134] The baseline interval range is matched from the interval mapping table by the baseline type and login time period.

[0135] The interval mapping table stores the interval time ranges of different baseline types that exceed the login time period. The parameters in the interval mapping table are set in advance by those skilled in the art based on actual conditions.

[0136] For example: midday peak: 1 to 3 hours, evening peak: 2 to 4 hours, late night: 3 to 5 hours, early morning: 5 to 7 hours. The interval between the base type and the login time period is used as the base interval range.

[0137] S66: Update login information by comparing the target interval duration with the baseline interval range.

[0138] The system determines whether the user is logged in by checking if the target interval falls within the baseline interval range, and updates the login information accordingly. If the target interval falls within the baseline interval range, the user is indeed playing, and the system continues to output login information. If the target interval does not fall within the baseline interval range, the user is not playing, and the system removes login information from the original login information to obtain new login information.

[0139] Methods for obtaining user profiles include: S70: Responds to the activity output signal and retrieves the activity item type based on the activity output signal and the game type.

[0140] Activity output signals refer to the duration and content of game-designed activities.

[0141] Event item type refers to the types of new items that appear in game events.

[0142] Based on the game platform, retrieve the tag type of the event item related to the game type from the event output signal as the event item type.

[0143] In this embodiment, the tag type of the event item needs to be identified in relation to the game when it is retrieved. For example, if the event item belongs to the category of increasing combat power in the game, then the tag type of the event item is combat power.

[0144] S71: Obtain the baseline duration range based on the type of event item.

[0145] The baseline duration range refers to the range of time during which event items in the game can be continuously obtained.

[0146] The duration range of events related to the type of event item is retrieved from the game platform as the baseline duration range.

[0147] S72: Use the login time points within the reference duration range as calibration time points.

[0148] The calibrated time point refers to the time point at which a user logs into the game within the benchmark time range.

[0149] The login time point within the baseline time range of the game login time is used as the calibration time point.

[0150] S73: Update user profiles by combining activity output signals, login information, and baseline duration range.

[0151] User profiles are updated by analyzing activity output signals, login information, and baseline duration ranges. Specific user profile update methods are detailed in S80 to S94.

[0152] Methods for updating user profiles include: S80: The average activity value is obtained by combining the activity output signal with the login information.

[0153] The average duration of a user's time logged into the game during the event period is the average duration of the event.

[0154] Retrieve the historical event times from the game platform, and retrieve the login duration for each event from the login information. Calculate the average of each login duration as the event average.

[0155] S81: Obtain the activity duration value based on the baseline duration range and login information.

[0156] The activity duration value refers to the length of time a user logs in and plays during the current game activity.

[0157] The login duration within the baseline duration range is retrieved from the login information and used as the activity duration value.

[0158] S82: Determine whether the activity duration value is less than the activity average value.

[0159] By determining whether the activity duration is less than the average activity duration, we can judge whether users like the type of items in the activity.

[0160] S83: If it is less than, remove the event item type from the game type to rebuild a new user profile.

[0161] If the activity duration is less than the average activity duration, it means that the user does not like the item type of the game activity. In this case, the same activity item type will be removed from the game type, and a new user profile will be reconstructed using the removed game types.

[0162] S84: If not less than, update the user profile by combining the game type and the type of event items.

[0163] If the activity duration is not less than the average activity duration, it means that the user likes the type of items in the game activity or likes to collect items. In this case, add an activity item type to the game type to get a new game type, and use the new game type to rebuild a new user profile.

[0164] Methods after obtaining user profiles include: S90: Retrieve a base single type from the game types.

[0165] A baseline single type refers to a specific tag among many tags for a game.

[0166] The gaming platform categorizes games into multiple type tags, using one of these tags as a base single type.

[0167] S91: Obtain the number of items that overlap with the baseline single type by comparing the activity item type with the base item type.

[0168] The number of overlaps refers to the number of times that an active item type overlaps with each of the base single types.

[0169] Compare the overlap between the activity item type and the baseline single type. When the baseline single type overlaps with the activity item type, count the overlapping baseline single type and use the count result as the overlap quantity.

[0170] S92: When the number of overlaps exceeds 1, the type of activity item that overlaps with the baseline single type is used as the activity type to be detected.

[0171] The activity type to be detected is the type that overlaps with the baseline single type and the activity item type.

[0172] If the number of overlapping game event items exceeds 1, it indicates that users prefer a large number of types. In this case, the event item tags that overlap with the baseline single type will be used as the event type to be detected.

[0173] S93: Update the activity output signal according to the detected activity type and re-output it. Obtain the activity duration value again based on the updated activity output signal and the reference duration range.

[0174] Based on the detection activity type, the game planner of the corresponding game type generates a new activity output signal and re-outputs the game activity. The login information after the activity output is then re-executed within the baseline duration range of the output activity to obtain a new activity duration value.

[0175] S94: Update the user profile based on the activity duration value and the detection activity type.

[0176] Continue outputting the original user profile by referring to the steps in S83 and S84 for the activity duration value, or rebuild a new user profile after adding the detection activity type in the game type.

[0177] Based on the same inventive concept, embodiments of the present invention provide a system for constructing precise user profiles for a game platform, comprising: The acquisition module is used to retrieve login information; The memory is used to store a program for constructing a precise user profile for a game platform; The processor is used to load and execute programs stored in memory.

[0178] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0179] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for constructing precise user profiles on a game platform, characterized in that, include: Collect user login information from the game platform; Retrieve the login time and login address from the login information; The login duration is obtained based on the login information and login time. Determine if adjacent login addresses are identical; If they match, the game type is retrieved from the game platform using the login duration and login information, and a user profile is built based on the game type. If they are inconsistent, adjacent but inconsistent login addresses will be defined as abnormal login addresses; Obtain the geographical route based on the abnormal login address; User profiles are built based on geographical routes, login duration, and login information.

2. The method for constructing precise user profiles for a game platform according to claim 1, characterized in that, Methods for building user profiles based on geographic routes, login duration, and login information include: Traffic type is determined based on geographical routes; The login date is obtained based on the login time point of the abnormal login address, and the login duration of the abnormal login address is defined as the abnormal login duration; The traffic type is determined by combining the duration of abnormal logins with the traffic type. Average spending information is obtained based on the marked traffic type and login date; Retrieve the consumption information corresponding to the login information from the game platform; Login information is updated based on consumption information and average consumption information, and the game type is retrieved again based on the updated login information and login duration to build a user profile.

3. The method for constructing precise user profiles for a game platform according to claim 2, characterized in that, Methods for updating login information based on consumption information and average consumption information include: The consumption coefficient is obtained based on the login duration and login information; Consumption information is derived from the product of the consumption coefficient and consumption information. Determine whether the detected consumption information is greater than the average consumption information; If the value is greater than the original value, continue outputting the original login information. If it is not greater than, then consumption trends can be obtained through consumption information; Login information will be updated based on consumption trends and abnormal login addresses.

4. The method for constructing precise user profiles for a game platform according to claim 3, characterized in that, Methods for updating login information based on consumption trends and unusual login addresses include: The difference between the detected consumption information and the average consumption information is calculated as the consumption deviation value; The duration of trend decline was determined based on consumption trends, abnormal login addresses, and login times. The amount of consumption decrease is obtained based on consumption trends and the duration of the trend decline. The product of the decrease in consumption and the duration of the downward trend is used as the accumulated consumption amount. The login information is updated based on the difference between the accumulated spending and the spending deviation value.

5. The method for constructing precise user profiles for a game platform according to claim 4, characterized in that, The methods after updating login information include: The arrival time is obtained by combining the marked traffic type with the abnormal login address; The login time period is determined based on the landing time. The landing and login information is obtained based on the landing time and login information; Baseline login information is derived based on login information and login time period; Update the login information by comparing the actual login information with the baseline login information.

6. The method for constructing precise user profiles for a game platform according to claim 5, characterized in that, Methods after updating login information also include: When the login time period is a preset baseline type, the average duration of the marker is obtained based on the login time period and the login duration. Retrieve the marked login time points that exceed the average marked duration from the landing login information; Retrieve the initial login time from the login information; The baseline time point is obtained based on the marked traffic type and the initial login time point; The duration between the baseline time point and the marked login time point is calculated as the target interval duration. The baseline interval range is obtained by combining the login time period and the baseline type. Update login information by comparing the target interval duration with the baseline interval range.

7. The method for constructing precise user profiles for a game platform according to claim 1, characterized in that, Methods for obtaining user profiles include: In response to the activity output signal, retrieve the activity item type based on the activity output signal and the game type; The baseline duration range is determined based on the type of event item. Use the login time points within the benchmark duration range as the calibration time points; Update user profiles by combining activity output signals, login information, and baseline duration range.

8. The method for constructing precise user profiles for a game platform according to claim 7, characterized in that, Methods for updating user profiles include: The average activity value is obtained by combining the activity output signal with login information. The activity duration value is obtained based on the baseline duration range and login information; Determine if the activity duration is less than the activity average. If it is less than that, then remove the event item type from the game type to rebuild a new user profile; If the value is not less than the specified value, the user profile will be updated based on the game type and the type of event items.

9. The method for constructing precise user profiles for a game platform according to claim 8, characterized in that, Methods after obtaining user profiles include: Retrieve a base single type from the game genres; The number of items that overlap with the baseline single type is obtained by comparing the activity item type with the standard single type. When the number of overlaps exceeds 1, the type of activity item that overlaps with the baseline single type is used as the activity type to be detected. The activity output signal is updated based on the activity type and re-output. The activity duration value is then obtained again based on the updated activity output signal and the reference duration range. Update user profiles based on activity duration and activity type.

10. A system for constructing precise user profiles for a game platform, characterized in that, include: The acquisition module is used to retrieve login information; A memory for storing a program that implements a method for constructing accurate user profiles for a game platform as described in any one of claims 1 to 9; The processor is used to load and execute programs stored in memory.

Images