Member management platform based on cloud

Abstract

The invention relates to the technical field of member information processing, in particular to a cloud-based member management platform, which comprises an information synchronization integration module, a data conflict judgment module, a track behavior collection module, an interest trend judgment module and a service intelligent push module. According to the method, through database retrieval and field comparison, member identity information is automatically corrected and unified, repetition and conflicts caused by multi-source data are completely eradicated, a member archive structure is combined with dynamic adjustment and historical data combination, uniqueness and traceability are enhanced, and in the data screening process, the data screening efficiency is improved. Dominant information is comprehensively identified depending on credibility levels, business attributes and parameter association, the situation that a single time sequence or source affects data accuracy is avoided, circulation of all parks and service items is tracked in real time in the member behavior management link, regional switching and service sequences are periodically combed, behavior links are finely restored, interest judgment is fused with multiple time sequence behavior changes, and the management efficiency is improved. The interest flow direction is actively recognized, and dynamic trend monitoring is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of member information processing, and in particular to a cloud-based member management platform. Background Art

[0002] Member information processing refers to the related technologies for collecting, storing, managing, analyzing and applying various types of member data through information technology means. This field covers member identity authentication, information security, data synchronization, member rights management, consumer behavior analysis, precision marketing, points system, customer relationship maintenance and other contents. With the development of the Internet and cloud computing, member information processing not only requires efficient data processing capabilities, but also focuses on the real-time, reliability and cross-platform application capabilities of data. It is widely used in commercial retail, entertainment and leisure, service industry, fitness, education and other industries.

[0003] Among them, the park member information management platform refers to a member information management platform built using cloud computing technology for various theme parks, electric carnivals, outdoor playgrounds and other park operating scenarios. The platform aims to achieve centralized management and multi-terminal synchronization of park member data, covering member registration, identity authentication, information maintenance, points and rights management, consumption data analysis, member interaction and event push and other functions, helping park operators to improve member service quality, optimize operational decisions, and enhance member stickiness and experience.

[0004] Existing technologies often rely on static information merging and manual archiving, resulting in duplication, omissions or identity mismatches in member files, making it difficult to establish a complete and consistent member information chain. Data screening methods are limited to update time or a single weight, and lack comprehensive judgment of multiple parameters, resulting in key data being easily overlooked or misselected, reducing data credibility. The member behavior recording method is single, and regional circulation and service experience trajectories are fragmented, making it difficult to capture interest dynamics and preference migration, resulting in a misalignment between recommended content and members' actual needs. The data mobility among multiple parks is high, and static integration methods are insufficient to support real-time rights management and push, causing operators to face management pressure and service fragmentation. Summary of the Invention

[0005] In order to solve the problems existing in the prior art, such as often relying on static information merging and manual archiving, resulting in duplication, omission or identity mismatch in member files, difficulty in establishing a complete and consistent member information chain, data screening methods limited to update time or single weight, lack of comprehensive judgment of multiple parameters, resulting in key data being easily ignored or misselected, reducing data credibility, single member behavior recording method, regional flow and service experience trajectory are fragmented, difficult to capture interest dynamics and preference migration, resulting in misalignment between recommended content and actual member needs, high data mobility in multiple parks, and static integration methods insufficient to support real-time rights management and push, resulting in operators facing management pressure and service fragmentation. The embodiment of the present invention provides a cloud-based member management platform. The technical solution is as follows:

[0006] On the one hand, it provides a cloud-based membership management platform, including:

[0007] The information synchronization and integration module is based on the member information of multiple parks. It searches the database to match names and contact information, determines whether the unique identification data conflicts, completes the information merger, adjusts the file structure, removes duplicates, and obtains a unique member file set;

[0008] The data conflict determination module compares the contact information, membership level and historical consumption amount based on the unique member profile set, determines the update time and the credibility of the data source, selects the priority parameter items, and obtains the conflict preferred parameter group;

[0009] The trajectory behavior collection module analyzes the member area switching behavior based on the conflict optimization parameter group, collects the area entry and exit times through behavior tracking, calculates the area switching frequency, organizes the access time points of each project, optimizes the time sequence, and obtains the periodic service trajectory sequence;

[0010] The interest trend identification module compares the item type switching interval and the area switching frequency based on the periodic service trajectory sequence, analyzes the parameter trend of the continuous time period, determines whether the interest shifts to a new type, filters the related data segments, and obtains the interest direction transfer characteristics.

[0011] On the other hand, the conflict preference parameter group includes the latest contact information, the latest membership level, the latest consumption amount, and the highest trusted source. The periodic service trajectory sequence includes timed regional flow, project access sequence, and service usage distribution. The interest direction transfer characteristics include changes in preferred projects, service selection trends, and activity participation directions.

[0012] On the other hand, the information synchronization and integration module includes:

[0013] The unique identification screening submodule is based on the member information of multiple parks. It matches the name and contact information fields with the unique identification of the unique number, ID number and mobile phone number in the database, determines the unique identification of each member data across multiple parks, and generates the number of unique identifications.

[0014] The field conflict detection submodule compares the data contents under the same unique identifier based on the number of unique attribution identifiers, and detects whether there is a name or contact information conflict between different parks or databases by judging the difference in name content and contact information, and obtains the number of conflicting items;

[0015] The information merging and adjusting submodule merges the conflicting member information according to the number of conflicting items, readjusts the data structure of the member files in the database, optimizes duplicate information, and obtains a unique member file set through field deduplication and structural adjustment.

[0016] On the other hand, the data conflict determination module includes:

[0017] The information update judgment submodule, based on the unique member profile set, compares the update time of each contact information, member level, and historical consumption amount, determines the update order of the same member information in the difference records, and identifies the distribution of the update time of each parameter to obtain the update time distribution interval;

[0018] The source trust screening submodule compares the trust levels of multiple data sources of the same member based on the update time distribution interval and the trust level of each information source, screens the data items with the best trust level, and obtains the number of trustworthy sources;

[0019] The preferred parameter generation submodule selects parameter items with timeliness and priority according to the number of trusted sources, the update time of each data and the source trust level, and obtains a conflicting preferred parameter group.

[0020] On the other hand, the trajectory behavior collection module includes:

[0021] The area handover collection submodule collects the area entry and exit time information of each member within the period based on the conflict optimization parameter group, records the movement process of members between areas, and establishes the cross-area handover information of each member;

[0022] The switching frequency calculation submodule counts the number of times the same member switches between different regions within a period based on the cross-region switching information, and calculates the region switching frequency of each member within the period;

[0023] The behavior sequence arrangement submodule calls the time point data of each park project visit according to the frequency of the area switching, arranges the behavior of each member in chronological order, optimizes the arrangement of the behavior data on the time axis, and obtains a periodic service trajectory sequence.

[0024] On the other hand, the frequency of region switching within a period of each member is calculated using the formula:

[0025]

[0026] Get the periodic area switching frequency characteristic value of the i-th member Among them, T p Represents the total time of the statistical period, n i Represents the total number of cross-region switching records generated by the i-th member in this period, represents the actual time interval experienced by the i-th member during the j-th switch, represents the spatial weight difference between the regions crossed by the i-th member in the j-th switch, represents the congestion coefficient of the corresponding area of ​​the i-th member in the j-th handover, Represents the average of the congestion coefficients of all zone switching records of the i-th member in the period.

[0027] On the other hand, the interest trend identification module includes:

[0028] The switching interval calculation submodule calculates the time interval between two consecutive project type changes based on the periodic service trajectory sequence and the project type switching time point of each member, and calculates the switching frequency changes between each area to obtain the type switching interval;

[0029] The change trend determination submodule calls the project type switching interval and the area switching frequency based on the type switching interval, analyzes the parameter change trend in the continuous time period, and obtains the interest change trend by comparing the project type changes and area flow in the different time periods;

[0030] The interest transfer feature screening submodule calls the project type and area switching distribution in each time period according to the interest change trend, compares the change direction of interest in consecutive time periods, screens the data segments associated with the interest change, and obtains the interest direction transfer feature.

[0031] On the other hand, the direction of change of the contrast interest in consecutive time periods is expressed using the formula:

[0032]

[0033] Calculate the interest change feature value ΔI, filter the data segments associated with interest changes, and obtain the interest direction transfer feature, where: represents the participation frequency of the kth project type in time period t, represents the participation frequency of the kth project type in time period t+1, represents the switching frequency of the kth region in time period t, represents the switching frequency of the kth region in time period t+1, n Q Represents the total number of project types or areas analyzed.

[0034] In another aspect, the platform further comprises:

[0035] The service intelligent push module screens the relevant service categories based on the interest direction transfer characteristics, determines the park service items that the member has not experienced, analyzes the correlation between the items and the current interest change trend, optimizes the recommendation order, and organizes the service content to obtain the point of interest push data;

[0036] The POI push data includes recommendations for interesting services, a list of unexperienced items, and personalized push configurations.

[0037] On the other hand, the service intelligent push module includes:

[0038] The service category screening submodule screens the park service categories associated with the interest change direction based on the interest direction transfer characteristics, counts the number of service categories involved, and obtains the service type distribution;

[0039] The experience difference judgment submodule compares and filters service items that have not been recorded as experiences based on the distribution of service types and the list of experienced items of each member and the service category, and counts the total number of unexperienced items corresponding to each member to obtain the number of unexperienced items;

[0040] The recommendation order optimization submodule analyzes the correlation between the unexperienced items and the interest trend based on the number of unexperienced items, the interest change trend and the item type, sorts the unexperienced items according to the interest transfer direction, and organizes the service content to obtain interest point push data.

[0041] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0042] Through database retrieval and field comparison, member identity information is automatically corrected and unified to eliminate duplication and conflict caused by multi-source data. The member file structure is combined with dynamic adjustment and historical data merging to enhance uniqueness and traceability. During the data screening process, the trust level, business attributes and parameter associations are relied upon to comprehensively identify the dominant information to avoid the impact of a single time series or source on data accuracy. The member behavior management link tracks the flow of various parks and service projects in real time, periodically sorts out regional switching and service sequence, and accurately restores the behavior chain. Interest discrimination integrates multiple time series behavior changes, actively identifies interest flows, realizes dynamic trend monitoring, and pushes personalized intelligent matching of interest tags and unexperienced projects. Recommended content is dynamically generated, which improves the high degree of adaptation between recommendations and actual interests. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 Schematic diagram of the platform of the present invention;

[0045] Figure 2 Schematic diagram of the platform framework of the present invention;

[0046] Figure 3 This is a flow chart of the information synchronization integration module of the present invention;

[0047] Figure 4 This is a flow chart of the data conflict determination module of the present invention;

[0048] Figure 5 This is a flow chart of the trajectory behavior collection module of the present invention;

[0049] Figure 6 This is a flow chart of the interest trend identification module of the present invention;

[0050] Figure 7 This is a flow chart of the intelligent push module of the present invention. DETAILED DESCRIPTION

[0051] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0052] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0053] In the embodiments of the present invention, the terms "image" and "picture" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same. The terms "of," "corresponding," and "corresponding" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same.

[0054] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0055] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0056] The embodiment of the present invention provides a cloud-based member management platform, such as Figure 1 As shown, the platform includes:

[0057] The information synchronization and integration module uses database search functions based on multi-park member information to match names and contact information fields, compare data content with the same unique identifier, determine whether there are name or contact information conflicts, perform information merging, and adjust the member file data structure to optimize duplicate items to obtain a unique member file set;

[0058] The data conflict determination module compares each contact information, membership level, and historical spending amount based on a unique member profile set. It determines the update time by sequence, compares the credibility of the source, analyzes the data content of each amount and membership level, and selects parameters with timeliness and priority to obtain the conflict-preferred parameter group.

[0059] The trajectory behavior collection module analyzes member area switching information based on the conflict optimization parameter group, collects the area entry and exit times within the cycle through behavior tracking, calculates the switching frequency between different areas, organizes the behavior data based on the time of each visit to the park project, optimizes the time series arrangement, and obtains the periodic service trajectory sequence;

[0060] The interest trend identification module compares the changes in item type switching intervals and area switching frequencies based on the periodic service trajectory sequence, analyzes the parameter change trends within continuous time periods, determines whether interest has shifted from the original item type to the new item type, and screens data segments associated with interest changes to obtain interest direction shift characteristics;

[0061] The service intelligent push module is based on the interest direction transfer characteristics, screens the service categories involved, determines the park service items that members have not experienced, analyzes the correlation between the items and the current interest change trends, optimizes the recommendation order, and organizes the service content to obtain interest point push data.

[0062] The conflict optimization parameter group includes the latest contact information, the latest membership level, the latest consumption amount, and the highest trusted source. The periodic service trajectory sequence includes the timed regional flow, project access sequence, and service usage distribution. The interest direction transfer characteristics include changes in preferred projects, service selection trends, and activity participation directions. The interest point push data includes interest service recommendations, a list of unexperienced projects, and personalized push configuration.

[0063] In Module 1, the same unique identifier refers to the unique number, ID number, mobile phone number and other fields generated when a member registers in multiple parks or databases, which are used as a mark to uniquely identify a member's identity. The basis for determining whether it is the same member, such as mobile phone number, ID number, and automatically assigned unique ID number; the member file data structure refers to the data table structure and fields used when storing member information in the database, including the storage format and arrangement of data such as name, contact information, registration time, member level, and historical consumption, such as database table structure design, field order and type, and member information stored in JSON format. In Module 2, timeliness and priority refer to the priority selection of the latest data (timeliness) among multiple parameter records about the same member, and the selection of data that is highly reliable or should be adopted first in business according to pre-set rules (priority), such as giving priority to the most recently updated contact information and data provided by the headquarters rather than the branch park. In Module 3, member area switching information refers to the movement, entry, exit, and switching records of members in different parks, different venues, venues or functional areas, which are usually automatically collected by devices such as card swiping, code scanning, and access control systems, such as the entry and exit records of moving from the water park area to the electric carnival area; the switching frequency between areas refers to the number of times a member crosses different parks or different amusement areas within a certain period of time, which is used to measure member activity and mobility characteristics, such as switching from area A to area B 4 times in one day; visiting park projects refers to the behavior of members entering specific amusement projects, facilities, venues, activities, etc., and records the actual participation of members in various entertainment projects, such as the check-in records of members entering the carousel in the morning and the bumper car in the afternoon. In Module 4, the project type switching interval refers to the time interval between two consecutive participations of members in different types of amusement projects, which is used to reflect the speed of changes in members' experience content, such as the time period between members switching from "water projects" to "electric projects"; the area switching frequency refers to the distribution of the number of times members switch areas within a certain period of time, which is used to analyze the scope of members' activities and preferences. For example, members move between different parks multiple times a day, which is a high frequency; the parameter change trend refers to the analysis of the above-mentioned switching interval, frequency and other data to determine the changing direction or pattern of members' behavior in the time series, reflecting the dynamic adjustment of members' interests and behaviors. For example, in the past week, members have shifted from preferring a single area to playing in multiple areas, showing an expansion trend.

[0064] like Figure 2 and Figure 3 As shown, the information synchronization and integration module includes:

[0065] The unique identification screening submodule is based on the member information of multiple parks. It matches the name and contact information fields with the unique identification of the unique number, ID number and mobile phone number in the database, determines the unique identification of each member data across multiple parks, and generates the number of unique identifications.

[0066] For each member data in the park database, the name, contact information, unique number, ID number and mobile phone number fields are extracted one by one. First, the extracted name field is directly matched with the unique identifier in the database, and the name field content is compared with the unique number in turn to see if they are the same. If the unique number in a record is exactly the same as that in other records in the database, for example, the member's unique number in Park A is "U001" and the unique number in Park B is also "U001", then the system determines that these two records belong to the same member. Next, the mobile phone number field is further compared, and the mobile phone number is matched from the first to the last digit. If the mobile phone numbers are exactly the same, for example, "1 If "3812341234" exists in the data of both parks and is exactly the same, it is confirmed that the two records belong to the same member, and the mobile phone number is included in the same unique identifier attribution set. For records with different mobile phone numbers but the same unique number or ID number, the system still regards them as part of the unique identifier attribution set and continues to perform subsequent detection operations. The above matching process is repeated for the data of all parks. When all data are matched, the number of unique attribution identifiers is generated based on the identified unique identifier set. For example, there are 1,200 member records in the entire database. After screening, 400 different unique identifier attribution sets are found, and the number of unique attribution identifiers generated is 400.

[0067] The field conflict detection submodule compares the data contents under the same unique identifier based on the number of unique attribution identifiers. By judging the differences in name content and contact information, it detects whether there are conflicts in names or contact information between different parks or databases, and obtains the number of conflicting items.

[0068] First, perform a character-by-character comparison on the name fields of all records in the set. For example, in a certain set of records, there are two records named "Li Lei" and "Li Lei". First, compare the first character "Li" with "Li". After they are the same, continue to compare the second character "Lei" with "Lei" and find that there is a difference, so it is marked as a name conflict. Subsequently, perform the same character-by-character comparison operation on the contact information fields in the same group of data. For example, the contact information fields are "13900001111" and "13900001112" respectively. Compare each digit from the first to the last digit one by one. When comparing the last digit "1" with "2", a difference is detected and it is marked as a contact information conflict. Repeat the above comparison operations on the name fields and contact information fields for all data in each unique identifier set. For records with differences, accumulate the number of conflict items. For example, there are 3 records in a certain set. By pairwise comparison of the name fields, 2 name conflicts are found, and by comparison of the contact information fields, 1 contact information conflict is found. Then the cumulative number of conflict items in this set is 3. Traverse all unique identifier sets and finally summarize all the numbers of conflict items. For example, within the entire database range, 600 name conflicts and 400 contact information conflicts are detected cumulatively, so the total number of conflict items is 1000.

[0069] The information merging and adjustment sub-module merges the conflicting member information according to the number of conflict items, re-adjusts the data structure of the member files in the database, optimizes the duplicate information, and obtains a unique set of member files through field deduplication and structure adjustment;

[0070] Perform merging processing on the conflicting member information group by group. First, for each group of conflicting unique identifier sets, extract all the different name fields and contact information fields, and perform screening processing according to the field deduplication rules. For example, in a certain group of sets, there are names "Wang Wu" and "Wang Wu". According to the character length and common Chinese character standards, preferentially select the name "Wang Wu" with a shorter character length and meeting the common standards as the retained name field after merging. Subsequently, compare the latest update dates of the contact information fields. For example, the contact information "13600001111" in this group was last updated on May 1, 2025, while "13600002222" was last updated on December 30, 2024. Directly compare the year, month, and day of the two dates, and confirm that "May 1, 2025" is later than "December 30, 2024". Therefore, preferentially retain "13600001111" as the contact information field after merging. After completing the field deduplication, adjust the data structure of the member file, delete other duplicate or conflicting fields in the set, only retain the above-screened name and contact information, and update the member file record again. For example, there were originally 4 records in the set, and only 1 record containing the deduplicated fields was retained after merging. The system writes the updated unique member file into the database, and repeats the above operations for all conflicting unique identifier sets in sequence. Finally, complete the merging and adjustment of all member information. For example, among 1200 records in the entire database, there are 400 groups of conflicts. After the merging and adjustment are completed, a total of 800 unique member files are generated.

[0071] As Figure 2 and Figure 4 shown, the data conflict determination module includes:

[0072] The information update judgment sub-module is based on the unique member file set. For each contact information, membership level, and historical consumption amount, by comparing their respective update times, judge the update order of the same member information in the different records, and identify the new and old distribution of the update times of each parameter to obtain the update time distribution interval;

[0073] First, extract the update time fields of the contact information, membership level, and historical consumption amount in each file one by one. For example, if a member's contact information is updated on May 1, 2025, the membership level is updated on April 20, 2025, and the historical consumption amount is updated on May 5, 2025, perform a pairwise comparison of the update time of the above three parameters. First, compare the update time of the contact information and membership level to determine that "May 1, 2025" is later than "April 20, 2025", and then continue to compare the contact information and historical consumption. The update time of the amount is determined to be later than "May 1, 2025". Based on this, the update time of the historical consumption amount is determined to be the latest, the contact information is second, and the membership level is the oldest. Then, the update time of each parameter is recorded, and the update time distribution interval of the member is defined based on the granularity of year, month, and day. For example, the earliest update time of the member is April 20, 2025, and the latest update time is May 5, 2025. Therefore, the update time distribution interval of the member's parameters is "April 20, 2025 to May 5, 2025".

[0074] The source trust screening submodule compares the trust levels of multiple data sources of the same member based on the update time distribution interval and the trust level of each information source, and selects the data items with the best trust level to obtain the number of trustworthy sources;

[0075] Extract the source field of each data. For example, a member's contact information comes from the "headquarters", the membership level comes from the "branch A", and the historical consumption amount comes from the "branch B". Compare each source field with the preset trust level rule in turn. The rule is set to the trust level of 3 for the headquarters, 2 for branch A, and 1 for branch B. Then, perform numerical comparison on the trust levels of the contact information, membership level, and historical consumption amount. First, judge that the trust level of the contact information source is 3, which is the highest level. Then judge that the trust level of the membership level source is 2. Finally, judge that the trust level of the historical consumption amount source is 1, which is the lowest level. Based on this, filter out the data items with the best trust level. For example, the contact information comes from the headquarters and has the highest trust level. Therefore, the contact information is identified as the most trustworthy data in the member file. Perform the same trustworthy source screening operation on all member files in turn, and summarize the number of trustworthy sources in the entire database.

[0076] The optimal parameter generation submodule selects the timeliness and priority parameters according to the number of trusted sources, the update time of each data item and the source trust level, and obtains the conflicting optimal parameter group;

[0077] First, for each parameter data of each member profile, the update time is compared in turn, and the latest data item is selected. For example, if a member's contact information has two records, updated on April 25, 2025, and May 1, 2025, respectively, the system directly selects the data on May 1, 2025 as the candidate. Then, it determines whether the source of this record has the highest credibility level. For example, if the source is headquarters and the credibility level is 3, it meets the optimal conditions, and this record is directly used as the preferred parameter. If the credibility level of the candidate record's source is low, the system continues to search for data with a higher credibility level in the next newest record. For example, if the next newest record comes from headquarters and is updated on April 25, 2025, this record is preferred. Then, the same screening process is performed on the member level and historical consumption amount in sequence. The latest and most credible data of each parameter are combined to generate a conflict-preferred parameter group, such as the contact information "13800009999", the member level "Gold Member", and the historical consumption amount "6000 yuan".

[0078] like Figure 2 and Figure 5 As shown, the trajectory behavior collection module includes:

[0079] The regional handover collection submodule collects the regional entry and exit time information of each member within the period based on the conflict optimization parameter group, records the movement process of members between regions, and establishes the cross-region handover information of each member;

[0080] First, extract the entry and exit times of each member in each park or functional area. For example, a member enters the children's play area at 10:00 on May 3, 2025 by swiping a card, leaves at 12:00, then enters the food and rest area at 14:00, and leaves at 15:30. First, extract all the entry and exit time records of the member involved, and match each entry record with the exit record one by one to ensure that the entry and exit records of the same visit cycle are complete. If an unmatched situation is found, such as an entry record but no exit time, the record will be marked as an abnormal state and will not be included in the scope of this data processing. Next, arrange all the records in chronological order. The effective entry and exit times are calculated, for example, the arrangement results are "Children's Play Area: 10:00-12:00", "Food Rest Area: 14:00-15:30", and then the member's movement process between areas is identified. By comparing the departure time of the previous area with the entry time of the next area, the interval between areas is calculated. For example, the departure time of the children's play area is 12:00, and the entry time of the food rest area is 14:00, and the interval between the two is 2 hours. Based on this, it is marked that the member has switched from the children's play area to the food rest area within this period. All entry and exit records of all members are traversed in turn to establish the cross-area switching information of each member.

[0081] The switching frequency calculation submodule counts the number of times the same member switches between different regions within a period based on the cross-region switching information, and calculates the region switching frequency of each member within the period;

[0082] Calculate the frequency of region switching for each member within a period using the formula:

[0083]

[0084] Get the periodic area switching frequency characteristic value of the i-th member Switching refers to moving, specifically referring to a member moving from Park A to Park B, where T p Represents the total time of the statistical period, n i Represents the total number of cross-region switching records generated by the i-th member in this period, represents the actual time interval experienced by the i-th member during the j-th switch, represents the spatial weight difference between the regions crossed by the i-th member in the j-th switch, represents the congestion coefficient of the corresponding area of ​​the i-th member in the j-th handover, represents the mean of the congestion coefficient of all regional switching records of the i-th member in the period;

[0085] The regional switching frequency characteristic value refers to a numerical indicator that can quantify the frequency and activity level of members' cross-regional movement by analyzing the actual cross-regional movement of individual members between different regions within the statistical period, comprehensively considering the time interval of each switch, the spatial weight differences between regions, and the relationship between the congestion coefficient of the switching region and the period mean. It reflects the intensity and regularity of members' cross-regional activities within the set period and is an important parameter to characterize the activity of their travel behavior and changes in regional preferences.

[0086] In order to avoid the negative differences from canceling each other out and improve the stability of the feature quantity, the square form of the crowding coefficient difference is used in the formula, T p The total time of the statistical cycle is determined by the cycle parameters set by the member management platform. Usually a 30-day service analysis cycle is selected, so T p =30 days;

[0087] n i The total number of cross-region switching records generated by the i-th member during the period is calculated based on the access card swiping, code scanning, and venue entry and exit system log records. According to system statistics, the member has three cross-region switchings within 30 days, so n i =3;

[0088] The actual time interval experienced by the i-th member during the j-th switch is calculated using the member's entry and exit timestamps before and after the switch. For example:

[0089] The first switch: From Park A to Park B, the time difference between entry and exit is 2 days, so sky;

[0090] The second switch: from Park B to Park C, the time difference between entry and exit is 5 days, so sky;

[0091] The third switch: from Park C to Park D, the time difference between entry and exit is 3 days, so sky;

[0092] The spatial weight difference between the areas crossed by the i-th member during the j-th switch is calculated based on factors such as the area, facility density, and passenger flow carrying capacity of each area. It is obtained by querying the regional indicator database maintained by the venue operation system. For example:

[0093] The spatial weight of Park A is 0.8, the spatial weight of Park B is 0.5, and the spatial weight difference is

[0094] The spatial weight of Park B is 0.5, the spatial weight of Park C is 0.7, and the spatial weight difference

[0095] The spatial weight of Park C is 0.7, the spatial weight of Park D is 0.3, and the spatial weight difference

[0096] The congestion coefficient of the corresponding area of ​​the i-th member in the j-th handover is calculated based on the passenger flow density data collected by the real-time passenger flow monitoring system in each area. For example:

[0097] The real-time crowd density of the target area B for the first switch is 1.0 person / square meter, so

[0098] The real-time crowd density of the second switching target area C is 1.2 people / square meter, so

[0099] The real-time crowd density of the target area D for the third switch is 1.1 people / square meter, so

[0100] The average of the crowding coefficients of all zone switching records of the i-th member in the period is obtained by calculating the average of the crowding coefficients of each switching record of the member:

[0101]

[0102] Calculate each switch item:

[0103] First switch:

[0104]

[0105] Second switch:

[0106]

[0107] The third switch:

[0108]

[0109] Calculate the square roots of each term:

[0110]

[0111] Sum:

[0112] 2.3+5.2+3.4=10.9;

[0113] Substituting into the formula:

[0114]

[0115] The results show that the regional switching frequency characteristic value of the i-th member within a 30-day period is 0.3633, indicating that the member has certain cross-regional active characteristics during this statistical period. The higher the value, the higher the frequency and intensity of cross-regional movement. This characteristic value is used as a quantitative indicator of the trajectory behavior aggregation module to provide a basic behavioral dynamic parameter basis for the subsequent process.

[0116] The behavior sequence arrangement submodule calls the time point data of each park project visit based on the frequency of regional switching, arranges the behavior of each member in chronological order, optimizes the arrangement of behavior data on the time axis, and obtains a periodic service trajectory sequence;

[0117] Call the specific time point of each visit of the member to the park project. For example, a member enters the slide area at 10:15 on May 3, 2025, enters the trampoline area at 10:50, enters the dining area at 14:15, and enters the performance area at 15:00. First, extract the entry time point information of all projects, and then arrange all the extracted time point data in chronological order to confirm that the access sequence is "slide area → trampoline area → dining area → performance area". Next, check the continuity of the access time to determine whether there are multiple visits in the same time period. For example, if the slide area access time record is from 10:15 to 10 :45, while the trampoline area visit time was recorded as 10:50 to 11:20. The interval between the time of leaving the slide area and the time of entering the trampoline area was calculated to be 5 minutes. It was determined that the member had a short rest or moved between the two visits, and there was no overlapping visit. Subsequently, all visit records were optimized and arranged in time series at the minute level to ensure the integrity and continuity of the entire behavior sequence on the time axis. For example, the final behavior sequence was organized as "10:15 slide area → 10:50 trampoline area → 14:15 dining area → 15:00 performance area", and the periodic service trajectory sequence of the member within the defined period was obtained.

[0118] like Figure 2 and Figure 6 As shown, the interest trend identification module includes:

[0119] The switching interval calculation submodule is based on the periodic service trajectory sequence. It calculates the time interval between two consecutive project type changes for each member's project type switching point, and statistics the switching frequency changes between each area to obtain the type switching interval.

[0120] The switching time points of each project type for each member are extracted one by one. For example, a member switches from the parent-child amusement area to the dining area at 10:00 on May 4, 2025, and switches from the dining area to the performance area at 12:30. The time points of each project type switching are extracted in turn and arranged in chronological order. Then, the time intervals between two consecutive project type changes are calculated one by one. Specifically, the interval length is determined by subtracting the previous switching time point from the latter switching time point. For example, the performance area switching time point of 12:30 minus the dining area switching time point of 10:00 is obtained to obtain a time interval of 2 hours and 30 minutes. The switching frequency of the member between different areas within a certain period is counted. For example, the member switches from the parent-child amusement area to the dining area within one day. There is one switch from the sub-amusement area to the dining area, and one switch from the dining area to the performance area, for a total of two switches. All switching frequencies are accumulated. If a member switches across areas more than five times in a day, it is recorded as a high-frequency switching interval. Otherwise, if it is less than or equal to two times, it is recorded as a low-frequency switching interval. The specific classification standard is set as 0-2 times as low frequency, 3-5 times as medium frequency, and 6 times or more as high frequency. Combining all switching intervals and switching frequencies, the type switching interval of the member is generated. For example, the type switching interval of the member on May 4, 2025 is "Parent-child amusement area → dining area: 2 hours" and "Dining area → performance area: 2 hours and 30 minutes", with a switching frequency of 2 times, which is classified as a low-frequency switching interval.

[0121] The change trend determination submodule is based on the type switching interval, calls the project type switching interval and the region switching frequency, analyzes the parameter change trend in the continuous time period, and obtains the interest change trend by comparing the project type changes and regional flow in different time periods;

[0122] For each member, the data of all project type switching intervals and area switching frequencies are called. For example, the project type switching intervals of a member on May 4 were 2 hours and 2 hours and 30 minutes respectively, and the area switching frequency was 2 times a day. The data were analyzed in sequence. First, the changes in project types in consecutive time periods were compared. Specifically, by comparing whether there were obvious changes in project types in consecutive days, for example, the member only visited the parent-child amusement area on May 3 and did not switch across types. On May 4, he switched from the parent-child amusement area to the dining area and then to the performance area. Based on this, it was determined whether the project types changed significantly in consecutive time periods. A change occurred from a single region to multiple regions within the same period. Subsequently, the regional flow within the same time period was compared. For example, the member did not switch regions on May 3, and the regional switching frequency was 2 times on May 4. The regional switching frequencies of the two days were compared and it was determined that the switching frequency had changed from 0 times to 2 times. Then, a comprehensive analysis was made of the common trends of project type changes and regional flow. According to the above comparison results, it was judged that the member's interest change trend was "from a single-region fixed activity to multi-region and multi-type activities". Finally, the member's interest change trend was recorded as "diversification tendency".

[0123] The interest transfer feature screening submodule calls the project type and region switching distribution in each time period according to the interest change trend, compares the change direction of interest in consecutive time periods, and screens the data segments associated with interest changes to obtain interest direction transfer features;

[0124] To compare the direction of interest changes in consecutive time periods, use the formula:

[0125]

[0126] Calculate the interest change feature value ΔI, filter the data segments associated with interest changes, and obtain the interest direction transfer feature, where: represents the participation frequency of the kth project type in time period t, represents the participation frequency of the kth project type in time period t+1, represents the switching frequency of the kth region in time period t, represents the switching frequency of the kth region in time period t+1, n Q Represents the total number of project types or areas analyzed;

[0127] The interest change characteristic value refers to the degree of change in the frequency of users' participation in different project types and the frequency of switching between different regions within two consecutive time periods. By quantifying the magnitude of the change, a specific numerical value is used to reflect the intensity of changes in user interest behavior over a period of time.

[0128] The interest change characteristic value takes the user's active frequency in various projects and the switching frequency in different areas as the observation objects, and calculates a comprehensive value by comparing the numerical differences of these data between consecutive time periods. This value can be used to determine whether there is a significant change or transfer in the user's interest points (such as the type of projects participated in and the active areas). For example, if a user mainly participates in projects A and B in the previous period, and frequently participates in projects C and D in the next period, and there are also major changes in the activity areas, then the interest change characteristic value will be larger. The higher the value, the more significant the transfer or change of the user's interest points; the lower the value, the basically stable interest structure and the small change.

[0129] Project Type Participation Frequency and Automatically collect data on members' card swiping records, code scanning and other behavioral data on each project, and count the number of entries into each project within the statistical period, quantifying it as participation frequency and area switching frequency. and The access control system and positioning equipment are used to collect the members’ entry and exit data in different areas in real time, and the number of switching times in each area is counted. The total number of project types or areas analyzed is n QThe system automatically identifies and collects statistics based on the project types and number of regions involved during the actual monitoring period. Taking the monitoring data of a member during the analysis period as an example, the specific values ​​collected by the system are as follows:

[0130] n Q =3, that is, there are three project types or regions involved in the analysis, numbered k=1, 2, 3;

[0131] Project Type Participation Frequency:

[0132] Second-rate, Second-rate;

[0133] Second-rate, Second-rate;

[0134] Second-rate, Second-rate;

[0135] Frequency of region switching:

[0136] Second-rate, Second-rate;

[0137] Second-rate, Second-rate;

[0138] Second-rate, Second-rate;

[0139] Computing molecules

[0140] (18-12)+(9-7)+(6-5)=6+2+1=9;

[0141] Calculate the denominator

[0142]

[0143] Calculate ΔI:

[0144]

[0145] The results show that the characteristic value of interest change is 3, reflecting that the total change in the frequency of members' participation in project types during the analysis period is large, and there are also certain changes in the frequency of regional switching. This value quantifies the intensity of members' interest changes in two consecutive time periods. This value is used to measure the amplitude and direction of interest changes. The larger the value, the more significant the changes in members' project selection and regional activities. The numerical results are used as quantitative indicators of interest changes and are further used to screen data segments associated with interest changes.

[0146] like Figure 2 and Figure 7 As shown, the service intelligent push module includes:

[0147] The service category screening submodule screens the park service categories associated with the interest change direction based on the interest direction transfer characteristics, counts the number of service categories involved, and obtains the distribution of service types;

[0148] The interest shift characteristics of each member are extracted. For example, if a member recently shifted from static viewing projects to dynamic experience projects, the interest change direction is read and all park service categories related to dynamic experience in the database are retrieved, such as self-controlled aircraft, track projects, indoor karting, indoor rock climbing, slides, backgammon, etc. All service categories that meet the interest change direction are extracted to form a list, and then the number of specific projects included in each service category is counted one by one. For example, the self-controlled aircraft category has two projects, "single-person self-controlled" and "double-person self-controlled". The track project has two projects, "small train track" and "through the tunnel". Indoor karting and indoor rock climbing each contain one project, and slides and backgammon each contain one project. The number of projects in the service categories is accumulated in sequence, and finally the service type distribution is summarized. For example, this screening involves 6 service categories, which contain a total of 7 specific projects. Based on this, the service type distribution is recorded as "dynamic experience: 7 projects".

[0149] The experience difference judgment submodule is based on the distribution of service types. It compares and filters the service items that have not been recorded as experiences according to the list of items experienced by each member and the service category. It also counts the total number of unexperienced items corresponding to each member to obtain the number of unexperienced items.

[0150] Extract the list of experience projects recorded by each member in the system. For example, the list of projects that a member has experienced includes "single-player self-driving", "small train track", and "indoor rock climbing". First, extract the list of all projects from the service type distribution, namely "single-player self-driving", "double-player self-driving", "small train track", "crossing the tunnel", "indoor go-kart", "slide", and "backgammon". Then compare the experienced projects with the above service projects in turn, confirm that "single-player self-driving", "small train track", and "indoor rock climbing" have been experienced, and mark these three items as experienced. Continue to compare "double-player self-driving", "crossing the tunnel", "indoor go-kart", "slide", and "backgammon". It is found that none of the projects are in the member's list of experienced projects, so they are marked as unexperienced projects. Then count the number of unexperienced projects. For example, if there are currently 5 unexperienced projects, the number of unexperienced projects for this member is recorded as 5. Perform the same comparison and statistical operations on all members in turn to generate a complete list of the number of unexperienced projects.

[0151] The recommendation order optimization submodule analyzes the correlation between unexperienced items and interest trends based on the number of unexperienced items, interest change trends, and item types. It then sorts the unexperienced items according to the direction of interest transfer and organizes the service content to obtain interest point push data.

[0152] The interest change trend of each member is extracted. For example, if a member's interest changes from static viewing to dynamic experience, all the items that the member has not experienced are read in sequence, such as "two-person automatic control", "crossing the tunnel", "indoor karting", "slide", and "backgammon". Then, the correlation between each unexperienced item and the member's interest change trend is analyzed one by one. By comparing whether the item category is consistent with the interest direction, for example, "indoor karting" and "two-person automatic control" belong to high-intensity dynamic experience and are judged to be strongly correlated. "Slide" and "backgammon" belong to medium-to-low-intensity dynamic experience and are also judged to be strongly correlated. "Crossing the tunnel" belongs to track items. The project also belongs to the category of dynamic experience. All unexperienced projects are sorted according to the degree of relevance. If the relevance is high, they are further sorted based on the participating groups and experience intensity that the projects are suitable for. For example, the "slide" suitable for families or beginners is ranked first, followed by "step by step", then "crossing the tunnel", then "two-person self-control", and finally "indoor karting". After the sorting is completed, the sorting results are integrated to generate service recommendation content for the member and form interest point push data, such as "first recommendation: slide, second recommendation: step by step, third recommendation: crossing the tunnel, fourth recommendation: two-person self-control, fifth recommendation: indoor karting".

[0153] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0154] In this disclosure, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0155] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0156] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0157] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0158] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of the device or unit, which can be electrical, mechanical or other forms.

[0159] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0160] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0161] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0162] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. Cloud-based membership management platform, characterized by: The platform includes: The information synchronization and integration module is based on the member information of multiple parks. It searches the database to match names and contact information, determines whether the unique identification data conflicts, completes the information merger, adjusts the file structure, removes duplicates, and obtains a unique member file set; The data conflict determination module compares the contact information, membership level and historical consumption amount based on the unique member profile set, determines the update time and the credibility of the data source, selects the priority parameter items, and obtains the conflict preferred parameter group; The trajectory behavior collection module analyzes the member area switching behavior based on the conflict optimization parameter group, collects the area entry and exit times through behavior tracking, calculates the area switching frequency, organizes the access time points of each project, optimizes the time sequence, and obtains the periodic service trajectory sequence; The interest trend identification module compares the item type switching interval and the area switching frequency based on the periodic service trajectory sequence, analyzes the parameter trend of the continuous time period, determines whether the interest shifts to a new type, filters the related data segments, and obtains the interest direction transfer characteristics.

2. The cloud-based membership management platform according to claim 1, characterized in that: The conflict optimization parameter group includes the latest contact information, the latest membership level, the latest consumption amount, and the highest trusted source. The periodic service trajectory sequence includes timed regional flow, project access sequence, and service usage distribution. The interest direction transfer characteristics include changes in preferred projects, service selection trends, and activity participation directions.

3. The cloud-based membership management platform according to claim 1, characterized in that: The information synchronization and integration module includes: The unique identification screening submodule is based on the member information of multiple parks. It matches the name and contact information fields with the unique identification of the unique number, ID number and mobile phone number in the database, determines the unique identification of each member data across multiple parks, and generates the number of unique identifications. The field conflict detection submodule compares the data contents under the same unique identifier based on the number of unique attribution identifiers, and detects whether there is a name or contact information conflict between different parks or databases by judging the difference in name content and contact information, and obtains the number of conflicting items; The information merging and adjusting submodule merges the conflicting member information according to the number of conflicting items, readjusts the data structure of the member files in the database, optimizes duplicate information, and obtains a unique member file set through field deduplication and structural adjustment.

4. The cloud-based membership management platform according to claim 1, characterized in that: The data conflict determination module includes: The information update judgment submodule, based on the unique member profile set, compares the update time of each contact information, member level, and historical consumption amount, determines the update order of the same member information in the difference records, and identifies the distribution of the update time of each parameter to obtain the update time distribution interval; The source trust screening submodule compares the trust levels of multiple data sources of the same member based on the update time distribution interval and the trust level of each information source, screens the data items with the best trust level, and obtains the number of trustworthy sources; The preferred parameter generation submodule selects parameter items with timeliness and priority according to the number of trusted sources, the update time of each data and the source trust level, and obtains a conflicting preferred parameter group.

5. The cloud-based membership management platform according to claim 1, characterized in that: The trajectory behavior collection module includes: The area handover collection submodule collects the area entry and exit time information of each member within the period based on the conflict optimization parameter group, records the movement process of members between areas, and establishes the cross-area handover information of each member; The switching frequency calculation submodule counts the number of times the same member switches between different regions within a period based on the cross-region switching information, and calculates the region switching frequency of each member within the period; The behavior sequence arrangement submodule calls the time point data of each park project visit according to the frequency of the area switching, arranges the behavior of each member in chronological order, optimizes the arrangement of the behavior data on the time axis, and obtains a periodic service trajectory sequence.

6. The cloud-based membership management platform according to claim 5, characterized in that: The frequency of region switching within each member's period is calculated using the formula: Get the periodic area switching frequency characteristic value of the i-th member Among them, T p Represents the total time of the statistical period, n i Represents the total number of cross-region switching records generated by the i-th member in this period, represents the actual time interval experienced by the i-th member during the j-th switch, represents the spatial weight difference between the regions crossed by the i-th member in the j-th switch, represents the congestion coefficient of the corresponding area of ​​the i-th member in the j-th handover, Represents the average of the congestion coefficients of all zone switching records of the i-th member in the period.

7. The cloud-based membership management platform according to claim 1, characterized in that: The interest trend identification module includes: The switching interval calculation submodule calculates the time interval between two consecutive project type changes based on the periodic service trajectory sequence and the project type switching time point of each member, and calculates the switching frequency changes between each area to obtain the type switching interval; The change trend determination submodule calls the project type switching interval and the area switching frequency based on the type switching interval, analyzes the parameter change trend in the continuous time period, and obtains the interest change trend by comparing the project type changes and area flow in the different time periods; The interest transfer feature screening submodule calls the project type and area switching distribution in each time period according to the interest change trend, compares the change direction of interest in consecutive time periods, screens the data segments associated with the interest change, and obtains the interest direction transfer feature.

8. The cloud-based membership management platform according to claim 7, characterized in that: The direction of change of the contrast interest in consecutive time periods is expressed using the formula: Calculate the interest change feature value ΔI, filter the data segments associated with interest changes, and obtain the interest direction transfer feature, where: represents the participation frequency of the kth project type in time period t, represents the participation frequency of the kth project type in time period t+1, represents the switching frequency of the kth region in time period t, represents the switching frequency of the kth region in time period t+1, n Q Represents the total number of project types or areas analyzed.

9. The cloud-based membership management platform according to claim 1, characterized in that: The platform also includes: The service intelligent push module screens the relevant service categories based on the interest direction transfer characteristics, determines the park service items that the member has not experienced, analyzes the correlation between the items and the current interest change trend, optimizes the recommendation order, and organizes the service content to obtain the point of interest push data; The POI push data includes recommendations for interesting services, a list of unexperienced items, and personalized push configurations.

10. The cloud-based membership management platform according to claim 9, characterized in that: The service intelligent push module includes: The service category screening submodule screens the park service categories associated with the interest change direction based on the interest direction transfer characteristics, counts the number of service categories involved, and obtains the service type distribution; The experience difference judgment submodule compares and filters service items that have not been recorded as experiences based on the distribution of service types and the list of items experienced by each member and the service category, and counts the total number of unexperienced items corresponding to each member to obtain the number of unexperienced items; The recommendation order optimization submodule analyzes the correlation between the unexperienced items and the interest trend based on the number of unexperienced items, the interest change trend and the item type, sorts the unexperienced items according to the interest transfer direction, and organizes the service content to obtain interest point push data.

Images