Business circle format optimization method and device, electronic equipment and computer readable storage medium

Abstract

The application provides a business format optimization method and device, electronic equipment and computer readable storage medium, and relates to the technical field of big data. On the one hand, the application identifies the resident users of the target business circle based on signaling data, unlike the existing passenger flow identification technology, which has a monitoring blind area and needs to be integrated with multi-source data. On the other hand, by identifying each group of people traveling together to the target business circle, the consumption capacity coefficient, access probability and resident time length evaluation model of each group are determined, and finally the target business circle is optimized by comprehensively considering the three factors. This not only guarantees the benefits brought by high-quality user groups, but also guarantees that the target business circle can retain users for a long time and improve the competitiveness of the target business circle in the competition among business circles, avoiding the problems of business format imbalance, brand tone dilution and the like caused by the optimization target being limited to short-term passenger flow or maximum rental income, and ensuring the long-term sustainable development of the business circle.

Description

Technical Field

[0001] This invention relates to the field of big data technology, and more specifically, to a method, apparatus, electronic device, and computer-readable storage medium for optimizing business district formats. Background Technology

[0002] With the acceleration of urbanization and the deepening of consumption upgrading trends, commercial complexes are playing an increasingly important role in modern urban life. Based on different consumer travel purposes, commercial districts can be broadly divided into two categories: destination-oriented commercial districts and leisure-oriented commercial districts. Destination-oriented commercial districts focus on meeting specific consumption goals, such as electronics stores or large supermarkets. Leisure-oriented commercial districts, on the other hand, offer consumers a comprehensive experience space integrating shopping, dining, leisure, entertainment, and social interaction.

[0003] In the operation and management of shopping districts, how to scientifically and rationally allocate the proportion of various business formats (such as retail, catering, and leisure and entertainment) has become a key challenge to improve customer traffic quality, extend dwell time, and enhance consumption conversion. An ideal business format distribution not only needs to match the preferences of the target customer group but also needs to consider the functional coordination of the overall space and the shaping of a long-term brand image. Therefore, dynamically optimizing the business format combination and achieving precise and intelligent resource allocation have become important research directions in current commercial real estate operations.

[0004] In existing technologies, business district planning mainly relies on dynamic business monitoring and data analysis methods, which include: collecting real-time customer flow data of each store through cameras, Wi-Fi probes or POS systems; evaluating business performance by combining tenant occupancy rates and sales per square meter (i.e., turnover / sales area); and using association rules to mine and analyze the customer flow linkage between different business formats (e.g., the driving effect of catering customer flow on cinema customer flow), thereby guiding the adjustment of leasing and spatial restructuring.

[0005] However, existing technologies are prone to monitoring blind spots, and the integration of multi-source data is difficult and costly. Furthermore, the optimization goals of existing technologies are limited to maximizing short-term customer flow or rental income, which can easily lead to problems such as imbalance in business structure and dilution of brand image, making it difficult to support the sustainable development of shopping districts. Summary of the Invention

[0006] The purpose of this invention is to provide a method, apparatus, electronic device, and computer-readable storage medium for optimizing business district formats, so as to improve the problems existing in the prior art.

[0007] The embodiments of the present invention can be implemented as follows: This invention provides a method for optimizing the business format of a commercial district, comprising: Based on the signaling dataset of the target business district within a specified time period, identify a number of resident users and identify at least one group of people traveling together from among these resident users. For each group, based on the travel trajectory information and personal information of each resident user in the group, the actual business structure of the target business district and at least one competing business district, the consumption capacity coefficient of the group and the evaluation model for the probability of visiting the target business district and the length of stay are determined. Based on the consumption capacity coefficient, visit probability, and dwell time assessment model of each group, the actual business structure of the target business district is optimized.

[0008] This invention also provides a business district format optimization device, comprising: The dwelling identification module is used to identify several dwelling users based on the signaling dataset of the target business district within a specified time period, and to identify at least one group of people traveling together from among the dwelling users. The calculation module is used to determine the consumption capacity coefficient of each group and the evaluation model of the probability of access to the target business district and the length of stay for each group, based on the travel trajectory information and personal information of each resident user in the group, the actual business structure of the target business district and at least one competing business district. The business format optimization module is used to optimize the actual business format structure of the target business district based on the consumption capacity coefficient, access probability, and dwell time evaluation model of each group.

[0009] This invention also provides an electronic device, including a memory and a processor. The memory stores a software program, and when the electronic device is running, the processor executes the software program to implement the above-mentioned business district optimization method.

[0010] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned business district optimization method.

[0011] Compared with existing technologies, embodiments of the present invention provide a method, apparatus, electronic device, and computer-readable storage medium for optimizing the business format of a commercial district. First, based on the signaling dataset of the target commercial district within a specified time period, a number of resident users are identified, and at least one group traveling together is identified from among these resident users. Then, for each group, based on the travel trajectory information and personal information of each resident user in the group, the actual business format structure of the target commercial district and at least one competing commercial district, the group's consumption capacity coefficient, as well as its probability of accessing the target commercial district and its dwell time evaluation model, are determined. Finally, based on the consumption capacity coefficient, access probability, and dwell time evaluation model of each group, the actual business format structure of the target commercial district is optimized.

[0012] On the one hand, this invention identifies resident users in a target business district based on signaling data, unlike existing customer flow identification technologies which have monitoring blind spots and require multi-source data integration. On the other hand, by identifying each group traveling together to the target business district, it determines the consumption capacity coefficient, access probability, and dwell time evaluation model for each group. Finally, it integrates these three factors to optimize the business format of the target business district. This ensures both the revenue from high-quality user groups and the ability of the target business district to retain users for a long time, thereby enhancing its competitiveness among competing business districts. It avoids problems such as unbalanced business format structure and dilution of brand image caused by limiting optimization targets to maximizing short-term customer flow or rental income, thus ensuring the long-term sustainable development of the business district. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a flowchart illustrating a business district optimization method provided in an embodiment of the present invention.

[0015] Figure 2 An undirected graph provided for an embodiment of the present invention.

[0016] Figure 3 This is a schematic diagram of a business district optimization device provided in an embodiment of the present invention.

[0017] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0020] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0021] It should be noted that, where there is no conflict, the features in the embodiments of the present invention can be combined with each other.

[0022] The business district optimization method provided in this invention can be applied to electronic devices, including but not limited to smartphones, personal laptops, personal computers, servers, and other computing devices.

[0023] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a business district optimization method provided in an embodiment of the present invention. The method includes the following steps S101 to S104.

[0024] S101. Based on the signaling dataset of the target business district within a specified time period, identify a number of resident users.

[0025] It is understandable that the signaling dataset can include all signaling data from all indoor and outdoor base stations within the target business area within a specified time period. The specified time period can be a specific day, week, month, or a certain period of time.

[0026] Signaling data is generated when a user terminal uses a mobile communication network and interacts with a base station (i.e., a public mobile communication base station, or simply a base station). A single signaling data item can carry a user code (representing the user terminal used by the user), a base station code (representing the base station that interacts with the user terminal), the user's location, and the interaction time (i.e., the timestamp of the signaling interaction).

[0027] In this embodiment, by identifying customer flow in the business district based on the signaling dataset, each resident user who has visited the target business district within a specified time period can be identified.

[0028] S102. Identify at least one group of people traveling together from among several resident users.

[0029] Optionally, for each user staying in each target business district, groups traveling together can be identified by combining the spatiotemporal overlap features in their travel trajectory information and communication behavior patterns.

[0030] S103. For each group, based on the travel trajectory information and personal information of each resident user in the group, the actual business structure of the target business district and at least one competing business district, determine the group's consumption capacity coefficient and the evaluation model for the probability of visiting the target business district and the length of stay.

[0031] Travel trajectory information can reflect the spatiotemporal trajectory of a resident user within a specified time period, or it can reflect the spatiotemporal trajectory of a resident user within a certain period before and after visiting a target business district (for example). The actual business structure of a business district reflects the current proportion of business types in that business district, such as the proportion of catering, retail, and leisure and entertainment.

[0032] S104. Based on the consumption capacity coefficient, access probability and dwell time evaluation model of each group, optimize the actual business structure of the target business district.

[0033] In this embodiment, the target business format structure of the target business district can be determined based on the consumption capacity coefficient, access probability, and dwell time evaluation model of each group. Then, the optimization strategy can be determined based on the target business format structure and the actual business format structure of the target business district.

[0034] The business district optimization method provided in this invention has two advantages. First, it identifies resident users in the target business district based on signaling data, enabling accurate and rapid identification of customer flow, unlike existing customer flow identification technologies which have monitoring blind spots and require multi-source data integration. Second, by identifying each group traveling together to the target business district, it determines the consumption capacity coefficient, access probability, and dwell time evaluation model for each group. Finally, it integrates these three factors to optimize the business district's business format. This approach ensures both the revenue generated by high-quality user groups and the long-term retention of users in the target business district, enhancing its competitiveness among competing business districts. It avoids problems such as business format imbalance and brand dilution caused by limiting optimization targets to maximizing short-term customer flow or rental income, thus ensuring the long-term sustainable development of the business district.

[0035] Traditional user association methods often rely on single-dimensional data, such as judging companions solely by location overlap. This can easily misclassify strangers encountered by chance or with similar commuting routes as real companions, leading to biased subsequent value assessments. This invention primarily combines spatiotemporal overlap features and communication behavior patterns from travel trajectory information to determine the probability of companionship between two resident users. Then, group identification is performed through mapping. That is, the implementation process of step S102 can include steps S1021-S1024.

[0036] S1021. Obtain the call list, travel trajectory information, and work and residence information for each resident user.

[0037] In this embodiment, the call list and travel trajectory information both belong to the same time period, while the work and residence information can include at least one workplace and at least one residence of the resident user.

[0038] S1022. For each pair of resident users, determine the probability of the two resident users traveling together based on their call lists, travel trajectory information, and work and residence information.

[0039] In this embodiment, for all users residing in the target business district, it is necessary to calculate the probability of pairing up between every pair of users. The process of calculating the probability of pairing up between two users may include steps a1 to a3.

[0040] Step a1: Based on the call lists, travel trajectory information, and work-residence information of the two resident users, determine the call frequency ratio, spatiotemporal trajectory overlap, and work-residence overlap.

[0041] For example, suppose two resident users are user A and user B. Based on their call lists, the total number of calls made by user A can be obtained. Total number of calls made by user B Number of calls between User A and User B The percentage of call frequency can be: .

[0042] In this embodiment, the spatiotemporal trajectory overlap degree can be obtained by performing spatial and temporal matching calculations on the travel trajectory information of two resident users, and is used to measure the spatiotemporal consistency of physical movement trajectories. The workplace-residence overlap degree is determined by comparing whether the residences and workplaces of two resident users are located in the same or adjacent geographical grids, representing the level of overlap in daily life circles, and reflecting the possibility of co-travel.

[0043] Step a2: Based on the travel trajectory information of the two resident users, extract the number of times they travel together on non-working days, and determine the degree of relationship based on the number of times they travel together.

[0044] In this embodiment, non-working days include statutory holidays and weekends. During these periods, users' travel purposes are more often related to social, leisure, and family activities, rather than rigid needs such as work commuting. When two users exhibit joint travel records with similar start and end points and highly overlapping trajectories on multiple non-working days, it indicates that they may have a close, intimate relationship, such as relatives or partners, and the likelihood of them traveling together is relatively high. The number of joint trips can be mapped to a relationship closeness of 0-1.

[0045] Step a3: Weight the call frequency ratio, spatiotemporal trajectory overlap, relationship closeness, and work-residence overlap to obtain the probability of two resident users forming a companionship.

[0046] In this embodiment, the first resident users With, the resident users Probability of pairing for:

[0047] In formula (1), Indicates the percentage of call frequency. Indicates the degree of overlap in spatiotemporal trajectories. Indicates the closeness of the relationship. Indicates the degree of overlap between work and residence. All are weighting coefficients.

[0048] S1023. Treat all resident users as user nodes. If the probability of two resident users forming a pair exceeds a set probability, then create an undirected edge between the two corresponding user nodes.

[0049] S1024. Based on all user nodes and all undirected edges, generate an undirected graph and identify at least one community from the undirected graph.

[0050] In this embodiment, two user nodes within a community are connected by at least one undirected edge, while the two communities are not connected in the undirected graph. Optionally, breadth-first search or a connected component algorithm can be used to search for each community in the undirected graph.

[0051] For example, please see Figure 2 The undirected graph shown can be used to identify four groups (group 1 to group 4). It should be noted that this example is for illustrative purposes only and is not intended to be limiting.

[0052] In an optional implementation, for each group, the process of determining the group's consumption capacity coefficient and the evaluation model for the probability of visiting the target business district and the length of stay in the target business district in step S103 above may include the following sub-steps S1031 to S1038.

[0053] S1031 obtains the geographical location and actual business structure of the target business district and at least one competing business district, as well as the terminal model, application usage data, travel trajectory information, age, and gender of each resident user in the group.

[0054] The aforementioned personal information includes the device model, application usage data, age, and gender.

[0055] S1032. Calculate the activity level of the corresponding resident user across multiple shopping apps from each application usage data.

[0056] S1033. Based on the usage data of each application, determine the preference distribution data of each resident user in the group.

[0057] Application usage data can include information such as the foreground runtime and launch frequency of multiple mobile applications. Therefore, based on the foreground runtime and launch frequency of each shopping app, the activity level of resident users on each shopping app can be determined. Furthermore, based on the application usage data page, the preference distribution data of resident users (e.g., liking food, liking reading, etc.) can be determined, or the preference distribution data page can be extracted from the resident user's personal profile.

[0058] S1034. Convert each preference distribution data into a preference vector, and convert each actual business structure into an actual business vector.

[0059] S1035. Obtain the starting point of the corresponding resident user's journey to the target business district from each travel trajectory information.

[0060] S1036. Determine the group's spending power coefficient based on the terminal model of each resident user within the group and their activity level on multiple shopping apps.

[0061] Optionally, the official selling price of each terminal model can be determined first, thus obtaining the terminal price for each user in the group. Then, based on the terminal prices of each user in the group and their activity levels on multiple shopping apps, the spending power score of each user in the group can be determined. Finally, the number of users in the group and the spending power score of each user are input into a preset ability assessment function to obtain the spending power coefficient of the group.

[0062] The expression for the ability assessment function is as follows:

[0063] In formula (2), group The consumption capacity coefficient group The number of users, group Chinese resident users Spending power score; This is a logarithmic gain operator based on group size, characterizing the enhancement effect of group consumption.

[0064] S1037. Based on the geographical location and actual business format vectors of the target business district and at least one competing business district, and the starting point and preference vector of each resident user in the group, determine the probability of the group visiting the target business district.

[0065] It should be understood that when assessing whether a group traveling together will choose to visit a specific business district, the judgment cannot be based solely on individual preferences or a single distance factor. Instead, both spatial accessibility and the attractiveness of the business district's offerings need to be considered. Therefore, the implementation process of step S1037 may include steps b1 to b6.

[0066] Step b1: For each business district in the target business district and all its competing business districts, determine the corresponding group collaboration center based on the geographical location and starting points of that business district.

[0067] The group collaboration center can be the weighted geometric center of each starting point. That is, the distance between each starting point and the business district can be calculated; then, based on each distance, the compromise weight of each resident user can be determined; and then, using the compromise weight of each resident user, the weighted average of each starting point can be obtained to get the group collaboration center corresponding to the business district.

[0068] Step b2: Based on the distance between the geographical location of the business district and the group's collaboration center, determine the probability of the group's compromise on the distance to the business district.

[0069] The distance compromise probability characterizes the degree of collaborative compromise among groups regarding the spatial distance of their trips. The formula for calculating the distance compromise probability can be:

[0070] In formula (3), group For the business district The probability of compromise at a distance; It is the attenuation factor; group For the business district The group collaboration center Represents the group collaboration center Distance to business district C.

[0071] Step b3: Sum all the preference vectors corresponding to the group and calculate the mean to obtain the group preference vector.

[0072] Step b4: Calculate the cosine similarity between the group preference vector and the actual business format vector of the business district to obtain the group's satisfaction with the business format of the business district.

[0073] In this embodiment, the cosine similarity between the group preference vector and the actual business format vector of the business district also reflects the attractiveness of the current actual business format structure of the business district to the group.

[0074] Step b5: Weight the probability of distance compromise and the business type satisfaction to obtain the group's overall selection probability for the business district.

[0075] The formula for calculating the overall selection probability is as follows:

[0076] In formula (4), group For the business district The overall selection probability, group For the business district Business satisfaction; These are all weighting coefficients, used to measure whether "proximity" or "access to good stores" is more important for a group.

[0077] It is understandable that each business district within a certain distance of the target business district (e.g., within 5km) can be considered a competing business district; or, other business districts located within the same administrative district as the target business district (e.g., Dongcheng District in Beijing) can be considered competing business districts; or, other business districts with a high degree of overlap in business structure with the target business district and a relatively close distance (e.g., less than 3km) can be considered competing business districts.

[0078] For both the target business district and each competing business district, steps b1 to b5 are performed to obtain the overall selection probability of the target business district and the overall selection probability of each competing business district.

[0079] Step b6: Based on the group's overall selection probability of the target business district and all its competing business districts, calculate the group's access probability of the target business district.

[0080] The formula for calculating the access probability is:

[0081] In formula (5), This refers to the target business district and the set of business districts comprised of its various competing business districts. For business districts Any business district in the middle, Indicates the target business district. group For the target business district The probability of access.

[0082] S1038. Based on the actual business format vector of the target business district and the preference vector, age, and gender of each resident user in the group, construct a stay duration assessment model for the group.

[0083] Optionally, the implementation process of step S1038 may include steps c1 to c3.

[0084] Step c1: Determine the group preference vector based on the preference vectors of each resident user in the group.

[0085] Step c2: Determine browsing preference factors based on the gender and age of each resident user in the group.

[0086] Here, the group preference vector is the average of the preference vectors of all resident users in the group. The browsing preference factor can be determined based on the age and gender of each resident user within a group using a specific mapping rule. For example: if the group... They were all young men. Too small (e.g., 0.4), if the group They are all young women, so Too large (e.g., 0.8), if the group They are all young women, so If the value is too large (e.g., 0.8), it should be noted that this example is for illustrative purposes only and is not intended to be limiting.

[0087] Step c3: Based on the preset basic dwell time, browsing preference factors, group preference vector and target business format vector corresponding to the business format information entropy, construct a dwell time evaluation model for the group.

[0088] Among them, the higher the business information entropy corresponding to the target business format vector, the richer the business formats are; the lower the business information entropy, the more singular the business formats are.

[0089] It can be understood that the dwell time assessment model is used to estimate the dwell time of groups based on the target business format vector within the target business district. In other words, in the dwell time assessment model, the target business format vector is a parameter to be solved. Therefore, the expression for the dwell time assessment model is:

[0090] In formula (6), Indicates the target business district Using target business format vectors At that time, the group The estimated duration of stay (i.e., the estimated duration of the visit). This represents the preset base stay duration, which is a statistical value. This represents the elasticity of wandering. Represents the target business format vector The corresponding business format information entropy.

[0091] group For the target business district Using target business format vectors Business satisfaction at that time, i.e. For the group The group preference vector and the target business format vector to be solved Cosine similarity between them: , For the group The group preference vector.

[0092] In an optional implementation, the consumption capacity coefficient, access probability, and dwell time assessment model of each group need to be combined to obtain the target business format vector. That is, the implementation process of step S104 above may include the following sub-steps S1041 to S1043.

[0093] S1041. Multiply and sum the consumption capacity coefficient, access probability and length of stay evaluation models of each group to construct the objective function.

[0094] In this embodiment, the objective function reflects the target business district's ability to retain high-quality users. The expression for the objective function is:

[0095] S1042. Maximize the objective function to obtain the target business vector.

[0096] As can be seen from the combined formulas (6) and (7), the objective function is a nonlinear function (because it includes Logit probability division, Entropy logarithm terms, and has strict linear constraints), so traditional differentiation methods (such as the Lagrange multiplier method) are not used.

[0097] In this embodiment, the objective function can be solved comprehensively using a genetic algorithm (suitable for iterative optimization) and / or sequential quadratic programming from heuristic algorithms to find a specific... , making The maximum. The process of solving the problem using genetic algorithms and sequential quadratic programming is existing technology and will not be elaborated upon here.

[0098] When genetic algorithms and sequential quadratic programming are used together, the objective function can be solved using genetic algorithms and sequential quadratic programming respectively, yielding two results. ,two The final target business vector is obtained by averaging.

[0099] This invention abandons the existing business format optimization that aims to maximize short-term customer flow or short-term revenue. Instead, it constructs an objective function based on the consumption capacity coefficient, access probability, and dwell time evaluation model of each group and solves for maximization. The resulting target business format vector maximizes the retention capacity of the target business district for high-quality user groups. After adjusting the business format according to the target business format vector, it helps the business district to move towards a balance and promote the sustainable development of the business district.

[0100] S1043. Determine the business format adjustment strategy based on the target business format vector and the actual business format vector corresponding to the actual business format structure of the target business district.

[0101] In this embodiment, the target business format vector and the actual business format vector respectively include the target proportion and actual proportion of various business format types. Therefore, the implementation process of step S1043 may include steps d1 to d5.

[0102] Step d1: Obtain the preset floating threshold corresponding to each business type.

[0103] Assume there is So, what are the different types of business formats? The K preset floating thresholds corresponding to each business type are: .

[0104] Step d2: Calculate the difference between the target percentage and the actual percentage for each business type to obtain the percentage gap for each business type.

[0105] In this embodiment, the first The gap in the proportion of various business types is: , For the first in the target business format vector One value, The first element in the actual business format vector of the target business district One value, It can be positive or negative. Among them: like This indicates that the first No adjustment is needed for this type of business; like This indicates that the first There are relatively few merchants of this type of business; we can consider increasing their numbers. like This indicates that the first There are too many merchants of this type; we should consider reducing their number.

[0106] Step d3: For the business type with the largest percentage gap, if the percentage gap of this business type is greater than the corresponding preset floating threshold, then at least one business type to be added is retrieved from the preset business type tag library.

[0107] Optionally, the business category tag library can be constructed using a two-level or three-level classification mechanism. For example, it can be divided into three levels: major category → intermediate category → minor category. The major and intermediate categories can be as shown in Table 1 below: Table 1

[0108] In Table 1, the subcategories under each category can be either functional tags or style-specific tags.

[0109] For example, satisfying and The largest business type is "Asian cuisine" in Table 1. Therefore, each subcategory under "Asian cuisine" can be retrieved as a potential business type category, such as Thai cuisine, Korean cuisine, Japanese cuisine, etc. It should be noted that the above are merely examples and are not intended to be limiting.

[0110] Step d4: For the business type with the smallest percentage gap, if the percentage gap of this business type is less than the opposite of the corresponding preset floating threshold, then obtain the customer traffic and turnover of each merchant belonging to this business type in the target business district.

[0111] Step d5: Normalize the customer traffic and turnover of each merchant and then perform a weighted average to obtain the comprehensive coefficient of each merchant. The merchant with the smallest comprehensive coefficient is selected as the target merchant to be withdrawn.

[0112] Alternatively, the three merchants with the lowest overall coefficient can be selected as potential merchants to be withdrawn.

[0113] It should be noted that the execution order of each step in the above method embodiments is not limited to that shown in the attached figures, and the execution order of each step shall be subject to the actual application situation.

[0114] In order to perform the corresponding steps in the above method embodiments and various possible implementations, an implementation method of a business district optimization device is given below.

[0115] Please see Figure 3 , Figure 3 A schematic diagram of the structure of a business district optimization device provided in an embodiment of the present invention is shown. The business district optimization device 200 includes: a dwell identification module 210, a calculation module 220, and a business district optimization module 230.

[0116] The dwelling identification module 210 is used to identify a number of dwelling users based on the signaling dataset of the target business district within a specified time period, and to identify at least one group of people traveling together from among the dwelling users. The calculation module 220 is used to determine the consumption capacity coefficient of each group, as well as the probability of access to the target business district and the dwell time evaluation model, based on the travel trajectory information and personal information of each resident user in the group, the actual business structure of the target business district and at least one competing business district. The business format optimization module 230 is used to optimize the actual business format structure of the target business district based on the consumption capacity coefficient, access probability and dwell time evaluation model of each group.

[0117] Optionally, the residency identification module 210 can be used to: obtain the call list, travel trajectory information, and work / residence information of each residency user; for each pair of residency users, determine the probability of them forming a group based on their call list, travel trajectory information, and work / residence information; treat all residency users as user nodes, and if the probability of two residency users forming a group exceeds a set probability, create an undirected edge between the corresponding two user nodes; generate an undirected graph based on all user nodes and all undirected edges, and identify at least one group from the undirected graph; and connect two user nodes in a group through at least one undirected edge.

[0118] Optionally, the residency identification module 210 can be used to: determine the call frequency ratio, spatiotemporal trajectory overlap, and work-residence overlap based on the call lists, travel trajectory information, and work-residence information of the two residency users respectively; extract the number of times they travel together on non-working days based on the travel trajectory information of the two residency users, and determine the relationship closeness based on the number of times they travel together; and perform a weighted summation of the call frequency ratio, spatiotemporal trajectory overlap, relationship closeness, and work-residence overlap to obtain the probability of the two residency users traveling together.

[0119] Optionally, the calculation module 220 can be specifically used to: obtain the geographical location and actual business structure of the target business district and at least one competing business district, as well as the terminal model, application usage data, travel trajectory information, age, and gender of each resident user in the group; calculate the activity level of the corresponding resident user on multiple shopping apps from each application usage data set; determine the preference distribution data of each resident user in the group based on each application usage data set; convert each preference distribution data set into a preference vector, and convert each actual business structure into an actual business vector; obtain the starting point of the corresponding resident user's journey to the target business district from each travel trajectory information set; determine the group's consumption capacity coefficient based on the terminal model of each resident user in the group and their activity level on multiple shopping apps; determine the group's access probability to the target business district based on the geographical location and actual business vector of the target business district and at least one competing business district, the starting point and preference vector of each resident user in the group; and construct a dwell time evaluation model for the group based on the actual business vector of the target business district and the preference vector, age, and gender of each resident user in the group.

[0120] Optionally, the calculation module 220 can be used to: determine the official selling price of each terminal model to obtain the terminal price of each resident user in the group; determine the consumption capacity score of each resident user in the group based on the terminal price of each resident user in the group and their activity on multiple shopping software; and input the number of users in the group and the consumption capacity score of each resident user into a preset capacity evaluation function to obtain the consumption capacity coefficient of the group.

[0121] Optionally, the calculation module 220 can be specifically used for: determining the group collaboration center corresponding to each business district in the target business district and all its competing business districts, based on the geographical location of the business district and each starting point; determining the group's distance compromise probability for the business district based on the distance between the geographical location of the business district and the group collaboration center; calculating the mean of all the group's corresponding preference vectors after summing them to obtain the group preference vector; calculating the cosine similarity between the group preference vector and the actual business format vector of the business district to obtain the group's business format satisfaction with the business district; weighted summing of the distance compromise probability and business format satisfaction to obtain the group's comprehensive selection probability for the business district; and calculating the group's access probability to the target business district based on the group's comprehensive selection probability for the target business district and all its competing business districts.

[0122] Optionally, the calculation module 220 can be used to: calculate the distance between each starting point and the business district; determine the compromise weight of each resident user based on each distance; and use the compromise weight of each resident user to perform a weighted average on each starting point to obtain the group collaboration center corresponding to the business district.

[0123] Optionally, the calculation module 220 can be used to: determine the group preference vector based on the preference vectors of each resident user in the group; determine the browsing preference factor based on the gender and age of each resident user in the group; and construct a group dwell time evaluation model based on the business information entropy corresponding to the preset basic dwell time, browsing preference factor, group preference vector and target business vector.

[0124] Optionally, the business format optimization module 230 can be used to: multiply and sum the consumption capacity coefficient, access probability and dwell time evaluation models of each group to construct an objective function; the objective function reflects the retention capacity of the target business district for high-quality user groups; maximize the objective function to obtain the target business format vector; and determine the business format adjustment strategy based on the target business format vector and the actual business format vector corresponding to the actual business format structure of the target business district.

[0125] Optionally, the target business format vector and the actual business format vector each include the target proportion and actual proportion of various business format types. The business format optimization module 230 can specifically be used to: obtain the preset floating threshold corresponding to each business format type; calculate the difference between the target proportion and the actual proportion for each business format type to obtain the proportion gap for each business format type; for the business format type with the largest proportion gap, if the proportion gap of this business format type is greater than the corresponding preset floating threshold, then retrieve at least one business format category to be added under this business format type from the preset business format tag library; for the business format type with the smallest proportion gap, if the proportion gap of this business format type is less than the negative of the corresponding preset floating threshold, then obtain the customer traffic and turnover of each merchant belonging to this business format type within the target business district; normalize the customer traffic and turnover of each merchant and then perform a weighted average to obtain the comprehensive coefficient of each merchant, and select the merchant with the smallest comprehensive coefficient as the target merchant to be withdrawn.

[0126] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the business district optimization device 200 described above can be referred to the corresponding process in the aforementioned method embodiments, and will not be repeated here.

[0127] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. The electronic device 300 includes a processor 310, a memory 320, and a bus 330, with the processor 310 connected to the memory 320 via the bus 330.

[0128] The memory 320 can be used to store software programs or firmware, for example, the software program or firmware corresponding to the aforementioned business district optimization device 200. The processor 310 executes various functional applications and data processing by running the software program stored in the memory 320 to realize the business district optimization method provided in the embodiments of the present invention.

[0129] The memory 320 may be, but is not limited to, RAM (Random Access Memory), ROM (Read Only Memory), FLASH (Flash Memory), PROM (Programmable Read-Only Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electric Erasable Programmable Read-Only Memory), etc.

[0130] The processor 310 can be an integrated circuit chip with signal processing capabilities, capable of executing software programs, such as the software program corresponding to the aforementioned business district optimization device 200. The processor 310 can be a general-purpose processor, including: CPU (Central Processing Unit), NP (Network Processor), SoC (System on Chip), etc.; it can also be: DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0131] Understandable. Figure 4 The structure shown is for illustrative purposes only; the electronic device 300 may also include components that are more advanced than those shown. Figure 4 The more or fewer components shown, or having the same Figure 4 The different configurations shown. Figure 4 The components shown can be implemented using hardware, software, or a combination thereof.

[0132] This invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the business district optimization method disclosed in the above embodiments. The computer-readable storage medium can be, but is not limited to, various media capable of storing program code, such as a USB flash drive, external hard drive, ROM, RAM, PROM, EPROM, EEPROM, FLASH disk, or optical disk.

[0133] In summary, this invention provides a method, apparatus, electronic device, and computer-readable storage medium for optimizing the business district. First, based on the signaling dataset of the target business district within a specified time period, several resident users are identified, and at least one group traveling together is identified from among these resident users. Then, for each group, based on the travel trajectory and personal information of each resident user within the group, and the actual business structure of the target business district and at least one competing business district, the group's spending power coefficient, as well as its probability of accessing the target business district and its dwell time evaluation model, are determined. Finally, based on the spending power coefficient, access probability, and dwell time evaluation model of each group, the actual business structure of the target business district is optimized. On the one hand, this invention identifies resident users in a target business district based on signaling data, unlike existing customer flow identification technologies which have monitoring blind spots and require multi-source data integration. On the other hand, by identifying each group traveling together to the target business district, it determines the consumption capacity coefficient, access probability, and dwell time evaluation model for each group. Finally, it integrates these three factors to optimize the business format of the target business district. This ensures both the revenue from high-quality user groups and the ability of the target business district to retain users for a long time, thereby enhancing its competitiveness among competing business districts. It avoids problems such as unbalanced business format structure and dilution of brand image caused by limiting optimization targets to maximizing short-term customer flow or rental income, thus ensuring the long-term sustainable development of the business district.

[0134] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for optimizing business district formats, characterized in that, include: Based on the signaling dataset of the target business district within a specified time period, identify a number of resident users and identify at least one group of people traveling together from among these resident users. For each group, based on the travel trajectory information and personal information of each resident user in the group, the actual business structure of the target business district and at least one competing business district, the consumption capacity coefficient of the group and the evaluation model for the probability of visiting the target business district and the length of stay are determined. Based on the consumption capacity coefficient, visit probability, and dwell time assessment model of each group, the actual business structure of the target business district is optimized.

2. The business district format optimization method according to claim 1, characterized in that, The step of identifying at least one group of travelers from a plurality of resident users includes: Obtain each resident user's call list, travel history information, and work / residence information; For each pair of resident users, the probability of the two resident users traveling together is determined based on their call lists, travel trajectory information, and work and residence information. All resident users are treated as user nodes. If the probability of two resident users forming a pair exceeds a set probability, an undirected edge is created between the two corresponding user nodes. An undirected graph is generated based on all user nodes and all undirected edges, and at least one community is identified from the undirected graph; two user nodes in the community are connected by at least one undirected edge.

3. The business district format optimization method according to claim 2, characterized in that, The step of determining the probability of two resident users traveling together based on their call lists, travel trajectory information, and work / residence information includes: Based on the call lists, travel trajectory information, and work-residence information of the two resident users, the call frequency ratio, spatiotemporal trajectory overlap, and work-residence overlap were determined respectively. Based on the travel trajectory information of two resident users, the number of times they traveled together on non-working days is extracted, and the degree of relationship is determined based on the number of times they traveled together. The probability of two resident users forming a companionship is obtained by weighting and summing the call frequency ratio, the spatiotemporal trajectory overlap, the relationship closeness, and the work-residence overlap.

4. The business district format optimization method according to claim 1, characterized in that, The steps of determining the group's spending power coefficient and the evaluation model for the probability of visiting and the duration of stay in the target business district, based on the travel trajectory information and personal information of each resident user in the group, the actual business structure of the target business district and at least one competing business district, include: Obtain the geographical location and actual business structure of the target business district and at least one competing business district, as well as the terminal model, application usage data, travel trajectory information, age, and gender of each resident user in the group; The activity level of the corresponding resident user across multiple shopping apps is statistically analyzed from each set of application usage data. Based on the usage data of each of the aforementioned applications, determine the preference distribution data of each resident user in the group; Each of the aforementioned preference distribution data is converted into a preference vector, and each actual business structure is converted into an actual business vector. Obtain the starting point of the corresponding resident user's journey to the target business district from each of the aforementioned travel trajectory information; The spending power coefficient of the group is determined based on the terminal model of each resident user in the group and their activity level on multiple shopping apps. Based on the geographical location and actual business format vector of the target business district and at least one competing business district, and the starting point and preference vector of each resident user in the group, the probability of the group visiting the target business district is determined. Based on the actual business format vector of the target business district and the preference vector, age, and gender of each resident user in the group, a stay duration evaluation model for the group is constructed.

5. The business district format optimization method according to claim 4, characterized in that, The step of determining the group's spending power coefficient based on the terminal model of each resident user within the group and their activity level on multiple shopping apps includes: Determine the official selling price for each terminal model to obtain the terminal price for each resident user in the group; Based on the terminal prices of each resident user in the group and their activity levels on multiple shopping apps, a spending power score is determined for each resident user in the group. The number of users in the group and the spending power score of each resident user are input into a preset ability assessment function to obtain the spending power coefficient of the group.

6. The business district format optimization method according to claim 4, characterized in that, The step of determining the probability of the group's access to the target business district based on the geographical location and actual business format vector of the target business district and at least one competing business district, and the starting point and preference vector of each resident user in the group, includes: For each business district in the target business district and all its competing business districts, a group collaboration center corresponding to the business district is determined based on the geographical location of the business district and each of the starting points; Based on the distance between the geographical location of the business district and the group's collaboration center, determine the probability of the group's compromise regarding the distance to the business district; The group preference vector is obtained by summing all the preference vectors corresponding to the group and calculating the mean. Calculate the cosine similarity between the group's preference vector and the actual business format vector of the business district to obtain the group's satisfaction with the business format of the business district; The group's overall selection probability for the business district is obtained by weighted summation of the distance compromise probability and the business type satisfaction. Based on the group's overall selection probability for the target business district and all its competing business districts, the group's access probability for the target business district is calculated.

7. The business district format optimization method according to claim 6, characterized in that, The step of determining the group collaboration center corresponding to the business district based on the geographical location of the business district and each of the starting points includes: Calculate the distance between each starting point and the business district; Based on each of the aforementioned interval distances, the compromise weight for each resident user is determined; By utilizing the compromise weights of each resident user, a weighted average is calculated for each starting point to obtain the group collaboration center corresponding to the business district.

8. The business district format optimization method according to claim 4, characterized in that, The step of constructing a dwell time assessment model for the group based on the actual business format vector of the target business district and the preference vectors, age, and gender of each resident user in the group includes: Based on the preference vectors of each resident user in the group, the group preference vector is determined. Based on the gender and age of each resident user in the group, a browsing preference factor is determined; Based on the preset basic dwell time, the browsing preference factor, the group preference vector, and the business information entropy corresponding to the target business vector, a dwell time evaluation model for the group is constructed.

9. The business district format optimization method according to claim 1, characterized in that, The step of optimizing the actual business structure of the target business district based on the consumption capacity coefficient, access probability, and dwell time evaluation model of each group includes: The consumption capacity coefficient, access probability, and dwell time evaluation models of each group are multiplied and summed to construct an objective function; the objective function reflects the retention ability of the target business district for high-quality user groups. The objective function is maximized to obtain the target business vector; The business format adjustment strategy is determined based on the target business format vector and the actual business format vector corresponding to the actual business format structure of the target business district.

10. The business district format optimization method according to claim 9, characterized in that, The target business format vector and the actual business format vector each include the target proportion and actual proportion of various business format types; The step of determining the business format adjustment strategy based on the target business format vector and the actual business format vector corresponding to the actual business format structure of the target business district includes: Obtain the preset floating threshold corresponding to each business type; Calculate the difference between the target percentage and the actual percentage for each business type to obtain the percentage gap for each business type; For the business type with the largest percentage gap, if the percentage gap of the business type is greater than the corresponding preset floating threshold, at least one business type to be added will be retrieved from the preset business type tag library. For the business type with the smallest percentage gap, if the percentage gap of this business type is less than the negative of the corresponding preset floating threshold, then obtain the customer traffic and turnover of each merchant belonging to this business type in the target business district. After normalizing the customer traffic and turnover of each merchant, a weighted average is calculated to obtain the comprehensive coefficient of each merchant, and the merchant with the smallest comprehensive coefficient is selected as the target merchant to be withdrawn.

11. A business district format optimization device, characterized in that, include: The dwelling identification module is used to identify several dwelling users based on the signaling dataset of the target business district within a specified time period, and to identify at least one group of people traveling together from among the dwelling users. The calculation module is used to determine the consumption capacity coefficient of each group and the evaluation model of the probability of access to the target business district and the length of stay for each group, based on the travel trajectory information and personal information of each resident user in the group, the actual business structure of the target business district and at least one competing business district. The business format optimization module is used to optimize the actual business format structure of the target business district based on the consumption capacity coefficient, access probability, and dwell time evaluation model of each group.

12. An electronic device, characterized in that, include: The electronic device includes a memory and a processor, wherein the memory stores a software program, and the processor executes the software program to implement the business district optimization method as described in any one of claims 1-10 when the electronic device is running.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the business district optimization method according to any one of claims 1-10.

Images