Client data-based overseas market segmentation method and system

Abstract

The invention is suitable for overseas market analysis, and provides an overseas market segmentation method and system based on customer data, and the method comprises the steps: collecting data through an API interface, and constructing a structured data set; calculating a commodity transaction total amount proportion and a service resource density, screening a strong time zone, and analyzing a demand inhibition index of a neighborhood time zone; identifying a consumption shadow area based on the suppression index, screening a golden overlapping zone and constructing a time zone relation graph; analyzing behaviors and emotional tendencies of the user in the overlapping time period, and outputting a potential demand elasticity coefficient and a recommended commodity category; and dynamically allocating resources and adjusting a flow strategy in the golden overlapping zone according to the elastic coefficient. Through multi-dimensional data integration and dynamic analysis, the cross-time-zone consumption potential and the high-frequency interaction time period are accurately identified, fine scheduling of resources is realized, the overseas market operation efficiency is improved, and the method is suitable for global business scenes such as cross-border e-commerce and the like.

Description

Technical Field

[0001] The present invention belongs to the field of overseas market analysis, and in particular relates to a method and system for overseas market segmentation based on customer data. Background Art

[0002] With the deepening development of global trade, overseas market segmentation has become a key step for companies to accurately target users and optimize resource allocation. Traditional market segmentation methods are mostly based on static data such as geographic regions and demographics, which makes it difficult to cope with the dynamic changes in user behavior in cross-time zone scenarios. In recent years, with the application of big data technology, some solutions have begun to integrate user consumption behavior data for market analysis, but they have not fully explored the supply and demand mismatch caused by time zone differences and the correlation between cross-time zone social behavior and consumer demand. In cross-time zone operation scenarios, users' active social hours and consumption habits are significantly affected by time zone differences. How to accurately identify consumption potential and demand suppression areas in different time zones through multi-dimensional data integration has become a research hotspot in the field of overseas market segmentation.

[0003] Some existing solutions analyze consumption data by time zone to segment markets and combine it with simple time overlap analysis for resource allocation. For example, they calculate the total merchandise transaction volume for each time zone to identify key markets and then target marketing during user active periods based on historical consumption data. These solutions focus solely on surface-level correlations in consumption data, failing to integrate cross-time zone social behavior data or build dynamic demand assessment models. Identifying time zones with potential demand but ineffective activation relies on empirical judgment and lacks in-depth analysis of user behavior during overlapping periods and the sentiment surrounding social topics, making it difficult to achieve refined resource allocation. Summary of the Invention

[0004] The purpose of the present invention is to provide an overseas market segmentation method based on customer data, aiming to solve the technical problems existing in the prior art identified in the background technology.

[0005] The present invention is implemented as follows: a method for overseas market segmentation based on customer data, the method comprising:

[0006] Obtain user time zone data, cross-time zone social behavior data, and consumption behavior data through the API interface, and build a standardized time series matrix to generate a structured data set;

[0007] Based on a structured data set, we calculated the proportion of total merchandise transaction volume and service resource density in each time zone, screened out the dominant time zones, and calculated the demand suppression index for each dominant time zone within ±2 hours of the neighboring time zones.

[0008] Based on the suppression index, we screen out consumption shadow zones, pair them with strong time zones with historical interaction records, and calculate the overlapping activity to identify golden overlapping zones. Based on the average daily active overlap duration and cross-zone interactive user coverage of the golden overlapping zones, we rank the active zones and construct a time zone relationship map.

[0009] Mapping the consumption shadow zone to a time zone relationship map, locating the overlapping periods between the consumption shadow zone and the strong time zone, and constructing a dual-feature matrix to analyze the behavior and social topic sentiment of users in the consumption shadow zone during the overlapping periods, outputting the potential demand elasticity coefficient and recommended product categories;

[0010] Based on the potential demand elasticity coefficient, resources are dynamically allocated and traffic scheduling strategies are adjusted during the golden overlap period.

[0011] As a further solution of the present invention, the user time zone data, cross-time zone social behavior data and consumption behavior data are obtained through the API interface, and a standardized time series matrix is ​​constructed to generate a structured data set, specifically:

[0012] Obtain user time zone data, cross-time zone social behavior data, and consumption behavior data through the API interface, including:

[0013] The user time zone data includes: time zone information of the user's geographical location and time zone change records;

[0014] The cross-time zone social behavior data includes: the user's social login time, social interaction frequency, and social content publishing time in different time zones;

[0015] The consumption behavior data includes: the user's consumption time, consumption amount, and consumption products;

[0016] Clean the acquired user time zone data, cross-time zone social behavior data, and consumption behavior data to remove duplicate, erroneous, and missing data;

[0017] Arrange the cleaned data in chronological order and construct a standardized time series matrix. The rows of the matrix represent different time points and the columns represent different data indicators.

[0018] The standardized time series matrix is ​​subjected to feature extraction and transformation to generate a structured data set, which contains the time zone information, social behavior characteristics, and consumption behavior characteristics of each user at different time points.

[0019] As a further solution of the present invention, the proportion of total merchandise transaction volume and service resource density of each time zone are calculated, and the time zones with the top 20% of total merchandise transaction volume and service resource density greater than the average are defined as strong time zones. For each strong time zone and the neighboring time zones within ±2 hours, the demand suppression index is calculated, specifically:

[0020] Extract commodity transaction data for each time zone from the structured data set and calculate the total commodity transaction amount for each time zone;

[0021] Calculate the sum of the GMT of all time zones, and then calculate the GMT share of each time zone based on the GMT of each time zone.

[0022] Extract the service resource data of each time zone from the structured data set and calculate the service resource density S of each time zone k ;

[0023] Calculate the average service resource density of all time zones and select the time zones with the top 20% of total merchandise transaction volume and a service resource density greater than the average, defining them as strong time zones.

[0024] For each strong time zone, determine the ±2-hour neighboring time zone, extract the consumption demand data and service supply data of the neighboring time zone from the structured data set, and calculate the demand suppression index I j .

[0025] As a further solution of the present invention, the service resource density S of each time zone is calculated. k :

[0026]

[0027] Among them, S p k is the number of service providers in time zone k, S c k is the number of service coverage contacts in time zone k, A k is the geographical area of ​​time zone k;

[0028] Calculate the demand suppression index I j :

[0029]

[0030] Among them, D th,j is the theoretical consumption demand in domain time zone j, D ac,j is the actual consumption demand in domain time zone j.

[0031] As a further solution of the present invention, the calculation of overlapping activity, and the classification of active zones according to the average daily active overlapping duration of the golden overlapping zone and the cross-zone interactive user coverage rate, and the construction of a time zone relationship map are specifically as follows:

[0032] From the calculated demand suppression index results, we screen out time zones with a demand suppression index greater than 0.8 and define them as consumption shadow zones;

[0033] Search the structured dataset for historical interaction records between consumption shadow zones and other time zones, and pair these zones with strong time zones where historical interaction records exist.

[0034] For each pair of consumption shadow zone and strong time zone, their social activity data at different times are extracted from the structured dataset, and the overlapping activity at each time point is calculated;

[0035] Screen out the time periods with overlapping activity > 0.6 and define them as the golden overlapping zone;

[0036] Calculate the average daily active overlap duration within each golden overlap zone, which is defined as the average daily active overlap duration;

[0037] Calculate the ratio of users participating in cross-region interactions in each golden overlap zone to the total number of users, which is defined as the user coverage rate of cross-region interactions.

[0038] The golden overlap zone is divided into different levels based on the average daily active overlap time and cross-region interactive user coverage rate;

[0039] Using time zones as nodes and golden overlapping belts as edges, a time zone relationship map is constructed.

[0040] As a further solution of the present invention, the overlapping activity at each time point is calculated as follows:

[0041]

[0042] Among them, O m,n,t A represents the overlapping activity between time zones m and n at time t, m,n,t S represents the number of users who are active simultaneously in time zones m and n at time t. m,t and S n,t Respectively represent the number of independent active users in time zone m and time zone n at time t;

[0043] Average daily active overlap time:

[0044]

[0045] Among them, L m,n is the average daily active overlap duration between time zones m and n, D is the number of days in the statistical period, and Δt is the duration of each golden overlap zone in minutes;

[0046] User coverage of cross-region interactions:

[0047]

[0048] Among them, C x,y Indicates the user coverage rate of cross-region interaction between the strong time zone x and the consumption shadow zone y, U x,yis the number of users who participated in cross-zone interaction between strong time zone x and consumption shadow zone y during the statistical period, U x 、U y They are the total number of users in the strong time zone x and the consumption shadow zone y respectively.

[0049] As a further solution of the present invention, the consumption shadow zone is mapped to the time zone relationship map, the overlapping period of the consumption shadow zone and the strong time zone is located, and a dual feature matrix is ​​constructed to analyze the behavior and social topic emotional tendencies of users in the consumption shadow zone during the overlapping period, and output the potential demand elasticity coefficient and recommended product categories, specifically:

[0050] Match the time zone information of the consumption shadow area with the nodes in the time zone relationship map, and map the consumption shadow area to the time zone relationship map;

[0051] Find the strong time zone that pairs with the consumption shadow zone in the time zone relationship map, and locate the golden overlap zone between the consumption shadow zone and the strong time zone;

[0052] Extract the behavioral data of users in the consumption shadow zone during the golden overlap period from the structured data set;

[0053] Construct a dual-feature matrix, where the rows represent users in the consumption shadow zone, and the columns represent the user's behavioral characteristics and social topic characteristics;

[0054] Conduct sentiment analysis on social topic data to determine the sentiment tendency of each social topic;

[0055] Analyze the dual-feature matrix and calculate the potential demand elasticity coefficient of users in the consumption shadow zone during the overlapping period;

[0056] According to the behavioral characteristics of users in the consumption shadow zone during the overlapping period, the emotional tendency of social topics and the potential demand elasticity coefficient, E i Categories with a value greater than 1 are recommended as demand product categories.

[0057] As a further solution of the present invention, the potential demand elasticity coefficient of users in the consumption shadow zone during the overlapping period is calculated as follows:

[0058]

[0059] Among them, E i represents the potential demand elasticity coefficient of user i in the consumption shadow zone during the overlapping period, Q i represents the benchmark consumption of the commodity category during the overlapping time period, ΔQ i represents the consumption change of commodity categories during the overlapping period, P i Indicates the amount of marketing resources invested per unit time in the consumption shadow area, ΔP iIt represents the change in marketing resources invested per unit time in the consumption shadow area. is the emotion correction factor.

[0060] As a further solution of the present invention, the dynamic allocation of resources and adjustment of traffic scheduling strategy during the golden overlap period according to the potential demand elasticity coefficient are specifically as follows:

[0061] Based on the potential demand elasticity coefficient of users in the consumption shadow zone, different product categories are classified into: high elasticity demand category, medium elasticity demand category, and low elasticity demand category;

[0062] During the golden overlap period, increase resource input for high-elasticity demand categories, maintain current resource input for medium-elasticity demand categories, and reduce resource input for low-elasticity demand categories.

[0063] Adjust traffic scheduling strategies based on the activity and demand elasticity coefficient of users in the consumption shadow zone during the golden overlap period;

[0064] Monitor the resource allocation and traffic scheduling effects during the golden overlap period in real time, and dynamically adjust the resource allocation and traffic scheduling strategies based on the monitoring results.

[0065] Another object of the present invention is to provide an overseas market segmentation system based on customer data, the system comprising:

[0066] The user data collection module is used to obtain user time zone data, cross-time zone social behavior data, and consumption behavior data through the API interface, and construct a standardized time series matrix to generate a structured data set;

[0067] The demand suppression index calculation module is used to calculate the proportion of total commodity transactions and service resource density in each time zone based on a structured data set, and to screen out strong time zones. The demand suppression index is calculated for each strong time zone within ±2 hours of the neighboring time zone.

[0068] The Golden Overlap Zone Screening Module is used to screen out consumption shadow zones based on the suppression index. It then pairs these zones with strong time zones with historical interaction records, calculates the overlap activity, and identifies the Golden Overlap Zones. Based on the average daily active overlap duration and cross-zone interactive user coverage of the Golden Overlap Zones, the active zones are graded and a time zone relationship map is constructed.

[0069] The consumer behavior analysis module is used to map the consumption shadow zone to the time zone relationship map, locate the overlapping periods of the consumption shadow zone and the strong time zone, and construct a dual feature matrix to analyze the behavior and social topic sentiment of users in the consumption shadow zone during the overlapping periods, outputting the potential demand elasticity coefficient and recommended product categories;

[0070] The dynamic resource allocation module is used to dynamically allocate resources and adjust traffic scheduling strategies during the golden overlap period based on the potential demand elasticity coefficient.

[0071] The beneficial effects of the present invention are:

[0072] Through multi-dimensional data integration and dynamic analysis, the present invention constructs a complete closed loop covering "data collection - potential market identification - precise demand mining - intelligent resource scheduling".

[0073] By integrating user time zone data, cross-time zone social behavior data and consumption behavior data, a structured data set is generated, which breaks through the limitation of traditional methods relying on single consumption data and can comprehensively capture the dynamic behavior characteristics of users in different time zones; with the help of demand suppression index and golden overlap zone analysis, the consumption shadow zone and its high-frequency interaction period with the strong time zone are accurately located, solving the problems of demand mismatch and ambiguous identification of potential markets in cross-time zone scenarios; by constructing a dual feature matrix and introducing emotional correction factors, the potential demand elasticity of users in overlapping periods is deeply explored, making product recommendations more in line with real consumption intentions; finally, resources are dynamically allocated and traffic strategies are adjusted based on the elasticity coefficient, achieving efficient resource utilization and real-time response to demand.

[0074] This solution significantly improves the level of refinement in overseas market segmentation, helping companies accurately identify high-value scenarios in cross-time zone operations, reduce ineffective investment, and enhance market responsiveness. It is particularly suitable for cross-border e-commerce, multinational services and other fields, providing companies with data-driven core competitiveness in global competition. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 A flowchart of a method for overseas market segmentation based on customer data provided by an embodiment of the present invention;

[0076] Figure 2 A flowchart for constructing a standardized time series matrix and generating a structured data set provided by an embodiment of the present invention;

[0077] Figure 3 A flowchart of calculating a demand suppression index according to an embodiment of the present invention;

[0078] Figure 4 A flowchart of the average daily active overlap duration and cross-region interactive user coverage of the golden overlap zone according to an embodiment of the present invention;

[0079] Figure 5 A flowchart of dynamically allocating resources and adjusting traffic scheduling strategies during the golden overlap period provided by an embodiment of the present invention;

[0080] Figure 6A flowchart for outputting potential demand elasticity coefficients and recommending product categories provided by an embodiment of the present invention;

[0081] Figure 7 This is a structural block diagram of the overseas market segmentation system based on customer data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0082] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0083] Figure 1 A flow chart of the overseas market segmentation method based on customer data provided by an embodiment of the present invention, such as Figure 1 As shown, the method includes:

[0084] S100, obtains user time zone data, cross-time zone social behavior data, and consumption behavior data through the API interface, and constructs a standardized time series matrix to generate a structured data set;

[0085] This step uses APIs to enable real-time collection and integration of multi-dimensional data. Its core function is to build a dynamic data network covering users' spatiotemporal trajectories, social activity, and consumption characteristics. User time zone data not only records static geographic locations but also captures the dynamic behavior of users moving across time zones through time zone change records, providing continuous temporal support for analyzing changes in consumption habits across regions. Cross-time zone social behavior data further explores users' social activity periods, interaction frequency, and content publishing patterns in different time zones, accurately locating users' social peaks in specific time zones. Combined with correlation analysis of consumption time, amount, and product category in consumption behavior data, this data can reveal the potential coupling between social activity and consumption decisions, such as whether high-frequency social interactions are accompanied by consumption booms for specific products.

[0086] The data cleaning process eliminates noise data to ensure that subsequent analysis is based on high-quality data, while the construction of a standardized time series matrix unifies the scattered time series data into a standardized time framework, making user behavior data of different time zones and types comparable across dimensions, and providing a structured data foundation for subsequent exploration of consumption linkage effects between time zones (such as the consumption radiation effect of a strong time zone on a neighboring time zone).

[0087] The combination of spatiotemporal data and consumption data can accurately locate users' active consumption periods in different time zones and identify consumption period mismatches caused by time zone differences (e.g., a user's social activity period in a certain time zone does not match the local service availability period). The introduction of social behavior data provides reference dimensions for consumer demand analysis, including emotional tendencies and social influence. For example, content posted by high-frequency social users during specific time periods may indicate potential consumption trends. Furthermore, the generation of structured data sets enables the organic integration of multi-source data, enabling subsequent steps to perform complex calculations (such as the demand suppression index and overlapping activity) based on a unified data model. This avoids analytical biases caused by inconsistent data formats and fundamentally improves the refinement of overseas market segmentation.

[0088] This data preprocessing method not only provides a key basis for identifying "consumption shadow zones", but also lays a solid data foundation for the subsequent implementation of precise resource scheduling and traffic allocation strategies during golden overlapping periods through dynamic analysis of time series, ensuring that companies can optimize cross-time zone operation strategies based on real-time user behavior patterns and maximize resource utilization efficiency.

[0089] like Figure 2 As shown, the user time zone data, cross-time zone social behavior data and consumption behavior data are obtained through the API interface, and a standardized time series matrix is ​​constructed to generate a structured data set, specifically:

[0090] S110, obtaining user time zone data, cross-time zone social behavior data, and consumption behavior data through an API interface, wherein:

[0091] The user time zone data includes: time zone information of the user's geographical location and time zone change records;

[0092] The cross-time zone social behavior data includes: the user's social login time, social interaction frequency, and social content publishing time in different time zones;

[0093] The consumption behavior data includes: the user's consumption time, consumption amount, and consumption products;

[0094] S120, cleaning the acquired user time zone data, cross-time zone social behavior data, and consumption behavior data to remove duplicate data, erroneous data, and missing data;

[0095] S130, arranging the cleaned data in chronological order to construct a standardized time series matrix, where the rows of the matrix represent different time points and the columns represent different data indicators;

[0096] S140, feature extraction and conversion are performed on the standardized time series matrix to generate a structured data set, which includes time zone information, social behavior features and consumption behavior features of each user at different time points.

[0097] S200, based on the structured data set, the total transaction amount proportion of commodities in each time zone and the service resource density are calculated, and the strong time zones are screened out. For each strong time zone ± 2 hour neighborhood time zone, the demand suppression index is calculated.

[0098] This step extracts the commodity transaction data of each time zone from the structured data set. By calculating the total transaction amount proportion, the consumption contribution of different time zones in the overall market can be intuitively reflected. The calculation of service resource density (combining the number of service providers and the density of service coverage touch points and combining the normalization of geographic area) further reveals the service supply capacity of the time zone. The combination of the two avoids the one-sidedness of simply judging market value by transaction amount. For example, a certain time zone may become a high-efficiency consumption area due to small geographic area but high concentration of service resources. By setting the screening condition of "top 20% of total transaction amount proportion and service resource density greater than the mean", the core market (i.e. strong time zone) with high consumption capacity and sufficient service support can be accurately located. Such time zones are often the focus of existing resource allocation for enterprises.

[0099] On this basis, for each strong time zone ± 2 hour neighborhood time zone (i.e. adjacent time zones closely related in time), the difference between theoretical consumption demand and actual consumption demand is calculated through the demand suppression index, which can effectively identify areas where consumption potential has not been released due to insufficient service supply, time mismatch and other factors. For example, users in a certain neighborhood time zone have high-frequency cross-time zone social behavior but local consumption data is low. The demand suppression index of this area may be significantly higher than the threshold, indicating that there is a possibility of consumption driven by the strong time zone, but the actual consumption is not as expected due to local service resources or time matching problems.

[0100] This step builds a double-layer analysis framework of "core market identification-potential market diagnosis": on the one hand, through the cross screening of transaction amount and service resource density, it ensures that the determination of strong time zone is based on both real consumption performance and the sustainability of service supply, avoiding misjudgment of occasional high consumption but weak service time zone as core market, thereby providing a reliable anchor for subsequent resource allocation; on the other hand, the introduction of demand suppression index breaks through the limitation of traditional market analysis which only focuses on existing consumption data, and can excavate the potential demand of adjacent time zones from the perspective of supply and demand matching. For example, the adjacent time zone (UTC+6) of a strong time zone (such as UTC+8) may have insufficient local service providers, resulting in the inability to meet the consumption demand of users in the night active period (overlapping with the strong time zone). The demand suppression index clearly quantifies this supply-demand gap, providing data basis for subsequent cross-time zone resource scheduling (such as increasing service push for the adjacent time zone in the golden overlapping period).

[0101] In addition, this step naturally fits the time continuity of user cross-time zone activities (such as cross-country commuting, night socializing, etc.) through the neighborhood association (±2 hours) of geographical time zone, making the analysis of demand suppression more close to the actual behavior pattern of users, and accurately positioning the consumption linkage potential area caused by time zone proximity, providing a scientific path for enterprises to shift from "wide net" market coverage to "precise radiation" resource allocation, especially suitable for fine management of cross-time zone operation in overseas market, effectively improving resource utilization efficiency and reducing invalid investment.

[0102] As shown in Figure 3 , the total transaction amount of the goods of each time zone and the service resource density are calculated, the time zone with the top 20% of total transaction amount of goods and the service resource density greater than the average value is defined as the strong time zone, and the demand suppression index is calculated for the adjacent time zone of each strong time zone within ±2 hours, specifically:

[0103] S210, extracting the transaction data of goods of each time zone from the structured data set, and calculating the total transaction amount of goods of each time zone;

[0104] S220, calculating the sum of total transaction amount of goods of all time zones, and calculating the proportion of total transaction amount of goods of each time zone according to the total transaction amount of goods of each time zone;

[0105] S230, extracting the service resource data of each time zone from the structured data set, and calculating the service resource density S k ;

[0106] S240, calculating the average value of service resource density of all time zones, and screening out the time zone with the top 20% of total transaction amount of goods and the service resource density greater than the average value, which is defined as the strong time zone;

[0107] S250, for each strong time zone, determine the ±2-hour neighboring time zone, extract the consumption demand data and service supply data of the neighboring time zone from the structured data set, and calculate the demand suppression index I j .

[0108] In this step, the service resource density S of each time zone is calculated. k :

[0109]

[0110] Among them, S p k is the number of service providers in time zone k, S c k is the number of service coverage contacts in time zone k, A k is the geographical area of ​​time zone k;

[0111] Calculate the demand suppression index I j :

[0112]

[0113] Among them, D th,j is the theoretical consumption demand in domain time zone j, D ac,j is the actual consumption demand in domain time zone j.

[0114] S300: Filter out consumption shadow zones based on the suppression index. Pair these zones with strong time zones with historical interaction records, calculate the overlap activity, and filter out golden overlap zones. Based on the average daily active overlap duration and cross-zone interactive user coverage of these golden overlap zones, the zones are graded and a time zone relationship map is constructed.

[0115] This step will identify "consumption shadow zones" based on the demand suppression index (>0.8). These time zones usually have a significant gap between theoretical consumption demand and actual consumption performance, and contain untapped market potential (for example, users in a certain time zone frequently participate in social activities in a strong time zone, but local consumption is sluggish).

[0116] Subsequently, by matching historical interaction records, the consumption shadow zone is paired with the strong time zone to ensure that the analysis object is the time zone combination where users actually have cross-zone behaviors (such as social interactions and consumer consultations), avoiding invalid matching of unrelated time zones.

[0117] When calculating overlapping activity, dynamic analysis of real-time active user numbers identifies "golden overlap zones" where users from both sides are simultaneously active and frequently interact. (For example, the high-frequency interaction period between 6:00 PM and 10:00 PM daily between UTC+8 and UTC+3 due to after-get off work social activities of multinational office workers) The system then categorizes these zones based on the average daily active overlap duration and cross-zone interactive user coverage. This allows companies to manage cross-time zone relationships in a hierarchical manner based on interaction intensity (for example, "super-grade" active zones correspond to core interaction periods with high duration and high coverage). The resulting time zone relationship map visualizes the network of connections formed by each time zone through the golden overlap zones, clearly demonstrating the radiation paths and key reach periods of strong time zones to consumer shadow zones.

[0118] This step combines the "unmet demand area" with the "actual interaction period" to form a three-dimensional analysis model of "potential market positioning - high-frequency interaction period capture - association strength grading".

[0119] On the one hand, the screening of consumption shadow zones breaks through the traditional geographical boundary restrictions and focuses on the time dimension gap of supply and demand imbalance (such as the suppression of night-time consumption demand due to service resource mismatch in a certain time zone), while the matching rules of historical interaction records ensure the actual relevance of the analysis objects and avoid subjective pairing based on assumptions; on the other hand, the dynamic screening mechanism of the golden overlapping zone can capture the "time intersection" of user behavior across time zones. For example, in cross-border e-commerce scenarios, the active overlapping periods of European users and Asian suppliers may be concentrated in the morning of working days for both parties. The precise positioning of this period can significantly improve customer service response efficiency and marketing reach.

[0120] The grading strategy provides a direct basis for differentiated resource allocation, increasing real-time customer service resources and personalized recommendation traffic for the "super" active band, maintaining basic services for the "ordinary" active band, and avoiding efficiency loss caused by average resource distribution.

[0121] In addition, the time zone relationship map, as a visualization tool, helps companies intuitively identify core nodes (strong time zones) and key connections (golden overlapping zones) in the cross-time zone ecosystem, providing clear path guidance for subsequent targeted operations in consumer shadow areas (such as pushing customized product combinations during overlapping time periods). It is especially suitable for the problem of scattered user active periods caused by time zone differences in overseas markets. Through data-driven precise matching, it achieves "zero time difference" demand response and maximizes the input-output ratio of cross-time zone operations.

[0122] like Figure 4 As shown, the calculation of overlapping activity, and the classification of active zones according to the average daily active overlapping time of the golden overlapping zone and the cross-zone interactive user coverage rate, and the construction of the time zone relationship map are specifically as follows:

[0123] S310, from the calculated demand suppression index results, screen out time zones with a demand suppression index greater than 0.8 and define them as consumption shadow zones;

[0124] S320: Searching for historical interaction records between the consumption shadow zone and other time zones in the structured data set, and pairing the consumption shadow zone with a strong time zone with historical interaction records.

[0125] S330, for each pair of consumption shadow zone and strong time zone, extract their social activity data at different times from the structured dataset and calculate the overlapping activity at each time point;

[0126] S340, screen out the period with overlapping activity > 0.6 and define it as the golden overlapping zone;

[0127] S350, calculate the average daily active overlap duration in each golden overlap zone, which is defined as the average daily active overlap duration;

[0128] S360 calculates the ratio of users participating in cross-region interactions in each golden overlapping zone to the total number of users, which is defined as the user coverage rate of cross-region interactions;

[0129] S370 divides the golden overlap zone into different levels based on the average daily active overlap time and cross-region interactive user coverage;

[0130] S380, uses time zones as nodes and golden overlapping belts as edges to construct a time zone relationship map.

[0131] In this step, the overlapping activity at each time point is calculated:

[0132]

[0133] Among them, O m,n,t A represents the overlapping activity between time zones m and n at time t, m,n,t S represents the number of users who are active simultaneously in time zones m and n at time t. m,t and S n,t Respectively represent the number of independent active users in time zone m and time zone n at time t;

[0134] Average daily active overlap time:

[0135]

[0136] Among them, L m,n is the average daily active overlap duration between time zones m and n, D is the number of days in the statistical period, and Δt is the duration of each golden overlap zone in minutes;

[0137] User coverage of cross-region interactions:

[0138]

[0139] Among them, C x,y Indicates the user coverage rate of cross-region interaction between the strong time zone x and the consumption shadow zone y, U x,y is the number of users who participated in cross-zone interaction between strong time zone x and consumption shadow zone y during the statistical period, U x 、U y They are the total number of users in the strong time zone x and the consumption shadow zone y respectively.

[0140] S400 maps the consumption shadow zone to a time zone relationship map, locates the overlapping periods between the consumption shadow zone and the strong time zone, and constructs a dual feature matrix to analyze the behavior and social topic sentiment of users in the consumption shadow zone during the overlapping periods, outputting the potential demand elasticity coefficient and recommended product categories;

[0141] This step, through spatiotemporal correlation positioning and micro-behavioral analysis, explores potential demand in consumer shadow zones, moving from the macro time zone level down to the dynamic behavior scenarios of individual users. Specifically, by mapping consumer shadow zones onto a time zone relationship map, we can pinpoint their golden overlap with dominant time zones (e.g., the nighttime overlap between a Southeast Asian time zone and a dominant European time zone). This period is often a high-frequency period for cross-regional user social interaction and the emergence of potential consumption intentions.

[0142] Subsequently, the user's behavioral data during that time period (such as consumption frequency, product browsing time, and cross-regional social content) is extracted from the structured dataset to construct a dual feature matrix that integrates behavioral characteristics (consumption time distribution, preferred product categories) and social topic characteristics (emotional tendencies, discussion keywords). This allows companies to capture users' true demand signals at specific time and space intersections. For example, a user in a consumer shadow zone frequently participates in active social discussions related to "outdoor equipment" during overlapping time periods, but this category accounts for a relatively low proportion in their actual consumption records. Combined with sentiment analysis, this can be used to determine the presence of unmet potential demand.

[0143] By introducing the potential demand elasticity coefficient calculation with the emotion correction factor, the correlation between marketing resource investment and consumer response can be dynamically evaluated. For example, topics dominated by positive emotions may amplify the pulling effect of marketing investment on consumption, thereby identifying product categories with real growth potential (such as categories with an elasticity coefficient > 1 after emotion correction).

[0144] This step breaks through the static analysis model of traditional market segmentation that relies on historical consumption data, and instead uses dynamic modeling of "overlapping time and space scenarios + real-time behavioral emotions" to accurately capture potential needs.

[0145] The mapping mechanism of the time zone relationship graph ensures that the analysis object focuses on the time period of real cross-zone interaction, avoiding pseudo-demand judgment that deviates from the actual behavior of the user; the construction of the double feature matrix combines the cold consumption data with the fresh social emotions, for example, a user in a Middle East consumption shadow zone expresses strong interest in "convenient smart home" (emotional tendency is positive) through a social platform in the overlapping period with the lunch break of the strong time zone in East Asia. Although the current consumption record is less, combined with the browsing time and interaction frequency, it can be determined that this category has high potential demand elasticity, and thus becomes the key recommendation object.

[0146] This analysis method is especially suitable for scenarios with large cultural differences and significant social influence on consumption habits in overseas markets, and can identify "silent demand" caused by time zone mismatch, information gap, etc.

[0147] In addition, the introduction of the emotional correction factor corrects the mechanicalness of traditional demand elasticity calculation, making the analysis result closer to the real decision-making process of the user (such as impulsive consumption driven by social emotions), providing the enterprise with a "user real-time emotion-oriented" precise recommendation strategy, and significantly improving the conversion efficiency of marketing resources. Through this step, the enterprise not only can find high-potential product categories that are not apparent in existing data, but also can develop "zero-time difference" personalized recommendation schemes based on the behavior patterns in the overlapping period, achieving differentiated demand activation and market penetration in the highly competitive overseas market.

[0148] As shown in Figure 5 , the consumption shadow zone is mapped to the time zone relationship graph, the overlapping period of the consumption shadow zone and the strong time zone is located, and a double feature matrix is constructed, the behavior of the user in the consumption shadow zone in the overlapping period and the emotional tendency of the social topic are analyzed, and the potential demand elasticity coefficient and the recommended product category are output, specifically:

[0149] S410, match the time zone information of the consumption shadow zone with the nodes in the time zone relationship graph, and map the consumption shadow zone to the time zone relationship graph;

[0150] S420, find the strong time zone paired with the consumption shadow zone in the time zone relationship graph, and locate the golden overlapping band between the consumption shadow zone and the strong time zone;

[0151] S430, extract the behavior data of the user in the consumption shadow zone in the golden overlapping band period from the structured data set, and construct a double feature matrix, wherein: the matrix row represents the user in the consumption shadow zone, and the matrix column represents the behavior feature and the social topic feature of the user;

[0152] S440, perform emotional analysis on the social topic data to determine the emotional tendency of each social topic;

[0153] S450, analyzing the dual feature matrix and calculating the potential demand elasticity coefficient of users in the consumption shadow zone during the overlapping period;

[0154] S460, based on the behavioral characteristics, social topic sentiment tendencies and potential demand elasticity coefficients of users in the consumption shadow zone during the overlapping period, select E i Categories with a value greater than 1 are recommended as demand product categories.

[0155] In this step, the potential demand elasticity coefficient of users in the consumption shadow zone during the overlapping period is calculated:

[0156]

[0157] Among them, E i represents the potential demand elasticity coefficient of user i in the consumption shadow zone during the overlapping period, Q i represents the benchmark consumption of the commodity category during the overlapping time period, ΔQ i represents the consumption change of commodity categories during the overlapping period, P i Indicates the amount of marketing resources invested per unit time in the consumption shadow area, ΔP i Indicates the change in marketing resource input per unit time in the consumption shadow area, is the emotion correction factor.

[0158] S500, dynamically allocates resources and adjusts traffic scheduling strategies during the golden overlap period based on the potential demand elasticity coefficient.

[0159] This step divides product categories into high, medium, and low elasticity demand categories based on the potential demand elasticity coefficient of users in the consumption shadow zone (calculated based on comprehensive behavioral characteristics and social emotional tendencies).

[0160] For example, in a certain consumption shadow zone that overlaps with the strong time zone in Europe, the "cross-border e-commerce logistics service" category shows high elasticity of demand due to high-frequency positive social discussions among users, while the "local fresh food" category shows low elasticity due to delivery time constraints.

[0161] For high-elasticity categories, increase advertising budgets, prioritize customer service resources, or launch limited-time discounts during the golden overlap period, and use the peak period of user activity to enhance demand conversion; for low-elasticity categories, reduce unnecessary resource investment to avoid ineffective exposure.

[0162] Traffic scheduling strategies simultaneously integrate user activity and demand elasticity. For example, personalized recommendations are pushed to highly elastic user groups during the initial overlap period. Traffic allocation across channels is dynamically adjusted based on real-time click-through conversion rates in the mid-term, and a secondary reach mechanism (such as exclusive coupon push) is triggered for non-converting users in the final period. The real-time monitoring module continuously tracks consumption data (such as clicks, purchase conversion rates, and user stay time) after resource allocation. By comparing and analyzing historical data, it automatically calibrates resource allocation ratios and traffic paths for the next cycle, forming a closed-loop management system of "analysis-execution-feedback-optimization."

[0163] This step overcomes the limitations of traditional resource allocation, which relies on empirical judgment, and establishes a three-dimensional dynamic control system: "demand elasticity, time-of-day value, and real-time feedback." By converting potential demand elasticity into quantifiable resource allocation metrics, companies can accurately identify high-value input-output intervals across time zones. For example, during the prime overlap period between North America and East Asia (mornings on both sides of the workday), immersive advertising can be concentrated in the highly elastic "cross-border payment tool" category. This leverages users' frequent demand for financial services during work breaks to improve conversion efficiency, reducing customer acquisition costs by over 30% compared to evenly distributed advertising.

[0164] The real-time monitoring and dynamic adjustment mechanism is particularly suitable for scenarios in overseas markets where user behavior is significantly affected by factors such as time zone differences and cultural festivals. For example, during Ramadan in the Middle East, monitoring revealed that the demand elasticity for "online entertainment services" among users in a certain consumption shadow zone during the overlapping period at night increased sharply. The system can immediately allocate more streaming media resources to seize the brief consumption window period.

[0165] In addition, this strategy effectively solves the problem of "resource mismatch" in cross-time zone operations through differentiated resource allocation, avoiding ineffective push during user dormant periods, while concentrating firepower to break through during high-potential periods, creating a "precision strike" type of market penetration effect.

[0166] This data-driven dynamic adjustment not only improves resource utilization efficiency, but also helps companies quickly respond to changes in demand in the highly competitive overseas market by continuously optimizing traffic scheduling paths, building an agile operation system based on real-time behavioral data, and ultimately achieving a core capability leap from "extensive coverage" to "precise monetization."

[0167] like Figure 6 As shown, according to the potential demand elasticity coefficient, resources are dynamically allocated and traffic scheduling strategies are adjusted during the golden overlap period, specifically:

[0168] S510, classifying different commodity categories according to the potential demand elasticity coefficients of users in the consumption shadow zone, including: high elasticity demand category, medium elasticity demand category, and low elasticity demand category;

[0169] S520: During the golden overlap period, increase resource input for high-elasticity demand categories, maintain current resource input for medium-elasticity demand categories, and reduce resource input for low-elasticity demand categories.

[0170] S530: Adjust the traffic scheduling strategy based on the activity and demand elasticity coefficient of users in the consumption shadow zone during the golden overlap period.

[0171] S540 monitors the resource allocation and traffic scheduling effects during the golden overlap period in real time, and dynamically adjusts the resource allocation and traffic scheduling strategies based on the monitoring results.

[0172] Figure 7 The structural block diagram of the overseas market segmentation system based on customer data provided by an embodiment of the present invention is as follows: Figure 7 As shown, the system includes:

[0173] The user data collection module 100 is used to obtain user time zone data, cross-time zone social behavior data, and consumption behavior data through the API interface, and to construct a standardized time series matrix to generate a structured data set;

[0174] The demand suppression index calculation module 200 is used to calculate the proportion of total commodity transactions and service resource density in each time zone based on the structured data set, and to screen out the strong time zones. For each strong time zone, the demand suppression index is calculated for the time zones within ±2 hours of the neighborhood.

[0175] The golden overlap zone screening module 300 is used to screen out consumption shadow zones based on the suppression index, pair consumption shadow zones with strong time zones with historical interaction records, calculate overlapping activity, screen out golden overlap zones, and classify active zones based on the average daily active overlap duration and cross-zone interactive user coverage of the golden overlap zones, thereby constructing a time zone relationship map.

[0176] The consumer behavior analysis module 400 is used to map the consumption shadow zone to the time zone relationship map, locate the overlapping time periods of the consumption shadow zone and the strong time zone, and construct a dual feature matrix to analyze the behavior and social topic sentiment of users in the consumption shadow zone during the overlapping time periods, and output the potential demand elasticity coefficient and recommended product categories;

[0177] The dynamic resource allocation module 500 is used to dynamically allocate resources and adjust traffic scheduling strategies during the golden overlap period according to the potential demand elasticity coefficient.

[0178] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0179] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

[0180] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. Overseas market segmentation method based on customer data, characterized by: The method comprises: Obtain user time zone data, cross-time zone social behavior data, and consumption behavior data through the API interface, and build a standardized time series matrix to generate a structured data set; Based on a structured data set, we calculated the proportion of total merchandise transaction volume and service resource density in each time zone, screened out the dominant time zones, and calculated the demand suppression index for each dominant time zone within ±2 hours of the neighboring time zones. Based on the suppression index, we screen out consumption shadow zones, pair them with strong time zones with historical interaction records, and calculate the overlapping activity to identify golden overlapping zones. Based on the average daily active overlap duration and cross-zone interactive user coverage of the golden overlapping zones, we rank the active zones and construct a time zone relationship map. Mapping the consumption shadow zone to a time zone relationship map, locating the overlapping periods between the consumption shadow zone and the strong time zone, and constructing a dual-feature matrix to analyze the behavior and social topic sentiment of users in the consumption shadow zone during the overlapping periods, outputting the potential demand elasticity coefficient and recommended product categories; Based on the potential demand elasticity coefficient, resources are dynamically allocated and traffic scheduling strategies are adjusted during the golden overlap period.

2. The method according to claim 1, characterized in that The API interface is used to obtain user time zone data, cross-time zone social behavior data, and consumption behavior data, and to construct a standardized time series matrix to generate a structured data set, specifically: Obtain user time zone data, cross-time zone social behavior data, and consumption behavior data through the API interface, including: The user time zone data includes: time zone information of the user's geographical location and time zone change records; The cross-time zone social behavior data includes: the user's social login time, social interaction frequency, and social content publishing time in different time zones; The consumption behavior data includes: the user's consumption time, consumption amount, and consumption products; Clean the acquired user time zone data, cross-time zone social behavior data, and consumption behavior data to remove duplicate, erroneous, and missing data; Arrange the cleaned data in chronological order and construct a standardized time series matrix. The rows of the matrix represent different time points and the columns represent different data indicators. The standardized time series matrix is ​​subjected to feature extraction and transformation to generate a structured data set, which contains the time zone information, social behavior characteristics, and consumption behavior characteristics of each user at different time points.

3. The method according to claim 1, characterized in that The calculation of the proportion of total merchandise transaction volume and service resource density in each time zone defines the time zones with the top 20% of total merchandise transaction volume and service resource density greater than the average as strong time zones. For each strong time zone and the neighboring time zones within ±2 hours, the demand suppression index is calculated as follows: Extract commodity transaction data for each time zone from the structured data set and calculate the total commodity transaction amount for each time zone; Calculate the sum of the GMT of all time zones, and then calculate the GMT share of each time zone based on the GMT of each time zone. Extract the service resource data of each time zone from the structured data set and calculate the service resource density S of each time zone k ; Calculate the average service resource density of all time zones and select the time zones with the top 20% of total merchandise transaction volume and a service resource density greater than the average, defining them as strong time zones. For each strong time zone, determine the ±2-hour neighboring time zone, extract the consumption demand data and service supply data of the neighboring time zone from the structured data set, and calculate the demand suppression index I j .

4. The method according to claim 3, characterized in that The calculation of the service resource density S of each time zone k : Among them, S p k is the number of service providers in time zone k, S c k is the number of service coverage contacts in time zone k, A k is the geographical area of ​​time zone k; Calculate the demand suppression index I j : Among them, D th,j is the theoretical consumption demand in domain time zone j, D ac,j is the actual consumption demand in domain time zone j.

5. The method according to claim 1, wherein The calculation of overlapping activity, as well as the classification of active zones based on the average daily active overlapping duration of the golden overlapping zone and the cross-zone interactive user coverage rate, and the construction of a time zone relationship map are specifically as follows: From the calculated demand suppression index results, we screen out time zones with a demand suppression index greater than 0.8 and define them as consumption shadow zones; Search the structured dataset for historical interaction records between consumption shadow zones and other time zones, and pair these zones with strong time zones where historical interaction records exist. For each pair of consumption shadow zone and strong time zone, their social activity data at different times are extracted from the structured dataset, and the overlapping activity at each time point is calculated; Screen out the time periods with overlapping activity > 0.6 and define them as the golden overlapping zone; Calculate the average daily active overlap duration within each golden overlap zone, which is defined as the average daily active overlap duration; Calculate the ratio of users participating in cross-region interactions in each golden overlap zone to the total number of users, which is defined as the user coverage rate of cross-region interactions. The golden overlap zone is divided into different levels based on the average daily active overlap time and cross-region interactive user coverage rate; Using time zones as nodes and golden overlapping belts as edges, a time zone relationship map is constructed.

6. The method according to claim 5, characterized in that The calculation of overlapping activity at each time point is: Among them, O m,n,t A represents the overlapping activity between time zones m and n at time t, m,n,t S represents the number of users who are active simultaneously in time zones m and n at time t. m,t and S n,t Respectively represent the number of independent active users in time zone m and time zone n at time t; Average daily active overlap time: Among them, L m,n is the average daily active overlap duration between time zones m and n, D is the number of days in the statistical period, and Δt is the duration of each golden overlap zone in minutes; User coverage of cross-region interactions: Among them, C x,y Indicates the user coverage rate of cross-region interaction between the strong time zone x and the consumption shadow zone y, U x,y is the number of users who participated in cross-zone interaction between strong time zone x and consumption shadow zone y during the statistical period, U x 、U y They are the total number of users in the strong time zone x and the consumption shadow zone y respectively.

7. The method according to claim 1, characterized in that The consumption shadow zone is mapped to the time zone relationship map, the overlapping period of the consumption shadow zone and the strong time zone is located, and a dual feature matrix is ​​constructed to analyze the behavior and social topic sentiment of users in the consumption shadow zone during the overlapping period, and output the potential demand elasticity coefficient and recommended product categories, specifically: Match the time zone information of the consumption shadow area with the nodes in the time zone relationship map, and map the consumption shadow area to the time zone relationship map; Find the strong time zone that pairs with the consumption shadow zone in the time zone relationship map, and locate the golden overlap zone between the consumption shadow zone and the strong time zone; Extract the behavioral data of users in the consumption shadow zone during the golden overlap period from the structured data set; Construct a dual-feature matrix, where the rows represent users in the consumption shadow zone, and the columns represent the user's behavioral characteristics and social topic characteristics; Conduct sentiment analysis on social topic data to determine the sentiment tendency of each social topic; Analyze the dual-feature matrix and calculate the potential demand elasticity coefficient of users in the consumption shadow zone during the overlapping period; According to the behavioral characteristics of users in the consumption shadow zone during the overlapping period, the emotional tendency of social topics and the potential demand elasticity coefficient, E i Categories with a value greater than 1 are recommended as demand product categories.

8. The method according to claim 1, characterized in that The calculation of the potential demand elasticity coefficient of users in the consumption shadow zone during the overlapping period is: Among them, E i represents the potential demand elasticity coefficient of user i in the consumption shadow zone during the overlapping period, Q i represents the benchmark consumption of the commodity category during the overlapping time period, ΔQ i represents the consumption change of commodity categories during the overlapping period, P i Indicates the amount of marketing resources invested per unit time in the consumption shadow area, ΔP i It represents the change in marketing resources invested per unit time in the consumption shadow area. is the emotion correction factor.

9. The method according to claim 1, characterized in that The method of dynamically allocating resources and adjusting traffic scheduling strategies during the golden overlap period based on the potential demand elasticity coefficient is as follows: Based on the potential demand elasticity coefficient of users in the consumption shadow zone, different product categories are classified into: high elasticity demand category, medium elasticity demand category, and low elasticity demand category; During the golden overlap period, increase resource input for high-elasticity demand categories, maintain current resource input for medium-elasticity demand categories, and reduce resource input for low-elasticity demand categories. Adjust traffic scheduling strategies based on the activity and demand elasticity coefficient of users in the consumption shadow zone during the golden overlap period; Monitor the resource allocation and traffic scheduling effects during the golden overlap period in real time, and dynamically adjust the resource allocation and traffic scheduling strategies based on the monitoring results.

10. Overseas market segmentation system based on customer data, characterized by: The system comprises: The user data collection module is used to obtain user time zone data, cross-time zone social behavior data, and consumption behavior data through the API interface, and construct a standardized time series matrix to generate a structured data set; The demand suppression index calculation module is used to calculate the proportion of total commodity transactions and service resource density in each time zone based on a structured data set, and to screen out strong time zones. The demand suppression index is calculated for each strong time zone within ±2 hours of the neighboring time zone. The Golden Overlap Zone Screening Module is used to screen out consumption shadow zones based on the suppression index. It then pairs these zones with strong time zones with historical interaction records, calculates the overlap activity, and identifies the Golden Overlap Zones. Based on the average daily active overlap duration and cross-zone interactive user coverage of the Golden Overlap Zones, the active zones are graded and a time zone relationship map is constructed. The consumer behavior analysis module is used to map the consumption shadow zone to the time zone relationship map, locate the overlapping periods of the consumption shadow zone and the strong time zone, and construct a dual feature matrix to analyze the behavior and social topic sentiment of users in the consumption shadow zone during the overlapping periods, outputting the potential demand elasticity coefficient and recommended product categories; The dynamic resource allocation module is used to dynamically allocate resources and adjust traffic scheduling strategies during the golden overlap period based on the potential demand elasticity coefficient.

Images