CRM customer ecosphere dynamic visual management system based on GIS

Abstract

The invention relates to the technical field of geographic information data, in particular to a GIS-based CRM customer ecosphere dynamic visual management system, which comprises the following steps: acquiring geographical location information, transaction data, customer relationship data and time sequence data of customers; constructing a customer ecosphere space model to analyze the geographic position information and the customer relationship data, obtaining space association strength and business association strength among customers, and generating a dynamic evolution matrix; performing real-time spatial clustering analysis on the transaction data, and identifying a client distribution hotspot area; based on the time sequence data and the geographic position information of the customer, constructing a customer space-time trajectory model, analyzing a space-time behavior pattern of the customer, and obtaining a space-time trajectory prediction result including a customer periodic behavior pattern and a space activity trend; and carrying out three-dimensional rendering on the spatial association strength, the business association strength, the customer distribution hotspot area and the spatio-temporal trajectory prediction result to obtain a three-dimensional visual interface.

Description

Technical Field

[0001] The present invention relates to the technical field of geographic information data, and particularly to a dynamic visualization management system for a CRM customer ecosystem based on GIS. Background Art

[0002] With the rapid development of information technology, customer relationship management systems have become important tools for enterprises to improve customer service quality and marketing efficiency. Traditional customer relationship management systems mainly store basic customer information through relational databases, use attribute-based classification methods to statically group customers, process customer service requests through linear business processes, and present the results of customer data analysis in the form of static reports. However, these systems have significant limitations in processing customer geographical location information. Usually, they can only perform simple address text storage and basic map point marking, lack professional spatial analysis capabilities, and cannot achieve advanced geographical information processing functions such as spatial relationship analysis, spatial clustering analysis, and geographical area optimization.

[0003] Although existing geographic information systems can provide basic map services and simple spatial query functions in commercial applications, their analysis capabilities are mainly limited to basic functions such as point marking, area division, route planning, and simple buffer analysis. When integrating with customer relationship management systems, they face technical obstacles such as inconsistent data formats, poor real-time performance, and single analysis dimension, resulting in the inability to effectively integrate spatial information and customer relationship data.

[0004] In terms of customer relationship modeling, existing technologies mainly adopt single-customer attribute models and linear relationship chain modeling methods, construct customer relationships through static grouping and transaction-based association methods, and the analysis dimensions focus on customer value analysis, purchase behavior analysis, life cycle management, and satisfaction evaluation. This modeling method lacks in-depth analysis of the overall relationship of the customer ecosystem, especially in aspects such as customer network relationship analysis, consideration of spatial influencing factors, capture of dynamic changes, and group effect analysis, and cannot construct a complete customer ecological network model. In addition, existing systems mainly rely on static charts and simple map displays in dynamic visualization, lack real-time dynamic update mechanisms and multi-dimensional interactive analysis capabilities, and cannot meet the actual needs of enterprises for real-time monitoring and dynamic management of the customer ecosystem.

[0005] Therefore, a dynamic visualization management system for a CRM customer ecosystem based on GIS is proposed. Summary of the Invention

[0006] The object of the present invention is to provide a GIS-based dynamic visualization management system for CRM customer ecosystem, which includes obtaining the geographical location information, transaction data, customer relationship data and time series data of customers; constructing a customer ecosystem spatial model to analyze the geographical location information and customer relationship data, obtaining the spatial association strength and business association strength between customers, and generating a dynamic evolution matrix; performing real-time spatial clustering analysis on the transaction data to identify the hot spots of customer distribution; constructing a customer spatio-temporal trajectory model based on the time series data and geographical location information of customers, analyzing the spatio-temporal behavior patterns of customers, and obtaining spatio-temporal trajectory prediction results including customer periodic behavior patterns and spatial activity trends; and performing three-dimensional rendering on the spatial association strength, business association strength, customer distribution hot spots and spatio-temporal trajectory prediction results to obtain a three-dimensional visualization interface.

[0007] To achieve the above object, the present invention provides the following technical solutions: A GIS-based dynamic visualization management system for CRM customer ecosystem, including: A data collection module for obtaining the geographical location information, transaction data, customer relationship data and time series data of customers; An ecosystem modeling module for constructing a customer ecosystem spatial model to analyze the geographical location information and customer relationship data, obtaining the spatial association strength and business association strength between customers, and generating a dynamic evolution matrix; A spatial clustering module for performing real-time spatial clustering analysis on the transaction data to identify the hot spots of customer distribution; A spatio-temporal trajectory prediction module for constructing a customer spatio-temporal trajectory model based on the time series data and geographical location information of customers, analyzing the spatio-temporal behavior patterns of customers, and obtaining spatio-temporal trajectory prediction results including customer periodic behavior patterns and spatial activity trends; A multi-dimensional dynamic visualization module for performing three-dimensional rendering on the dynamic evolution matrix, customer distribution hot spots and spatio-temporal trajectory prediction results to obtain a three-dimensional visualization interface, supporting multi-touch interaction, early warning mechanism and real-time adjustment of time slices, spatial range boundaries and analysis dimension levels.

[0008] Preferably, the geographical location information includes the longitude and latitude coordinate data, administrative division code, detailed address information, altitude data and geographical boundary range parameters of customers; the transaction data includes transaction amount value, transaction frequency, transaction timestamp, commodity category identifier, transaction channel information, payment method code and transaction status mark; the customer relationship data includes recommendation relationship graph, partner association table, social network connection information, customer level, credit score data and contact frequency statistics; and the time series data includes customer historical behavior record sequence, system access timestamp set, activity cycle identifier, seasonal mark and trend change index.

[0009] Preferably, the customer ecosystem space model includes: A spatial relationship layer, which obtains the geometric topological relationship of customer spatial distribution based on longitude and latitude coordinate data, administrative division codes, and altitude data, and generates a spatial association intensity; A business association layer, which performs graph theory analysis on the recommendation relationship graph, partner association table, and social network connection information, and adjusts it in combination with customer level and credit score data to obtain the business association intensity between customers; A spatio-temporal evolution layer, which performs time series analysis on the spatial association intensity and business association intensity based on historical time series data, and constructs a dynamic evolution matrix of the customer ecosystem.

[0010] Preferably, the specific process for obtaining the customer distribution hot spot area is as follows: A data preprocessing layer, which standardizes the transaction amount value, transaction frequency, and longitude and latitude coordinate data, and combines the transaction channel information and commodity category identifier to obtain the transaction density, and generates a normalized transaction density vector; A spatial density calculation layer, which performs spatial interpolation calculation on the transaction density vector, and combines the geographical boundary range parameter to obtain the customer activity density value within the region; A dynamic clustering layer, which performs spatial clustering analysis on the customer activity density value, identifies the spatial regions exceeding the density threshold, and performs time series analysis in combination with the transaction timestamp to generate a set of coordinates of the customer distribution hot spot area with time attributes and geometric boundaries; A hot spot area output layer, which analyzes the set of coordinates of the identified customer distribution hot spot area in combination with the dynamic evolution matrix, customer level classification, and credit score data, and generates a customer distribution hot spot area, including geographical boundaries, intensity levels, time characteristics, and customer composition information.

[0011] Preferably, the spatio-temporal trajectory prediction module includes: A trajectory feature extraction layer, which extracts features from the historical behavior record sequence, system access timestamp set, and longitude and latitude coordinate data, obtains the position change pattern of the customer in different time periods, and generates a trajectory feature vector; A customer periodic behavior pattern recognition layer, which identifies the spatial movement behavior of the customer based on the trajectory feature vector, activity cycle identifier, and seasonal marker, obtains the daily activity area, movement path preference, and time regularity of the customer, and generates customer periodic behavior pattern parameters, including activity cycle frequency, spatial preference area, and behavior stability index; A spatial activity trend prediction layer, which combines the customer distribution hot spot area to predict the future spatial activity trend of the customer, and generates a spatial activity trend including probability distribution and confidence interval.

[0012] Preferably, the three-dimensional rendering process includes: Convert the spatial association intensity, business association intensity, customer distribution hot spot areas, and spatio-temporal trajectory prediction results into a three-dimensional coordinate system, perform a projection transformation from geographic coordinates to screen coordinates using longitude and latitude coordinate data and altitude data, and generate the position information of spatial objects in the three-dimensional scene; generate a set of geometric body parameters based on the spatial object position information and customer level classification; assign material attributes and texture coordinates to the set of geometric body parameters according to spatio-temporal trajectory prediction and product category identification; the set of geometric body parameters combines ambient light, diffuse light, and specular light, adjusts the light intensity and color temperature according to seasonal markings, and generates three-dimensional rendered image data, and the three-dimensional rendered image data is output to a three-dimensional visualization interface for display.

[0013] Preferably, the early warning mechanism includes: monitoring the change rates of the spatial association intensity and the business association intensity for thresholds, identifying changes in customer relationships that exceed the normal fluctuation range, and generating abnormal event identifiers; combining customer levels, credit score data, and historical abnormal events, evaluating the risk levels of the identified abnormal event identifiers, and generating a risk assessment report including the degree of risk, the scope of influence, and the processing priority; according to the risk assessment report and user permission settings, display real-time warning information through the three-dimensional visualization interface.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. By constructing a spatial model of the customer ecosystem, the present invention realizes the in-depth fusion analysis of geographical location information and customer relationship data. The present invention constructs a geometric topological relationship network based on customer longitude and latitude, administrative division, and altitude data to obtain the spatial association intensity between customers; analyzes the recommendation relationship graph, partner table, and social network information through graph theory methods, and combines weight information such as customer levels and credit scores to obtain the business association intensity between customers. This not only improves the accuracy of customer relationship modeling but also supports the identification of more complex cross-regional and cross-business logic customer behaviors, providing a scientific analysis basis for market promotion, customer stratification management, and strategic customer expansion.

[0015] 2. By constructing a customer spatio-temporal trajectory model and integrating time series data and spatial position information, the present invention can accurately analyze and predict the spatio-temporal behavior patterns of customers. It includes multiple layers of modules such as trajectory feature extraction, periodic behavior identification, and spatial trend prediction. By modeling multi-dimensional data such as customer historical behaviors, system access records, and position information, it can identify customers' daily activity areas, mobile path preferences, and their behavior patterns, and generate behavior pattern parameters such as activity cycles, spatial preferences, and stability. Based on the above model, further integrating the dynamic changes of hot spot areas, it can predict the possible future activity areas of customers and give confidence analysis, providing strong support for precision marketing, regional network point selection, and personalized service configuration, and effectively improving the customer reach efficiency and business response ability.

[0016] 3. By constructing a multi-dimensional dynamic visualization module and a three-dimensional stereoscopic rendering mechanism, the present invention realizes the real-time visualization of customer ecosystem data in a three-dimensional coordinate system, significantly improving the information transmission efficiency of customer management and the intuitiveness of user operations. Through geometric body generation, texture mapping, and dynamic lighting modeling, combined with environmental parameters and customer attribute rendering effects, the realistic expression of spatial objects is achieved. Users can freely adjust time slices, spatial ranges, and dimensional levels through multi-touch interaction. At the same time, a built-in risk warning mechanism can generate warning prompts in real time and dynamically mark them in the three-dimensional view through monitoring of the change in association strength and risk assessment of abnormal events, enhancing the decision-makers' perception ability and response efficiency to changes in customer relationships. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic structural diagram of a dynamic visualization management system for a GIS-based CRM customer ecosystem provided by the present invention; Figure 2 It is a schematic diagram of the dynamic visualization management process of a GIS-based CRM customer ecosystem provided by the present invention; Figure 3 It is a schematic diagram of the dynamic visualization of the customer ecosystem provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0019] Embodiment 1: The present invention provides a dynamic visualization management system for a GIS-based CRM customer ecosystem. For specific reference, please refer to Figure 1 and Figure 2 . The technical solutions are as follows: A data acquisition module for obtaining the geographical location information, transaction data, customer relationship data, and time series data of customers; An ecosystem modeling module for constructing a customer ecosystem space model to analyze the geographical location information and customer relationship data, obtaining the spatial association strength and business association strength between customers, and generating a dynamic evolution matrix; A spatial clustering module for performing real-time spatial clustering analysis on transaction data to identify hotspots of customer distribution; A spatio-temporal trajectory prediction module, which is used to construct a customer spatio-temporal trajectory model based on the customer's time series data and geographical location information, analyze the customer's spatio-temporal behavior patterns, and obtain spatio-temporal trajectory prediction results including the customer's periodic behavior patterns and spatial activity trends; A multi-dimensional dynamic visualization module, which is used to perform three-dimensional rendering on the dynamic evolution matrix, customer distribution hot spot areas, and spatio-temporal trajectory prediction results, obtain a three-dimensional visualization interface, and support multi-touch interaction, an early warning mechanism, and real-time adjustment of time slices, spatial range boundaries, and analysis dimension levels.

[0020] The geographical location information includes the customer's latitude and longitude coordinate data, administrative division code, detailed address information, altitude data, and geographical boundary range parameters; the transaction data includes the transaction amount value, transaction frequency, transaction timestamp, commodity category identifier, transaction channel information, payment method code, and transaction status flag; the customer relationship data includes a recommendation relationship graph, a partner association table, social network connection information, customer level, credit score data, and contact frequency statistics; the time series data includes the customer's historical behavior record sequence, system access timestamp set, activity cycle identifier, seasonal marker, and trend change indicator.

[0021] In this embodiment, through a standardized multi-dimensional data structure, the system can achieve high-precision spatial positioning and comprehensive customer portrait construction, provide a high-quality data basis for subsequent spatial analysis and relationship mining, ensure data consistency and comparability, and improve the accuracy and reliability of the system analysis results.

[0022] The customer ecosystem space model includes: A spatial relationship layer, which uses the Delaunay triangulation algorithm to process latitude and longitude coordinates, combines with the Voronoi diagram to divide the customer influence area, realizes cross-regional association analysis through the administrative division code, and uses altitude data to calculate the three-dimensional spatial distance, and finally generates a 0-1 standardized spatial association strength; A business association layer, which uses a graph neural network algorithm to analyze the transitivity and influence of the recommendation relationship graph, uses a collaborative filtering algorithm to process the partner association table to identify potential cooperation opportunities, calculates the node importance of social network connection information through the PageRank algorithm, and uses the customer level and credit score as weight factors to adjust the association strength, and generates a business association strength reflecting the business closeness; A spatio-temporal evolution layer, which uses time series analysis methods to perform trend analysis and periodic detection on the historical changes of the spatial association strength and business association strength, constructs a Markov transition matrix to describe the state transition probability of the customer ecosystem, and forms a dynamic evolution matrix that can reflect the dynamic change law of customer relationships.

[0023] In this embodiment, the spatial model of the three-layer architecture can comprehensively depict the complex relationship network of the customer ecosystem from three dimensions: space, business, and time. The construction of the dynamic evolution matrix enables the system to have the ability to predict the changing trend of customer relationships, providing a scientific basis for enterprises to formulate customer relationship management strategies and significantly improving the intelligence level of the CRM system.

[0024] The specific process for obtaining the customer distribution hot spot area is as follows: The data preprocessing layer performs logarithmic transformation and Z-score standardization on the transaction amount, smooths the transaction frequency using a sliding window, converts the longitude and latitude coordinates to planar coordinates through Gaussian projection, and calculates the transaction density value of each spatial point by combining the transaction channel weight coefficient and the commodity category influence factor to generate a normalized transaction density vector. The spatial density calculation layer uses the kernel density estimation method, selects the Gaussian kernel function for spatial interpolation, adaptively adjusts the bandwidth parameter according to the customer distribution density, and generates continuous customer activity density values in combination with geographical boundary constraint conditions. The dynamic clustering layer uses the DBSCAN density clustering algorithm, sets the density threshold to the mean plus 1.5 times the standard deviation, and the minimum number of samples to 5, identifies high-density spatial regions, introduces a time window mechanism to analyze the time stability of hot spot areas, and generates a set of coordinates of customer distribution hot spot areas with time attributes and geometric boundaries. The hot spot area output layer analyzes the set of coordinates of the identified customer distribution hot spot areas in combination with the dynamic evolution matrix, customer level classification, and credit scoring data to generate customer distribution hot spot areas, including geographical boundaries, intensity levels, time characteristics, and customer composition information.

[0025] In this embodiment, the hot spot recognition mechanism can accurately capture the spatial aggregation pattern of customer activities. The dynamic clustering algorithm ensures the real-time and accuracy of hot spot areas. The spatio-temporal analysis function enables the system to predict the evolution trend of hot spot areas, providing precise spatial decision support for enterprises' market layout, resource allocation, and marketing strategy formulation.

[0026] The spatio-temporal trajectory prediction module includes: The trajectory feature extraction layer processes the historical behavior sequence using a convolutional neural network to extract the spatial features of the customer access pattern, analyzes the temporal dependence relationship of the time stamp sequence through the LSTM network, calculates the moving speed, direction, and stay time in combination with the change of longitude and latitude coordinates, and generates a 128-dimensional trajectory feature vector, including three types of features: spatial preference, time pattern, and movement mode. The customer periodic behavior pattern recognition layer uses the spectral clustering algorithm to analyze the trajectory feature vectors, combines the fast Fourier transform to identify the main frequency components in the activity period identifier, extracts the periodic patterns of seasonal markers through the seasonal decomposition method, and identifies the customer's core activity area (high-frequency access points within a radius of 500 meters), main movement paths (path combinations with a probability greater than 0.3), and time regularity (time interval distribution of repeated behaviors), generating customer periodic behavior pattern parameters including activity period frequency (daily / weekly / monthly level), geographical coordinates and area of the spatial preference area, and behavior stability index (a value between 0 and 1, where 1 represents complete regularity). The spatial activity trend prediction layer constructs a Markov chain model to predict the customer's future location. Combining the current hot spot area distribution and historical trajectory patterns, it uses the Monte Carlo method to generate 1000 simulation predictions, calculates the access probability distribution of each spatial grid, sets a 95% confidence interval, and outputs the spatial activity trend including a probability heat map, confidence interval boundaries, and uncertainty assessment.

[0027] In this embodiment, a complete process from feature extraction to pattern recognition and then to trend prediction is realized. The deep learning method can capture complex non-linear spatio-temporal relationships. The recognition of periodic behavior patterns helps to understand customer behavior rules, and the probabilistic prediction results provide a quantitative decision-making basis for enterprises, significantly improving the accuracy and practicality of customer behavior prediction.

[0028] The three-dimensional rendering process includes: The three-dimensional rendering process first performs coordinate system conversion, converting the spatial association intensity, business association intensity, customer distribution hot spot area, and spatio-temporal trajectory prediction results from geographical coordinates to a unified three-dimensional Cartesian coordinate system. It uses the Mercator projection or equiangular projection to convert the latitude and longitude coordinate data into plane coordinates, combines the altitude data to construct the Z-axis information, and realizes the projection transformation from geographical coordinates to screen coordinates through a perspective transformation matrix, generating accurate three-dimensional scene spatial object position information.

[0029] Based on the spatial object position information and customer level classification, a geometric body system is constructed. Different geometric body types (such as spheres, cubes, cylinders, etc.) are used according to the customer level, and the size parameters of the geometric bodies are adjusted through level weights, generating a set of geometric body parameters including vertex coordinates, face indices, normal vectors, and texture coordinates.

[0030] The material rendering system adjusts the transparency attribute according to the confidence level of the spatio-temporal trajectory prediction results, and assigns different color themes and texture patterns based on the commodity category identifier; the set of geometric body parameters combines ambient light, diffuse light, and specular light, adjusts the light intensity and color temperature according to the seasonal marker, generates three-dimensional rendering image data, and outputs the three-dimensional rendering image data to a three-dimensional visualization interface for display.

[0031] In this embodiment, the three-dimensional rendering process significantly improves the accuracy and expression ability of spatial information display by introducing multi-source data fusion and visualization technology. By uniformly converting the spatial association strength, business association strength, customer distribution hotspots, and spatio-temporal trajectory prediction results into a three-dimensional Cartesian coordinate system, the position deviation problem of the original geographical data in the visualization process is effectively solved, ensuring the true restoration of spatial objects in the three-dimensional scene. At the same time, a customer level classification and geometric body type mapping mechanism is introduced to facilitate users to intuitively identify high-value customer groups. The material rendering system further adjusts the transparency in combination with the confidence level of the spatio-temporal trajectory prediction results, effectively improving the richness of image information and the classification and recognition ability. In addition, by introducing a lighting system and a seasonal adjustment mechanism, the visual immersion and temporal sense of the three-dimensional rendered image are enhanced. Finally, through the three-dimensional visualization interface display, the user's understanding efficiency and decision-making support ability for the multi-dimensional data structure are greatly improved.

[0032] The early warning mechanism includes: monitoring the change rates of the spatial association strength and business association strength with thresholds to identify changes in customer relationships that exceed the normal fluctuation range and generating abnormal event identifiers; combining customer levels, credit score data, and historical abnormal events to evaluate the risk levels of the identified abnormal event identifiers and generating a risk assessment report including the degree of risk, scope of impact, and processing priorities; and displaying real-time early warning information through the three-dimensional visualization interface according to the risk assessment report and user permission settings.

[0033] In this embodiment, by monitoring the real-time change rates of the spatial association strength and business association strength, abnormal fluctuations in the customer relationship network are effectively identified, and abnormal event identifiers are generated in a timely manner, providing technical support for the early detection of potential risks. Combining with the real-time display function of the three-dimensional visualization interface, users can intuitively grasp the distribution and association of abnormal events in space, effectively enhancing the risk perception ability and decision-making response speed, and overall improving the intelligent early warning and visualization management level of the system in complex business scenarios.

[0034] The GIS-based CRM customer ecosystem dynamic visualization management system provided by the present invention integrates multiple advanced technologies such as geographic information technology, graph neural network, deep learning and three-dimensional visualization, and realizes multi-dimensional dynamic analysis of customer relationships, transaction behaviors and spatial distribution. Through the comprehensive collection and modeling of customer geographic location information, transaction data, relationship network and temporal behavior, the system can build a customer ecosystem spatial model with clear structure and high data consistency, accurately reflecting the spatial and business connections between customers. The spatial relationship analysis method of Delaunay triangulation and Voronoi diagram is introduced, and the business association analysis mechanism of graph neural network and collaborative filtering algorithm is combined, so that the correlation strength between space and business can be objectively quantified, and a dynamic evolution matrix is constructed through Markov model in the time dimension to effectively capture the evolution trend of customer relationship network. On this basis, the system further identifies customer distribution hot spots through density estimation and DBSCAN algorithm, and combines seasonality and behavior periodicity analysis to realize periodic identification of customer behavior and prediction of spatial activity trends. The convolutional neural network and LSTM model are used to deeply explore the spatiotemporal trajectory characteristics of customers, which improves the accuracy and practicality of prediction. The 3D visualization rendering module enhances the expressiveness and operability of multi-dimensional data through coordinate conversion, geometric construction and material lighting rendering. Users can intuitively grasp the customer structure and dynamic changes through the visualization interface. The system early warning mechanism monitors abnormal fluctuations in customer relationships with high sensitivity, and generates risk assessment reports based on customer credit ratings. It realizes real-time display and priority processing of risks through a visualization interface, effectively improving the intelligent early warning capability and data-driven strategic decision-making capability of the CRM system, and providing solid technical support for enterprises to achieve efficient customer relationship management and refined operations in complex environments.

[0035] Compared with the traditional CRM system and the existing GIS system, the present invention has significant advantages, as shown in Table 1.

[0036] Table 1 Comprehensive performance index table

[0037] Traditional CRM systems use relational databases to store basic customer information and perform data retrieval through SQL queries. Customer analysis mainly relies on static grouping methods, such as the RFM model (last purchase time, purchase frequency, purchase amount) to stratify customer value. Business processes use linear workflow engines to handle standard processes such as sales funnels and customer service requests. Data presentation uses reporting tools such as Crystal Reports and SSRS to generate static charts and tables. Geographic information processing is limited to text address storage and simple postal code area statistics. It lacks professional spatial analysis functions and cannot perform advanced processing such as spatial relationship calculation and geographic clustering analysis. It also does not support real-time geographic location tracking and spatial behavior pattern recognition.

[0038] Existing GIS systems have the ability to process spatial data and support the storage and analysis of vector data (points, lines, polygons) and raster data. The core functions include spatial queries (buffer analysis, overlay analysis), network analysis (shortest path, service area analysis), geostatistical analysis (interpolation, trend analysis), etc. Geographic data is managed using spatial databases such as PostGIS and Oracle Spatial, and maps are published through standard services such as WMS and WFS. In terms of visualization, functions such as thematic map production and 3D scene display are provided. However, there are obvious limitations in commercial applications: it is difficult to integrate with enterprise business systems, and data format conversion is complex; there is a lack of real-time data stream processing capabilities, and most are offline batch processing analyses; the ability to model customer relationships is weak, and it is unable to effectively integrate non-spatial business attribute data; the interactivity is limited, and it is difficult to support dynamic multi-dimensional analysis requirements.

[0039] Example 2: The present invention can be applied to the customer relationship management scenario of large retail chain enterprises to achieve intelligent management of complex customer ecosystems. In such applications, the CRM customer ecosystem dynamic visualization management system based on GIS can integrate the data of numerous stores distributed in multiple cities of the retail enterprise, online mall transaction records, and membership system information, aiming to deeply explore customer value, optimize store layout and marketing strategies, and significantly improve customer loyalty, referring to Figure 3 .

[0040] Comprehensively obtain multi-dimensional information required to support customer ecosystem analysis. For the retail scenario, geographical location information not only includes the home and work addresses registered by customers for membership and their corresponding latitude and longitude coordinate data, administrative division codes, but also covers the precise locations of each offline store, the geographical boundary range parameters of the service coverage area, and the altitude data of specific regions. Transaction data meticulously records every consumption behavior of customers, including the transaction amount value, transaction frequency, specific transaction timestamp, category identifier of the purchased goods (such as fresh food, daily necessities, or household appliances), the channel where the transaction occurred (such as a specific store, official APP, or mini-program mall), as well as the payment method code and transaction status marker. At the same time, customer relationship data is crucial for understanding the mutual influence among customers, including the recommendation relationship graph among members, the partner association table with banks or local service providers cooperating with the retail enterprise, the interaction information of customers on the brand's official social media, the customer level divided according to consumption behavior, the internally credit score data comprehensively evaluated, and the contact frequency statistics between customers and the enterprise. Finally, time series data captures the dynamics of customer behavior, such as the historical purchase record sequence of customers, the set of timestamps for accessing the APP or browsing goods, the cycle identifier for participating in activities such as "Membership Day", the purchase behavior marker for seasonal goods, and the trend change indicators reflecting the monthly average consumption amount or activity level change of customers.

[0041] Based on the above - collected data, the ecological circle modeling module deeply integrates and analyzes the geographical location information and customer relationship data of retail customers to construct a spatial model of the customer ecological circle, thereby obtaining the spatial association intensity and business association intensity among customers, and generating a dynamic evolution matrix reflecting their dynamic changes. First, the spatial relationship layer analyzes the spatial proximity relationship and aggregation degree between customers and stores, and between customers and customers based on the latitude and longitude coordinate data of customers, store locations, and administrative division codes, using algorithms such as Delaunay triangulation. For example, it identifies customer groups living in the same community and frequently visiting the same branch store, thus quantitatively generating the spatial association intensity. Secondly, the business association layer processes the membership recommendation relationship graph through graph theory analysis (such as graph neural networks), identifies core recommenders and community structures, and at the same time analyzes the co - purchase behavior patterns of customers, and adjusts the weights in combination with data such as customer levels and credit scores to generate the business association intensity reflecting the closeness of business connections between customers. For example, two customers who are both high - level members, have a recommendation relationship, and have similar consumption preferences will have a relatively high business association intensity. A remarkable feature of the present invention lies in its spatio - temporal evolution layer. It does not view spatial or business associations in isolation, but based on historical time - series data, conducts trend analysis and periodic detection on the historical changes of these two association intensities. For example, it constructs a Markov transition matrix to describe the state transition probability of the customer ecological circle, thereby forming a dynamic evolution matrix that can reflect the dynamic evolution law of the customer relationship network. This analysis method that deeply couples spatial distribution, business connection, and temporal dynamics enables retail enterprises to gain insights into the structural changes, influence transfer, and life - cycle evolution of the customer ecological circle (such as customer groups in a specific business district or a certain membership community), going beyond traditional static snapshot - based customer analysis, and providing a dynamic and in - depth perspective for understanding the customer ecosystem.

[0042] Subsequently, the spatial clustering module processes the real - time transaction data to identify the hot spots of customer distribution. This process starts from the data pre - processing layer, where data such as transaction amounts are standardized, geographical coordinates are uniformly projected, and transaction density is calculated by combining transaction channels and commodity categories to generate a normalized transaction density vector. The spatial density calculation layer uses methods such as kernel density estimation for spatial interpolation to generate a continuous density map of customer transaction activity or specific commodity demand within the region. The dynamic clustering layer uses algorithms such as DBSCAN to cluster the customer activity density values, identify regions where the transaction density far exceeds the surrounding areas, such as the concentrated purchase points of residents in newly developed real estate projects or consumption hot spots during the lunch break in office areas, and analyze the temporal change laws of these hot spots in combination with the transaction timestamp, such as formation, development, and recession cycles. Finally, the hot spot area output layer provides detailed information about these areas, including geographical boundaries, intensity levels, time characteristics (such as only active on weekends), and customer composition (such as main customer levels, frequently purchased commodity categories), and can predict their future development trends in combination with the aforementioned dynamic evolution matrix.

[0043] Based on the customer's time series data and geographical location information, the spatio-temporal trajectory prediction module constructs a customer spatio-temporal trajectory model, deeply analyzes the customer's spatio-temporal behavior patterns, and obtains prediction results including the customer's periodic behavior patterns and spatial activity trends. The trajectory feature extraction layer first processes the customer's historical behavior records (such as APP check-ins, store sign-ins) and location data, extracts features such as location changes, moving speeds, and staying durations at different times, and generates multi-dimensional trajectory feature vectors. The customer periodic behavior pattern recognition layer then identifies the customer's daily activity areas (such as frequently visited stores), habitual mobile path preferences, and temporal regularity of behaviors based on these feature vectors and activity period identifiers (such as promotion periods). Particularly important is its spatial activity trend prediction layer, which combines the dynamic changes in the identified customer distribution hotspots and the customer historical trajectory patterns to predict the customer's possible future spatial activity trends. For example, it predicts the stores that customers in a certain area are most likely to visit during the next large-scale promotion, and gives the corresponding probability distribution and confidence interval. This mechanism that combines real-time transaction-driven hotspot discovery with forward-looking trajectory guidance based on historical behaviors enables retail enterprises to shift from passive response to market changes to active prediction and layout. For example, it can not only mark in real time which store has a sharp increase in customer flow due to promotions, but also predict the areas that a specific customer group is most likely to flow to at a certain future time point, thus supporting the enterprise to optimize resource allocation in advance or push more precise marketing information.

[0044] All these analysis results are finally integrated and presented by the multi-dimensional dynamic visualization module. This module is responsible for three-dimensional rendering of the calculated spatial association intensity (for example, represented by the thickness of connection lines in a three-dimensional view), business association intensity (for example, distinguished by the size or color of nodes), customer distribution hotspots (for example, shown as highlighted areas or protruding heat maps on a three-dimensional map), and spatio-temporal trajectory prediction results (for example, showing customer flows with dynamic arrows or streamlines), so as to obtain an interactive three-dimensional visualization interface. In a retail scenario, each store can be represented as a three-dimensional building model, customers can be displayed as geometric bodies of different forms according to their value or level, and the complex associations between customers are clearly shown through dynamically changing connection lines. Managers can freely zoom in, rotate, and roam the three-dimensional scene through multi-touch and other means, and can adjust the time slice in real time to observe dynamic changes (such as comparing the migration of customer hotspots last month and this month), or change the spatial range boundary to focus on a specific city or business district, as well as adjust the analysis dimension level.

[0045] The three-dimensional rendering process first performs coordinate system conversion, unifying various analysis results from the geographic coordinate system to the three-dimensional Cartesian coordinate system. It constructs the Z-axis information by using an appropriate projection method and combining elevation data, and achieves accurate projection to the screen coordinates through the perspective transformation matrix. Subsequently, based on the classification of spatial object position information and customer levels, etc., a set of geometric body parameters including vertex coordinates, patch indices, normal vectors, and texture coordinates is generated. For example, customers of different levels can be represented by geometric bodies of different sizes or shapes. The material rendering system adjusts the transparency attribute of the object according to the confidence level of the spatio-temporal trajectory prediction result, or assigns a unique color theme and texture pattern to it based on the category identifier of the goods frequently purchased by the customer. Combining multiple lighting effects such as ambient light, diffuse light, and specular light, and the lighting intensity and color temperature can be adjusted according to seasonal markers (such as bright colors for summer promotions and warm colors for winter sales). Finally, three-dimensional rendering image data with a sense of reality and information-carrying ability is generated and output to the three-dimensional visualization interface for users to make decisions.

[0046] The system continuously monitors the change rates of the spatial association intensity (for example, monitoring whether the geographical aggregation degree of a certain core customer group begins to loosen) and the business association intensity (for example, paying attention to whether the consumption frequency or amount of large customers shows a significant decline) against thresholds. Once the change exceeds the preset normal fluctuation range, an abnormal event identifier will be automatically generated. Subsequently, the system combines the customer level, credit score data, and the historical abnormal event records of this customer or region to evaluate the risk levels of these identified abnormal event identifiers, and outputs a risk assessment report including the risk degree, possible impact scope, and recommended handling priorities. For example, it prompts that there is a risk of loss of a certain high-value customer group or a risk of shrinkage in a certain regional market. These early warning messages will be prominently displayed in real time through the three-dimensional visualization interface according to the user permission settings. For example, the customer groups or geographical areas with abnormalities on the 3D map will be highlighted in a specific color (such as red), accompanied by a concise risk description and an interaction entry, allowing managers to immediately view the details. This way of combining the complex customer ecosystem analysis results, intuitive immersive three-dimensional visualization interaction, and real-time risk closed-loop early warning management greatly improves the retail enterprise's sensitivity to customer relationship changes and potential market risks and the decision-making response efficiency, forming a complete management closed-loop from data insight to risk identification, then to visualization early warning and decision-making assistance.

[0047] Through the collaborative work of the above-mentioned modules, in the specific application of retail chain enterprises, the present invention effectively overcomes many limitations existing in the prior art. Traditional customer relationship management systems can often only perform simple address storage and map point marking, lacking professional spatial analysis capabilities, and it is difficult to achieve advanced geographic information processing such as spatial relationship analysis and spatial clustering. In the commercial application of existing geographic information systems, although they can provide basic map services, they often face obstacles such as inconsistent data formats, poor real-time performance, and single analysis dimensions when integrating with CRM systems, resulting in the inability to effectively integrate spatial information and customer relationship data. In addition, in terms of customer relationship modeling, previous technologies mostly adopted single-customer attribute models and static linear relationship chains, lacking in-depth analysis of the overall relationship of the customer ecosystem, spatial influencing factors, dynamic change capture, and group effects, unable to construct a complete customer ecological network model, and the visualization display also mostly relied on static charts, lacking real-time dynamic update and multi-dimensional interaction capabilities. The CRM customer ecosystem dynamic visualization management system based on GIS proposed by the present invention realizes in-depth fusion analysis of geographical location information and customer relationship data by constructing a customer ecosystem spatial model; by constructing a customer spatio-temporal trajectory model, integrating time series data and spatial location information, it can accurately analyze the spatio-temporal behavior patterns of customers and make predictions; and by constructing a multi-dimensional dynamic visualization module, a three-dimensional stereoscopic rendering mechanism, and an early warning function, it significantly improves the information transmission efficiency of customer management, the intuitiveness of user operations, and the risk response ability.

[0048] In summary, in the application of the retail scenario, by constructing a customer ecosystem spatial model, the present invention enables retail enterprises to deeply integrate the geographical locations of customers with their transaction behaviors, social relationships, etc., quantify the spatial and business association intensities between customers, thereby accurately identifying core customer communities and key influencing nodes, providing a solid scientific basis for refined customer stratification and regional differentiated marketing, and significantly improving the accuracy and depth of customer relationship modeling. By analyzing the periodic behavior patterns of customers and predicting future spatial activity trends with the customer spatio-temporal trajectory model, retail enterprises can carry out more accurate personalized information push, optimize store location, commodity layout, and promotion resource allocation, effectively improving the marketing response rate and operational efficiency. At the same time, the multi-dimensional dynamic three-dimensional visualization interface integrated with a real-time early warning mechanism converts complex customer ecological data into intuitive and easy-to-understand interactive views, enabling managers to quickly grasp the overall market situation and customer dynamics, actively monitor and respond to potential customer churn or market risks, thereby enhancing the foresight of decision-making and the initiative of risk prevention and control.

[0049] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A GIS-based dynamic visualization management system for CRM customer ecosystem, characterized in that, It includes: A data acquisition module, which is used to obtain the geographical location information, transaction data, customer relationship data, and time series data of customers; An ecological circle modeling module, which is used to construct a customer ecological circle space model to analyze the geographical location information and customer relationship data, obtain the spatial association intensity and business association intensity between customers, and generate a dynamic evolution matrix; A spatial clustering module, which is used to perform real-time spatial clustering analysis on transaction data to identify hot spots of customer distribution; A spatio-temporal trajectory prediction module, which is used to construct a customer spatio-temporal trajectory model based on the time series data and geographical location information of customers, analyze the spatio-temporal behavior patterns of customers, and obtain spatio-temporal trajectory prediction results including customer periodic behavior patterns and spatial activity trends; A multi-dimensional dynamic visualization module, which is used to perform three-dimensional rendering on the dynamic evolution matrix, hot spots of customer distribution, and spatio-temporal trajectory prediction results to obtain a three-dimensional visualization interface, and support multi-touch interaction, an early warning mechanism, and real-time adjustment of time slices, spatial range boundaries, and analysis dimension levels.

2. The dynamic visualization management system of a CRM customer ecosystem based on GIS according to claim 1, characterized in that: The geographical location information includes the longitude and latitude coordinate data, administrative division code, detailed address information, altitude data, and geographical boundary range parameters of customers; the transaction data includes transaction amount values, transaction frequencies, transaction timestamps, commodity category identifiers, transaction channel information, payment method codes, and transaction status marks; the customer relationship data includes a recommendation relationship graph, a partner association table, social network connection information, customer levels, credit score data, and contact frequency statistics; the time series data includes a sequence of customer historical behavior records, a set of system access timestamps, activity period identifiers, seasonal marks, and trend change indicators.

3. The dynamic visualization management system of a CRM customer ecosystem based on GIS according to claim 2, characterized in that: The customer ecological circle space model includes: A spatial relationship layer, which obtains the geometric topological relationship of customer spatial distribution based on the longitude and latitude coordinate data, administrative division code, and altitude data, and generates the spatial association intensity; A business association layer, which performs graph theory analysis on the recommendation relationship graph, partner association table, and social network connection information, and adjusts them in combination with customer levels and credit score data to obtain the business association intensity between customers; A spatio-temporal evolution layer, which performs temporal analysis on the spatial association intensity and business association intensity based on historical time series data, and constructs a dynamic evolution matrix of the customer ecological circle.

4. A GIS-based CRM customer ecosystem dynamic visualization management system according to claim 3, characterized in that: The specific process for obtaining the hot spots of customer distribution is as follows: A data preprocessing layer, which standardizes the transaction amount values, transaction frequencies, and longitude and latitude coordinate data, combines the transaction channel information and commodity category identifiers to obtain the transaction density, and generates a normalized transaction density vector; A spatial density calculation layer, which performs spatial interpolation calculation on the transaction density vector, and combines the geographical boundary range parameters to obtain the customer activity density value within the region; A dynamic clustering layer, which performs spatial clustering analysis on the customer activity density value, identifies the spatial regions exceeding the density threshold, and performs temporal analysis in combination with the transaction timestamp to generate a set of coordinates of hot spots of customer distribution with time attributes and geometric boundaries; Hotspot area output layer, which analyzes the set of coordinates of the identified customer distribution hotspot areas by combining the dynamic evolution matrix, customer level classification, and credit scoring data, and generates customer distribution hotspot areas, including geographical boundaries, intensity levels, time characteristics, and customer composition information.

5. The dynamic visualization management system of a CRM customer ecosystem based on GIS according to claim 2, characterized in that: The spatio-temporal trajectory prediction module includes: Trajectory feature extraction layer, which extracts features from the historical behavior record sequence, system access timestamp set, and longitude and latitude coordinate data, obtains the position change pattern of the customer in different time periods, and generates a trajectory feature vector; Customer periodic behavior pattern recognition layer, which identifies the spatial movement behavior of the customer based on the trajectory feature vector, activity cycle identifier, and seasonal marker, obtains the daily activity area, movement path preference, and time regularity of the customer, and generates customer periodic behavior pattern parameters, including activity cycle frequency, spatial preference area, and behavior stability index; Spatial activity trend prediction layer, which combines the customer distribution hotspot area to predict the future spatial activity trend of the customer, and generates a spatial activity trend including probability distribution and confidence interval.

6. A GIS-based CRM customer ecosystem dynamic visualization management system according to claim 1, characterized in that: The specific process of the three-dimensional rendering includes: Converting the spatial association intensity, business association intensity, customer distribution hotspot area, and spatio-temporal trajectory prediction results into a three-dimensional coordinate system, performing a projection transformation from geographical coordinates to screen coordinates using longitude and latitude coordinate data and altitude data, and generating the position information of spatial objects in the three-dimensional scene; generating a set of geometry parameters based on the position information of spatial objects and customer level classification; assigning material attributes and texture coordinates to the set of geometry parameters according to spatio-temporal trajectory prediction and product category identifier; the set of geometry parameters combines ambient light, diffuse light, and specular light, adjusts the light intensity and color temperature according to the seasonal marker, and generates three-dimensional rendering image data, and the three-dimensional rendering image data is output to the three-dimensional visualization interface for display.

7. A GIS-based CRM customer ecosystem dynamic visualization management system according to claim 1, characterized in that: The early warning mechanism includes: monitoring the change rate of the spatial association intensity and business association intensity for thresholds, identifying changes in customer relationships that exceed the normal fluctuation range, and generating an abnormal event identifier; combining customer level, credit scoring data, and historical abnormal events to evaluate the risk level of the identified abnormal event identifier, and generating a risk assessment report including risk degree, impact range, and processing priority; according to the risk assessment report and user permission settings, real-time warning information is displayed through the three-dimensional visualization interface.

Images