Intelligent shopping guide recommendation method and system based on big data

Abstract

The invention provides an intelligent shopping guide recommendation method and system based on big data, and relates to the technical field of intelligent shopping guide. The method comprises the following steps: constructing a commodity network topological structure diagram; performing buying power fluctuation band quantification, decision-making tolerance attenuation curve modeling and shopping path tolerance analysis on the target customer, and establishing a three-dimensional value model of the target customer; generating a dynamic optimized commodity permeation pool by adopting a commodity association mining strategy; generating an optimal shopping guide route sequence by using a space-time fusion field construction technology and a global dynamic path optimization algorithm; and according to the shopping guide route deviation and the shopping cart commodity type deviation, generating a dynamically adjusted shopping guide instruction. According to the invention, through multi-level technology cooperation, the person and goods yard matching efficiency of a retail scene is significantly improved, and efficient and pleasant immersive shopping experience is realized; an intelligent shopping guide mode with precise recommendation ability, space-time resource optimization efficiency and humanistic care attributes is constructed, and core infrastructure support is provided for entity retail digital transformation.

Description

Technical Field

[0001] The present invention relates to the field of intelligent shopping guide technology, and in particular to an intelligent shopping guide recommendation method based on big data and an intelligent shopping guide recommendation system based on big data. Background Art

[0002] As the digital transformation of the retail industry accelerates, traditional shopping malls face a significant challenge from online e-commerce, and consumers are increasingly demanding more efficient and personalized shopping experiences. Traditional shopping guide models, which rely on manual experience, suffer from issues such as low recommendation accuracy, homogeneous services, and illogical route planning, making them unable to meet the diverse needs of modern consumers. For example, consumers in large shopping malls often prolong their shopping time due to difficulty finding their desired products or receiving recommendations that don't match their needs. This leads to a poor shopping experience and impacts mall traffic and conversion rates.

[0003] To address these issues, intelligent shopping guide technologies based on big data and artificial intelligence have become a research hotspot. Existing solutions, including some that analyze historical user consumption data to make product recommendations, often rely on a single-dimensional analysis of purchasing preferences, ignoring various dynamic factors. This results in recommendations that lack timeliness and adaptability. Furthermore, they struggle to process the high-dimensional, large-scale product data found in shopping malls, resulting in inadequate exploration of product relationships and weakly correlated recommendation lists. Furthermore, shopping guide route planning often only considers physical distance, which can lead to redundant routes or deviations from actual user needs, reducing shopping guide efficiency.

[0004] In this context, there is an urgent need for an intelligent shopping guide recommendation method to achieve accurate recommendations and efficient shopping guides, improve consumers' shopping experience, and enhance the core competitiveness of shopping malls. Summary of the Invention

[0005] The present invention provides a big data-based intelligent shopping guide recommendation method and system to solve the defects in the prior art.

[0006] In one aspect, the present invention provides an intelligent shopping guide recommendation method based on big data, comprising: Collect historical shopping records and product category information from the mall, use a high-dimensional correlation calculation model, and construct a product network topology diagram.

[0007] Obtain the target customer's historical consumption records and current shopping demand list, quantify the target customer's purchasing power fluctuation band, model the decision patience attenuation curve, and analyze the shopping path tolerance to establish a three-dimensional value model for the target customer.

[0008] Based on the commodity network topology diagram and the current shopping demand list, a commodity association mining strategy is used to generate a dynamic optimal commodity penetration pool.

[0009] Obtain the spatial coordinates of each commodity in the commodity penetration pool and the three-dimensional map of the shopping mall, combine it with the three-dimensional value model, use the space-time fusion field construction technology and the global dynamic path optimization algorithm to generate the optimal shopping guide route sequence.

[0010] Acquire user positioning signals and dynamic changes in shopping cart items in real time, and generate dynamically adjusted shopping guide instructions based on shopping guide route deviations and shopping cart item type deviations.

[0011] According to the big data-based intelligent shopping guide recommendation method provided by the present invention, the process of constructing a product network topology diagram includes: Clean the mall's historical shopping records and extract product combination information from the complete transaction flow.

[0012] Define a product association tensor, calculate the first-order association strength between any two products, and expand on the multi-order association relationship. The first-order association strength represents the geometric mean of the conditional probability and the lift index. Multi-order association relationships represent high-level collaborative purchasing patterns within a product portfolio.

[0013] Multi-order association relationships are stored in the form of three-dimensional tensors, and topological edge weights are generated through tensor decomposition.

[0014] The connection edges whose topological edge weights are lower than the preset association strength threshold are removed to obtain a commodity network topology graph, in which nodes represent commodities and edges represent the association relationship between multiple commodity combinations.

[0015] According to the present invention, a smart shopping guide recommendation method based on big data is provided. Historical consumption records include basic transaction data, product details, behavior trajectory data, and return and exchange records. The current shopping wish list represents a list of planned purchases actively entered by the target customer.

[0016] According to the big data-based intelligent shopping guide recommendation method provided by the present invention, the process of quantifying the purchasing power fluctuation band of target customers includes: Based on the target customers' historical consumption records, calculate the arithmetic mean and standard deviation of the target customers' single shopping consumption amounts.

[0017] Identify the time-periodic characteristics of target customers' consumption behavior and construct a fluctuation function combined with a time decay factor.

[0018] With the volatility function value as the center line, a dynamic volatility band interval is generated according to the standard deviation multiple.

[0019] By real-time monitoring of the cumulative amount of consumption in the target customer's shopping demand list, the fluctuation range is dynamically adjusted.

[0020] According to the big data-based intelligent shopping guide recommendation method provided by the present invention, the process of modeling the decision patience decay curve of target customers includes: Analyze the historical behavior records of target customers, calculate the average time it takes for target customers to browse products and make purchase decisions, and map the average time to the initial patience value.

[0021] A hyperbolic attenuation function model is established to make the initial patience value show nonlinear attenuation characteristics as the decision time goes by.

[0022] The attenuation rate coefficient is dynamically adjusted based on current environmental parameters, including mall crowd density and promotion intensity.

[0023] When the deviation between the actual decision time of the target customer and the predicted value exceeds the preset deviation threshold, the attenuation function parameters are modified.

[0024] According to the big data-based intelligent shopping guide recommendation method provided by the present invention, the process of analyzing the shopping path tolerance of target customers includes: Extract the abandonment records of target customers in their shopping history and fit the benchmark threshold of target customers' tolerance.

[0025] A spatial distance tolerance conversion model is established based on the benchmark threshold, and the tolerance energy function is calculated.

[0026] It takes a long time to integrate this shopping in real time and dynamically update the path tolerance threshold.

[0027] According to the present invention, a smart shopping guide recommendation method based on big data is provided, wherein the process of generating a dynamic preferred product penetration pool by adopting a product association mining strategy includes: In the product network topology diagram, starting from the target customer's current shopping demand list, the penetration rate of all candidate products that can be reached through direct association paths is calculated.

[0028] Conduct multi-demand collaborative correction to detect the second-order impact of the product combination in the target customer's current shopping demand list on the candidate products.

[0029] Monitor the decay rate of the commodity association strength in the pool in real time, and screen the initial commodity pool based on the dynamically preset penetration threshold.

[0030] According to the big data-based intelligent shopping guide recommendation method provided by the present invention, the process of generating the optimal shopping guide route sequence includes: The spatial coordinates and value weight of each commodity in the commodity penetration pool are mapped into a four-dimensional space-time field.

[0031] The user's three-dimensional value model is superimposed to form a comprehensive constraint field to establish a space-time optimization equation and obtain a continuous path curve. The comprehensive constraint field includes the value constraint defined by the current fluctuation band interval, the maximum stay time determined by the decision patience value, and the spatial movement restriction converted from the path tolerance.

[0032] The continuous path curve is discretized into a shopping guide node sequence, which is sorted by spatial accessibility and output as a shopping guide route instruction set to obtain the optimal shopping guide route sequence.

[0033] According to the big data-based intelligent shopping guide recommendation method provided by the present invention, the process of generating dynamically adjusted shopping guide instructions includes: The target customer's location coordinates and shopping cart item list are obtained in real time, and a product decision entropy monitoring model is established based on the offset between the real-time location coordinates and shopping cart item list and the optimal shopping guide route sequence and product penetration pool.

[0034] When any deviation exceeds a preset threshold, dynamic replanning conditions are triggered. The dynamic replanning conditions include recalculating the gradient direction of the space-time field based on the latest data and generating a set of correction instructions including steering instructions, speed adjustments, and product replacement suggestions.

[0035] On the other hand, the present invention also provides an intelligent shopping guide recommendation system based on big data, comprising: The product association topology construction module is used to collect historical shopping records and product category information of the mall, and use a high-dimensional association calculation model to construct a product network topology diagram.

[0036] The user three-dimensional value modeling module is used to obtain the target customer's historical consumption records and current shopping demand list, quantify the target customer's purchasing power fluctuation band, model the decision-making patience attenuation curve, and analyze the shopping path tolerance to establish a three-dimensional value model for the target customer.

[0037] The dynamic commodity penetration pool generation module is used to generate a dynamic optimal commodity penetration pool based on the commodity network topology diagram and the current shopping demand list using a commodity association mining strategy.

[0038] The space-time path optimization module is used to obtain the spatial coordinates of each commodity in the commodity penetration pool and the three-dimensional map of the shopping mall. Combined with the three-dimensional value model, it uses the space-time fusion field construction technology and the global dynamic path optimization algorithm to generate the optimal shopping guide route sequence.

[0039] The dynamic instruction adjustment module is used to obtain user positioning signals and dynamic change data of shopping cart items in real time, and generate dynamically adjusted shopping guide instructions based on the deviation of shopping guide routes and the deviation of shopping cart item types.

[0040] The present invention provides a big data-based intelligent shopping guide recommendation method and system. This method overcomes the limitations of traditional two-by-two recommendations through high-level scenario-based association mining, enabling accurate cross-category demand prediction. It also integrates purchasing power fluctuations, decision patience decay, and path tolerance into a three-dimensional value assessment system. This system has the adaptive ability to perceive users' real-time spending power and behavioral tendencies. Using spatiotemporal fusion field technology, it unifies product value, physical movement paths, and time constraints into a continuous optimization problem, generating personalized shopping guide routes that maximize user value. Finally, it leverages dynamic entropy change monitoring and a dual-deviation response mechanism to achieve closed-loop instruction tuning, thoroughly resolving the pain point of recommendation failure caused by user interest drift or path deviation. Through multi-layered technology collaboration, this method significantly improves the efficiency of matching people, goods, and places in retail scenarios. On the user side, it reduces the cognitive burden and ineffective walking losses of selecting a large number of products, achieving an efficient and enjoyable immersive shopping experience. On the merchant side, it boosts the penetration rate of long-tail products and boosts sales conversion, while optimizing inventory turnover through data-driven dynamic pricing. Ultimately, this intelligent shopping guide model combines precise recommendation capabilities, spatiotemporal resource optimization efficiency, and humanistic care, providing core infrastructure support for the digital transformation of physical retail. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 This is a flow chart of an intelligent shopping guide recommendation method based on big data provided by an embodiment of the present invention; Figure 2 This is a structural diagram of an intelligent shopping guide recommendation system based on big data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0044] The following combination Figure 1-Figure 2 The present invention describes an intelligent shopping guide recommendation method and system based on big data.

[0045] Figure 1This is a flow chart of an intelligent shopping guide recommendation method based on big data provided by an embodiment of the present invention.

[0046] like Figure 1 As shown, an embodiment of the present invention provides an intelligent shopping guide recommendation method and system based on big data. The execution subject may be an intelligent shopping guide recommendation method based on big data, and the method includes: Collect historical shopping records and product category information from the mall, use a high-dimensional correlation calculation model, and construct a product network topology diagram. The process includes: Perform data cleaning on the mall's historical shopping records, extract the product combination information in the complete transaction flow, and construct a transaction matrix grouped by transaction day. The formula is expressed as:

[0047] Where N represents the number of valid transaction orders, and M represents the total number of products after deduplication.

[0048] Define the product association tensor ,in Indicates the maximum correlation order.

[0049] Calculate the first-order correlation strength between any two items and expand the multi-order correlation relationship. The first-order correlation strength represents the geometric mean of the conditional probability and the lift index. The formula is expressed as:

[0050]

[0051]

[0052] Multi-order associations represent high-order collaborative purchasing patterns in a product portfolio, and the formula is:

[0053]

[0054]

[0055] Where, Represents a multi-order association relationship. represents the conditional probability, Indicates the degree of improvement, represents the weight adjustment factor, represents the smoothing constant, m represents the correlation order, represents the combined product set, Q represents the set of all products in the mall, Indicates that the product is included The transaction set, Indicates that the product is included The transaction set, Indicates product combination support, Represents the correlation strength amplification factor.

[0056] The multi-order association relationship is stored in the form of a three-dimensional tensor, and the topological edge weight is generated by tensor decomposition. The formula is expressed as:

[0057] Where, represents the product of tensors along the third dimension, represents the eigenvector of the mth-order association, represents the Frobenius norm.

[0058] Remove the connection edges whose topological edge weight is lower than the preset association strength threshold to obtain the commodity network topology graph , V represents the set of commodity nodes, and E represents the edge set.

[0059] Obtain the target customer's historical consumption records and current shopping demand list, quantify the target customer's purchasing power fluctuation band, model the decision patience attenuation curve, and analyze the shopping path tolerance to establish a three-dimensional value model for the target customer.

[0060] Historical consumption records include basic transaction data, product details, behavioral trajectory data, and return and exchange records. Basic transaction data includes transaction ID, timestamp, payment amount, and payment method. Product details include inventory information, category, unit price, quantity, and discount rate. Behavioral trajectory data includes shelf time, try-on and trial records, and abandoned shopping cart records. Return and exchange records include returned product inventory information, reasons, and refund amounts. The current shopping wish list represents the list of items the target customer proactively enters for purchase.

[0061] The process of quantifying the purchasing power fluctuation band of target customers includes: Based on the target customers' historical consumption records, calculate the arithmetic mean and standard deviation of the target customers' single shopping consumption amounts.

[0062] Identify the time-periodic characteristics of target customers' consumption behavior and construct a fluctuation function combined with the time decay factor. The formula is expressed as follows:

[0063] It represents the purchasing power benchmark, and the formula is:

[0064] It represents the fluctuation amplitude, and the formula is:

[0065] Where, represents the quantile function, Indicates the single transaction amount. represents the valid transaction set of the target customer, t represents the current time, Indicates the latest consumption timestamp, represents the attenuation strength, represents the first consumption time of the target customer, f represents the consumption frequency, B represents the total number of consumptions, represents the i-th consumption amount, represents the mean of all consumption.

[0066] Taking the volatility function value as the center line, the dynamic volatility band interval is generated according to the standard deviation multiple. The formula is expressed as:

[0067] Where s represents the confidence coefficient.

[0068] By real-time monitoring of the cumulative amount of consumption in the target customer's shopping demand list, the fluctuation range is dynamically adjusted.

[0069] The process of modeling the decision patience decay curve for target customers includes: Analyze the historical behavior records of target customers, calculate the average time it takes for target customers to browse products and make purchase decisions, and map the average time to the initial patience value.

[0070] A hyperbolic attenuation function model is established to make the initial patience value show nonlinear attenuation characteristics as the decision time passes. The formula of the hyperbolic attenuation function model is expressed as:

[0071] Where, Indicates the initial patience value, represents the attenuation rate coefficient, t represents the current time, Indicates the starting time of entering the decision-making state, Represents the nonlinear decay exponent.

[0072] The attenuation rate coefficient is dynamically adjusted based on current environmental parameters, including mall crowd density and promotion intensity.

[0073] When the deviation between the target customer's actual decision time and the predicted value exceeds the preset deviation threshold, the attenuation function parameters are modified. The formula is expressed as follows:

[0074] Where, represents the learning rate, Indicates the actual decision-making time, Indicates the prediction decision time.

[0075] The process of analyzing the shopping path tolerance of target customers includes: Extract the abandonment records of target customers in their shopping history and fit the benchmark threshold of target customers' tolerance.

[0076] A spatial distance tolerance conversion model is established based on the benchmark threshold, and the tolerance energy function is calculated. The formula is expressed as follows:

[0077] Where, represents the tolerance energy integral of path Γ, represents a parameterized path, represents the distance field function, q represents the tolerance coefficient, Represents the distance field gradient norm.

[0078] It takes a long time to integrate this shopping in real time and dynamically update the path tolerance threshold.

[0079] Based on the product network topology diagram and the current shopping demand list, a product association mining strategy is used to generate a dynamic optimal product penetration pool. The process includes: In the product network topology, starting from the target customer's current shopping demand list, the penetration rate of all candidate products that can be reached through direct association paths is calculated. The formula is expressed as:

[0080] Where, represents the penetration rate of candidate product C, C represents the candidate product identifier, Indicates the direct association strength, that is, the product The geometric mean of the probability of association with C and the lift, Represents the gradient logarithm of the associated potential energy field, characterizing the commodity The direction and intensity of the impact on other commodities.

[0081] Conduct multi-demand collaborative correction to detect the second-order impact of the product combination in the target customer's current shopping demand list on the candidate products.

[0082] Monitor the decay rate of the correlation strength of commodities in the pool in real time, and screen the initial commodity pool based on the dynamically preset penetration threshold. The formula is expressed as:

[0083] Where P represents the dynamic optimization commodity penetration pool, represents the permeability threshold, represents the permeability decay rate, Indicates the lower limit of the attenuation rate tolerance, Indicates the emergency inclusion conditions, represents the mixed partial derivative of the correlation strength, represents the synergy gain threshold.

[0084] Obtain the spatial coordinates of each product in the product penetration pool and the 3D map of the mall. Combined with the 3D value model, the optimal shopping guide route sequence is generated using the space-time fusion field construction technology and the global dynamic path optimization algorithm. The process includes: The spatial coordinates and value weight of each commodity in the commodity penetration pool are mapped into a four-dimensional space-time field.

[0085] The user's three-dimensional value model is superimposed to form a comprehensive constraint field to establish a spatiotemporal optimization equation, and a continuous path curve is obtained. The comprehensive constraint field includes the value constraint defined by the current fluctuation band interval, the maximum stay time determined by the decision patience value, and the spatial movement restriction converted from the path tolerance. The spatiotemporal optimization equation formula is expressed as:

[0086] Where, represents the total space-time loss of path Γ, represents the current parameterized path, t represents the current time, represents the lower limit of integration, that is, the moment when path planning begins, Indicates the upper limit of points, that is, the time when you plan to complete shopping. represents the spatial friction coefficient, that is, the walking loss cost per unit distance, represents the space-time field gradient, represents the value conversion rate, represents the real-time purchase probability, Represents dynamic purchasing power.

[0087] The continuous path curve that minimizes the space-time friction loss is solved by the variational method. The space-time friction loss includes path distance loss and value decay loss.

[0088] The continuous path curve is discretized into a shopping guide node sequence, and the sequence is sorted by spatial accessibility and output as a shopping guide route instruction set to obtain the optimal shopping guide route sequence, which is expressed as follows:

[0089] Where, represents the optimal shopping guide route sequence, K represents the number of path segments, represents the path node coordinates, i.e. the shelf location of the kth recommended product, represents the time point, i.e. the planned time to arrive at the kth node, is the Lagrangian density, which represents the path loss function per unit time.

[0090] Real-time acquisition of user location signals and dynamic changes in shopping cart items. Based on deviations in shopping guide routes and shopping cart item types, dynamically adjusted shopping guide instructions are generated. The process includes: The target customer's location coordinates and shopping cart item list are obtained in real time. A product decision entropy monitoring model is established based on the offset between the real-time location coordinates and shopping cart item list and the optimal shopping guide route sequence and the product penetration pool. The formula is expressed as follows:

[0091] In the formula, C represents the candidate product identifier, Represents a collection of unpurchased items. represents the purchase probability of candidate products, represents the path deviation coupling coefficient, represents the target customer's moving speed vector, represents the planned path velocity vector.

[0092] When any deviation exceeds a preset threshold, the dynamic replanning condition is triggered. The dynamic replanning condition includes recalculating the spatiotemporal field gradient direction based on the latest data and generating a correction instruction set including steering instructions, speed adjustment, and product replacement suggestions. The dynamic replanning condition formula is expressed as:

[0093] Where, Indicates the monitoring time window, represents the real-time decision entropy, represents the historical decision entropy, represents the entropy change threshold, Represents a logical operator, Indicates the real-time location of the target customer. represents the planned path position, Indicates the path offset distance.

[0094] In summary, this embodiment provides a big data-based intelligent shopping guide recommendation method. Through high-level scenario-based association mining, it breaks through the limitations of traditional two-by-two recommendations and achieves accurate cross-category demand prediction. It also integrates purchasing power fluctuations, decision patience decay, and path tolerance into a three-dimensional value assessment system. It has the adaptive ability to perceive users' real-time consumption capacity and behavioral tendencies. Based on spatiotemporal fusion field technology, it unifies the modeling of product value, physical movement routes, and time constraints as a continuous optimization problem, generating personalized shopping guide routes that maximize user value. Finally, it achieves closed-loop instruction tuning through dynamic entropy change monitoring and a dual-deviation response mechanism, completely resolving the pain point of recommendation failure caused by user interest drift or path deviation. Through the synergy of multiple layers of technology, it significantly improves the efficiency of matching people, goods, and places in retail scenarios. On the user side, it reduces the cognitive burden of massive product selection and the loss of ineffective walking, achieving an efficient and enjoyable immersive shopping experience. On the merchant side, it activates the penetration rate of long-tail products and the conversion of related sales, and optimizes inventory turnover through data-driven dynamic pricing. Ultimately, it constructs an intelligent shopping guide model that combines precise recommendation capabilities, spatiotemporal resource optimization efficiency, and humanistic care, providing core infrastructure support for the digital transformation of physical retail.

[0095] Based on the same general inventive concept, the present invention also protects an intelligent shopping guide recommendation system based on big data. The intelligent shopping guide recommendation system based on big data provided by the present invention is described below. The intelligent shopping guide recommendation system based on big data described below and the intelligent shopping guide recommendation method based on big data described above can be referenced to each other.

[0096] Figure 2 This is a structural diagram of an intelligent shopping guide recommendation system based on big data provided by an embodiment of the present invention.

[0097] like Figure 2 As shown, an intelligent shopping guide recommendation system based on big data includes a memory, a processor, and a computer program stored in the memory and runnable on the processor. The processor includes a product association topology construction module, a user three-dimensional value modeling module, a dynamic product penetration pool generation module, a spatiotemporal path optimization module, and a dynamic instruction adjustment module.

[0098] The commodity association topology construction module is used to collect historical shopping records and commodity category information of the mall, and use a high-dimensional association calculation model to construct a commodity network topology diagram.

[0099] The user three-dimensional value modeling module is used to obtain the target customer's historical consumption records and current shopping demand list, quantify the target customer's purchasing power fluctuation band, model the decision patience attenuation curve, and analyze the shopping path tolerance, and establish a three-dimensional value model for the target customer.

[0100] The dynamic commodity penetration pool generation module is used to generate a dynamic optimal commodity penetration pool based on the commodity network topology diagram and the current shopping demand list using a commodity association mining strategy.

[0101] The space-time path optimization module is used to obtain the spatial coordinates of each commodity in the commodity penetration pool and the three-dimensional map of the shopping mall. Combined with the three-dimensional value model, it uses the space-time fusion field construction technology and the global dynamic path optimization algorithm to generate the optimal shopping guide route sequence.

[0102] The dynamic instruction adjustment module is used to obtain user positioning signals and dynamic change data of shopping cart items in real time, and generate dynamically adjusted shopping guide instructions based on the deviation of shopping guide routes and the deviation of shopping cart item types.

[0103] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An intelligent shopping guide recommendation method based on big data, characterized in that: include: Collect historical shopping records and product category information from the mall, and use a high-dimensional correlation calculation model to construct a product network topology diagram; Obtain the target customer's historical consumption records and current shopping needs list, quantify the target customer's purchasing power fluctuation band, model the decision patience decay curve, and analyze the shopping path tolerance to establish a three-dimensional value model for the target customer; Based on the commodity network topology diagram and the current shopping demand list, a commodity association mining strategy is adopted to generate a dynamic optimal commodity penetration pool; Obtain the spatial coordinates of each commodity in the commodity penetration pool and the three-dimensional map of the shopping mall, combine the three-dimensional value model, and use the space-time fusion field construction technology and the global dynamic path optimization algorithm to generate the optimal shopping guide route sequence; Acquire user positioning signals and dynamic changes in shopping cart items in real time, and generate dynamically adjusted shopping guide instructions based on shopping guide route deviations and shopping cart item type deviations.

2. The intelligent shopping guide recommendation method based on big data according to claim 1, characterized in that: The process of constructing a product network topology diagram includes: Clean the shopping mall's historical shopping records and extract product combination information from the complete transaction flow; Define a product association tensor, calculate the first-order association strength between any two products, and expand the multi-order association relationship. The first-order association strength represents the geometric mean of the conditional probability and the lift index; the multi-order association relationship represents the high-order collaborative purchasing pattern in the product combination; Storing the multi-order association relationship in a three-dimensional tensor form and generating topological edge weights by tensor decomposition; The connection edges whose topological edge weights are lower than the preset association strength threshold are removed to obtain a commodity network topology graph, in which nodes represent commodities and edges represent the association relationship between multiple commodity combinations.

3. The intelligent shopping guide recommendation method based on big data according to claim 1, characterized in that: The historical consumption records include basic transaction data, commodity details data, behavior trajectory data and return and exchange records; the current shopping demand list represents a list of planned purchase commodities actively input by the target customer.

4. The intelligent shopping guide recommendation method based on big data according to claim 1, characterized in that: The process of quantifying the purchasing power fluctuation band of target customers includes: Based on the target customer's historical consumption records, calculate the arithmetic mean and standard deviation of the target customer's single shopping consumption amount; Identify the time-periodic characteristics of target customers' consumption behavior and construct a fluctuation function that combines the time decay factor; With the volatility function value as the center line, the dynamic volatility band interval is generated according to the standard deviation multiple; By real-time monitoring of the cumulative consumption amount in the target customer's shopping demand list, the fluctuation range is dynamically adjusted.

5. The intelligent shopping guide recommendation method based on big data according to claim 1, characterized in that: The process of modeling the decision patience decay curve for target customers includes: Analyze the target customer's historical behavior records, calculate the average time it takes for the target customer to browse products and make a purchase decision, and map the average time to an initial patience value; Establishing a hyperbolic decay function model so that the initial patience value exhibits nonlinear decay characteristics as the decision time passes; Dynamically adjust the attenuation rate coefficient based on current environmental parameters, including mall crowd density and promotion intensity; When the deviation between the actual decision time of the target customer and the predicted value exceeds the preset deviation threshold, the attenuation function parameters are modified.

6. The intelligent shopping guide recommendation method based on big data according to claim 1, characterized in that: The process of analyzing the shopping path tolerance of target customers includes: Extract the abandonment records of target customers’ shopping history and fit the benchmark threshold of target customers’ tolerance; Establishing a spatial distance tolerance conversion model based on the benchmark threshold and calculating a tolerance energy function; It takes a long time to integrate this shopping in real time and dynamically update the path tolerance threshold.

7. The intelligent shopping guide recommendation method based on big data according to claim 1, characterized in that: The process of generating a dynamic optimal product penetration pool using a product association mining strategy includes: In the product network topology, starting from the target customer's current shopping list, calculate the penetration rate of all candidate products that can be reached through direct connection paths; Conduct multi-demand collaborative correction to detect the second-order impact of the product combination in the target customer's current shopping demand list on the candidate products; Monitor the decay rate of the commodity association strength in the pool in real time, and screen the initial commodity pool based on the dynamically preset penetration threshold.

8. The intelligent shopping guide recommendation method based on big data according to claim 1, characterized in that: The process of generating the optimal shopping guide route sequence includes: Map the spatial coordinates and value weight of each commodity in the commodity penetration pool into a four-dimensional space-time field; The user's three-dimensional value model is superimposed to form a comprehensive constraint field to establish a spatiotemporal optimization equation to obtain a continuous path curve. The comprehensive constraint field includes the value constraint defined by the current fluctuation band interval, the maximum stay time determined by the decision patience value, and the spatial movement limit converted from the path tolerance. The continuous path curve is discretized into a shopping guide node sequence, and the nodes are sorted by spatial accessibility and output as a shopping guide route instruction set to obtain an optimal shopping guide route sequence.

9. The intelligent shopping guide recommendation method based on big data according to claim 1, characterized in that: The process of generating dynamically adjusted shopping guide instructions includes: Obtain the target customer's location coordinates and shopping cart item list in real time, and establish a product decision entropy monitoring model based on the offset between the real-time location coordinates and shopping cart item list and the optimal shopping guide route sequence and product penetration pool; When any deviation exceeds a preset threshold, a dynamic replanning condition is triggered. The dynamic replanning condition includes recalculating the spatiotemporal field gradient direction based on the latest data and generating a correction instruction set including steering instructions, speed adjustment and product replacement suggestions.

10. An intelligent shopping guide recommendation system based on big data, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, it implements the intelligent shopping guide recommendation method based on big data as described in any one of claims 1 to 9.

Images