An intelligent optimization method and system for store operation

Abstract

The present application discloses an intelligent optimization method and system for store operation, including: obtaining the behavior data of users in the store; analyzing the cognitive level of users through a preset dynamic user stratification model according to the behavior data to obtain multiple user levels; based on the behavior data, analyzing the interests of each user level through a preset user interest map construction model to construct an interest map corresponding to each user level; generating a corresponding initial operation plan through a preset operation plan generation model according to the interest map; evaluating the effect of the initial operation plan through a preset plan effect evaluation model to obtain the plan effect; optimizing the initial operation plan through a preset operation plan optimization model according to the plan effect to obtain an optimized operation plan; by dynamically optimizing the operation plan, the present application can timely adapt to market changes and user needs, and improve the competitiveness of the store.

Description

Technical Field

[0001] The present invention relates to the technical field of store operation, and particularly to an intelligent optimization method and system for store operation. Background Art

[0002] Current education product stores face challenges such as a wide variety of products, significant age differences and demand differences among service objects. How to make corresponding product recommendations and activity plans for children of different ages is crucial for improving the service quality and economic benefits of the store. However, the existing methods for formulating store operation plans lack in-depth mining and accurate analysis of user behavior data, making it difficult to accurately grasp the cognitive level and interests of users, resulting in a lack of pertinence in the operation plans; relying on empirical judgment in plan formulation, lacking systematicness and scientificity, and being unable to adapt to market changes and dynamic changes in user needs in a timely manner; classifying users using a single standard, unable to make dynamic adjustments according to the actual situation of users, and unable to meet the ever-changing needs of users.

[0003] To solve at least one of the above problems, the present application proposes an intelligent optimization method and system for store operation. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the main purpose of the present invention is to provide an intelligent optimization method and system for store operation, which can effectively solve the problems in the background art. The specific technical solutions of the present invention are as follows:

[0005] An intelligent optimization method for store operation includes:

[0006] Obtaining the behavior data of users in the store;

[0007] Analyzing the cognitive level of users through a preset dynamic user stratification model according to the behavior data to obtain multiple user levels;

[0008] Analyzing the interest tags of users based on the behavior data, and analyzing the interest frequency of each user level through a preset user interest map construction model to construct an interest map corresponding to each user level, wherein the user level in the interest map is the root node, the interest tag is the child node, and the corresponding root node and child node are connected through the interest frequency;

[0009] Screening out the core interests from the interest map, and mapping them in a preset activity template library through a preset operation plan generation model to generate a corresponding initial operation plan;

[0010] Evaluating the effect of the initial operation plan through a preset plan effect evaluation model to obtain the plan effect;

[0011] According to the effects of the above solution, through a preset operation plan optimization model, combined with non-core interests in the interest graph, optimize the activity sequence and activity time of the initial operation plan to obtain an optimized operation plan, so as to achieve intelligent optimization of store operations.

[0012] Specifically, based on the behavioral data, analyze the cognitive level of users through a preset dynamic user stratification model to obtain multiple user levels, including:

[0013] According to the user age in the behavioral data, divide multiple age intervals along the time axis to obtain age groups;

[0014] Based on the age groups, analyze the cognitive level of each group of users to obtain the corresponding cognitive index value for each group of users;

[0015] Combined with the cognitive index values, merge according to the similarity between groups being greater than a preset similarity threshold, and split according to the standard deviation of the cognitive index values within a group being greater than a preset standard deviation threshold. Dynamically adjust the age groups through a preset dynamic user stratification model to obtain multiple user levels.

[0016] Specifically, the step of dividing multiple age intervals along the time axis according to the user age in the behavioral data to obtain age groups includes:

[0017] According to the user age in the behavioral data, calculate the corresponding age density;

[0018] Generate multiple seed points according to the age density;

[0019] Based on the seed points, divide the time axis into multiple age intervals to obtain initial age intervals, where each seed point corresponds to an initial age interval;

[0020] Set overlapping regions between every two adjacent initial age intervals according to a preset interval length;

[0021] Allocate according to the user density in the overlapping regions to the corresponding initial age intervals to obtain age groups.

[0022] Specifically, based on the behavioral data, analyze the interest tags of users, and analyze the interest frequencies of each user level through a preset user interest graph construction model to construct an interest graph corresponding to each user level. Among them, the user level in the interest graph is the root node, and the interest tag is the child node, and the corresponding root node and child node are connected through the interest frequency, including:

[0023] According to the behavioral data, extract keywords through a preset keyword extraction model to obtain high-frequency behavioral keywords;

[0024] Map the high-frequency behavior keywords to corresponding tags through a preset keyword mapping model to obtain the user's interest tags;

[0025] According to the interest tags and combined with the user hierarchy, analyze the interest frequencies of each user hierarchy through a preset user interest graph construction model, and construct an interest graph corresponding to each user hierarchy.

[0026] Specifically, the step of analyzing the interest frequencies of each user hierarchy through a preset user interest graph construction model according to the interest tags and combined with the user hierarchy to construct an interest graph corresponding to each user hierarchy includes:

[0027] According to the user hierarchy corresponding to the interest tags, obtain the interest frequencies by analyzing the operation frequencies and durations of the user for different interest tags;

[0028] Through a preset user interest graph construction model, use the user hierarchy as the root node, the interest tags as the child nodes, and the interest frequencies as the weights of the edges to connect the corresponding root nodes and child nodes to obtain an initial interest graph;

[0029] Combined with real-time user data, when the change value of the interest frequency within the same user hierarchy is greater than a preset first threshold, update the initial interest graph to obtain an interest graph corresponding to each user hierarchy.

[0030] Specifically, the step of screening out the core interests from the interest graph, mapping through a preset operation plan generation model in a preset activity template library, and generating a corresponding initial operation plan includes:

[0031] According to the weights of the edges in the interest graph, select the interest tags with interest frequency values greater than a preset second threshold as the core interests;

[0032] Based on the core interests, map through a preset operation plan generation model and a preset activity template library to obtain corresponding operation activities;

[0033] Optimize the operation activities through a preset activity optimization model to obtain a corresponding initial operation plan.

[0034] Specifically, the step of optimizing the operation activities through a preset activity optimization model to obtain a corresponding initial operation plan includes:

[0035] According to the operation activities, arrange the activity sequence and duration through a preset activity arrangement model to obtain an initial activity arrangement;

[0036] Within a preset time period, optimize the initial activity arrangement through a preset activity optimization model to obtain a corresponding initial operation plan.

[0037] Specifically, according to the effects of the said solution, through a preset operation plan optimization model, combined with non-core interests in the interest graph, optimize the activity sequence and activity time of the initial operation plan to obtain an optimized operation plan, so as to realize the intelligent optimization of store operation, including:

[0038] Calculate the effect score value of the initial operation plan according to the solution effect;

[0039] When the effect score value is less than a preset third threshold, combined with non-core interests in the interest graph, optimize the activity sequence and activity time of the initial operation plan through a preset operation plan optimization model to obtain an optimized operation plan.

[0040] Specifically, when the effect score value is less than a preset third threshold, combined with non-core interests in the interest graph, optimize the activity sequence and activity time of the initial operation plan through a preset operation plan optimization model to obtain an optimized operation plan, including:

[0041] When the effect score value is less than a preset third threshold, combined with non-core interests in the interest graph, generate a candidate plan through a preset strategy generation model;

[0042] According to the interest graph, sort the activities in the initial operation plan and the candidate plan to obtain the activity sequence;

[0043] Based on the activity sequence, optimize the activity time through a preset operation plan optimization model to obtain an optimized operation plan.

[0044] A store operation intelligent optimization system for implementing the said store operation intelligent optimization method, including:

[0045] A data acquisition module, which acquires the behavior data of users in the store;

[0046] A user grading module, according to the said behavior data, analyzes the cognitive level of users through a preset dynamic user stratification model to obtain multiple user levels;

[0047] An interest graph construction module, based on the said behavior data, analyzes the interest tags of users, analyzes the interest frequency of each user level through a preset user interest graph construction model, and constructs an interest graph corresponding to each user level, wherein the user level in the interest graph is the root node, the interest tag is the child node, and the corresponding root node and child node are connected through the interest frequency;

[0048] An operation plan generation module, screens out core interests from the said interest graph, and maps them in a preset activity template library through a preset operation plan generation model to generate a corresponding initial operation plan;

[0049] The operation plan evaluation module evaluates the effect of the initial operation plan through a preset plan effect evaluation model to obtain the plan effect;

[0050] The operation plan optimization module optimizes the activity sequence and activity time of the initial operation plan according to the plan effect through a preset operation plan optimization model, combined with non-core interests in the interest map, to obtain an optimized operation plan, so as to realize intelligent optimization of store operation.

[0051] Compared with the prior art, the present application has the following beneficial effects:

[0052] The present application constructs a corresponding user interest map by combining user dynamic stratification, formulates personalized operation plans for different levels of users, and dynamically optimizes the operation plan according to the plan effect, can more accurately understand the characteristics and needs of different user levels, the constructed interest map can reflect the interest changes of users in real time, making the operation plan more in line with the actual interests of users; through the collaborative work of multiple models, the generated operation plan is more scientific and reasonable, can effectively improve the operation efficiency and effect; dynamically optimizing the operation plan according to the plan effect can timely adapt to market changes and user needs, and enhance the competitiveness of the store Description of the Drawings

[0053] Figure 1 It is a working flow chart of an intelligent optimization method for store operation in Embodiment 1 of the present invention;

[0054] Figure 2 It is a schematic diagram of interest map construction in Embodiment 1 of the present invention;

[0055] Figure 3 It is a schematic diagram of the mapping between core interests and activity template libraries in Embodiment 1 of the present invention;

[0056] Figure 4 It is a schematic diagram of the structure of an intelligent optimization system for store operation in Embodiment 2 of the present invention. Detailed Embodiments

[0057] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the detailed embodiments of the present invention with reference to the accompanying drawings of the specification.

[0058] Many specific details are set forth in the following description in order to fully understand the present invention, but the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar promotions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0059] Second, the "one embodiment" or "embodiment" referred to herein means a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that are mutually exclusive of other embodiments.

[0060] Embodiment 1:

[0061] This embodiment provides an intelligent optimization method for store operation, as Figure 1 shown, the intelligent optimization method for store operation includes:

[0062] S101. Obtain the behavior data of users in the store;

[0063] S102. Analyze the cognitive level of users through a preset dynamic user stratification model according to the behavior data to obtain multiple user levels;

[0064] S103. Analyze the interest tags of users based on the behavior data, and analyze the interest frequency of each user level through a preset user interest map construction model to construct an interest map corresponding to each user level. Among them, the user level in the interest map is the root node, and the interest tag is the sub-node, and the corresponding root node and sub-node are connected through the interest frequency;

[0065] S104. Screen out the core interests from the interest map, and map them in a preset activity template library through a preset operation plan generation model to generate a corresponding initial operation plan;

[0066] S105. Evaluate the effect of the initial operation plan through a preset plan effect evaluation model to obtain the plan effect;

[0067] S106. According to the plan effect, through a preset operation plan optimization model, combine the non-core interests in the interest map to optimize the activity sequence and activity time of the initial operation plan to obtain an optimized operation plan, so as to realize the intelligent optimization of store operation.

[0068] In this embodiment, users are dynamically stratified, corresponding interest maps are constructed, and personalized operation plans are formulated in combination with the interest maps, so that the operation plans can accurately reach the target user group; compared with the existing single operation plan formulation method, through dynamic adjustment of user stratification, interest maps, and operation plans, it can be optimized according to real-time data and plan effects to ensure that store operation always adapts to market changes and user needs.

[0069] In this embodiment, cameras are arranged in the store to capture behavioral data of users, such as their walking paths, staying areas, and products of interest. The store's sales system is used to collect users' purchase records, including the types and quantities of products purchased and the purchase time. The learning robot in the store can obtain users' operation records and the usage time of each function. Behavioral data such as users' browsing records, search records, and favorite products on the online platform are collected online. By comprehensively and accurately obtaining users' behavioral data, a sufficient data basis is provided for analyzing users' characteristics and needs.

[0070] Specifically, according to the collected behavioral data, users are dynamically stratified. The users are stratified through a preset dynamic user stratification model to obtain multiple user levels. The preset dynamic user stratification model can be a machine learning model such as a clustering analysis model or a decision tree model. The user stratification results are dynamically adjusted. Considering the differences in cognitive levels among users of different ages through user stratification, by analyzing the age information in the behavioral data and combining the users' behavioral performances in the store, such as their understanding of product information, the depth of their inquiries, and their operation records on the learning robot, the dynamic user stratification model is used to divide users into different levels according to their cognitive levels. Through precise user stratification, the store can formulate more targeted marketing strategies and service methods for users with different cognitive levels. For example, for children with a higher cognitive level, more difficult and more professional learning activities and learning products can be provided; for children with a lower cognitive level, more basic learning content and products can be provided, thereby improving the operation effect and user satisfaction.

[0071] Specifically, after stratifying the users, an interest analysis is conducted on each layer of users. Through a preset user interest graph construction model, an interest graph corresponding to each user level is constructed. By extracting keywords from the users' behavioral data and mapping them to interest tags, and then combining the user levels and information such as the operation frequency and duration of users on the interest tags, a graph that can reflect the interest preferences of different user levels is constructed. The constructed interest graph can intuitively display the interest preferences of different user levels, helping the store clearly understand the interest points of the target users. When the store plans activities, displays products, or pushes information, it can more accurately meet the users' needs according to the interest graph, improving user participation and purchase intention. For example, if it is found through the interest graph that a certain user level has a strong interest in a certain type of electronic product, the store can increase the display and promotional activities of related products in that area.

[0072] Meanwhile, according to the constructed interest graph, a corresponding operation plan is generated. Through a preset operation plan generation model, based on the constructed interest graph, interest tags with high weight values are selected as the core interests, which are mapped to the preset activity template library to generate corresponding operation activities, and then further optimized to obtain an initial operation plan. The generated initial operation plan can closely revolve around the user interests, with high pertinence and feasibility, quickly and effectively attracting user attention, improving the effect of operation activities, and greatly increasing the probability of success compared with the traditional operation plan formulated based on experience.

[0073] Specifically, after the initial operation plan is generated, the effect of the initial operation plan is evaluated through a preset plan effect evaluation model. Using the preset evaluation indicators and algorithms, the effect after the implementation of the initial operation plan is quantitatively evaluated from multiple dimensions to obtain the plan effect. The preset plan effect evaluation model can be the analytic hierarchy process, the fuzzy comprehensive evaluation method, etc. The plan effect evaluation model in this embodiment is specifically the analytic hierarchy process. The correlation relationship between the evaluation indicators and the initial operation plan is established, and the judgment matrix is constructed by combining the effects of the activities corresponding to the indicators to obtain the weight of each evaluation indicator. The comprehensive score of the plan effect is calculated through weighted summation to obtain the corresponding plan effect. Through objective and scientific evaluation, the actual effect of the initial operation plan can be evaluated, the advantages and disadvantages of the plan can be found, so as to optimize the plan and avoid blindly adjusting the operation plan. According to the plan effect evaluation result, when the effect does not meet the expectation, combined with information such as non-core interests in the interest graph, a candidate plan is generated using the operation plan optimization model, and the activity sequence and time are optimized to obtain a better operation plan. By continuously optimizing the operation plan, the operation effect of the store can be continuously improved, adapting to the dynamic adjustment of market changes and user needs, keeping the store operation at a relatively high level all the time, and improving the competitiveness and profitability of the store.

[0074] This application constructs a corresponding user interest graph in combination with user dynamic stratification, formulates personalized operation plans for different levels of users, and dynamically optimizes the operation plans according to the plan effect, which can more accurately understand the characteristics and needs of different user levels. The constructed interest graph can reflect the interest changes of users in real time, making the operation plan more in line with the actual interests of users. Through the collaborative work of multiple models, the generated operation plan is more scientific and reasonable, which can effectively improve the operation efficiency and effect. Dynamically optimizing the operation plan according to the plan effect can timely adapt to market changes and user needs, and enhance the competitiveness of the store.

[0075] Furthermore, according to the behavior data, the cognitive level of users is analyzed through a preset dynamic user stratification model to obtain multiple user levels, including:

[0076] S201. Divide multiple age ranges according to the user age in the behavior data along the time axis to obtain age groups.

[0077] S202. Analyze the cognitive level of each group of users according to the age groups to obtain the corresponding cognitive index values for each group of users.

[0078] S203. Combine the cognitive index values, merge according to the similarity between groups being greater than a preset similarity threshold, and split according to the standard deviation of the cognitive index values within a group being greater than a preset standard deviation threshold. Dynamically adjust the age groups through a preset dynamic user stratification model to obtain multiple user levels.

[0079] In this embodiment, users are grouped according to the user age in the collected user behavior data. Multiple age ranges are divided along the time axis to obtain age groups. Considering that users in different age groups often have differences in consumption habits, hobbies, etc., by grouping users by age, users with similar characteristics can be initially classified into one category, which can intuitively reflect the distribution of different age stages.

[0080] Specifically, based on the age groups, analyze the behavior performance of each group of users in the store. These behaviors can reflect the understanding and cognitive level of users towards products, services, etc. Quantify the cognitive level of users into cognitive index values to facilitate the comparison and differentiation of the cognitive levels of users in different age groups. Set a series of indicators for measuring the cognitive level and assign corresponding weights to each indicator. Calculate the cognitive index value of each user through the weighted sum of the quantified data of the user on each indicator for measuring the cognitive level and the corresponding weight. For example, set the weight of the viewing time of the product manual to 0.3, the weight of the complexity of the questions consulted to the store clerk to 0.4, the weight of the staying time in the product area to 0.2, and the weight of the understanding and reaction to the activity rules to 0.1. Calculate the cognitive index value of each user according to the set indicators and weights, and calculate the average of the cognitive index values of all users in each age group to obtain the corresponding cognitive index value for each group of users. Calculating the cognitive index value helps the store to more accurately understand the cognitive characteristics of users in different age groups and provides a basis for formulating targeted marketing strategies and service plans.

[0081] Specifically, after obtaining the cognitive index values corresponding to each age group, by comparing the similarity between groups (including cosine similarity, Euclidean distance, Pearson correlation coefficient, etc. between the cognitive index values corresponding to age groups) and the dispersion degree of the cognitive index values within the group (standard deviation), the initial age groups are optimized and adjusted using a preset dynamic user stratification model. The preset dynamic user stratification model can be a machine learning model such as a clustering analysis model or a decision tree model. In this embodiment, the dynamic user stratification model is specifically a random forest model. The random forest model is trained with a large amount of historical grouping data to obtain a pre-trained dynamic user stratification model. The discrete degree of each age group, the similarity between groups, and the cognitive index values within the group are input into the pre-trained dynamic user stratification model. When the similarity between groups is relatively high, it indicates that the cognitive levels of these two groups of users are relatively similar and can be merged into one group. When the standard deviation of the cognitive index values within the group is large, it means that the cognitive levels of the users within the group vary greatly and need to be further split into different levels to more accurately reflect the true situation of the users. The model outputs the user stratification result after merging or splitting. By dynamically adjusting the user stratification, the grouping result can more accurately reflect the differences in the cognitive levels of users, avoiding the irrationality caused by simply grouping by age, enabling the store to formulate more refined and personalized operation strategies for different user levels with different cognitive levels.

[0082] Exemplarily, by calculating the cosine similarity of the cognitive index between age groups, the similarity between age groups, that is, the similarity between groups, is obtained. The calculated similarity between groups is compared with a preset similarity threshold. The preset similarity threshold is set to 0.95. When the similarity of 0.98 between age groups A and B is greater than the threshold, these two age groups are merged into a new group. At the same time, the standard deviation of the cognitive index values within each age group is calculated, and the calculated standard deviation within the group is compared with a preset standard deviation threshold. The preset standard deviation threshold is set to 0.8. When the standard deviation of the cognitive index values within the group, 0.96, is greater than the threshold, it indicates that the cognitive levels of the users within this group vary greatly and this group needs to be split. After the merge and split operations, the age groups are updated through the preset dynamic user stratification model to obtain multiple user levels.

[0083] Further, the dividing the user ages in the behavior data into multiple age intervals according to the time axis to obtain age groups includes:

[0084] S301. Calculate the corresponding age density according to the user age in the behavior data;

[0085] S302. Generate multiple seed points according to the age density;

[0086] S303. Divide the time axis into multiple age intervals according to the seed points to obtain initial age intervals, where each seed point corresponds to an initial age interval;

[0087] S304. Set overlapping regions between every two adjacent initial age intervals according to a preset interval length;

[0088] S305. Assign to the corresponding initial age interval according to the user density of the overlapping region to obtain age groups.

[0089] In this embodiment, the age density is calculated based on the user age to measure the distribution of users in different age groups among the overall users. By calculating the age density, it can be known which age groups have relatively more users and which have relatively fewer users, providing a basis for determining the seed points and dividing the age intervals. Based on the statistics of the user frequencies in different age groups, the central tendency and distribution characteristics of the user age are reflected.

[0090] Specifically, sort the calculated age density data, find the age intervals with higher age density, select a representative age in each age interval with higher age density as the seed point. In this embodiment, take the middle value of the interval. For example, in the 10 - 15 age interval, select 12 years old as the seed point; in the 15 - 20 age interval, select 17 years old as the seed point; divide the time axis based on the seed points, and divide the users into different intervals according to age. Each interval corresponds to a seed point. By corresponding the seed points with the age intervals, it can be ensured that each age interval is centered around a representative age, so that the users within the same interval have certain similarities in age. Centered on each seed point, expand the interval according to the width of the preset initial age interval. In this embodiment, expand 2 years to the left and right of the seed point as the width of the initial age interval to obtain the initial age intervals including 10 - 14 years old and 15 - 19 years old.

[0091] Meanwhile, between every two adjacent initial age intervals, an overlapping area is set according to a preset interval length. The length of the overlapping area is set according to business requirements and data characteristics. By setting the overlapping area, the flexibility and accuracy of age grouping are increased, which can better reflect the continuity of user age characteristics and avoid missing some users with transitional characteristics due to simple age interval division. According to the user density in the overlapping area, the overlapping area is assigned to the corresponding age interval. The user density characteristics in the overlapping area are compared with those in the adjacent initial age intervals. According to the comparison results, the users in the overlapping area are assigned to the initial age interval with more similar user density characteristics. For example, if the age interval of the overlapping area is 13 - 16 years old, when the user density of users who purchase a specific product in the overlapping area is closer to the user density in the 10 - 14 - year - old interval, the users in the overlapping area are assigned to the 10 - 14 - year - old age interval; otherwise, they are assigned to the 15 - 19 - year - old interval. After all the users in the overlapping area are assigned, the final age grouping is obtained. By assigning according to the user density in the overlapping area, the age grouping result can be further optimized, making the users in each age grouping more similar in behavior characteristics, providing a more reliable basis for the store to formulate targeted operation strategies, and improving the effectiveness and accuracy of operation strategies.

[0092] Furthermore, the interest tags of users are analyzed based on the behavioral data. By using a preset user interest graph construction model, the interest frequency of each user level is analyzed, and an interest graph corresponding to each user level is constructed. Among them, the user level in the interest graph is the root node, and the interest tag is the child node. The corresponding root node and child node are connected through the interest frequency, including:

[0093] S401. According to the behavioral data, high - frequency behavioral keywords are extracted through a preset keyword extraction model.

[0094] S402. Through a preset keyword mapping model, the high - frequency behavioral keywords are mapped to the corresponding tags to obtain the interest tags of users.

[0095] S403. According to the interest tags and combined with the user level, through a preset user interest graph construction model, the interest frequency of each user level is analyzed, and an interest graph corresponding to each user level is constructed.

[0096] In this embodiment, high-frequency behavior keywords are identified from behavior data through a preset keyword extraction model. The keyword extraction model is specifically a TF-IDF model. Calculate the term frequency of each word, that is, the number of times the word appears divided by the total number of words; then calculate the inverse document frequency, which reflects the rarity of a word. It is obtained by dividing the total number of documents by the number of documents containing the word and then taking the logarithm. Finally, multiply the term frequency and inverse document frequency of each word to obtain the TF-IDF value of the word. The higher the TF-IDF value of a word, the more important it is in this document and the relatively less common it is in other documents. Select the words with higher weights as high-frequency behavior keywords; by screening keywords, the behavior information of users can be highly condensed, and complex behavior data can be simplified into a representative set of words.

[0097] Specifically, according to the extracted keywords, through a preset keyword mapping model, the preset keyword mapping model can be a model such as a rule-based mapping model, a deep learning model, etc. The preset keyword mapping model in this embodiment is specifically a mapping model based on mapping rules, map the high-frequency behavior keywords to corresponding labels to obtain the user's interest labels, map the keywords to a predefined interest label system. The keyword mapping model maps specific behavior keywords to interest labels by establishing a corresponding mapping relationship between keywords and interest labels. For example, when the high-frequency behavior keywords are "eraser" and "pencil", map the high-frequency behavior keywords to the interest label "stationery"; combined with the analyzed user interest labels, construct a corresponding interest graph through a preset user interest graph construction model. The preset user interest graph construction model can be a model such as an association rule mining model, a graph embedding model, a deep learning model, etc. Take the user level as the root node of the graph and the interest labels as the sub-nodes, and use the weight of the edge to represent the degree of attention of the user to the interest label. By combining the user level and interest labels to construct an interest graph, the interest preferences and differences of different user levels can be clearly presented, helping the store to formulate personalized operation strategies for different user groups.

[0098] Further, analyzing the interest frequency of each user level through a preset user interest graph construction model according to the interest label combined with the user level, and constructing an interest graph corresponding to each user level, includes:

[0099] S501. According to the user level corresponding to the interest label, obtain the interest frequency by analyzing the operation frequency and duration of the user for different interest labels;

[0100] S502. Through a preset user interest graph construction model, take the user level as the root node, the interest label as the sub-node, and use the interest frequency as the weight of the edge to connect the corresponding root node and sub-node to obtain an initial interest graph;

[0101] S503. When combining real-time user data, when the interest frequency change value within the same user level is greater than a preset first threshold, update the initial interest map to obtain an interest map corresponding to each user level.

[0102] In this embodiment, the operation frequency and duration of users in each user level for different interest tags are analyzed respectively to obtain the interest frequency. The operation behavior of users on the products or contents associated with different interest tags reflects their attention degree to this interest. The operation frequency reflects the participation activity of users in a specific interest, and the operation duration reflects the concentration depth of users on this interest. By comprehensively analyzing the data of these two dimensions, the interest degree of users for each interest tag can be quantified. The operation frequency and operation duration of users on the corresponding interest tag are weighted and summed to calculate the interest frequency corresponding to each interest tag. By calculating the interest frequency, the association strength between the user level and the interest tag can be accurately measured.

[0103] Specifically, as Figure 2 shown, through a preset user interest map construction model, the user interest map construction model in this embodiment is specifically a graph embedding model. According to the calculated interest frequency, an interest map is constructed. The user level is used as the root node, the interest tag is used as the child node, and the interest frequency is used as the weight of the edge to obtain the corresponding interest map. By constructing the interest map, the interest distribution and interest strength of different user levels are reflected. The interests of users are not static. With the passage of time, the change of the market environment, and the emergence of new products or services, the interests of users will change. By real-time monitoring user data, calculating the change value of the interest frequency, and comparing it with the preset first threshold, when the change value exceeds the preset first threshold, it indicates that the interests of users at the corresponding level have changed significantly. Update the interest map. By real-time updating the interest map, the store can always keep up with the dynamic changes of users' interests and adjust the operation strategy in a timely manner.

[0104] Further, screening out the core interests from the interest map and mapping them in a preset activity template library through a preset operation plan generation model to generate a corresponding initial operation plan, including:

[0105] S601. According to the weight of the edge in the interest map, select the interest tags with interest frequency values greater than a preset second threshold as the core interests;

[0106] S602. Based on the core interests, map through a preset operation plan generation model and a preset activity template library to obtain corresponding operation activities;

[0107] S603. Optimize the operation activities through a preset activity optimization model to obtain a corresponding initial operation plan.

[0108] In this embodiment, according to the weights of the edges in the interest graph, the interest tags are screened, and the interest tags with an interest frequency value (the value corresponding to the interest frequency) greater than a preset second threshold are selected as the core interests. The weights of the edges in the interest graph reflect the user's attention degree and interest intensity for different interest tags. The preset second threshold is set according to the actual business situation and data characteristics. All the interest tags in the interest graph are traversed to obtain the weight value of each interest tag, that is, the interest frequency value. The weight value of each interest tag is compared with the preset second threshold. When the weight value of a certain interest tag is greater than the preset second threshold, it is selected as the core interest. By selecting the core interests, it can help the operator quickly grasp the main interest points of the user group and avoid dispersing energy when formulating the operation plan. Guided by the core interests, the operation plan can be more targeted, directly hit the key needs of users, improve the effectiveness and attractiveness of the operation plan, and thus better attract users to participate and improve user satisfaction.

[0109] Specifically, as Figure 3 shown, the interest tags with weights greater than 0.5 are selected as the core interests. According to the selected core interests, through the mapping between the preset operation plan generation model and the preset activity template library, the corresponding operation activities are obtained. The preset activity template library stores a variety of already designed operation activity templates suitable for different scenarios and interest points. The operation plan generation model is specifically a random forest model, which is trained through a large amount of historical data to obtain a pre-trained operation plan generation model. The core interests are input into the pre-trained operation plan generation model. The operation plan generation model searches and matches in the activity template library according to the learned mapping relationship to obtain the corresponding operation activities. By generating operation activities related to the core interests, it saves the time and cost of designing operation activities, can ensure that the generated operation activities have high feasibility and attractiveness, and improves the quality and success rate of the operation activities.

[0110] At the same time, the effects of the operation activities are evaluated, and according to the evaluation results, the initially generated operation activities are optimized from multiple dimensions through a preset activity optimization model. The preset activity optimization model includes, but is not limited to, the particle swarm optimization model. By optimizing the operation activities through the activity optimization model, it can ensure that the operation plan has higher feasibility and effectiveness in actual execution. The optimized operation plan can better balance the activity cost and expected benefits, improve user participation and satisfaction, enhance the competitiveness of the activity in the market, and thus bring better operation effects and economic benefits to the store.

[0111] Further, the step of optimizing the operation activities through the preset activity optimization model to obtain the corresponding initial operation plan includes:

[0112] S701. Arrange the activity sequence and duration according to the operation activities through a preset activity arrangement model to obtain an initial activity arrangement.

[0113] S702. Optimize the initial activity arrangement through a preset activity optimization model within a preset time period to obtain a corresponding initial operation plan.

[0114] In this embodiment, according to multiple matched operation activities, arrange the sequence and duration of the activities. Different operation activities have different characteristics and goals. Reasonably arranging the activity sequence and duration can make them cooperate with each other to achieve the best effect. Plan the activity process through a preset activity arrangement model. The preset activity arrangement model includes, but is not limited to, a genetic algorithm model. Train the genetic algorithm model through a large number of historical operation activities to obtain a pre-trained activity arrangement model, and arrange the activity sequence and duration of multiple activities to make the activity effect optimal. Clearly define each sub-activity included in the operation activity and its goal, and classify them according to the nature of the sub-activities. For example, promotional activities focus on the scope and speed of information dissemination, experience activities emphasize user participation and interactivity, and promotional activities focus on the purchase conversion rate. According to different users corresponding to different times, formulate the activity sequence and duration. Through a reasonable initial activity arrangement, it can ensure that the operation activities are carried out in an orderly manner, the sub-activities cooperate with each other, improve the attractiveness and participation of the activities, avoid the blindness of activity implementation, and enhance the overall effect of the operation activities.

[0115] Exemplarily, based on the law of users' online activity, start an online preheating promotion at 7 pm on weekdays for 3 consecutive days to fully cover the target user group; then carry out an offline learning experience activity on weekends for 2 days, which conforms to the characteristics of users' offline participation time; then launch an online time-limited discount activity at 8 pm on weekdays next week for 1 day to promote purchases using the free time after users finish school; finally, conduct a post-sales feedback activity within one week after the activity ends to consolidate user relationships. Through such an arrangement, an initial activity arrangement is obtained.

[0116] Specifically, based on the initial activity arrangement, the initial activity arrangement is optimized through a preset activity optimization model. The preset activity optimization model includes, but is not limited to, a particle swarm optimization model. The particle swarm optimization model is trained with a large amount of activity execution data to obtain a pre-trained activity optimization model. The model performs optimization steps such as sequential optimization and duration optimization on multiple activities, searches for the optimal activity arrangement plan, and obtains an optimized initial operation plan. During a preset time period, real-time data during the activity execution process is analyzed to adjust aspects such as activity sequence, duration, and resource allocation to achieve better operation effects. According to the scale and nature of the operation activities, a suitable time period is determined for optimization analysis. For a one-month promotion activity, the optimization evaluation is carried out on a weekly basis. During the activity execution process, various data is collected in real time, and the collected data is input into the activity optimization model. The model analyzes the reasons for low user participation or unmet sales expectations. According to the analysis results, the activity optimization model uses algorithms to optimize and adjust the initial activity arrangement. Optimizing the initial activity arrangement within a preset time period can timely discover problems during the activity execution process and make adjustments according to the actual situation, making the operation plan more in line with market changes and user needs, improving the success rate and operation effects of the activity, maximizing the benefits of the operation activity, and bringing better economic benefits and market competitiveness to the store.

[0117] Exemplarily, when the purchase conversion rate of the online time-limited discount activity is low, the model suggests increasing the discount intensity, extending the activity duration by half a day, and at the same time optimizing the activity page design to improve the loading speed. When the number of participants in the offline new product experience activity is small, the model suggests adding a round of targeted publicity and promotion before the activity and adjusting the activity time to a more suitable period on weekend afternoons. Through these adjustments, an optimized initial operation plan is obtained.

[0118] Furthermore, according to the effect of the plan, through a preset operation plan optimization model, combined with non-core interests in the interest graph, the activity sequence and activity time of the initial operation plan are optimized to obtain an optimized operation plan to achieve intelligent optimization of store operations, including:

[0119] S801. Calculate the effect score value of the initial operation plan according to the plan effect;

[0120] S802. When the effect score value is less than a preset third threshold, combine non-core interests in the interest graph and optimize the activity sequence and activity time of the initial operation plan through a preset operation plan optimization model to obtain an optimized operation plan.

[0121] In this embodiment, according to the effect of the solution, the effect score value of the initial operation solution is calculated. By collecting various types of data related to the implementation of the solution, a comprehensive effect score value is calculated; the indicators for evaluating the effect of the operation solution are determined, including the growth of sales, the change in customer flow, user participation, etc. According to the importance of each indicator for the store operation and combined with the business objectives of the store, corresponding weights are assigned to each indicator; the collected data is weighted and calculated according to the set weights to obtain the effect score value; the calculated effect score value is compared with a preset third threshold. When the effect score value is less than the preset third threshold, the initial operation solution is optimized through a preset operation solution optimization model to obtain an optimized operation solution. The preset operation solution optimization model can be a decision tree model, a reinforcement learning model, etc. The preset operation solution optimization model can conduct scientific analysis based on multiple factors, quickly generate effective optimization strategies, improve the effect of the operation solution, and the optimized operation solution can better adapt to market changes and user needs, which helps to improve the operation efficiency and competitiveness of the store and realize the intelligent optimization of store operation.

[0122] Specifically, when it is determined that optimization is needed, the initial operation solution and related data are input into the preset operation solution optimization model. The model analyzes the solution and finds the key factors affecting the solution effect; for example, the model analyzes that the preferential intensity of the promotion activity in the initial operation solution is not attractive enough to users, resulting in the unmet expected growth of sales; the selection of the activity publicity channel is inappropriate, resulting in limited increase in customer flow; based on the analysis results of the model, specific optimization strategies are generated; for the problem of insufficient preferential intensity of the promotion activity, the model recommends increasing the discount rate, launching full reduction activities or gift strategies; for the publicity channel problem, the model recommends adjusting the publicity channel combination and increasing the publicity investment on the social media platforms where the target user group is active. The initial operation solution is adjusted according to the generated optimization strategies to obtain an optimized operation solution.

[0123] Furthermore, when the effect score value is less than the preset third threshold, the activity sequence and activity time of the initial operation solution are optimized through the preset operation solution optimization model in combination with the non-core interests in the interest graph to obtain an optimized operation solution, including:

[0124] S901. When the effect score value is less than the preset third threshold, in combination with the non-core interests in the interest graph, a candidate solution is generated through a preset strategy generation model;

[0125] S902. According to the interest graph, the activities in the initial operation solution and the candidate solution are sorted to obtain the activity sequence;

[0126] S903. Based on the activity sequence, the activity time is optimized through the preset operation solution optimization model to obtain an optimized operation solution.

[0127] In this embodiment, when the effect score value is less than the preset third threshold, it indicates that the effect of the operation plan does not meet the expectation. Then, there are deficiencies in the excavation and satisfaction of the user's interests in the original plan. The interest map includes core interests and non-core interests. The non-core interests also reflect the potential needs of users. Analyze these non-core interests through a preset strategy generation model, and use the preset rules and algorithms in the model, combined with factors such as market dynamics and user behavior patterns, to generate candidate plans that can better attract users and improve the plan effect; the preset strategy generation model includes, but is not limited to, a genetic algorithm model. Use a large amount of historical data to train the genetic algorithm model to obtain a pre-trained strategy generation model. The model evaluates the activity effect of the corresponding activity in combination with non-core interests according to mechanisms such as inheritance, mutation, and selection, and generates candidate plans by mining non-core interests, providing new ideas and directions for optimizing the operation plan, expanding the coverage of operation activities, meeting the more diverse needs of users, and avoiding ignoring the potential interest points of users by only carrying out activities around core interests.

[0128] Specifically, according to the interest map, sort the activities in the initial operation plan and the candidate plan to obtain the activity order, which reflects the degree of attention of users to different interests and the correlation relationship between interests. Sort the activities according to the interest map, so that the activities are carried out in turn according to the priority and correlation logic of user interests, improving user participation and activity effect; through reasonable activity sorting, the coherence and attractiveness of operation activities can be improved, and activities are carried out according to the user interest logic, making it easier for users to participate, and improving the user's attention and participation enthusiasm for the activities.

[0129] Furthermore, according to the sorted activity order, optimize the activity time through a preset operation plan optimization model to obtain an optimized operation plan. The preset operation plan optimization model can be a decision tree model, a reinforcement learning model, etc. The operation plan optimization model in this embodiment is specifically a particle swarm optimization model. Use a large amount of historical activity data to train the particle swarm optimization model to obtain a pre-trained operation plan optimization model. According to the previously determined activity order, analyze the characteristics and target user groups of each activity, input the data into the preset operation plan optimization model, and the model optimizes and adjusts the activity time to search for the optimal solution to obtain an optimized operation plan. By optimizing the activity time, user participation and activity effect can be improved. At the same time, the reasonable arrangement of activity time also helps the smooth connection between activities, improves operation efficiency, and thus realizes the intelligent optimization of store operation, bringing better economic benefits to the store.

[0130] Embodiment 2:

[0131] In this embodiment, as Figure 4, a store operation intelligent optimization system is provided to implement the described store operation intelligent optimization method, including:

[0132] A data acquisition module that acquires the behavior data of users in the store;

[0133] A user grading module that analyzes the cognitive level of users through a preset dynamic user stratification model based on the behavior data to obtain multiple user levels;

[0134] An interest graph construction module that analyzes the interest tags of users based on the behavior data, analyzes the interest frequencies of each user level through a preset user interest graph construction model, and constructs an interest graph corresponding to each user level. Among them, the user level in the interest graph is the root node, the interest tag is the child node, and the corresponding root node and child node are connected through the interest frequency;

[0135] An operation plan generation module that screens out the core interests from the interest graph, maps them in a preset activity template library through a preset operation plan generation model, and generates a corresponding initial operation plan;

[0136] An operation plan evaluation module that evaluates the effect of the initial operation plan through a preset plan effect evaluation model to obtain the plan effect;

[0137] An operation plan optimization module that optimizes the activity sequence and activity time of the initial operation plan according to the plan effect through a preset operation plan optimization model, combined with the non-core interests in the interest graph, to obtain an optimized operation plan to achieve intelligent optimization of store operations.

[0138] In this embodiment, the data acquisition module includes components such as a data collection interface and a data storage unit. This module is responsible for collecting various behavior data of users in the store, providing a data basis for user analysis, operation plan formulation, etc. By integrating data from different sources, the comprehensiveness and accuracy of the data are ensured so that the system can comprehensively understand the behavior performance of users in the store; the user grading module includes a data preprocessing unit, a stratification model calculation unit, and a result output unit. This module analyzes the cognitive level of users according to the behavior data, using a preset dynamic user stratification model, and divides users into multiple different levels, so that more targeted operation strategies can be formulated for users at different levels, improving the operation effect and resource utilization efficiency.

[0139] Specifically, the interest graph construction module includes an interest tag extraction unit, a frequency calculation unit, and a graph construction unit. This module constructs an interest graph that reflects the interest characteristics of different user levels through the analysis of behavior data, so as to formulate personalized operation plans according to the interest distribution and preference degree of users; the operation plan generation module includes an interest weight screening unit, an activity mapping unit, and a plan generation unit. This module generates an initial operation plan that conforms to the user's interest characteristics according to the interest graph by using a preset operation plan generation model, enabling the operation activities to accurately target the core interests of users and improving user participation and operation effects.

[0140] Specifically, the operation plan evaluation module includes an evaluation index setting unit, a data collection unit, and an effect evaluation unit. This module objectively and comprehensively evaluates the implementation effect of the initial operation plan through a preset plan effect evaluation model, providing a direction for plan optimization to timely discover the problems and deficiencies in the operation plan; the operation plan optimization module includes a candidate plan generation unit, an activity sorting unit, and a time optimization unit. This module optimizes the initial operation plan according to the evaluation results of the operation plan, fully considering the non-core interests in the interest graph, generating a more comprehensive and reasonable operation plan, and improving the overall effect of the operation plan by optimizing the activity order and time, realizing the intelligent optimization of store operation and improving the operation efficiency of the store.

[0141] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent optimization method for store operation, characterized in that, Including: Obtaining the behavior data of users in the store; According to the behavior data, analyzing the cognitive level of users through a preset dynamic user stratification model to obtain multiple user levels; Based on the behavior data, analyzing the interest tags of users, and analyzing the interest frequency of each user level through a preset user interest map construction model to construct an interest map corresponding to each user level. Among them, the user level in the interest map is the root node, the interest tag is the child node, and the interest frequency is used as the weight of the edge to connect the corresponding root node and child node. The interest frequency is calculated by weighted summation of the operation frequency and operation duration of the user for the interest tag. The interest map is updated according to the change value of the interest frequency to reflect the interest change of the user in real time; Screening out the core interests from the interest map, and mapping them in a preset activity template library through a preset operation plan generation model to generate a corresponding initial operation plan; Evaluating the effect of the initial operation plan through a preset plan effect evaluation model to obtain the plan effect; According to the plan effect, through a preset operation plan optimization model, combining the non-core interests in the interest map to optimize the activity sequence and activity time of the initial operation plan to obtain an optimized operation plan to achieve intelligent optimization of store operation.

2. The intelligent optimization method for store operation according to claim 1, wherein The step of analyzing the cognitive level of users through a preset dynamic user stratification model according to the behavior data to obtain multiple user levels includes: Dividing multiple age intervals according to the user age in the behavior data along the time axis to obtain age groups; According to the age groups, analyzing the cognitive level of each group of users to obtain the corresponding cognitive index value of each group of users; Combining the cognitive index values, merging according to the inter-group similarity being greater than a preset similarity threshold, and splitting according to the standard deviation of the cognitive index values within the group being greater than a preset standard deviation threshold. Dynamically adjusting the age groups through a preset dynamic user stratification model to obtain multiple user levels.

3. The intelligent optimization method for store operation according to claim 2, wherein The step of dividing multiple age intervals according to the user age in the behavior data along the time axis to obtain age groups includes: Calculating the corresponding age density according to the user age in the behavior data; Generating multiple seed points according to the age density; According to the seed points, dividing the time axis into multiple age intervals to obtain initial age intervals, where each seed point corresponds to an initial age interval; Setting an overlapping area between every two adjacent initial age intervals according to a preset interval length; Allocating according to the user density of the overlapping area to the corresponding initial age interval to obtain age groups.

4. The intelligent optimization method for store operation according to claim 1, characterized in that The step of analyzing the interest tags of users based on the behavior data, and analyzing the interest frequency of each user level through a preset user interest map construction model to construct an interest map corresponding to each user level. Among them, the user level in the interest map is the root node, the interest tag is the child node, and the corresponding root node and child node are connected through the interest frequency, includes: Extracting keywords through a preset keyword extraction model according to the behavior data to obtain high-frequency behavior keywords; Map the high-frequency behavior keywords to corresponding tags through a preset keyword mapping model to obtain the user's interest tags; Based on the interest tags and combined with the user hierarchy, analyze the interest frequencies of each user hierarchy through a preset user interest graph construction model, and construct an interest graph corresponding to each user hierarchy.

5. The intelligent optimization method for store operation according to claim 4, characterized in that, The step of analyzing the interest frequencies of each user hierarchy through a preset user interest graph construction model based on the interest tags and combined with the user hierarchy to construct an interest graph corresponding to each user hierarchy includes: Based on the user hierarchy corresponding to the interest tags, obtain the interest frequencies by analyzing the operation frequencies and durations of users for different interest tags; Through a preset user interest graph construction model, use the user hierarchy as the root node, the interest tags as the child nodes, and the interest frequencies as the weights of the edges to connect the corresponding root nodes and child nodes to obtain an initial interest graph; Combined with real-time user data, when the change value of the interest frequency within the same user hierarchy is greater than a preset first threshold, update the initial interest graph to obtain an interest graph corresponding to each user hierarchy.

6. The intelligent optimization method for store operation according to claim 1, wherein The step of screening out the core interests from the interest graph and mapping them through a preset operation plan generation model in a preset activity template library to generate a corresponding initial operation plan includes: Based on the weights of the edges in the interest graph, select the interest tags with interest frequency values greater than a preset second threshold as the core interests; Based on the core interests, map through a preset operation plan generation model and a preset activity template library to obtain corresponding operation activities; Optimize the operation activities through a preset activity optimization model to obtain a corresponding initial operation plan.

7. An intelligent optimization method for store operation according to claim 6, characterized in that, The step of optimizing the operation activities through a preset activity optimization model to obtain a corresponding initial operation plan includes: Based on the operation activities, arrange the activity sequence and duration through a preset activity arrangement model to obtain an initial activity arrangement; Within a preset time period, optimize the initial activity arrangement through a preset activity optimization model to obtain a corresponding initial operation plan.

8. The intelligent optimization method for store operation according to claim 1, wherein The step of optimizing the activity sequence and activity time of the initial operation plan according to the plan effect through a preset operation plan optimization model and combined with the non-core interests in the interest graph to obtain an optimized operation plan to achieve intelligent optimization of store operations includes: Calculate the effect score value of the initial operation plan according to the plan effect; When the effect score value is less than a preset third threshold, optimize the activity sequence and activity time of the initial operation plan through a preset operation plan optimization model combined with the non-core interests in the interest graph to obtain an optimized operation plan.

9. The intelligent optimization method for store operation according to claim 8, characterized in that, The step of optimizing the activity sequence and activity time of the initial operation plan through a preset operation plan optimization model combined with the non-core interests in the interest graph to obtain an optimized operation plan when the effect score value is less than a preset third threshold includes: When the effect score value is less than a preset third threshold, combine the non-core interests in the interest graph and generate candidate plans through a preset strategy generation model; Sort the activities in the initial operation plan and candidate plans according to the interest graph to obtain the activity sequence; Based on the activity sequence, optimize the activity time through a preset operation plan optimization model to obtain an optimized operation plan.

10. An intelligent optimization system for store operations, characterized in that, A method for intelligent optimization of store operations according to any one of claims 1 to 9, comprising: A data acquisition module that acquires the behavior data of users in the store; A user grading module that analyzes the cognitive level of users through a preset dynamic user stratification model according to the behavior data to obtain multiple user levels; An interest graph construction module that analyzes the interest tags of users based on the behavior data, analyzes the interest frequencies of each user level through a preset user interest graph construction model, and constructs an interest graph corresponding to each user level. Among them, the user level in the interest graph is the root node, the interest tag is the child node, and the corresponding root node and child node are connected through the interest frequency; An operation plan generation module that screens out the core interests from the interest graph and maps them in a preset activity template library through a preset operation plan generation model to generate a corresponding initial operation plan; An operation plan evaluation module that evaluates the effect of the initial operation plan through a preset plan effect evaluation model to obtain the plan effect; An operation plan optimization module that optimizes the activity sequence and activity time of the initial operation plan according to the plan effect, through a preset operation plan optimization model, in combination with the non-core interests in the interest graph, to obtain an optimized operation plan to achieve intelligent optimization of store operations.

Images