Business online distribution strategy making method and system based on large model, and storage medium

Abstract

The invention relates to the technical field of data processing, and discloses a business online distribution strategy making method and system based on a large model and a storage medium, and the method comprises the steps: constructing the large model, enabling the large model to be in communication connection with a business line, operating the large model to obtain business update data, and making a distribution strategy. The system and the storage medium both correspond to the method. According to the application, the business line real-time data is automatically tracked and compared through a large model, information in multiple aspects such as products, markets, customers, competitors and channels is covered, the limitation of manual data processing in traditional distribution strategy making is effectively solved, data changes are quickly perceived to generate business update data, and the efficiency is improved. And the distribution strategy is reorganized and formulated based on the feature extraction and the distributed sales layout, so that the efficiency, the accuracy and the scientificity of the formulation of the distribution strategy are greatly improved, and an enterprise can more flexibly adapt to the market change, optimize the resource allocation and the channel layout and enhance the online distribution competitiveness.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and specifically to a method, system, and storage medium for formulating an online business distribution strategy based on a large model. Background Art

[0002] In today's highly competitive business environment, the formulation of an online business distribution strategy is crucial for the success of an enterprise. When the inventor was researching the traditional distribution strategy formulation, it was found that the existing distribution strategy formulation often relies on manual experience analysis and has many limitations, which are specifically manifested in the following aspects:

[0003] 1. It is difficult for humans to comprehensively and accurately process a large amount of business data, including product-related information, market dynamics, changes in customer needs, and competitor actions, etc., which easily leads to information omission or analysis deviation. For example, in the face of complex market and customer data, manual analysis may not be able to promptly capture subtle but critical changes in consumption trends, thus missing the best opportunity to adjust the distribution strategy.

[0004] 2. The traditional method of formulating a distribution strategy is inefficient and time-consuming, and it cannot quickly adapt to the rapidly changing market environment. When sudden changes occur in the market, such as a competitor suddenly launching a highly competitive new product or opening up a new distribution channel, the distribution strategy formulated manually is difficult to react and optimize in a timely manner.

[0005] 3. The traditional method lacks systematicness and scientificity, and often lacks accurate basis in channel resource allocation and layout adjustment, which may cause resource waste or unreasonable allocation, affecting the market coverage and sales performance of the enterprise.

[0006] In summary, there is an urgent need for a new technical solution for formulating an online business distribution strategy based on a large model to enhance the competitiveness and adaptability of enterprises in online business distribution. Summary of the Invention

[0007] The purpose of the present application is to provide a method, system, and storage medium for formulating an online business distribution strategy based on a large model to solve the technical problems raised in the above background art.

[0008] To achieve the above purpose, the present application discloses the following technical solutions:

[0009] In the first aspect, the present application discloses a method for formulating an online business distribution strategy based on a large model, and the method includes:

[0010] S1: Construct a large model for formulating a distribution strategy and communicatively connect the large model with the business line; wherein, the large model is constructed based on historical business data and the corresponding historical distribution strategies;

[0011] S2: Run the large model to obtain business update data, and formulate a distribution strategy based on the business update data; the business update data at least includes updated product-related data, market and customer data, competitor data, and channel data; formulating the distribution strategy based on the business update data means: extracting features from the business update data, and reorganizing the distributed sales layout based on the results of feature extraction to obtain the corresponding distribution strategy, where the distributed sales layout is used to represent the resource distribution method and channel structure of distribution.

[0012] Preferably, the construction process of the large model includes:

[0013] Collect and analyze the historical business data and the corresponding historical distribution strategies to generate corresponding strategy generation relationships; among them, the strategy generation relationship is used to represent the corresponding relationship between business data and distribution strategies;

[0014] Store the strategy generation relationship in a preset large model, and use the historical business data and the corresponding historical distribution strategies to train the large model to formulate distribution strategies based on business data and combined with the strategy generation relationship.

[0015] Preferably, the collection process of the historical business data and the corresponding historical distribution strategies includes:

[0016] Collect corresponding business data from different data sources on the business line;

[0017] Screen the collected business data corresponding to the distribution strategy, and save the combination of the business data and the corresponding distribution strategy.

[0018] Preferably, the generation process of the strategy generation relationship includes:

[0019] Preferably, the generation process of the strategy generation relationship includes:

[0020] Obtain the combined and saved business data, the corresponding distribution strategy, and the influence factor of the change in business data on the distribution strategy, where the influence factor is extracted based on the influence of the change in historical business data on the distribution strategy;

[0021] Use feature extraction technology to extract the business features and strategy features of the business data and the corresponding distribution strategy, and correspondingly generate the strategy generation relationship between the business features and the strategy features;

[0022] The strategy generation relationship is specifically:

[0023]

[0024] Among them, there are n pieces of data in the business data D and m strategies in the distribution strategy S. For the business data, its business feature vector is extracted and there are k eigenvalues; for the distribution strategy, its strategy feature vector is extracted and there are l eigenvalues. Based on finding the maximum value, the corresponding strategy generation relationship r is obtained i,j , and α is the influence factor of the change in business data on the distribution strategy.

[0025] Preferably, the process of obtaining the business update data includes:

[0026] Using the large model to track the real-time data of the business line and conduct real-time comparison. When there is data update, the corresponding business update data is generated;

[0027] The process of generating the business update data includes:

[0028] Comparing the historical product-related data with the real-time generated related data. When there is data change, the corresponding updated product-related data is generated and output;

[0029] Comparing the historical market and customer data with the real-time market and customer data. When there is data change, the corresponding updated market and customer data is generated and output;

[0030] Comparing the historical competitor data with the real-time competitor data. When there is data change, the corresponding updated competitor data is generated and output;

[0031] Comparing the historical channel data with the real-time channel data. When there is data change, the corresponding updated channel data is generated and output;

[0032] Defining any one or more of the output updated product-related data, market and customer data, competitor data, and channel data as business update data and then outputting.

[0033] Preferably, the feature extraction of the business update data is specifically as follows:

[0034] Obtain and parse the business update data;

[0035] Perform dimensionality reduction processing on the business update data to obtain low-dimensional data, and perform feature extraction on the low-dimensional data to obtain business update features.

[0036] Preferably, the recombination of the distributed sales layout based on the results of feature extraction is specifically to perform the following steps:

[0037] A1: Based on the strategy generation relationship, match the distribution strategy corresponding to the business update features;

[0038] A2: Adjust the distribution ratio of channel resources based on updated market and customer data;

[0039] A3: Re-plan the channel structure based on updated market and customer data, and determine the direction of channel expansion or contraction;

[0040] A4: Add new types of sales channels or optimize the existing channel portfolio based on updated product-related data;

[0041] A5: Evaluate the differences in customer needs in different regions based on updated market and customer data, and re-allocate sales human and material resources to the corresponding regions;

[0042] A6: When based on updated competitor data, make differential layout adjustments to the channels with a high overlap with competitors;

[0043] A7: Generate a new distributed sales layout based on the adjustments in steps A2 to A6;

[0044] In step A1, it is carried out based on a strategy update formula, where the specific strategy update formula is:

[0045] S 0 =W 1 x 1 +W 2 x 2 +W 3 x 3

[0046] Among them, S 0 represents the preliminary basic distribution strategy, x 1 represents the updated market and customer data, x 2 represents the updated product-related data, x 3 represents the updated competitor data, W 1 、W 2 、W 3 are respectively the strategy generation relationship matrices corresponding to the market and customer data, product-related data, and competitor data obtained through learning and training based on historical business data and strategies.

[0047] Preferably, the obtaining of the corresponding distribution strategy is specifically:

[0048] Merge the reorganized distributed sales layout into the distribution strategy obtained in step A1 and then output;

[0049] Determine the corresponding strategy execution department based on the reorganized distributed sales layout, and release the corresponding distribution strategy to this strategy execution department for execution.

[0050] Second aspect, the present application discloses a system for formulating an online distribution strategy for a business based on a large model. This system is applicable to the method for formulating an online distribution strategy for a business based on a large model as described above. The system includes:

[0051] A large model construction module, which is configured to: construct a large model for formulating a distribution strategy and communicatively connect the large model to the business line; wherein, the large model is constructed based on historical business data and corresponding historical distribution strategies;

[0052] A large model operation module, which is configured to: run the large model to obtain business update data and formulate a distribution strategy based on the business update data; the business update data at least includes product-related data with updates, market and customer data, competitor data, and channel data; formulating the distribution strategy based on the business update data means: extracting features from the business update data and reorganizing the distributed sales layout based on the results of feature extraction to obtain the corresponding distribution strategy, and the distributed sales layout is used to represent the resource distribution method and channel structure of distribution;

[0053] The large model construction module is communicatively connected to the large model operation module.

[0054] Third aspect, the present application discloses a storage medium, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the method for formulating an online distribution strategy for a business based on a large model as described above.

[0055] Advantageous effects: The method, system, and storage medium for formulating an online distribution strategy for a business based on a large model of the present application utilize a large model constructed based on historical business data and corresponding historical distribution strategies, and in combination with its communicative connection to the business line, achieve efficient acquisition and accurate analysis of business update data. Among them, the large model automatically tracks real-time data of the business line and compares them, covering various aspects of information such as products, markets, customers, competitors, and channels, effectively solving the limitations of manual data processing in traditional distribution strategy formulation, quickly detecting data changes to generate business update data, and then formulating a distribution strategy based on feature extraction and reorganization of the distributed sales layout, greatly improving the efficiency, accuracy, and scientific nature of distribution strategy formulation, enabling enterprises to more flexibly adapt to market changes, optimize resource allocation and channel layout, and enhance online distribution competitiveness. Description of the Drawings

[0056] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those skilled in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0057] Figure 1 It is a flowchart of a method for formulating an online distribution strategy for a business based on a large model provided by an embodiment of the present application;

[0058] Figure 2 It is a block diagram of the structure of a system for formulating an online distribution strategy for a business based on a large model provided by an embodiment of the present application. Detailed implementation manners

[0059] The following will clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of them. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0060] In this article, the term "including" is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such a process, method, article or device. Without more limitations, the elements defined by the statement "including..." do not exclude the existence of additional identical elements in the process, method, article or device including the said elements.

[0061] The first aspect of this embodiment discloses a method for formulating an online distribution strategy for a business based on a large model as shown in Figure 1 The method includes:

[0062] S1: Construct a large model for formulating a distribution strategy and communicatively connect the large model with the business line; wherein, the large model is constructed based on historical business data and the corresponding historical distribution strategies;

[0063] S2: Run the large model to obtain business update data, and formulate a distribution strategy based on the business update data; the business update data at least includes updated product-related data, market and customer data, competitor data, and channel data; formulating a distribution strategy based on the business update data is: extracting features from the business update data, and reorganizing the distributed sales layout based on the results of the feature extraction to obtain the corresponding distribution strategy, and the distributed sales layout is used to represent the resource distribution method and channel structure of the distribution.

[0064] Through the above, this embodiment utilizes a large model constructed based on historical business data and corresponding historical distribution strategies, and combines its communication connection with the business line to achieve efficient acquisition and accurate analysis of business update data. Among them, the large model automatically tracks and compares real-time data of the business line, covering various aspects of information such as products, markets, customers, competitors, and channels, effectively solving the limitations of manual data processing in traditional distribution strategy formulation, quickly detecting data changes to generate business update data, and then formulating distribution strategies based on feature extraction and reorganization of distributed sales layouts, greatly improving the efficiency, accuracy, and scientific nature of distribution strategy formulation, enabling enterprises to more flexibly adapt to market changes, optimize resource allocation and channel layouts, and enhance online distribution competitiveness.

[0065] Specifically, the construction process of the large model includes:

[0066] Collect and analyze historical business data and corresponding historical distribution strategies to generate corresponding strategy generation relationships; among them, the strategy generation relationship is used to represent the corresponding relationship between business data and distribution strategies;

[0067] Store the strategy generation relationship in a preset large model, and use the historical business data and corresponding historical distribution strategies to train the large model to formulate distribution strategies based on business data and combined with the strategy generation relationship.

[0068] Through the above, this embodiment stores and trains the strategy generation relationship generated by collecting and analyzing historical business data and strategies in the large model, realizing that the large model intelligently formulates distribution strategies based on the corresponding relationship between business data and strategies. Let the large model learn the appropriate distribution strategy models under different business data, so that it can more accurately match and generate effective strategies when facing new business data, improving the accuracy and reliability of strategy formulation, reducing the mistakes of manual experience judgment, providing a scientific and reasonable decision-making basis for enterprise online distribution, and helping enterprises better explore the market and meet customer needs. It should be noted that this embodiment uses existing large model technologies to implement the construction of the large model.

[0069] Specifically, the collection process of historical business data and corresponding historical distribution strategies includes:

[0070] Collect corresponding business data from different data sources on the business line;

[0071] Screen the collected business data corresponding to the distribution strategies, and save the combination of the business data and the corresponding distribution strategies.

[0072] Through the above, this embodiment collects and filters business data corresponding to distribution strategies from multiple data sources on the business line, achieving the construction of a high-quality historical business data and strategy combination library. It ensures that the data used for large model training is targeted and relevant, avoiding interference from irrelevant data in model learning. Precise data collection and filtering help the large model capture the internal relationship between business data and distribution strategies more accurately, making the generated strategy generation relationship more in line with the actual business scenario, improving the effectiveness of the large model in formulating distribution strategies, and ensuring the adaptability and feasibility of the enterprise's online distribution strategies. It can be understood that the combination of business data and corresponding distribution strategies from different data sources in this embodiment is based on the common knowledge of those skilled in the art, and the purpose of the combination is to provide an association basis for business data and distribution strategies for the generation of strategy generation relationships.

[0073] Specifically, the process of generating the strategy generation relationship includes:

[0074] Obtain the combined and saved business data, the corresponding distribution strategies, and the impact factor of changes in business data on distribution strategies, which is extracted based on the impact of changes in historical business data on distribution strategies;

[0075] Use feature extraction technology to extract the business features and strategy features of business data and corresponding distribution strategies, and correspondingly generate the strategy generation relationship between business features and strategy features;

[0076] The strategy generation relationship is specifically:

[0077]

[0078] Among them, there are n data in business data D and m strategies in distribution strategy S. For business data, extract its business feature vector And there are k eigenvalues; for distribution strategies, extract its strategy feature vector And there are l eigenvalues. Based on finding the maximum value, obtain the corresponding strategy generation relationship r i,j , and α is the impact factor of changes in business data on distribution strategies.

[0079] It can be understood that the impact of changes in business data on distribution strategies is different. Based on this, this embodiment quantifies the impact of changes in historical business data on distribution strategies to obtain the impact factor, and further optimizes the original strategy generation relationship based only on business features and strategy features based on this impact factor, so as to optimize the strategy generation relationship to provide more accurate data reference. Based on the maximum value calculation among them, the optimal matching of business data and distribution strategies is achieved, and a verification method is provided for the subsequent updated distribution strategies.

[0080] Through the above, this embodiment uses feature extraction technology to extract the features of business data and distribution strategies to generate a strategy generation relationship, achieving the in-depth understanding and learning of the relationship between business data and strategies by the large model. Through precise feature extraction, the association pattern between business features and strategy features is clarified, enabling the large model to quickly locate the strategy corresponding to the similar feature combination when facing new business updated data, improving the speed and quality of strategy generation. This learning mechanism based on feature relationships enhances the intelligence and adaptability of the large model, providing more efficient and flexible support for the formulation of enterprise online distribution strategies. It should be noted that this embodiment uses existing feature extraction technology to extract the corresponding business features and strategy features, aiming to provide a computational basis for the reorganization of the distributed sales layout in the following text.

[0081] Specifically, the process of obtaining business updated data includes:

[0082] Using the large model to track the real-time data of the business line and conduct real-time comparison, and generating corresponding business updated data when there is data update;

[0083] The process of generating business updated data includes:

[0084] Comparing the historical product-related data with the real-time generated related data, and generating and outputting the corresponding updated product-related data when there is data change;

[0085] Comparing the historical market and customer data with the real-time market and customer data, and generating and outputting the corresponding updated market and customer data when there is data change;

[0086] Comparing the historical competitor data with the real-time competitor data, and generating and outputting the corresponding updated competitor data when there is data change;

[0087] Comparing the historical channel data with the real-time channel data, and generating and outputting the corresponding updated channel data when there is data change;

[0088] Defining any one or more of the output updated product-related data, market and customer data, competitor data, and channel data as business updated data and then outputting.

[0089] Through the above, this embodiment utilizes the function of the large model to track and compare real-time data of business lines, achieving the timely and accurate acquisition of business update data. Whether it is changes in product-related, market and customer, competitor, or channel data, the large model can quickly detect them and generate corresponding update data. This overcomes the drawback that the traditional method is difficult to capture market changes in a timely manner, enabling enterprises to master market dynamic information in the first place, providing a strong basis for quickly adjusting the distribution strategy, ensuring that enterprises maintain a sensitive response ability in the highly competitive online market, and seizing the market initiative. It can be understood that this embodiment uses existing data comparison technologies to compare historical business data and real-time business data, aiming to achieve real-time tracking of business data and provide accurate update references for formulating distribution strategies based on business lines.

[0090] Specifically, the feature extraction of business update data is as follows:

[0091] Obtain and parse business update data;

[0092] Perform dimensionality reduction processing on the business update data to obtain low-dimensional data, and perform feature extraction on the low-dimensional data to obtain business update features.

[0093] Through the above, this embodiment utilizes the dimensionality reduction processing and feature extraction of business update data to achieve the transformation of complex business update data into key business update features that can be used to formulate distribution strategies. Dimensionality reduction processing reduces data complexity, and feature extraction focuses on key information, enabling the large model to process data more efficiently and accurately grasp the core key points of the data. This helps to improve the accuracy and efficiency of subsequent feature matching strategies and reorganizing sales layouts, enabling enterprises to make more reasonable decisions based on key business changes more accurately when formulating distribution strategies, and optimizing resource allocation and channel planning. It should be noted that as is well known to those skilled in the art, there is a large amount of redundancy in business update data. Using existing data dimensionality reduction technologies to perform dimensionality reduction on the original business update data provides a more refined data basis for subsequent data processing.

[0094] Specifically, and reorganize the distributed sales layout based on the results of feature extraction, specifically by performing the following steps:

[0095] A1: Based on the policy generation relationship, match the distribution strategy corresponding to the business update feature;

[0096] A2: Based on the updated market and customer data, adjust the distribution ratio of channel resources;

[0097] A3: Based on the updated market and customer data, re-plan the channel structure and determine the direction of channel expansion or contraction;

[0098] A4: Based on the updated product-related data, add new sales channel types or optimize the existing channel combination;

[0099] A5: Based on the updated market and customer data, evaluate the differences in customer needs in different regions, and reallocate the sales human and material resources to the corresponding regions;

[0100] A6: When based on the updated competitor data, make differential layout adjustments to the channels with a high overlap with competitors;

[0101] A7: Based on the adjustments in steps A2 to A6, generate a new distributed sales layout.

[0102] As a preferred implementation manner of this embodiment, this embodiment implements step A1 based on a policy update formula, where the policy update formula is specifically:

[0103] S 0 = W 1 x 1 + W 2 x 2 + W 3 x 3

[0104] Among them, S 0 represents the preliminary basic distribution strategy, x 1 represents the updated market and customer data (such as comprehensive quantitative indicators such as the market size change rate, customer preference change score, etc.), x 2 represents the updated product-related data (such as the product innovation degree score, the product function upgrade range, etc., quantitatively represented), x 3 represents the updated competitor data (such as quantifiable parameters such as the competitor market share change rate, the competitive product differentiation index, etc.), W 1 、W 2 、W 3 are respectively the policy generation relationship matrices corresponding to the market and customer data, product-related data, and competitor data, and are obtained through learning and training of historical business data and policies, and their dimensions are determined according to specific policy dimensions and data characteristics.

[0105] As a preferred implementation manner of this embodiment, this embodiment implements step A2 based on a proportional adjustment formula, where the proportional adjustment formula is specifically:

[0106]

[0107] Among them, y 1 = [y 1,1 , y 1,2 ,…y 1,n is the calculated channel resource distribution ratio, S 0,i is the i-th element corresponding to S 0 ki is an adjustment coefficient related to the allocation of the i-th channel resource fitted based on historical data and business logic, and in this embodiment, the total allocation is controlled to be equal to 1 based on the existing normalization technology.

[0108] As a preferred implementation manner of this embodiment, this embodiment is based on the channel reconstruction condition y 2 = f 2 (S 0 , x 1 ) to implement step A3, where the channel reconstruction condition is specifically:

[0109] A preset channel reconstruction threshold, when the result after comprehensively considering S 0 and x 1 is greater than or equal to the channel reconstruction threshold, channel reconstruction is performed, and this reconstruction includes at least expanding channels and shrinking channels. It should be noted that the method of comprehensively considering S 0 and x 1 can be but is not limited to the weight assignment calculation based on the common knowledge of those skilled in the art.

[0110] As a preferred implementation manner of this embodiment, this embodiment is based on the channel optimization condition y 3 = f 3 (S 0 , x 2 ) to implement step A4, where the channel optimization condition is specifically:

[0111] A preset channel optimization threshold, when the result after comprehensively considering S 0 and x 2 is greater than or equal to the channel optimization threshold, channel optimization is performed, and this reconstruction includes at least adding channels and deleting channels. It should be noted that the method of comprehensively considering S 0 and x 2 can be but is not limited to the weight assignment calculation based on the common knowledge of those skilled in the art. It can be understood that the channel reconstruction condition and the channel optimization condition of this embodiment are adjustments to the channels based on different business update data.

[0112] As a preferred implementation manner of this embodiment, this embodiment is based on the resource allocation formula to implement step A5, where the resource allocation formula is specifically:

[0113]

[0114] where y 4 is the regional resource allocation matrix, k 4,kl is the adjustment coefficient of the resource type l in region k fitted based on historical data and business logic, R k is the total amount of resources that can be allocated in region k, and this total amount of resources includes the total amount of human or material resources.

[0115] As a preferred implementation manner of this embodiment, this embodiment implements step A6 based on a difference adjustment formula, where the difference adjustment formula is specifically:

[0116] y 5 =[y 5,m =f 5 (S 0 ,x 3 )=S 0,m +k 5,m x 3

[0117] Where y 5 represents the differential layout of the overlapping channels with competitors, y 5,m represents the calculated differential adjustment parameter for the m-th overlapping channel with competitors, and k 5,m is the coefficient of differential adjustment for the m-th channel fitted based on historical data and business logic.

[0118] Based on steps A1 to A6, the final distribution strategy is determined using a strategy determination formula, where the strategy determination formula is specifically:

[0119]

[0120] Where V is a weight vector fitted based on historical data and business logic and is used to incorporate each part of the adjustment strategy into the final distribution strategy S, × represents element-wise multiplication operation, and Y is the new distributed sales layout obtained.

[0121] It should be noted that f 1 (S 0 ,x 1 ), f 2 (S 0 ,x 1 ), f 3 (S 0 ,x 2 ), f 4 (S 0 ,x 1 ) and f 5 (S 0 ,x 3 ) are functions for generating corresponding channel adjustment strategies based on different business update characteristics, and are all constructed based on the strategy generation relationship and the well-known business logic in the art. They can be used centrally in this embodiment or independently based on requirements.

[0122] Through the above, this embodiment realizes the accurate and automatic reorganization of the sales layout according to market changes by using a series of distributed sales layout adjustment steps based on policy-generated relationships and various types of updated data. From the adjustment of the distribution ratio of channel resources, structural planning, product channel optimization to regional resource allocation and differential layout to deal with competitors, the distributed sales layout is reshaped comprehensively and systematically. It enables the enterprise's distribution strategy to closely fit the changes in the market, products, customers and competitive situation, improves the rationality and effectiveness of the sales layout, and enhances the enterprise's competitiveness and market coverage ability in the online distribution channel. It should be noted that the feature extraction of the business update data in this embodiment corresponds to the feature extraction of the foregoing business data.

[0123] Specifically, the corresponding distribution strategy is obtained, specifically:

[0124] Output after incorporating the reorganized distributed sales layout into the distribution strategy obtained in step A1;

[0125] Determine the corresponding policy execution department based on the reorganized distributed sales layout, and publish the corresponding distribution strategy to this policy execution department for execution.

[0126] Through the above, this embodiment realizes the complete generation and effective execution of the distribution strategy by incorporating the reorganized distributed sales layout into the matching strategy and determining the execution department. It ensures that the formulated distribution strategy takes into account both the basic strategy corresponding to the business update characteristics and the optimized sales layout. The clear allocation of the policy execution department guarantees the rapid implementation of the strategy and avoids the disconnection between strategy formulation and execution. This coherent process from formulation to execution improves the efficiency and coordination of the enterprise's online distribution operation, and promotes the enterprise to achieve sales goals and strategic plans in market competition.

[0127] The second aspect of this embodiment discloses a Figure 2 business online distribution strategy formulation system based on a large model as shown. This system is applicable to the business online distribution strategy formulation method based on a large model as described above. The system includes:

[0128] A large model construction module, which is configured to: construct a large model for formulating a distribution strategy and communicate this large model with the business line; wherein, the large model is constructed based on historical business data and the corresponding historical distribution strategy;

[0129] Large model operation module, which is configured to: run a large model to obtain business update data, and formulate a distribution strategy based on the business update data; the business update data at least includes product-related data, market and customer data, competitor data, and channel data with updates; formulating a distribution strategy based on the business update data is to extract features from the business update data, and reorganize the distributed sales layout based on the results of feature extraction to obtain the corresponding distribution strategy, and the distributed sales layout is used to represent the resource distribution method and channel structure of distribution;

[0130] The large model construction module is communicatively connected to the large model operation module.

[0131] It should be noted that the system for formulating an online business distribution strategy based on a large model in this embodiment corresponds to the aforementioned method for formulating an online business distribution strategy based on a large model. Therefore, for the content not specifically described in the system for formulating an online business distribution strategy based on a large model in this embodiment, such as but not limited to function definition, working principle, and technical effect, etc., reference can be made to the aforementioned method for formulating an online business distribution strategy based on a large model, and this text will not elaborate here.

[0132] The third aspect of this embodiment discloses a storage medium, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the method for formulating an online business distribution strategy based on a large model as described above.

[0133] Similarly, it should be noted that the storage medium in this embodiment corresponds to the aforementioned method for formulating an online business distribution strategy based on a large model. Therefore, for the content not specifically described in the storage medium in this embodiment, such as but not limited to function definition, working principle, and technical effect, etc., reference can be made to the aforementioned method for formulating an online business distribution strategy based on a large model, and this text will not elaborate here.

[0134] In summary, the method, system, and storage medium for formulating an online business distribution strategy based on a large model in this embodiment utilize a large model constructed based on historical business data and corresponding historical distribution strategies, and in combination with its communication connection with the business line, achieve efficient acquisition and accurate analysis of business update data. Among them, the large model automatically tracks real-time data of the business line and compares them, covering multiple aspects of information such as products, markets, customers, competitors, and channels, effectively solving the limitations of manual data processing in traditional distribution strategy formulation, quickly detecting data changes to generate business update data, and then formulating a distribution strategy based on feature extraction and reorganization of the distributed sales layout, greatly improving the efficiency, accuracy, and scientific nature of distribution strategy formulation, enabling enterprises to more flexibly adapt to market changes, optimize resource allocation and channel layout, and enhance online distribution competitiveness.

[0135] In the embodiments provided in the present application, it should be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code, or any appropriate combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, microcontroller, microprocessor, or other electronic units designed to implement the functions described herein, or a combination thereof. For software implementation, part or all of the processes of the embodiments can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable storage medium or transmitted as one or more instructions or codes on a computer-readable storage medium. The computer-readable storage medium includes computer storage media and communication media, where the communication media includes any medium that facilitates the transfer of a computer program from one place to another. The storage media can be any available medium that can be accessed by a computer. The computer-readable storage medium can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM, or other optical disk storage, magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer.

[0136] Finally, it should be noted that the above are only the preferred embodiments of the present application and are not used to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for formulating business online distribution strategy based on a large model, characterized in that: The method includes: S1: construct a big model for formulating distribution strategies, and connect the big model to the business line for communication; wherein the big model is constructed based on historical business data and corresponding historical distribution strategies; S2: Run the large model to obtain business update data, and formulate a distribution strategy based on the business update data; the business update data at least includes updated product-related data, market and customer data, competitor data and channel data; the distribution strategy formulated based on the business update data is: extract features from the business update data, and reorganize the distributed sales layout based on the result of the feature extraction to obtain a corresponding distribution strategy, wherein the distributed sales layout is used to characterize the resource distribution method and channel structure of the distribution.

2. The method for formulating business online distribution strategy based on a large model according to claim 1 is characterized in that: The construction process of the large model includes: Collecting and analyzing the historical business data and the corresponding historical distribution strategies, and generating corresponding strategy generation relationships; wherein the strategy generation relationships are used to characterize the corresponding relationship between business data and distribution strategies; The strategy generation relationship is stored in a preset big model, and the big model is trained using the historical business data and the corresponding historical distribution strategy to formulate a distribution strategy based on the business data and in combination with the strategy generation relationship.

3. The method for formulating business online distribution strategy based on a large model according to claim 2, characterized in that: The process of collecting the historical business data and the corresponding historical distribution strategy includes: Collect corresponding business data from different data sources on the business line; The collected business data is screened according to the distribution strategy, and the business data is combined with the corresponding distribution strategy and then saved.

4. The method for formulating business online distribution strategy based on a large model according to claim 3 is characterized in that: The generation process of the strategy generation relationship includes: Obtain the business data and corresponding distribution strategies saved in the combination and the impact factor of the change of business data on the distribution strategy, the impact factor is extracted based on the impact of the change of historical business data on the distribution strategy; Extracting business features and strategy features of business data and corresponding distribution strategies using feature extraction technology, and generating strategy generation relationships between the business features and the strategy features accordingly; The strategy generation relationship is specifically: Among them, there are n data in the business data D, and m strategies in the distribution strategy S. For the business data, extract its business feature vector And there are k eigenvalues; for the distribution strategy, extract its strategy eigenvector And there are l eigenvalues, based on finding the maximum value, the corresponding strategy generation relationship r is obtained i,j , and α is the impact factor of changes in business data on distribution strategies.

5. The method for formulating business online distribution strategy based on a big model according to claim 1, characterized in that: The process of acquiring the service update data includes: The large model is used to track the real-time data of the business line and perform real-time comparison, and when there is data update, corresponding business update data is generated; The process of generating the service update data includes: Compare historical product-related data with real-time related data, and generate and output corresponding updated product-related data when there is data change; Compare historical market and customer data with real-time market and customer data, and generate and output corresponding updated market and customer data when there are data changes; Compare historical competitor data with real-time competitor data, and generate and output corresponding updated competitor data when there are data changes; Compare historical channel data with real-time channel data, and generate and output corresponding updated channel data when there is data change; Any one or more of the updated product-related data, market and customer data, competitor data, and channel data outputted is defined as business update data and then outputted.

6. The method for formulating business online distribution strategy based on a large model according to claim 2, characterized in that: The feature extraction of the business update data is specifically as follows: Acquire and parse the business update data; The business update data is subjected to dimensionality reduction processing to obtain low-dimensional data, and features are extracted from the low-dimensional data to obtain business update features.

7. The method for formulating business online distribution strategy based on a large model according to claim 6, characterized in that: The distributed sales layout is reorganized based on the result of feature extraction, specifically by performing the following steps: A1: Generate relationships based on the strategy and match distribution strategies corresponding to business update characteristics; A2: Adjust the distribution ratio of channel resources based on updated market and customer data; A3: Re-plan the channel structure based on updated market and customer data and determine the direction of channel expansion or contraction; A4: Add new sales channel types or optimize existing channel combinations based on updated product-related data; A5: Based on updated market and customer data, evaluate the differences in customer needs in different regions and reallocate sales manpower and material resources to the corresponding regions; A6: Based on updated competitor data, differentiated layout adjustments are made to channels that have high overlap with competitors; A7: Generate a new distributed sales layout based on the adjustments in steps A2 to A6; In step A1, the strategy is updated based on a formula, wherein the strategy update formula is specifically: S0=W1x1+W2x2+W3x3 Among them, S0 represents the preliminary basic distribution strategy, x1 represents the updated market and customer data, x2 represents the updated product-related data, x3 represents the updated competitor data, W1, W2, and W3 are the strategy generation relationship matrices of the corresponding market and customer data, product-related data, and competitor data obtained based on the learning and training of historical business data and strategies.

8. The method for formulating business online distribution strategy based on a big model according to claim 7 is characterized in that: The corresponding distribution strategy is specifically: Incorporate the reorganized distributed sales layout into the distribution strategy obtained in step A1 and output it; Based on the reorganized distributed sales layout, the corresponding strategy execution department is determined, and the corresponding distribution strategy is issued to the strategy execution department for execution.

9. A business online distribution strategy formulation system based on a big model, the system is applicable to the business online distribution strategy formulation method based on a big model as described in any one of claims 1 to 8, characterized in that: The system includes: A big model building module, wherein the big model building module is configured to: build a big model for formulating a distribution strategy, and connect the big model to the business line for communication; wherein the big model is built based on historical business data and corresponding historical distribution strategies; A large model operation module, wherein the large model operation module is configured to: operate the large model to obtain business update data, and formulate a distribution strategy based on the business update data; the business update data at least includes updated product-related data, market and customer data, competitor data and channel data; the formulation of a distribution strategy based on the business update data is: extracting features from the business update data, and reorganizing the distributed sales layout based on the feature extraction results to obtain a corresponding distribution strategy, wherein the distributed sales layout is used to characterize the resource distribution mode and channel structure of the distribution; The large model building module is communicatively connected with the large model running module.

10. A storage medium, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method for formulating business online distribution strategy based on a large model as described in any one of claims 1 to 8.

Images