CN118674479A - Marketing strategy intelligent generation method and device, electronic equipment and storage medium - Google Patents

Abstract

The present invention provides a marketing strategy intelligent generation method, device, electronic device and storage medium, which relates to the field of artificial intelligence technology, including: obtaining communication records of target customers and product information of products to be recommended; performing label mining on the communication records based on a large language model to generate label information of the target customers; performing label mining on the product information based on a large language model to generate label information of the products to be recommended; performing information fusion on the label information of the target customers and the label information of the products to be recommended based on a large language model to generate a marketing strategy for recommending the products to be recommended to the target customers. The method and device provided by the present invention can accurately locate the customer needs of different customers, recommend products that better meet customer needs, improve marketing efficiency, and also improve customer experience.

Description

Marketing strategy intelligent generation method, device, electronic device and storage medium

Technical Field

The present invention relates to the field of artificial intelligence technology, and in particular to a marketing strategy intelligent generation method, device, electronic device and storage medium.

Background Art

In scenarios such as finance and e-commerce, the market environment is highly competitive, and account managers need to use all available resources and tools to improve marketing efficiency and effectiveness.

In related technologies, account managers are faced with massive amounts of product information and customer data. They are unable to accurately and comprehensively understand the characteristics of different products, nor are they able to deeply analyze customers' consumption habits and preferences. This makes it impossible for account managers to accurately identify customer needs, resulting in low marketing efficiency and poor customer experience.

Summary of the invention

The present invention provides a marketing strategy intelligent generation method, device, electronic device and storage medium, which are used to solve the technical problems of being unable to accurately locate customer needs and low marketing efficiency.

The present invention provides a marketing strategy intelligent generation method, comprising:

Obtain communication records with target customers and product information of products to be recommended;

Performing label mining on the communication records based on a large language model to generate label information of the target customer;

Performing label mining on the product information based on a large language model to generate label information of the product to be recommended;

Based on a large language model, information fusion is performed on the label information of the target customer and the label information of the product to be recommended, so as to generate a marketing strategy for recommending the product to be recommended to the target customer.

In some embodiments, the performing tag mining on the communication records based on a large language model to generate tag information of the target customer includes:

Performing tag mining on the communication records based on a large language model to generate real-time tag information of the target customer;

The real-time label information and historical label information of the target customer are fused based on a large language model to generate label information of the target customer.

In some embodiments, after fusing the real-time tag information and the historical tag information of the target customer based on the large language model to generate the tag information of the target customer, the method further includes:

Based on the tag information of the target customer, generating a customer profile of the target customer;

The customer portrait is analyzed based on a large language model to generate a customer analysis result of the target customer; the customer analysis result includes at least one of a product preference analysis result and a purchase intention analysis result of the target customer.

In some embodiments, after fusing the real-time tag information and the historical tag information of the target customer based on the large language model to generate the tag information of the target customer, the method further includes:

Based on the tag information of the target customer, generate a marketing opportunity portrait of the target customer;

The marketing opportunity portrait is analyzed based on a large language model to generate marketing opportunity recommendation information corresponding to the target customer.

In some embodiments, after performing label mining on the product information based on a large language model to generate label information of the product to be recommended, the method further includes:

Based on a large language model, the product information is compared with the product information of the associated products corresponding to the product to be recommended to generate product comparison information of the product to be recommended.

In some embodiments, the step of fusing the label information of the target customer and the label information of the product to be recommended based on a large language model to generate a marketing strategy for recommending the product to be recommended to the target customer includes: matching the label information of the target customer and the label information of the product to be recommended to obtain information on the degree of matching between the target customer and the product to be recommended;

Based on a large language model, the matching degree information, the customer analysis results of the target customers and the marketing opportunity recommendation information corresponding to the target customers are integrated to generate a marketing strategy for recommending the product to be recommended to the target customers.

In some embodiments, the method further comprises:

Building a product knowledge base containing product information of the product to be recommended;

Information is extracted from the product knowledge base based on a large language model to generate a product information base corresponding to the product knowledge base; and product knowledge answers corresponding to product knowledge questions are generated based on the large language model and the product information base.

The present invention provides a marketing strategy intelligent generation device, comprising:

The information acquisition module is used to obtain the communication records of target customers and the product information of the products to be recommended;

A customer tag mining module, used to perform tag mining on the communication records based on a large language model to generate tag information of the target customer;

A product label mining module, used to perform label mining on the product information based on a large language model to generate label information of the product to be recommended;

The strategy generation module is used to fuse the label information of the target customer and the label information of the product to be recommended based on a large language model, and generate a marketing strategy for recommending the product to be recommended to the target customer.

The present invention provides an electronic device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the method for intelligently generating marketing strategies is implemented when the processor executes the computer program.

The present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, and the computer program implements the marketing strategy intelligent generation method when executed by a processor.

The intelligent generation method, device, electronic device and storage medium of marketing strategy provided by the present invention obtain the communication records of target customers and the product information of the products to be recommended; perform label mining on the communication records based on a large language model to generate the label information of the target customers; perform label mining on the product information based on a large language model to generate the label information of the products to be recommended; perform information fusion on the label information of the target customers and the label information of the products to be recommended based on the large language model to generate a marketing strategy for recommending the products to be recommended to the target customers; on the one hand, label mining is performed on the communication records of the target customers and the product information of the products to be recommended through the large language model, and on the other hand, information fusion is performed on the mined label information, and the powerful computing power and learning ability of the large language model for complex features are utilized, so that data can be better understood and utilized, and useful information can be more effectively extracted from the data, which not only improves the accuracy of data mining such as communication records and product information, but also processes large-scale massive data, and finally automatically generates high-quality marketing strategies, which can accurately locate the customer needs of different customers, recommend products that better meet customer needs, improve marketing efficiency, and improve customer experience.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

FIG. 1 is one of the flow charts of the method for intelligently generating marketing strategies provided by the present invention.

FIG. 2 is a second flow chart of the method for intelligently generating marketing strategies provided by the present invention.

FIG3 is a schematic diagram of the structure of the marketing strategy intelligent generation device provided by the present invention.

FIG. 4 is a schematic diagram of the overall architecture of the marketing system provided by the present invention.

FIG. 5 is a schematic diagram of the component logic architecture of the marketing system provided by the present invention.

FIG. 6 is a schematic diagram of the structure of an electronic device provided by the present invention.

DETAILED DESCRIPTION

In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

It should be noted that the terms "first", "second", etc. in the present invention are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units or modules is not necessarily limited to those steps or units or modules that are clearly listed, but may include other steps or units or modules that are not clearly listed or inherent to these processes, methods, products or devices.

In the technical solution of the present invention, the collection, storage, use, processing, transmission, provision and disclosure of personal information involved are in compliance with the provisions of relevant laws and regulations, necessary confidentiality measures are taken, and do not violate public order and good morals.

Marketing is the process of transferring goods or services from producers to consumers. It is a series of activities carried out by enterprises or other organizations to meet consumer needs. Marketing personnel need to understand the preferences of customers (consumers), recommend suitable products to customers, and encourage customers to purchase products. As the amount of product information and customer data increases, account managers will use recommendation platforms in related technologies to conduct marketing.

The recommendation platforms in related technologies usually analyze and organize single data to achieve single functions, such as data regularization, customer profiling, recommendation sorting and result delivery. They lack in-depth mining and correlation mining of massive data, and rely on the business experience of account managers in the marketing process, which makes it impossible for account managers to accurately identify customer needs, resulting in low marketing efficiency and poor customer experience.

In order to solve the deficiencies in the related art, FIG1 is one of the flow charts of the marketing strategy intelligent generation method provided by the present invention. As shown in FIG1 , the method includes step 110 , step 120 , step 130 and step 140 .

Step 110: Obtain the communication records of the target customers and the product information of the products to be recommended.

Specifically, the execution subject of the marketing strategy intelligent generation method provided in the embodiment of the present invention is a marketing strategy intelligent generation device. The device can be implemented by software, such as a marketing strategy intelligent generation program or marketing assistant software running in a computer; it can also be a device that executes the marketing strategy intelligent generation method, such as a mobile terminal, a tablet computer, a desktop computer or a server.

The target customers are individuals or organizations to whom products (including goods or services) need to be recommended.

Communication records refer to text records or voice records generated during the communication process with customers. For example, text chat records can be obtained by communicating with customers through instant messaging software, email, etc., or voice chat records can be obtained by communicating with customers through voice calls or video calls. These records can all be used as communication records. It should be noted that before processing the personal information involved in the communication records, the customer's authorization must be obtained, and the processing and confidentiality must be carried out in accordance with the requirements of relevant laws and regulations.

The recommended product refers to the commodity or service recommended to the customer. Product information refers to information describing the basic identification, technical specifications and functional characteristics of the product. For example, the product information of the recommended product can be recorded in the form of a product manual.

Step 120: Perform tag mining on the communication records based on the large language model to generate tag information of the target customers.

Specifically, the Large Language Model (LLM) is a deep learning model trained on massive text data. It can not only generate natural language text, but also deeply understand the meaning of the text and handle various natural language tasks such as text summarization, question answering, translation, etc.

The communication records can be converted into text data, for example, the voice records can be transcribed by voice-to-text or optical character recognition (OCR) and other technologies to obtain the corresponding text data. The text data corresponding to the communication records is input into a large language model, so that the large language model automatically extracts keywords or tags from the text data to obtain the tag information of the target customer.

The label information of the target customers may include information on the target customers' product preferences, consumption habits, purchase intentions, etc., and may also include information on the target customers' industry background and occupational information, etc.

Step 130: Perform label mining on product information based on a large language model to generate label information of products to be recommended.

Specifically, the text data corresponding to the product information can be input into a large language model, so that the large language model automatically extracts keywords or tags from the text data to obtain tag information of the product to be recommended.

The label information of the product to be recommended may include information on the functional characteristics, applicable objects, product identification and sales price of the product to be recommended.

Step 140: Based on the large language model, information fusion is performed on the label information of the target customer and the label information of the product to be recommended, and a marketing strategy for recommending the product to be recommended to the target customer is generated.

Specifically, the label information of target customers and the label information of products to be recommended can be input into a large language model, so that the large language model can use its powerful computing power and ability to learn complex features to better understand and utilize the label information of target customers and the label information of products to be recommended, accurately locate customer needs, and generate a marketing strategy to recommend products to target customers.

Marketing strategy refers to the plans and actions developed and executed to achieve marketing goals, covering a wide range of activities and decisions to promote the sales of products or services and to establish and maintain customer relationships. Marketing strategy can specifically include information such as customer demand analysis, product competitiveness analysis, marketing timing analysis, marketing speech recommendations, and marketing pricing recommendations.

Large language models can be output in text form to provide prompts to account managers, or to generate corresponding to-do items or reminder texts.

The intelligent generation method of marketing strategies provided by the embodiment of the present invention obtains the communication records of target customers and the product information of the products to be recommended; performs label mining on the communication records based on a large language model to generate the label information of the target customers; performs label mining on the product information based on the large language model to generate the label information of the products to be recommended; performs information fusion on the label information of the target customers and the label information of the products to be recommended based on the large language model to generate a marketing strategy for recommending the products to be recommended to the target customers; on the one hand, the communication records of the target customers and the product information of the products to be recommended are subjected to label mining through the large language model, and on the other hand, the information fusion is performed on the label information obtained by mining, and the powerful computing power and learning ability of the large language model for complex features are utilized, so that the data can be better understood and utilized, and useful information can be more effectively extracted from the data, which not only improves the accuracy of data mining such as communication records and product information, but also processes large-scale massive data, and finally automatically generates high-quality marketing strategies, which can accurately locate the customer needs of different customers, recommend products that better meet customer needs, improve marketing efficiency, and improve customer experience.

It should be noted that each implementation of the present invention can be freely combined, the order can be changed, or it can be executed separately, and does not need to rely on or depend on a fixed execution order.

In some embodiments, label mining is performed on communication records based on a large language model to generate label information of target customers, including:

Conduct tag mining on communication records based on large language models to generate real-time tag information for target customers;

Based on a large language model, the real-time label information and historical label information of the target customer are fused to generate the label information of the target customer.

Specifically, the communication records can be chat records with target customers in the recent period of time, etc. The communication records are mined by tags based on the large language model to generate real-time tag information of the target customers. The real-time tag information can be used to represent the product preferences, consumption habits, purchase intentions, etc. of the target customers in the recent period of time.

In addition, in the marketing system or business system, historical tag information of the target customer can also be saved. The historical tag information can include inherent tag information generated according to the basic information of the target customer, and can also include tag information of the target customer in the past time period.

By fusing the real-time tag information and historical tag information of target customers through a large language model, we can understand the deep semantics and intrinsic connections in the real-time tag information and historical tag information, thereby generating tag information that can comprehensively reflect customer needs.

The intelligent generation method of marketing strategies provided in the embodiment of the present invention generates label information of target customers by fusing real-time label information and historical label information of target customers through a large language model, which can accurately locate customer needs, improve marketing efficiency, and improve customer experience.

In some embodiments, after the real-time tag information and historical tag information of the target customer are fused based on the large language model to generate the tag information of the target customer, the method further includes:

Generate customer profiles of target customers based on their label information;

The customer portrait is analyzed based on a large language model to generate customer analysis results for target customers; the customer analysis results include at least one of product preference analysis results and purchase intention analysis results for target customers.

Specifically, customer profiling usually refers to building customer information and feature models based on customer behavior, preferences, and data using machine learning and natural language processing technologies. This approach can help understand customer interests, behavior patterns, and personality traits, thereby providing support for personalized recommendations, marketing, and customer experience optimization.

Large language models or other algorithm models can be used to process the target customer's label information to generate a customer profile of the target customer. The customer profile can reflect the customer's consumption interests, consumption behavior patterns, and personality characteristics.

Use a large language model to analyze customer portraits, achieve deep mining of label information, extract more effective information, and generate customer analysis results for target customers. Customer analysis results include at least one of the target customers' product preference analysis results and purchase intention analysis results. Product preference analysis results are used to indicate information such as the degree of customer preference for various products. Purchase intention analysis results are used to indicate information such as the possibility of customers purchasing various products.

The customer analysis results can be displayed in the marketing system or business system to provide prompts to account managers, so that account managers can quickly and deeply understand the customers through the customer analysis results.

The intelligent generation method of marketing strategies provided by the embodiment of the present invention uses a large language model to analyze customer portraits and generate customer analysis results of target customers, so that account managers can understand customers without relying on business experience, thereby improving marketing efficiency and customer experience.

In some embodiments, after the real-time tag information and historical tag information of the target customer are fused based on the large language model to generate the tag information of the target customer, the method further includes:

Generate marketing opportunity profiles of target customers based on their tag information;

Analyze marketing opportunity portraits based on large-scale language models and generate marketing opportunity recommendation information corresponding to target customers.

Specifically, marketing timing refers to the time at a specific point in time or under certain conditions that is most conducive to conducting marketing activities to achieve the best results.

Large language models or other algorithm models can be used to process target customers' label information to generate marketing opportunity profiles for target customers. Marketing opportunity profiles can indicate information such as the distribution of time periods when customers may have the need to purchase products.

Use large language models to analyze marketing opportunity profiles, achieve deep mining of tag information, extract more effective information, and generate marketing opportunity recommendation information corresponding to target customers. Marketing opportunity recommendation information can be used to indicate the recommended marketing opportunities for target customers based on information such as customer purchasing habits, seasonal demand changes, and preference changes.

The intelligent generation method of marketing strategies provided by the embodiment of the present invention uses a large language model to analyze the marketing opportunity portrait and generates marketing opportunity recommendation information corresponding to the target customers, so that the account manager can choose the best time to market to the customer, thereby improving marketing efficiency.

In some embodiments, after performing label mining on product information based on a large language model to generate label information of the product to be recommended, the method further includes:

Based on a large language model, the product information is compared with the product information of the related products corresponding to the product to be recommended, and product comparison information of the product to be recommended is generated.

Specifically, the associated product corresponding to the product to be recommended may be a product that is in competition with the product to be recommended.

The product information of the product to be recommended and the product information of the associated products corresponding to the product to be recommended can be input into the large language model, so that the large language model can compare the product information of different products from multiple different dimensions to generate product comparison information of the product to be recommended.

In addition, the large-scale language model can also accurately capture the customer's intention for product comparison and the names of the products that need to be compared based on the previously input target customer's label information and multi-round conversation information, refine and display the same information and difference information of different products, and generate product comparison information that is more in line with customer needs.

The intelligent marketing strategy generation method provided by the embodiment of the present invention uses a large language model to compare product information with product information of related products corresponding to the product to be recommended, and generates product comparison information of the product to be recommended, so that customer managers can recommend suitable products to customers more accurately, thereby improving marketing efficiency.

In some embodiments, information fusion is performed on the tag information of the target customer and the tag information of the product to be recommended based on the large language model to generate a marketing strategy for recommending the product to be recommended to the target customer, including:

Match the target customer's label information with the label information of the product to be recommended to obtain the matching degree information between the target customer and the product to be recommended;

Based on a large language model, the matching degree information, the customer analysis results of the target customers and the marketing opportunity recommendation information corresponding to the target customers are integrated to generate a marketing strategy to recommend the products to be recommended to the target customers.

Specifically, a large language model or other algorithm model can be used to match the target customer's label information with the label information of the product to be recommended, and obtain the matching degree information between the target customer and the product to be recommended. The matching degree information is used to measure the matching degree of the product to be recommended to meet the customer's needs, and can be represented by a matching score. The higher the matching score, the more suitable the product to be recommended is for recommendation to the target customer; the lower the matching score, the less suitable the product to be recommended is for recommendation to the target customer.

The matching degree information, customer analysis results of target customers and marketing opportunity recommendation information corresponding to target customers are input into a large language model, which will understand and mine the matching degree information, customer analysis results and marketing opportunity recommendation information to generate more effective marketing strategies.

The marketing strategy intelligent generation method provided by the embodiment of the present invention utilizes a large language model to fuse matching degree information, customer analysis results of target customers, and marketing opportunity recommendation information corresponding to the target customers, thereby generating more effective and accurate marketing strategies and improving marketing efficiency.

In some embodiments, the method further comprises:

Build a product knowledge base containing product information of products to be recommended;

Extract information from the product knowledge base based on a large language model to generate a product information base corresponding to the product knowledge base;

Generate product knowledge answers corresponding to product knowledge questions based on large language models and product information databases.

Specifically, product information of multiple products including the product to be recommended can be collected to build a product knowledge base, and a large language model is used to extract information from the product knowledge base to generate a product information base corresponding to the product knowledge base.

The product knowledge base contains various types of data, most of which are unstructured data. The product information database obtains structured data after information extraction through a large language model, which is convenient for computer processing.

The product information database can be connected to a large language model to obtain a product question-answering system. Account managers can input product knowledge questions into the large language model, which will output product knowledge answers after understanding the questions and extracting the answers.

The marketing strategy intelligent generation method provided by the embodiment of the present invention can generate product knowledge answers corresponding to product knowledge questions through a large language model and a product information database, so that customer managers can quickly understand product information through knowledge questions and answers, thereby improving marketing efficiency.

Figure 2 is the second flow chart of the intelligent generation method of marketing strategies provided by the present invention. As shown in Figure 2, the method can be applied to banking systems, taking customer data as the entry point, integrating unstructured chat data such as instant messaging software and applications (APP), and converting it into text data through speech-to-text, OCR and other algorithm technologies, and importing it into a large language model. The large language model will perform real-time tag mining based on the content of the conversation, save the mined tags in a tag library, and generate basic customer portraits and marketing opportunity portraits in combination with the inherent customer tags of the banking system, which are used for customer analysis and display and actual marketing recommendations in the large language model respectively.

The large model also needs to extract labels from product information and establish a product information database. Then, the product recommendation system combines the customer label database with the product information database to make accurate product recommendations and calculate purchase rates.

The product information database and other private knowledge are deployed locally in the bank to realize the knowledge question and answer function of local bank data.

The method provided by the embodiment of the present invention combines customer, product and bank information, so that the account manager can quickly understand the customer and formulate the most reasonable marketing plan in combination with the recommendation plan of the big model.

The device provided by the present invention is described below. The device described below and the method described above can be referenced to each other.

FIG3 is a schematic diagram of the structure of a marketing strategy intelligent generation device provided by the present invention. As shown in FIG3 , the device includes:

The information acquisition module 310 is used to acquire the communication records of the target customers and the product information of the products to be recommended;

A customer tag mining module 320 is used to perform tag mining on communication records based on a large language model to generate tag information of target customers;

A product label mining module 330, for performing label mining on product information based on a large language model to generate label information of products to be recommended;

The strategy generation module 340 is used to fuse the label information of the target customer and the label information of the product to be recommended based on the large language model, and generate a marketing strategy for recommending the product to be recommended to the target customer.

The intelligent generation device of marketing strategies provided by the embodiment of the present invention obtains the communication records of target customers and the product information of the products to be recommended; performs label mining on the communication records based on a large language model to generate the label information of the target customers; performs label mining on the product information based on the large language model to generate the label information of the products to be recommended; performs information fusion on the label information of the target customers and the label information of the products to be recommended based on the large language model to generate a marketing strategy for recommending the products to be recommended to the target customers; on the one hand, the communication records of the target customers and the product information of the products to be recommended are subjected to label mining through the large language model, and on the other hand, the information fusion is performed on the label information obtained by mining, and the powerful computing power and learning ability of the large language model for complex features are utilized, so that the data can be better understood and utilized, and useful information can be more effectively extracted from the data, which not only improves the accuracy of data mining such as communication records and product information, but also processes large-scale massive data, and finally automatically generates high-quality marketing strategies, which can accurately locate the customer needs of different customers, recommend products that better meet customer needs, improve marketing efficiency, and improve customer experience.

In some embodiments, the customer tag mining module is used to:

Conduct tag mining on communication records based on large language models to generate real-time tag information for target customers;

Based on a large language model, the real-time label information and historical label information of the target customer are fused to generate the label information of the target customer.

In some embodiments, the apparatus further comprises:

The portrait analysis module is used to generate customer portraits of target customers based on their label information;

The customer portrait is analyzed based on a large language model to generate customer analysis results for target customers; the customer analysis results include at least one of the product preference analysis results and the purchase intention analysis results of the target customers.

In some embodiments, the portrait analysis module is further used to:

Generate marketing opportunity profiles of target customers based on their tag information;

Analyze marketing opportunity portraits based on large-scale language models and generate marketing opportunity recommendation information corresponding to target customers.

In some embodiments, the apparatus further comprises:

The product comparison module is used to compare product information with product information of related products corresponding to the product to be recommended based on a large language model to generate product comparison information of the product to be recommended.

In some embodiments, the policy generation module is used to:

Match the target customer's label information with the label information of the product to be recommended to obtain the matching degree information between the target customer and the product to be recommended;

Based on a large language model, the matching degree information, the customer analysis results of the target customers and the marketing opportunity recommendation information corresponding to the target customers are integrated to generate a marketing strategy to recommend the products to be recommended to the target customers.

In some embodiments, the apparatus further comprises:

A knowledge question and answer module is used to build a product knowledge base containing product information of the products to be recommended;

Extract information from the product knowledge base based on a large language model to generate a product information base corresponding to the product knowledge base;

Generate product knowledge answers corresponding to product knowledge questions based on large language models and product information databases.

The following uses a marketing system connected to a banking system as an example to illustrate the marketing strategy intelligent generation method provided by the present invention.

FIG4 is a schematic diagram of the overall architecture of the marketing system provided by the present invention. As shown in FIG4 , the marketing system is connected with instant communication software, a product recommendation system, a product knowledge database and a large language model to provide marketing strategy intelligent generation services.

Instant messaging software is used to provide information such as recording files, real-time text, and call attributes of communications with customers.

The product recommendation system is used to match based on label information provided by large language models.

The bank's external data mainly includes customers' basic labels, transaction records, communication records and other data.

The product knowledge database provides services related to the bank's product knowledge, including knowledge import, knowledge retrieval and document retrieval.

Large language models can be used to perform summary generation, to-do generation, intent recognition, feature extraction, multi-round dialogue, interaction optimization, and product information extraction.

Marketing strategy intelligent generation services can include customer profiling, speech recommendations, customer analysis, tag mining, summary, to-do list, product consultation, product comparison and product recommendations, etc.

Among them, label mining refers to improving the understanding of unstructured data by integrating multi-dimensional unstructured data and utilizing the generative information extraction capabilities of large models, making label extraction more accurate and richer.

Customer analysis refers to the process of outputting the product categories that customers tend to purchase based on the customer's transaction data, behavioral data, and related communication records in the banking system, combined with customer tags and customer stratification and grouping mined by big models, through the analysis and summary capabilities of big models.

Intention analysis refers to the integration and analysis of multi-dimensional customer data, using rule matching or a combination of large and small models to enrich customer portraits, build recommendation models, explore product features preferred by customers, build customer intention analysis models, and analyze customer product purchase probabilities.

Product comparison refers to the use of the text understanding and text generation capabilities of the big model based on the understanding of multi-round conversation data to accurately capture the customer's intention for product comparison and the names of the products that need to be compared, retrieve product information from the knowledge base, and use the big model to summarize, refine and display the comparison results.

To-do items refer to using past communication records with customers and mined tag data to build customer time profiles, predict customer free time, and use big model technology to recommend marketing time to help account managers increase the probability of successful customer contact. Use the generation capabilities of big models to generate reminder texts. When multiple reminders overlap, use different prompt templates to issue instructions to the big model. For example, when a birthday is approaching and a product is due to expire, the big model can first send birthday greetings and then use a smooth transition tone to switch to the expiration reminder.

Speech recommendation refers to identifying customer pain points, product comparison needs, emotional soothing, and service interspersion needs, providing appropriate speech, and realizing related product marketing.

Product consultation refers to activating the traditional enterprise knowledge application capabilities and realizing the interactive upgrade of knowledge question and answer. Traditional knowledge can be transmitted and used immediately, and questions can be answered immediately, thus improving the intelligent experience of knowledge question and answer application throughout the entire chain.

FIG5 is a schematic diagram of the component logic architecture of the marketing system provided by the present invention. As shown in FIG5 , the marketing strategy intelligent generation method provided by the present invention can be implemented by constructing different component layers and setting different components in different component layers.

The marketing system can include access layer, application layer, core capability layer, data layer and basic layer.

The access layer can be integrated into the customer relationship management system (CRM) of the banking system, or into the instant communication software or client of the banking system. The access layer is mainly used to provide an operation interface for the account manager.

The application layer can adopt a front-end and back-end separation technical architecture design, where the front-end provides functional services and the back-end implements functions by calling modules in the core capability layer.

The core capability layer may include a call module, a knowledge base module, a recommendation engine module, a speech recognition module, an image recognition module, and a large language model.

The data layer can store customer basic data, customer behavior data, customer transaction data, label analysis data, product knowledge data, etc.

The basic layer can provide hardware and software support, including servers, storage, load balancers, operating systems, networks, and backup software.

FIG6 is a schematic diagram of the structure of an electronic device provided by the present invention. As shown in FIG6 , the electronic device may include: a processor (Processor) 610, a communication interface (Communications Interface) 620, a memory (Memory) 630 and a communication bus (Communications Bus) 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logic command in the memory 630 to execute the method described in the above embodiment, for example:

Obtain the communication records of target customers and product information of the products to be recommended; perform label mining on the communication records based on a large language model to generate label information of the target customers; perform label mining on the product information based on a large language model to generate label information of the products to be recommended; perform information fusion on the label information of the target customers and the label information of the products to be recommended based on a large language model to generate a marketing strategy for recommending the products to be recommended to the target customers.

In addition, the logic commands in the above-mentioned memory can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on such an understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several commands to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

The processor in the electronic device provided in the embodiment of the present invention can call the logic instructions in the memory to implement the above method. Its specific implementation method is consistent with the implementation method of the aforementioned method and can achieve the same beneficial effects, which will not be repeated here.

An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the method provided in the above embodiments is implemented.

Its specific implementation is consistent with the aforementioned method implementation and can achieve the same beneficial effects, so it will not be repeated here.

An embodiment of the present invention provides a computer program product, including a computer program, and when the computer program is executed by a processor, the above method is implemented.

The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM/RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

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

Claims (10)

1. A marketing strategy intelligent generation method, characterized by comprising: Obtain communication records with target customers and product information of products to be recommended; Performing label mining on the communication records based on a large language model to generate label information of the target customer; Performing label mining on the product information based on a large language model to generate label information of the product to be recommended; Based on a large language model, information fusion is performed on the label information of the target customer and the label information of the product to be recommended, so as to generate a marketing strategy for recommending the product to be recommended to the target customer. 2. The method for intelligently generating marketing strategies according to claim 1, wherein the step of performing tag mining on the communication records based on a large language model to generate tag information of the target customers comprises: Performing tag mining on the communication records based on a large language model to generate real-time tag information of the target customer; The real-time label information and historical label information of the target customer are fused based on a large language model to generate label information of the target customer. 3. The method for intelligently generating marketing strategies according to claim 2, characterized in that after the real-time tag information and historical tag information of the target customer are fused based on a large language model to generate the tag information of the target customer, the method further comprises: Based on the tag information of the target customer, generating a customer profile of the target customer; The customer portrait is analyzed based on a large language model to generate a customer analysis result of the target customer; the customer analysis result includes at least one of a product preference analysis result and a purchase intention analysis result of the target customer. 4. The method for intelligently generating marketing strategies according to claim 2, characterized in that after the real-time tag information and historical tag information of the target customer are fused based on a large language model to generate the tag information of the target customer, the method further comprises: Based on the tag information of the target customer, generate a marketing opportunity portrait of the target customer; The marketing opportunity portrait is analyzed based on a large language model to generate marketing opportunity recommendation information corresponding to the target customer. 5. The method for intelligently generating marketing strategies according to claim 1, characterized in that after performing label mining on the product information based on a large language model to generate label information of the product to be recommended, the method further comprises: Based on a large language model, the product information is compared with the product information of the associated products corresponding to the product to be recommended to generate product comparison information of the product to be recommended. 6. The method for intelligently generating a marketing strategy according to claim 1, characterized in that the step of fusing the label information of the target customer and the label information of the product to be recommended based on a large language model to generate a marketing strategy for recommending the product to be recommended to the target customer comprises: Matching the label information of the target customer with the label information of the product to be recommended to obtain information on the degree of matching between the target customer and the product to be recommended; Based on a large language model, the matching degree information, the customer analysis results of the target customers and the marketing opportunity recommendation information corresponding to the target customers are integrated to generate a marketing strategy for recommending the product to be recommended to the target customers. 7. The method for intelligently generating marketing strategies according to any one of claims 1 to 6, characterized in that the method further comprises: Building a product knowledge base containing product information of the product to be recommended; Information is extracted from the product knowledge base based on a large language model to generate a product information base corresponding to the product knowledge base; and product knowledge answers corresponding to product knowledge questions are generated based on the large language model and the product information base. 8. A marketing strategy intelligent generation device, characterized by comprising: The information acquisition module is used to obtain the communication records of target customers and the product information of the products to be recommended; A customer tag mining module, used to perform tag mining on the communication records based on a large language model to generate tag information of the target customer; A product label mining module, used to perform label mining on the product information based on a large language model to generate label information of the product to be recommended; The strategy generation module is used to fuse the label information of the target customer and the label information of the product to be recommended based on a large language model, and generate a marketing strategy for recommending the product to be recommended to the target customer. 9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the marketing strategy intelligent generation method as described in any one of claims 1 to 7 when executing the computer program. 10. A non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for intelligently generating marketing strategies as described in any one of claims 1 to 7 is implemented.