User behavior habit analysis system and method based on big data fusion

Abstract

The invention discloses a user behavior habit analysis system and method based on big data fusion, and relates to the technical field of intelligent interaction, and the system carries out the preliminary screening of sales personnel based on the actual product demands, analyzes the value degree of an order based on the actual product demands of a customer, and analyzes the purchasing stage and emergency degree of the customer based on the customer dialogue record. The purchase intention of the customer is analyzed based on the purchase stage and the urgency degree, the priority of the customer is analyzed based on the value degree of the order and the purchase intention of the customer, and matching of a plurality of sales is performed based on the priority of the customer. According to the invention, through real-time dialogue analysis, the purchase intention of the customer is quantified, and the optimal sales is allocated according to the priority of the customer, so that the accurate and preferential delivery of sales resources is realized, and the customer experience is improved. Therefore, the purchase transaction rate and the customer satisfaction are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent interaction technology, and in particular to a user behavior habit analysis system and method based on big data fusion. Background Technology

[0002] In the field of foreign trade customer management and sales automation, existing technologies generally suffer from superficial language understanding and lack of context when handling cross-language and cross-cultural customer inquiries. Traditional systems mostly rely on keyword matching and direct translation technologies. Therefore, they lack a cultural dimension in language processing and cannot understand the business culture habits behind different languages. The lack of cultural context can easily lead to misjudgment of the urgency and true intentions of customers. Furthermore, different regions may have different terms or expressions for the same product, which can also cause traditional keyword matching systems to fail and misclassify customer needs. These limitations in language understanding directly lead to a chain of errors in subsequent steps, ultimately resulting in serious inefficiency in the allocation of sales resources. To address the aforementioned issues, this invention provides a user behavior habit analysis system and method based on big data fusion. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a user behavior habit analysis system and method based on big data fusion. This invention quantifies customer purchase intentions through real-time analysis of dialogues and allocates the most suitable sales personnel according to customer priorities, thereby achieving precise and prioritized allocation of sales resources, which improves the purchase conversion rate and customer satisfaction.

[0004] To achieve the above objectives, this invention provides a user behavior habit analysis method based on big data fusion, comprising the following specific steps: Step 1: The customer initiates a conversation. The language processing module collects the customer's conversation records and identifies the customer's actual product needs based on the characteristics of the language used by the customer and the habits of the target market. Based on the actual product needs, the sales personnel are initially screened. Step 2: Analyze the value of the order based on the customer's actual product needs; Step 3: Analyze the customer's purchasing stage and urgency based on customer dialogue records, and analyze the customer's purchasing intention based on the purchasing stage and urgency. Step 4: Analyze customer priority based on order value and customer purchase intention, and match several sales based on customer priority; Step 5: Match the best salesperson based on the customer's historical inquiries about the order and the corresponding salesperson's knowledge of the product.

[0005] Preferably, step one includes the following specific steps: Step 11: The customer initiates a conversation. The language processing module collects the customer's conversation records and automatically performs language recognition, entity recognition, and intent classification. The customer's conversation records are converted into structured data tags. The entity recognition includes product name, product model, purchase quantity, target market, company name, and contact person name. The intent classification includes inquiries, requests for samples, technical consultations, orders, and order reminders. Step 12: Generate a customer profile based on the characteristics of the customer's language, target market habits, customer type, and historical purchasing information. The characteristics of the customer's language include language expression style, spelling preferences, and word order. The target market habits include target regions and related market characteristics. The related market characteristics include regional regulations, preferences, and logistics. The historical purchasing information includes the customer's historical purchased products, historical purchase quantities, and historical purchase times. Step 13: Based on the customer profile, infer the customer's purchasing pattern, adaptive needs, and business development stage. The customer purchasing pattern is the purchasing characteristics based on customer type, the adaptive needs are the purchasing needs based on geographical location, and the business development stage is the extended needs based on historical purchasing information. Identify the customer's actual product needs based on the customer's purchasing pattern, adaptive needs, and business development stage. The actual product needs include the actual product name, actual product model, actual purchase quantity, and actual target market. Step 14: Construct actual product demand tags based on actual product needs, and construct tags for each salesperson based on the salesperson database. The tags for each salesperson include the product line they are responsible for, the language they are responsible for, the type of customer they are responsible for, and their current workload, i.e., the number of customers or intent orders they have been assigned. Calculate the matching degree between the actual product demand tags and the tags for each salesperson, and filter out salespeople whose matching degree exceeds the preset threshold.

[0006] Preferably, step two includes the following specific steps: Step 21: Obtain the product price and purchase quantity based on the customer's actual product needs; obtain the product profit margin by dividing the difference between the product price and the product cost by the product price; obtain the purchase scale by dividing the product quantity and the product price by the maximum historical total purchase amount; and obtain the product profitability by weighted summing of the product profit margin and the purchase scale. Step 22: Calculate customer rating based on customer size and credit risk, obtain customer importance score based on the ratio of customer's historical total purchase amount to historical total sales, and obtain customer stability by weighted summation of customer rating score and customer importance score. Step 23: Analyze the ease of production based on the degree of customization; subtract the standard production cycle of the product from the difference between the customer's required delivery date and the current date to obtain the buffer time; obtain the delivery capacity based on the ratio of the buffer time to the standard production cycle; and obtain the operational feasibility by weighted summation of the ease of production and delivery capacity. Step 24: Obtain the order value by weighting and summing the product profitability, customer stability, and operational feasibility.

[0007] Preferably, step three includes the following specific steps: Step 31: Obtain communication frequency, response time, and conversation duration based on customer dialogue records; obtain a communication frequency assessment value based on the ratio of communication frequency to standard communication frequency; obtain a response duration assessment value based on the ratio of standard response time to response duration; obtain a conversation duration assessment value based on the ratio of conversation duration to standard conversation duration; and obtain an order urgency assessment value by weighted summation of the communication frequency assessment value, response duration assessment value, and conversation duration assessment value. Step 32: Based on customer dialogue records, obtain the customer's procurement stage, which includes information collection stage, demand definition stage, product comparison stage, decision-making stage and transaction stage. Obtain the total number of customers and the final number of transactions for each historical procurement stage. Obtain the historical procurement transaction rate based on the ratio of the final number of transactions to the total number of customers. Obtain the corresponding procurement transaction rate based on the procurement stage in which the customer is located. Step 33: Weight the order urgency assessment value and the purchase completion rate to obtain the customer's purchase intention assessment value.

[0008] Preferably, step four includes the following specific steps: Step 41: Calculate the customer's priority level by weighted summation of the order's value assessment and the customer's purchase intention assessment; Step 42: Obtain the screened sales personnel and determine their competency level based on their product knowledge, communication skills, and customer management. Step 43: Select at least three sales representatives who are highly matched with the customer's priority level and the salesperson's ability level, and set them as pre-selected sales representatives.

[0009] Preferably, step five includes the following specific steps: Step 51: Obtain the pre-selected sales staff's understanding of each part of the product. The sales staff's understanding of each part of the product includes the frequency of introduction and the number of words in the introduction of each part in the past. Specifically, the word count understanding value is obtained by dividing the average number of words in the introduction of the corresponding part by the average number of words in the introduction of each part. The frequency understanding value is obtained by dividing the average frequency of introduction of the corresponding part by the average frequency of introduction of each part. The word count understanding value and the frequency understanding value are weighted and summed to obtain the understanding of the corresponding part. Step 52: Obtain the frequency of customer inquiries about each part of the product during the dialogue phase, as well as the proportion of historical inquiries about the corresponding question to all historical inquiries about that part. The smaller the proportion, the fewer people know about the question. Divide the proportion of historical inquiries about the corresponding question to all historical inquiries about that part by the average question proportion and then take the reciprocal to obtain the difficulty of solving the question. Divide the frequency of inquiries about each part of the product during the dialogue phase by the standard inquiry frequency to obtain the inquiry frequency value. Multiply the average difficulty of solving the question for the corresponding part by the inquiry frequency value to obtain the customer awareness value for the corresponding part. Step 53: Obtain the customer perception value and sales perception status of the corresponding part. Divide the sales perception status of the corresponding part by the customer perception value to obtain the predicted understanding value of the corresponding part. Add up the predicted understanding values ​​of all parts and average them to obtain the customer predicted understanding value of the product. Step 54: Select the pre-selected sale corresponding to the largest customer forecast understanding value and set it as the customer's optimal sale.

[0010] This invention also provides a user behavior habit analysis system based on big data fusion, including: The product demand analysis module is used to collect customer dialogue records through the language processing module and identify the customer's actual product needs based on the characteristics of the language used by the customer and the habits of the target market. The initial sales screening module is used to initially screen sales personnel based on actual product needs; The order value analysis module is used to analyze the value of orders based on customers' actual product needs. The purchase intention analysis module is used to analyze customers' purchasing stages and urgency based on customer conversation records, and to analyze customers' purchase intentions based on purchasing stages and urgency. The customer priority analysis module is used to analyze customer priority based on order value and customer purchase intention. The optimal sales allocation module analyzes customer priority based on order value and customer purchase intention, matches several sales representatives based on customer priority, and selects the optimal sales representative based on the customer's historical inquiries about the order and the corresponding sales representative's knowledge of the order's products.

[0011] The present invention also provides an electronic device, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes the above-described user behavior habit analysis method based on big data fusion by calling the computer program stored in the memory.

[0012] The present invention also provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described user behavior habit analysis method based on big data fusion.

[0013] Compared with existing technologies, this invention involves a customer initiating a dialogue, a language processing module collecting customer dialogue records, identifying the customer's actual product needs based on the characteristics of the language used by the customer and the habits of the target market, conducting preliminary screening of sales personnel based on actual product needs, analyzing the value of the order based on the customer's actual product needs, analyzing the customer's procurement stage and urgency based on the customer's dialogue records, analyzing the customer's purchase intention based on the procurement stage and urgency, analyzing the customer's priority based on the order value and the customer's purchase intention, and assigning the salesperson with the highest matching degree to follow up based on the customer's priority. The beneficial effects of this invention are: by analyzing dialogues in real time, the purchase intention of customers can be quantified, and the most suitable sales can be allocated according to the customer's priority, so as to achieve accurate and prioritized allocation of sales resources, thereby improving the purchase conversion rate and customer satisfaction. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of the user behavior habit analysis method based on big data fusion according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating step one of the user behavior habit analysis method based on big data fusion in an embodiment of the present invention. Figure 3 This is a flowchart illustrating step two of the user behavior habit analysis method based on big data fusion in an embodiment of the present invention. Figure 4 This is a schematic diagram of the user behavior habit analysis system framework based on big data fusion, as described in an embodiment of the present invention. Detailed Implementation

[0016] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.

[0017] like Figure 1 As shown, this embodiment of the invention provides a user behavior habit analysis method based on big data fusion, including the following specific steps: Step 1: The customer initiates a conversation. The language processing module collects the customer's conversation records and identifies the customer's actual product needs based on the characteristics of the language used by the customer and the habits of the target market. Based on the actual product needs, the sales personnel are initially screened. like Figure 2 As shown, in this embodiment, step one includes the following specific steps: Step 11: The customer initiates a conversation. The language processing module collects the customer's conversation records and uses a natural language processing engine to automatically perform language recognition, entity recognition, and intent classification. The customer's conversation records are converted into structured data tags. The entity recognition includes product name, product model, purchase quantity, target market, company name, and contact person name. The intent classification includes inquiries, requests for samples, technical consultations, orders, order reminders, etc. Preferably, all customer communication channels are integrated through API interfaces or SDKs, and the system captures the original, unstructured dialogue text, timestamps, and channel sources; Step 12: Generate a customer profile based on the characteristics of the customer's language, target market habits, customer type, and historical purchasing information. The characteristics of the customer's language include language expression style, spelling preferences, and word order. Language expression style is divided into expressive and introverted styles. Expressive styles often use a large number of exclamation marks and exaggerated adjectives. The sentiment analysis model needs to be calibrated for these language characteristics to avoid misjudging cultural enthusiasm or complaints. Introverted styles may also use very restrained language. The sentiment analysis model needs to identify the true intensity of adverbs of degree such as "somewhat," "slightly," and "relatively" in specific cultural contexts. Their actual severity is usually higher than the literal meaning. Meanwhile, expressive styles often have clear intentions, focusing on capturing the specific requirements of the customer. Introverted types tend to conceal their intentions within the dialogue, focusing on understanding the customer's potential needs based on context. Different regions have different spelling preferences, such as standard American English spelling versus standard British English spelling. Differences in word order lead customers to place key information differently when expressing themselves. The target market habits include the target region and related market characteristics. Related market characteristics include regional regulations, preferences, and logistics. For example, regulations might include the EU requiring CE certification; preferences might include some regions' consumer preference for environmentally friendly packaging; and logistics might include commonly used shipping routes and delivery times to the target region. Customer types are categorized as large retailers, manufacturers, wholesalers, end customers, etc. Historical purchasing information includes the customer's historical purchased products, quantities, and purchase dates. This embodiment takes into account the unique ways, habits, and subtexts that customers express their needs, intentions, and emotions in different language and cultural backgrounds. By distinguishing the communication styles of customers from different cultural backgrounds, it can more accurately determine their true intentions and emotional states. Step 13: Infer customer purchasing patterns, adaptive needs, and business development stages based on customer profiles. The customer purchasing pattern refers to the purchasing characteristics based on customer type. For the same product, different types of customers have completely different purchasing patterns. For example, when ordering 10,000 stainless steel screws, small businesses might use them for equipment maintenance, with low purchasing frequency and a focus on unit price, while large manufacturing enterprises might use them for mass production, with stable demand and a focus on delivery cycle and quality consistency. By analyzing customer types, their purchasing motives can be further differentiated. Combined with the business development stage, the actual product demand can be predicted, thereby providing accurate product recommendations and customized service solutions to improve the conversion rate and customer satisfaction. The adaptive needs refer to purchasing needs based on geographical location. The customer's geographical location determines that the product needs to adapt to the local environment. For example, when ordering the same specifications of solar panels, high-temperature areas require high-temperature resistant models with dust-proof coatings, while cold regions require low-light efficiency optimized models with anti-icing and snow accumulation designs. The business development stage refers to extended needs based on historical purchasing information. Changes in customer purchasing patterns reflect their business development status. For example, a customer who previously ordered 10,000 parts per month... A customer suddenly inquires about the price of 50,000 units. This may indicate they have secured a major client or a new project. When providing products, they require greater supply stability and may also have extended needs such as installment delivery or extended payment terms. Based on the customer's purchasing model, adaptability requirements, and business development stage, we identify their actual product needs. These actual product needs include the actual product name, model, quantity, and target market. For example, a European automotive electronics system integrator initially inquires about 5,000 industrial-grade wireless modules. Given the automotive industry and the nature of wireless modules, it can be inferred that they are used in connected car systems. Therefore, they require automotive-grade certification, high reliability, and long lifespan support. Given Europe's geographical location, they need to comply with local automotive industry standards. This initial inquiry of 5,000 units may be for a pre-research project of a new car model or small-batch trial production, requiring sample support, small-batch supply capabilities, and a future price roadmap for mass production. Therefore, the final calculated actual needs of this customer are: the customer is ordering 5,000 automotive-grade, highly reliable industrial-grade wireless modules for a new connected car project for a European automaker, requiring the supplier to have technical support capabilities and a smooth transition from small-batch to large-scale production. Step 14: Construct actual product demand tags based on actual product needs, and construct tags for each salesperson based on the salesperson database. The tags for each salesperson include the product line they are responsible for, the language they are responsible for, the type of customer they are responsible for, and their current workload, i.e., the number of customers or intent orders they have been assigned. Calculate the matching degree between the actual product demand tags and the tags for each salesperson, and filter out salespeople whose matching degree exceeds the preset threshold. Preferably, the product line matching score can be obtained by dividing the number of intersections between the customer's required product line and the product line the salesperson is responsible for by the number of the customer's required product lines. If the customer's requirement is only one product line and the salesperson is responsible for that product line, the score is 1; otherwise, it is 0. The language matching score checks whether the language required by the customer is in the salesperson's language list. If it is, the score is 1; otherwise, it is 0. Alternatively, it can be divided into levels between 0 and 1 based on proficiency. The customer type matching score can be obtained by dividing the number of intersections between the customer type and the customer types the salesperson is proficient in by the number of customer types. If the customer has only one type, the score is 1; otherwise, it is 0. The workload matching score can be obtained by dividing 1 by the current workload by the specified maximum workload. The lower the workload, the higher the score. The matching score between the customer and the salesperson is obtained by weighted summing of the product line matching score, language matching score, customer type matching score, and workload matching score.

[0018] Step 2: Analyze the value of the order based on the customer's actual product needs; like Figure 3 As shown, in this embodiment, step two includes the following specific steps: Step 21: Obtain the product price and purchase quantity based on the customer's actual product needs; obtain the product profit margin by dividing the difference between the product price and the product cost by the product price; obtain the purchase scale by dividing the product quantity and the product price by the maximum historical total purchase amount; and obtain the product profitability by weighted summing of the product profit margin and the purchase scale. Step 22: Customer rating is conducted based on customer size and credit risk. Large customers are defined as those with annual revenue exceeding 100 million RMB or more than 1,000 employees; medium-sized customers are defined as those with annual revenue between 10 million and 100 million RMB or 100-1,000 employees; and small customers are defined as those with annual revenue between 1 million and 10 million RMB or 10-100 employees. Low credit risk is defined as occasional minor delinquencies (<30 days) with stable operations and good debt repayment ability; medium credit risk is defined as those with a history of delinquencies (30-90 days) or some operational pressure; and high credit risk is defined as those with frequent delinquencies (>90 days) or tight cash flow and high debt ratio. Customer ratings are divided into A, B, C, and D, with scores corresponding to 1, 0.7, 0.4, and 0, respectively. The customer rating matrix is ​​shown in Table 1. Table 1:

[0019] Customer importance score is obtained by the ratio of customer's historical total purchase amount to historical total sales amount, and customer stability is obtained by weighted summation of customer level score and customer importance score; Step 23: Analyze the ease of production based on the degree of customization. The degree of customization is divided into standard products, modular customization, and deep customization. Standard products are products that do not require any modifications and can be used directly. The ease of production level is Level 1, with a score of 1. Modular customization involves modifying non-functional parts such as color, packaging, and labels, or selecting combinations from the threshold configuration options. The ease of production level is Level 2, with a score of 0.6. Deep customization involves modifications to core functions or structures. The ease of production level is Level 3, with a score of 0.1. The buffer time is obtained by subtracting the standard production cycle of the product from the difference between the customer's required delivery date and the current date. The delivery capacity is obtained by the ratio of the buffer time to the standard production cycle. The operational feasibility is obtained by weighted summing of the ease of production level and delivery capacity. Step 24: Obtain the order value by weighting and summing the product profitability, customer stability, and operational feasibility.

[0020] Step 3: Analyze the customer's purchasing stage and urgency based on customer dialogue records, and analyze the customer's purchasing intention based on the purchasing stage and urgency. In this embodiment, step three includes the following specific steps: Step 31: Obtain communication frequency, response time, and conversation duration based on customer conversation records. Obtain a communication frequency assessment value based on the ratio of communication frequency to standard communication frequency. Obtain a response duration assessment value based on the ratio of standard response time to response duration. Obtain a conversation duration assessment value based on the ratio of conversation duration to standard conversation duration. Obtain an order urgency assessment value by weighted summing of the communication frequency assessment value, response duration assessment value, and conversation duration assessment value. Standard communication frequency, standard response time, and standard conversation duration can be obtained by averaging all historical user conversation records. Step 32: Based on customer dialogue records, obtain the customer's procurement stage, which includes information collection stage, demand clarification stage, product comparison stage, decision-making stage, and transaction stage. Obtain the total number of customers and the final number of transactions for each historical procurement stage. Obtain the historical procurement transaction rate based on the ratio of the final number of transactions to the total number of customers. Obtain the corresponding procurement transaction rate based on the customer's current procurement stage. The information collection stage is characterized by the customer initially understanding the product, inquiring about basic information, and not yet forming a clear procurement intention. The demand clarification stage is characterized by the customer beginning to express specific needs, such as specifications, quantity, and budget. The solution comparison stage is characterized by the customer communicating with multiple suppliers, comparing quotations, samples, and services. The decision-making stage is characterized by the customer showing a strong procurement intention, inquiring about details such as delivery time, payment, and logistics. The transaction stage is characterized by the customer confirming the order and entering the contract and payment process. Step 33: Weight the order urgency assessment value and the purchase completion rate to obtain the customer's purchase intention assessment value.

[0021] Step 4: Analyze customer priority based on order value and customer purchase intention, and match several sales based on customer priority; In this embodiment, step four includes the following specific steps: Step 41: Obtain the customer's priority level by weighted summing of the order value assessment and the customer's purchase intention assessment. Quantify the customer's priority by weighted summing of the order value and the customer's purchase intention. Step 42: Obtain the screened sales personnel. Based on the sales personnel's product knowledge, communication skills, and customer management, obtain the sales personnel's ability level. The sales personnel's ability level is determined periodically by sales volume and average customer evaluation scores. In this embodiment, for example, the sales personnel's ability level is obtained by: obtaining the ratio of sales volume to the average sales volume, and simultaneously obtaining the ratio of customer evaluation scores to the average scores. The two ratios are weighted and summed to obtain the sales personnel's ability level. Based on the weighted ratio of sales volume (business results) and customer evaluation (service quality), the sales ability is dynamically evaluated. Step 43: Select at least three sales representatives with the highest matching degree based on the customer's priority level and the salesperson's ability level, and set them as pre-selected sales representatives; the matching degree is obtained by dividing the customer's priority level by the salesperson's ability level, obtaining the selection value, obtaining the difference between the selection value and the selection threshold, and taking the reciprocal to obtain the matching degree.

[0022] Step 5: Match the best salesperson based on the customer's historical inquiries about the order and the salesperson's knowledge of the products in the order; In this embodiment, step five includes the following specific steps: Step 51: Obtain the pre-selected sales staff's understanding of each part of the product. This includes the frequency and word count of historical descriptions of each part. Specifically, the word count is calculated by dividing the average word count of the description of each part by the average word count of all parts. The frequency is calculated by dividing the average frequency of the description of each part by the average frequency of the description of all parts. The word count and frequency are then weighted and summed to obtain the overall understanding of each part. By using historical word count and frequency, the sales staff's familiarity with each part of the product can be quantified. Step 52: Obtain the frequency of customer inquiries about each part of the product during the dialogue phase, as well as the proportion of historical inquiries about the corresponding question to all historical inquiries about that part. The smaller the proportion, the fewer people are aware of the question. Divide the proportion of historical inquiries about the corresponding question to all historical inquiries about that part by the average question proportion and then take the reciprocal to obtain the difficulty of solving the question. Divide the frequency of inquiries about each part of the product during the dialogue phase by the standard inquiry frequency to obtain the inquiry frequency value. Multiply the average difficulty of solving the question for the corresponding part by the inquiry frequency value to obtain the customer's awareness value for the corresponding part. Combine the inquiry frequency and the rarity of the question (reciprocal proportion) to calculate the complexity of the customer's question. Step 53: Obtain the customer perception value and sales perception status of the corresponding part. Divide the sales perception status of the corresponding part by the customer perception value to obtain the predicted understanding value of the corresponding part. Add up the predicted understanding values ​​of all parts and average them to obtain the customer predicted understanding value of the product. Step 54: Select the pre-selected sales corresponding to the highest customer predicted understanding value as the customer's optimal sales, compare the sales understanding with the difficulty of the customer's questions, predict the understanding level, and select the sales with the highest score. This embodiment matches sales personnel with appropriate capabilities based on customer priority. High-priority customers are assigned to experienced sales personnel first, while low-priority customers are followed up and continuously nurtured by junior sales personnel to ensure efficient use of resources and improved conversion rates. At the same time, it matches and compares sales personnel's knowledge with the difficulty of customer issues, predicts comprehension levels, and selects the sales personnel with the highest scores, which greatly improves the ability to solve customer problems.

[0023] The process of determining the weights and thresholds in this implementation can be as follows: acquire several dialogues as test dialogues, input the dialogue records into each step of this embodiment to obtain the final allocated sales, follow up on the actual final allocated sales to obtain the final conversion rate, and determine the weights through data-driven methods (such as regression analysis or entropy weight method) to reflect the relative importance of each indicator; the thresholds are set based on historical data quantiles, cluster analysis, or business rules to define the effective critical values ​​for matching. Both support dynamic adjustment to ensure accurate model matching and adaptability to business changes, while emphasizing transparency and flexibility. In the initial stage, expert experience can be combined to solve the cold start problem, and in the later stage, data optimization can be used to avoid overfitting.

[0024] like Figure 4 As shown, embodiments of the present invention also provide a user behavior habit analysis system based on big data fusion, including: The product demand analysis module is used to collect customer dialogue records through the language processing module and identify the customer's actual product needs based on the characteristics of the language used by the customer and the habits of the target market. The initial sales screening module is used to initially screen sales personnel based on actual product needs; The order value analysis module is used to analyze the value of orders based on customers' actual product needs. The purchase intention analysis module is used to analyze customers' purchasing stages and urgency based on customer conversation records, and to analyze customers' purchase intentions based on purchasing stages and urgency. The customer priority analysis module is used to analyze customer priority based on order value and customer purchase intention. The optimal sales allocation module is used to analyze customer priority based on order value and customer purchase intention, and match several sales based on customer priority; it also matches the best sales based on the customer's historical inquiries about the order and the corresponding salesperson's knowledge of the order's products.

[0025] This invention also provides an electronic device, including a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes the above-described user behavior habit analysis method based on big data fusion by calling the computer program stored in the memory.

[0026] The electronic device can vary considerably depending on its configuration or performance. It may include one or more Central Processing Units (CPUs) and one or more memories, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the user behavior habit analysis method based on big data fusion provided in the above-described embodiment. The electronic device may also include other components for implementing its functions; for example, it may have wired or wireless network interfaces and input / output interfaces for data input and output. Further details are omitted in this embodiment.

[0027] This invention also provides a computer-readable storage medium storing instructions that, when a computer program is run on a computer device, cause the computer device to execute the aforementioned user behavior habit analysis method based on big data fusion.

[0028] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, electronic devices, and computer storage media described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] In the several embodiments provided by this invention, it is understood that each block in the flowchart or block diagram may represent a module, program segment, or part of code. The module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the figures. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved.

[0031] In the several embodiments provided by this invention, it should be understood that the disclosed systems, electronic devices, computer storage media, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed between each other may be through some interfaces, and the indirect coupling or communication connection of devices or units may be electrical, mechanical, or other forms.

[0032] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0033] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for executing all or part of the steps of the methods described in the various embodiments of the present invention through a computer device (which may be a personal computer, a server, or a network device, etc.). The aforementioned storage medium includes: USB flash drive, mobile hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and other media that can store program code.

[0034] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A user behavior habit analysis method based on big data fusion, characterized in that, The specific steps include the following: Step 1: The customer initiates a conversation. The language processing module collects the customer's conversation records and identifies the customer's actual product needs based on the characteristics of the language used by the customer and the habits of the target market. Based on the actual product needs, the sales personnel are initially screened. Step 2: Analyze the value of the order based on the customer's actual product needs; Step 3: Analyze the customer's purchasing stage and urgency based on customer dialogue records, and analyze the customer's purchasing intention based on the purchasing stage and urgency. Step 4: Analyze customer priority based on order value and customer purchase intention, and match several sales based on customer priority; Step 5: Match the best salesperson based on the customer's historical inquiries about the order and the corresponding salesperson's knowledge of the product.

2. The user behavior habit analysis method based on big data fusion according to claim 1, characterized in that, Step one includes the following specific steps: Step 11: The customer initiates a conversation. The language processing module collects the customer's conversation records and automatically performs language recognition, entity recognition, and intent classification. The customer's conversation records are converted into structured data tags. The entity recognition includes product name, product model, purchase quantity, target market, company name, and contact person name. Step 12: Generate a customer profile based on the characteristics of the customer's language, target market habits, customer type, and historical purchasing information. The characteristics of the customer's language include language expression style, spelling preferences, and word order. The target market habits include target regions and related market characteristics. The related market characteristics include regional regulations, preferences, and logistics. The historical purchasing information includes the customer's historical purchased products, historical purchase quantities, and historical purchase times. Step 13: Based on the customer profile, infer the customer's purchasing pattern, adaptive needs, and business development stage. The customer purchasing pattern is the purchasing characteristics based on customer type, the adaptive needs are the purchasing needs based on geographical location, and the business development stage is the extended needs based on historical purchasing information. Identify the customer's actual product needs based on the customer's purchasing pattern, adaptive needs, and business development stage. The actual product needs include the actual product name, actual product model, actual purchase quantity, and actual target market. Step 14: Construct actual product demand tags based on actual product needs, and construct tags for each salesperson based on the salesperson database. The tags for each salesperson include the product line they are responsible for, the language they are responsible for, the type of customer they are responsible for, and their current workload. Calculate the matching degree between the actual product demand tags and the tags for each salesperson, and filter out salespeople whose matching degree exceeds a preset threshold.

3. The user behavior habit analysis method based on big data fusion according to claim 2, characterized in that, Step two includes the following specific steps: Step 21: Obtain the product price and purchase quantity based on the customer's actual product needs; obtain the product profit margin by dividing the difference between the product price and the product cost by the product price; obtain the purchase scale by dividing the product quantity and the product price by the maximum historical total purchase amount; and obtain the product profitability by weighted summing of the product profit margin and the purchase scale. Step 22: Calculate customer rating based on customer size and credit risk, obtain customer importance score based on the ratio of customer's historical total purchase amount to historical total sales, and obtain customer stability by weighted summation of customer rating score and customer importance score. Step 23: Analyze the ease of production based on the degree of customization; subtract the standard production cycle of the product from the difference between the customer's required delivery date and the current date to obtain the buffer time; obtain the delivery capacity based on the ratio of the buffer time to the standard production cycle; and obtain the operational feasibility by weighted summation of the ease of production and delivery capacity. Step 24: Obtain the order value by weighting and summing the product profitability, customer stability, and operational feasibility.

4. The user behavior habit analysis method based on big data fusion according to claim 3, characterized in that, Step three includes the following specific steps: Step 31: Obtain communication frequency, response time, and conversation duration based on customer dialogue records; obtain a communication frequency assessment value based on the ratio of communication frequency to standard communication frequency; obtain a response duration assessment value based on the ratio of standard response time to response duration; obtain a conversation duration assessment value based on the ratio of conversation duration to standard conversation duration; and obtain an order urgency assessment value by weighted summation of the communication frequency assessment value, response duration assessment value, and conversation duration assessment value. Step 32: Based on customer dialogue records, obtain the customer's procurement stage, which includes information collection stage, demand definition stage, product comparison stage, decision-making stage and transaction stage. Obtain the total number of customers and the final number of transactions for each historical procurement stage. Obtain the historical procurement transaction rate based on the ratio of the final number of transactions to the total number of customers. Obtain the corresponding procurement transaction rate based on the procurement stage in which the customer is located. Step 33: Weight the order urgency assessment value and the purchase completion rate to obtain the customer's purchase intention assessment value.

5. The user behavior habit analysis method based on big data fusion according to claim 4, characterized in that, Step four includes the following specific steps: Step 41: Calculate the customer's priority level by weighted summation of the order's value assessment and the customer's purchase intention assessment; Step 42: Obtain the screened sales personnel and determine their competency level based on their product knowledge, communication skills, and customer management. Step 43: Select at least three sales representatives who are highly matched with the customer's priority level and the salesperson's ability level, and set them as pre-selected sales representatives.

6. The user behavior habit analysis method based on big data fusion according to claim 5, characterized in that, Step five includes the following specific steps: Step 51: Obtain the pre-selected sales staff's understanding of each part of the product. The sales staff's understanding of each part of the product includes the frequency of introduction and the number of words in the introduction of each part in the past. Specifically, the word count understanding value is obtained by dividing the average number of words in the introduction of the corresponding part by the average number of words in the introduction of each part. The frequency understanding value is obtained by dividing the average frequency of introduction of the corresponding part by the average frequency of introduction of each part. The word count understanding value and the frequency understanding value are weighted and summed to obtain the understanding of the corresponding part. Step 52: Obtain the frequency of customer inquiries about each part of the product during the dialogue phase, as well as the proportion of historical inquiries about the corresponding question to all historical inquiries about that part. The smaller the proportion, the fewer people know about the question. Divide the proportion of historical inquiries about the corresponding question to all historical inquiries about that part by the average question proportion and then take the reciprocal to obtain the difficulty of solving the question. Divide the frequency of inquiries about each part of the product during the dialogue phase by the standard inquiry frequency to obtain the inquiry frequency value. Multiply the average difficulty of solving the question for the corresponding part by the inquiry frequency value to obtain the customer awareness value for the corresponding part. Step 53: Obtain the customer perception value and sales perception status of the corresponding part. Divide the sales perception status of the corresponding part by the customer perception value to obtain the predicted understanding value of the corresponding part. Add up the predicted understanding values ​​of all parts and average them to obtain the customer predicted understanding value of the product. Step 54: Select the pre-selected sale corresponding to the largest customer forecast understanding value and set it as the customer's optimal sale.

7. A user behavior habit analysis system based on big data fusion, used to implement the user behavior habit analysis method based on big data fusion as described in any one of claims 1-6, characterized in that, include: The product demand analysis module is used to collect customer dialogue records through the language processing module and identify the customer's actual product needs based on the characteristics of the language used by the customer and the habits of the target market. The initial sales screening module is used to initially screen sales personnel based on actual product needs; The order value analysis module is used to analyze the value of orders based on customers' actual product needs. The purchase intention analysis module is used to analyze customers' purchasing stages and urgency based on customer conversation records, and to analyze customers' purchase intentions based on purchasing stages and urgency. The customer priority analysis module is used to analyze customer priority based on order value and customer purchase intention. The optimal sales allocation module analyzes customer priority based on order value and customer purchase intention, matches several sales representatives based on customer priority, and selects the optimal sales representative based on the customer's historical inquiries about the order and the corresponding sales representative's knowledge of the order's products.

8. An electronic device, characterized in that, include: A memory and a processor, wherein the memory stores a computer program that can be called by the processor, and the processor executes the user behavior habit analysis method based on big data fusion as described in any one of claims 1-6 by calling the computer program stored in the memory.

9. A computer-readable storage medium, characterized in that, The system stores instructions that, when executed on a computer, cause the computer to perform the user behavior habit analysis method based on big data fusion as described in any one of claims 1-6.

Images