Product dynamic pricing method, device, equipment and medium

Abstract

The invention relates to the field of artificial intelligence, can be applied to business system platforms of financial science and technology, medical health and the like, and discloses a product dynamic pricing method, device and equipment and a medium, and the method comprises the steps: collecting multi-dimensional data related to a to-be-priced product from a plurality of channels; performing data preprocessing and feature extraction on the multi-dimensional data, and extracting key features for dynamic pricing from the multi-dimensional data; inputting the key features into a pre-trained pricing model, and generating a corresponding initial pricing result according to the key features; and obtaining risk monitoring data and client interaction data in a preset time period, adjusting the initial pricing result according to the risk monitoring data and the client interaction data, and generating a dynamic pricing result. The multi-dimensional data is automatically priced through the pricing model, and the pricing result is dynamically adjusted by acquiring risk monitoring and customer feedback, so that the product pricing efficiency and flexibility are improved, and the requirements of customers and markets are quickly and dynamically met.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a product dynamic pricing method, device, equipment and medium. Background Art

[0002] At present, with the increasing complexity of the business environment, pricing strategies play an increasingly important role in product operations. In the operation of traditional products, product pricing often only considers the cost and market positioning of the product itself. However, this pricing method is not suitable for the pricing needs of health-related products that are increasingly concerned by consumers.

[0003] For example, in the financial field, the pricing of health insurance products mainly relies on the experience and manual analysis of actuaries, involving a large amount of manual calculation and analysis to determine the premium price, which is time-consuming and inefficient. For example, in the medical and health field, hospitals or physical examination institutions usually launch various physical examination package products, but the pricing of physical examination packages is usually fixed, and a one-time price is set according to the package content (such as examination items, service scope, etc.). This pricing method lacks flexibility and is difficult to adjust according to the specific needs of customers and market changes.

[0004] Therefore, the current traditional pricing methods for health products still have problems such as low efficiency and lack of flexibility that need to be addressed urgently. Summary of the Invention

[0005] In view of the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide a product dynamic pricing method, device, equipment and medium that can be applied to the medical field, financial technology or other related fields. Its main purpose is to improve the efficiency and flexibility of product pricing, realize efficient personalized pricing, and quickly and dynamically adapt to customer and market needs.

[0006] The technical solutions of the present invention are as follows:

[0007] A first aspect of the present invention provides a product dynamic pricing method, comprising:

[0008] Collect multi-dimensional data related to the product to be priced from multiple channels;

[0009] performing data preprocessing and feature extraction on the multi-dimensional data, and extracting key features for dynamic pricing from the multi-dimensional data;

[0010] Inputting the key features into a pre-trained pricing model to generate corresponding initial pricing results based on the key features;

[0011] The risk monitoring data and customer interaction data within a preset time period are obtained, and the initial pricing result is adjusted according to the risk monitoring data and customer interaction data to generate a dynamic pricing result.

[0012] A second aspect of the present invention provides a product dynamic pricing device, comprising:

[0013] A data collection module, used to collect multi-dimensional data associated with the product to be priced from multiple channels;

[0014] A data processing module, configured to perform data preprocessing and feature extraction on the multi-dimensional data, and extract key features for dynamic pricing from the multi-dimensional data;

[0015] A pricing output module, configured to input the key features into a pre-trained pricing model and generate corresponding initial pricing results based on the key features;

[0016] The dynamic adjustment module is used to obtain risk monitoring data and customer interaction data within a preset time period, adjust the initial pricing result according to the risk monitoring data and customer interaction data, and generate a dynamic pricing result.

[0017] A third aspect of the present invention provides a computer device comprising at least one processor; and

[0018] a memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the above-mentioned product dynamic pricing method.

[0020] A fourth aspect of the present invention provides a non-volatile computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by one or more processors, the one or more processors can execute the above-mentioned product dynamic pricing method.

[0021] Beneficial effects: The present invention discloses a method, apparatus, device and medium for dynamic pricing of products. Compared with the prior art, the embodiments of the present invention collect multi-dimensional data associated with the product to be priced from multiple channels; perform data preprocessing and feature extraction on the multi-dimensional data, and extract key features for dynamic pricing from the multi-dimensional data; input the key features into a pre-trained pricing model, and generate corresponding initial pricing results based on the key features; obtain risk monitoring data and customer interaction data within a preset time period, and adjust the initial pricing results based on the risk monitoring data and customer interaction data to generate dynamic pricing results. On the basis of automatically pricing multi-dimensional data through the pricing model, risk monitoring and customer interaction feedback are also obtained to dynamically adjust the pricing results, thereby improving product pricing efficiency and flexibility, and realizing efficient personalized pricing, thereby quickly and dynamically adapting to customer and market needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the solutions in the present invention, a brief introduction is given below to the drawings required for use in describing the embodiments of the present invention. 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 any creative work.

[0023] Figure 1 A schematic diagram of an application environment of the product dynamic pricing method provided by an embodiment of the present invention;

[0024] Figure 2 A flow chart of a product dynamic pricing method provided by an embodiment of the present invention;

[0025] Figure 3 A flowchart of step S202 in the product dynamic pricing method provided by an embodiment of the present invention;

[0026] Figure 4 A flow chart of step S203 in the product dynamic pricing method provided by an embodiment of the present invention;

[0027] Figure 5 A flow chart of step S204 in the product dynamic pricing method provided by an embodiment of the present invention;

[0028] Figure 6 A flowchart of optimizing and updating the pricing model in the dynamic product pricing method provided by an embodiment of the present invention;

[0029] Figure 7 A schematic diagram of the functional modules of a device for dynamic product pricing according to an embodiment of the present invention;

[0030] Figure 8A schematic diagram of the hardware structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0031] To make the objectives, technical solutions, and effects of the present invention more clear and distinct, the present invention is further described in detail below. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. The embodiments of the present invention are described below with reference to the accompanying drawings.

[0032] The product dynamic pricing method provided by the embodiment of the present invention can be applied to Figure 1 In an application environment, the system includes a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired and / or wireless communication links, etc.

[0033] The user may use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications may be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as knowledge reading applications, web browser applications, search applications, instant messaging tools, email clients, and / or social platform software (for example only).

[0034] The first terminal device 101 , the second terminal device 102 , and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0035] The server 105 may be a server that provides various services, such as a backend server that provides support for the content browsed by the user using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (for example only). The backend server may analyze and process the received user requests and other data, and feed back the processing results (such as web pages, information, or data obtained or generated according to the user request) to the terminal device. The server 105 may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server 105 may also be a server for a distributed system, or a server combined with a blockchain.

[0036] It should be noted that the product dynamic pricing method provided in the embodiments of the present application can generally be executed by the first terminal device 101, the second terminal device 102, or the third terminal device 103. Accordingly, the product dynamic pricing apparatus provided in the embodiments of the present invention can also be provided in the first terminal device 101, the second terminal device 102, or the third terminal device 103. Alternatively, the product dynamic pricing method provided in the embodiments of the present invention can generally be executed by the server 105. Accordingly, the product dynamic pricing apparatus provided in the embodiments of the present invention can generally be provided in the server 105.

[0037] It should be understood that the numbers of the above terminal devices, networks and servers are merely illustrative and any number of terminal devices, networks and servers may be provided as required.

[0038] like Figure 2 As shown, the product dynamic pricing method provided by the embodiment of the present invention specifically includes the following steps:

[0039] S201. Collect multi-dimensional data associated with the product to be priced from multiple channels.

[0040] In this embodiment, data of different dimensions related to the product to be priced is collected through multiple channels. The product to be priced can be health-related products, such as health insurance, physical examination packages, health consulting services, etc., or other types of products, such as property insurance, life insurance, etc. When the product to be priced is health-related, the multi-dimensional data collected in this embodiment includes customer basic data, health data, medical data, claims data, and market data. Among them, customer basic data can be obtained from the customer relationship management system (CRM) to obtain basic customer information such as age, gender, occupation, and income level; health data can be obtained from medical partners, wearable devices, or electronic health records (EHR) to obtain customer health data such as physical examination reports, medical records, chronic disease status, and lifestyle habits (exercise, diet, smoking, etc.); medical data can be obtained from hospitals and clinics. Medical records, prescription drug usage, surgical records, etc.; claims data can be obtained from the insurance claims system to obtain claims history, including claim amount, claim frequency, and claim reason; market data can be obtained from market research institutions, industry reports, or competitor analysis to obtain market pricing information, competitor strategies, and changes in customer demand.

[0041] Specifically, data can be obtained in real time through API interfaces, data streams or IoT devices, or data can be imported in batches from databases or data warehouses on a regular basis to obtain multi-dimensional data from different sources covering different aspects, thereby ensuring that the input information of the pricing model is comprehensive and can more accurately describe customer characteristics and market environment.

[0042] For example, for health insurance products in the financial sector, customers' health data (such as physical examination reports, chronic disease history), claims data (historical claim amounts, claim frequency, claim reasons), medical data (medical records, surgical records) and market data (such as competitor pricing) can be collected to provide a basis for personalized pricing of health insurance products.

[0043] For physical examination package products in the medical and health field, customers' health data (such as physical examination needs and past physical examination records) and market data (such as market pricing trends and changes in customer demand) can be collected to optimize the pricing of physical examination packages.

[0044] S202: Perform data preprocessing and feature extraction on the multi-dimensional data, and extract key features for dynamic pricing from the multi-dimensional data.

[0045] In this embodiment, data preprocessing operations such as cleaning, standardization, and encoding are performed on the collected multi-dimensional data to remove noise, fill missing values, correct outliers, and convert the data into a unified format and range, such as normalization or standardization. This improves data quality and ensures the accuracy and consistency of the input data. Furthermore, key features highly correlated with dynamic pricing are extracted from the large amount of multi-dimensional data, thereby removing other features that are ineffective in pricing. This reduces the amount of data processed by the subsequent model and improves pricing accuracy.

[0046] For example, for health insurance products in the financial sector, customers' health data and claims data are cleaned and standardized to ensure data quality, and key features such as health risk level, medical treatment time distribution, medical expense distribution, etc. are extracted for subsequent personalized product pricing.

[0047] For physical examination package products in the medical and health field, by cleaning and standardizing customer health data and market data, key features such as historical physical examination frequency, historical physical examination fees, market price change trends, etc. are extracted to optimize the pricing of physical examination packages, so as to improve the scientific nature and market adaptability of package pricing.

[0048] S203: Input the key features into a pre-trained pricing model, and generate corresponding initial pricing results based on the key features.

[0049] In this embodiment, a pricing model is pre-built. Specifically, ensemble learning methods (such as random forests and gradient boosting trees) and deep learning methods (such as convolutional neural networks and long-short-term memory networks) can be used in combination with traditional insurance actuarial models to build an initial pricing model. The initial pricing model is then adjusted by collecting training data to optimize model performance and obtain a trained pricing model. The trained pricing model can quickly and efficiently generate personalized initial pricing plans based on key input features such as the customer's health status, lifestyle, risk preferences, and market demand, thereby meeting the needs of different customers while also improving pricing efficiency and market adaptability.

[0050] For example, for health insurance products in the financial sector, key features such as a customer's health risk level, chronic disease probability, and historical claims amount are extracted and fed into a trained pricing model to generate initial insurance pricing. For example, for a customer with high blood pressure and a history of claims, the model might assign a higher initial premium.

[0051] For health checkup package products in the healthcare sector, key features such as abnormalities in a customer's physical examination indicators and lifestyle habits are input into the pricing model to generate an initial package price. For example, for a young customer with normal physical examination indicators, the model may assign a lower initial package price.

[0052] S204: Obtain risk monitoring data and customer interaction data within a preset time period, adjust the initial pricing result according to the risk monitoring data and customer interaction data, and generate a dynamic pricing result.

[0053] In this embodiment, to further enhance pricing flexibility and adaptability, risk monitoring data and customer interaction data within a preset time period are obtained after generating the initial pricing results. For example, the system monitors changes in a customer's health risks (e.g., abnormal health indicators) and market dynamics (e.g., fluctuations in medical costs) in real time. After generating the initial pricing results, risk monitoring data for a preset time period, such as the past month, three months, or six months, is obtained as a basis for adjustment. Furthermore, after generating the initial pricing results, the results are pushed to the customer for review. The customer can view the currently generated initial pricing plan in real time and provide interaction data (e.g., premium acceptance, coverage requirements, etc.) through an interactive interface. Furthermore, the system can analyze customer renewal and cancellation behavior data to comprehensively obtain customer interaction data. Based on the risk monitoring data and customer interaction data, the initial pricing results are adjusted, such as by offering short-term discounts or increasing coverage when customer feedback is poor, or by increasing health insurance premiums when monitored risks increase, thereby generating the final dynamic pricing results. By further dynamically adjusting the initial pricing results based on the efficient generation of the initial pricing results, the pricing plan can quickly respond to market changes and customer feedback, better adapting to the needs of different customer groups and the market.

[0054] For example, for health insurance products in the financial sector, after generating initial pricing results (including premiums and coverage) through a pricing model, real-time monitoring of changes in customers' health risks (such as abnormal health indicators) and market dynamics (such as fluctuations in medical costs) is carried out. Customer feedback on premiums and coverage is collected through a customer service platform. Premiums are dynamically adjusted based on real-time monitoring data and customer feedback. For example, premiums may be appropriately reduced for customers whose health improves, and appropriately increased for customers with increased health risks. These dynamic adjustments ensure the scientific nature and customer acceptance of pricing plans, thereby improving market competitiveness.

[0055] For medical health checkup package products, after generating the initial pricing results (including package prices and package items) through the pricing model, we monitor changes in customers' health data and market trends in real time, and collect customer feedback on the price and content of the medical checkup packages through the customer service platform. We dynamically adjust the price and content of the medical checkup packages based on real-time monitoring data and customer feedback. For example, we add additional examination items for customers with higher health needs; provide discounted packages for price-sensitive customers, and so on. Through dynamic adjustments, we optimize the compatibility between the packages and customers.

[0056] In the above embodiment, the present invention discloses a method for dynamic pricing of products, which collects multi-dimensional data associated with the product to be priced from multiple channels; performs data preprocessing and feature extraction on the multi-dimensional data, and extracts key features for dynamic pricing from the multi-dimensional data; inputs the key features into a pre-trained pricing model, and generates corresponding initial pricing results based on the key features; obtains risk monitoring data and customer interaction data within a preset time period, and adjusts the initial pricing results based on the risk monitoring data and customer interaction data to generate dynamic pricing results. On the basis of automatically pricing the multi-dimensional data through the pricing model, risk monitoring and customer interaction feedback are also obtained to dynamically adjust the pricing results, thereby improving product pricing efficiency and flexibility, achieving efficient personalized pricing, and thus quickly and dynamically adapting to customer and market needs.

[0057] In one embodiment, Figure 3 As shown, step S202 includes:

[0058] S301: performing data cleaning on the multi-dimensional data to remove noise data in the multi-dimensional data;

[0059] S302: Integrate the cleaned multi-dimensional data to obtain standardized data in a unified format;

[0060] S303: Extract features from the standardized data and perform correlation analysis on the extracted features to obtain correlations between features of different dimensions and the product to be priced;

[0061] S304. Filter the features of all dimensions according to the correlation results to obtain key features for dynamic pricing.

[0062] In this embodiment, when preprocessing the collected multidimensional data, the noise data in the multidimensional data is first identified and removed through data cleaning, that is, errors, outliers or inconsistent data points in the data. Specifically, statistical methods (such as box plots, Z-scores) can be used to identify outliers in the data. For example, for the customer's income data, the Z-score is calculated, and data points exceeding 3 times the standard deviation are regarded as outliers; or the logical consistency in the data can be checked. For example, the customer's age should not be a negative number, the income should not be a negative value, etc. The identified noise data can be directly deleted or corrected. For example, for outliers, the median or mean can be used to replace them. For missing data, interpolation, mean filling, median filling or machine learning methods are used to fill in the missing data, etc. Data cleaning improves the accuracy and consistency of the data, reduces the impact of noise data on subsequent model predictions, and improves the prediction ability.

[0063] After removing noise data through data cleaning, since multi-dimensional data comes from different channels, the data format may be inconsistent or there may be inconsistencies in time, space, etc. Therefore, multi-dimensional data from different channels is integrated. For example, data from different sources are aligned according to a common key (such as customer ID), and then data in different formats are converted to a unified format. For example, the date format is unified into YYYY-MM-DD, etc., so as to obtain standardized numbers with a unified format. Data from different sources and different formats are merged and converted to make them consistent and operational.

[0064] Afterwards, feature engineering is used to convert the original multi-dimensional data into features that are more suitable for model processing. Specifically, principal component analysis (PCA), cluster analysis and other methods can be used to extract the intrinsic features of the data in each dimension, and correlation analysis such as the Pearson correlation coefficient and the Spearman rank correlation coefficient can be used to calculate the correlation between the features of different dimensions and the products to be priced. According to the results of the correlation analysis, the key features for dynamic pricing are screened from the features of all dimensions, that is, the key features that have an important impact on the pricing target are selected from all features, such as selecting features with a correlation higher than a certain threshold, or sorting the correlation results of all features from large to small, selecting features with a preset proportion before sorting, etc. Through feature screening, the key features that have an important impact on the pricing target are selected to improve the predictive ability of the model.

[0065] In one embodiment, Figure 4 As shown, step S203 includes:

[0066] S401: Input the key features into a pre-trained pricing model, fuse the key features, and predict several pricing indicators;

[0067] S402: Perform actuarial pricing processing according to the plurality of pricing indicators to generate corresponding initial pricing results.

[0068] In this embodiment, the pricing model is a pre-trained machine learning model (random forest, gradient boosting tree, support vector machine, etc.) or a deep learning model (convolutional neural network, long short-term memory network, etc.), which is trained and optimized through historical data so that the pricing model can make pricing predictions for the input key features. Specifically, the key features are input into the pre-trained pricing model, and the filtered key features are fused, for example, by feature concatenation, weighted summation, or through the feature interaction mechanism within the model. The model processes the fused features through its internal algorithm logic and outputs several pricing indicators, which may include basic premium, risk premium, discount factor, etc. Based on the predicted pricing indicators, actuarial processing is performed in combination with the existing insurance actuarial system. For example, actuarial pricing rules are set in advance based on industry standards, company policies, and risk assessment results, which may include risk adjustment factors, discount policies, minimum premium limits, etc. The predicted pricing indicators (such as basic premium, risk premium, discount factor, etc.) are combined with the actuarial pricing rules to calculate, thereby obtaining the final initial pricing result, which serves as the basis for subsequent dynamic adjustments. Through actuarial pricing processing, we ensure that pricing results comply with industry standards and company policies, and improve the scientificity and rationality of the pricing process.

[0069] In one embodiment, Figure 5 As shown, step S204 includes:

[0070] S501. Monitor market dynamics and customer health changes within a preset time period to obtain corresponding risk monitoring data;

[0071] S502: receiving feedback information input by the customer after viewing the initial pricing result, and obtaining corresponding customer interaction data;

[0072] S503: Match the risk monitoring data and customer interaction data with preset dynamic adjustment rules to obtain a corresponding personalized adjustment strategy;

[0073] S504: Adjust the initial pricing result based on the personalized adjustment strategy to generate a corresponding dynamic pricing result.

[0074] In this embodiment, when making dynamic pricing adjustments, market dynamics and changes in customer health are monitored. Market dynamics refer to changes in external market factors that affect pricing, such as interest rate fluctuations, competitor strategy adjustments, etc., and customer health changes refer to changes in customer health status, such as new disease diagnoses, abnormal fluctuations in health indicators, etc. Specifically, a real-time data interface (API) can be set up or data can be crawled regularly to obtain market dynamics data, such as interest rate changes, competitor pricing adjustments, industry policy changes, etc., from financial market data platforms, industry reports, competitor websites, etc., to ensure that the latest market dynamics information can be obtained in a timely manner; customer health data, such as new physical examination reports, changes in chronic disease indicators, changes in lifestyle habits, etc., can be obtained from medical institutions, health management platforms, wearable devices, etc. through real-time data interfaces or regular data synchronization, so as to obtain the customer's latest health status in a timely manner. Comprehensive monitoring of market dynamics and customer health changes as risk monitoring data can be used to timely identify risk factors and provide a basis for dynamic adjustments.

[0075] In addition, by setting up customer feedback channels such as online customer service, mobile applications, questionnaires, etc., we collect feedback from customers after they view the initial pricing results, such as their acceptance of the price, their needs for coverage, and their needs for package content. For example, users may feedback that the initial premium for health insurance products is too high, and suggest increasing the coverage or providing installment payment options, or raise questions about certain protection terms and require further explanation or adjustment; for physical examination package products, they may feedback that the package price is too high, and hope to reduce some unnecessary physical examination items or provide coupons, etc.

[0076] Based on the collected and monitored risk monitoring data and customer interaction data, the system matches them with pre-set dynamic adjustment rules. These rules, which include price adjustment ranges, coverage adjustments, and package optimization, are pre-set based on factors such as company policies, market strategies, and risk assessment results. During dynamic pricing adjustments, the system matches risk monitoring data and customer interaction data with the pre-set dynamic adjustment rules to determine the appropriate adjustment measures, such as reducing premiums or increasing coverage. This generates a personalized adjustment strategy, ensuring that the adjusted pricing plan meets customer needs and adapts to market changes. Based on the personalized adjustment strategy, the initial pricing results are adjusted. These adjustments may include price adjustments, coverage optimization, package content changes, coupons, and promotional offers. The final dynamic pricing results are then generated and notified to customers through the system. By implementing personalized dynamic pricing adjustments based on pre-set dynamic adjustment rules based on risk monitoring data and customer interaction data, the system meets customer needs and market changes while ensuring that price adjustments are within feasible limits, preventing price adjustments from harming the company's interests or failing to meet regulatory requirements.

[0077] Optionally, when making dynamic pricing adjustments based on risk monitoring and feedback, customers can view the pricing plan generation process and adjustment basis in real time through the online platform, and obtain personalized health advice (such as how to reduce health risks and improve lifestyle habits), making the pricing process transparent and convenient, and achieving in-depth interaction between customers and the pricing system, which not only provides personalized pricing plans but also health advice, significantly improving the customer experience.

[0078] In one embodiment, the method further comprises:

[0079] The performance indicators of the pricing model are collected, and whether the preset update conditions are met is confirmed based on the performance indicators. If so, the pricing model is optimized and updated according to the model update strategy.

[0080] In this embodiment, to ensure the performance of the pricing model, appropriate performance indicators are selected based on the characteristics of the pricing model and business requirements, that is, quantitative indicators for evaluating the performance of the pricing model, such as prediction accuracy (such as mean square error, R 2 The performance of the pricing model is analyzed by collecting data on performance indicators, such as the accuracy of the model (e.g., cross-validation error), model generalization ability (e.g., cross-validation error), response time (e.g., model prediction delay), etc. And based on the model performance requirements, preset update conditions are set. For example, when the model's prediction accuracy is lower than the preset threshold (e.g., R 2 When the value is lower than 0.8, or the generalization ability of the model significantly decreases, an update is triggered. The collected performance indicator data is regularly checked to determine whether the preset update conditions are met. If the conditions are met, the model update process begins. Alternatively, in other embodiments, in addition to triggering model updates based on performance indicators, update conditions based on time (such as automatic updates every quarter) or data volume (such as automatic updates every 1,000 new data items accumulated) may also be used. This embodiment is not limited to this.

[0081] After the update is triggered, the pricing model is optimized and updated according to the preset model update strategy. The specific model update strategy can be, for example, retraining the model, fine-tuning model parameters, introducing new features or data, etc. After the update is completed, the updated pricing model is verified to ensure that its performance meets expectations. For example, the prediction accuracy of the model is evaluated through a test set, or the performance of the model in actual business is evaluated through A / B testing, etc. By regularly collecting performance indicators and triggering model updates according to conditions, it is possible to ensure that the pricing model always maintains good performance, can adapt to market changes, changes in customer demand and changes in data distribution in a timely manner, improve prediction accuracy and generalization capabilities, and thus maintain the rationality and reliability of the pricing plan.

[0082] In one embodiment, Figure 6As shown, the pricing model is optimized and updated according to the model update strategy, including:

[0083] S601: Collect historical pricing cases of the pricing model and construct an optimized data set based on the historical pricing cases, wherein the optimized data set includes pricing status data and pricing adjustment actions;

[0084] S602: Randomly initialize the policy network, sample pricing status data from the optimized data set and input it into the pricing model, control the pricing model to perform corresponding pricing adjustment actions according to the initialization policy and calculate corresponding reward values, and repeat the above sampling process to obtain multiple sampling sequences;

[0085] S603: Update the parameters of the policy network based on the cumulative rewards of each sampling sequence, and optimize the pre-built objective function according to the parameters of the policy network before and after the update;

[0086] S604, looping through the above pricing and policy updating processes until the objective function satisfies a preset condition and an optimal policy network is obtained;

[0087] S605: Based on the optimal policy network, control the pricing model to output pricing according to the optimal policy to optimize pricing results.

[0088] In this embodiment, after a model update is triggered, the pricing model is optimized and updated using a reinforcement learning algorithm. Specifically, reinforcement learning is performed using a policy network to optimize the pricing model. The policy network is a deep learning model that directly learns the mapping relationship from state to action, i.e., the policy function. Using the policy gradient method, the parameters of the policy network can be optimized, enabling the intelligent agent, i.e., the pricing model, to achieve higher cumulative rewards in the environment.

[0089] When updating the pricing model through the policy network, historical pricing cases are first extracted from the pricing system, including customer characteristics, initial pricing results, market dynamics data, customer feedback, and final pricing adjustment results. The data in these historical pricing cases is then divided into two parts: pricing status data and pricing adjustment actions. Pricing status data refers to the environmental state during the pricing process, including customer characteristics (such as age, health status, income level, etc.), market environment (such as competitor pricing, market interest rates, etc.), and initial pricing results (such as premium price, coverage, etc.). Pricing adjustment actions refer to the adjustment measures taken based on the pricing status data, such as price adjustment and coverage optimization. Based on these two parts of data, an optimized dataset is constructed for training and optimizing the policy network.

[0090] When training and optimizing the policy network, the network's parameters are randomly initialized, and a reward function is defined. For example, if a customer accepts the pricing proposal, the reward is positive; if the customer rejects the proposal, the reward is negative; if the pricing proposal results in higher profits, the reward increases; if the pricing proposal results in a decrease in market share, the reward decreases, and so on. Under the current policy, a set of state, action, and reward sequences are sampled from the environment. Specifically, pricing state data is sampled from the optimization dataset and input into the pricing model. Based on the policy network's current policy, a pricing adjustment action is selected, such as adjusting the premium or coverage. The reward value is calculated based on the results of the pricing adjustment (such as customer acceptance or profit change). The reward value reflects the effectiveness of the pricing adjustment action. This sampling and adjustment process is repeated to generate several sampling sequences, each of which contains a series of pricing state data, adjustment actions, and reward values. The reward value calculation provides direct feedback for the policy network's learning, helping to optimize the policy.

[0091] The cumulative reward value is calculated for each sampling sequence, reflecting the overall effect of the pricing adjustments taken during the entire sampling sequence. The policy gradient is then calculated based on the cumulative reward value. The policy network parameters are then updated using gradient ascent. A pre-built objective function, which measures the performance of the policy network, is then optimized based on the parameters before and after the update. The optimal policy network is obtained by iterating through the sampling, adjustment, reward calculation, and policy network parameter updates until the objective function meets pre-set conditions, such as maximizing the cumulative reward value or terminating the loop when a pre-set number of iterations has been reached. The optimal policy network is then used to control the pricing output of the pricing model. Specifically, for a new pricing case, pricing status data is input into the pricing model, which, guided by the optimal policy, selects the optimal pricing adjustment action, thereby optimizing the pricing results. The pricing model is then optimized and updated using a reinforcement learning algorithm, ensuring that the pricing solutions output by the optimized pricing model better meet customer needs and market changes, further enhancing the flexibility and adaptability of dynamic pricing.

[0092] It should be noted that there is not necessarily a certain order between the above steps. A person skilled in the art can understand, based on the description of the embodiments of the present invention, that in different embodiments, the above steps may have different execution orders, that is, they may be executed in parallel, or may be executed interchangeably, etc.

[0093] Further references Figure 7 , as a response to the above Figure 2 The present invention provides an embodiment of a product dynamic pricing device, which is similar to the embodiment of the present invention. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0094] like Figure 7 As shown, the product dynamic pricing device 70 described in this embodiment includes:

[0095] A data collection module 701 is used to collect multi-dimensional data associated with the product to be priced from multiple channels;

[0096] A data processing module 702 is configured to perform data preprocessing and feature extraction on the multi-dimensional data, and extract key features for dynamic pricing from the multi-dimensional data;

[0097] The pricing output module 703 is used to input the key features into a pre-trained pricing model and generate corresponding initial pricing results based on the key features;

[0098] The dynamic adjustment module 704 is configured to obtain risk monitoring data and customer interaction data within a preset time period, adjust the initial pricing result according to the risk monitoring data and customer interaction data, and generate a dynamic pricing result.

[0099] The module referred to in the present invention refers to a series of computer program instruction segments that can perform specific functions. It is more suitable for describing the execution process of the product dynamic pricing method than a program. For the specific implementation methods of each module, please refer to the corresponding method embodiments above, which will not be repeated here.

[0100] In one embodiment, the data processing module 702 includes:

[0101] Performing data cleaning on the multi-dimensional data to remove noise data in the multi-dimensional data;

[0102] Integrate the cleaned multi-dimensional data to obtain standardized data in a unified format;

[0103] Extracting features from the standardized data and performing correlation analysis on the extracted features to obtain correlations between features of different dimensions and the product to be priced;

[0104] Based on the correlation results, the key features for dynamic pricing are screened from the features of all dimensions.

[0105] In one embodiment, the pricing output module 703 includes:

[0106] An input prediction unit, configured to input the key features into a pre-trained pricing model, fuse the key features, and predict a plurality of pricing indicators;

[0107] The actuarial output unit is used to perform actuarial pricing processing according to the multiple pricing indicators and generate corresponding initial pricing results.

[0108] In one embodiment, the dynamic adjustment module 704 includes:

[0109] The risk monitoring unit is used to monitor market dynamics and changes in customer health within a preset time period and obtain corresponding risk monitoring data;

[0110] a feedback interaction unit, configured to receive feedback information input by a customer after viewing the initial pricing result, and obtain corresponding customer interaction data;

[0111] A strategy matching unit, configured to match the risk monitoring data and customer interaction data with preset dynamic adjustment rules to obtain a corresponding personalized adjustment strategy;

[0112] A dynamic adjustment unit is used to adjust the initial pricing result based on the personalized adjustment strategy to generate a corresponding dynamic pricing result.

[0113] In one embodiment, the apparatus 70 further comprises:

[0114] The model update module is used to collect performance indicators of the pricing model, confirm whether the preset update conditions are met based on the performance indicators, and if so, optimize and update the pricing model according to the model update strategy.

[0115] In one embodiment, the model updating module includes:

[0116] a collection and construction unit, configured to collect historical pricing cases of the pricing model and construct an optimized data set based on the historical pricing cases, wherein the optimized data set includes pricing status data and pricing adjustment actions;

[0117] a sampling unit configured to randomly initialize the policy network, sample pricing state data from the optimized data set and input the data into the pricing model, control the pricing model to select and execute corresponding pricing adjustment actions according to the initialization policy and calculate corresponding reward values, and repeat the above sampling process to obtain a plurality of sampling sequences;

[0118] A strategy updating unit, configured to update the parameters of the strategy network based on the cumulative rewards of each sampling sequence, and optimize the pre-constructed objective function according to the parameters of the strategy network before and after the update;

[0119] A loop control unit, configured to loop through the above-mentioned pricing and policy updating processes until an optimal policy network is obtained when the objective function satisfies a preset condition;

[0120] The model optimization unit is used to control the pricing model to output pricing according to the optimal strategy based on the optimal strategy network, so as to optimize the pricing result.

[0121] In one embodiment, the multi-dimensional data includes: customer basic data, health data, medical data, claims data and / or market data.

[0122] In the above embodiment, the present invention discloses a product dynamic pricing device, which collects multi-dimensional data associated with the product to be priced from multiple channels; performs data preprocessing and feature extraction on the multi-dimensional data, and extracts key features for dynamic pricing from the multi-dimensional data; inputs the key features into a pre-trained pricing model, and generates corresponding initial pricing results based on the key features; obtains risk monitoring data and customer interaction data within a preset time period, and adjusts the initial pricing results based on the risk monitoring data and customer interaction data to generate dynamic pricing results. On the basis of automatically pricing the multi-dimensional data through the pricing model, risk monitoring and customer interaction feedback are also obtained to dynamically adjust the pricing results, thereby improving product pricing efficiency and flexibility, achieving efficient personalized pricing, and thus quickly and dynamically adapting to customer and market needs.

[0123] Another embodiment of the present invention provides a computer device, such as Figure 8 As shown, the computer device 80 includes:

[0124] One or more processors 801 and memory 802, Figure 8 In the description, a processor 801 is used as an example. The processor 801 and the memory 802 can be connected via a bus or other means. Figure 8 The bus connection is taken as an example.

[0125] The processor 801 is used to complete various control logics of the computer device 80. It can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a single-chip microcomputer, an ARM (Acorn RISC Machine) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination of these components. In addition, the processor 801 can also be any traditional processor, microprocessor, or state machine. The processor 801 can also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors combined with a DSP, and / or any other such configuration.

[0126] Memory 802, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions corresponding to the dynamic product pricing method in the embodiments of the present invention. Processor 801 executes the non-volatile software programs, instructions, and modules stored in memory 802 to execute various functional applications and data processing of computer device 80, thereby implementing the dynamic product pricing method in the above-mentioned method embodiments.

[0127] The memory 802 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of the computer device 80, etc. In addition, the memory 802 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 802 may optionally include a memory remotely located relative to the processor 801, and these remote memories may be connected to the computer device 80 via a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. One or more units are stored in the memory 802, and when executed by one or more processors 801, the steps of the product dynamic pricing method in any of the above-mentioned method embodiments are executed.

[0128] In the above embodiment, the present invention discloses a computer device that collects multi-dimensional data associated with the product to be priced from multiple channels; performs data preprocessing and feature extraction on the multi-dimensional data, extracts key features for dynamic pricing from the multi-dimensional data; inputs the key features into a pre-trained pricing model, and generates corresponding initial pricing results based on the key features; obtains risk monitoring data and customer interaction data within a preset time period, adjusts the initial pricing results based on the risk monitoring data and customer interaction data, and generates dynamic pricing results. On the basis of automatically pricing the multi-dimensional data through the pricing model, risk monitoring and customer interaction feedback are also obtained to dynamically adjust the pricing results, thereby improving product pricing efficiency and flexibility, achieving efficient personalized pricing, and thus quickly and dynamically adapting to customer and market needs.

[0129] An embodiment of the present invention provides a non-volatile computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by one or more processors, the steps of the product dynamic pricing method in any of the above method embodiments are executed.

[0130] In the above embodiment, the present invention discloses a non-volatile computer-readable storage medium, which collects multi-dimensional data associated with the product to be priced from multiple channels; performs data preprocessing and feature extraction on the multi-dimensional data, extracts key features for dynamic pricing from the multi-dimensional data; inputs the key features into a pre-trained pricing model, and generates corresponding initial pricing results based on the key features; obtains risk monitoring data and customer interaction data within a preset time period, adjusts the initial pricing results based on the risk monitoring data and customer interaction data, and generates dynamic pricing results. On the basis of automatically pricing the multi-dimensional data through the pricing model, risk monitoring and customer interaction feedback are also obtained to dynamically adjust the pricing results, thereby improving product pricing efficiency and flexibility, achieving efficient personalized pricing, and thus quickly and dynamically adapting to customer and market needs.

[0131] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0132] The present invention can be used in a wide variety of general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like. The present invention can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present invention can also be practiced in distributed computing environments in which tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0133] In summary, the present invention discloses a method, device, equipment and medium for dynamic pricing of a product, wherein the method includes: collecting multi-dimensional data associated with the product to be priced from multiple channels; performing data preprocessing and feature extraction on the multi-dimensional data, and extracting key features for dynamic pricing from the multi-dimensional data; inputting the key features into a pre-trained pricing model, and generating corresponding initial pricing results based on the key features; obtaining risk monitoring data and customer interaction data within a preset time period, and adjusting the initial pricing results based on the risk monitoring data and customer interaction data to generate dynamic pricing results. On the basis of automatically pricing the multi-dimensional data through the pricing model, risk monitoring and customer interaction feedback are also obtained to dynamically adjust the pricing results, thereby improving product pricing efficiency and flexibility, achieving efficient personalized pricing, and thus quickly and dynamically adapting to customer and market needs.

[0134] Of course, those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware (such as a processor, controller, etc.) through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes in the above-described method embodiments. The storage medium can be a memory, a magnetic disk, a floppy disk, a flash memory, an optical storage device, etc.

[0135] It should be noted that if any software tools or components not developed by our company appear in the examples of this application, they are for illustration purposes only and do not represent actual use. It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications shall fall within the scope of protection of the appended claims.

Claims

1. A product dynamic pricing method, characterized in that: include: Collect multi-dimensional data related to the product to be priced from multiple channels; performing data preprocessing and feature extraction on the multi-dimensional data, and extracting key features for dynamic pricing from the multi-dimensional data; Inputting the key features into a pre-trained pricing model to generate corresponding initial pricing results based on the key features; The risk monitoring data and customer interaction data within a preset time period are obtained, and the initial pricing result is adjusted according to the risk monitoring data and customer interaction data to generate a dynamic pricing result.

2. The product dynamic pricing method according to claim 1, characterized in that: The performing data preprocessing and feature extraction on the multi-dimensional data, and extracting key features for dynamic pricing from the multi-dimensional data, includes: Performing data cleaning on the multi-dimensional data to remove noise data in the multi-dimensional data; Integrate the cleaned multi-dimensional data to obtain standardized data in a unified format; Extracting features from the standardized data and performing correlation analysis on the extracted features to obtain correlations between features of different dimensions and the product to be priced; Based on the correlation results, the key features for dynamic pricing are screened from the features of all dimensions.

3. The product dynamic pricing method according to claim 1, characterized in that: Inputting the key features into a pre-trained pricing model and generating corresponding initial pricing results based on the key features includes: Inputting the key features into a pre-trained pricing model, fusing the key features and predicting several pricing indicators; Actuarial pricing processing is performed based on the several pricing indicators to generate corresponding initial pricing results.

4. The product dynamic pricing method according to claim 1, characterized in that: The acquiring of risk monitoring data and customer interaction data within a preset time period, adjusting the initial pricing result according to the risk monitoring data and customer interaction data, and generating a dynamic pricing result, includes: Monitor market dynamics and customer health changes within a preset time period to obtain corresponding risk monitoring data; receiving feedback information input by the customer after viewing the initial pricing result, and obtaining corresponding customer interaction data; Matching the risk monitoring data and customer interaction data with preset dynamic adjustment rules to obtain corresponding personalized adjustment strategies; The initial pricing result is adjusted based on the personalized adjustment strategy to generate a corresponding dynamic pricing result.

5. The product dynamic pricing method according to claim 1, characterized in that: The method also includes: The performance indicators of the pricing model are collected, and whether the preset update conditions are met is confirmed based on the performance indicators. If so, the pricing model is optimized and updated according to the model update strategy.

6. The product dynamic pricing method according to claim 5, characterized in that: Optimizing and updating the pricing model according to the model update strategy includes: Collecting historical pricing cases of the pricing model, and constructing an optimized data set based on the historical pricing cases, wherein the optimized data set includes pricing status data and pricing adjustment actions; Randomly initializing the policy network, sampling pricing state data from the optimized data set and inputting it into the pricing model, controlling the pricing model to select and execute corresponding pricing adjustment actions according to the initialization strategy and calculate corresponding reward values, and repeating the above sampling process to obtain multiple sampling sequences; Update the parameters of the policy network based on the cumulative rewards of each sampling sequence, and optimize the pre-built objective function according to the parameters of the policy network before and after the update; The above pricing and policy updating process is executed cyclically until the optimal policy network is obtained when the objective function satisfies the preset conditions; Based on the optimal policy network, the pricing model is controlled to output pricing according to the optimal policy to optimize the pricing result.

7. The product dynamic pricing method according to any one of claims 1 to 6, characterized in that: The multi-dimensional data includes: customer basic data, health data, medical data, claims data and / or market data.

8. A product dynamic pricing device, characterized in that: include: A data collection module, used to collect multi-dimensional data associated with the product to be priced from multiple channels; A data processing module, configured to perform data preprocessing and feature extraction on the multi-dimensional data, and extract key features for dynamic pricing from the multi-dimensional data; A pricing output module, configured to input the key features into a pre-trained pricing model and generate corresponding initial pricing results based on the key features; The dynamic adjustment module is used to obtain risk monitoring data and customer interaction data within a preset time period, adjust the initial pricing result according to the risk monitoring data and customer interaction data, and generate a dynamic pricing result.

9. A computer device, characterized in that: comprising at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the product dynamic pricing method according to any one of claims 1 to 7.

10. A non-volatile computer-readable storage medium, characterized in that: The non-volatile computer-readable storage medium stores computer-executable instructions, which, when executed by one or more processors, enable the one or more processors to execute the product dynamic pricing method according to any one of claims 1 to 7.

Images