Method, device and equipment for analyzing user vehicle repurchase intention and storage medium

Abstract

The application relates to a user vehicle repurchase willingness analysis method and device, equipment and a storage medium, wherein the method comprises the following steps: acquiring a vehicle repurchase willingness test data set, and extracting at least one effective feature from the test data set; inputting the at least one effective feature into a pre-trained cold start model to obtain a user repurchase vehicle probability and a logistic regression coefficient; and obtaining a willingness analysis result based on the user repurchase vehicle probability and a pre-trained hot start model. The application introduces a user response to operation access and an attention mechanism, so that the user repurchase willingness analysis result is more accurate, marketing side crowd selection is facilitated, operation cost is reduced, and a single rate is improved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, apparatus, device, and storage medium for analyzing users' willingness to repurchase vehicles. Background Technology

[0002] Car dealerships primarily acquire customers through in-store visits, phone calls, auto shows, and referrals from friends. While these customers are high-potential, the channels are too limited and user information is easily obtained by competitors, making the conversion rate uncontrollable.

[0003] Currently, related technologies can create a multi-view car image database for car angle recognition, car grade recognition, and car brand recognition based on car styling images; create a multi-view car styling database for styling scoring to automate the evaluation of car exterior styling; in addition, related technologies can also group features by utilizing feature correlation, vertically split the dataset using the grouped feature sets, perform independent model prediction, and then perform model fusion.

[0004] However, existing technologies do not incorporate dynamic data from user reviews, do not consider changes in user focus, and have limited application scenarios, which urgently need to be addressed. Summary of the Invention

[0005] This application provides a method, apparatus, device, and storage medium for analyzing users' willingness to repurchase vehicles, in order to solve the problems of existing technologies not incorporating dynamic data of user evaluation, not considering changes in user concerns, and having narrow application scenarios.

[0006] The first aspect of this application provides a method for analyzing users' vehicle repurchase intentions, applied in the offline training phase. The method includes the following steps: obtaining a first training sample set and a second training sample set of vehicle repurchase intentions; extracting at least one first effective feature and at least one second effective feature from the first training sample set and the second training sample set, respectively; training a preset logistic regression algorithm based on the at least one first effective feature to obtain a trained cold-start model; and training a preset coefficient of variation method based on the at least one second effective feature to obtain a trained hot-start model, thereby obtaining intention analysis results based on the cold-start model and the hot-start model.

[0007] Optionally, in one embodiment of this application, obtaining the first and second training sample sets of vehicle repurchase intention includes: obtaining non-interactive data of the target vehicle, wherein the non-interactive data includes at least one of vehicle basic attribute data, vehicle owner basic attribute data, vehicle maintenance data, and vehicle upkeep data; obtaining interactive data of the target user, wherein the interactive data includes at least one of APP browsing data, private domain conversation data, and vehicle owner return visit data; generating the first test sample set based on the non-interactive data; and generating the second test sample set based on the interactive data.

[0008] Optionally, in one embodiment of this application, the step of extracting at least one first effective feature and at least one second effective feature from the first training sample set and the second training sample set respectively includes: preprocessing the indicator data of the first training sample set and the second training sample set respectively to obtain the at least one first effective feature and at least one second effective feature after data cleaning, data integration, data transformation and data reduction operations.

[0009] A second aspect of this application provides a method for analyzing users' vehicle repurchase intentions, applied in an online analysis phase. The method includes the following steps: obtaining a test dataset of vehicle repurchase intentions and extracting at least one effective feature from the test dataset; inputting the at least one effective feature into a pre-trained cold-start model to obtain the user's vehicle repurchase probability and logistic regression coefficient; and obtaining the intention analysis result based on the user's vehicle repurchase probability and the pre-trained hot-start model.

[0010] Optionally, in one embodiment of this application, obtaining the intention analysis result based on the user's vehicle repurchase probability and the pre-trained hot-start model includes: conducting marketing outreach to the user based on the user's vehicle repurchase probability to obtain user response data; and inputting the user response data into the hot-start model to obtain the intention analysis result.

[0011] Optionally, in one embodiment of this application, the step of inputting the user response data into the hot start model to obtain the intention analysis result includes: optimizing the feature weights of the coefficient of variation method in the hot start model based on the logistic regression coefficient, the preset attention mechanism, and the intention analysis result.

[0012] A third aspect of this application provides an analysis device for user vehicle repurchase intention, applied in an offline training phase, comprising: a first acquisition module for acquiring a first training sample set and a second training sample set of vehicle repurchase intention; an extraction module for extracting at least one first effective feature and at least one second effective feature from the first training sample set and the second training sample set, respectively; and a training module for training a preset logistic regression algorithm based on the at least one first effective feature to obtain a trained cold-start model, and training a preset coefficient of variation method based on the at least one second effective feature to obtain a trained hot-start model, so as to obtain intention analysis results based on the cold-start model and the hot-start model.

[0013] Optionally, in one embodiment of this application, the acquisition module includes: a first acquisition unit, configured to acquire non-interactive data of the target vehicle, wherein the non-interactive data includes at least one of vehicle basic attribute data, vehicle owner basic attribute data, vehicle maintenance data, and vehicle upkeep data; a second acquisition unit, configured to acquire interactive data of the target user, wherein the interactive data includes at least one of APP browsing data, private domain conversation data, and vehicle owner return visit data; a first generation unit, configured to generate a first test sample set based on the non-interactive data; and a second generation unit, configured to generate a second test sample set based on the interactive data.

[0014] Optionally, in one embodiment of this application, the extraction module includes: a preprocessing unit, used to preprocess the indicator data of the first training sample set and the second training sample set respectively, to obtain at least one first effective feature and at least one second effective feature after data cleaning, data integration, data transformation and data reduction operations.

[0015] A fourth aspect of this application provides an analysis device for users' vehicle repurchase intentions, applied in an online analysis phase, comprising: a second acquisition module for acquiring a test dataset of vehicle repurchase intentions and extracting at least one effective feature from the test dataset; a first analysis module for inputting the at least one effective feature into a pre-trained cold-start model to obtain the user's vehicle repurchase probability and logistic regression coefficient; and a second analysis module for obtaining intention analysis results based on the user's vehicle repurchase probability and the pre-trained hot-start model.

[0016] Optionally, in one embodiment of this application, the second analysis module includes: a reach unit, used to reach out to users for marketing based on the probability of users repurchasing vehicles, and obtain user response data; and a response unit, used to input the user response data into the hot start model to obtain the intention analysis result.

[0017] Optionally, in one embodiment of this application, the response unit is specifically used to optimize the feature weights of the coefficient of variation method in the hot start model based on the logistic regression coefficients, the preset attention mechanism, and the intention analysis results.

[0018] A fifth aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for analyzing user vehicle repurchase intention as described in the above embodiments.

[0019] A sixth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for analyzing user vehicle repurchase intentions.

[0020] Therefore, the embodiments of this application have the following beneficial effects:

[0021] The embodiments of this application can obtain a test dataset of vehicle repurchase intentions and extract at least one effective feature from the test dataset; input the at least one effective feature into a pre-trained cold start model to obtain the user's vehicle repurchase probability and logistic regression coefficient; and obtain the intention analysis results based on the user's vehicle repurchase probability and the pre-trained warm start model. This application, by introducing a user response and attention mechanism to operational outreach, makes the user repurchase intention analysis results more accurate, facilitates audience selection by the marketing side, reduces operational costs, and increases the conversion rate. Therefore, it solves the problems of existing technologies that do not incorporate dynamic data from user evaluations, do not consider changes in user focus, and have narrow application scenarios.

[0022] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0023] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0024] Figure 1 This is a flowchart illustrating a method for analyzing user vehicle repurchase intentions during the offline training phase, according to an embodiment of this application.

[0025] Figure 2 A schematic diagram illustrating the execution logic of a method for analyzing user vehicle repurchase intention during the offline training phase, provided as an embodiment of this application;

[0026] Figure 3A schematic diagram illustrating the execution logic of a method for analyzing a user's willingness to repurchase a vehicle, provided as an embodiment of this application;

[0027] Figure 4 This is a flowchart illustrating a method for analyzing user vehicle repurchase intentions during the online analysis phase, according to an embodiment of this application.

[0028] Figure 5 This is an example diagram of an analysis device for analyzing user vehicle repurchase intentions during the offline training phase, according to an embodiment of this application.

[0029] Figure 6 This is an example diagram of an analysis device for analyzing user vehicle repurchase intentions applied in the online analysis phase according to an embodiment of this application;

[0030] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0031] Among them, 10-analysis device for user vehicle repurchase intention applied in the offline training stage, 20-analysis device for user vehicle repurchase intention applied in the online analysis stage, 100-first acquisition module, 200-extraction module, 300-training module, 400-second acquisition module, 500-first analysis module, 600-second analysis module, 701-memory, 702-processor, and 703-communication interface. Detailed Implementation

[0032] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0033] The following describes a method, apparatus, device, and storage medium for analyzing user vehicle repurchase intentions according to embodiments of this application, with reference to the accompanying drawings. Addressing the problems mentioned in the background art, this application provides a method for analyzing user vehicle repurchase intentions. In this method, a test dataset of vehicle repurchase intentions is obtained, and at least one effective feature is extracted from the test dataset. The at least one effective feature is input into a pre-trained cold-start model to obtain the user's vehicle repurchase probability and logistic regression coefficient. Based on the user's vehicle repurchase probability and the pre-trained warm-start model, the intention analysis result is obtained. This application introduces a user response and attention mechanism to operational outreach, making the user repurchase intention analysis results more accurate, facilitating audience selection by the marketing side, reducing operational costs, and increasing the conversion rate. Therefore, it solves the problems of existing technologies that do not incorporate dynamic data from user evaluations, do not consider changes in user focus, and have limited application scenarios.

[0034] Specifically, Figure 1 This is a flowchart illustrating a method for analyzing user vehicle repurchase intentions during the offline training phase, as provided in an embodiment of this application.

[0035] like Figure 1 As shown, the method for analyzing a user's willingness to repurchase a vehicle includes the following steps:

[0036] In step S101, the first training sample set and the second training sample set of vehicle repurchase intention are obtained.

[0037] The embodiments of this application can first collect a sample dataset, which includes user data (such as age, gender, whether they own multiple vehicles), user driving behavior data (such as usage time, usage frequency, fuel consumption, fuel balance, accelerator pedal data, brake pedal data, and engine speed), user touchpoint behavior data (such as activity participation, private domain conversations, APP access, user feedback), vehicle data (total mileage, vehicle age, frequency of malfunctions, frequency of maintenance, frequency of visits to service stations), and post-operation effect data.

[0038] User data, user driving behavior data, user touchpoint behavior data, and vehicle data are mainly used for training and prediction of the subsequent intention analysis model, while performance data is mainly used for evaluating the model's performance.

[0039] In practice, the embodiments of this application may also supplement the sample data as needed, such as adding reputation data and transaction data.

[0040] Furthermore, based on the collected sample dataset, the first and second training sample sets related to users' vehicle repurchase intentions are obtained, providing data support for the construction of subsequent repurchase intention analysis models.

[0041] Optionally, in one embodiment of this application, obtaining the first training sample set and the second training sample set of vehicle repurchase intention includes: obtaining non-interactive data of the target vehicle, wherein the non-interactive data includes at least one of vehicle basic attribute data, vehicle owner basic attribute data, vehicle maintenance data, and vehicle upkeep data; obtaining interactive data of the target user, wherein the interactive data includes at least one of APP browsing data, private domain conversation data, and vehicle owner return visit data; generating the first test sample set based on the non-interactive data; and generating the second test sample set based on the interactive data.

[0042] It should be noted that the embodiments of this application collect data from various channels, and classify, organize and design indicators for the acquired data to obtain non-interactive data of vehicles and interactive data of users.

[0043] The vehicle's non-interactive data includes static attribute data such as basic vehicle attributes and owner attributes, as well as dynamic vehicle data such as maintenance and repair data. User interactive data includes app browsing data, private chat data, and owner follow-up data, such as... Figure 2 As shown.

[0044] Therefore, embodiments of this application can generate a first test sample set based on non-interactive data and a second test sample set based on interactive data, thereby providing data support for subsequent model training.

[0045] In step S102, at least one first effective feature and at least one second effective feature are extracted from the first training sample set and the second training sample set, respectively.

[0046] After obtaining the first and second training sample sets of vehicle repurchase intentions, embodiments of this application can further perform preprocessing operations such as data cleaning, data integration, data transformation, and data reduction on the first and second training sample sets, and perform feature processing on the preprocessed data, such as... Figure 3 As shown, this allows us to obtain effective features.

[0047] Optionally, in one embodiment of this application, at least one first effective feature and at least one second effective feature are extracted from the first training sample set and the second training sample set, respectively, including: preprocessing the indicator data of the first training sample set and the second training sample set to obtain at least one first effective feature and at least one second effective feature after data cleaning, data integration, data transformation and data reduction operations.

[0048] It should be noted that after obtaining the first and second training sample sets, it is necessary to first observe the format, missing values, outliers and noise of the data in the training sample sets, and then perform preprocessing operations such as data cleaning, handling missing values, handling outliers, normalization, standardization, discretization or data transformation on the data in the training sample sets.

[0049] Furthermore, for each type of data in the preprocessed training sample set, corresponding features are generated, as detailed below:

[0050] 1. User data may include the following feature types:

[0051] Age, gender, whether they own multiple vehicles, and user type (individual user / corporate user);

[0052] 2. User driving behavior data may include the following feature types:

[0053] Average daily number of vehicle usages per user, average monthly number of vehicle usage days per user, distribution of vehicle usage times per user, average fuel consumption over the past 30 days, average daily number of rapid accelerations, average daily number of rapid decelerations, average daily number of sharp turns, average daily number of dangerous braking actions, average daily number of high engine speeds, average daily carbon dioxide emissions, average daily fatigue driving time, average daily nighttime driving time, and average daily driving time in congested areas.

[0054] 3. User touchpoint behavior data may include the following feature types: number of times participating in offline activities, number of times viewing trade-in policies, number of times vehicle evaluations, number of times viewing new models, number of times participating in WeChat questionnaire surveys, and number of times responding to WeChat group / trade-in topics;

[0055] 4. Vehicle data may include the following feature types: total vehicle mileage, vehicle age, number of vehicle fault types, average number of vehicle faults per day, number of times a single vehicle fault occurs, average number of times a user enters the station per month, and number of days between the last entry and the last entry.

[0056] 5. Maintenance / repair data may include the following feature types:

[0057] Number of repairs in the past 30 days, number of maintenance services in the past 30 days, repair costs in the past 30 days, maintenance costs in the past 30 days, number of days since the last repair, and number of days since the last maintenance.

[0058] After generating the corresponding features, embodiments of this application may further perform feature engineering and processing operations such as feature selection, feature construction, feature scaling, feature dimensionality reduction, and feature importance evaluation on all generated features based on actual business needs, to obtain a first effective feature and at least one second effective feature.

[0059] Therefore, this application optimizes the number of features by preprocessing the data and denoising the features, thereby greatly improving the quality of the data.

[0060] In step S103, a preset logistic regression algorithm is trained based on at least one first effective feature to obtain a trained cold start model, and a preset coefficient of variation method is trained based on at least one second effective feature to obtain a trained hot start model, so as to obtain the intention analysis results based on the cold start model and the hot start model.

[0061] After obtaining the first and second effective features, further, for the cold start case, the embodiments of this application can use the first effective features to train the logistic regression algorithm to obtain a trained cold start model, and use the AUC objective function to compare the difference between the predicted user repurchase data and the actual repurchase data to optimize the cold start model; under the warm start case, the second effective features and the MSE objective function are used to train the coefficient of variation method to obtain a trained warm start model.

[0062] Therefore, the embodiments of this application use AUC and MSE as objective functions and the weights of each model as decision variables to solve for the optimal weight vector corresponding to the objective function and optimize the model, effectively ensuring the accuracy and reliability of predicting users' vehicle repurchase intentions.

[0063] The method for analyzing user vehicle repurchase intentions in the offline training phase, as proposed in this application, involves obtaining a first and a second training sample set of vehicle repurchase intentions; extracting at least one first effective feature and at least one second effective feature from the first and second training sample sets, respectively; training a preset logistic regression algorithm based on at least one first effective feature to obtain a trained cold-start model; and training a preset coefficient of variation method based on at least one second effective feature to obtain a trained hot-start model. The intention analysis results are then obtained based on the cold-start and hot-start models. This application improves the accuracy of user repurchase intention analysis results by selecting appropriate training data to train the logistic regression algorithm and the coefficient of variation method, facilitating audience selection for marketing purposes and reducing operating costs.

[0064] Figure 4 This is a flowchart illustrating a method for analyzing user vehicle repurchase intentions during the online analysis phase, as provided in an embodiment of this application.

[0065] like Figure 4 As shown, the method for analyzing a user's willingness to repurchase a vehicle includes the following steps:

[0066] In step S401, a test dataset of vehicle repurchase intentions is obtained, and at least one valid feature is extracted from the test dataset.

[0067] In the embodiments of this application, a test dataset containing relevant data on users' willingness to repurchase vehicles can first be collected. The dataset is then subjected to preprocessing operations such as data cleaning, data transformation, and data reduction, as well as operations such as feature selection, feature construction, feature scaling, feature dimensionality reduction, and feature importance evaluation, thereby obtaining effective features of the test dataset and providing reliable data for the analysis of vehicle repurchase intentions.

[0068] In step S402, at least one valid feature is input into the pre-trained cold start model to obtain the probability of a user repurchasing a vehicle and the logistic regression coefficient.

[0069] After extracting effective features, in the case of a cold start, embodiments of this application can input the effective features into a pre-trained cold start model, thereby generating the probability of a user repurchasing a vehicle and the logistic regression coefficient through a logistic regression algorithm.

[0070] In step S403, the intention analysis results are obtained based on the probability of users repurchasing vehicles and the pre-trained hot start model.

[0071] After obtaining the probability of a user repurchasing a vehicle and the logistic regression coefficient, the embodiments of this application, in the case of a hot start, can send advertising content to users based on the probability of a user repurchasing a vehicle, analyze user repurchase demand-related behaviors, such as increased click data and increased inquiry questions, and collect user response data in a timely manner. Thus, through user response data, the probability of a user repurchasing a vehicle, and the pre-trained hot start model, a user repurchase intention score is obtained, and the intention analysis results are obtained.

[0072] Optionally, in one embodiment of this application, the intention analysis result is obtained based on the user's vehicle repurchase probability and a pre-trained hot-start model, including: conducting marketing outreach to users based on the user's vehicle repurchase probability to obtain user response data; and inputting the user response data into the hot-start model to obtain the intention analysis result.

[0073] It should be noted that the embodiments of this application can select users with a high repurchase intention based on the probability of users repurchasing vehicles, and conduct marketing outreach to users in batches according to the existing operational resources (such as pushing advertisements on the APP, initiating private domain hot topics, telephone follow-ups, offline invitations, etc.) to obtain the corresponding user data. The response data after the user outreach is then input into the hot start model to mine the repurchase intention score of existing car owners through the scoring model and obtain the intention analysis results.

[0074] Optionally, in one embodiment of this application, user response data is input into a warm-start model to obtain intention analysis results, including: optimizing the feature weights of the coefficient of variation method in the warm-start model based on logistic regression coefficients, a preset attention mechanism, and user intention analysis results.

[0075] In the embodiments of this application, in the case of a hot start, the feature weights of the coefficient of variation method are derived from the logistic regression coefficients and the response data.

[0076] In actual implementation, the feature weight construction can incorporate an attention mechanism, namely, whether users respond highly to certain brands or models. For brands or models that users respond highly to, relevant advertising content can be frequently sent to observe changes in user response data, and the feature weights of the coefficient of variation method under hot start can be adjusted based on user response to relevant content.

[0077] Therefore, the embodiments of this application obtain user repurchase intentions based on existing vehicle and user data and through a preset model; in addition, the embodiments of this application introduce a dynamic repurchase intention evaluation model, which adjusts feature weights through user feedback on operational advertisements, effectively improving the accuracy of user repurchase intention prediction.

[0078] The method for analyzing user vehicle repurchase intentions in the online analysis phase, as proposed in this application, involves obtaining a test dataset of vehicle repurchase intentions and extracting at least one effective feature from it. This effective feature is then input into a pre-trained cold-start model to obtain the user's vehicle repurchase probability and logistic regression coefficient. Based on the user's vehicle repurchase probability and the pre-trained warm-start model, the intention analysis results are obtained. This application introduces a user response and attention mechanism to operational outreach, making the user repurchase intention analysis results more accurate, facilitating audience selection for marketing, reducing operational costs, and increasing the conversion rate.

[0079] Secondly, the analysis apparatus for analyzing user vehicle repurchase intention according to an embodiment of this application is described with reference to the accompanying drawings.

[0080] Figure 5 This is a block diagram of a device for analyzing user vehicle repurchase intentions during the offline training phase, as described in this application embodiment.

[0081] like Figure 5 As shown, the analysis device 10 for user vehicle repurchase intention includes: a first acquisition module 100, an extraction module 200, and a training module 300.

[0082] The first acquisition module 100 is used to acquire the first training sample set and the second training sample set of vehicle repurchase intention.

[0083] The extraction module 200 is used to extract at least one first effective feature and at least one second effective feature from the first training sample set and the second training sample set, respectively.

[0084] The training module 300 is used to train a preset logistic regression algorithm based on at least one first effective feature to obtain a trained cold start model, and to train a preset coefficient of variation method based on at least one second effective feature to obtain a trained hot start model, so as to obtain intention analysis results based on the cold start model and the hot start model.

[0085] Optionally, in one embodiment of this application, the acquisition module 100 includes: a first acquisition unit, a second acquisition unit, a first generation unit, and a second generation unit.

[0086] The first acquisition unit is used to acquire non-interactive data of the target vehicle, wherein the non-interactive data includes at least one of the following: vehicle basic attribute data, vehicle owner basic attribute data, vehicle maintenance data, and vehicle upkeep data.

[0087] The second data acquisition unit is used to acquire the target user's interaction data, which includes at least one of the following: APP browsing data, private domain conversation data, and car owner return visit data.

[0088] The first generation unit is used to generate the first test sample set based on non-interactive data.

[0089] The second generation unit is used to generate a second set of test samples based on the interaction data.

[0090] Optionally, in one embodiment of this application, the extraction module 200 includes: a preprocessing unit, used to preprocess the index data of the first training sample set and the second training sample set respectively, to obtain at least one first effective feature and at least one second effective feature after data cleaning, data integration, data transformation and data reduction operations.

[0091] The device for analyzing user vehicle repurchase intentions during the offline training phase, as proposed in this application, includes a first acquisition module for acquiring a first training sample set and a second training sample set of vehicle repurchase intentions; an extraction module for extracting at least one first effective feature and at least one second effective feature from the first training sample set and the second training sample set, respectively; and a training module for training a preset logistic regression algorithm based on at least one first effective feature to obtain a trained cold-start model, and training a preset coefficient of variation method based on at least one second effective feature to obtain a trained hot-start model, thereby obtaining intention analysis results based on the cold-start model and the hot-start model. This application, by selecting appropriate training data to train the logistic regression algorithm and the coefficient of variation method, makes the user repurchase intention analysis results more accurate, facilitates audience selection by the marketing side, and reduces operating costs.

[0092] Figure 6 This is a block diagram of a device for analyzing user vehicle repurchase intentions in the online analysis phase, as described in this application embodiment.

[0093] like Figure 6 As shown, the user vehicle repurchase intention analysis device 20 includes: a second acquisition module 400, a first analysis module 500, and a second analysis module 600.

[0094] The second acquisition module 400 is used to acquire a test dataset of vehicle repurchase intentions and extract at least one valid feature from the test dataset.

[0095] The first analysis module 500 is used to input at least one effective feature into the pre-trained cold start model to obtain the probability of users repurchasing vehicles and the logistic regression coefficient.

[0096] The second analysis module 600 is used to obtain intention analysis results based on the probability of users repurchasing vehicles and the pre-trained hot start model.

[0097] Optionally, in one embodiment of this application, the second analysis module 300 includes: an access unit and a response unit.

[0098] The "reach unit" is used to reach out to users with marketing efforts based on their probability of repeat vehicle purchases and to obtain user response data.

[0099] The response unit is used to input user response data into the warm-start model to obtain intention analysis results.

[0100] Optionally, in one embodiment of this application, the response unit is specifically used to optimize the feature weights of the coefficient of variation method in the hot start model based on the logistic regression coefficients, the preset attention mechanism, and the intention analysis results.

[0101] It should be noted that the explanation of the aforementioned method embodiment for analyzing users' willingness to repurchase vehicles also applies to the device for analyzing users' willingness to repurchase vehicles in this embodiment, and will not be repeated here.

[0102] The analysis device for user vehicle repurchase intention proposed in the online analysis stage according to the embodiments of this application obtains a test dataset of vehicle repurchase intention and extracts at least one effective feature from the test dataset; inputs the at least one effective feature into a pre-trained cold start model to obtain the user's vehicle repurchase probability and logistic regression coefficient; and obtains the intention analysis result based on the user's vehicle repurchase probability and the pre-trained hot start model. This application, by introducing a user response and attention mechanism to operational outreach, makes the user repurchase intention analysis results more accurate, facilitates audience selection by the marketing side, reduces operating costs, and increases the conversion rate.

[0103] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include:

[0104] The memory 701, the processor 702, and the computer program stored on the memory 701 and executable on the processor 702.

[0105] When the processor 702 executes the program, it implements the method for analyzing users' willingness to repurchase vehicles provided in the above embodiments.

[0106] Furthermore, electronic devices also include:

[0107] Communication interface 703 is used for communication between memory 701 and processor 702.

[0108] The memory 701 is used to store computer programs that can run on the processor 702.

[0109] The memory 701 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0110] If the memory 701, processor 702, and communication interface 703 are implemented independently, then the communication interface 703, memory 701, and processor 702 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 7 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0111] Optionally, in a specific implementation, if the memory 701, processor 702, and communication interface 703 are integrated on a single chip, then the memory 701, processor 702, and communication interface 703 can communicate with each other through an internal interface.

[0112] The processor 702 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0113] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described method for analyzing a user's willingness to repurchase a vehicle.

[0114] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0115] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least one, such as two, three, etc., unless otherwise explicitly specified.

[0116] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0117] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0118] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0119] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0120] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0121] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for analyzing users' willingness to repurchase vehicles, characterized in that, When applied to the offline training phase, the method includes the following steps: Obtain the first and second training sample sets of vehicle repurchase intentions; Extract at least one first effective feature and at least one second effective feature from the first training sample set and the second training sample set, respectively; The preset logistic regression algorithm is trained based on at least one first effective feature to obtain a trained cold start model. The difference between the predicted user repurchase data and the actual repurchase data is compared using the AUC objective function to optimize the cold start model. The preset coefficient of variation method is trained based on at least one second effective feature and the MSE objective function to obtain a trained hot start model. The intention analysis results are obtained based on the cold start model and the hot start model. When applied to the online analysis phase, the method includes the following steps: Obtain a test dataset of vehicle repurchase intention, and extract at least one valid feature from the test dataset; The at least one effective feature is input into the pre-trained cold start model to obtain the probability of users repurchasing vehicles and the logistic regression coefficient; Based on the probability of users repurchasing vehicles and the pre-trained hot start model, the intention analysis results are obtained; The first and second training sample sets for obtaining vehicle repurchase intentions include: Obtain non-interactive data of the target vehicle, wherein the non-interactive data includes at least one of the following: vehicle basic attribute data, vehicle owner basic attribute data, vehicle maintenance data, and vehicle upkeep data; Acquire interaction data of target users, wherein the interaction data includes at least one of APP browsing data, private domain conversation data, and car owner return visit data; The first training sample set is generated based on the non-interactive data; The second training sample set is generated based on the interaction data.

2. The method according to claim 1, characterized in that, The step of extracting at least one first effective feature and at least one second effective feature from the first training sample set and the second training sample set, respectively, includes: The indicator data of the first and second training sample sets are preprocessed respectively to obtain at least one first effective feature and at least one second effective feature after data cleaning, data integration, data transformation and data reduction operations.

3. The method according to claim 1, characterized in that, The intention analysis results obtained based on the user's repurchase vehicle probability and the pre-trained hot-start model include: Based on the probability of users repurchasing vehicles, marketing outreach is conducted to users to obtain user response data; The user response data is input into the hot start model to obtain the intention analysis results.

4. The method according to claim 3, characterized in that, The step of inputting the user response data into the hot-start model to obtain the intention analysis results includes: Based on the logistic regression coefficients, the preset attention mechanism, and the intention analysis results, the feature weights of the coefficient of variation method in the hot start model are optimized.

5. An analysis device for users' vehicle repurchase intention, characterized in that, Applied to the offline training phase, including: The first acquisition module is used to acquire the first and second training sample sets of vehicle repurchase intentions. The extraction module is used to extract at least one first effective feature and at least one second effective feature from the first training sample set and the second training sample set, respectively; The training module is used to train a preset logistic regression algorithm based on at least one first effective feature to obtain a trained cold start model, and to optimize the cold start model by comparing the difference between user repurchase prediction data and actual repurchase data using the AUC objective function. The module is also used to train a preset coefficient of variation method based on at least one second effective feature and the MSE objective function to obtain a trained hot start model, so as to obtain the intention analysis results based on the cold start model and the hot start model. When applied to the online analysis phase, the device includes: The second acquisition module is used to acquire the test dataset of vehicle repurchase intention and extract at least one effective feature from the test dataset; The first analysis module is used to input the at least one effective feature into the pre-trained cold start model to obtain the probability of user repeat purchase of vehicles and the logistic regression coefficient. The second analysis module is used to obtain the intention analysis results based on the user's repurchase vehicle probability and the pre-trained hot start model; The first acquisition module includes: The first acquisition unit is used to acquire non-interactive data of the target vehicle, wherein the non-interactive data includes at least one of the following: vehicle basic attribute data, vehicle owner basic attribute data, vehicle maintenance data, and vehicle upkeep data. The second collection unit is used to acquire the interaction data of the target user, wherein the interaction data includes at least one of APP browsing data, private domain conversation data, and car owner return visit data; The first generation unit is used to generate the first training sample set based on the non-interactive data; The second generation unit is used to generate the second training sample set based on the interaction data.

6. An electronic device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, the processor executing the program to implement the method for analyzing user vehicle repurchase intention as described in any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the method for analyzing user vehicle repurchase intention as described in any one of claims 1-4.

Images