CN111681051A - Purchasing intention degree prediction method, device, storage medium and terminal - Google Patents
Purchasing intention degree prediction method, device, storage medium and terminal Download PDFInfo
- Publication number
- CN111681051A CN111681051A CN202010512538.6A CN202010512538A CN111681051A CN 111681051 A CN111681051 A CN 111681051A CN 202010512538 A CN202010512538 A CN 202010512538A CN 111681051 A CN111681051 A CN 111681051A
- Authority
- CN
- China
- Prior art keywords
- machine learning
- model
- learning model
- training
- samples
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Game Theory and Decision Science (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method, a device, a storage medium and a terminal for predicting purchase intention, wherein the method comprises the steps of obtaining target access data capable of representing online matching behaviors of a user to be predicted; extracting a characteristic value of the target access data under a preset characteristic dimension, and inputting the characteristic value of the target access data into a prediction model so that the prediction model classifies the characteristic value of the target access data; and then the purchase intention degree of the user to be predicted can be obtained through the prediction model. Therefore, the method for predicting the purchase intention can simply and effectively mine the high-purchase intention latent customers, so that subsequent marketing resources can be utilized to the maximum extent, and the automobile sales volume under the limited marketing cost is increased.
Description
Technical Field
The invention relates to the technical field of automobile internet marketing, in particular to a purchasing intention degree prediction method, a purchasing intention degree prediction device, a storage medium and a terminal.
Background
The automobile internet marketing field is the cross field of automobile sales and internet sales. On the basis of the continuous development of the internet technology, in order to meet different vehicle using requirements of different users, the service of online customizing of the automobile is generated.
The automobile online customization opens a channel for the configuration sheet to be selected by the user, and the user can go to a 4s store to place the order online or directly place the order online after the internet matching is completed. The process of matching products is an important channel for a user to know the automobile, and few or few users can directly decide to buy the automobile after matching once.
Therefore, how to find a user who really has a purchasing intention from a plurality of users who select products becomes a problem which needs to be solved urgently at present.
Disclosure of Invention
In view of the above, in order to solve the above problems, the present invention provides a method, an apparatus, a storage medium, and a terminal for predicting an intention to purchase, and the technical solution is as follows:
a purchasing intent prediction method, the method comprising:
acquiring target access data capable of representing the online matching behavior of a user to be predicted;
extracting a characteristic value of the target access data under a preset characteristic dimension, and inputting the characteristic value of the target access data into a prediction model so that the prediction model classifies the characteristic value of the target access data, wherein the prediction model is obtained by training a machine learning model by taking access data of historical user online matching behaviors as samples in advance;
and acquiring the purchase intention degree of the user to be predicted output by the prediction model.
Preferably, the process of training the machine learning model to obtain the prediction model by taking access data of the historical user online matching behavior as a sample in advance includes:
taking access data representing the online matching behavior of the historical user as a sample, and taking conversion data representing the purchase intention behavior of the historical user as a label of the sample;
extracting a characteristic value of the sample under the preset characteristic dimension, and inputting the characteristic value and the label of a part of samples used for the current training in the sample into a machine learning model so that the machine learning model adjusts model parameters by fitting the sample;
and calculating a loss function value of the machine learning model, repeatedly inputting the characteristic value and the label of part of samples used for the current training into the general machine learning model when the loss function value does not accord with a preset first training end condition, and taking the trained machine learning model as a prediction model when the loss function value accords with the first training end condition.
Preferably, the samples comprise training samples and test samples;
inputting the characteristic values and the labels of the part of samples used for the training into a machine learning model, wherein the method comprises the following steps:
inputting the characteristic values and the labels of the part of samples used for the current training in the training samples into a machine learning model;
before the trained machine learning model is used as the prediction model, the method further comprises the following steps:
inputting the characteristic value of the test sample into a trained machine learning model so that the trained machine learning model classifies the characteristic value of the test sample;
and acquiring the purchase intention output by the trained machine learning model and the label of the test sample, calculating the accuracy of the trained machine learning model, and taking the trained machine learning model as a prediction model when the accuracy meets a preset second training end condition.
Preferably, before extracting the feature value of the sample under the preset feature dimension, the method further includes:
and screening qualified samples meeting preset forward conversion conditions from the samples.
Preferably, the method further comprises:
and adding the user to be predicted, the purchase intention of which accords with the preset high purchase intention condition, into the recommendation form, and outputting.
A purchase intention prediction apparatus, the apparatus comprising: the system comprises a data acquisition module, a prediction module and an intention acquisition module, wherein the prediction module comprises a model training unit;
the model training unit is used for training a machine learning model to obtain a prediction model by taking access data of historical user online matching behaviors as samples in advance;
the data acquisition module is used for acquiring target access data capable of representing the online matching behavior of the user to be predicted;
the prediction module is used for extracting a characteristic value of the target access data under a preset characteristic dimension and inputting the characteristic value of the target access data into the prediction model so that the prediction model classifies the characteristic value of the target access data;
and the intention acquisition module is used for acquiring the purchase intention of the user to be predicted output by the prediction model.
Preferably, the model training unit is specifically configured to:
taking access data representing the online matching behavior of the historical user as a sample, and taking conversion data representing the purchase intention behavior of the historical user as a label of the sample;
extracting a characteristic value of the sample under the preset characteristic dimension, and inputting the characteristic value and the label of a part of samples used for the current training in the sample into a machine learning model so that the machine learning model adjusts model parameters by fitting the sample;
and calculating a loss function value of the machine learning model, repeatedly inputting the characteristic value and the label of part of samples used for the current training into the general machine learning model when the loss function value does not accord with a preset first training end condition, and taking the trained machine learning model as a prediction model when the loss function value accords with the first training end condition.
Preferably, the samples comprise training samples and test samples;
the model training unit is configured to input feature values and labels of a part of the samples used for the current training into a general machine learning model, and is specifically configured to:
inputting the characteristic values and the labels of the part of samples used for the current training in the training samples into a general machine learning model;
a model training unit further to:
inputting the characteristic value of the test sample into a trained machine learning model so that the trained machine learning model classifies the characteristic value of the test sample;
and acquiring the purchase intention output by the trained machine learning model and the label of the test sample, calculating the accuracy of the trained machine learning model, and taking the trained machine learning model as a prediction model when the accuracy meets a preset second training end condition.
A computer readable storage medium having stored thereon computer instructions which, when executed, perform the steps of any of the purchasing intention prediction methods.
A terminal comprising a memory and a processor, the memory having stored thereon computer instructions executable on the processor, the processor executing the computer instructions to perform any of the steps of the purchasing intent prediction method.
According to the purchase intention prediction method, the purchase intention prediction device, the storage medium and the terminal, the target access data of the user to be predicted during online selection can be obtained, and the purchase intention of the user to be predicted is predicted by processing the characteristic value of the target access data under the preset characteristic dimension through the prediction model. Therefore, the method for predicting the purchase intention can simply and effectively mine the high-purchase intention latent customers, so that subsequent marketing resources can be utilized to the maximum extent, and the automobile sales volume under the limited marketing cost is increased.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flowchart of a method for predicting purchase intention according to an embodiment of the present disclosure;
FIG. 2 is a flowchart of a portion of a method for predicting an intention to purchase according to an embodiment of the present disclosure;
FIG. 3 is a schematic diagram illustrating a prediction model using a SHAP interpreter according to an embodiment of the present application;
FIG. 4 is a schematic diagram of an influence result of a prediction model provided in an embodiment of the present application;
fig. 5 is a schematic structural diagram of a purchase intention prediction apparatus according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The existing scheme of internet automobile marketing is still put in advertisements, and attracts users to leave information such as contact information (called withholding for short) through information such as activities, preferential offers and the like. Then customer service personnel make calls one by one to follow up, confirm the interested vehicle type and the vehicle purchasing requirement and recommend the user to a nearby 4s store or a dealer. As the service customized on line for the automobile is still in the initial development stage, the marketing scheme specially aiming at the service is similar to the traditional automobile Internet marketing scheme, or the offline activities attract fans to register an account number and a landing platform, and then return visit is carried out on the reserved user by telephone, or the return visit is selectively carried out according to the online activity degree of the user.
Of course, the prediction of buying intent for supervised learning based online customization of automobiles has temporarily been without relevant research. In internet marketing in other fields, machine learning is used, but all methods use historical deal data, historical deal information of users, language judgment emotional tendency in user communication records, and the like. The traditional internet marketing is mainly recommended, which is to recommend commodities suitable for users from countless commodities, and the application can find users who really have purchasing intention from numerous users who select products.
The embodiment of the application provides a purchase intention prediction method, and a method flow chart of the method is shown in fig. 1, and the method comprises the following steps:
and S10, acquiring target access data capable of representing the online matching behavior of the user to be predicted.
In the embodiment of the application, most users of the matching products can consider whether to buy the automobile or not through behaviors of clicking and checking the webpage and the like on the matching webpage for many times. Therefore, the internet webpage can transmit some behaviors of clicking the webpage by the user back through the buried point information, and the behaviors are stored in the server or the storage medium. In sum, the automobile manufacturer or the sales organization can legally obtain the access behavior of the user in the matching process.
And S20, extracting the characteristic value of the target access data under the preset characteristic dimension, and inputting the characteristic value of the target access data into a prediction model so that the prediction model classifies the characteristic value of the target access data, wherein the prediction model is obtained by training a machine learning model by taking the access data of the on-line matching behavior of the historical user as a sample in advance.
In the embodiment of the application, at least one field can be selected from fields corresponding to the online matching behavior of the user as the preset characteristic dimension, so that the data content of the target access data under the preset characteristic dimension is extracted as the characteristic value of the target access data.
Most of the existing machine learning applications in internet marketing are focused on product recommendation of e-commerce platforms, and historical transaction data, historical transaction information of users, language judgment emotional tendency in user communication records and the like are used for judging products possibly needed by the users, or one or more products are used for matching the requirements of the users in a vertical marketing link. However, this has no practical value for the vertical marketing of the automobile internet for the online customization service of the automobile, because the related product brand, the automobile series are fixed, the user can select the configuration in the automobile series according to the needs, and the user can get on and off the order on the internet or go to the off-line dealer for further negotiation and understanding after the selection is completed, which needs to classify the users hierarchically.
Therefore, according to the method and the device, the purchase intention of the new user is predicted by utilizing the access data of the historical user to generate the prediction model, and therefore subsequent marketing can be accurately put.
It should be noted that the preset feature dimension used for extracting the target visit data is the same as the preset feature dimension used for training the prediction model in advance. And the construction of the preset characteristic dimension and the time window for acquiring the target access data can be finished in a self-defined way according to different scenes and service operation periods.
It should be noted that the general machine learning model may be any machine learning model capable of completing a classification task, and may be any one of a logistic regression model, a random forest model, an SVM model, and a LightGBM model.
In the specific implementation process, the process of training a general machine learning model by taking access data of the historical user online matching behavior as a sample in step S20 in advance to obtain a prediction model includes the following steps, and a flowchart of the method is shown in fig. 2:
s201, using the access data representing the on-line matching behavior of the historical user as a sample, and using the conversion data representing the purchase intention behavior of the historical user as a label of the sample.
In the embodiment of the application, some important characteristic dimensions which can summarize user behaviors and operation preferences are selected based on the matching behaviors of users, and then the access data of historical users are calculated by taking the users as units of characteristic values, so that the conversion data of the users, such as information of capital reservation, card establishment, sales and the like, is tracked in an internet marketing system. Therefore, in the embodiment of the present application, conversion data such as "fund retention", "card establishment", "sales", and the like may be used as a label of a sample in units of a user, and the label can represent the purchase intention degree of the user, that is, the degree of purchase intention possessed by the user. For example, the label of conversion data such as "reserve", "build card", "sell" and the like may be set to a positive number, and the magnitude of the value indicates the degree of purchase intention, whereas the label of a sample without conversion data may be set to 0, indicating that there is no purchase intention.
In practical applications, the specimen and the label of the specimen may be stored and used in the form of structured data.
S202, extracting characteristic values of the samples under a preset characteristic dimension, and inputting the characteristic values and labels of part of the samples used for training to the machine learning model so that the machine learning model adjusts model parameters by fitting the samples.
In the embodiment of the present application, a machine learning model is taken as an example of a logistic regression model, a part of samples used for each training is sequentially selected from samples, and a feature value and a label of the part of samples are input to the logistic regression model during each training. The logistic regression model after training can be used as a prediction model to judge the purchase intention of the user, and then marketers can accurately deliver the marketing intention to the user according to the intention of the user.
The logistic regression model is explained below:
establishing a logistic regression model shown in the following formula (1)
P(y＝1|x；θ)＝hθ(x)＝g(θTx) (1)
Wherein the content of the first and second substances,
wherein x isiFor feature input in the model, y is a label item, and y is 1 meaning that the subscriber line reserves or builds a card or sells, θiFitting of model to xiThe coefficient of (a).
In order to prevent overfitting of the model, the embodiment of the present application introduces a regularization term containing λ into the loss function J (θ) of the logistic regression model, where the loss function J (θ) is shown in the following formula (2):
wherein m is sample data size, J is the number of features in the logistic regression function, and when J (theta) is minimum, we obtain the required parameter thetaiAnd a hyperparameter λ.
Further, the embodiment of the application can further adopt a cross-check mode to perform on the parameter thetaiAnd a hyper-parameter lambda. When the hyper-parameter lambda is small, the model is easy to over-fit; when the hyper-parameter λ is larger, the degree of model overfitting may decrease while increasing the error rate of classification to some extent.
In addition, in some other embodiments, before extracting the feature value of the sample at the preset feature dimension, qualified samples meeting the preset forward conversion condition may be screened from the sample. Therefore, the qualified sample is used as the basis for obtaining the prediction model through subsequent training.
Specifically, the unsatisfactory sample can be removed from a practical perspective, for example, the corresponding date of the card establishment information of some users is before accessing the online matching platform, which indicates that the user is a high-intention user before accessing the online matching platform and should not be taken as a qualified sample of forward conversion.
And S203, calculating a loss function value of the machine learning model, and when the loss function value does not meet a preset first training end condition, returning to execute the step S202 of inputting the characteristic values and the labels of the part of samples used for the current training into the machine learning model, and taking the trained machine learning model as a prediction model until the loss function value meets the first training end condition.
In the embodiment of the application, if the function value of the loss function is greater than the preset loss function threshold value after the training is finished, the machine learning model continues to be trained, and the next training is started until the loss function value of the machine learning model after the training is finished is not greater than the loss function threshold value.
In other embodiments, the samples may be divided into training samples and test samples.
In this case, the step S202 of "inputting the feature values and labels of the partial samples used for the current training in the samples into the general machine learning model" includes:
and inputting the characteristic values and the labels of the part of samples used for the current training in the training samples into a general machine learning model.
Before the step S203 "using the trained machine learning model as the prediction model", the method further includes the following steps:
inputting the characteristic value of the test sample into the trained machine learning model so that the trained machine learning model classifies the characteristic value of the test sample;
and acquiring the purchase intention corresponding to the test sample output by the trained machine learning model and the label of the test sample, calculating the accuracy of the trained machine learning model, and taking the trained machine learning model as a prediction model when the accuracy meets a preset second training end condition.
In the embodiment of the application, the samples are divided into training samples and testing samples, the training samples are used for training the machine learning model to obtain the prediction model, and the testing samples are used for verifying the prediction accuracy of the prediction model. Specifically, if the purchase intention degree predicted by the prediction model for a test sample can represent the label of the test sample, the prediction model predicts the test sample accurately, otherwise, the prediction model is not accurate. The prediction model can be represented by the proportion of the number of samples predicted accurately by the prediction model in the test samples to the total number of samples in the test samples.
For the convenience of understanding of the present application, the following describes the processes of off-line training logistic regression model and on-line prediction in the present application by taking an application scenario as an example:
1. and training the logistic regression model off line.
Sample and label input: 1800 pieces of user behavior log records are selected and matched online through an online buried point pull internet, 20 fields are extracted according to a service scene and an invention target, offline funding/card building/sales information is obtained, and matching is carried out by taking a user as an id, so that 20646 pieces of data are obtained.
Sample treatment: according to the model, 12 fields are selected as characteristic inputs of the model from 20 fields, and 10713 data worth of prediction are selected from 20646 pieces of data.
Decomposing a training sample and a test sample: 10713 pieces of data are divided into two parts of training samples and test samples, wherein the test samples account for 20 percent of the total number, and meanwhile, the proportion of positive samples reserved/built/sold in the training samples and the test samples is ensured to be consistent by taking off-line reserved/built/sold information as a layering standard.
And (3) feature standardization: the input 12 features are standardized, and the influence on the model caused by the size difference among different dimensional features is eliminated, so that the hyper-parameter adjustment of the model is performed.
Model fitting and parameter adjustment: introducing a hyperparameter C after generating a logistic regression model, wherein
Fitting the training set by using cross-folding inspection, searching the optimal hyper-parameter for C from 0.1 to 100 by grid search, and finally obtaining theta0Is-0.54, theta1To theta12Respectively 1.03e-03, -6.02e-04,1.01e-01,7.26e-05, -1.23e-01, -8.58e-04, -8.26e-04,3.99e-03, -2.45e-02-3.78e-05,3.63e-01,5.22e-03
And (3) testing results of the model: and (3) performing prediction on a test set (2143 pieces), wherein the prediction is 284 pieces of results of card building, and the result is 1501 pieces of correct results with the correct rate of 70%. The model is saved for subsequent invocation and prediction of new data.
Model interpretation of SHAP: the generated logistic regression model was interpreted on the test samples using the SHAP interpreter, and the results are shown in FIG. 3. It can be seen that the ft3, ft1 features have a significant effect on the positive prediction of the model, while ft5, ft6 have a significant effect on the negative prediction of the model. This also coincides with the coefficients to which the model is fitted. With ft3 having the strongest influence on the model.
Ft3 was used as a study object to examine the effect on the model, as shown in FIG. 4. It can be seen that the positive predictive contribution capability of ft3 to the model is linearly positively correlated with its own value, which is related to the model form of the logistic regression itself.
2. And (4) online prediction.
Target access data entry and processing: similar to the process of training the logistic regression model, 20551 pieces of data are obtained.
Prediction of purchase intention degree: calling out the prediction model of the off-line training, and predicting on the target access data to obtain corresponding purchase intention, wherein 1272 prediction results of high intention are obtained.
The method and the system solve the problem of how to accurately put limited marketing resources to users with high purchase intention in the Internet marketing system with the online automobile customization service. The method for predicting the purchasing intention degree has the advantages that the purchasing intention degree can be layered for a user who does not have deep knowledge before. In addition, the method has another advantage that more user data can be continuously acquired according to the accumulation of the data, and then the iterative model is retrained according to the data, so that the accuracy of the prediction model can be continuously improved along with the increase of samples. According to the method and the device, the purchase intention of the user is layered to automatically generate results through calculation of the logistic regression model, manual screening is not needed, and a large amount of human resources are saved.
And S30, acquiring the purchase intention degree of the user to be predicted output by the prediction model.
In the embodiment of the application, the prediction probability output by the prediction model is the purchase intention degree of the user to be predicted, and the purchase intention degree can represent the degree of purchase intention of the user to be predicted. If the predetermined high-intention-to-purchase condition is satisfied, for example, the intention degree of purchase is higher than a predetermined intention degree threshold, the user to be predicted may be considered as a high-intention user. Of course, the intention range can be set for different intention types, namely withholding, card building and selling, so that accurate putting is realized.
In other embodiments, the user to be predicted whose purchase intention meets the preset high purchase intention condition may be added to the recommendation form and output.
Specifically, whether the user to be predicted is located in a recommendation form obtained by previous prediction or not can be judged firstly; if yes, indicating that the user to be predicted has been recommended, and directly ending; if not, in order to reduce the workload of the marketing department, in the embodiment of the application, the users with high purchase intention degree are added into the recommendation form and pushed to the marketing department, the recommendation form at least comprises the identity information and the purchase intention degree of the users to be predicted, in addition, the ordering can be further carried out according to the sequence of the purchase intention degree from high to low, and the marketing department carries out accurate pushing according to the requirements and the service scenes.
According to the purchasing intention prediction method provided by the embodiment of the application, the target access data of the user to be predicted in online selection can be obtained, and the purchasing intention of the user to be predicted is predicted by processing the characteristic value of the target access data under the preset characteristic dimension through the prediction model. Based on the method and the system, the high-intention potential customers can be simply and effectively mined, so that subsequent marketing resources can be utilized to the maximum extent, and the automobile sales volume under the limited marketing cost is increased.
Based on the method for predicting the purchasing intention provided by the above embodiment, an embodiment of the present application further provides a device for executing the method for predicting the purchasing intention, and a schematic structural diagram of the device is shown in fig. 5, and the device includes: the system comprises a data acquisition module 10, a prediction module 20 and an intention acquisition module 30, wherein the prediction module comprises a model training unit 201;
the model training unit 201 is used for training the machine learning model by taking access data of the historical user online matching behavior as a sample in advance to obtain a prediction model;
the data acquisition module 10 is used for acquiring target access data capable of representing the online matching behavior of the user to be predicted;
the prediction module 20 is configured to extract a feature value of the target access data under a preset feature dimension, and input the feature value of the target access data into the prediction model, so that the prediction model classifies the feature value of the target access data;
and the intention acquisition module 30 is used for acquiring the purchase intention of the user to be predicted, which is output by the prediction model.
Optionally, the model training unit 201 is specifically configured to:
taking access data representing the online matching behavior of the historical user as a sample, and taking conversion data representing the purchase intention behavior of the historical user as a label of the sample;
extracting a characteristic value of the sample under a preset characteristic dimension, and inputting the characteristic value and the label of a part of samples used for the training in the sample into a machine learning model so that the machine learning model adjusts model parameters by fitting the sample;
and calculating a loss function value of the machine learning model, repeatedly inputting the characteristic value and the label of part of samples used for the current training in the samples into the machine learning model when the loss function value does not accord with a preset first training end condition, and taking the trained machine learning model as a prediction model until the loss function value accords with the first training end condition.
Optionally, the samples include training samples and test samples;
the model training unit 201 is configured to input feature values and labels of a part of samples used for the current training in the samples into a general machine learning model, and is specifically configured to:
inputting the characteristic values and labels of part of samples used for the current training in the training samples into a general machine learning model;
the model training unit 201 is further configured to:
inputting the characteristic value of the test sample into the trained machine learning model so that the trained machine learning model classifies the characteristic value of the test sample; and calculating the accuracy of the trained machine learning model based on the purchase intention corresponding to the test sample output by the trained machine learning model and the label of the test sample, and executing the trained machine learning model as a prediction model when the accuracy accords with a preset second training end condition.
Optionally, the model training unit 201 is further configured to:
and screening qualified samples meeting the preset forward conversion conditions from the samples.
Optionally, the intention acquisition module 30 is further configured to:
and adding the user to be predicted, the purchase intention of which accords with the preset high purchase intention condition, into the recommendation form, and outputting.
The purchasing intention degree prediction device provided by the embodiment of the application can acquire the target access data of the user to be predicted in online selection, and predicts the purchasing intention degree of the user to be predicted by processing the characteristic value of the target access data under the preset characteristic dimension by using the prediction model. Based on the method and the system, the high-intention potential customers can be simply and effectively mined, so that subsequent marketing resources can be utilized to the maximum extent, and the automobile sales volume under the limited marketing cost is increased.
Embodiments of the present invention also provide a computer-readable storage medium having stored thereon computer instructions, which, when executed, perform the steps of the purchasing intention prediction method. The computer-readable storage medium may include, for example, a non-volatile (non-volatile) or non-transitory (non-transitory) memory, and may also include an optical disc, a mechanical hard disk, a solid state hard disk, and the like.
The embodiment of the invention also provides a terminal, which comprises a memory and a processor, wherein the memory is stored with computer instructions capable of running on the processor, and the processor executes the steps of the purchasing intention prediction method when running the computer instructions. The terminal includes, but is not limited to, a mobile phone, a computer, a tablet computer and other terminal devices.
The processor may be a Central Processing Unit (CPU), an application-specific integrated circuit (ASIC), a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device.
For example, the computer instructions stored in the memory may be executed by the processor to implement all or part of the functions of the purchase intention degree prediction method in the above embodiments. In addition, when all or part of the functions in the above embodiments are implemented by computer instructions, the computer instructions may also be stored in a storage medium such as a server, another computer, a magnetic disk, an optical disk, a flash disk, or a removable hard disk, and stored in a memory of the local terminal by downloading or copying, or by performing version update on a system of the local terminal, and when the computer instructions in the memory are executed by a processor, all or part of the functions in the above embodiments may be implemented.
The method, the device, the storage medium and the terminal for predicting the purchasing intention provided by the invention are described in detail, a specific example is applied in the description to explain the principle and the implementation mode of the invention, and the description of the embodiment is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
It should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
It is further noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include or include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method for predicting an intention to purchase, the method comprising:
acquiring target access data capable of representing the online matching behavior of a user to be predicted;
extracting a characteristic value of the target access data under a preset characteristic dimension, and inputting the characteristic value of the target access data into a prediction model so that the prediction model classifies the characteristic value of the target access data, wherein the prediction model is obtained by training a machine learning model by taking access data of historical user online matching behaviors as samples in advance;
and acquiring the purchase intention degree of the user to be predicted output by the prediction model.
2. The method of claim 1, wherein the process of training a machine learning model to obtain the prediction model by taking access data of historical user online matching behaviors as samples in advance comprises:
taking access data representing the online matching behavior of the historical user as a sample, and taking conversion data representing the purchase intention behavior of the historical user as a label of the sample;
extracting a characteristic value of the sample under the preset characteristic dimension, and inputting the characteristic value and the label of a part of samples used for the current training in the sample into a machine learning model so that the machine learning model adjusts model parameters by fitting the sample;
and calculating a loss function value of the machine learning model, repeatedly inputting the characteristic value and the label of part of samples used for the current training into the general machine learning model when the loss function value does not accord with a preset first training end condition, and taking the trained machine learning model as a prediction model when the loss function value accords with the first training end condition.
3. The method of claim 2, wherein the samples comprise training samples and test samples;
inputting the characteristic values and the labels of the part of samples used for the training into a general machine learning model, wherein the method comprises the following steps:
inputting the characteristic values and the labels of the part of samples used for the current training in the training samples into a machine learning model;
before the trained machine learning model is used as the prediction model, the method further comprises the following steps:
inputting the characteristic value of the test sample into a trained machine learning model so that the trained machine learning model classifies the characteristic value of the test sample;
and acquiring the purchase intention output by the trained machine learning model and the label of the test sample, calculating the accuracy of the trained machine learning model, and taking the trained machine learning model as a prediction model when the accuracy meets a preset second training end condition.
4. The method of claim 2, wherein before extracting the feature values of the sample at the preset feature dimension, the method further comprises:
and screening qualified samples meeting preset forward conversion conditions from the samples.
5. The method of claim 1, further comprising:
and adding the user to be predicted, the purchase intention of which accords with the preset high purchase intention condition, into the recommendation form, and outputting.
6. A purchase intention prediction apparatus, characterized in that the apparatus comprises: the system comprises a data acquisition module, a prediction module and an intention acquisition module, wherein the prediction module comprises a model training unit;
the model training unit is used for training a machine learning model to obtain a prediction model by taking access data of historical user online matching behaviors as samples in advance;
the data acquisition module is used for acquiring target access data capable of representing the online matching behavior of the user to be predicted;
the prediction module is used for extracting a characteristic value of the target access data under a preset characteristic dimension and inputting the characteristic value of the target access data into the prediction model so that the prediction model classifies the characteristic value of the target access data;
and the intention acquisition module is used for acquiring the purchase intention of the user to be predicted output by the prediction model.
7. The apparatus according to claim 6, wherein the model training unit is specifically configured to:
taking access data representing the online matching behavior of the historical user as a sample, and taking conversion data representing the purchase intention behavior of the historical user as a label of the sample;
extracting a characteristic value of the sample under the preset characteristic dimension, and inputting the characteristic value and the label of a part of samples used for the current training in the sample into a machine learning model so that the machine learning model adjusts model parameters by fitting the sample;
and calculating a loss function value of the machine learning model, repeatedly inputting the characteristic value and the label of part of samples used for the current training into the general machine learning model when the loss function value does not accord with a preset first training end condition, and taking the trained machine learning model as a prediction model when the loss function value accords with the first training end condition.
8. The apparatus of claim 7, wherein the samples comprise training samples and test samples;
the model training unit is configured to input feature values and labels of a part of the samples used for the current training into a general machine learning model, and is specifically configured to:
inputting the characteristic values and the labels of the part of samples used for the current training in the training samples into a general machine learning model;
a model training unit further to:
inputting the characteristic value of the test sample into a trained machine learning model so that the trained machine learning model classifies the characteristic value of the test sample;
and acquiring the purchase intention output by the trained machine learning model and the label of the test sample, calculating the accuracy of the trained machine learning model, and taking the trained machine learning model as a prediction model when the accuracy meets a preset second training end condition.
9. A computer readable storage medium having stored thereon computer instructions, wherein the computer instructions when executed perform the steps of the purchasing intention prediction method according to any one of claims 1 to 5.
10. A terminal comprising a memory and a processor, the memory having stored thereon computer instructions executable on the processor, wherein the processor executes the computer instructions to perform the steps of the purchasing intention prediction method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010512538.6A CN111681051B (en) | 2020-06-08 | 2020-06-08 | Purchase intention prediction method and device, storage medium and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010512538.6A CN111681051B (en) | 2020-06-08 | 2020-06-08 | Purchase intention prediction method and device, storage medium and terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111681051A true CN111681051A (en) | 2020-09-18 |
| CN111681051B CN111681051B (en) | 2023-09-26 |
Family
ID=72454942
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010512538.6A Active CN111681051B (en) | 2020-06-08 | 2020-06-08 | Purchase intention prediction method and device, storage medium and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111681051B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112258242A (en) * | 2020-11-02 | 2021-01-22 | 上海汽车集团股份有限公司 | Form configuration item data pushing method and device |
| CN112884449A (en) * | 2021-03-12 | 2021-06-01 | 北京乐学帮网络技术有限公司 | User guiding method, device, computer equipment and storage medium |
| CN113190599A (en) * | 2021-06-30 | 2021-07-30 | 平安科技（深圳）有限公司 | Processing method, device and equipment for application user behavior data and storage medium |
| CN113763032A (en) * | 2021-08-03 | 2021-12-07 | 北京光速斑马数据科技有限公司 | Commodity purchase intention identification method and device |
| CN113807921A (en) * | 2021-09-17 | 2021-12-17 | 深圳市数聚湾区大数据研究院 | Data commodity recommendation method and device, electronic equipment and computer readable storage medium |
| CN115309737A (en) * | 2022-10-11 | 2022-11-08 | 深圳市明源云客电子商务有限公司 | Visitor intention analysis method and system, terminal device and readable storage medium |
| CN117217852A (en) * | 2023-08-03 | 2023-12-12 | 广州兴趣岛信息科技有限公司 | Behavior recognition-based purchase willingness prediction method and device |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130268468A1 (en) * | 2012-04-09 | 2013-10-10 | 24/7 Customer, Inc. | Method and apparatus for intent modeling and prediction |
| US20150134401A1 (en) * | 2013-11-09 | 2015-05-14 | Carsten Heuer | In-memory end-to-end process of predictive analytics |
| CN105389639A (en) * | 2015-12-15 | 2016-03-09 | 上海汽车集团股份有限公司 | Logistics transportation route planning method, device and system based on machine learning |
| CN105488697A (en) * | 2015-12-09 | 2016-04-13 | 焦点科技股份有限公司 | Potential customer mining method based on customer behavior characteristics |
| JP2016076109A (en) * | 2014-10-07 | 2016-05-12 | 国立大学法人九州工業大学 | Device and method for predicting customers's purchase decision |
| US20160188726A1 (en) * | 2014-12-31 | 2016-06-30 | TCL Research America Inc. | Scalable user intent mining using a multimodal restricted boltzmann machine |
| CN107993088A (en) * | 2017-11-20 | 2018-05-04 | 北京三快在线科技有限公司 | A kind of Buying Cycle Forecasting Methodology and device, electronic equipment |
| CN109741112A (en) * | 2019-01-10 | 2019-05-10 | 博拉网络股份有限公司 | A kind of user's purchase intention prediction technique based on mobile big data |
| TWM589312U (en) * | 2019-10-08 | 2020-01-11 | 富邦人壽保險股份有限公司 | Product purchase evaluation system based on user web browsing history |
-
2020
- 2020-06-08 CN CN202010512538.6A patent/CN111681051B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130268468A1 (en) * | 2012-04-09 | 2013-10-10 | 24/7 Customer, Inc. | Method and apparatus for intent modeling and prediction |
| US20150134401A1 (en) * | 2013-11-09 | 2015-05-14 | Carsten Heuer | In-memory end-to-end process of predictive analytics |
| JP2016076109A (en) * | 2014-10-07 | 2016-05-12 | 国立大学法人九州工業大学 | Device and method for predicting customers's purchase decision |
| US20160188726A1 (en) * | 2014-12-31 | 2016-06-30 | TCL Research America Inc. | Scalable user intent mining using a multimodal restricted boltzmann machine |
| CN105488697A (en) * | 2015-12-09 | 2016-04-13 | 焦点科技股份有限公司 | Potential customer mining method based on customer behavior characteristics |
| CN105389639A (en) * | 2015-12-15 | 2016-03-09 | 上海汽车集团股份有限公司 | Logistics transportation route planning method, device and system based on machine learning |
| CN107993088A (en) * | 2017-11-20 | 2018-05-04 | 北京三快在线科技有限公司 | A kind of Buying Cycle Forecasting Methodology and device, electronic equipment |
| CN109741112A (en) * | 2019-01-10 | 2019-05-10 | 博拉网络股份有限公司 | A kind of user's purchase intention prediction technique based on mobile big data |
| TWM589312U (en) * | 2019-10-08 | 2020-01-11 | 富邦人壽保險股份有限公司 | Product purchase evaluation system based on user web browsing history |
Non-Patent Citations (1)
| Title |
|---|
| 张李义;李一然;文璇;: "新消费者重复购买意向预测研究", no. 11 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112258242A (en) * | 2020-11-02 | 2021-01-22 | 上海汽车集团股份有限公司 | Form configuration item data pushing method and device |
| CN112884449A (en) * | 2021-03-12 | 2021-06-01 | 北京乐学帮网络技术有限公司 | User guiding method, device, computer equipment and storage medium |
| CN112884449B (en) * | 2021-03-12 | 2024-05-14 | 北京乐学帮网络技术有限公司 | User guiding method, device, computer equipment and storage medium |
| CN113190599A (en) * | 2021-06-30 | 2021-07-30 | 平安科技（深圳）有限公司 | Processing method, device and equipment for application user behavior data and storage medium |
| CN113763032A (en) * | 2021-08-03 | 2021-12-07 | 北京光速斑马数据科技有限公司 | Commodity purchase intention identification method and device |
| CN113763032B (en) * | 2021-08-03 | 2023-08-04 | 北京光速斑马数据科技有限公司 | Commodity purchase intention recognition method and device |
| CN113807921A (en) * | 2021-09-17 | 2021-12-17 | 深圳市数聚湾区大数据研究院 | Data commodity recommendation method and device, electronic equipment and computer readable storage medium |
| CN113807921B (en) * | 2021-09-17 | 2023-11-24 | 深圳市数聚湾区大数据研究院 | Data commodity recommendation method and device, electronic equipment and computer readable storage medium |
| CN115309737A (en) * | 2022-10-11 | 2022-11-08 | 深圳市明源云客电子商务有限公司 | Visitor intention analysis method and system, terminal device and readable storage medium |
| CN117217852A (en) * | 2023-08-03 | 2023-12-12 | 广州兴趣岛信息科技有限公司 | Behavior recognition-based purchase willingness prediction method and device |
| CN117217852B (en) * | 2023-08-03 | 2024-02-27 | 广州兴趣岛信息科技有限公司 | Behavior recognition-based purchase willingness prediction method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111681051B (en) | 2023-09-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111681051A (en) | Purchasing intention degree prediction method, device, storage medium and terminal | |
| CN110008973B (en) | Model training method, method and device for determining target user based on model | |
| US20150213111A1 (en) | Obtaining social relationship type of network subjects | |
| CN111061979B (en) | User tag pushing method and device, electronic equipment and medium | |
| US20140067472A1 (en) | System and Method For Segmenting A Customer Base | |
| CN111666275B (en) | Data processing method and device, electronic equipment and storage medium | |
| CN112001754A (en) | User portrait generation method, device, equipment and computer readable medium | |
| CN112380449B (en) | Information recommendation method, model training method and related device | |
| CN110827086A (en) | Product marketing prediction method and device, computer equipment and readable storage medium | |
| CN112598472A (en) | Product recommendation method, device, system, medium and program product | |
| CN110362702B (en) | Picture management method and equipment | |
| CN111680213B (en) | Information recommendation method, data processing method and device | |
| CN111178972A (en) | Message pushing method and device, storage medium and equipment | |
| US20230351418A1 (en) | Business opportunity information recommendation server and method therefor | |
| CN110738529A (en) | User diffusion method and device, readable storage medium and electronic equipment | |
| CN112785391B (en) | Recommendation processing method and device, intelligent equipment and storage medium | |
| CN115168700A (en) | Information flow recommendation method, system and medium based on pre-training algorithm | |
| CN113378071A (en) | Advertisement recommendation method and device, electronic equipment and storage medium | |
| CN112070530A (en) | Online evaluation method and related device of advertisement prediction model | |
| CN113554448A (en) | User loss prediction method and device and electronic equipment | |
| CN117217852B (en) | Behavior recognition-based purchase willingness prediction method and device | |
| CN115222486B (en) | Article recommendation model training method, article recommendation method, device and storage medium | |
| CN115660733A (en) | Sales prediction system and method based on artificial intelligence | |
| CN117726403A (en) | Product recommendation method and device, storage medium and electronic equipment | |
| CN116049560A (en) | Information processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |