Method, device and electronic equipment for predicting object behavior

Abstract

The application discloses a kind of object behavior prediction method, device and electronic equipment, it is related to big data technical field.The method comprises: obtaining the object information of target object, wherein the object information at least includes the behavior data of target object and transaction data;Object information is cleaned and handled, at least one target object feature of target object in current stage is obtained, wherein current stage is any one stage in at least one stage corresponding to target object in customer life cycle, customer life cycle characterizes the relationship between the behavior data of target object and target business;At least one target object feature is used to predict the object behavior of target object in current stage, and prediction result is obtained.The present application solves the technical problem of low accuracy of customer behavior judgment caused by incomplete customer analysis in the prior art.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of big data, in particular to a method and device for predicting object behavior and an electronic device. BACKGROUND

[0002] With the rapid development of modern science and technology, the updating speed of knowledge is accelerating, and people's lifestyle and consumption patterns are gradually changing. Under the new situation, enterprises must update their marketing strategies to survive and develop in the fierce market competition.

[0003] Currently, enterprises are more likely to consider products when formulating marketing strategies, that is, different management measures are implemented according to the characteristics of each stage of the product in the market, and there are problems of insufficient consideration of customer needs and insufficient comprehensive analysis of customers.

[0004] In view of the above problems, no effective solution has been proposed so far. SUMMARY

[0005] The embodiments of the present application provide a method and device for predicting object behavior and an electronic device to at least solve the technical problem of low accuracy of customer behavior judgment caused by insufficient comprehensive analysis of customers in the prior art.

[0006] According to an aspect of an embodiment of the present application, a method for predicting object behavior is provided, comprising: obtaining object information of a target object, wherein the object information at least includes behavior data and transaction data of the target object; performing cleaning processing on the object information to obtain at least one target object feature of the target object in a current stage, wherein the current stage is any one of at least one stage corresponding to a customer life cycle of the target object, and the customer life cycle represents the relationship between the behavior data of the target object and a target business; and predicting the object behavior of the target object in the current stage based on the at least one target object feature to obtain a prediction result.

[0007] Further, the method for predicting object behavior further comprises: performing cleaning processing on the object information of the target object to obtain cleaned object information; determining target object information corresponding to the current stage from the cleaned object information; and performing feature extraction on the target object information to obtain at least one target object feature.

[0008] Further, the method for predicting object behavior further comprises: determining a target feature label corresponding to the target object in the current stage according to the at least one target object feature, wherein the target feature label at least includes a static label and a dynamic label; and predicting the object behavior of the target object in the current stage based on the target feature label to obtain a prediction result.

[0009] Further, the method for predicting the object behavior further comprises: obtaining object features corresponding to the plurality of objects in each stage, wherein the object features of the plurality of objects correspond to the business of the corresponding stage; analyzing the association degree between the object features and at least one stage to determine the feature label corresponding to each stage.

[0010] Further, the method for predicting the object behavior further comprises: analyzing the association degree between the static features in the object features and at least one stage to determine the static label corresponding to each stage, wherein the static features represent inherent attribute information of the plurality of objects; analyzing the association degree between the dynamic features in the object features and at least one stage to determine the dynamic label corresponding to each stage, wherein the dynamic features represent features that change with the object behavior of the plurality of objects.

[0011] Further, the method for predicting the object behavior further comprises: obtaining business information of at least one business corresponding to the current stage; determining the association degree between the business information and the target feature label; and determining target business information corresponding to the target object from the at least one business corresponding to the current stage according to the association degree, wherein the association degree between the target business information and the target feature label is greater than a preset association degree, and the target business information represents the intended business of the target object.

[0012] According to another aspect of the embodiments of the present application, a device for predicting object behavior is also provided, comprising: an obtaining module configured to obtain object information of a target object, wherein the object information at least includes behavior data and transaction data of the target object; a processing module configured to perform cleaning processing on the object information to obtain at least one target object feature of the target object in a current stage, wherein the current stage is any one of at least one stage corresponding to a customer life cycle of the target object, and the customer life cycle represents the relationship between the behavior data of the target object and a target business; and a prediction module configured to predict the object behavior of the target object in the current stage based on the at least one target object feature to obtain a prediction result.

[0013] According to another aspect of the embodiments of the present application, a computer readable storage medium is also provided, and the computer readable storage medium stores a computer program, wherein the computer program is configured to execute the above-mentioned method for predicting the object behavior when running.

[0014] According to another aspect of the embodiments of the present application, an electronic device is also provided, and the electronic device comprises one or more processors; a memory configured to store one or more programs, which make the one or more processors implement a program for running when the one or more programs are executed by the one or more processors, wherein the program is configured to execute the above-mentioned method for predicting the object behavior when running.

[0015] According to another aspect of the embodiments of the present application, there is also provided a computer program product comprising computer programs / instructions which, when executed by a processor, implement the above-mentioned object behavior prediction method.

[0016] In the embodiments of the present application, the object behavior is predicted by object features, and the object information of a target object is first acquired, and then the object information is cleaned to obtain at least one target object feature of the target object in a current stage, and the object behavior of the target object in the current stage is predicted based on the at least one target object feature to obtain a prediction result. The object information at least includes behavior data and transaction data of the target object, and the current stage is any one of at least one stage of the target object in a customer life cycle, and the customer life cycle represents the relationship between the behavior data of the target object and a target business.

[0017] In the above process, a large amount of data of the target object is collected to comprehensively reflect the situation of the customer, and the data cleaning is performed to improve the analysis efficiency. In addition, the extracted object features are applied to each stage of the customer life cycle to predict the customer behavior, and the comprehensive analysis of the customer is realized. Based on the prediction result, the customer relationship can be further mined, maintained and developed to provide accurate service for the customer, thereby prolonging the customer life cycle.

[0018] Therefore, the technical solution of the present application achieves the purpose of deeply and comprehensively understanding the customer, thereby realizing the technical effect of improving the accuracy of the customer behavior judgment, and further solving the technical problem of low accuracy of the customer behavior judgment caused by the incomplete customer analysis in the prior art. BRIEF DESCRIPTION OF DRAWINGS

[0019] The accompanying drawings, which are included to provide a further understanding of the present application and are incorporated in and constitute a part of this application, illustrate embodiments of the present application and serve to explain the present application. In the drawings:

[0020] Figure 1 is a flowchart of an optional object behavior prediction method according to an embodiment of the present application;

[0021] Figure 2 is a schematic diagram of an optional object behavior prediction model according to an embodiment of the present application;

[0022] Figure 3 is a schematic diagram of an optional object behavior prediction device according to an embodiment of the present application;

[0023] Figure 4 is a schematic diagram of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to make the personnel in the technical field better understand the present application scheme, the technical scheme in the embodiment of the present application will be clearly and completely described below in combination with the drawings in the embodiment of the present application. Obviously, the described embodiment is only a part of the embodiment of the present application, not all. Based on the embodiment in the present application, all other embodiments obtained by the person skilled in the art without creative labor should belong to the scope of protection of the present application.

[0025] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0026] It should be noted that the relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, analyzed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties. For example, an interface is provided between the system and the relevant user or institution. Before obtaining the relevant information, the interface needs to send a request to the aforementioned user or institution, and after receiving the consent information feedback from the aforementioned user or institution, the relevant information is obtained.

[0027] Embodiment 1

[0028] According to the embodiment of the present application, a method embodiment of a method for predicting object behavior is provided. It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in a different order.

[0029] Figure 1 is a flowchart of an optional method for predicting object behavior according to an embodiment of the present application, as shown in Figure 1 The method comprises the following steps:

[0030] Step S101, obtaining object information of a target object.

[0031] In the above steps, the target object can be an object to be predicted, and the object information includes at least behavior data and transaction data of the object. To comprehensively reflect the object, the object information can be obtained from the local memory or the Internet. In addition, the object information can also include basic data of the object, such as age, gender, risk and return data, and can also include Internet data of the object, such as third-party payment data, e-commerce platform shopping data, life payment data, social platform data such as WeChat and Weibo, and Internet credit data.

[0032] It should be noted that in the above process, by obtaining the object information of the target object, data reflecting the object can be obtained, which provides a data basis for subsequent analysis of the target object.

[0033] In step S102, the object information is cleaned to obtain at least one target object feature of the target object in the current stage.

[0034] In the above steps, after obtaining a large amount of data of the target object, the data needs to be sorted and summarized, that is, the object information is cleaned, so that valuable information can be obtained from a large amount of data, that is, at least one target object feature of the target object in the current stage. Optionally, taking the relationship between the business of the financial institution and the customer as an example, the customer life cycle can be divided into five stages: accurate promotion, accurate customer acquisition, accurate tracking, accurate service, and full-process risk prevention and control. The foregoing current stage can be any one of the five stages.

[0035] It should be noted that since the amount of information obtained by the target object is huge, the behavior of the target object is very scattered, therefore, by cleaning the object information, the amount of invalid information can be reduced, and the efficiency of analyzing the target object is improved.

[0036] In step S103, the object behavior of the target object in the current stage is predicted based on at least one target object feature, and a prediction result is obtained.

[0037] In the above steps, different target object features can be extracted based on different business needs and different customer life cycle stages, and the object behavior of the target object in the current stage can be predicted to obtain a prediction result.

[0038] Optionally, according to the mining demand of potential customers of the credit card business of the financial institution, at least one target object feature is extracted in the precise customer acquisition stage of the customer life cycle, the object behavior of the target object at the current stage can be predicted, so as to develop potential credit card customers. For example, when the target object shops on the e-commerce platform, the object often adopts the payment method of installment payment, and the object feature of the target object can be installment payment. Through behavior prediction of the object with the feature of installment payment, the possibility of the target object to handle the credit card can be obtained, so as to improve the ability of the financial institution to market the credit card and expand the scale of the credit card customer group. Similarly, the whole process risk control can also be carried out in the customer life cycle, and abnormal information can be obtained in time to reduce the loss of the financial institution and the customer.

[0039] It should be noted that by predicting the behavior of each stage of the customer life cycle, the customer relationship can be mined, maintained and developed, thereby prolonging the customer life cycle.

[0040] Based on the scheme defined in the above steps S101 to S103, it can be known that in the embodiment of the present application, the object behavior is predicted by the object feature, the object information of the target object is first obtained, then the object information is cleaned to obtain at least one target object feature of the target object at the current stage, and then the object behavior of the target object at the current stage is predicted based on the at least one target object feature to obtain a prediction result. The object information at least includes the behavior data and transaction data of the target object, the current stage is any one of at least one stage corresponding to the customer life cycle of the target object, and the customer life cycle represents the relationship between the behavior data of the target object and the target business.

[0041] It is easy to note that in the above process, by collecting a large amount of data of the target object, the situation of the customer can be fully reflected, and by data cleaning, the analysis efficiency can be improved. In addition, according to different business demands, the extracted object features are applied to each stage of the customer life cycle to predict the customer behavior, realizing the comprehensive analysis of the customer; based on the prediction result, the customer relationship can be further mined, maintained and developed, so as to provide accurate service for the customer, thereby prolonging the customer life cycle.

[0042] As can be seen, by the technical scheme of the present application, the purpose of deep and comprehensive understanding of the customer is achieved, thereby realizing the technical effect of improving the accuracy of the customer behavior judgment, and further solving the technical problem of low accuracy of the customer behavior judgment caused by the incomplete customer analysis in the prior art.

[0043] In an alternative embodiment, in the process of cleaning the object information to obtain at least one target object feature of the target object at the current stage, the object information of the target object is first cleaned to obtain cleaned object information, then the target object information corresponding to the current stage is determined from the cleaned object information, and then the target object information is feature extracted to obtain at least one target object feature.

[0044] Optionally, Figure 2 is a schematic diagram of an alternative object behavior prediction model according to an embodiment of the present application, as Figure 2 shown, a sandglass model is provided, which consists of three parts of data acquisition, customer portrait and behavior prediction. The upper part of the model is the source of the acquired data, i.e. the object information of the target object can be acquired through terminal data, client data, online bank data, telephone bank data, Internet data, etc.; the middle part is the process of forming the customer portrait, i.e. the customer portrait is formed through data cleaning, label definition and feature engineering; the lower part is behavior prediction, which applies the data analyzed in the foregoing to the customer life cycle, i.e. predicts the behavior of each stage of the customer life cycle, including promotion, customer acquisition, tracking, service and risk prevention and control throughout the whole process.

[0045] Specifically, in the present embodiment, the target object can be a customer of a financial institution, and after acquiring the customer information, the customer information is cleaned to obtain cleaned customer information. Wherein, the acquisition of customer information can be the acquisition of information within a unit time, and the source of the acquired data can also be flexibly updated according to the demand. In addition, the cleaning of the customer information can be filtering the coding errors, format errors, incomplete records, data values out of the specified range and data that are not helpful to the analysis results in the customer information, and at the same time classifying the filtered information, the classified information mainly includes five dimensions of demographic information, customer value, relationship, risk assessment and interest. The specific interpretation of the dimensions is shown in Table 1.

[0046] Table 1 Interpretation of Five Dimensions of Customer Basic Information

[0047]

[0048]

[0049] Further, after processing the customer information, information corresponding to the current stage is determined from the customer information, for example, in the precise customer acquisition stage of the customer life cycle, the information corresponding to the current stage can be information in four dimensions of demographic information, customer value, risk assessment, and interest and hobby; then, feature extraction is performed on the customer information in the four dimensions to obtain at least one target object feature, for example, when a customer uses a mobile phone to shop on an e-commerce platform, the customer often uses a payment method of installment payment, at this time, the financial institution has a demand for mining potential customers of credit card business, and can select channel product cross-class information in the interest and hobby dimension corresponding to the feature of the customer using a mobile phone to consume on a third-party platform in installment.

[0050] It should be noted that by cleaning the object information of the target object, data positively affecting the analysis result is retained, thereby improving the analysis efficiency.

[0051] In an optional embodiment, in the process of predicting the object behavior of the target object in the current stage based on the at least one target object feature to obtain a prediction result, first, the target feature label corresponding to the target object in the current stage is determined according to the at least one target object feature, and then the object behavior of the target object in the current stage is predicted based on the target feature label to obtain a prediction result. The target feature label at least includes a static label and a dynamic label.

[0052] In an optional embodiment, before the target feature label corresponding to the target object in the current stage is determined according to the at least one target object feature, the object features of a plurality of objects corresponding to each stage are first obtained, and then the association degree between the object features and at least one stage is analyzed to determine the feature label corresponding to each stage. The object features of the plurality of objects correspond to the business of the corresponding stage.

[0053] Optionally, before the target feature label corresponding to the target object in the current stage is determined according to the at least one target object feature, 100 customer information data stored by the financial institution is selected as sample data, and the sample data is input into the hourglass model as shown in Figure 2 After the foregoing data cleaning process, the object features of a plurality of objects corresponding to each stage can be obtained, and then the label definition process is performed, that is, the clustering analysis algorithm is adopted and the five-dimensional data obtained by the foregoing classification is used to label the 100 sample customers. For example, for the data in the demographic information dimension and the customer value dimension in the five dimensions, the feature label as shown in Table 2 can be obtained by the clustering analysis algorithm.

[0054] Table 2 Feature Label Example Table

[0055] Dimension information Feature label Age (demographic dimension) Young people (18-40 years old) Middle-aged people (40-65 years old) Old people (over 66 years old) Annual income (customer value dimension) High-income people (over 400,000 per year) Middle-income people (10-40 per year) Low-income people (less than 100,000 per year)

[0056] Specifically, the clustering analysis is to divide a group of data into different groups according to the similarity and difference in different dimensions. The purpose of the clustering analysis is to make the similarity between the same group of data as large as possible and the difference as small as possible, and the similarity between different groups of data as small as possible and the difference as large as possible. For example, when analyzing the consumption ability of 10 customers, among which, 3 customers have a consumption amount of 1 yuan, 2 customers have a consumption amount of 6 yuan, 2 customers have a consumption amount of 7 yuan, and 3 customers have a consumption amount of 10 yuan, the analysis result obtained by using the clustering method is that there are 3 customers with low consumption ability, the consumption ability is 1 yuan, there are 3 customers with high consumption ability, the consumption ability is 10 yuan, and there are 4 customers with medium consumption ability, the consumption ability is 6.5 yuan.

[0057] Further, the degree of association between the characteristics of the object and at least one stage is analyzed to determine the characteristic label corresponding to each stage. For example, in the precise customer acquisition stage of the customer life cycle, the financial institution has a demand for mining potential customers of credit card business. After cleaning and preliminary analysis of the information of 100 sample customers, the highly similar behavior characteristics of the sample customers who have applied for credit cards can be extracted. The degree of association between these variable characteristics and whether the customer has applied for a credit card is analyzed, and the relatively larger degree of association is selected as the characteristic label used for predicting the to-be-predicted customer. Optionally, as shown in Table 3, five variable characteristics are selected as the characteristic label used for predicting the to-be-predicted customer.

[0058] Table 3: Variable characteristic example table

[0059] Variable characteristics Interpretation Age Between 18 and 50 years old Gender Male or female Monthly income Monthly average income in the past year Monthly consumption Monthly average consumption in the past year Credit qualification Credit score (five levels from high to low)

[0060] It should be noted that the more the number of similar behavior characteristics selected according to the degree of association, the wider the range covered, the more the customer information contained is directional, and the more accurate the final prediction result will be.

[0061] Further, the characteristic label selected based on the information of 100 sample customers and used for predicting the to-be-predicted customer is used as the target characteristic label. The object behavior of the target object in the current stage is predicted to obtain a prediction result, that is, the possibility of the customer with the characteristic to apply for a credit card business is predicted, and a conclusion is drawn on whether the customer with the characteristic applies for a credit card business.

[0062] It should be noted that by analyzing a plurality of objects, the characteristic label used for predicting the to-be-predicted customer, i.e., the target characteristic label, can be obtained, which can quickly mine, maintain, and develop customers in each stage of the customer life cycle, and thus achieve the purpose of precise marketing.

[0063] In one optional embodiment, in the process of analyzing the correlation between object features and at least one stage to determine the feature label corresponding to each stage, the correlation between static features in the object features and at least one stage is first analyzed to determine the static label corresponding to each stage. Then, the correlation between dynamic features in the object features and at least one stage is analyzed to determine the dynamic label corresponding to each stage. Static features represent the inherent attribute information of multiple objects, while dynamic features represent features that change with the object behavior of multiple objects.

[0064] Specifically, such as Figure 2 As shown, after completing the label definition process, the feature engineering process is performed. Based on the feature labels, a customer profile is built. Target feature labels can be divided into static labels and dynamic labels. Static labels can be labels corresponding to fixed, unchanging features such as gender, while dynamic labels can be labels corresponding to variable features such as age and spending power. Therefore, in the aforementioned process of obtaining the feature labels (target feature labels) used for predicting customers by analyzing multiple objects, we can analyze static and dynamic features separately. Specifically, we analyze the correlation between static features and at least one stage to determine the static label corresponding to each stage, and then analyze the correlation between the dynamic features in the object features and at least one stage to determine the dynamic label corresponding to each stage.

[0065] Optionally, in the process of building the customer feature engineering of the financial institution, in combination with all stages of the customer life cycle, it can be divided into six dimensions of customer value classification, customer behavior preference, customer potential demand, key focus customer, customer risk score, and business marketing expansion. Each dimension can develop corresponding sub-labels. Among them, the classification methods and principles of the six dimensions are as follows: customer value classification, according to the contribution and activity of the customer to the financial institution, the customer is analyzed, which can be used for the reasonable and effective allocation of resources, each customer has only one in this class of label, and is subdivided into: mass customers, potential customers, growing customers, outstanding customers, etc.; customer behavior preference, through the customer's transaction channel and transaction type, the customer's characteristics are analyzed, which is convenient for the customer manager to dig and understand the customer, and is subdivided into: transaction channel preference, transaction type preference, etc.; customer potential demand, according to the past behavior of the customer, from the specific product, the potential customer is mined, and the precise marketing is carried out, and is subdivided into: investment and financial potential customers, credit card potential customers, etc.; key focus customer, in the process of daily operation, the primary focus of the customer group, including fraud customers, blacklist customers, information incomplete customers, and information untruthful customers; customer risk score, showing the results of various credit scoring models of the bank, and is subdivided into: high-risk customers, general-risk customers, and low-risk customers; business marketing expansion, developed for supporting a certain or a certain type of marketing activity of the business department, and is subdivided into: non-effective active households, and lost customers.

[0066] It should be noted that by analyzing the static features and dynamic features of the object, a customer feature engineering with certain universality is built, the customer is labeled, a diversified and multi-angle customer portrait is provided, and multi-aspect analysis of the customer is realized.

[0067] In an optional embodiment, in the process of predicting the object behavior of the target object in the current stage based on the target feature label to obtain a prediction result, first, the business information of at least one business corresponding to the current stage is acquired, then the association degree between the business information and the target feature label is determined, and then the target business information corresponding to the target object is determined from the at least one business corresponding to the current stage according to the association degree, wherein the association degree between the target business information and the target feature label is greater than a preset association degree, and the target business information represents the intended business of the target object.

[0068] Optionally, in the precise customer acquisition stage of the customer life cycle, the financial institution has different business needs, such as bank card business, credit card business, etc. The target feature label determined by the foregoing is associated with the bank card business information and the credit card business information for correlation degree analysis. For example, when a customer uses a mobile phone to shop on an e-commerce platform, the customer often uses a payment method of installment payment, which corresponds to the feature of the customer using a mobile phone to make installment consumption on a third-party platform. The feature label corresponding to the feature can be a credit card potential customer label. The label and the bank card business information and the credit card business information are associated for correlation degree analysis. Optionally, the correlation degree has a threshold, that is, the preset correlation degree can be 90%. When the correlation degree between the target feature label and the business information is greater than 90%, it can be determined that the target object is interested in the credit card business.

[0069] Optionally, in some other embodiments, after the five feature labels of age, gender, monthly income, monthly consumption, and credit quality of the foregoing 100 sample customer data shown in Table 3 are selected, a label group shown in Table 4 is formed. Through the hourglass model shown in Table 4, it can be predicted whether these customers will apply for a credit card. Part of the prediction results are shown in Table 5. Among them, the age range of the customer is 18-50 years old; 0 in the gender label indicates that the gender is female, and 1 indicates that the gender is male; the credit quality is divided into five levels, and the larger the number is, the better the credit quality is. Optionally, the information of whether the credit card has been applied for is counted, and 0 indicates that the credit card has not been applied for, and 1 indicates that the credit card has been applied for. Figure 2

[0070] Table 4: Customer sample information data example table

[0071]

[0072] Table 5: Model prediction result example table

[0073]

[0074] It should be noted that the customers whose prediction values are 1 are further screened, which are the target customers of personal credit card business that the marketing personnel are looking for, and the purpose of precise customer acquisition is achieved.

[0075] It can be seen that, through the technical scheme of the present application, the purpose of understanding the customer in depth and comprehensively is achieved, so that the technical effect of improving the accuracy of customer behavior judgment is realized, and the technical problem of low accuracy of customer behavior judgment caused by incomplete customer analysis in the prior art is solved.

[0076] Embodiment 2

[0077] According to the embodiment of the present application, an embodiment of an object behavior prediction device is provided, wherein,​Figure 3 is a schematic diagram of an optional object behavior prediction device according to an embodiment of the present application, as shown, the device comprises: an acquisition module 301, configured to acquire object information of a target object, wherein the object information at least comprises behavior data and transaction data of the target object; a processing module 302, configured to perform cleaning processing on the object information to obtain at least one target object feature of the target object in a current stage, wherein the current stage is any one of at least one stage corresponding to a customer life cycle of the target object, and the customer life cycle represents a relationship between the behavior data of the target object and a target business; and a prediction module 303, configured to predict an object behavior of the target object in the current stage based on the at least one target object feature to obtain a prediction result. Figure 3

[0078] It should be noted that the acquisition module 301, the processing module 302 and the prediction module 303 correspond to steps S101-S103 in the above embodiment, and the three modules have the same examples and application scenarios as the corresponding steps, but are not limited to the content disclosed in the above embodiment 1.

[0079] Optionally, the processing module comprises: a cleaning module, configured to perform cleaning processing on the object information of the target object to obtain cleaned object information; a first determination module, configured to determine target object information corresponding to the current stage from the cleaned object information; and an extraction module, configured to perform feature extraction on the target object information to obtain at least one target object feature.

[0080] Optionally, the prediction module comprises: a second determination module, configured to determine a target feature label corresponding to the target object in the current stage according to the at least one target object feature, wherein the target feature label at least comprises a static label and a dynamic label; and a first prediction module, configured to predict the object behavior of the target object in the current stage based on the target feature label to obtain a prediction result.

[0081] Optionally, the object behavior prediction device further comprises: a first acquisition module, configured to acquire object features of a plurality of objects in each stage, wherein the object features of the plurality of objects correspond to a business of the corresponding stage; and a third determination module, configured to analyze an association degree between the object features and at least one stage to determine a feature label corresponding to each stage.

[0082] ​Optionally, the third determining module comprises: a fourth determining module, configured to analyze the association degree between static features in the object features and the at least one stage, and determine a static label corresponding to each stage, wherein the static features represent inherent attribute information of the plurality of objects; and a fifth determining module, configured to analyze the association degree between dynamic features in the object features and the at least one stage, and determine a dynamic label corresponding to each stage, wherein the dynamic features represent features that change with the object behaviors of the plurality of objects.

[0083] Optionally, the first predicting module comprises: a second obtaining module, configured to obtain service information of at least one service corresponding to the current stage; a sixth determining module, configured to determine the association degree between the service information and the target feature label; and a seventh determining module, configured to determine target service information corresponding to the target object from the at least one service corresponding to the current stage according to the association degree, wherein the association degree between the target service information and the target feature label is greater than a preset association degree, and the target service information represents an intended service of the target object.

[0084] Embodiment 3

[0085] According to another aspect of the embodiments of the present application, a computer readable storage medium is also provided, and the computer readable storage medium stores a computer program, wherein the computer program is configured to execute the object behavior prediction method when running.

[0086] Embodiment 4

[0087] According to another aspect of the embodiments of the present application, an electronic device is also provided, and the electronic device comprises: Figure 4 is a schematic diagram of an optional electronic device according to the embodiments of the present application, as Figure 4 shown, the electronic device comprises one or more processors; a memory, configured to store one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement a program for running, wherein the program is configured to execute the object behavior prediction method when running.

[0088] Embodiment 5

[0089] According to another aspect of the embodiments of the present application, a computer program product is also provided, comprising computer programs / instructions, which, when executed by a processor, implement the object behavior prediction method.

[0090] The above-mentioned serial numbers of the embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0091] In the above-mentioned embodiments of the present application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0092] In several embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. Among them, the above-mentioned device embodiments are only schematic, for example, the division of the units can be a logical function division, and actual implementation can have another division mode, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be through some interface, indirect coupling or communication connection between units or modules, which can be electrical or other forms.

[0093] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or they can be distributed to multiple units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0094] In addition, each functional unit in each embodiment of the present application can be integrated in a processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0095] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0096] The above is only the preferred embodiment of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should be considered as the protection scope of the present application.

Claims

1. A method of predicting behavior of an object, characterized by, The method comprises the following steps: obtaining object information of a target object, wherein the object information at least comprises behavior data and transaction data of the target object; cleaning the object information to obtain at least one target object feature of the target object in a current stage, wherein the current stage is any one of at least one stage corresponding to a customer life cycle of the target object, and the customer life cycle represents a relationship between the behavior data of the target object and a target business; predicting an object behavior of the target object in the current stage based on the at least one target object feature to obtain a prediction result; wherein a target feature label is obtained based on the at least one target object feature; wherein the object behavior of the target object in the current stage is predicted based on the target feature label to obtain the prediction result, comprising: obtaining business information of at least one business corresponding to the current stage; determining an association degree between the business information and the target feature label; determining target business information corresponding to the target object from the at least one business corresponding to the current stage according to the association degree, wherein the association degree between the target business information and the target feature label is greater than a preset association degree, and the target business information represents an intended business of the target object; wherein the object behavior of the target object in the current stage is predicted based on the at least one target object feature to obtain a prediction result, comprising: determining a target feature label corresponding to the target object in the current stage according to the at least one target object feature, wherein the target feature label at least comprises a static label and a dynamic label; predicting the object behavior of the target object in the current stage based on the target feature label to obtain the prediction result; wherein, before determining the target feature label corresponding to the target object in the current stage according to the at least one target object feature, the method further comprises: obtaining object features of a plurality of objects in each stage, wherein the object features of the plurality of objects correspond to businesses in the corresponding stages; analyzing an association degree between the object features and the at least one stage to determine a feature label corresponding to each stage; wherein the feature label is calculated by a clustering analysis algorithm.

2. The method of claim 1, wherein, cleaning the object information to obtain at least one target object feature of the target object in a current stage, comprising: cleaning the object information of the target object to obtain cleaned object information; determining target object information corresponding to the current stage from the cleaned object information; extracting features from the target object information to obtain the at least one target object feature.

3. The method of claim 1, wherein, analyzing an association degree between the object features and the at least one stage to determine a feature label corresponding to each stage, comprising: analyzing a degree of association between static features in the object features and the at least one stage, to determine a static label corresponding to each stage, wherein the static features represent inherent attribute information of the plurality of objects; analyzing a degree of association between dynamic features in the object features and the at least one stage, to determine a dynamic label corresponding to each stage, wherein the dynamic features represent features that change with object behaviors of the plurality of objects.

4. A device for predicting behavior of an object, characterized by, comprising: an acquisition module configured to acquire object information of a target object, wherein the object information at least includes behavior data and transaction data of the target object; a processing module configured to perform cleaning processing on the object information, to obtain at least one target object feature of the target object in a current stage, wherein the current stage is any one of at least one stage corresponding to a customer life cycle of the target object, and the customer life cycle represents a relationship between the behavior data of the target object and a target business; a prediction module configured to predict an object behavior of the target object in the current stage based on the at least one target object feature, to obtain a prediction result; wherein the device is further configured to obtain a target feature label based on the at least one target object feature; wherein the device is further configured to acquire business information of at least one business corresponding to the current stage, to determine a degree of association between the business information and the target feature label, and to determine target business information corresponding to the target object from the at least one business corresponding to the current stage according to the degree of association, wherein a degree of association between the target business information and the target feature label is greater than a preset degree of association, and the target business information represents an intended business of the target object; wherein the prediction module is further configured to determine a target feature label corresponding to the target object in the current stage according to the at least one target object feature, wherein the target feature label at least includes a static label and a dynamic label, and to predict the object behavior of the target object in the current stage based on the target feature label, to obtain the prediction result; wherein, before determining the target feature label corresponding to the target object in the current stage according to the at least one target object feature, the device is further configured to acquire object features of a plurality of objects in each stage, wherein the object features of the plurality of objects correspond to businesses in the corresponding stages, to analyze a degree of association between the object features and the at least one stage, and to determine a feature label corresponding to each stage; wherein the feature label is calculated by a clustering analysis algorithm.

5. A computer readable storage medium, characterized in that, The computer readable storage medium stores a computer program, wherein the computer program is configured to execute the object behavior prediction method in any one of claims 1 to 3 when running.

6. An electronic device, comprising: The electronic device includes one or more processors; A memory for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement a program for running, wherein the program is configured to perform the object behavior prediction method of any one of claims 1 to 3 when running.

7. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instructions, when executed by the processor, implement the object behavior prediction method of any one of claims 1 to 3.

Images