CN110874758A - Potential customer prediction method, device, system, electronic equipment and storage medium - Google Patents
Potential customer prediction method, device, system, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN110874758A CN110874758A CN201811024999.8A CN201811024999A CN110874758A CN 110874758 A CN110874758 A CN 110874758A CN 201811024999 A CN201811024999 A CN 201811024999A CN 110874758 A CN110874758 A CN 110874758A
- Authority
- CN
- China
- Prior art keywords
- prediction
- customer information
- potential customer
- sample
- quasi
- 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.)
- Pending
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
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a potential customer prediction method, a potential customer prediction device, a potential customer prediction system, electronic equipment and a storage medium, wherein the potential customer prediction method comprises the following steps: training a plurality of quasi-prediction models by adopting sample potential customer information; calculating the recall rates of the plurality of quasi-prediction models, and taking the quasi-prediction model with the highest accuracy as a prediction model when the recall rates of the plurality of quasi-prediction models are greater than a preset threshold; and taking the potential customer information to be predicted as the input of the prediction model, wherein the output of the prediction model is the predicted value of the potential customer information to be predicted, and the predicted value indicates the uniprobability of the potential customer. The method and the device improve the accuracy of the prediction of the potential customers.
Description
Technical Field
The invention relates to the technical field of computer application, in particular to a potential customer prediction method, device, system, electronic equipment and storage medium.
Background
The sales identifies the potential customers by the customer management system tracking the potential customer information, the quality of the potential customers depends on the experience judgment of sales personnel in the tracking process, but the tracking effect is greatly influenced due to the difference of sales capacity and large liquidity, and the disturbance to the unintended potential customers is serious. The main technology for judging the quality of potential customers at present is as follows:
the prior art is simple analysis and manual judgment, and lacks deep application and mining on business understanding and data, however, the method has the following defects: 1) the identification of the potential client is only the presentation of data summarization inside the client management system, and is only the fragment information; 2) the potential customer quality completely depends on manual judgment, and resource waste and cost increase are caused by a large amount of follow-up and conversation; 3) the need for specialized data mining teams to support is time consuming and the features employed result in poor model accuracy for only internal data (potential customers are not yet clients, data integrity is insufficient) generation.
Disclosure of Invention
The present invention is directed to a method, apparatus, system, electronic device, and storage medium for predicting potential customers that overcome the limitations and disadvantages of the related art, which may result in one or more of the problems due to limitations and disadvantages of the related art.
According to an aspect of the present invention, there is provided a potential customer prediction method, including:
training a plurality of quasi-prediction models by adopting sample potential customer information;
calculating the recall rates of the plurality of quasi-prediction models, and taking the quasi-prediction model with the highest accuracy as a prediction model when the recall rates of the plurality of quasi-prediction models are greater than a preset threshold; and
and taking the potential customer information to be predicted as the input of the prediction model, wherein the output of the prediction model is the predicted value of the potential customer information to be predicted, and the predicted value indicates the uniprobability of the potential customer.
Optionally, the sample potential customer information and/or the potential customer information to be predicted come from multiple platforms and form potential customer information of the same potential customer based at least on matching of identification information of the potential customer.
Optionally, the training the plurality of quasi-predictive models with the sample potential customer information includes:
and performing characteristic processing on the sample potential customer information.
Optionally, the characterizing the sample potential customer information comprises:
and performing at least one of abnormal value processing, missing value processing, normalization processing and discretization processing on the sample potential customer information.
Optionally, after performing the feature processing on the sample potential customer information, the method further includes:
and if the sample potential customer information after the characteristic processing is a discrete field, carrying out chi-square inspection on the sample potential customer information after the characteristic processing.
Optionally, after performing the feature processing on the sample potential customer information, the method further includes:
and if the sample potential customer information after the characteristic processing is the continuous field, performing F test on the sample potential customer information after the characteristic processing.
Optionally, the potential customer information to be predicted is subjected to feature processing in the same manner as the sample potential customer information.
Optionally, the plurality of quasi-predictive models are binary models.
Optionally, the two classification models include at least two of a logistic regression model, a support vector machine model, and a neural network model.
According to another aspect of the present invention, there is also provided a potential customer prediction apparatus, including:
the training module is used for training a plurality of quasi-prediction models by adopting the sample potential customer information;
the model selection module is used for calculating the recall rates of the multiple quasi-prediction models, and when the recall rates of the multiple quasi-prediction models are larger than a preset threshold value, the quasi-prediction model with the highest accuracy rate is used as the prediction model; and
and the prediction module is used for taking the information of the potential customers to be predicted as the input of the prediction model, and the output of the prediction model is the predicted value of the information of the potential customers to be predicted, wherein the predicted value indicates the uniprobability of the potential customers.
According to still another aspect of the present invention, there is also provided an electronic apparatus, including: a processor; a storage medium having stored thereon a computer program which, when executed by the processor, performs the steps as described above.
According to yet another aspect of the present invention, there is also provided a storage medium having stored thereon a computer program which, when executed by a processor, performs the steps as described above.
Compared with the prior art, the invention has the advantages that:
on one hand, the training of a plurality of prediction models is combined, and when the recall rate is greater than a preset threshold value, the prediction model with the highest accuracy rate is adopted for prediction, so that the prediction accuracy rate of potential customers is increased; on the other hand, the predicted value of the potential customer prediction for indicating the singleton probability of the potential customer is output by the prediction model, so that the working personnel can conveniently carry out subsequent work based on the predicted value, and the resource waste and the cost increase caused by follow-up and conversation are relieved.
Drawings
The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.
FIG. 1 shows a flow diagram of a potential customer prediction method according to an embodiment of the invention.
Fig. 2 is a system architecture diagram illustrating a method of latent customer prediction in accordance with a specific embodiment of the present invention.
Fig. 3 shows a block diagram of a potential customer prediction apparatus according to an embodiment of the present invention.
Fig. 4 is a block diagram of a potential customer prediction apparatus according to a specific embodiment of the present invention.
Fig. 5 schematically illustrates a computer-readable storage medium in an exemplary embodiment of the invention.
Fig. 6 schematically shows an electronic device in an exemplary embodiment of the invention.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Furthermore, the drawings are merely schematic illustrations of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus their repetitive description will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor devices and/or microcontroller devices.
The flow charts shown in the drawings are merely illustrative and do not necessarily include all of the steps. For example, some steps may be decomposed, and some steps may be combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.
FIG. 1 shows a flow diagram of a potential customer prediction method according to an embodiment of the invention. Referring to fig. 1, the potential customer prediction method includes the steps of:
step S110: training a plurality of quasi-prediction models by adopting sample potential customer information;
step S120: calculating the recall rates of the plurality of quasi-prediction models, and taking the quasi-prediction model with the highest accuracy as a prediction model when the recall rates of the plurality of quasi-prediction models are greater than a preset threshold; and
step S130: and taking the potential customer information to be predicted as the input of the prediction model, wherein the output of the prediction model is the predicted value of the potential customer information to be predicted, and the predicted value indicates the uniprobability of the potential customer.
In the method for predicting the potential customers according to the exemplary embodiment of the present invention, on one hand, training of a plurality of prediction models is combined, and when the recall rate is greater than a predetermined threshold, the prediction model with the highest accuracy is used for prediction, so as to increase the accuracy of prediction of the potential customers; on the other hand, the predicted value of the potential customer prediction for indicating the singleton probability of the potential customer is output by the prediction model, so that the working personnel can conveniently carry out subsequent work based on the predicted value, and the resource waste and the cost increase caused by follow-up and conversation are relieved.
The prospective customer information may include follow-up information for the prospective customer, customer information for the prospective customer, and the like. The potential customer information may primarily include a continuous field and a discrete field.
The above steps are further described with reference to the system architecture diagram of a potential customer prediction method according to an embodiment of the present invention shown in fig. 2.
Specifically, the sample potential customer information and the potential customer information to be predicted can obtain the potential customer information from a plurality of platforms.
For example, the potential customer information may be obtained from a potential customer information log 201 of the customer management system; the potential customer information may also be obtained from potential customer information sources 202 of other platforms (such as e-commerce, search engines, vertical websites, telecommunications, etc.), whereby the potential customer information may be refined based on the identity of the potential customer.
In the step S110, the potential customer information may be obtained from the potential customer information log 201 and the potential customer information source 202 of the customer management system as the sample potential customer information 203.
Further, the training of the plurality of quasi-predictive models by using the sample potential customer information 203 in step S110 further includes: the sample potential customer information 203 is feature processed.
Characterizing the sample potential customer information 203 may include: and performing at least one of abnormal value processing, missing value processing, normalization processing and discretization processing on the sample potential customer information 203.
The abnormal value processing is mainly directed to the sample potential customer information 203 of the continuous field, the abnormal value processing of the continuous field firstly calculates the mean value and the standard deviation of the continuous field of each potential customer, and if the value of the potential customer information of the current continuous field is greater than (mean value + M standard deviation), the value of the potential customer information of the current continuous field is corrected to (mean value + M standard deviation). The average value + M standard deviation is used as a critical value in the abnormal value processing, and M is a constant of 3 or more.
In missing value processing, sample potential customer information 203 of the missing discrete field is assigned to-999; the sample potential customer information 203 of the continuous type field positively correlated with the predicted value is assigned to 0; the sample potential customer information 203 for the continuous type field negatively correlated with the predicted value is assigned as a critical value among the above-mentioned abnormal values.
In the normalization process, the sample potential customer information 203 of each continuous field is assigned x-min (x)/(max) (x) -min (x)), where x is the value of the sample potential customer information 203 of the current continuous field, min (x) is the minimum value of the sample potential customer information 203 of each potential customer in the continuous field, and max (x) is the maximum value of the sample potential customer information 203 of each potential customer in the continuous field.
In the discrete processing, the sample potential customer information 203 of the discrete type field, such as the region, is set to (a)01，a02，a03) Wherein, it is equivalent to determine whether the current region is a01Whether the current region is a02Whether the current region is a03Three values (e.g., 1 if the value is assigned, and 0 if the value is not assigned) form a feature.
Further, the above feature processing may further include encrypting the field name, and converting the field name from a regular named field to a field name without business meaning. For example, MD5 (fifth version of the message digest algorithm) encryption may be performed on field names required for matching potential customer data between the potential customer information log 201 and different potential customer information sources 202, such as ID, cell phone number, name, device number, etc.
Further, in order to increase the accuracy of model training, after the step of performing feature processing on the sample potential customer information 203, the step of selecting the sample potential customer information 203 after feature processing is further included.
Specifically, if the sample potential customer information 203 after feature processing is a discrete field, the chi-square test is performed on the sample potential customer information 203 after feature processing.
In the chi-square checking step, an actual distribution of the sample potential customer information 203 of the feature-processed discrete field and the label (for example, whether the sample potential customer information 203 is marked as an order or not) is first calculated, then an expected distribution of the sample potential customer information 203 of the feature-processed discrete field and the label is calculated, a chi-square value and a degree of freedom of the sample potential customer information 203 of the feature-processed discrete field are calculated based on the actual distribution and the expected distribution, and whether the sample potential customer information 203 of the feature-processed discrete field is significantly related to the label (significance of the chi-square value) is determined based on the chi-square value table. That is, in this step, only the sample potential customer information 203 of the discrete field having a significant relationship with the label may be retained, so as to reduce the redundant computation of the sample potential customer information 203 with a small correlation in the model, and speed up the model training efficiency.
Specifically, if the sample potential customer information 203 after feature processing is a continuation type field, the F-test is performed on the sample potential customer information 203 after feature processing.
In the F-test step, a first standard deviation of the sample potential customer information 203 after feature processing as a continuation type field is calculated first, then a second standard deviation of a predicted value of a mark (for example, whether the mark of the sample potential customer information 203 is a continuation type field, the mark of the continuation type field is 1, and the mark of the continuation type field is 0) is calculated, an F value is calculated as the first standard deviation/the second standard deviation, and it is determined whether the sample potential customer information 203 after feature processing is a continuation type field and the predicted value of the mark is significant according to an F value table. That is, in this step, only the sample potential customer information 203 that is a continuous field having a significant relationship with the label may be retained, so as to reduce the redundant computation of the sample potential customer information 203 with a small correlation in the model, and speed up the model training efficiency.
The plurality of quasi-predictive models are then trained using the feature-processed and filtered sample potential customer information 203, as at reference numeral 205. The plurality of quasi-predictive models may be binary models. The two classification models may include at least two of a logistic regression model, a support vector machine model, and a neural network model. The reason for the wide application of the logistic regression model is mainly because of the dominant characteristic of the probability expression, the solving speed of the logistic regression model is high, and the logistic regression model is convenient to apply. When the selection set of the logistic regression model is not changed, but only when the level of each variable is changed (for example, the travel time is changed), the selection probability of each selection branch in the new environment can be conveniently solved. According to the characteristics of the logistic regression model, the reduction or increase of the selection branches does not affect the selection probability ratio of other selections, so that the selection branches needing to be removed can be directly removed from the logistic regression model, and the newly added selection branches can be added into the logistic regression model to be directly used for prediction. The support vector machine model maps a sample space into a high-dimensional or infinite-dimensional feature space (Hilbert space) through a nonlinear mapping p, so that the problem of nonlinear divisibility in the original sample space is converted into the problem of linear divisibility in the feature space. In general, the complexity of calculation is increased, and even a "dimension disaster" is caused, so that people have little need for a lot of attention. The general dimensionality increase brings complexity to calculation, and the support vector machine model ingeniously solves the problem: by applying the expansion theorem of the kernel function, the explicit expression of the nonlinear mapping is not required to be known; because the linear learning machine is built in the high-dimensional feature space, the computational complexity is hardly increased compared with the linear model, and the 'dimensionality disaster' is avoided to some extent. The BP neural network model in the neural network model is a multilayer network for carrying out weight training on a nonlinear differentiable function. The method has the greatest characteristic that the high nonlinear mapping from pm space to yn space (the number of output nodes) consisting of m mode vectors p of input neurons can be realized for the system only by means of sample data without establishing a mathematical model of the system. The BP algorithm is proposed for solving the weight coefficient optimization of the multilayer forward neural network. Other two-classification models can be adopted in the invention, and are not described again.
Further, the quasi-prediction models with the highest accuracy are used as the prediction models when the recall rates of the quasi-prediction models are greater than a preset threshold (for example, close to 100%) by calculating the recall rates of the quasi-prediction models (the recall rate is predicted to be positive and is actually the number of positive samples/the number of all positive samples). Therefore, the subsequent prediction can be carried out by using the prediction model with the highest accuracy rate while considering the recall rate. In addition, the present invention can also calculate the sample number and positive sample data of different prediction segments ([0.0,0.1], [0.1,0.2], …, [0.9,1.0]) of each quasi-prediction model for analysis application.
Further, the characteristic processing steps of the potential customer information to be predicted 204 may be the same as those of the sample potential customer information 203. Then, as indicated by reference numeral 206, the potential customer information to be predicted is taken as an input of the prediction model, and an output of the prediction model is a predicted value 207 of the potential customer information to be predicted, wherein the predicted value 207 indicates a probability that a potential customer is unifonn.
The following describes a potential customer prediction device provided by the present invention with reference to fig. 3. Fig. 3 shows a block diagram of a potential customer prediction apparatus according to an embodiment of the present invention. The potential customer prediction device 300 includes a training module 310, a model selection module 320, and a prediction module 330.
The training module 310 is configured to train a plurality of quasi-predictive models using the sample potential customer information;
the model selection module 320 is configured to calculate recall rates of the multiple quasi-prediction models, and when the recall rates of the multiple quasi-prediction models are greater than a preset threshold, use the quasi-prediction model with the highest accuracy as the prediction model; and
the prediction module 330 is configured to use the information of the potential customer to be predicted as an input of the prediction model, and an output of the prediction model is a predicted value of the information of the potential customer to be predicted, where the predicted value indicates a uniprobability of the potential customer.
The prospective customer information may include follow-up information for the prospective customer, customer information for the prospective customer, and the like. The potential customer information may primarily include a continuous field and a discrete field.
In the potential customer prediction device according to the exemplary embodiment of the present invention, on one hand, training of a plurality of prediction models is combined, and when the recall rate is greater than a predetermined threshold, the prediction model with the highest accuracy is used for prediction, so as to increase the accuracy of prediction of a potential customer; on the other hand, the predicted value of the potential customer prediction for indicating the singleton probability of the potential customer is output by the prediction model, so that the working personnel can conveniently carry out subsequent work based on the predicted value, and the resource waste and the cost increase caused by follow-up and conversation are relieved.
Further, referring to fig. 4, fig. 4 shows a block diagram of a potential customer prediction apparatus according to a specific embodiment of the present invention. The potential customer prediction device 400 includes a training module 410, a model selection module 420, and a prediction module 430. The functions of the modules are the same as the training module 310, the model selection module 320, and the prediction module 330 in fig. 3. Unlike fig. 3, the training module 410 includes a feature processing module 411 and a feature selection module 412.
The feature processing module 411 is used for performing feature processing on the sample potential customer information. Characterizing the sample potential customer information may include: and performing at least one of abnormal value processing, missing value processing, normalization processing and discretization processing on the sample potential customer information.
The feature selection module 412 selects the sample potential customer information after feature processing. The feature selection module 412 includes a discrete verification module 4121 and a continuous verification module 4122.
Specifically, if the sample potential customer information after feature processing is a discrete field, the discrete verification module 4121 performs chi-square verification on the sample potential customer information after feature processing.
In the chi-square checking step, an actual distribution of the sample potential customer information and the label of the discrete field after feature processing (for example, whether the sample potential customer information is marked as a bill) is first calculated, then an expected distribution of the sample potential customer information and the label of the discrete field after feature processing is calculated, a chi-square value and a degree of freedom of the sample potential customer information of the discrete field after feature processing are calculated based on the actual distribution and the expected distribution, and whether the sample potential customer information of the discrete field after feature processing is significantly related to the label (significance of the chi-square value) is determined based on the chi-square value table. In other words, in this step, only the sample potential customer information of the discrete field having a significant relationship with the label may be retained, so as to reduce the redundant computation of the sample potential customer information with a small correlation in the model, and accelerate the model training efficiency.
Specifically, if the sample potential customer information after feature processing is the continuation type field, the continuation check module 4122 performs the F-check on the sample potential customer information after feature processing.
In the F-test step, a first standard deviation of the sample potential customer information after feature processing as the continuous field is calculated, then a second standard deviation of the predicted value of the mark (for example, whether the mark of the sample potential customer information is a singleton, the singleton mark is 1, and the singleton mark is 0) is calculated, an F value is calculated as the first standard deviation/the second standard deviation, and whether the sample potential customer information after feature processing is the continuous field and the predicted value of the mark is significant is determined according to an F value table. In other words, in this step, only the sample potential customer information of the continuous field having a significant relationship with the label may be retained, so as to reduce the redundant computation of the sample potential customer information with a small correlation in the model, and speed up the model training efficiency.
In an exemplary embodiment of the present invention, there is also provided a computer-readable storage medium having stored thereon a computer program which, when executed by, for example, a processor, can implement the steps of the electronic prescription flow processing method described in any one of the above embodiments. In some possible embodiments, aspects of the present invention may also be implemented in the form of a program product comprising program code for causing a terminal device to perform the steps according to various exemplary embodiments of the present invention described in the above-mentioned electronic prescription flow processing method section of this specification, when the program product is run on the terminal device.
Referring to fig. 5, a program product 500 for implementing the above method according to an embodiment of the present invention is described, which may employ a portable compact disc read only memory (CD-ROM) and include program code, and may be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited in this regard and, in the present document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The computer readable storage medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable storage medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the tenant computing device, partly on the tenant device, as a stand-alone software package, partly on the tenant computing device and partly on a remote computing device, or entirely on the remote computing device or server. In the case of remote computing devices, the remote computing devices may be connected to the tenant computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).
In an exemplary embodiment of the invention, there is also provided an electronic device that may include a processor and a memory for storing executable instructions of the processor. Wherein the processor is configured to execute the steps of the electronic prescription flow processing method in any one of the above embodiments via execution of the executable instructions.
As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as a system, method or program product. Thus, various aspects of the invention may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, microcode, etc.) or an embodiment combining hardware and software aspects that may all generally be referred to herein as a "circuit," module "or" system.
An electronic device 600 according to this embodiment of the invention is described below with reference to fig. 6. The electronic device 600 shown in fig. 6 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.
As shown in fig. 6, the electronic device 600 is embodied in the form of a general purpose computing device. The components of the electronic device 600 may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 that connects the various system components (including the storage unit 620 and the processing unit 610), a display unit 640, and the like.
Wherein the storage unit stores program code executable by the processing unit 610 to cause the processing unit 610 to perform steps according to various exemplary embodiments of the present invention described in the above-mentioned electronic prescription flow processing method section of the present specification. For example, the processing unit 610 may perform the steps as shown in fig. 1.
The storage unit 620 may include readable media in the form of volatile memory units, such as a random access memory unit (RAM)6201 and/or a cache memory unit 6202, and may further include a read-only memory unit (ROM) 6203.
The memory unit 620 may also include a program/utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.
The electronic device 600 may also communicate with one or more external devices 700 (e.g., keyboard, pointing device, bluetooth device, etc.), with one or more devices that enable a tenant to interact with the electronic device 600, and/or with any devices (e.g., router, modem, etc.) that enable the electronic device 600 to communicate with one or more other computing devices. Such communication may occur via an input/output (I/O) interface 650. Also, the electronic device 600 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the Internet) via the network adapter 660. The network adapter 660 may communicate with other modules of the electronic device 600 via the bus 630. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiment of the present invention may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which may be a personal computer, a server, or a network device, etc.) to execute the above-mentioned electronic prescription flow processing method according to the embodiment of the present invention.
Compared with the prior art, the invention has the advantages that:
on one hand, the training of a plurality of prediction models is combined, and when the recall rate is greater than a preset threshold value, the prediction model with the highest accuracy rate is adopted for prediction, so that the prediction accuracy rate of potential customers is increased; on the other hand, the predicted value of the potential customer prediction for indicating the singleton probability of the potential customer is output by the prediction model, so that the working personnel can conveniently carry out subsequent work based on the predicted value, and the resource waste and the cost increase caused by follow-up and conversation are relieved.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.
Claims (12)
1. A method for predicting potential customers, comprising:
training a plurality of quasi-prediction models by adopting sample potential customer information;
calculating the recall rates of the plurality of quasi-prediction models, and taking the quasi-prediction model with the highest accuracy as a prediction model when the recall rates of the plurality of quasi-prediction models are greater than a preset threshold; and
and taking the potential customer information to be predicted as the input of the prediction model, wherein the output of the prediction model is the predicted value of the potential customer information to be predicted, and the predicted value indicates the uniprobability of the potential customer.
2. The prospective customer prediction method of claim 1 wherein the sample prospective customer information and/or the prospective customer information to be predicted come from multiple platforms and form prospective customer information for the same prospective customer based at least on a match of identifying information of the prospective customer.
3. The method of latent customer prediction according to claim 1, wherein training a plurality of quasi-predictive models using sample latent customer information comprises:
and performing characteristic processing on the sample potential customer information.
4. The method of prospective customer prediction according to claim 3 wherein characterizing the sample prospective customer information comprises:
and performing at least one of abnormal value processing, missing value processing, normalization processing and discretization processing on the sample potential customer information.
5. The method of prospective customer prediction according to claim 3, wherein after the characterizing the sample prospective customer information, further comprising:
and if the sample potential customer information after the characteristic processing is a discrete field, carrying out chi-square inspection on the sample potential customer information after the characteristic processing.
6. The method of prospective customer prediction according to claim 3, wherein after the characterizing the sample prospective customer information, further comprising:
and if the sample potential customer information after the characteristic processing is the continuous field, performing F test on the sample potential customer information after the characteristic processing.
7. The method of any of claims 3 to 6, wherein the potential customer information to be predicted is characterized in the same manner as the sample potential customer information.
8. The method of predicting potential customers of any one of claims 1 to 6, wherein the plurality of quasi-predictive models are binary models.
9. The method of latent customer prediction according to claim 8, wherein the two classification models comprise at least two of a logistic regression model, a support vector machine model, and a neural network model.
10. A prospective customer prediction apparatus comprising:
the training module is used for training a plurality of quasi-prediction models by adopting the sample potential customer information;
the model selection module is used for calculating the recall rates of the multiple quasi-prediction models, and when the recall rates of the multiple quasi-prediction models are larger than a preset threshold value, the quasi-prediction model with the highest accuracy rate is used as the prediction model; and
and the prediction module is used for taking the information of the potential customers to be predicted as the input of the prediction model, and the output of the prediction model is the predicted value of the information of the potential customers to be predicted, wherein the predicted value indicates the uniprobability of the potential customers.
11. An electronic device, characterized in that the electronic device comprises:
a processor;
a memory having stored thereon a computer program which, when executed by the processor, performs the steps of any of claims 1 to 9.
12. A storage medium, having stored thereon a computer program which, when executed by a processor, performs the steps of any of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811024999.8A CN110874758A (en) | 2018-09-03 | 2018-09-03 | Potential customer prediction method, device, system, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811024999.8A CN110874758A (en) | 2018-09-03 | 2018-09-03 | Potential customer prediction method, device, system, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110874758A true CN110874758A (en) | 2020-03-10 |
Family
ID=69716886
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811024999.8A Pending CN110874758A (en) | 2018-09-03 | 2018-09-03 | Potential customer prediction method, device, system, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110874758A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021139432A1 (en) * | 2020-10-13 | 2021-07-15 | 平安科技（深圳）有限公司 | Artificial intelligence-based user rating prediction method and apparatus, terminal, and medium |
| CN113157763A (en) * | 2021-01-04 | 2021-07-23 | 北京汇达城数科技发展有限公司 | Accurate identification system and method for user with specified behavior information |
-
2018
- 2018-09-03 CN CN201811024999.8A patent/CN110874758A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021139432A1 (en) * | 2020-10-13 | 2021-07-15 | 平安科技（深圳）有限公司 | Artificial intelligence-based user rating prediction method and apparatus, terminal, and medium |
| CN113157763A (en) * | 2021-01-04 | 2021-07-23 | 北京汇达城数科技发展有限公司 | Accurate identification system and method for user with specified behavior information |
| CN113157763B (en) * | 2021-01-04 | 2023-10-13 | 北京汇达城数科技发展有限公司 | Accurate identification system and method for user with specified behavior information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110070391B (en) | Data processing method and device, computer readable medium and electronic equipment | |
| CN111950738B (en) | Machine learning model optimization effect evaluation method, device, terminal and storage medium | |
| CN110349009B (en) | Multi-head lending default prediction method and device and electronic equipment | |
| CN112270545A (en) | Financial risk prediction method and device based on migration sample screening and electronic equipment | |
| CN110995459B (en) | Abnormal object identification method, device, medium and electronic equipment | |
| CN110599335A (en) | User financial risk assessment method and device based on multiple models | |
| CN110705719A (en) | Method and apparatus for performing automatic machine learning | |
| US20190026805A1 (en) | Issue resolution utilizing feature mapping | |
| US10824956B1 (en) | System and method for price estimation of reports before execution in analytics | |
| CN112015562A (en) | Resource allocation method and device based on transfer learning and electronic equipment | |
| CN112348662B (en) | Risk assessment method and device based on user occupation prediction and electronic equipment | |
| CN111178687A (en) | Financial risk classification method and device and electronic equipment | |
| CN112508723A (en) | Financial risk prediction method and device based on automatic preferential modeling and electronic equipment | |
| CN111191677B (en) | User characteristic data generation method and device and electronic equipment | |
| CN110874758A (en) | Potential customer prediction method, device, system, electronic equipment and storage medium | |
| CN111210109A (en) | Method and device for predicting user risk based on associated user and electronic equipment | |
| US20220083519A1 (en) | Explanative analysis for records with missing values | |
| CN112488865A (en) | Financial risk prediction method and device based on financial time nodes and electronic equipment | |
| US20230206114A1 (en) | Fair selective classification via a variational mutual information upper bound for imposing sufficiency | |
| US20230169389A1 (en) | Domain adaptation | |
| US11809375B2 (en) | Multi-dimensional data labeling | |
| CN115221892A (en) | Work order data processing method and device, storage medium and electronic equipment | |
| US20220366300A1 (en) | Data drift mitigation in machine learning for large-scale systems | |
| US20140164069A1 (en) | Generating Global Optimized Strategies For Information Requests, Proposals, And Statements of Work Within a Time Period Across Hierarchical Entity Boundaries | |
| US11481212B2 (en) | Automatic identification of reference data |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: Room 221, 2 / F, block C, 18 Kechuang 11th Street, Daxing District, Beijing, 100176 Applicant after: Jingdong Technology Holding Co.,Ltd. Address before: Room 221, 2 / F, block C, 18 Kechuang 11th Street, Daxing District, Beijing, 100176 Applicant before: Jingdong Digital Technology Holding Co.,Ltd. Address after: Room 221, 2 / F, block C, 18 Kechuang 11th Street, Daxing District, Beijing, 100176 Applicant after: Jingdong Digital Technology Holding Co.,Ltd. Address before: Room 221, 2 / F, block C, 18 Kechuang 11th Street, Daxing District, Beijing, 100176 Applicant before: JINGDONG DIGITAL TECHNOLOGY HOLDINGS Co.,Ltd. Address after: Room 221, 2 / F, block C, 18 Kechuang 11th Street, Daxing District, Beijing, 100176 Applicant after: JINGDONG DIGITAL TECHNOLOGY HOLDINGS Co.,Ltd. Address before: Room 221, 2nd floor, Block C, 18 Kechuang 11th Street, Daxing Economic and Technological Development Zone, Beijing, 100176 Applicant before: BEIJING JINGDONG FINANCIAL TECHNOLOGY HOLDING Co.,Ltd. |