CN114549071A - Marketing strategy determination method and device, computer equipment and storage medium - Google Patents
Marketing strategy determination method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN114549071A CN114549071A CN202210149807.6A CN202210149807A CN114549071A CN 114549071 A CN114549071 A CN 114549071A CN 202210149807 A CN202210149807 A CN 202210149807A CN 114549071 A CN114549071 A CN 114549071A
- Authority
- CN
- China
- Prior art keywords
- marketing
- user
- characteristic information
- target user
- purchase probability
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a device for determining a marketing strategy, computer equipment and a storage medium, wherein the method comprises the following steps: acquiring user characteristic information of a target user and characteristic information of a product to be marketed; inputting the user characteristic information and the product characteristic information into a prediction model obtained by pre-training, and outputting purchase probability increment of a target user under different marketing strategies; and determining the marketing strategy suitable for the target user according to the purchase probability increment of the target user under different marketing strategies. According to the method, the user characteristic information and the product characteristic information are used as input through a pre-trained prediction model, the purchase probability increment of the target user under different marketing strategies is predicted, a specific quantification result of the marketing strategies on the purchase probability improvement is determined, and the corresponding marketing strategies are determined according to the quantification result, so that a more accurate marketing strategy is determined.
Description
Technical Field
The invention relates to the technical field of deep learning, in particular to a method and a device for determining a marketing strategy, computer equipment and a storage medium.
Background
In the product promotion process, in order to improve the attraction of the product to the user, the operator usually adopts some marketing strategies, such as consumption subsidies, to enhance the attraction and stickiness of the user to the product. Especially in the field of O2O, in order to ensure the stickiness and usage frequency of the user to the APP platform, the operator usually adopts a series of strategies to attract new users to register and stabilize the consumption habits of old users.
The APP in the shared traffic field is taken as an example, the APP special-shared riding card is a specific card type of the shared platform, the price of the APP special-shared riding card is consistent with that of the ordinary riding card, and only in order to attract users to transfer from other platforms to own platforms and retain registered users, the number of daily active users of the APP platform is improved. The special card type is set and the user who purchased the card type is given a certain subsidy amount. If the user purchases the card type, the user can enjoy more substantial price when using the service on the platform within a period of time, and for the platform, the activity and the retention probability of the user on the APP are improved, and the platform is given more service users.
For a platform operator, the adoption of the marketing strategy can theoretically enhance the attraction and stickiness of a user to the platform, but any marketing strategy generates corresponding cost, and how to obtain the maximum conversion rate under the condition of adopting low cost is an urgent problem to be solved.
The existing subsidy scheme only can play a role in enhancing the user viscosity, but cannot determine the specific quantitative results brought by the marketing strategies to the conversion rate improvement, so that a more accurate marketing means cannot be provided for specific people.
Disclosure of Invention
In view of the above drawbacks of the prior art, the technical problem to be solved by the present invention is that a more accurate marketing strategy cannot be determined due to the fact that the conversion rate is increased by the marketing strategy cannot be quantified in the prior art.
In order to achieve the above object, the present invention provides a method for determining a marketing strategy, including: acquiring user characteristic information of a target user and characteristic information of a product to be marketed; inputting the user characteristic information and the product characteristic information into a pre-trained prediction model, and outputting purchase probability increment of the target user under different marketing strategies, wherein the purchase probability increment is the increment of the purchase probability of the target user after the marketing strategy is adopted relative to the purchase probability without adopting the marketing strategy; and determining the marketing strategy suitable for the target user according to the purchase probability increment of the target user under different marketing strategies.
In a preferred embodiment of the present invention, the predicting model adopts an incremental model, and the inputting the user characteristic information and the product characteristic information into the predicting model obtained by pre-training and outputting the purchase probability increment of the target user under different marketing strategies includes: copying the user characteristic information and the product characteristic information into a plurality of groups of input data; inputting one of the multiple sets of input data into a control group, and respectively inputting other sets of input data into multiple processing groups, wherein the control group is an input data set which does not adopt marketing strategies, the processing groups are input data sets which adopt marketing strategies, and each processing group corresponds to one marketing strategy; forecasting the purchase probability of the input data of the control group to obtain a first purchase probability which does not adopt a marketing strategy; predicting purchase probability of the input data of each processing group to obtain a plurality of second purchase probabilities under different marketing strategies; and obtaining the purchase probability increment by utilizing the difference between the second purchase probability and the first purchase probability.
In a preferred embodiment of the present invention, the determining a marketing strategy suitable for the target user according to the purchase probability increment of the target user under different marketing strategies includes: and determining the marketing strategy with the largest purchase probability increment as the marketing strategy of the target user.
In a preferred embodiment of the present invention, the determining a marketing strategy suitable for the target user according to the purchase probability increment of the target user under different marketing strategies includes: acquiring a user type of the target user, wherein the user type is determined according to the activity degree of the user; acquiring preset marketing operation cost; and determining the marketing strategy of the target user according to the user type, the marketing operation cost and the purchase probability increment.
In a preferred embodiment of the present invention, the obtaining the user type of the target user includes: acquiring historical behavior characteristic data of the target user; dividing the target users into corresponding user types according to the historical behavior characteristic data, wherein the user types comprise: active users, silent users, and attrition users.
In a preferred embodiment of the present invention, the determining the marketing strategy of the target user according to the user type, the marketing operation cost and the purchase probability increment includes: determining a target relation table according to the user type, wherein the target relation table records the corresponding relation between the probability increment threshold and the marketing operation cost; determining a probability increment threshold of the target user by using the target relation table and the marketing operation cost; judging whether the purchase probability increment of the target user under different marketing strategies is smaller than the probability increment threshold value; and when the probability increment is smaller than the probability increment threshold, selecting the marketing strategy with the minimum cost.
In a preferred embodiment of the present invention, the prediction model is obtained by training the following steps: dividing an experimental target population into a plurality of treatment groups and a control group, wherein each treatment group adopts a marketing strategy, and the control group does not adopt the marketing strategy; recording a conversion result, and recording the successfully purchased users as positive samples and recording the users not purchased as negative samples; acquiring user characteristic information and product characteristic information of all users; and training to obtain the prediction model by using the conversion result, the user characteristic information and the product characteristic information as training samples.
In order to achieve the above object, the present invention further provides a device for determining a marketing strategy, including: the acquisition module is used for acquiring the user characteristic information of a target user and the characteristic information of a product to be marketed; the detection module is used for inputting the user characteristic information and the product characteristic information into a pre-trained prediction model and outputting purchase probability increment of the target user under different marketing strategies, wherein the purchase probability increment is the increment of the purchase probability of the target user after the marketing strategy is adopted relative to the purchase probability of the target user without the marketing strategy; and the determining module is used for determining the marketing strategy suitable for the target user according to the purchase probability increment of the target user under different marketing strategies.
To achieve the above object, the present invention also provides a computer device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to perform the method of determining marketing strategies described above.
To achieve the above object, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions for causing a computer to execute the marketing strategy determination method as described above.
The device or the method provided by the invention has the following technical effects:
according to the embodiment of the invention, the pre-trained prediction model is utilized, the user characteristic information and the product characteristic information are used as input, the purchase probability increment of the target user under different marketing strategies is predicted, the specific quantitative result of the marketing strategies on the purchase probability improvement is determined, and the corresponding marketing strategies are determined according to the quantitative result, so that the more accurate marketing strategies are determined.
The conception, the specific structure and the technical effects of the present invention will be further described with reference to the accompanying drawings to fully understand the objects, the features and the effects of the present invention.
Drawings
FIG. 1 is a flow chart of a preferred embodiment of a method for determining marketing strategies in accordance with an embodiment of the present invention;
FIG. 2 is a flow chart of a purchase probability calculation corresponding to a subsidy policy in an embodiment of the invention;
FIG. 3 is a schematic diagram of an incremental model of an embodiment of the invention;
FIG. 4 is a schematic diagram of a target relationship table of an embodiment of the present invention;
FIG. 5 is a flow chart of a model training process of an embodiment of the present invention;
FIG. 6 is a schematic diagram of a preferred embodiment of a marketing strategy determination apparatus according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of the computer device of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the drawings only show the components related to the present invention rather than the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
Some exemplary embodiments of the invention have been described for illustrative purposes, and it is to be understood that the invention may be practiced otherwise than as specifically described.
The invention provides a method for determining a marketing strategy, which can be applied to marketing of various products, including marketing and popularization of software service products such as APP, websites and the like, and can also be used for marketing and popularization of specific physical products. Correspondingly, the marketing strategy can also be adjusted according to needs, for example, the marketing strategy can be a subsidy strategy, a cash back strategy, a discount strategy and the like, each type of marketing strategy can also adopt different strength, for example, the subsidy limit is divided into a plurality of types, and the discount strategy can be divided into a plurality of discounts and the like.
The method for determining the marketing strategy can be executed through the background server, and the corresponding marketing strategy is calculated for each user and then issued to the user, so that the marketing purpose is realized.
As shown in fig. 1, a method for determining a marketing strategy according to an embodiment of the present invention includes:
and S101, acquiring user characteristic information of a target user and characteristic information of a product to be marketed.
The target user in this embodiment may refer to a user whose marketing strategy needs to be determined currently, and each user may be used as the target user, and the marketing strategy corresponding to the user is determined by the method of this embodiment. The user characteristic information may be attribute characteristics of the user, such as age, gender, constellation and the like, or may be user behavior characteristics, such as how long the user has not purchased or used the target product, a time difference between the last time of purchase or use and the current time, and the like. The product to be marketed can be software or a physical product in the above, such as an APP platform, and also can be a product such as a membership card promoted on the platform. The product characteristic information may refer to information such as the type of the product. Taking platform marketing of shared vehicles as an example, attribute characteristics (such as age, gender, constellation and the like) of a user, user behavior characteristics (such as how long the user has not ridden, the number of days since the last ride and the like), and card type characteristics (such as 7-day riding card, 30-day riding card and the like) can be obtained.
As an aspect of the embodiment of the present invention, in order to improve the accuracy of the determined marketing strategy, in the embodiment of the present invention, a scenario characteristic and a context characteristic may also be obtained, where the scenario characteristic may refer to a key event that occurs when the user uses the product, and the context characteristic may refer to information such as a current environment. Or taking a platform of a shared vehicle as an example, the scene characteristics may be the number of times of code scanning unlocking failure in a riding scene, the number of times of click in a card purchasing scene, and the like, and the real-time context characteristics may be a city, a region where the city is located, the current weather, and the like. The scene characteristics and the context characteristics are used as the input of the prediction model together, so that the accuracy of the prediction of the increment of the conversion rate (increment of the purchase probability) can be improved.
Step S102, inputting the user characteristic information and the product characteristic information into a pre-trained prediction model, and outputting purchase probability increment of the target user under different marketing strategies, wherein the purchase probability increment is the increment of the purchase probability of the target user after the marketing strategy is adopted relative to the purchase probability of the target user without the marketing strategy.
In this embodiment, the user characteristic information and the product characteristic information are input into a prediction model for prediction, and the prediction model outputs purchase probability increments of the target user under different marketing strategies. When the prediction model is trained, different marketing strategies are adopted for different users for training, namely, the marketing strategies are written into the model as curing parameters of the prediction model instead of being used as input data, and only the user characteristic information and the product characteristic information are used as input to carry out purchase probability increment prediction under the corresponding marketing strategies. Of course, in the embodiment of the present invention, the prediction model may be trained after adding different marketing strategies in the offline training process.
Specifically, different predictive models may be trained according to different marketing strategy types. For example, taking the subsidy policy as an example, different subsidy amounts can be set, and as different marketing policies, a prediction model capable of predicting product purchase probability increments corresponding to the different subsidy amounts is trained; taking a discount strategy as an example, different discounts can be set as different marketing strategies, and a prediction model capable of predicting purchase probability increments corresponding to the different discounts is trained.
In the embodiment of the invention, the purchase probability increment predicted by the prediction model is the promotion amount relative to the purchase probability of the marketing strategy which is not adopted, specifically, the prediction model can predict the purchase probability under each marketing strategy, then predict the purchase probability under the marketing strategy which is not adopted, and the two probabilities are subtracted to obtain the purchase probability increment, so that the influence condition of the marketing strategy on the purchase probability can be intuitively reflected, and the most appropriate marketing strategy is selected.
And S103, determining a marketing strategy suitable for the target user according to the purchase probability increment of the target user under different marketing strategies.
After the purchase probability increment under different marketing strategies is predicted, the marketing strategy suitable for the target user can be determined. Specifically, the marketing strategy corresponding to the maximum value of the purchase probability increment may be used as the marketing strategy of the target user, and the corresponding marketing means may be executed for the target user. Wherein, the larger the value of the increment of the purchase probability is, the maximum purchase probability of the target user under the marketing strategy is represented, and the promotion of the marketing strategy to the purchase probability is larger. On the other hand, from the viewpoint of cost, when the influence on the purchase probability is considerable, the marketing strategy with the lowest cost can be selected, and the marketing cost can be reduced. No matter which mode is adopted, the oriented marketing to specific users can be realized, and the maximum conversion rate is realized.
According to the embodiment of the invention, the pre-trained prediction model is utilized, the user characteristic information and the product characteristic information are used as input, the purchase probability increment of the target user under different marketing strategies is predicted, the specific quantitative result of the marketing strategies on the purchase probability improvement is determined, and the corresponding marketing strategies are determined according to the quantitative result, so that the more accurate marketing strategies are determined.
As an optional implementation manner of the embodiment of the present invention, in the embodiment of the present invention, the using an incremental model for the prediction model, inputting the user characteristic information and the product characteristic information into a prediction model obtained by pre-training, and outputting a purchase probability increment of the target user under different marketing strategies includes:
and step S1, copying the user characteristic information and the product characteristic information into a plurality of groups of input data.
In this embodiment, the user characteristic information and the product characteristic information are copied into a plurality of groups, and the user characteristic information and the product characteristic information in each group of data are the same, so that a plurality of groups of input data are formed, and are conveniently input into the control group and the plurality of processing groups. The number of sets of input data copied is equal to the sum of the number of processing sets and the number of control sets.
Step S2, inputting one of the input data sets into a control group, and inputting the other input data sets into a plurality of processing groups, wherein the control group is an input data set that does not adopt a marketing strategy, the processing groups are input data sets that adopt a marketing strategy, and each processing group corresponds to a marketing strategy.
In this embodiment, the prediction model is divided into two groups, one group is a control group, and the other group is a processing group, where the control group is a group, and the processing group may be multiple groups, and each group corresponds to one marketing strategy. And predicting each group of input data respectively to obtain the purchase probability under the corresponding marketing strategy.
And step S3, carrying out purchase probability prediction on the input data of the control group to obtain a first purchase probability which does not adopt the marketing strategy.
And step S4, carrying out purchase probability prediction on the input data of each processing group to obtain a plurality of second purchase probabilities under different marketing strategies.
And step S5, obtaining the purchase probability increment by using the difference between the second purchase probability and the first purchase probability.
As shown in fig. 2, for different marketing strategies (patch 1, patch 2 … …, patch 4), a plurality of purchase probabilities lift are obtained by the prediction model, and then the highest purchase probability is selected from the plurality of purchase probabilities as the optimal marketing strategy. Because the purchase probability without adopting the marketing strategy is a fixed value, the maximum purchase probability corresponding to the marketing strategy corresponds to the maximum value of the purchase probability increment.
In the embodiment of the invention, an incremental model, namely a TwoModel, is adopted, model training is shown in fig. 3, a traffic group is used as a processing group, a control group is used as a control group, and the two groups are respectively predicted to obtain corresponding purchase probabilities G and G'.
As an optional implementation manner, in the embodiment of the present invention, the determining, according to the purchase probability increment of the target user under different marketing strategies, a marketing strategy suitable for the target user includes: and determining the marketing strategy with the largest purchase probability increment as the marketing strategy of the target user.
In the embodiment of the invention, the marketing strategy corresponding to the maximum purchase probability increment is taken as the marketing strategy of the target user, so that the probability of purchasing products by the user is maximum, and accurate marketing is realized.
As a further optional implementation manner, the determining the marketing strategy suitable for the target user according to the purchase probability increment of the target user under different marketing strategies includes: acquiring a user type of the target user, wherein the user type is determined according to the activity degree of the user; acquiring preset marketing operation cost; and determining the marketing strategy of the target user according to the user type, the marketing operation cost and the purchase probability increment.
Specifically, the obtaining the user type of the target user includes: acquiring historical behavior characteristic data of the target user; dividing the target users into corresponding user types according to the historical behavior characteristic data, wherein the user types comprise: active users, silent users, and attrition users.
In this embodiment, each user can calculate a purchase probability increment through the prediction model, and then obtain the marketing strategy corresponding to the user, but in the actual operation process, the marketing strategy has cost constraints, and under certain cost control, the total profit of the platform can be maximized by maximizing the transformation and promotion proportion. Therefore, after the model is predicted, based on the purchase probability, the operation setting cost and the user type of the user, a user history hierarchy may also be formed (the user history hierarchy refers to dividing the user into different user groups, such as active users, silent users, lost users, and the like, according to the user's history behavior characteristics, such as card purchase times, riding times, active days, and the like), and different distribution strategies are formulated for different users in the hierarchy to adjust the marketing strategies of the user.
Further optionally, in the above embodiment, the determining the marketing strategy of the target user according to the user type, the marketing operation cost, and the purchase probability increment includes: determining a target relation table according to the user type, wherein the target relation table records the corresponding relation between the probability increment threshold and the marketing operation cost; determining a probability increment threshold of the target user by using the target relation table and the marketing operation cost; judging whether the purchase probability increment of the target user under different marketing strategies is smaller than the probability increment threshold value; and when the probability increment is smaller than the probability increment threshold, selecting the marketing strategy with the minimum cost.
In the embodiment of the present invention, a target relationship table corresponding to each user may be determined according to statistical data, where the target relationship table records a corresponding relationship between a probability increment threshold and a marketing operation cost, specifically as shown in fig. 4, in this embodiment, an estimation may be performed on the cost, and a subsidy threshold P _ value is set at the same time, if a purchase rate increment of a user under all subsidies is less than P _ value, it is considered that the minimum subsidy amount can be given to the user, after different parameters are tried for determining the P _ value, the cost is determined according to the cost, for example, the operation setting cost is 4.87 yuan, and as shown in fig. 4, a value of P _ value is 0.025.
According to the embodiment of the invention, the purchase probability increment is associated with the cost, so that a marketing strategy corresponding to a larger purchase probability increment is selected under the condition of reasonable cost control, and accurate marketing is realized.
As an alternative, the prediction model is obtained by training the following steps: dividing an experimental target population into a plurality of treatment groups and a control group, wherein each treatment group adopts a marketing strategy, and the control group does not adopt the marketing strategy; recording a conversion result, and recording the successfully purchased users as positive samples and recording the users not purchased as negative samples; acquiring user characteristic information and product characteristic information of all users; and training to obtain the prediction model by using the conversion result, the user characteristic information and the product characteristic information as training samples.
In the embodiment of the invention, random experiment setting is firstly carried out: since the sample data required by the model needs to satisfy the cia (conditional index assignment) hypothesis, random experiments were performed in the previous period, the users were divided into different experimental groups (each experimental group is a homogeneous user), and each group was given different marketing strategies, for example, different subsidies are set, as shown in fig. 5, and the transformation of the users is observed. Collecting training samples: and acquiring user data participating in random experiments and real conversion conditions of the user, marking the user as a positive sample if the user successfully purchases the product, and regarding the user as a negative sample if the user does not successfully purchase the product. Then, by using the collected sample data, the conversion rate-purchase probability is counted, and then the incremental model is trained to obtain a trained prediction model.
According to the embodiment of the invention, based on the user characteristic information and the product characteristic information, the personalized matching between the user and the marketing strategy is realized, and meanwhile, a distribution scheme based on cost control is designed, so that the marketing effect and the overall income are maximized under the condition of certain cost.
In another aspect of the embodiments of the present invention, there is provided a marketing strategy determining apparatus, which may be used to execute the method shown in fig. 1, as shown in fig. 6, and includes:
an obtaining module 601, configured to obtain user characteristic information of a target user and characteristic information of a product to be marketed;
a detection module 602, configured to input the user characteristic information and the product characteristic information into a pre-trained prediction model, and output a purchase probability increment of the target user under different marketing strategies, where the purchase probability increment is an increase of a purchase probability of the target user after a marketing strategy is adopted relative to a purchase probability of the target user without the marketing strategy;
the determining module 603 is configured to determine a marketing strategy suitable for the target user according to the purchase probability increment of the target user under different marketing strategies.
According to the embodiment of the invention, the pre-trained prediction model is utilized, the user characteristic information and the product characteristic information are used as input, the purchase probability increment of the target user under different marketing strategies is predicted, the specific quantitative result of the marketing strategies on the purchase probability improvement is determined, and the corresponding marketing strategies are determined according to the quantitative result, so that the more accurate marketing strategies are determined.
Optionally, the prediction model adopts an incremental model, and the detection module includes: the copying unit is used for copying the user characteristic information and the product characteristic information into a plurality of groups of input data; the input unit is used for inputting one of the multiple sets of input data into a control group, and inputting the other sets of input data into a plurality of processing groups respectively, wherein the control group is an input data set which does not adopt marketing strategies, the processing groups are input data sets which adopt marketing strategies, and each processing group corresponds to one marketing strategy; the first prediction unit is used for predicting the purchase probability of the input data of the control group to obtain a first purchase probability without adopting a marketing strategy; the second prediction unit is used for predicting the purchase probability of the input data of each processing group to obtain a plurality of second purchase probabilities under different marketing strategies; and the calculating unit is used for obtaining the purchasing probability increment by utilizing the difference between the second purchasing probability and the first purchasing probability.
Optionally, the determining module is specifically configured to determine the marketing strategy with the largest purchase probability increment as the marketing strategy of the target user.
Optionally, the determining module includes: a first obtaining unit, configured to obtain a user type of the target user, where the user type is determined according to an activity level of the user; the second acquisition unit is used for acquiring preset marketing operation cost; and the determining unit is used for determining the marketing strategy of the target user according to the user type, the marketing operation cost and the purchase probability increment.
Optionally, the first obtaining unit includes: the acquisition subunit is used for acquiring historical behavior characteristic data of the target user; a dividing subunit, configured to divide the target user into corresponding user types according to the historical behavior feature data, where the user types include: active users, silent users, and attrition users.
Optionally, the determining unit includes: the first determining subunit is used for determining a target relation table according to the user type, wherein the target relation table records a corresponding relation between a probability increment threshold and marketing operation cost; a second determining subunit, configured to determine a probability increment threshold of the target user by using the target relation table and the marketing operation cost; the judging subunit is used for judging whether the purchase probability increment of the target user under different marketing strategies is smaller than the probability increment threshold value; and the selecting subunit is used for selecting the marketing strategy with the minimum cost when the probability increment is smaller than the probability increment threshold.
Optionally, the apparatus for determining a marketing strategy further includes a training module, configured to train the prediction model by: dividing an experimental target population into a plurality of treatment groups and a control group, wherein each treatment group adopts a marketing strategy, and the control group does not adopt the marketing strategy; recording a conversion result, recording the users who successfully buy as positive samples, and recording the users who do not buy as negative samples; acquiring user characteristic information and product characteristic information of all users; and training to obtain the prediction model by using the conversion result, the user characteristic information and the product characteristic information as training samples.
In an embodiment of the present invention, a computer device is also provided, an internal structure diagram of which may be as shown in fig. 7. The computer device comprises a processor and a memory which are connected through a system bus, and also comprises a display screen and an input device. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The computer program is executed by a processor to realize the fire detection method of the battery charging cabinet, the computer equipment also comprises a display screen and an input device, the display screen can be a liquid crystal display screen or an electronic ink display screen, the input device of the computer equipment can be a touch layer covered on the display screen, also can be a key, a track ball or a touch pad arranged on the shell of the computer equipment, and also can be an external keyboard, a touch pad and the like.
Those skilled in the art will appreciate that the architecture shown in fig. 7 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the prior art according to the concept of the present invention should be within the scope of protection defined by the claims.
Claims (10)
1. A method for determining a marketing strategy, comprising:
acquiring user characteristic information of a target user and characteristic information of a product to be marketed;
inputting the user characteristic information and the product characteristic information into a pre-trained prediction model, and outputting purchase probability increment of the target user under different marketing strategies, wherein the purchase probability increment is the increment of the purchase probability of the target user after adopting the marketing strategy relative to the purchase probability without adopting the marketing strategy;
and determining the marketing strategy suitable for the target user according to the purchase probability increment of the target user under different marketing strategies.
2. The marketing strategy determination method of claim 1, wherein the prediction model adopts an incremental model, the inputting the user characteristic information and the product characteristic information into the pre-trained prediction model and outputting the purchase probability increment of the target user under different marketing strategies comprises:
copying the user characteristic information and the product characteristic information into a plurality of groups of input data;
inputting one of the multiple sets of input data into a control group, and respectively inputting other sets of input data into multiple processing groups, wherein the control group is an input data set which does not adopt marketing strategies, the processing groups are input data sets which adopt marketing strategies, and each processing group corresponds to one marketing strategy;
carrying out purchase probability prediction on the input data of the control group to obtain a first purchase probability which does not adopt a marketing strategy;
predicting purchase probability of the input data of each processing group to obtain a plurality of second purchase probabilities under different marketing strategies;
and obtaining the purchasing probability increment by utilizing the difference between the second purchasing probability and the first purchasing probability.
3. The method for determining marketing strategies according to claim 1, wherein the determining marketing strategies suitable for the target users according to the purchase probability increment of the target users under different marketing strategies comprises:
and determining the marketing strategy with the largest purchase probability increment as the marketing strategy of the target user.
4. The method for determining marketing strategies according to claim 1, wherein the determining marketing strategies suitable for the target users according to the purchase probability increment of the target users under different marketing strategies comprises:
acquiring a user type of the target user, wherein the user type is determined according to the activity degree of the user;
acquiring preset marketing operation cost;
and determining the marketing strategy of the target user according to the user type, the marketing operation cost and the purchase probability increment.
5. The marketing strategy determination method of claim 4, wherein the obtaining of the user type of the target user comprises:
acquiring historical behavior characteristic data of the target user;
dividing the target users into corresponding user types according to the historical behavior characteristic data, wherein the user types comprise: active users, silent users, and attrition users.
6. The method for determining marketing strategy of claim 4, wherein the determining marketing strategy of the target user according to the user type, marketing operation cost and the purchase probability increment comprises:
determining a target relation table according to the user type, wherein the target relation table records the corresponding relation between the probability increment threshold and the marketing operation cost;
determining a probability increment threshold of the target user by using the target relation table and the marketing operation cost;
judging whether the purchase probability increment of the target user under different marketing strategies is smaller than the probability increment threshold value;
and when the probability increment is smaller than the probability increment threshold, selecting the marketing strategy with the minimum cost.
7. The marketing strategy of claim 1, wherein the predictive model is trained by:
dividing an experimental target population into a plurality of treatment groups and a control group, wherein each treatment group adopts a marketing strategy, and the control group does not adopt the marketing strategy;
recording a conversion result, recording the users who successfully buy as positive samples, and recording the users who do not buy as negative samples;
acquiring user characteristic information and product characteristic information of all users;
and training to obtain the prediction model by using the conversion result, the user characteristic information and the product characteristic information as training samples.
8. An apparatus for determining a marketing strategy, comprising:
the acquisition module is used for acquiring the user characteristic information of a target user and the characteristic information of a product to be marketed;
the detection module is used for inputting the user characteristic information and the product characteristic information into a pre-trained prediction model and outputting purchase probability increment of the target user under different marketing strategies, wherein the purchase probability increment is the increment of the purchase probability of the target user after the marketing strategy is adopted relative to the purchase probability of the target user without the marketing strategy;
and the determining module is used for determining the marketing strategy suitable for the target user according to the purchase probability increment of the target user under different marketing strategies.
9. A computer device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to perform the method of determining a marketing strategy of any of claims 1-7.
10. A computer-readable storage medium storing computer instructions for causing a computer to perform the method of determining a marketing strategy of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210149807.6A CN114549071A (en) | 2022-02-18 | 2022-02-18 | Marketing strategy determination method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210149807.6A CN114549071A (en) | 2022-02-18 | 2022-02-18 | Marketing strategy determination method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114549071A true CN114549071A (en) | 2022-05-27 |
Family
ID=81676299
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210149807.6A Pending CN114549071A (en) | 2022-02-18 | 2022-02-18 | Marketing strategy determination method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114549071A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116304374A (en) * | 2023-05-19 | 2023-06-23 | 云印技术(深圳)有限公司 | Customer matching method and system based on package data |
-
2022
- 2022-02-18 CN CN202210149807.6A patent/CN114549071A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116304374A (en) * | 2023-05-19 | 2023-06-23 | 云印技术(深圳)有限公司 | Customer matching method and system based on package data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109993652B (en) | Loan credit risk assessment method and device | |
| Keane et al. | Exploring the usefulness of a nonrandom holdout sample for model validation: Welfare effects on female behavior | |
| CN111340244B (en) | Prediction method, training method, device, server and medium | |
| Hsieh et al. | Incorporating multiple uncertainties into projections of Chinese private car sales and stock | |
| CN110706054A (en) | Scheme pushing method and device, computer device and readable storage medium | |
| CN114549071A (en) | Marketing strategy determination method and device, computer equipment and storage medium | |
| CN113407854A (en) | Application recommendation method, device and equipment and computer readable storage medium | |
| CN113379532A (en) | Credit consciousness level prediction method, device, equipment and storage medium | |
| Chakroborty et al. | A psychophysical ordered response model of time perception and service quality: Application to level of service analysis at toll plazas | |
| CN110796379A (en) | Risk assessment method, device and equipment of business channel and storage medium | |
| CN116503092A (en) | User reservation intention recognition method and device, electronic equipment and storage medium | |
| US20220366437A1 (en) | Method and system for deep reinforcement learning and application at ride-hailing platform | |
| CN112819507A (en) | Service pushing method and device, electronic equipment and readable storage medium | |
| CN114169906B (en) | Electronic coupon pushing method and device | |
| CN112101576B (en) | Distributed device usage machine learning model aggregation system | |
| Xu et al. | Effect of Data Imbalance on the Performance of Pavement Deterioration Models | |
| CN113393330B (en) | Financial wind control management system based on block chain | |
| CN116485457A (en) | Coupon issuing method and device, computer device and readable storage medium | |
| CN113256034B (en) | Information delivery method and device, electronic equipment and storage medium | |
| CN114819908A (en) | Method and device for determining material purchasing negotiation time, computer equipment and storage medium | |
| CN116668527A (en) | Content pushing processing method and device, electronic equipment and storage medium | |
| Connor et al. | Predicting the outcomes of policy diffusion from US states to federal law | |
| Mukherjee | Boosting renewable energy technology uptake in Ireland: A machine learning approach | |
| Alghanmi et al. | Population Fusion Transformer for Subnational Population Forecasting | |
| CN117036053A (en) | Training method of renewal behavior prediction model, renewal behavior prediction method and product |
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 |