CN113034171B - Business data processing method and device, computer and readable storage medium - Google Patents

Business data processing method and device, computer and readable storage medium Download PDF

Info

Publication number
CN113034171B
CN113034171B CN202110076600.6A CN202110076600A CN113034171B CN 113034171 B CN113034171 B CN 113034171B CN 202110076600 A CN202110076600 A CN 202110076600A CN 113034171 B CN113034171 B CN 113034171B
Authority
CN
China
Prior art keywords
test
service
strategy
user cluster
resource consumption
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.)
Active
Application number
CN202110076600.6A
Other languages
Chinese (zh)
Other versions
CN113034171A (en
Inventor
李池洋
朱志华
蔡政
林晓健
任宇堃
李成龙
蔡越
邓颖
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tencent Technology Shenzhen Co Ltd
Original Assignee
Tencent Technology Shenzhen Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Tencent Technology Shenzhen Co Ltd filed Critical Tencent Technology Shenzhen Co Ltd
Priority to CN202110076600.6A priority Critical patent/CN113034171B/en
Publication of CN113034171A publication Critical patent/CN113034171A/en
Application granted granted Critical
Publication of CN113034171B publication Critical patent/CN113034171B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G06Q30/0241—Advertisements
    • G06Q30/0242—Determining effectiveness of advertisements
    • 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
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
    • 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)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Accounting & Taxation (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Theoretical Computer Science (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Data Mining & Analysis (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The embodiment of the application discloses a business data processing method, a business data processing device, a computer and a readable storage medium, which relate to the field of block chain technology and artificial intelligence, and the method comprises the following steps: under a first prediction condition, predicting first prediction total resource consumption of the first user cluster aiming at the test service object based on the first service strategy; under a second prediction condition, predicting second predicted total resource consumption of a second user cluster aiming at the test service object based on a second service strategy; determining a consumption difference index of the test business object according to the first prediction total resource consumption and the second prediction total resource consumption, acquiring a transaction data volume index of the test business object, and determining a release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data volume index. By the method and the device, the accuracy of strategy decision of the test service object can be improved.

Description

Business data processing method and device, computer and readable storage medium
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for processing service data, a computer, and a readable storage medium.
Background
As the advertising system evolves from Cost Per Click (CPC) advertising to current optimized cost per Action (ocap) advertising, the experimental platform of the advertising generally uses the consumption index as an index for measuring the effect of the a/B experiment to determine the update decision of the business strategy of the advertising. The A/B experiment comprises an experiment group and a control group, consumption refers to income brought to a platform (namely a popularization party) by advertisement putting, and the scientificity of the experiment effect can be influenced by measuring the A/B experiment effect through the consumption index. On one hand, if the experimental group overestimates the estimated Click Rate (pCTR) or the estimated conversion Rate (pCVR), the consumption index is increased, because the actual conversion is not increased, the advertisement cost is increased, for the cpap advertisement, the conversion cost is adjusted downward by the price adjustment system, and the downward conversion cost acts on all the traffic of the whole advertisement, that is, the consumption of the experimental group and the control group is decreased at the same time, so that although the consumption index is increased compared with that of the control group, the consumption index of the experimental group cannot be accurately shown to be better than that of the control group. On the other hand, if the experimental group only increases the advertisement queue length to cause the increase of the second-order fee deduction, thereby causing the increase of consumption, the conversion cost of the advertisement is increased in this way, and the price adjusting system can adjust the conversion cost down to offset the increase of consumption to a certain extent, so that the consumption index cannot accurately determine the quality of the business strategy of the experimental group and the quality of the business strategy of the comparison group. Alternatively, if the experimental group only increases the idle consumption, so that the consumption of the advertisement increases, this may result in a decrease in the revenue for the advertiser (i.e., the owner of the advertisement). Therefore, determining whether to release the business strategy of the experimental group through the consumption index may result in lower accuracy and scientificity of the strategy decision for testing the business object.
Disclosure of Invention
The embodiment of the application provides a business data processing method, a business data processing device, a computer and a readable storage medium, which can improve the accuracy and the scientificity of strategy decision of a test business object.
One aspect of the embodiments of the present application provides a method for processing service data, where the method includes:
under a first prediction condition, predicting first prediction total resource consumption generated by a first user cluster aiming at a test service object based on a first service strategy; the first prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the first service strategy;
under a second prediction condition, predicting second predicted total resource consumption generated by the second user cluster aiming at the test service object based on a second service strategy; the second prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on a second service strategy;
determining a consumption difference index of the test business object according to the first prediction total resource consumption and the second prediction total resource consumption, acquiring a transaction data volume index of the test business object, and determining a release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data volume index; and the volume releasing decision result is used for indicating the business strategy for performing the volume releasing processing between the first business strategy and the second business strategy and the volume releasing mode of the business strategy for performing the volume releasing processing.
In one aspect, an embodiment of the present application provides a service data processing apparatus, where the apparatus includes:
the first consumption obtaining module is used for predicting first predicted total resource consumption of the first user cluster aiming at the test service object based on the first service strategy under a first prediction condition; the first prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the first service strategy;
the second consumption obtaining module is used for predicting second predicted total resource consumption of the second user cluster aiming at the test service object based on the second service strategy under a second prediction condition; the second prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on a second service strategy;
the index generation module is used for determining a consumption difference index of the test business object according to the first prediction total resource consumption and the second prediction total resource consumption, acquiring a transaction data volume index of the test business object, and determining a release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data volume index; and the volume releasing decision result is used for indicating the business strategy for performing the volume releasing processing between the first business strategy and the second business strategy and the volume releasing mode of the business strategy for performing the volume releasing processing.
Wherein, the device still includes:
the object delivery module is used for delivering the test service object to the first user cluster based on the first service strategy and delivering the test service object to the second user cluster based on the second service strategy under the actual test condition;
the first data acquisition module is used for acquiring first actual total resource consumption and first actual conversion data volume generated by the first user cluster aiming at the test service object based on the first service strategy;
the second data acquisition module is used for acquiring a second actual conversion data volume generated by the second user cluster aiming at the test service object based on a second service strategy;
the first consumption acquisition module includes:
a first consumption determining unit, configured to, under a first prediction condition, determine, based on a first business policy, a first actual total resource consumption as a first predicted total resource consumption generated by the first user cluster for the test business object if the test business object belongs to a non-correction type object;
and the second consumption determining unit is used for determining first predicted total resource consumption of the first user cluster aiming at the test service object under the first service strategy according to the first actual conversion data volume, the second actual conversion data volume and the first actual total resource consumption if the test service object belongs to the correction type object.
The number of the users in the first user cluster is the first user number, and the number of the users in the second user cluster is the second user number;
the second consumption determination unit includes:
the first weighting subunit is used for weighting the first actual conversion data volume and the second actual conversion data volume based on the first user number and the second user number to obtain a basic conversion data volume;
the first adjusting subunit is used for determining a first data volume adjusting parameter according to the basic conversion data volume and the first actual conversion data volume;
and the first prediction subunit is configured to perform adjustment processing on the first actual total resource consumption based on the first data amount adjustment parameter, so as to obtain first predicted total resource consumption, which is generated by the first user cluster for the test service object under the first service policy.
Wherein, the device still includes:
the object delivery module is further configured to deliver the test service object to the first user cluster based on the first service policy and to deliver the test service object to the second user cluster based on the second service policy under the actual test condition;
the first data acquisition module is further used for acquiring a first actual conversion data volume generated by the first user cluster aiming at the test service object based on the first service strategy;
the second data obtaining module is further configured to obtain, based on a second service policy, a second actual conversion data amount and a second actual total resource consumption, which are generated by the second user cluster for the test service object;
the second consumption acquisition module includes:
a third consumption determining unit, configured to determine, under a second prediction condition and based on a second service policy, a second actual total resource consumption as a second predicted total resource consumption generated by the second user cluster for the test service object if the test service object belongs to the non-correction type object;
and the fourth consumption determining unit is used for determining second predicted total resource consumption of the second user cluster aiming at the test service object under the second service strategy according to the first actual conversion data volume, the second actual conversion data volume and the second actual total resource consumption if the test service object belongs to the correction type object.
The number of users in the first user cluster is the first user number, and the number of users in the second user cluster is the second user number;
the fourth consumption determining unit includes:
the second weighting subunit is used for weighting the first actual conversion data volume and the second actual conversion data volume based on the first user number and the second user number to obtain a basic conversion data volume;
the second adjustment subunit is used for determining a second data volume adjustment parameter according to the basic conversion data volume and the second actual conversion data volume;
and the second prediction subunit is configured to perform adjustment processing on the second actual total resource consumption based on the second data amount adjustment parameter, so as to obtain second predicted total resource consumption, which is generated by the second user cluster for the test service object under the second service policy.
The number of the users in the first user cluster is the first user number, and the number of the users in the second user cluster is the second user number;
in determining a consumption difference index of the test business object according to the first predicted total resource consumption and the second predicted total resource consumption, the index generating module includes:
a unit consumption acquiring unit, configured to determine a ratio of the first predicted total resource consumption to the first user amount as a first predicted unit resource consumption, and determine a ratio of the second predicted total resource consumption to the second user amount as a second predicted unit resource consumption;
and the consumption index generating unit is used for acquiring a total difference value between the first prediction unit resource consumption and the second prediction unit resource consumption, acquiring a flow ratio between the second user quantity and the first user quantity, and determining the consumption difference index of the test business object based on the total difference value and the flow ratio.
In obtaining a transaction data volume index of a test service object, the index generation module includes:
the conversion counting unit is used for counting the total conversion number of the first user cluster and the second user cluster aiming at the test service object; the total conversion number is used for representing the number of conversions in the interaction behaviors of the first user cluster and the second user cluster aiming at the test service object together;
the transaction index generation unit is used for acquiring a target conversion data volume and determining a transaction data volume index of the test service object based on the total conversion number and the target conversion data volume; the target conversion data volume refers to the conversion data volume that all parties of the test business object expect to achieve for the test business object.
Wherein, in determining the release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data volume index, the index generating module comprises:
a standard acquisition unit for acquiring a release amount processing standard; the release processing standard comprises at least two index ranges and a release decision corresponding to each index range;
and the decision determining unit is used for searching a target index range to which the consumption difference index and the transaction data volume index belong in the release processing standard, and determining a release decision corresponding to the target index range as a release decision result of the test business object between the first business strategy and the second business strategy.
Wherein, in determining the release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data volume index, the index generating module comprises:
a difference obtaining unit, configured to obtain a first predicted resource difference value between the first predicted total resource consumption and the transaction data amount index, and obtain a second predicted resource difference value between the second predicted total resource consumption and the transaction data amount index;
a first result determining unit, configured to determine that a release decision result of the test service object between the first service policy and the second service policy is, if the second predicted resource difference value is less than or equal to the first predicted resource difference value and the consumption difference index and the transaction data amount index satisfy the policy release condition, a release result for the second service policy;
and the second result determining unit is used for determining that the release decision result of the test service object between the first service strategy and the second service strategy is a release canceling result aiming at the second service strategy if the second predicted resource difference value is greater than the first predicted resource difference value or the consumption difference index and the transaction data volume index do not meet the strategy release condition.
Wherein, the device still includes:
the strategy updating module is used for acquiring a local user from the first user cluster based on the release decision result if the release decision result is the determined release result aiming at the second service strategy, and updating the test service object released to the local user from the first service strategy to the second service strategy;
the policy updating module is further configured to update the test service object delivered to the second user cluster from the second service policy to the first service policy if the release decision result is a release cancellation result for the second service policy.
In terms of obtaining local users from the first user cluster based on the release decision result, the policy updating module includes:
the proportion determining unit is used for acquiring the sum of the first user quantity and the second user quantity to obtain the total number of the users and determining the proportion of the second user quantity in the total number of the users;
the quantity obtaining unit is used for obtaining a first release level to which the user proportion belongs, determining a second release level according to the first release level, and determining the decision quantity according to the second release level and the total number of the users;
the user determining unit is used for determining the quantity of the to-be-updated users from the first user cluster according to the quantity of the to-be-updated users and determining the M to-be-updated users as local users; m is the amount to be put.
Wherein, the device still includes:
the contract calling module is used for calling the intelligent contract based on the release decision result and executing a release processing function corresponding to the release decision result based on the intelligent contract;
and the volume processing module is used for executing a volume processing process corresponding to the volume decision result on the first service strategy and the second service strategy through the volume processing function.
In one aspect, an embodiment of the present application provides a computer device, including a processor, a memory, and an input/output interface;
the processor is connected to the memory and the input/output interface, respectively, where the input/output interface is configured to receive data and output data, the memory is configured to store a computer program, and the processor is configured to call the computer program, so that a computer device including the processor executes the service data processing method in an aspect of the embodiment of the present application.
In one aspect, an embodiment of the present application provides a computer-readable storage medium, where a computer program is stored, where the computer program is suitable for being loaded and executed by a processor, so that a computer device having the processor executes a service data processing method in one aspect of the embodiment of the present application.
An aspect of an embodiment of the present application provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. The computer instructions are read by a processor of a computer device from a computer-readable storage medium, and the computer instructions are executed by the processor to cause the computer device to perform the method provided in the various alternatives in one aspect of the embodiments of the application.
The embodiment of the application has the following beneficial effects:
in this embodiment of the application, the computer device may predict, based on the first business policy, a first predicted total resource consumption, generated by the first user cluster for the test business object, under a first prediction condition; the first prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the first service strategy; under a second prediction condition, predicting second predicted total resource consumption of a second user cluster aiming at the test service object based on a second service strategy; the second prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on a second service strategy; determining a consumption difference index of the test business object according to the first prediction total resource consumption and the second prediction total resource consumption, acquiring a transaction data volume index of the test business object, and determining a release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data volume index. When a test business object is put in by a second business strategy instead of the first business strategy, the consumption difference index represents the income which can be obtained by a popularization party putting in the test business object when the business strategy of the test business object is updated, the transaction data volume index represents the income which can be obtained by all parties of the test business object when the business strategy of the test business object is updated, and the put decision result of the test business object between the first business strategy and the second business strategy is jointly carried out according to the consumption difference index and the transaction data volume index, so that the income between the popularization party and all parties of the test business object can be considered in multiple ways, the income balance between the popularization party and all parties is improved, different business strategies can be better evaluated, and the accuracy and the scientificity of strategy decision of the test business object are improved. The business strategy determined by the scheme of the application has the advantages of low consumption, good promotion effect on users and the like, is favorable for saving the cost of a promotion party, saving system promotion resources and the like, and is also favorable for improving the income brought to all parties by promotion and test business objects.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a network interaction architecture diagram of service data processing provided in an embodiment of the present application;
FIG. 2 is a schematic diagram illustrating a policy decision scenario for testing a business object according to an embodiment of the present application;
fig. 3 is a flowchart of a method for processing service data according to an embodiment of the present application;
FIG. 4 is a schematic diagram of an advertisement pricing scenario provided by an embodiment of the present application;
fig. 5 is a schematic diagram of a specific flow of service data processing provided in an embodiment of the present application;
FIG. 6 is a schematic diagram of a policy prediction scenario for cost-per-click advertising provided by an embodiment of the present application;
FIG. 7a is a schematic diagram of a policy prediction scenario for optimizing behavioral bidding advertisements according to an embodiment of the present application;
FIG. 7b is a schematic diagram of a pricing process scenario for a calibration-type object according to an embodiment of the present application;
fig. 8 is a schematic diagram of a service data processing apparatus according to an embodiment of the present application;
fig. 9 is a schematic structural diagram of a computer device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The method and the device for processing the business data can use a block chain technology to store and process the data generated in the embodiment, can also trigger the acquisition and processing of the data associated with the test business object and the generation process of the index based on an artificial intelligence technology, and can perform the decision on the first business strategy and the second business strategy by processing the data generated by the test business object under the first business strategy and the second business strategy, wherein the process may involve a large amount of data to ensure the universality and the credibility of the result.
Among them, artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result. In other words, artificial intelligence is a comprehensive technique of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence. Artificial intelligence is the research of the design principle and the implementation method of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making.
The artificial intelligence technology is a comprehensive subject and relates to the field of extensive technology, namely the technology of a hardware level and the technology of a software level. The artificial intelligence base technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.
The Big data (Big data) refers to a data set which cannot be captured, managed and processed by a conventional software tool within a certain time range, and is a massive, high-growth rate and diversified information asset which can have stronger decision power, insight discovery capability and process optimization capability only by a new processing mode. With the advent of the cloud era, big data has attracted more and more attention, and the big data needs special technology to effectively process a large amount of data within a tolerance elapsed time. The method is suitable for technologies of big data, including a large-scale parallel processing database, data mining, a distributed file system, a distributed database, a cloud computing platform, the Internet and an extensible storage system.
In the embodiment of the present application, please refer to fig. 1, where fig. 1 is a network interaction architecture diagram for service data processing provided in the embodiment of the present application. When data processing is performed on a test service object to determine a release decision result of a service policy (i.e., a second service policy) to be updated when the service policy of the test service object is updated, or when a service policy associated with the test service object is updated from a first service policy to a second service policy, the embodiment of the present application may include three conditions, such as an actual test condition, a first prediction condition, and a second prediction condition, where the actual test condition may be considered as a test condition corresponding to a current state, that is, the test service object is delivered to a first user cluster based on the first service policy, and the test service object is delivered to a second user cluster based on the second service policy; the first prediction condition can be regarded as a prediction condition corresponding to the no-policy state, namely, the test service object is delivered to the first user cluster and the second user cluster based on the first service policy; the second prediction condition may be considered as a prediction condition corresponding to the full policy state, that is, the test service object is delivered to the first user cluster and the second user cluster based on the second service policy. The computer device may divide the user cluster into an experimental group and a comparison group, the comparison group is marked as a first user cluster, the experimental group is marked as a second user cluster, the comparison group is used for indicating a group which does not receive an updated service policy (i.e., a second service policy), the experimental group is used for indicating a group which receives the updated service policy, and data generated by the experimental group and the comparison group for different service policies are compared to obtain a release decision result of the test service object between the first service policy and the second service policy.
As shown in fig. 1, the computer device 101 may deliver a test service object to users, and the users may be divided into a first user cluster 102 and a second user cluster 103, where the first user cluster 102 is regarded as a control group, the second user cluster 103 is regarded as an experiment group, the computer device 101 may deliver the test service object to a first user included in the first user cluster 102 based on a first service policy, such as the first user 102a, and deliver the test service object to a second user included in the second user cluster 103 based on a second service policy, such as the second user 102b and the second user 102c, and use of the service policy is regarded as an actual test condition, under which, data under the first prediction condition and data under the second prediction condition are predicted, for example, the computer device 101 may predict, under the first prediction condition, a first predicted total resource consumption of the first user cluster for the test service object based on the first service policy; a second predicted total resource consumption for the test business object by the second cluster of users may be predicted based on the second business policy under a second prediction condition. And determining the advantages and disadvantages of the first business strategy and the second business strategy according to the data generated under the first prediction condition and the data generated under the second prediction condition so as to make a decision on the business strategy of the test business object.
Specifically, please refer to fig. 2, where fig. 2 is a schematic diagram of a policy decision scenario for testing a service object according to an embodiment of the present application. As shown in fig. 2, the a/B test is performed on the test service object, and the resource consumption of the test service object under the two service policies is predicted according to the test result, so that the release decision is performed on the two service policies. Specifically, as shown in fig. 2, under the actual test condition 202, the user cluster 202 is divided into a first user cluster 2021 and a second user cluster 2022, the computer device 201 delivers the test service object to the first user cluster 2021 based on a first service policy, and delivers the test service object to the second user cluster 2022 based on a second service policy, where the first service policy is a service policy that needs to be replaced, and the second service policy is a service policy that needs to replace the first service policy. On the basis of the actual test condition 202, it is assumed that the computer device 201 launches the test service object to the first user cluster 2021 and the second user cluster 2022 respectively based on the first service policy, the launch condition is written as a first prediction condition 203, and data generated by the test service object under the first prediction condition 203, such as first predicted total resource consumption, is predicted according to data generated by the test service object under the actual test condition 202; on the basis of the actual test condition 202, it is assumed that the computer device 202 puts the test service object into the first user cluster 2021 and the second user cluster 2022 respectively based on the second service policy, the putting condition is referred to as a second prediction condition 204, and data generated by the test service object under the second prediction condition 204, such as second prediction total resource consumption, is predicted according to data generated by the test service object under the actual test condition 202.
The computer device 201 may determine the consumption difference index of the test service object according to the first predicted total resource consumption under the first prediction condition 203 and the second predicted total resource consumption under the second prediction condition 204, so that the consumption difference index may accurately and truly reflect the experimental effect of the a/B test, and quantify the change of the first service policy and the second service policy in consumption, where the consumption difference index may represent the degree of influence on the popularizing party of the test service object. Further, the computer device may obtain a transaction data volume index of the test business object, where the transaction data volume index is used to represent that, on the basis of the conversion data volume expected by all parties of the test business object, the change situations of the first business strategy and the second business strategy on the expected transaction total amount are quantized to represent the degree of influence on all parties of the test business object. The first business strategy and the second business strategy are subjected to release decision based on the consumption difference index and the transaction data volume, the benefits between the popularizing party and all parties of the test business object can be considered in multiple ways, the benefit balance between the popularizing party and all parties is improved, different business strategies can be better evaluated, and therefore the accuracy and the scientificity of the strategy decision of the test business object are improved.
It is understood that the computer device in the embodiment of the present application includes, but is not limited to, a terminal device or a server. In other words, the computer device may be a server or a terminal device, or may be a system of a server and a terminal device. The above-mentioned terminal device may be an electronic device, including but not limited to a mobile phone, a tablet computer, a desktop computer, a notebook computer, a palm-top computer, an Augmented Reality/Virtual Reality (AR/VR) device, a helmet-mounted display, a wearable device, a smart speaker, a digital camera, a camera, and other Mobile Internet Devices (MID) with network access capability. The above-mentioned server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a Network service, cloud communication, a middleware service, a domain name service, a security service, a Content Delivery Network (CDN), a big data and artificial intelligence platform, and the like.
Optionally, the data related in the embodiment of the present application may be stored in a computer device, or the data may be stored based on a cloud storage technology, which is not limited herein.
Where partial nouns referred to in this application can be explained as follows:
A/B test: the method comprises the steps of dividing an experimental sample into two mutually exclusive user groups of an experimental group and a comparison group through random sampling, marking the comparison group as a first user cluster, marking the experimental group as a second user cluster, keeping the original strategy (namely the first service strategy) unchanged by the users of the comparison group, using a new strategy by the users of the experimental group, and finally comparing the difference of the two groups of users to obtain the effect of the service strategy.
Experimental groups: and the group receiving the service policy processing in the A/B test is the second user cluster.
Control group: and the group which does not receive the service strategy processing in the A/B test is the first user cluster.
Consumption: which refers to the income brought by the advertising to the popularizing parties.
And (3) transformation: after an owner of an advertisement puts the advertisement, it is expected that a user performs user behavior based on the put advertisement, such as downloading an application (app), activating the app, placing an order, purchasing or traveling, and the like, wherein in different scenarios, a conversion refers to different user behaviors.
Conversion (conversion Rate, CVR): represents the average number of conversions per click, i.e., the number of conversions divided by the number of clicks.
Click-Through Rate (CTR): the number of unit clicks which occur averagely per exposure is represented, namely the number of clicks is divided by the number of exposures, wherein the exposure means that after the advertisement is delivered, users can see the advertisement, and one user can see the advertisement, namely one exposure.
Estimated transformation Rate (pCVR): an estimate of conversion rate by the advertiser.
Estimated click-Through Rate (pCTR): an estimate of click-through rate by the advertiser.
Cost of transformation (Cost Per Action, CPA): the cost, i.e., consumption divided by the number of conversions, required for an ad to acquire each conversion.
Target cost Per Action (targetCPA): the conversion cost expected by all parties to the advertisement, i.e., the cost expected by all parties to the advertisement to obtain a conversion by placing the advertisement.
oCPA advertisement: the advertisement is bid in the optimization behavior, all parties of the advertisement can preset a target conversion cost, and the popularizing party of the advertisement is automatically optimized based on the putting target and the bidding effect, so that the efficiency and the input-output ratio of the advertisement of all parties of the advertisement are continuously improved.
Adjusting the price factor: for the oCPA advertisement, the price adjusting system adjusts the advertisement bid through the price adjusting factor according to the relationship between the current conversion cost and the target conversion cost of all parties of the advertisement, so that the final actual cost of the advertisement is close to the target conversion cost as much as possible.
Second-order deduction: the advertisement system is a fee deduction rule for deducting the fee of the advertisement which is ranked the second in the advertisement queue at present.
And (3) idle consumption: refers to the consumption resulting from an advertisement for which no conversion has occurred.
Further, please refer to fig. 3, where fig. 3 is a flowchart of a method for processing service data according to an embodiment of the present application. As shown in fig. 3, when the service policy used when the test service object is released is updated from the first service policy to the second service policy, object benefits of the test service object generated in the first service policy and the second service policy respectively need to be tested, and if the second object benefit corresponding to the second service policy is greater than the first object benefit corresponding to the first service policy, it indicates that the second service policy has a better effect than the first service policy, and the second service policy of the test service object can be released; if the second object benefit corresponding to the second service policy is less than or equal to the first object benefit corresponding to the first service policy, it means that the service policy of the test service object is updated from the first service policy to the second service policy, and the object benefit generated by the test service object cannot be improved, so the second service policy of the test service object is not subjected to release processing.
In order to make a better decision on the first service policy and the second service policy, an index may be generated according to data generated by the test service object under the first service policy and the second service policy, and a release decision may be made on the first service policy and the second service policy based on the index. The method and the device for testing the user cluster have three conditions, wherein one condition is an actual test condition for actually testing the first user cluster and the second user cluster, the other condition is a first prediction condition for predicting based on the actual test condition, and the other condition is a second prediction condition for predicting based on the actual test condition. The first prediction condition is equivalent to that under the actual test condition, the test service objects acquired by the first user cluster and the second user cluster are realized based on a first service policy, and the first prediction condition can be considered as a policy-free state; the second prediction condition is equivalent to that under the actual test condition, the test service objects acquired by the first user cluster and the second user cluster are realized based on the second service policy, and the second prediction condition can be considered as a full policy state.
In other words, in the embodiment of the method depicted in fig. 3, the service data processing procedure includes the following steps:
step S301, under a first prediction condition, predicting first prediction total resource consumption generated by the first user cluster aiming at the test service object based on the first service strategy.
In the embodiment of the application, the computer device can acquire the service users associated with the test service object, divide the service users into two groups, and respectively use the two groups as a first user cluster and a second user cluster; optionally, the computer device may obtain the number of experiments, select the experimental users from the service users based on the number of experiments, and divide the experimental users into two groups, which are respectively used as the first user cluster and the second user cluster; optionally, the computer device may select the first user cluster from the service users, and select the second user cluster from the service users. The computer device may obtain a traffic proportion, determine the first user cluster and the second user cluster based on the traffic proportion, where a ratio between the number of the second users included in the second user cluster and the number of the first users included in the first user cluster is the traffic proportion, and for example, the traffic proportion may be denoted as p: q, wherein q and p can be positive integers or decimal numbers, and when q and p are decimal numbers, the sum of q and p is 1. Further, the computer device may determine a first user cluster based on a first service policy, and determine a second user cluster based on a second service policy, for example, the first service policy includes first drop user information, such as age information, gender information, and address information of the first drop user, the second service policy includes second drop user information, such as age information, gender information, and address information of the second drop user, the computer device obtains service users under 35 years old from the service users, selects the second user cluster from the service users under 35 years old, and determines the service users except the second user cluster as the first user cluster, assuming that the age information of the first drop user in the first service policy is full age, the gender information of the first drop user is full gender, and the gender information of the second drop user in the second service policy is under 35 years old; or the computer equipment acquires service users under 35 years old from the service users, selects a second user cluster from the service users under 35 years old, and selects a first user cluster from the service users except the second user cluster based on the flow ratio.
The computer device may launch a test service object to a first user cluster based on a first service policy, launch the test service object to a second user cluster based on a second service policy, determine the launch mode as the current state of the test service object, and record a condition indicated by the launch mode as an actual test condition for the first service policy and the second service policy, where the actual test condition is used to represent a condition for actually launching and testing the first service policy and the second service policy, in other words, under the actual test condition, the first user cluster actually receives the test service object launched based on the first service policy, and the second user cluster actually receives the test service object launched based on the second service policy. Further, the computer device may monitor data generated by the test business object under actual test conditions. Specifically, under the actual test condition, the computer device may release the test service object to the first user cluster based on the first service policy, and release the test service object to the second user cluster based on the second service policy; based on a first service strategy, acquiring first actual total resource consumption and first actual conversion data volume generated by a first user cluster aiming at a test service object; and acquiring a second actual conversion data volume generated by the second user cluster aiming at the test service object based on the second service strategy. For example, assuming that the test service object is an advertisement, the advertisement may be delivered to the user, so that the user may perform a click operation or a view operation on the advertisement.
Further, under actual test conditions, the computer device presets a first service policy and a second service policy associated with the test service object, respectively launches the test service object to the first user cluster and the second user cluster based on the first service policy, records the preset state as a no-policy state, and records a condition indicated by the preset launching mode as a first prediction condition. Under a first prediction condition, if the test service object belongs to a non-correction type object, the computer device may determine, based on a first service policy, a first actual total resource consumption as a first predicted total resource consumption generated by the first user cluster for the test service object; and if the test service object belongs to the correction type object, determining first predicted total resource consumption of the first user cluster aiming at the test service object under the first service strategy according to the first actual conversion data volume, the second actual conversion data volume and the first actual total resource consumption. The correction type object refers to that when the business strategy for launching the test business object is updated, the computer device can adaptively adjust the data of the test business object, so that the data of the test business object can accord with the expected value of all parties of the test business object.
For example, it is assumed that the test service object is an cpa advertisement, the cpa advertisement is a correction type object, and it is assumed that a service policy associated with the cpa advertisement is changed from a first service policy to a second service policy, where the second service policy is to increase a conversion cost of the cpa advertisement on the basis of the first service policy, at this time, a process of adjusting a price of the cpa advertisement based on a price adjustment factor may be shown in fig. 4, where fig. 4 is a schematic view of an advertisement price adjustment scenario provided in an embodiment of the present application. As shown in fig. 4, assuming that the target conversion cost of the cpa advertisement is 40 yuan, the computer device puts the test service object into the experimental group (i.e., the second user group) and the control group (i.e., the first user group) respectively based on the first service policy, assuming that the traffic ratio between the experimental group and the control group is 1, in this case, the computer device acquires the data in the table 401, and obtains that the first experiment consumption of the experimental group for the test service object (i.e., the cpa advertisement) is 400 yuan, the first experiment conversion number is 10, and the first experiment conversion cost is 40 yuan; the consumption of a first control of a control group for a test service object is 400 yuan, the conversion number of the first control is 10, and the conversion cost of the first control is 40 yuan; the total first population consumption of the control group and the experimental group was 800 yen, the first population conversion number was 20, and the first population conversion cost was 40 yen, and it was found that the target conversion cost of the cpap was in the achievement state. When the business strategy associated with the oCPA advertisement is updated to a second business strategy from a first business strategy by the computer equipment, the conversion cost of the oCPA advertisement put into the experimental group is increased from 40 yuan to 60 yuan, the conversion cost of the oCPA advertisement put into the control group is not changed, the computer equipment acquires the data in the table 402, and the second experiment consumption of the experimental group for the test business object (namely the oCPA advertisement) is 600 yuan, the second experiment conversion number is 10, and the second experiment conversion cost is 60 yuan; the consumption of a second control of the control group for the test service object is 400 yuan, the conversion number of the second control is 10, and the conversion cost of the second control is 40 yuan; the second total consumption of the control group and the experimental group is 1000 yuan, the second total conversion number is 20, and the second total conversion cost is 50 yuan.
As can be seen from table 402, the average cost of the opca advertisement increases from 40 yuan to 50 yuan, the computer device performs a price adjustment process on the data in table 402 based on a price adjustment factor, that is, performs a price adjustment process on the average cost to a target conversion cost, and adjusts the second overall conversion cost 50 yuan to a third overall conversion cost 40 yuan based on a price adjustment factor, that is, performs an overall downward adjustment on the conversion cost of the opca, so that the average cost between the experimental group under the second business strategy and the control group under the first business strategy can meet the target conversion cost, the third experimental conversion cost of the experimental group for the test business object is adjusted to 48 yuan, and the third experimental consumption of the experimental group for the test business object is 480 yuan in case that the number of conversion of the third experiment is still 10; the computer equipment adjusts the price of the third contrast conversion cost of the experimental group for the test business object into 32 yuan, and under the condition that the number of the third contrast conversion is still 10, the consumption of the third contrast of the contrast group for the test business object is 320 yuan; in this case, the third total conversion cost for the test business object in the experimental group and the control group is 40 yuan, the third total conversion number is 20, and the third total consumption is 800 yuan, thereby obtaining the data shown in table 403. Through the above process, the pricing processing of the oCPA advertisement is realized, the table 403 is data after the pricing processing, and it can be known from the table 403 that after the business strategy is changed, the consumption of the experimental group is increased by 50% compared with the consumption of the control group, and the overall consumption is unchanged.
The number of users in the first user cluster is the first user number, and the number of users in the second user cluster is the second user number. When the computer device obtains the first predicted total resource consumption, the computer device may perform weighting processing on the first actual conversion data volume and the second actual conversion data volume based on the first user number and the second user number to obtain a basic conversion data volume; determining a first data volume adjusting parameter according to the basic conversion data volume and the first actual conversion data volume; and adjusting the first actual total resource consumption based on the first data volume adjustment parameter to obtain first predicted total resource consumption generated by the first user cluster for the test service object under the first service strategy. The computer device performs weighting processing on the first actual conversion data volume and the second actual conversion data volume based on the first user number and the second user number to obtain a basic conversion data volume, determines a total number of users based on the first user number and the second user number, determines a first user weight based on the first user number and the total number of users, determines a second user weight based on the second user number and the total number of users, and performs weighting processing on the first actual conversion data volume and the second actual conversion data volume based on the first user weight and the second user weight to obtain the basic conversion data volume. For example, the traffic ratio between the second number of users and the first number of users is p: q, determining the ratio of the first user quantity in the total user quantity as the first user weight, which can be recorded as q/(q + p); the ratio of the second user number to the total user number is determined as a second user weight, which may be denoted as p/(q + p). Further, the computer device may perform price adjustment processing on the data predicted to be generated under the first prediction condition based on the price adjustment factor, and determine a first data amount adjustment parameter according to the basic conversion data amount and the first actual conversion data amount, wherein the first data amount adjustment parameter is used for indicating the price adjustment degree of the data predicted to be generated under the first prediction condition by the computer device.
Step S302, under a second prediction condition, predicting second prediction total resource consumption of the second user cluster aiming at the test service object based on the second service strategy.
In the embodiment of the application, under an actual test condition, a computer device puts a test service object into a first user cluster based on a first service strategy, puts the test service object into a second user cluster based on a second service strategy, and can also obtain a first actual conversion data volume generated by the first user cluster aiming at the test service object based on the first service strategy; and acquiring a second actual conversion data volume and a second actual total resource consumption generated by the second user cluster aiming at the test service object based on the second service strategy.
Under a second prediction condition, if the test service object belongs to the non-correction type object, the computer device may determine, based on a second service policy, a second actual total resource consumption as a second predicted total resource consumption generated by the second user cluster for the test service object; and if the test service object belongs to the correction type object, determining second predicted total resource consumption of the second user cluster aiming at the test service object under the second service strategy according to the first actual conversion data volume, the second actual conversion data volume and the second actual total resource consumption.
The number of the users in the first user cluster is the first user number, and the number of the users in the second user cluster is the second user number. When the computer device obtains the second predicted total resource consumption, the second predicted total resource consumption generated by the second user cluster for the test service object under the second service policy can be determined according to the first actual conversion data volume, the second actual conversion data volume and the second actual total resource consumption. Specifically, the computer device may perform weighting processing on the first actual conversion data volume and the second actual conversion data volume based on the first user number and the second user number to obtain a basic conversion data volume; determining a second data volume adjusting parameter according to the basic conversion data volume and the second actual conversion data volume; and adjusting the second actual total resource consumption based on the second data volume adjustment parameter to obtain a second predicted total resource consumption of the second user cluster aiming at the test service object under the second service strategy. The computer device can conduct price adjustment processing on the data predicted and generated under the second prediction condition based on the price adjustment factor, and determines a second data quantity adjusting parameter according to the basic conversion data quantity and the second actual conversion data quantity, wherein the second data quantity adjusting parameter is used for indicating the price adjustment degree of the computer device on the data predicted and generated under the second prediction condition.
Step S303, determining a consumption difference index of the test business object according to the first prediction total resource consumption and the second prediction total resource consumption, acquiring a transaction data volume index of the test business object, and determining a release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data volume index.
In this embodiment of the present application, the computer device may determine a ratio of the first predicted total resource consumption to the first user amount as the first predicted unit resource consumption, and determine a ratio of the second predicted total resource consumption to the second user amount as the second predicted unit resource consumption; the method comprises the steps of obtaining a total difference value between the first prediction unit resource consumption and the second prediction unit resource consumption, obtaining a flow ratio between the second user quantity and the first user quantity, and determining a consumption difference index of a test service object based on the total difference value and the flow ratio.
Further, the computer device may count a total number of conversions of the first user cluster and the second user cluster for the test service object in common, where the total number of conversions is used to represent a number of conversions occurring in an interaction behavior of the first user cluster and the second user cluster for the test service object in common. Specifically, the computer device may monitor the interaction behaviors of the first user cluster and the second user cluster with respect to the test service object, and count the conversions if the conversions occur in the interaction behaviors to obtain the total conversion number of the first user cluster and the second user cluster with respect to the test service object. The computer equipment can acquire conversion targets sent by all parties of the test business object, and determines the interactive behaviors belonging to the conversion targets in the interactive behaviors as the converted interactive behaviors; alternatively, the computer device may obtain the conventional conversion behavior, determine an interactive behavior belonging to the conventional conversion behavior in the interactive behaviors as an interactive behavior in which conversion occurs, and the like, which is not limited herein. For example, the test service object is a game promotion advertisement, the computer device obtains a conversion target sent by all parties of the test service object, such as a game downloading behavior, a game registration behavior, a game using behavior and the like, monitors the interactive behaviors of the first user cluster and the second user cluster for the game promotion advertisement, and counts the number of the interactive behaviors belonging to the conversion target in the interactive behaviors to obtain the total conversion number of the first user cluster and the second user cluster for the test service object.
Further, the computer device may obtain a target conversion data volume, and determine a transaction data volume index of the test service object based on the total conversion number and the target conversion data volume, where the target conversion data volume is a conversion data volume that all parties of the test service object expect to reach for the test service object. And the transaction data volume index represents the influence of the second business strategy on all the parties of the test business object on the basis of the first business strategy.
Further, the computer device may obtain a play processing criterion that includes at least two index ranges and a play decision corresponding to each index range. In the release processing standard, a target index range to which the consumption difference index and the transaction data amount index belong is searched, and a release decision corresponding to the target index range is determined as a release decision result of the test business object between the first business strategy and the second business strategy.
Optionally, the computer device may obtain a first predicted resource difference value between the first predicted total resource consumption and the transaction data amount index, and obtain a second predicted resource difference value between the second predicted total resource consumption and the transaction data amount index, and specifically, the computer device may determine a difference value between a ratio of the first predicted total resource consumption to the transaction data amount index and a standard consumption ratio as the first predicted resource difference value; and determining a difference value between a ratio of the second predicted total resource consumption to the transaction data volume index and a standard consumption proportion as a second predicted resource difference value, wherein the standard consumption proportion is used for representing a value expected to be reached by all parties of the test business object to the ratio between the predicted total resource consumption and the transaction data volume index, such as 1. If the second predicted resource difference value is smaller than or equal to the first predicted resource difference value and the consumption difference index and the transaction data volume index meet the strategy release condition, determining that the release decision result of the test business object between the first business strategy and the second business strategy is the release decision result aiming at the second business strategy; and if the second predicted resource difference value is greater than the first predicted resource difference value or the consumption difference index and the transaction data amount index do not meet the policy release condition, determining that the release decision result of the test service object between the first service policy and the second service policy is a release canceling result aiming at the second service policy. When the second predicted resource difference value is less than or equal to the first predicted resource difference value, the second predicted total resource consumption can better meet the expectation of all parties of the test business object, and therefore the second business strategy corresponding to the second predicted total resource consumption can be subjected to volume-releasing processing.
The release decision result is used to indicate the service policy for release processing between the first service policy and the second service policy, and the release mode of the service policy for release processing.
In this embodiment of the application, the computer device may predict, based on the first business policy, a first predicted total resource consumption, generated by the first user cluster for the test business object, under a first prediction condition; the first prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the first service strategy; under a second prediction condition, predicting second predicted total resource consumption generated by the second user cluster aiming at the test service object based on a second service strategy; the second prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the second service strategy; determining a consumption difference index of the test business object according to the first prediction total resource consumption and the second prediction total resource consumption, acquiring a transaction data volume index of the test business object, and determining a release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data volume index. When a test business object is put in by a second business strategy instead of the first business strategy, the benefits which can be obtained by a popularizing party putting in the test business object when the business strategy of the test business object is updated are represented by the consumption difference index, the benefits which can be obtained by all parties of the test business object when the business strategy of the test business object is updated are represented by the transaction data volume index, and the release decision result of the test business object between the first business strategy and the second business strategy is jointly performed according to the consumption difference index and the transaction data volume index, so that the benefits between the popularizing party and all parties of the test business object can be considered in multiple ways, the benefit balance between the popularizing party and all parties is improved, different business strategies can be better evaluated, and the accuracy and the scientificity of strategy decision on the test business object are improved. The business strategy determined by the scheme of the application has the advantages of low consumption, good promotion effect on users and the like, is favorable for saving the cost of a promotion party, saving system promotion resources and the like, and is also favorable for improving the income brought to all parties by promotion and test business objects.
Further, please refer to fig. 5, wherein fig. 5 is a schematic diagram of a specific flow of service data processing according to an embodiment of the present application. As shown in fig. 5, the method includes the steps of:
step S501, a test service object is delivered to a first user cluster based on a first service strategy, a test service object is delivered to a second user cluster based on a second service strategy, the current state is determined, and a condition indicating the current state is recorded as an actual test condition.
In the embodiment of the application, the computer device obtains a first user cluster and a second user cluster, puts a test service object into the first user cluster based on a first service strategy, puts the test service object into the second user cluster based on a second service strategy, generates the current state of the test service object, and records a condition indicating the current state of the test service object as an actual test condition. The process may refer to the specific description in step S301 in fig. 3, and is not described herein again. For example, if the test service object is an advertisement object, the first service policy of the advertisement object is a policy of delivering the advertisement object based on the push channel 1 and the push channel 2, and the conversion cost of delivering the advertisement object is the conversion cost 1, the second service policy is a policy of delivering the advertisement object based on the push channel 2 and the push channel 3, and the conversion cost of delivering the advertisement object is the conversion cost 2, the computer device delivers the advertisement object to the first user cluster based on the first service policy, and delivers the advertisement object to the second user cluster based on the second service policy, where the first user cluster may receive the advertisement object in the push channel 1 and the push channel 2, the conversion cost of the received advertisement object is the conversion cost 1, the second user cluster may receive the advertisement object in the push channel 2 and the push channel 3, and the conversion cost of the received advertisement object is the conversion cost 2.
Optionally, the computer device may generate an actual test launching block according to the test service object, the first service policy, the second service policy, the first user cluster, the second user cluster, and the like, and add the actual test launching block to the policy management block chain after the actual test launching block passes the consensus.
Step S502, under the actual test condition, acquiring the actual service data generated by the test service object.
In the embodiment of the application, the computer device obtains actual service data generated by the test service object under actual test conditions. The actual business data may include a first actual total resource consumption, a first actual conversion data amount, and a first actual conversion number, which are generated by the first user cluster for the test business object, where the first actual total resource consumption = the first actual conversion data amount × the first actual conversion number; the actual business data may further include a second actual total resource consumption, a second actual conversion data amount, and a second actual conversion number, which are generated by the second user cluster for the test business object, where the second actual total resource consumption = the second actual conversion data amount × the second actual conversion number.
Optionally, the computer device may obtain an object type of the test service object, and if the object type is a non-correction type, that is, the test service object is a non-correction type object, such as a CPC advertisement, the computer device may obtain actual service data, which is generated by the test service object under an actual test condition, where the actual service data includes first actual total resource consumption, which is generated by the first user cluster for the test service object, and second actual total resource consumption, which is generated by the second user cluster for the test service object. Specifically, please refer to fig. 6, where fig. 6 is a schematic diagram of a policy prediction scenario of a cost-to-click advertisement according to an embodiment of the present application. As shown in fig. 6, the traffic ratio between the second user cluster and the first user cluster is p: q, under an actual test condition (i.e. a condition indicating a current state), the computer device obtains a first actual total resource consumption of the first user cluster q for the test service object, and obtains a second actual total resource consumption of the second user cluster p for the test service object. If the object type is a calibration type, that is, the test service object is a calibration type object, such as an cpa advertisement, the computer device may obtain actual service data, which is generated by the test service object under an actual test condition, and includes a first actual total resource consumption, a first actual conversion data amount, and a first actual conversion number, which are generated by the first user cluster for the test service object, and a second actual total resource consumption, a second actual conversion data amount, and a second actual conversion number, which are generated by the second user cluster for the test service object. Specifically, please refer to fig. 7a, fig. 7a is a schematic diagram of a strategy prediction scenario for optimizing a behavior bid advertisement according to an embodiment of the present application. As shown in fig. 7a, the traffic ratio between the second user cluster and the first user cluster is p: q, the computer device obtains actual service data generated by the test service object under the actual test condition (i.e. the condition indicating the current state) under the actual test condition, where the actual service data includes first actual total resource consumption of the first user cluster q for the test service object and second actual total resource consumption of the second user cluster p for the test service object.
Optionally, the computer device may generate an actual data management block according to actual service data generated by the test service object under the actual test condition, and add the actual data management block to the policy management block chain after the actual data management block passes the consensus. Optionally, the actual test launching block may not be generated in step S501, the actual data management block may not be generated in step S502, the computer device directly generates the actual test block according to the test service object, the first service policy, the second service policy, the first user cluster, the second user cluster, the actual service data, and the like, and adds the actual test block to the policy management block chain after the actual test block passes the consensus.
Step S503, presetting a first prediction condition, and predicting first predicted service data generated by the test service object under the first prediction condition based on the actual service data.
In this embodiment of the present application, a computer device presets a first prediction condition under an actual test condition, where the first prediction condition is a condition for delivering a test service object to a first user cluster and a second user cluster, respectively, based on a first service policy, and the first prediction condition is used to indicate a condition of a no-policy state. The computer device may predict, based on the actual business data, first predicted business data generated by the test business object under a first prediction condition, such as a first predicted total resource consumption generated by the first user cluster for the test business object. If the test service object belongs to a non-correction type object, such as a CPC advertisement, as shown in fig. 6, the non-correction type object is not affected by the pricing factor, and therefore, the experimental group (i.e., the second user cluster) and the comparison group (i.e., the first user cluster) are not interfered with each other, the computer device presets a first prediction condition, because the service policy associated with the first user cluster under the actual test condition is the first service policy, and the service policy associated with the first user cluster under the first prediction condition is also the first service policy, under the first prediction condition, a first predicted total resource consumption generated by the first user cluster for the test service object is equal to a first actual total resource consumption generated by the first user cluster for the test service object under the actual test condition, and the computer device determines the first actual total resource consumption as the first predicted total resource consumption generated by the first user cluster for the test service object. Wherein the first actual total resource consumption is denoted C 01 Let the second actual total resource consumption be denoted as C 10 The first predicted total resource consumption is denoted as C 00 Determining a first actual total resource consumption as a first predicted total resource consumption, i.e. C 00 ＝C 01 。
If the test service object belongs to a calibration type object, such as an oCPA advertisement, as shown in FIG. 7a, the calibration type object is affected by the price adjustment factor, and therefore, it can be considered that the experimental group and the control group will interfere with each other and participate in the processReferring to fig. 4, the service policy associated with the comparison group (i.e., the first user cluster) under the actual test condition is the first service policy, the service policy associated under the first prediction condition is also the first service policy, and due to the change of the price adjustment factor, the first actual total resource consumption of the first user cluster under the actual test condition is different from the first predicted total resource consumption under the first prediction condition, that is, C 00 And C 01 Are not equal. Referring to fig. 7b specifically, fig. 7b is a schematic diagram of a pricing process scenario of a correction type object according to an embodiment of the present application. As shown in fig. 7b, under the actual test condition, the computer device obtains the first actual total resource consumption C of the first user cluster 01 And a second actual total resource consumption C of the second user cluster 10 Wherein, the first actual conversion data amount generated by the first user cluster aiming at the test service object is recorded as cpa 0 And recording a second actual conversion data volume generated by the second user cluster aiming at the test service object as cap 1 The computer device may perform weighting processing on the first actual conversion data amount and the second actual conversion data amount according to the first user amount and the second user amount to obtain a basic conversion data amount, and may refer to the basic conversion data amount as cpa 2 Assuming that the traffic ratio between the second user cluster and the first user cluster is p: and q, assuming that q and p are decimals and the sum of q and p is 1, the determination process of the basic conversion data amount can be shown in formula (1):
cpa 2 ＝cpa 0 *q+cpa 1 *p ①
under the first prediction condition, the business strategy associated with the first user cluster is always the first business strategy, the business strategy associated with the second user cluster is updated to the first business strategy from the second business strategy, and the first prediction average conversion data volume at the moment is obtained to be cpa 0 Adjusting the first predicted average conversion data amount to enable the first predicted average conversion data amount to be equal to cpa 0 Adjusted to cpa 2 Then the first predicted total resource consumption is adjusted accordingly. Wherein, according to the basic conversion data quantity and the first actual conversionThe data volume is converted, a first data volume adjustment parameter is determined, the first actual total resource consumption is adjusted based on the first data volume adjustment parameter, so as to obtain a first predicted total resource consumption generated by the first user cluster for the test service object under the first service policy, and a generation process of the first predicted total resource consumption may be as shown in formula (2):
C 00 ＝C 01 *cpa 2 /cpa 0 ②
wherein, in the formula (2), the cpa 2 /cpa 0 Adjusting a parameter for the determined first data quantity, based on the first data quantity adjustment parameter, for a first actual total resource consumption C 01 Adjusting to obtain a first predicted total resource consumption C 00 . Further, the first predicted total resource consumption may be simplified and expressed by equation (3):
C 00 ＝C 01 *cpa 2 /cpa 0 ＝conv 0 *cpa 2 ③
wherein, conv 0 For a first actual number of conversions, i.e., first actual number of conversions = first actual total resource consumption/first actual amount of conversion data, the computer device may determine a first predicted total resource consumption based on the first actual number of conversions and the amount of basic conversion data, as seen in equation (3).
Step S504, a second prediction condition is preset, and second predicted service data generated by the test service object under the second prediction condition is predicted based on the actual service data.
In this embodiment of the present application, a second prediction condition is preset by a computer device under an actual test condition, where the second prediction condition is a condition for delivering a test service object to a first user cluster and a second user cluster respectively based on a second service policy, and the second prediction condition is used to indicate a condition of a full policy state. The computer device may predict second predicted traffic data generated by the test traffic object under a second prediction condition based on the actual traffic data, such as a second predicted total resource consumption generated by the second user cluster for the test traffic object. Wherein if the testing industryIf the service object belongs to a non-correction type object, such as a CPC advertisement, as shown in fig. 6, the non-correction type object is not affected by the pricing factor, and therefore the experimental group (i.e., the second user cluster) and the control group (i.e., the first user cluster) are not interfered with each other, the computer device presets a second prediction condition, and because the service policy associated with the second user cluster under the actual test condition is the second service policy, and the service policy associated under the second prediction condition is the second service policy, under the second prediction condition, the second predicted total resource consumption generated by the second user cluster for the test service object is equal to the second actual total resource consumption generated by the second user cluster for the test service object under the actual test condition, and the computer device determines the second actual total resource consumption as the second predicted total resource consumption generated by the second user cluster for the test service object. Wherein the first actual total resource consumption is denoted C 01 Let the second actual total resource consumption be denoted as C 10 Let the second predicted total resource consumption be denoted as C 11 Determining a second actual total resource consumption as a second predicted total resource consumption, i.e. C 11 ＝C 10 。
If the test service object belongs to a calibration type object, such as an cpa advertisement, as shown in fig. 7a, the calibration type object is affected by the price adjustment factor, and therefore, it can be considered that the experimental group and the comparison group interfere with each other, as shown in fig. 4, the service policy associated with the experimental group (i.e., the second user cluster) under the actual test condition is the second service policy, and the service policy associated under the second prediction condition is the second service policy, and due to the price adjustment factor variation, the second actual total resource consumption of the second user cluster under the actual test condition is not equal to the second predicted total resource consumption under the second prediction condition, that is, C 11 And C 10 Not equal. Specifically, as shown in fig. 7b, under the actual test condition, the computer device obtains the first actual total resource consumption C of the first user cluster 01 And a second actual total resource consumption C of the second user cluster 10 Wherein the first user cluster is generated for the test service objectAn actual transformation data volume is recorded as cpa 0 And recording a second actual conversion data quantity generated by the second user cluster aiming at the test service object as cap 1 The computer device may perform weighting processing on the first actual conversion data volume and the second actual conversion data volume according to the first user number and the second user number to obtain a basic conversion data volume, and the basic conversion data volume is denoted as cpa 2 Assuming that the traffic ratio between the second user cluster and the first user cluster is p: and q, assuming that q and p are decimal numbers and the sum of q and p is 1, the determination process of the basic conversion data amount can be seen in formula (1).
Under a second prediction condition, the service strategy associated with the second user cluster is always the second service strategy, the service strategy associated with the first user cluster is updated from the first service strategy to the second service strategy, and a second prediction average conversion data volume at the moment is obtained and is cpa 1 Adjusting the second predicted average conversion data amount to be cpa 1 Adjusted to cpa 2 Then the second predicted total resource consumption is adjusted accordingly. Determining a second data volume adjustment parameter according to the basic conversion data volume and the second actual conversion data volume, and adjusting the second actual total resource consumption based on the second data volume adjustment parameter to obtain a second predicted total resource consumption of the second user cluster for the test service object under the second service policy, where the generation process of the second predicted total resource consumption may be shown in formula (4):
C 11 ＝C 10 *cpa 2 /cpa 1 ④
wherein, in the formula (4), the cpa 2 /cpa 1 Adjusting the parameter for the determined second amount of data, based on the second amount of data adjusting parameter, to a second actual total resource consumption C 10 Adjusting to obtain a second predicted total resource consumption C 11 . Further, the second predicted total resource consumption may be simplified and expressed by equation (5):
C 11 ＝C 10 *cpa 2 /cpa 1 ＝conv 1 *cpa 2 ⑤
wherein, conv 1 For the second actual conversion number, i.e., the second actual conversion number = second actual total resource consumption/second actual conversion data amount, according to equation (5), the computer device may determine the second predicted total resource consumption according to the second actual conversion number and the basic conversion data amount.
Step S505, determining a consumption difference index of the test service object based on the first predicted service data and the second predicted service data.
In this embodiment, the computer device may determine a ratio of the first predicted total resource consumption to the first user amount as a first predicted unit resource consumption, determine a ratio of the second predicted total resource consumption to the second user amount as a second predicted unit resource consumption, obtain a total difference value between the first predicted unit resource consumption and the second predicted unit resource consumption, and determine a consumption difference index based on the total difference value, where the total difference value in the full policy state compared to the no policy state may be represented as formula (6):
Figure BDA0002907743740000261
further, the computer device may obtain a flow ratio between the second number of users and the first number of users, and determine a consumption difference index of the test service object based on the total difference value F' and the flow ratio, where the consumption difference index may be represented by formula (7):
Figure BDA0002907743740000262
wherein, assuming that p and q are decimals, the sum of p and q is 1, and the consumption difference index can be further expressed as formula (8):
Figure BDA0002907743740000263
wherein, for the non-correction type object, the consumption difference index can be expressed as formula (9):
Figure BDA0002907743740000264
as can be seen from the formulas (7), (8) and (9), when the conversions of the first and second user clusters are both 0, the consumption difference index is 0, and the idle consumption can be eliminated, so that when and only when the conversion number is increased, the object benefit of the second service policy is considered to be increased on the basis of the object benefit of the first service policy, and the accuracy of the release decision can be improved by performing the release decision on the first and second service policies through the consumption difference index.
Step S506, the transaction data volume index of the test business object is obtained.
In this embodiment, the computer device may count a total conversion number of the first user cluster and the second user cluster for the test service object, count the total conversion number as conv, obtain a target conversion data amount, count the target conversion data amount as target _ cpa, and determine a transaction data amount index of the test service object based on the total conversion number and the target conversion data amount, where a generation process of the transaction data amount index may be as shown in formula (10):
GMV＝conv*target_cpa (10)
the transaction data amount index may be a sum of the conversion number of each advertisement multiplied by a target conversion data amount, and reflects a distribution value of user traffic, and the larger the GMV is, the higher the value of the user traffic is, the transaction data amount index is only related to the conversion number of the test service object and the target conversion data amount, and may reflect an object benefit that all parties of the test service object may obtain.
Step S507, determining a release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data amount index.
In the embodiment of the present application, the computer device obtains a quantitative play processing criterion, where the quantitative play processing criterion includes at least two index ranges and a quantitative play decision corresponding to each index range, where the quantitative play processing criterion may be shown in table 1:
TABLE 1
Figure BDA0002907743740000271
Figure BDA0002907743740000281
As shown in table 1 above, each row may be considered as an index range and a release decision corresponding to the index range, and in the release processing criteria, a target index range to which the consumption difference index and the transaction data amount index belong is searched. Specifically, the computer device may obtain a first predicted resource difference value between the first predicted total resource consumption and the transaction data amount index, optionally, the first predicted resource difference value may be a ratio of the first predicted total resource consumption to the transaction data amount index, a difference between the ratio of the first predicted total resource consumption to the transaction data amount index and a standard consumption ratio, a difference between the first predicted total resource consumption and the transaction data amount index, or the like, which is not limited herein; optionally, the second predicted resource difference value may be a ratio of the second predicted total resource consumption to the transaction data amount index, or a difference between the ratio of the second predicted total resource consumption to the transaction data amount index and the standard consumption ratio, or a difference between the second predicted total resource consumption and the transaction data amount index, and the like, which is not limited herein. The computer device may find a target index range to which the consumption difference index, the transaction data amount index, the first predicted resource difference value, and the second predicted resource difference value belong in the put processing criterion. And the computer equipment determines the release decision corresponding to the target index range as a release decision result of the test service object between the first service strategy and the second service strategy. Wherein, the second predicted resource difference is less than or equal to the first predicted resource difference, indicating that the difference between the consumption and transaction data volume indicators is reduced. For example, if the computer device obtains that the consumption difference index is a positive number, the transaction data amount index on the second business strategy is higher than the transaction data amount index on the first business strategy, and the value of large disk consumption/GMV is greater than 1, it determines that the release decision result is the release determination result for the second business strategy.
And step S508, performing volume releasing processing on the first service strategy and the second service strategy based on the volume releasing decision result.
In the embodiment of the application, if the release decision result is a release determination result for the second service policy, the local user is acquired from the first user cluster based on the release decision result, and the test service object released to the local user is updated from the first service policy to the second service policy; and if the release decision result is a release canceling result aiming at the second service strategy, updating the test service object delivered to the second user cluster into the first service strategy from the second service strategy, namely, the computer equipment respectively delivers the test service object to the first user cluster and the second user cluster on the basis of the first service strategy.
When the computer device obtains the local users from the first user cluster based on the volume decision result, the sum of the first user number and the second user number can be obtained to obtain the total number of the users, and the user proportion of the second user number in the total number of the users is determined; acquiring a first lofting level to which the proportion of users belongs, determining a second lofting level according to the first lofting level, and determining decision quantity according to the second lofting level and the total number of users; determining the quantity of the to-be-released quantity based on the decision quantity and the second user quantity, acquiring M users to be updated from the first user cluster according to the quantity of the to-be-released quantity, and determining the M users to be updated as local users; m is the amount to be put. For example, there are several release levels of 1%, 5%, 10%, 20%, 50%, and 100%, the computer device determines that the user proportion of the second user amount in the total user amount is 7%, determines that the first release level to which the user proportion "7%" belongs is between 5% and 10%, determines that the second release level is 10% according to the first release level, determines the decision amount according to the second release level and the total user amount, at this time, the decision amount is a product of the total user amount and the second release level, determines the release amount based on the decision amount and the second user amount, the release amount is a difference between the decision amount and the second user amount, the computer device obtains M users to be updated from the first user cluster according to the release amount, determines the M users to be updated as local users, and M is the release amount. If the release decision result is a release canceling result for the second service policy, it indicates that the object benefit of the second service policy is smaller than the object benefit of the first service policy, that is, the second service policy has a poorer effect than the first service policy, and therefore, the use of the second service policy can be directly cancelled, and the test service object is delivered to the first user cluster and the second user cluster based on the first service policy. If the release decision result is the determined release result for the second service policy, the computer device may directly release the test service object to the first user cluster based on the second service policy, or may obtain a local user from the first user cluster, and gradually release the second service policy based on the release level, thereby continuously detecting the second service policy and improving the accuracy of the release decision for the service policy.
Optionally, the computer device may call an intelligent contract based on the play decision result, and execute a play processing function corresponding to the play decision result based on the intelligent contract; and executing a release processing process corresponding to the release decision result on the first service strategy and the second service strategy through a release processing function, wherein the release processing process is used for expressing a process of releasing a test service object based on the first service strategy and the second service strategy, and the release processing process comprises the number of users associated with the first service strategy and the number of users associated with the second service strategy when the test service object is released.
In the embodiment of the application, the computer device may predict, based on the first business strategy, first predicted total resource consumption, generated by the first user cluster for the test business object, under the first prediction condition; the first prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the first service strategy; under a second prediction condition, predicting second predicted total resource consumption of a second user cluster aiming at the test service object based on a second service strategy; the second prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the second service strategy; and determining a consumption difference index of the test business object according to the first predicted total resource consumption and the second predicted total resource consumption, acquiring a transaction data volume index of the test business object, and determining a release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data volume index. When a test business object is put in by a second business strategy instead of the first business strategy, the consumption difference index represents the income which can be obtained by a popularization party putting in the test business object when the business strategy of the test business object is updated, the transaction data volume index represents the income which can be obtained by all parties of the test business object when the business strategy of the test business object is updated, and the put decision result of the test business object between the first business strategy and the second business strategy is jointly carried out according to the consumption difference index and the transaction data volume index, so that the income between the popularization party and all parties of the test business object can be considered in multiple ways, the income balance between the popularization party and all parties is improved, different business strategies can be better evaluated, and the accuracy and the scientificity of strategy decision of the test business object are improved.
The first business strategy and the second business strategy are subjected to an experiment strategy, the strategy is multiplied by 1.01 times, 1.03 times and 1.05 times on the basis of the existing pcvr value artificially by an experiment group, and the existing consumption indexes, the consumption difference indexes in the application and the transaction data volume indexes are detected as far as possible, so that the difference of each index can be obtained, and the experiment strategy is specifically shown in table 2:
TABLE 2
Multiplying original pcvr by 1.01 Multiplying original pcvr by 1.03 Multiplying original pcvr by 1.05
Consumption of the experimental group 5.56M 5.60M 5.63M
Consumption of control group 5.53M 5.53M 5.53M
Consumption promotion 0.48％ 1.35％ 1.88％
Predicted consumption by the experimental group 5.52M 5.55M 5.53M
Control group predicted consumption 5.53M 5.53M 5.53M
Consumption difference index -0.20％ 0.42％ -0.08％
Experimental group GMV 5.37M 5.41M 5.40M
Control group GMV 5.41M 5.41M 5.41M
Transaction data volume index -0.80％ -0.07％ -0.14％
As can be seen from table 2, the experimental effect is evaluated according to the existing consumption index, and when the multiple of the artificial increase of the pcvr is higher, the consumption of the experimental group is increased more than that of the control group, because the experimental group overestimates the pcvr, the consumption is increased, but the conversion rate and the like are not changed in practice, so that the actually obtained conversion is not increased by the second business strategy, and based on the present application, it can be seen that no matter how many times the pcvr is increased, the consumption difference index and the transaction data amount index are not obviously increased, and the personal strategy for overestimating the pcvr is obviously not a good strategy.
Therefore, if the consumption index is used for evaluating the experimental effect, whether the experiment is really beneficial or not can be judged wrongly, and the consumption difference index and the transaction data volume index in the application consider the profits of the business strategy on the promotion party and all the parties of the business object test, so that whether one strategy is really beneficial or not can be evaluated more accurately, the release decision can be correctly carried out on the business strategy, and the accuracy of the release decision of the business strategy is improved.
Further, please refer to fig. 8, wherein fig. 8 is a schematic diagram of a service data processing apparatus according to an embodiment of the present application. The service data processing apparatus may be a computer program (including program code, etc.) running in a computer device, for example, the service data processing apparatus may be an application software; the apparatus may be used to perform the corresponding steps in the methods provided by the embodiments of the present application. As shown in fig. 8, the service data processing apparatus 800 may be used in a computer device in the embodiment corresponding to fig. 3, and specifically, the apparatus may include: a first consumption obtaining module 11, a second consumption obtaining module 12 and an index generating module 13.
The first consumption obtaining module 11 is configured to predict, under a first prediction condition, first predicted total resource consumption, which is generated by the first user cluster for the test service object, based on the first service policy; the first prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the first service strategy;
a second consumption obtaining module 12, configured to predict, under a second prediction condition, second predicted total resource consumption, which is generated by a second user cluster for the test service object, based on a second service policy; the second prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the second service strategy;
and the index generation module 13 is configured to determine a consumption difference index of the test service object according to the first predicted total resource consumption and the second predicted total resource consumption, acquire a transaction data amount index of the test service object, and determine a release decision result of the test service object between the first service policy and the second service policy according to the consumption difference index and the transaction data amount index.
Wherein, the apparatus 800 further comprises:
the object delivery module 14 is configured to deliver the test service object to the first user cluster based on the first service policy and deliver the test service object to the second user cluster based on the second service policy under the actual test condition;
the first data acquisition module 15 is configured to acquire, based on a first service policy, first actual total resource consumption and a first actual conversion data amount that are generated by a first user cluster for a test service object;
the second data obtaining module 16 is configured to obtain, based on a second service policy, a second actual conversion data amount generated by the second user cluster for the test service object;
the first consumption acquisition module 11 includes:
a first consumption determining unit 111, configured to determine, under a first prediction condition and based on a first service policy, a first actual total resource consumption as a first predicted total resource consumption generated by the first user cluster for the test service object if the test service object belongs to a non-correction type object;
a second consumption determining unit 112, configured to determine, according to the first actual conversion data amount, the second actual conversion data amount, and the first actual total resource consumption, a first predicted total resource consumption, which is generated by the first user cluster for the test service object under the first service policy, if the test service object belongs to the correction type object.
The number of users in the first user cluster is the first user number, and the number of users in the second user cluster is the second user number;
the second consumption determining unit 112 includes:
a first weighting subunit 1121, configured to perform weighting processing on the first actual conversion data amount and the second actual conversion data amount based on the first user amount and the second user amount to obtain a basic conversion data amount;
a first adjusting subunit 1122, configured to determine a first data amount adjusting parameter according to the basic conversion data amount and the first actual conversion data amount;
the first prediction subunit 1123 is configured to perform adjustment processing on the first actual total resource consumption based on the first data amount adjustment parameter, so as to obtain a first predicted total resource consumption, which is generated by the first user cluster for the test service object under the first service policy.
Wherein, the apparatus 800 further comprises:
the object launching module 14 is further configured to launch the test service object to the first user cluster based on the first service policy and launch the test service object to the second user cluster based on the second service policy under the actual test condition;
the first data obtaining module 15 is further configured to obtain, based on the first service policy, a first actual conversion data amount generated by the first user cluster for the test service object;
the second data obtaining module 16 is further configured to obtain, based on a second service policy, a second actual conversion data amount and a second actual total resource consumption, which are generated by the second user cluster for the test service object;
the second consumption acquiring module 12 includes:
a third consumption determining unit 121, configured to determine, under a second prediction condition and based on a second service policy, a second actual total resource consumption as a second predicted total resource consumption generated by the second user cluster for the test service object if the test service object belongs to the non-correction type object;
a fourth consumption determining unit 122, configured to determine, according to the first actual conversion data amount, the second actual conversion data amount, and the second actual total resource consumption, a second predicted total resource consumption, which is generated by the second user cluster for the test service object under the second service policy, if the test service object belongs to the correction type object.
The number of the users in the first user cluster is the first user number, and the number of the users in the second user cluster is the second user number;
the fourth consumption determining unit 122 includes:
a second weighting subunit 1221, configured to perform weighting processing on the first actual conversion data amount and the second actual conversion data amount based on the first user amount and the second user amount, to obtain a basic conversion data amount;
a second adjusting subunit 1222, configured to determine a second data amount adjusting parameter according to the basic conversion data amount and the second actual conversion data amount;
a second predicting subunit 1223, configured to perform adjustment processing on the second actual total resource consumption based on the second data volume adjustment parameter, to obtain a second predicted total resource consumption, which is generated by the second user cluster for the test service object under the second service policy.
The number of users in the first user cluster is the first user number, and the number of users in the second user cluster is the second user number;
in determining a consumption difference index of the test business object according to the first predicted total resource consumption and the second predicted total resource consumption, the index generating module 13 includes:
a unit consumption obtaining unit 131, configured to determine a ratio of the first predicted total resource consumption to the first user amount as a first predicted unit resource consumption, and determine a ratio of the second predicted total resource consumption to the second user amount as a second predicted unit resource consumption;
a consumption index generating unit 132, configured to obtain a total difference between the first predicted unit resource consumption and the second predicted unit resource consumption, obtain a traffic ratio between the second number of users and the first number of users, and determine a consumption difference index of the test service object based on the total difference and the traffic ratio.
In obtaining the transaction data amount index of the test service object, the index generating module 13 includes:
the conversion counting unit 133 is configured to count the total conversion number of the first user cluster and the second user cluster for the test service object; the total conversion number is used for representing the number of conversions in the interaction behaviors of the first user cluster and the second user cluster aiming at the test service object together;
a transaction index generating unit 134, configured to obtain a target conversion data amount, and determine a transaction data amount index of the test service object based on the total conversion number and the target conversion data amount; the target translation data volume refers to the translation data volume that all parties of the test business object expect to achieve for the test business object.
Wherein, in determining the release decision result of the test service object between the first service policy and the second service policy according to the consumption difference index and the transaction data amount index, the index generating module 13 includes:
a standard acquisition unit 135 for acquiring a release amount processing standard; the release processing standard comprises at least two index ranges and a release decision corresponding to each index range;
the decision determining unit 136 is configured to search a target index range to which the consumption difference index and the transaction data amount index belong in the release processing standard, and determine a release decision corresponding to the target index range as a release decision result of the test service object between the first service policy and the second service policy.
Wherein, in determining the release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data amount index, the index generating module 13 includes:
a difference obtaining unit 137, configured to obtain a first predicted resource difference value between the first predicted total resource consumption and the transaction data amount index, and obtain a second predicted resource difference value between the second predicted total resource consumption and the transaction data amount index;
a first result determining unit 138, configured to determine that a release decision result of the test service object between the first service policy and the second service policy is, for a release result of the second service policy, if the second predicted resource difference value is less than or equal to the first predicted resource difference value and the consumption difference index and the transaction data amount index meet the policy release condition;
a second result determining unit 139, configured to determine that the release decision result of the test service object between the first service policy and the second service policy is a release cancellation result for the second service policy if the second predicted resource difference value is greater than the first predicted resource difference value or the consumption difference indicator and the transaction data amount indicator do not satisfy the policy release condition.
Wherein the apparatus 800 further comprises:
a policy updating module 17, configured to, if the release decision result is a release determination result for the second service policy, obtain a local user from the first user cluster based on the release decision result, and update the test service object delivered to the local user from the first service policy to the second service policy;
the policy updating module 17 is further configured to update the test service object delivered to the second user cluster from the second service policy to the first service policy if the release decision result is a release cancellation result for the second service policy.
Wherein, in terms of obtaining the local users from the first user cluster based on the volume decision result, the policy updating module 17 includes:
a proportion determining unit 171, configured to obtain a sum of the first user quantity and the second user quantity to obtain a user total number, and determine a user proportion of the second user quantity in the user total number;
the quantity obtaining unit 172 is configured to obtain a first release level to which the user proportion belongs, determine a second release level according to the first release level, and determine a decision quantity according to the second release level and the total number of users;
a user determining unit 173, configured to determine the number of to-be-released users based on the decision number and the second number of users, obtain M users to be updated from the first user cluster according to the number of to-be-released users, and determine the M users to be updated as local users; m is the quantity of the volume to be placed.
Wherein the apparatus 800 further comprises:
the contract calling module 18 is used for calling an intelligent contract based on the release decision result and executing a release processing function corresponding to the release decision result based on the intelligent contract;
and the play processing module 19 is configured to execute a play processing procedure corresponding to the play decision result on the first service policy and the second service policy through a play processing function.
The embodiment of the application provides a business data processing device, which can run in computer equipment and can predict first predicted total resource consumption of a first user cluster generated aiming at a test business object based on a first business strategy under a first prediction condition; the first prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the first service strategy; under a second prediction condition, predicting second predicted total resource consumption of a second user cluster aiming at the test service object based on a second service strategy; the second prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the second service strategy; determining a consumption difference index of the test business object according to the first prediction total resource consumption and the second prediction total resource consumption, acquiring a transaction data volume index of the test business object, and determining a release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data volume index. When a test business object is put in by a second business strategy instead of the first business strategy, the benefits which can be obtained by a popularizing party putting in the test business object when the business strategy of the test business object is updated are represented by the consumption difference index, the benefits which can be obtained by all parties of the test business object when the business strategy of the test business object is updated are represented by the transaction data volume index, and the release decision result of the test business object between the first business strategy and the second business strategy is jointly performed according to the consumption difference index and the transaction data volume index, so that the benefits between the popularizing party and all parties of the test business object can be considered in multiple ways, the benefit balance between the popularizing party and all parties is improved, different business strategies can be better evaluated, and the accuracy and the scientificity of strategy decision on the test business object are improved.
Referring to fig. 9, fig. 9 is a schematic structural diagram of a computer device according to an embodiment of the present disclosure. As shown in fig. 9, the computer device in the embodiment of the present application may include: one or more processors 901, memory 902, and input-output interface 903. The processor 901, the memory 902, and the input/output interface 903 are connected by a bus 904. The memory 902 is used for storing a computer program comprising program instructions, and the input/output interface 903 is used for receiving data and outputting data, such as data interaction between a computer device and a user terminal; processor 901 is operative to execute program instructions stored in memory 902.
The processor 901 may perform the following operations:
under a first prediction condition, predicting first prediction total resource consumption of the first user cluster aiming at the test service object based on the first service strategy; the first prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the first service strategy;
under a second prediction condition, predicting second predicted total resource consumption generated by the second user cluster aiming at the test service object based on a second service strategy; the second prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the second service strategy;
and determining a consumption difference index of the test business object according to the first predicted total resource consumption and the second predicted total resource consumption, acquiring a transaction data volume index of the test business object, and determining a release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data volume index.
In some possible embodiments, the processor 901 may be a Central Processing Unit (CPU), and the processor may also be other general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, and the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 902 may include a read-only memory and a random access memory, and provides instructions and data to the processor 901 and the input/output interface 903. A portion of the memory 902 may also include non-volatile random access memory. For example, the memory 902 may also store device type information.
In a specific implementation, the computer device may execute, through each built-in functional module thereof, the implementation manner provided in each step in fig. 3 or fig. 5, which may be referred to specifically for the implementation manner provided in each step in fig. 3 or fig. 5, and is not described herein again.
The embodiment of the present application provides a computer device, including: the system comprises a processor, an input/output interface and a memory, wherein the processor acquires a computer program in the memory, and executes each step of the method shown in the figure 3 to perform service data processing operation. The embodiment of the application realizes that under a first prediction condition, the first predicted total resource consumption of the first user cluster generated aiming at the test service object is predicted based on the first service strategy; the first prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the first service strategy; under a second prediction condition, predicting second predicted total resource consumption generated by the second user cluster aiming at the test service object based on a second service strategy; the second prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the second service strategy; determining a consumption difference index of the test business object according to the first prediction total resource consumption and the second prediction total resource consumption, acquiring a transaction data volume index of the test business object, and determining a release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data volume index. When a test business object is put in by a second business strategy instead of the first business strategy, the consumption difference index represents the income which can be obtained by a popularization party putting in the test business object when the business strategy of the test business object is updated, the transaction data volume index represents the income which can be obtained by all parties of the test business object when the business strategy of the test business object is updated, and the put decision result of the test business object between the first business strategy and the second business strategy is jointly carried out according to the consumption difference index and the transaction data volume index, so that the income between the popularization party and all parties of the test business object can be considered in multiple ways, the income balance between the popularization party and all parties is improved, different business strategies can be better evaluated, and the accuracy and the scientificity of strategy decision of the test business object are improved.
An embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored, where the computer program is suitable for being loaded by the processor and executing the service data processing method provided in each step in fig. 3 or fig. 5, and for details, reference may be made to an implementation manner provided in each step in fig. 3 or fig. 5, which is not described herein again. In addition, the beneficial effects of the same method are not described in detail. For technical details not disclosed in embodiments of the computer-readable storage medium referred to in the present application, reference is made to the description of embodiments of the method of the present application. By way of example, a computer program can be deployed to be executed on one computer device or on multiple computer devices at one site or distributed across multiple sites and interconnected by a communication network.
The computer-readable storage medium may be the service data processing apparatus provided in any of the foregoing embodiments or an internal storage unit of the computer device, such as a hard disk or a memory of the computer device. The computer readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Memory Card (SMC), a Secure Digital (SD) card, a flash card (flash card), and the like, provided on the computer device. Further, the computer-readable storage medium may also include both an internal storage unit and an external storage device of the computer device. The computer-readable storage medium is used for storing the computer program and other programs and data required by the computer device. The computer readable storage medium may also be used to temporarily store data that has been output or is to be output.
Embodiments of the present application also provide a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instruction from the computer readable storage medium, and executes the computer instruction, so that the computer device executes the method provided in the various optional manners in fig. 3 or fig. 5, when a test business object is put in a second business strategy instead of the first business strategy, the profit which can be obtained by a popularizing party putting in the test business object when the business strategy of the test business object is updated is represented by the consumption difference index, the profit which can be obtained by all parties of the test business object when the business strategy of the test business object is updated is represented by the transaction data volume index, and the yield between the first business strategy and the second business strategy of the test business object is jointly considered according to the consumption difference index and the transaction data volume index, so that the profit balance between the popularizing party and all parties of the test business object can be better considered, and the profit balance between the popularizing party and all parties can be improved, and different business strategies can be better evaluated, and the accuracy and scientific strategy of the test business object can be better evaluated.
The terms "first," "second," and the like in the description and claims of embodiments of the present application and in the drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "comprises" and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product, or apparatus that comprises a list of steps or elements is not limited to the listed steps or modules, but may alternatively include other steps or modules not listed or inherent to such process, method, apparatus, product, or apparatus.
Those of ordinary skill in the art will appreciate that the various illustrative components and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the components and steps of the various examples have been described herein generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The method and the related apparatus provided by the embodiments of the present application are described with reference to the flowchart and/or the structural diagram of the method provided by the embodiments of the present application, and each flow and/or block of the flowchart and/or the structural diagram of the method, and the combination of the flow and/or block in the flowchart and/or the block diagram can be specifically implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable business data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable business data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block or blocks of the block diagram. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable business data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block or blocks of the block diagram. These computer program instructions may also be loaded onto a computer or other programmable business data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above disclosure is only for the purpose of illustrating the preferred embodiments of the present application and is not to be construed as limiting the scope of the present application, so that the present application is not limited thereto, and all equivalent variations and modifications can be made to the present application.

Claims (14)

1. A method for processing service data, the method comprising:
under an actual test condition, a test service object is delivered to a first user cluster based on a first service strategy, the test service object is delivered to a second user cluster based on a second service strategy, and a first prediction condition and a second prediction condition are preset under the actual test condition;
under the first prediction condition, predicting first prediction total resource consumption generated by the first user cluster aiming at the test business object based on the first business strategy and the object type of the test business object; the first prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the first service strategy;
under the second prediction condition, predicting second predicted total resource consumption of the second user cluster aiming at the test service object based on the second service strategy and the object type of the test service object; the second prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the second service strategy;
determining a consumption difference index of the test business object according to the first predicted total resource consumption and the second predicted total resource consumption;
counting the total conversion number of the first user cluster and the second user cluster aiming at the test service object; the total conversion number is used for representing the number of conversions in the interaction behaviors of the first user cluster and the second user cluster aiming at the test service object together;
acquiring a target conversion data volume, and determining a transaction data volume index of the test service object based on the total conversion number and the target conversion data volume; the target conversion data volume is the conversion data volume which is expected to be achieved by all the parties of the test service object aiming at the test service object;
determining a release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data amount index; the volume releasing decision result is used for indicating the business strategy for performing the volume releasing processing between the first business strategy and the second business strategy and the volume releasing mode of the business strategy for performing the volume releasing processing.
2. The method of claim 1, wherein the method further comprises:
acquiring first actual total resource consumption and first actual conversion data volume generated by the first user cluster aiming at the test service object based on the first service strategy;
acquiring a second actual conversion data volume generated by the second user cluster aiming at the test service object based on the second service strategy;
the predicting, under the first prediction condition, a first predicted total resource consumption of the first user cluster for the test business object based on the first business strategy and the object type of the test business object includes:
under the first prediction condition, if the object type of the test service object is a non-correction type object, determining the first actual total resource consumption as a first predicted total resource consumption generated by the first user cluster for the test service object based on the first service policy;
and if the object type of the test service object is a correction type object, determining first predicted total resource consumption of the first user cluster aiming at the test service object under the first service strategy according to the first actual conversion data volume, the second actual conversion data volume and the first actual total resource consumption.
3. The method of claim 2, wherein the number of users in the first cluster of users is a first number of users and the number of users in the second cluster of users is a second number of users;
the determining, according to the first actual conversion data amount, the second actual conversion data amount, and the first actual total resource consumption, a first predicted total resource consumption generated by the first user cluster for the test service object under the first service policy includes:
based on the first user quantity and the second user quantity, weighting the first actual conversion data quantity and the second actual conversion data quantity to obtain a basic conversion data quantity;
determining a first data volume adjusting parameter according to the basic conversion data volume and the first actual conversion data volume;
and adjusting the first actual total resource consumption based on the first data volume adjustment parameter to obtain a first predicted total resource consumption generated by the first user cluster for the test service object under the first service policy.
4. The method of claim 1, wherein the method further comprises:
acquiring a first actual conversion data volume generated by the first user cluster aiming at the test service object based on the first service strategy;
acquiring a second actual conversion data volume and a second actual total resource consumption generated by the second user cluster aiming at the test service object based on the second service strategy;
the predicting, under the second prediction condition, a second predicted total resource consumption of the second user cluster for the test business object based on the second business strategy and the object type of the test business object includes:
under the second prediction condition, if the object type of the test service object is a non-correction type object, determining the second actual total resource consumption as a second predicted total resource consumption generated by the second user cluster for the test service object based on the second service policy;
and if the object type of the test service object is a correction type object, determining second predicted total resource consumption of the second user cluster aiming at the test service object under the second service strategy according to the first actual conversion data volume, the second actual conversion data volume and the second actual total resource consumption.
5. The method of claim 4, wherein the number of users in the first cluster of users is a first number of users and the number of users in the second cluster of users is a second number of users;
determining, according to the first actual conversion data volume, the second actual conversion data volume, and the second actual total resource consumption, a second predicted total resource consumption generated by the second user cluster for the test service object under the second service policy, including:
based on the first user quantity and the second user quantity, weighting the first actual conversion data quantity and the second actual conversion data quantity to obtain a basic conversion data quantity;
determining a second data volume adjusting parameter according to the basic conversion data volume and the second actual conversion data volume;
and adjusting the second actual total resource consumption based on the second data volume adjustment parameter to obtain a second predicted total resource consumption generated by the second user cluster for the test service object under the second service policy.
6. The method of claim 1, wherein the number of users in the first cluster of users is a first number of users and the number of users in the second cluster of users is a second number of users;
determining a consumption difference index of the test service object according to the first predicted total resource consumption and the second predicted total resource consumption, including:
determining the ratio of the first predicted total resource consumption to the first user quantity as a first predicted unit resource consumption, and determining the ratio of the second predicted total resource consumption to the second user quantity as a second predicted unit resource consumption;
acquiring a total difference value between the first prediction unit resource consumption and the second prediction unit resource consumption, acquiring a flow ratio between the second user quantity and the first user quantity, and determining a consumption difference index of the test service object based on the total difference value and the flow ratio.
7. The method of claim 1, wherein said determining a release decision result for the test business object between the first business strategy and the second business strategy based on the consumption difference indicator and the transaction data volume indicator comprises:
acquiring a release processing standard; the release processing standard comprises at least two index ranges and a release decision corresponding to each index range;
and searching a target index range to which the consumption difference index and the transaction data volume index belong in the release processing standard, and determining a release decision corresponding to the target index range as a release decision result of the test service object between the first service strategy and the second service strategy.
8. The method of claim 1, wherein said determining a release decision result for the test business object between the first business strategy and the second business strategy based on the consumption difference indicator and the transaction data volume indicator comprises:
acquiring a first predicted resource difference value between the first predicted total resource consumption and the transaction data volume index, and acquiring a second predicted resource difference value between the second predicted total resource consumption and the transaction data volume index;
if the second predicted resource difference value is less than or equal to the first predicted resource difference value, and the consumption difference index and the transaction data volume index meet a policy release condition, determining that a release decision result of the test service object between the first service policy and the second service policy is a release decision result for the second service policy;
if the second predicted resource difference value is greater than the first predicted resource difference value, or the consumption difference index and the transaction data volume index do not satisfy the policy release condition, determining that a release decision result of the test service object between the first service policy and the second service policy is a release cancellation result for the second service policy.
9. The method of claim 1, wherein the method further comprises:
if the release decision result is a release determination result for the second service policy, acquiring a local user from the first user cluster based on the release decision result, and updating the test service object released to the local user from the first service policy to the second service policy;
and if the release decision result is a release canceling result aiming at the second service strategy, updating the test service object released to the second user cluster into the first service strategy from the second service strategy.
10. The method of claim 9, wherein the obtaining local users from the first cluster of users based on the volume decision result comprises:
obtaining the sum of the first user number and the second user number to obtain the total number of users, and determining the user proportion of the second user number in the total number of users;
acquiring a first lofting level to which the user proportion belongs, determining a second lofting level according to the first lofting level, and determining decision quantity according to the second lofting level and the total number of the users;
determining the quantity of the amount to be updated based on the decision quantity and the second user quantity, acquiring M users to be updated from a first user cluster according to the quantity to be updated, and determining the M users to be updated as local users; and M is the quantity of the amount to be put.
11. The method of claim 1, wherein the method further comprises:
calling an intelligent contract based on the release decision result, and executing a release processing function corresponding to the release decision result based on the intelligent contract;
and executing a release processing process corresponding to the release decision result on the first service strategy and the second service strategy through the release processing function.
12. A service data processing apparatus, characterized in that the apparatus comprises:
the object delivery module is used for delivering a test service object to a first user cluster based on a first service strategy under an actual test condition, delivering the test service object to a second user cluster based on a second service strategy, and presetting a first prediction condition and a second prediction condition under the actual test condition;
a first consumption obtaining module, configured to predict, under the first prediction condition, a first predicted total resource consumption, which is generated by the first user cluster for the test service object, based on the first service policy and the object type of the test service object; the first prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the first service strategy;
a second consumption obtaining module, configured to predict, under the second prediction condition, a second predicted total resource consumption, which is generated by the second user cluster for the test service object, based on the second service policy and the object type of the test service object; the second prediction condition is a condition for respectively delivering the test service object to the first user cluster and the second user cluster based on the second service strategy;
the index generation module is used for determining the consumption difference index of the test business object according to the first predicted total resource consumption and the second predicted total resource consumption;
the index generation module is further used for counting the total conversion number of the first user cluster and the second user cluster for the test service object; the total conversion number is used for representing the number of conversions in the interaction behaviors of the first user cluster and the second user cluster aiming at the test service object together;
the index generation module is further used for acquiring a target conversion data volume and determining a transaction data volume index of the test service object based on the total conversion number and the target conversion data volume; the target conversion data volume refers to the conversion data volume which is expected to be achieved by all parties of the test business object aiming at the test business object;
the index generation module is further used for determining a release decision result of the test business object between the first business strategy and the second business strategy according to the consumption difference index and the transaction data amount index; the release decision result is used for indicating the service strategy for performing release processing between the first service strategy and the second service strategy and the release mode of the service strategy for performing release processing.
13. A computer device comprising a processor, a memory, an input output interface;
the processor is connected to the memory and the input/output interface respectively, wherein the input/output interface is used for receiving data and outputting data, the memory is used for storing a computer program, and the processor is used for calling the computer program to enable the computer device to execute the method of any one of claims 1-11.
14. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program adapted to be loaded and executed by a processor to cause a computer device having the processor to perform the method of any of claims 1-11.
CN202110076600.6A 2021-01-20 2021-01-20 Business data processing method and device, computer and readable storage medium Active CN113034171B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110076600.6A CN113034171B (en) 2021-01-20 2021-01-20 Business data processing method and device, computer and readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110076600.6A CN113034171B (en) 2021-01-20 2021-01-20 Business data processing method and device, computer and readable storage medium

Publications (2)

Publication Number Publication Date
CN113034171A CN113034171A (en) 2021-06-25
CN113034171B true CN113034171B (en) 2022-11-18

Family

ID=76460364

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110076600.6A Active CN113034171B (en) 2021-01-20 2021-01-20 Business data processing method and device, computer and readable storage medium

Country Status (1)

Country Link
CN (1) CN113034171B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113448872B (en) * 2021-07-22 2022-04-29 网易（杭州）网络有限公司 Test execution method, device, equipment and storage medium
CN113448876B (en) * 2021-08-31 2021-11-19 腾讯科技（深圳）有限公司 Service testing method, device, computer equipment and storage medium

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109308552A (en) * 2017-07-26 2019-02-05 北京嘀嘀无限科技发展有限公司 A kind of business game appraisal procedure, device and electronic equipment
CN111324533B (en) * 2020-02-17 2022-10-18 支付宝(杭州)信息技术有限公司 A/B test method and device and electronic equipment

Also Published As

Publication number Publication date
CN113034171A (en) 2021-06-25

Similar Documents

Publication Publication Date Title
CN107332902B (en) The user of online customer service system asks distribution method, device and computing device
CN109389431B (en) Method and device for distributing coupons, electronic equipment and readable storage medium
US10783002B1 (en) Cost determination of a service call
CN113034171B (en) Business data processing method and device, computer and readable storage medium
WO2018157808A1 (en) Method and system for evaluating internet credit
CN110287250B (en) User grade quantification method and device
CN112017042A (en) Resource quota determining method and device based on tweed distribution and electronic equipment
CN112148992A (en) Content pushing method and device, computer equipment and storage medium
CN110599240A (en) Application preference value determination method, device and equipment and storage medium
CN110569190B (en) Transaction pressure testing method and device, electronic device and readable storage medium
CN111681112A (en) Method and device for managing release strategy and electronic equipment
CN114298728A (en) Data processing method and related device
CN107507023B (en) Information delivery method and device
CN112884529A (en) Advertisement bidding method, device, equipment and medium
CN109905333B (en) Media information processing method, device and storage medium
CN110288366B (en) Evaluation method and device of resource distribution model
CN113592263A (en) Resource return increment prediction method and device based on dynamic resource return increase ratio
CN113689237A (en) Method and device for determining to-be-launched media resource and media resource processing model
CN109285035B (en) Method, device, equipment and storage medium for predicting application retention data
CN115170167A (en) Advertisement diagnosis method, device, computer equipment and storage medium
KR101818013B1 (en) Advertisement media management method and management apparatus
CN112836971A (en) Quota resource determination method and device, electronic equipment and storage medium
CN112070564A (en) Advertisement pulling method, device and system and electronic equipment
CN112669091A (en) Data processing method, device and storage medium
KR101771537B1 (en) Advertisement media management method and management apparatus

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
REG Reference to a national code

Ref country code: HK

Ref legal event code: DE

Ref document number: 40045495

Country of ref document: HK

GR01 Patent grant
GR01 Patent grant