CN111737646B - Advertisement promotion effect evaluation data processing method, system and storage medium - Google Patents
Advertisement promotion effect evaluation data processing method, system and storage medium Download PDFInfo
- Publication number
- CN111737646B CN111737646B CN202010413678.8A CN202010413678A CN111737646B CN 111737646 B CN111737646 B CN 111737646B CN 202010413678 A CN202010413678 A CN 202010413678A CN 111737646 B CN111737646 B CN 111737646B
- Authority
- CN
- China
- Prior art keywords
- advertisement
- information
- price information
- user
- acquired
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0273—Determination of fees for advertising
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- General Engineering & Computer Science (AREA)
- Marketing (AREA)
- Mathematical Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Mathematical Physics (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Algebra (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an effect evaluation data processing method, system and storage medium for advertisement promotion, wherein the method comprises the following steps: acquiring user information, advertisement channel information, advertisement click times and advertisement exposure times acquired in real time; acquiring advertisement unit price information according to the advertisement channel information; predicting advertisement total price information according to the advertisement unit price information, the advertisement click times and the advertisement exposure times; calculating average income information of each user according to the advertisement total price information and the user information; and evaluating the advertisement promotion effect according to the average income information of each user. The invention can improve the timeliness of data acquisition and the timeliness of advertisement popularization effect evaluation. The invention can be widely applied to the technical field of big data.
Description
Technical Field
The invention relates to the technical field of big data, in particular to an effect evaluation data processing method, system and storage medium for advertisement promotion.
Background
Advertising refers to the promotion of products, services, techniques, cultures, events and the like of the advertising system through newspapers, broadcasts, televisions or networks, so that more people and organizations can know and accept the advertising system, and the purposes of propaganda and popularization are achieved.
With the rapid development of the mobile internet, the market competition of advertisements is also more and more vigorous, various types of APP are more and more difficult to acquire new users needing to conduct advertising, and many advertisements are not covered due to the expensive cost of the topology and the limited advertising income. In the prior art, market personnel monitor the popularization effect through a statistical system, however, a large number of popularization optimizing personnel are needed in the mode, and the efficiency is low. As an important index for evaluating the advertisement putting effect, the investment return rate is used, but the income generated by advertisements in the investment return rate cannot be seen in time and can be calculated after waiting for the second day of data return, so that the timeliness of popularization evaluation is seriously affected.
Disclosure of Invention
In order to solve one of the above technical problems to a certain extent, the present invention aims to: provided are an advertisement promotion effect evaluation data processing method, system and storage medium, which can effectively improve the timeliness of advertisement promotion effect evaluation.
A first aspect of an embodiment of the present invention provides:
the advertisement promotion effect evaluation data processing method comprises the following steps:
acquiring user information, advertisement channel information, advertisement click times and advertisement exposure times acquired in real time;
acquiring advertisement unit price information according to the advertisement channel information;
predicting advertisement total price information according to the advertisement unit price information, the advertisement click times and the advertisement exposure times;
calculating average income information of each user according to the advertisement total price information and the user information;
and evaluating the advertisement promotion effect according to the average income information of each user.
Further, the advertisement total price information is predicted according to the advertisement price information, the advertisement click times and the advertisement exposure times through a prediction model, and the prediction model further comprises the following training steps before prediction:
acquiring user information, advertisement channel information, advertisement click times, advertisement exposure times, advertisement price information and advertisement total price information;
the user information, the advertisement channel information, the advertisement click times, the advertisement exposure times and the advertisement unit price information are used as input information of a prediction model, and the advertisement total price information is used as output information of the prediction model;
training the prediction model through the input information and the output information.
Further, the prediction model includes a linear model and a tree model, and the predicting advertisement total price information according to the advertisement unit price information, the advertisement click times and the advertisement exposure times includes:
predicting first advertisement sub-total price information through a linear model according to the advertisement unit price information, the advertisement click times and the advertisement exposure times;
predicting second advertisement sub-total price information through a tree model according to the advertisement unit price information, the advertisement click times and the advertisement exposure times;
and aggregating the first advertisement sub-total price information and the second advertisement sub-total price information to generate advertisement total price information.
Further, the acquiring user information, advertisement channel information, advertisement click times and advertisement exposure times acquired in real time specifically comprises:
user information, advertisement channel information, advertisement click times and advertisement exposure times which are acquired in real time in a first preset time period are acquired, wherein the first preset time period is a time range smaller than 24 hours.
Further, the acquiring user information, advertisement channel information, advertisement click times and advertisement exposure times acquired in real time specifically comprises:
user information, advertisement channel information, advertisement click times and advertisement exposure times acquired in real time at intervals of a second preset time period.
Further, the acquiring the user information, the advertisement channel information, the advertisement click times and the advertisement exposure times acquired in real time includes:
acquiring user information acquired in real time;
and acquiring the real-time acquired advertisement channel information, advertisement click times and advertisement exposure times corresponding to the user information meeting the preset requirements.
Further, after the step of acquiring the user information, the advertisement channel information, the advertisement click times and the advertisement exposure times acquired in real time, the method further comprises the following steps:
and filtering the user information, the advertisement channel information, the advertisement click times and the advertisement exposure times which are acquired in real time.
A second aspect of an embodiment of the invention provides:
an advertising effectiveness evaluation data processing system, comprising:
the first acquisition module is used for acquiring user information, advertisement channel information, advertisement click times and advertisement exposure times acquired in real time;
the second acquisition module is used for acquiring the advertising unit price information according to the advertising channel information;
the prediction module is used for predicting advertisement total price information according to the advertisement unit price information, the advertisement click times and the advertisement exposure times;
the calculating module is used for calculating average income information of each user according to the advertisement total price information and the user information;
and the evaluation module is used for evaluating the advertisement promotion effect according to the average income information of each user.
A third aspect of embodiments of the present invention provides:
an advertising effectiveness evaluation data processing system, comprising:
at least one memory for storing a program;
and the at least one processor is used for loading the program to execute the advertising promotion effect evaluation data processing method.
A fourth aspect of an embodiment of the invention provides:
a computer readable storage medium having stored therein processor executable instructions which when executed by a processor are for implementing the advertising promotion effectiveness assessment data processing method.
The embodiment of the invention has the beneficial effects that: according to the embodiment of the invention, the user information, the advertisement channel information, the advertisement click times and the advertisement exposure times which are acquired in real time are acquired, the advertisement unit price information is acquired according to the advertisement channel information, then the advertisement total price information is predicted according to the advertisement unit price information, the advertisement click times and the advertisement exposure times, then the average income information of each user is calculated according to the advertisement total price information and the user information, and finally the advertisement popularization effect is estimated according to the average income information of each user, so that the timeliness of data acquisition is improved, and the timeliness of advertisement popularization effect estimation is improved.
Drawings
FIG. 1 is a flowchart of a method for processing effectiveness evaluation data of advertisement promotion according to an embodiment of the present invention;
FIG. 2 is a block diagram of an application system according to one embodiment of the present invention.
Detailed Description
The invention will now be described in further detail with reference to the drawings and to specific examples. The step numbers in the following embodiments are set for convenience of illustration only, and the order between the steps is not limited in any way, and the execution order of the steps in the embodiments may be adaptively adjusted according to the understanding of those skilled in the art.
In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is to be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments and can be combined with one another without conflict.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of the present application only and is not intended to be limiting of the present application.
First, terms appearing in the present application are explained:
ARPU: average revenue per user, which refers to the ratio of revenue generated by a new user of a channel to the number of new people in the channel.
Existing APP advertisements can be classified into show advertisements, search advertisements, and download advertisements by type. Advertisement hit modes of different platforms of different advertisements are different, and common charging modes comprise:
CPM: cost Per Thousand Impression thousands of people affect the cost and are used for pricing information flow advertisements.
CPC: cost Per Click, for showing the pricing of the advertisement.
CPA: cost Per Action, single conversion Cost, user App download, product use receipt and other promotion advertisement price requiring definite conversion Action.
CPT: cost PerTime, cost Per unit Time, is used for pricing of the vertical industry platform display advertisement space, and is similar to outdoor advertisements such as subway advertisements, elevator advertisements and the like.
CPS: the Cost Per sales, sales commission mode is used for discount class sites and promotion class products, and the promotion platform is divided directly according to sales transformation. As a demand side, if the popularization effect can be accurately measured and evaluated, a large amount of exposure cost can be saved, so that the income of the platform is improved as a whole. The income of the advertisement is settled by the advertisement provider, whether the flow is purchased in a channel is determined by predicting the income of the advertisement space channel through an advertisement charging mode and combining the corresponding cost, the predicted income needs to be established with a model of indexes such as income, characteristics and the like, and the income is predicted according to the trained model and characteristic index data, wherein the characteristics comprise the click number, the exposure number, the advertisement space, the advertisement materials and the platform.
Referring to fig. 1, an embodiment of the present invention provides an advertisement promotion effect evaluation data processing method, which may be applied to a control server, where the control server may communicate with a data acquisition end, a data storage end, and a data processing end, respectively.
The implementation comprises the steps of S11-S15:
s11, acquiring user information, advertisement channel information, advertisement click times and advertisement exposure times acquired in real time; the user information, the advertisement channel information, the advertisement click times and the advertisement exposure times in the step are data acquired by the data acquisition terminal for each advertisement promotion APP. The data collected by the data collection end is stored in the data storage end, so that the workload of the data collection end is reduced.
In some embodiments, the acquiring the user information, the advertisement channel information, the advertisement click number and the advertisement exposure number, which are acquired in real time, specifically includes:
acquiring user information acquired in real time; the user information is information which can identify the identity of a viewer by clicking, viewing or browsing corresponding advertisement information on the APP platform.
And acquiring the real-time acquired advertisement channel information, advertisement click times and advertisement exposure times corresponding to the user information meeting the preset requirements. The preset requirement can be used to determine whether the user information is newly added user information in the APP.
According to the embodiment, the corresponding real-time collected advertisement channel information, advertisement click times and advertisement exposure times are obtained through the collected user information meeting the preset requirements, so that the evaluation of the subsequent advertisement popularization effect is more accurate.
In some embodiments, the acquiring the user information, the advertisement channel information, the advertisement click number and the advertisement exposure number, which are acquired in real time, specifically includes:
user information, advertisement channel information, advertisement click times and advertisement exposure times which are acquired in real time in a first preset time period are acquired, wherein the first preset time period is a time range smaller than 24 hours. Specifically, all data in a preset time period before the current moment are acquired. Even if the data in the current preset time period is acquired last time, the data can be acquired again at the present time, so that data omission is avoided.
In some embodiments, the acquiring the user information, the advertisement channel information, the advertisement click number and the advertisement exposure number acquired in real time may further be:
user information, advertisement channel information, advertisement click times and advertisement exposure times acquired in real time at intervals of a second preset time period. The second preset time period may be 5 minutes, 10 minutes, or 15 minutes. In the implementation, after the data acquisition is failed, the data is acquired again in a second preset time period between the time point of the data acquisition and the time point of the data acquisition, so that the data statistics omission caused by overlong interval time is avoided.
In some embodiments, after the step of acquiring the user information, the advertisement channel information, the advertisement click number and the advertisement exposure number acquired in real time, the method further comprises the following steps:
filtering the user information, the advertisement channel information, the advertisement click times and the advertisement exposure times which are acquired in real time. The present implementation specifically filters redundant data to reduce the data throughput of the data storage unit.
S12, acquiring advertising unit price information according to the advertising channel information; the advertising unit price information corresponds to the type of each advertising APP, the advertising type and the advertising mode type. The advertisement channel information comprises the type of advertisement promotion APP, the advertisement type and the advertisement promotion mode type. The advertising unit price information may be ARPU, CPM, CPC, CPA, CPT and CPS.
S13, predicting advertisement total price information according to the advertisement unit price information, the advertisement click times and the advertisement exposure times;
in some embodiments, the advertisement total price information is predicted by a prediction model according to the advertisement price information, the advertisement click times and the advertisement exposure times, and the prediction model further comprises the following training steps before the prediction is performed:
acquiring user information, advertisement channel information, advertisement click times, advertisement exposure times, advertisement price information and advertisement total price information; the user information, the advertisement channel information, the advertisement click times, the advertisement exposure times, the advertisement unit price information and the advertisement total price information in the step can be data acquired in real time, can be data pre-existing at a data storage end, can also be data acquired in real time and can also be data acquired and stored at the data storage end before.
The user information, the advertisement channel information, the advertisement click times, the advertisement exposure times and the advertisement unit price information are used as input information of a prediction model, and the advertisement total price information is used as output information of the prediction model; specifically, the user information, the advertisement channel information, the number of advertisement clicks, the number of advertisement exposures, and the advertisement price information are stored in input subsets of a plurality of training sets, respectively, and the advertisement total price information is stored in output subsets corresponding to the input subsets.
Training the prediction model through the input information and the output information. Specifically, the prediction model is trained through the input subset and the output subset respectively, so that the accuracy of an output result of the prediction model is improved.
In some embodiments, the prediction model includes a linear model and a tree model, and predicts advertisement total price information according to advertisement unit price information, advertisement click times and advertisement exposure times, and specifically includes the following steps:
predicting first advertisement sub-total price information through a linear model according to the advertisement unit price information, the advertisement click times and the advertisement exposure times;
predicting second advertisement sub-total price information through a tree model according to the advertisement unit price information, the advertisement click times and the advertisement exposure times;
and aggregating the first advertisement sub-total price information and the second advertisement sub-total price information to generate advertisement total price information.
In the embodiment, the prediction of the total price information is performed by adopting two modes, and then the two prediction results are fused to be used as the advertisement total price information of the current prediction, so that the prediction error caused by a single prediction model is reduced.
S14, calculating average income information of each user according to the advertisement total price information and the user information; the step is specifically to calculate average income information of each user corresponding to the newly added user.
S15, evaluating the advertisement promotion effect according to the average income information of each user. Specifically, the advertisement promotion effect is evaluated by adding average income information of each user corresponding to the user.
In summary, in the above embodiment, the user information, the advertisement channel information, the advertisement click frequency and the advertisement exposure frequency that are acquired in real time are acquired, the advertisement unit price information is acquired according to the advertisement channel information, then the advertisement total price information is predicted according to the advertisement unit price information, the advertisement click frequency and the advertisement exposure frequency, then the average income information of each user is calculated according to the advertisement total price information and the user information, and finally the advertisement popularization effect is estimated according to the average income information of each user, so that the timeliness of data acquisition is improved, and the timeliness of advertisement popularization effect estimation is improved.
In some embodiments, the above embodiments may also be applied to a system as shown in fig. 2, where the system shown in fig. 2 communicates with a control server, and specifically, the working principle of the system in fig. 2 includes:
the data acquisition terminal is used for acquiring user related information, advertisement channel related information, buried point information of user events and advertisement income related data, wherein the buried point information of the user events comprises click events, exposure events and login events.
The data storage end can use Maxcomputer of the big data processing service of the Arian cloud for constructing a complete data warehouse. Firstly, data collected by a data acquisition end is washed with stepping Maxcomputer to remove abnormal data, and then some coarse-grained data or broad tables are summarized from each dimension, wherein the method specifically comprises the following steps:
in the first step, when the data of the user is less, only some data such as equipment numbers, registration time, mobile phone models and the like are available in the user data, but the user is sourced from channels, the user and the channels are associated to endow the attribute data of the channels to the user, and a user broad table is obtained.
And secondly, aggregating the behavior data, the click number and the exposure number of the user according to the channels to obtain the click number and the exposure number of the granularity of the channels.
The prediction model training end is used for acquiring sample data from the data storage end, training the prediction model mechanical energy through the sample data, wherein the sample data comprises the click number, the exposure number and the income of each advertisement position of each channel, the number of new people in each channel, the creative, the material type, the platform type, the product type and the operating system type of the channel, the income is used as output data of the prediction model, and other information is used as input data of the prediction model.
And the model prediction end is used for acquiring real-time data from the data storage end and predicting advertisement popularization income of the current day through the real-time data. The information of the real-time data is input data of the sample data, namely residual data of the sample data except for income data. In the prediction process, the data is predicted once per hour, in order to ensure the integrity of the data, avoid the delay of the data at the data acquisition end, delay the extraction for one hour, and prevent the data from falling off in each small time slot, and adopt the total data of each extraction day, but the data volume is too large to ensure that the extraction time is too long, so that the data acquisition caliber is reduced to the minimum, namely, only the behavior data related to a new user is extracted. When the time is near the 23-day, the input data is compared with the sample of the day to help analyze the cause of the prediction model error.
In the predictive model training process, the linear model and the tree model are trained respectively. The linear model is similar to the existing scheme, and is assumed that the income is proportional to the click number, except that the linear model of the implementation refers to the click unit price of the previous day and also refers to the click unit price of the previous two days, and more dimensions, such as platform type, product type, advertising creative and the like, are introduced, so that the defects of the existing scheme can be relieved to a certain extent. The tree model is a gradient lifting tree model which is directly trained by using a sample of one week, the last day of sample data of one week is segmented out to be used as a test, the data of the first days are used as training, and rmse and R are adopted 2 The training and testing effect of the model is evaluated by the value, and the error is finally converged through multiple iterations. The linear model has a certain accuracy, but is greatly influenced by fluctuation, is sensitive to empty values, and shows a condition that is accurate for a while and inaccurate for a while. The generalization capability of the tree model is good, the adaptation to fluctuation of input data is good, but the accuracy is not as good as that of the linear model. In the embodiment, two models are fused together, and the advantages are complementary according to the respective characteristics, thereby obtainingIs better than two separate models.
In the model prediction module, firstly, real-time data is downloaded from a data storage end according to partition parameters, if the data is not downloaded, an alarm message is pushed and the data continues to be downloaded after waiting for 5 minutes, the process is repeated for 10 times until the data is downloaded successfully, the data is read to a memory after the data is downloaded successfully, the data is used as input data of a prediction model after certain preprocessing, a model output in a prediction model training end is stored in a text file, the latest model file is loaded in the prediction module, the input data is substituted into the model operation to obtain a prediction result, the result data is converted in a certain format and then is inserted into an Adb database, and meanwhile, a part of the result data is uploaded to a Maxcomputer.
The Adb database is an analytical database, can be fully compatible with MySQL protocol and SQL 2003, can be used like MySQL, integrates the advantages of distributed and elastic calculation and cloud calculation, and is improved in large scale in aspects of scale, usability, reliability, safety and the like so as to meet the requirements of real-time data warehouses in different scenes, and supports larger-scale concurrent access, faster reading and writing capability, intelligent mixed query load management and the like.
In addition, in the system of fig. 2, a timing scheduling and redundant data processing end is also provided, the timing scheduling end is mainly implemented by a system crontab, and the dependency relationship between the called modules is implemented by related logic in respective scripts. The redundant data cleaning can delete data before one week, and because all data can be provided with the date and hour identification, useless data can be cleaned directly according to the identification.
In some embodiments, the application process of the method embodiment in the system shown in fig. 2 is:
firstly, sample data is downloaded, the sample data needs to be preprocessed, the whole column is discarded for the characteristics with more vacancy values, the numerical type characteristics are converted into floating point number types, for example, income, click number, exposure number and number of new people, and the category type characteristics need to be encoded into onehot form, for example, product type, platform type, advertisement position, material type and creative type, because the category type characteristics are qualitative data and cannot be relatively large, and cannot participate in operation, for example, three types of products (A, B and C) are included in the product type, and the product type characteristics are decomposed into 3 characteristics: whether the data is the product A, whether the data is the product B and whether the data is the product C are directly input into the gradient lifting tree model for training by utilizing a lightgbm tool kit, and the trained tree model is stored as a text file and named for standby by date and hour. The linear model is built on the assumption that the income of each advertisement is in linear relation with the clicking, the error of the linear model is influenced by fluctuation of the clicking unit price, and the fluctuation of the clicking unit price does not collect relevant characteristic data; for the granularity of the advertisement position, an index Q is defined, wherein Q=mean (|c (i) -c (i-1) |), c (i) represents the clicking unit price on the i th day, Q is defined as the average value of the absolute value of the difference between the unit price on the i th day and the unit price on the i-1 th day, the fluctuation degree of the unit price of different advertisement positions can be measured by comparing the Q values of different advertisement positions, the Q values of the advertisement positions are dynamically observed, if the absolute value of the deviation of the clicking unit price of a certain advertisement position in the last two days is larger than the Q value, the reference function of the clicking unit price of the advertisement position is discarded, and the tree model is selected for prediction. If the absolute value of the deviation of the click unit price in the near two days is smaller than the Q value, the unit price is selected as the reference unit price to be predicted by using a linear model. Finally, the prediction results of the two models are aggregated according to channels to obtain the prediction result of income so as to improve the defects of the existing models.
The embodiment of the invention provides an advertising promotion effect evaluation data processing system corresponding to the method of FIG. 1, which comprises the following steps:
the first acquisition module is used for acquiring user information, advertisement channel information, advertisement click times and advertisement exposure times acquired in real time;
the second acquisition module is used for acquiring the advertising unit price information according to the advertising channel information;
the prediction module is used for predicting advertisement total price information according to the advertisement unit price information, the advertisement click times and the advertisement exposure times;
the calculating module is used for calculating average income information of each user according to the advertisement total price information and the user information;
and the evaluation module is used for evaluating the advertisement promotion effect according to the average income information of each user.
The content of the method embodiment of the invention is suitable for the system embodiment, the specific function of the system embodiment is the same as that of the method embodiment, and the achieved beneficial effects are the same as those of the method.
The embodiment of the invention provides an effect evaluation data processing system for advertisement promotion, which comprises the following components:
at least one memory for storing a program;
and the at least one processor is used for loading the program to execute the advertising promotion effect evaluation data processing method.
The content of the method embodiment of the invention is suitable for the system embodiment, the specific function of the system embodiment is the same as that of the method embodiment, and the achieved beneficial effects are the same as those of the method.
In addition, the embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium stores instructions executable by a processor, and the instructions executable by the processor are used for realizing the advertising promotion effect evaluation data processing method.
While the preferred embodiment of the present invention has been described in detail, the present invention is not limited to the embodiments described above, and various equivalent modifications and substitutions can be made by those skilled in the art without departing from the spirit of the present invention, and these equivalent modifications and substitutions are intended to be included in the scope of the present invention as defined in the appended claims.
Claims (10)
1. The advertisement promotion effect evaluation data processing method is characterized by comprising the following steps of:
acquiring user information, advertisement channel information, advertisement click times and advertisement exposure times acquired in real time, wherein the advertisement channel information comprises advertisement popularization APP types, advertisement types and advertisement popularization mode types;
acquiring advertising unit price information according to the advertising channel information, wherein the advertising unit price information comprises ARPU, CPM, CPC, CPA, CPT and CPS;
predicting advertisement total price information according to the advertisement unit price information, the advertisement click times and the advertisement exposure times;
calculating average income information of each user according to the advertisement total price information and the user information;
and evaluating the advertisement promotion effect according to the average income information of each user.
2. The method for processing effect evaluation data of advertisement promotion according to claim 1, wherein advertisement total price information is predicted by a prediction model based on the advertisement price information, the advertisement click times and the advertisement exposure times, the prediction model further comprising the training step of:
acquiring user information, advertisement channel information, advertisement click times, advertisement exposure times, advertisement price information and advertisement total price information;
the user information, the advertisement channel information, the advertisement click times, the advertisement exposure times and the advertisement unit price information are used as input information of a prediction model, and the advertisement total price information is used as output information of the prediction model;
training the prediction model through the input information and the output information.
3. The method for processing effect evaluation data of advertisement promotion according to claim 2, wherein the prediction model includes a linear model and a tree model, and the predicting advertisement total price information based on the advertisement price information, the advertisement click times, and the advertisement exposure times includes:
predicting first advertisement sub-total price information through a linear model according to the advertisement unit price information, the advertisement click times and the advertisement exposure times;
predicting second advertisement sub-total price information through a tree model according to the advertisement unit price information, the advertisement click times and the advertisement exposure times;
and aggregating the first advertisement sub-total price information and the second advertisement sub-total price information to generate advertisement total price information.
4. The method for processing the effect evaluation data of advertisement promotion according to claim 1, wherein the acquiring of the user information, the advertisement channel information, the advertisement click times and the advertisement exposure times collected in real time is specifically as follows:
user information, advertisement channel information, advertisement click times and advertisement exposure times which are acquired in real time in a first preset time period are acquired, wherein the first preset time period is a time range smaller than 24 hours.
5. The method for processing advertisement promotion effect evaluation data according to claim 4, wherein the acquiring of the user information, the advertisement channel information, the advertisement click number and the advertisement exposure number acquired in real time specifically comprises:
user information, advertisement channel information, advertisement click times and advertisement exposure times acquired in real time at intervals of a second preset time period.
6. The method for processing advertisement promotion effect evaluation data according to claim 1, wherein the acquiring user information, advertisement channel information, advertisement click times and advertisement exposure times acquired in real time comprises:
acquiring user information acquired in real time;
and acquiring the real-time acquired advertisement channel information, advertisement click times and advertisement exposure times corresponding to the user information meeting the preset requirements.
7. The method for processing advertisement promotion effect evaluation data according to claim 1, further comprising the steps of, after the step of acquiring user information, advertisement channel information, advertisement click number and advertisement exposure number acquired in real time:
and filtering the user information, the advertisement channel information, the advertisement click times and the advertisement exposure times which are acquired in real time.
8. An advertising effectiveness evaluation data processing system, comprising:
the first acquisition module is used for acquiring user information, advertisement channel information, advertisement click times and advertisement exposure times acquired in real time, wherein the advertisement channel information comprises advertisement popularization APP types, advertisement types and advertisement popularization mode types;
the second acquisition module is used for acquiring the advertising unit price information according to the advertising channel information, wherein the advertising unit price information comprises ARPU, CPM, CPC, CPA, CPT and CPS;
the prediction module is used for predicting advertisement total price information according to the advertisement unit price information, the advertisement click times and the advertisement exposure times;
the calculating module is used for calculating average income information of each user according to the advertisement total price information and the user information;
and the evaluation module is used for evaluating the advertisement promotion effect according to the average income information of each user.
9. An advertising effectiveness evaluation data processing system, comprising:
at least one memory for storing a program;
at least one processor for loading the program to perform an advertising effectiveness evaluation data processing method according to any one of claims 1-7.
10. A computer readable storage medium having stored therein processor executable instructions which, when executed by a processor, are for implementing a method of advertisement promotion effectiveness evaluation data processing according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010413678.8A CN111737646B (en) | 2020-05-15 | 2020-05-15 | Advertisement promotion effect evaluation data processing method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010413678.8A CN111737646B (en) | 2020-05-15 | 2020-05-15 | Advertisement promotion effect evaluation data processing method, system and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111737646A CN111737646A (en) | 2020-10-02 |
| CN111737646B true CN111737646B (en) | 2023-04-21 |
Family
ID=72647217
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010413678.8A Active CN111737646B (en) | 2020-05-15 | 2020-05-15 | Advertisement promotion effect evaluation data processing method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111737646B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113269595B (en) * | 2021-06-23 | 2022-01-11 | 南京网觉软件有限公司 | Advertisement delivery method and system based on real-time prediction ROI |
| CN114022186A (en) * | 2021-10-12 | 2022-02-08 | 深圳市思为软件技术有限公司 | Data processing method and related device |
| CN115796959B (en) * | 2022-12-02 | 2023-07-04 | 嘉兴凌巨网络科技有限公司 | Advertisement putting effect detection method based on data acquisition and analysis |
| CN116468486B (en) * | 2023-04-26 | 2023-12-12 | 浙江联欣科技有限公司 | Popularization optimizing management system based on Internet platform |
| CN116739669B (en) * | 2023-08-16 | 2023-10-27 | 成都一心航科技有限公司 | System and method for monitoring ocpx advertisements in real time |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010005259A2 (en) * | 2008-07-10 | 2010-01-14 | 엔에이치엔비지니스플랫폼 주식회사 | A method and system for providing advertisement based on use per time |
| CN109360012A (en) * | 2018-08-22 | 2019-02-19 | 中国平安人寿保险股份有限公司 | The selection method and device, storage medium, electronic equipment of advertisement dispensing channel |
| CN109583721A (en) * | 2018-11-16 | 2019-04-05 | 北京奇虎科技有限公司 | A kind of appraisal procedure of channel, device and electronic equipment |
| CN110866779A (en) * | 2019-10-17 | 2020-03-06 | 上海易点时空网络有限公司 | Data processing method, device, equipment and storage medium based on different channel advertisement delivery |
-
2020
- 2020-05-15 CN CN202010413678.8A patent/CN111737646B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010005259A2 (en) * | 2008-07-10 | 2010-01-14 | 엔에이치엔비지니스플랫폼 주식회사 | A method and system for providing advertisement based on use per time |
| CN109360012A (en) * | 2018-08-22 | 2019-02-19 | 中国平安人寿保险股份有限公司 | The selection method and device, storage medium, electronic equipment of advertisement dispensing channel |
| CN109583721A (en) * | 2018-11-16 | 2019-04-05 | 北京奇虎科技有限公司 | A kind of appraisal procedure of channel, device and electronic equipment |
| CN110866779A (en) * | 2019-10-17 | 2020-03-06 | 上海易点时空网络有限公司 | Data processing method, device, equipment and storage medium based on different channel advertisement delivery |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111737646A (en) | 2020-10-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111737646B (en) | Advertisement promotion effect evaluation data processing method, system and storage medium | |
| US10387900B2 (en) | Methods and apparatus for self-adaptive time series forecasting engine | |
| CN109840782B (en) | Click rate prediction method, device, server and storage medium | |
| CN109657138A (en) | A kind of video recommendation method, device, electronic equipment and storage medium | |
| CN108399564B (en) | Credit scoring method and device | |
| US7698420B2 (en) | Consolidating web visitor behavior statistics for sales prospect qualification | |
| CN111371672A (en) | Message pushing method and device | |
| CN110866698A (en) | Device for assessing service score of service provider | |
| CN111327514A (en) | WeChat image-text message group sending method, system, server and storage medium | |
| CN111414410A (en) | Data processing method, device, equipment and storage medium | |
| CN110163683B (en) | Value user key index determination method, advertisement delivery method and device | |
| CN112395499B (en) | Information recommendation method and device, electronic equipment and storage medium | |
| CN112070564B (en) | Advertisement pulling method, device and system and electronic equipment | |
| CN112734227A (en) | Big data decision system and method | |
| CN114285896B (en) | Information pushing method, device, equipment, storage medium and program product | |
| CN113643061A (en) | Intelligent delivery system based on big data machine learning | |
| CN114595473A (en) | Access data processing method and device, electronic equipment and computer readable medium | |
| CN112686705B (en) | Method and device for predicting sales effect data and electronic equipment | |
| KR20130054691A (en) | Method and server for advertisement by evaluation of user group | |
| CN108959324B (en) | Method and device for estimating multimedia display resource inventory and storage medium | |
| CN111752984B (en) | Information processing method, device and storage medium | |
| CN116308536B (en) | Advertisement flow data acquisition and analysis system | |
| CN109614328B (en) | Method and apparatus for processing test data | |
| CN112752252B (en) | Cell home location identification method and device | |
| CN106959865A (en) | Data acquisition and information-pushing method, device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |