CN113034167A - User interest analysis method and advertisement delivery method based on user behaviors - Google Patents
User interest analysis method and advertisement delivery method based on user behaviors Download PDFInfo
- Publication number
- CN113034167A CN113034167A CN201911348434.XA CN201911348434A CN113034167A CN 113034167 A CN113034167 A CN 113034167A CN 201911348434 A CN201911348434 A CN 201911348434A CN 113034167 A CN113034167 A CN 113034167A
- Authority
- CN
- China
- Prior art keywords
- user
- type
- interest
- advertisement
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
- G06Q30/0244—Optimization
-
- 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
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
- G06Q30/0271—Personalized advertisement
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Biology (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a user interest analysis method and an advertisement delivery method based on user behaviors, and relates to the field of crowd behavior characteristic analysis, wherein the method comprises the following steps: acquiring user data according to business requirements, and synchronously setting advertisement delivery types; step two: collecting user behavior data, and setting an access type, an access interval and a click advertisement type as sample characteristics; step three: taking the sample characteristics as an input layer of the neural network model, taking the advertisement putting type as an output layer of the neural network model, and training the neural network model to obtain an interest classification model; step four: inputting user behavior data to be classified into an interest classification model, and outputting user interest types; step five: and calculating the probability of the interest type of the user by using a Bayesian formula based on the interest type of the user and the advertisement putting type. According to the invention, the user interest is classified according to the user behavior data, and better data support is provided for advertisement putting, website operation and the like.
Description
Technical Field
The invention relates to the field of crowd behavior characteristic analysis, in particular to a user interest analysis method and an advertisement putting method based on user behaviors.
Background
In order to provide a better advertising effect or a data decision basis for advertisement delivery or website operation more accurately, the interest of the user needs to be analyzed, and the win-win situation between the user and the merchant is realized based on the user interest analysis.
At present, as shown in fig. 1, advertisement delivery is mainly performed by counting advertisement position traffic and advertisement effect, setting optimization according to personnel experience, and finally processing according to personnel optimization results; the user information mostly adopts first party data, and the first party data has various problems of inaccurate data, mismatching of data and real user behaviors and the like, so that the data can not be really used or the effect of using the data is poor; in addition, the accuracy of the users is difficult to verify by traditional user classification, so that the final effect is difficult to evaluate, or the effect is improved within 5%, and no good effect is achieved.
Disclosure of Invention
In view of the above-mentioned shortcomings of the prior art, an object of the present invention is to provide a user interest analysis method and an advertisement delivery method based on user behavior, which classify the user interests according to the behavior data of users accessing advertisements, websites, etc., so as to provide better data support for advertisement delivery, website operations, etc., and reduce the costs of advertisement delivery, website operations, etc.
The invention provides a user interest analysis method based on user behaviors, which comprises the following steps:
the method comprises the following steps: acquiring user data according to business requirements, and synchronously setting advertisement delivery types;
step two: collecting user behavior data, and setting an access type, an access interval and a click advertisement type as sample characteristics;
step three: taking the sample characteristics as an input layer of the neural network model, taking the advertisement putting type as an output layer of the neural network model, and training the neural network model to obtain an interest classification model;
step four: inputting user behavior data to be classified into an interest classification model, and outputting user interest types;
step five: and calculating the probability of the interest type of the user by using a Bayesian formula based on the interest type of the user and the advertisement putting type.
In an embodiment of the invention, after the neural network model is trained to obtain the interest classification model, the advertisement delivery type is optimized according to the user interest type.
In an embodiment of the present invention, the user behavior data includes access type data, access interval data, and click advertisement type data for websites and APPs.
In an embodiment of the present invention, the advertisement delivery types include e-commerce, game, finance, and mobile phone.
In an embodiment of the present invention, the neural network model includes an Input layer Input, an hidden layer Hide, and an Output layer Output, where the number of units of the Input layer Input is equal to the sample feature number, and the number of units of the Output layer Output is equal to the advertisement type number.
In an embodiment of the present invention, the neural network model uses a sigmoid function as a stimulus function, where the sigmoid function:
where e is the natural logarithm and x is the input sample characteristic.
In an embodiment of the present invention, the bayesian formula is:
wherein, A is the user interest type output by the interest classification model, and B is the set advertisement putting type.
An advertisement delivery method based on user behavior, the method being implemented based on claims 1-7, the method comprising the steps of:
the method comprises the following steps: calculating the probability of the interest type of the user by using a Bayesian formula based on the interest type of the user and the advertisement putting type;
step two: and judging whether the probability of the interest type of the user meets the interest classification requirement of the user, if so, releasing the corresponding type of advertisement, and otherwise, discarding the corresponding type of advertisement.
As described above, the user interest analysis method based on user behavior of the present invention has the following beneficial effects: according to the method and the system, the user interest classification is carried out according to the user access behavior data, and better data support is provided for advertisement putting, website operation and the like.
Drawings
Fig. 1 shows a flow chart of advertisement delivery disclosed in the prior art of the present invention.
FIG. 2 is a flowchart illustrating a method for obtaining an interest classification model according to an embodiment of the present invention.
Fig. 3 shows a flowchart of advertisement delivery disclosed in the embodiment of the present invention.
Fig. 4 is a schematic structural diagram of a neural network model disclosed in the embodiment of the present invention.
Fig. 5 shows a schematic diagram of the excitation function disclosed in the embodiment of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
As shown in fig. 2, the present invention provides a user interest analysis method based on user behavior, which includes the following steps:
the method comprises the following steps: acquiring user data according to business requirements, and synchronously setting advertisement delivery types;
step two: collecting user behavior data, and setting an access type, an access interval and a click advertisement type as sample characteristics;
step three: taking the sample characteristics as an input layer of the neural network model, taking the advertisement putting type as an output layer of the neural network model, and training the neural network model to obtain an interest classification model;
step four: inputting user behavior data to be classified into an interest classification model, and outputting user interest types;
step five: and calculating the probability of the interest type of the user by using a Bayesian formula based on the interest type of the user and the advertisement putting type.
The user behavior data comprises access type data, access interval data and click advertisement type data of websites and APPs; the categories of advertisement delivery types are selected from the common categories of Internet Architecture Board (IAB), including e-commerce, games, finance, and cell phone, etc.
Specifically, the neural network model comprises an Input layer Input, an hidden layer Hide and an Output layer Output, wherein the unit number of the Input layer Input is equal to the sample characteristic number, and the unit number of the Output layer Output is equal to the advertisement type number;
the Output of the Input layer serves as the Input of the hidden layer Hide, and the Output of the hidden layer Hide serves as the Input of the Output layer Output; as shown in fig. 4, the same formula is:
Ehide＝∑(weight*Eout)*Outhide*(1-Outhide)
wherein, weight is the weight of each sample feature, and the weight is w in fig. 4.
The more the number of units of each hidden layer Hide is, the higher the classification precision is, but the lower the calculation performance is, so the relation between the precision and the performance needs to be synthesized, and the optimization is performed according to the classification result.
As shown in fig. 5, the neural network model uses sigmoid function as excitation function, which is:
where e is the natural logarithm and x is the input sample characteristic.
The specific training parameters of the interest classification model are as follows:
(1) defaulting an automatic initialization weight matrix;
(2) defaults to a perceptron neural network containing an implied layer Hide;
(3) the default number of hidden layer Hide units is 50;
(4) the default learning rate is 0.13;
(5) normalization processing is carried out by default;
(6) the default error precision is controlled to be 0.02;
(7) the default maximum number of iterations is 5000;
finally, training an interest classification model under the condition of meeting the precision; after training the neural network model to obtain an interest classification model, optimizing the advertisement putting type according to the user interest type; namely, the original advertisement putting type can be replaced by the user interest type output by the interest classification model.
As shown in fig. 3, an advertisement delivery method based on user behavior is provided, where when a user accesses a website or APP, there are two attributes of user attribute preference and user type preference, and the user interest type and advertisement delivery type are used as prior distribution and sample information to perform final calculation to obtain posterior distribution, that is, a bayesian formula is used to calculate the probability of the type to which the user belongs (one user may have multiple classifications);
the Bayesian formula is as follows:
wherein, A is the user interest type output by the interest classification model, and B is the set advertisement putting type.
And then judging whether the probability of the interest type of the user meets the interest classification requirement of the user, if so, releasing the corresponding type of advertisement, and otherwise, discarding the corresponding type of advertisement.
The first embodiment is as follows: and in order to verify whether the interest classification of the user is accurate, verifying by adopting an a/b test method.
The scheme a is that operators select the current optimal position for putting;
and b, directly selecting general advertisements for delivery, but selecting the user types required by the project for delivery (e-commerce types), analyzing the e-commerce crowd by using the interest classification model and recommending the delivery.
The final dosing results were as follows:
| click rate (%) | Cost of clicks (yuan) | Cost of thousands of showings (Yuan) | |
| Scheme a | 0.61 | 1.02 | 6.25 |
| Scheme b | 1.68 | 0.27 | 4.47 |
The final putting result shows that the click rate is improved by 175%, the click cost is reduced by 73.5%, and the display cost is reduced by 39.8%.
In conclusion, the invention classifies the user interests according to the user access behavior data, provides better data support for advertisement putting, website operation and the like, and reduces the cost of advertisement putting, website operation and the like. Therefore, the invention effectively overcomes various defects in the prior art and has high industrial utilization value.
The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.
Claims (8)
1. A user interest analysis method based on user behaviors is characterized by comprising the following steps:
the method comprises the following steps: acquiring user data according to business requirements, and synchronously setting advertisement delivery types;
step two: collecting user behavior data, and setting an access type, an access interval and a click advertisement type as sample characteristics;
step three: taking the sample characteristics as an input layer of the neural network model, taking the advertisement putting type as an output layer of the neural network model, and training the neural network model to obtain an interest classification model;
step four: inputting user behavior data to be classified into an interest classification model, and outputting user interest types;
step five: and calculating the probability of the interest type of the user by using a Bayesian formula based on the interest type of the user and the advertisement putting type.
2. The user interest analysis method based on user behavior according to claim 1, wherein: and training the neural network model to obtain an interest classification model, and then optimizing the advertisement putting type according to the user interest type.
3. The user interest analysis method based on user behavior according to claim 1, wherein: the user behavior data comprises access type data, access interval data and click advertisement type data of websites and APPs.
4. The user interest analysis method based on user behavior according to claim 1, wherein: the advertisement delivery types include e-commerce, games, finance and cell phone.
5. The user interest analysis method based on user behavior according to claim 1, wherein: the neural network model comprises an Input layer Input, an hidden layer Hide and an Output layer Output, wherein the unit number of the Input layer Input is equal to the sample characteristic number, and the unit number of the Output layer Output is equal to the advertisement type number.
6. The user interest analysis method based on user behavior according to claim 1 or 5, wherein: the neural network model adopts a sigmoid function as a stimulus function, wherein the sigmoid function is as follows:
where e is the natural logarithm and x is the input sample characteristic.
7. The user interest analysis method based on user behavior according to claim 1, wherein: the Bayesian formula is as follows:
wherein, A is the user interest type output by the interest classification model, and B is the set advertisement putting type.
8. An advertisement delivery method based on user behavior, the method being implemented based on claims 1-7, characterized in that the method comprises the following steps:
the method comprises the following steps: calculating the probability of the interest type of the user by using a Bayesian formula based on the interest type of the user and the advertisement putting type;
step two: and judging whether the probability of the interest type of the user meets the interest classification requirement of the user, if so, releasing the corresponding type of advertisement, and otherwise, discarding the corresponding type of advertisement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911348434.XA CN113034167A (en) | 2019-12-24 | 2019-12-24 | User interest analysis method and advertisement delivery method based on user behaviors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911348434.XA CN113034167A (en) | 2019-12-24 | 2019-12-24 | User interest analysis method and advertisement delivery method based on user behaviors |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113034167A true CN113034167A (en) | 2021-06-25 |
Family
ID=76451823
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911348434.XA Pending CN113034167A (en) | 2019-12-24 | 2019-12-24 | User interest analysis method and advertisement delivery method based on user behaviors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113034167A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114187036A (en) * | 2021-11-30 | 2022-03-15 | 深圳市喂车科技有限公司 | Internet advertisement intelligent recommendation management system based on behavior characteristic recognition |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345512A (en) * | 2013-07-06 | 2013-10-09 | 北京品友互动信息技术有限公司 | Online advertising click-through rate forecasting method and device based on user attribute |
| CN105590240A (en) * | 2015-12-30 | 2016-05-18 | 合一网络技术(北京)有限公司 | Discrete calculating method of brand advertisement effect optimization |
| CN105631711A (en) * | 2015-12-30 | 2016-06-01 | 合一网络技术(北京)有限公司 | Advertisement putting method and apparatus |
| CN109214856A (en) * | 2018-07-27 | 2019-01-15 | 百度在线网络技术（北京）有限公司 | The method for digging and device, computer equipment and readable medium that user is intended to |
| CN109711872A (en) * | 2018-12-14 | 2019-05-03 | 中国平安人寿保险股份有限公司 | Advertisement placement method and device based on big data analysis |
| CN109816421A (en) * | 2018-12-13 | 2019-05-28 | 深圳壹账通智能科技有限公司 | Advertisement machine launches contents controlling method, device, computer equipment and storage medium |
| CN109858972A (en) * | 2019-02-13 | 2019-06-07 | 重庆金窝窝网络科技有限公司 | The prediction technique and device of ad click rate |
-
2019
- 2019-12-24 CN CN201911348434.XA patent/CN113034167A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345512A (en) * | 2013-07-06 | 2013-10-09 | 北京品友互动信息技术有限公司 | Online advertising click-through rate forecasting method and device based on user attribute |
| CN105590240A (en) * | 2015-12-30 | 2016-05-18 | 合一网络技术(北京)有限公司 | Discrete calculating method of brand advertisement effect optimization |
| CN105631711A (en) * | 2015-12-30 | 2016-06-01 | 合一网络技术(北京)有限公司 | Advertisement putting method and apparatus |
| CN109214856A (en) * | 2018-07-27 | 2019-01-15 | 百度在线网络技术（北京）有限公司 | The method for digging and device, computer equipment and readable medium that user is intended to |
| CN109816421A (en) * | 2018-12-13 | 2019-05-28 | 深圳壹账通智能科技有限公司 | Advertisement machine launches contents controlling method, device, computer equipment and storage medium |
| CN109711872A (en) * | 2018-12-14 | 2019-05-03 | 中国平安人寿保险股份有限公司 | Advertisement placement method and device based on big data analysis |
| CN109858972A (en) * | 2019-02-13 | 2019-06-07 | 重庆金窝窝网络科技有限公司 | The prediction technique and device of ad click rate |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114187036A (en) * | 2021-11-30 | 2022-03-15 | 深圳市喂车科技有限公司 | Internet advertisement intelligent recommendation management system based on behavior characteristic recognition |
| CN114187036B (en) * | 2021-11-30 | 2022-10-11 | 深圳市喂车科技有限公司 | Internet advertisement intelligent recommendation management system based on behavior characteristic recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107391760B (en) | User interest recognition methods, device and computer readable storage medium | |
| WO2020135535A1 (en) | Recommendation model training method and related apparatus | |
| CN110622196B (en) | Evaluating a model dependent on aggregated historical data | |
| CN104866969A (en) | Personal credit data processing method and device | |
| CN110753920A (en) | System and method for optimizing and simulating web page ranking and traffic | |
| WO2018149337A1 (en) | Information distribution method, device, and server | |
| CN107766929A (en) | model analysis method and device | |
| CN106776925B (en) | Method, server and system for predicting gender of mobile terminal user | |
| CN108764332A (en) | A kind of Channel Quality analysis method, computing device and storage medium | |
| CN107633257B (en) | Data quality evaluation method and device, computer readable storage medium and terminal | |
| WO2019108810A1 (en) | Display and analysis system for media content | |
| CN111695084A (en) | Model generation method, credit score generation method, device, equipment and storage medium | |
| CN111797320A (en) | Data processing method, device, equipment and storage medium | |
| Atienza-Barthelemy et al. | Relationship between ideology and language in the Catalan independence context | |
| CN115345530A (en) | Market address recommendation method, device and equipment and computer readable storage medium | |
| CN109614982A (en) | Product analysis method, apparatus, computer equipment and storage medium | |
| Hilman et al. | Socioeconomic biases in urban mixing patterns of US metropolitan areas | |
| CN118014622B (en) | Advertisement pushing method and system based on user portrait | |
| CN111754287A (en) | Article screening method, apparatus, device and storage medium | |
| CN114782123A (en) | Credit assessment method and system | |
| CN105389714B (en) | Method for identifying user characteristics from behavior data | |
| KR101028810B1 (en) | Apparatus and method for analyzing advertisement target | |
| CN103854206A (en) | Method and device for analyzing group characteristics | |
| CN113034167A (en) | User interest analysis method and advertisement delivery method based on user behaviors | |
| CN112686690A (en) | Data processing method and device, electronic equipment and computer readable storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210625 |
|
| RJ01 | Rejection of invention patent application after publication |