CN105912686A - Search engine marketing bid method and system based on machine learning - Google Patents
Search engine marketing bid method and system based on machine learning Download PDFInfo
- Publication number
- CN105912686A CN105912686A CN201610240342.XA CN201610240342A CN105912686A CN 105912686 A CN105912686 A CN 105912686A CN 201610240342 A CN201610240342 A CN 201610240342A CN 105912686 A CN105912686 A CN 105912686A
- Authority
- CN
- China
- Prior art keywords
- user
- characteristic value
- advertisement
- conversion rate
- participle
- 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
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000010801 machine learning Methods 0.000 title claims abstract description 37
- 238000006243 chemical reaction Methods 0.000 claims abstract description 44
- 238000005516 engineering process Methods 0.000 claims abstract description 19
- 238000004364 calculation method Methods 0.000 claims description 12
- 230000011218 segmentation Effects 0.000 claims description 12
- 238000012216 screening Methods 0.000 claims description 9
- 238000007477 logistic regression Methods 0.000 claims description 8
- 238000003058 natural language processing Methods 0.000 claims description 4
- 238000012847 principal component analysis method Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 230000008878 coupling Effects 0.000 abstract 1
- 238000010168 coupling process Methods 0.000 abstract 1
- 238000005859 coupling reaction Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/955—Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a search engine marketing bid method and system based on machine learning; the method comprises the following steps: using the machine learning technology to process user history search information and user click advertisement trace information in advance, thus obtaining training data; using training data to train a model, thus obtaining a prediction model used for predicting a keyword conversion rate, wherein the user history search information comprises user search keywords and searched advertisement documents; using the prediction module to calculate an object keyword conversion rate; using an object keyword cost value and conversion rate to calculate, thus obtaining an object keyword bid. The method can improve coupling level between the keyword bid and advertisement popularization efficiency, thus improving advertisement investment return rate of advertisers.
Description
Technical Field
The invention relates to the technical field of search engine marketing, in particular to a search engine marketing bidding method and a search engine marketing bidding system based on machine learning.
Background
Search engine bid advertising is currently the most successful business model in internet advertising, and the core technology behind it is bid keyword technology. Advertisers can promote flag-related products or services on internet search engines by purchasing keywords related to their promotional content on the search engines.
However, after an advertiser purchases a keyword at a certain price, the final advertisement promotion efficiency often cannot reach the degree matching with the keyword bid, that is, the return on investment of the advertisement is low.
In summary, it can be seen that how to improve the matching degree between the keyword bid price and the advertisement promotion efficiency so as to improve the advertisement investment return rate of the advertiser is a problem to be solved at present.
Disclosure of Invention
In view of this, the present invention provides a search engine marketing bidding method and system based on machine learning, which improves the matching degree between the keyword bidding and the advertisement promotion efficiency, thereby improving the return rate of advertisement investment of advertisers. The specific scheme is as follows:
a search engine marketing bidding method based on machine learning comprises the following steps:
processing historical search information of a user and trace information of advertisement clicking of the user by utilizing a machine learning technology in advance to obtain training data, and performing model training by utilizing the training data to obtain a prediction model for predicting the conversion rate of the keywords; wherein the user history search information includes a user search keyword and a searched advertisement document;
calculating the conversion rate of the target keyword by using the prediction model;
and calculating the bid of the target keyword by using the cost value of the target keyword and the conversion rate.
Preferably, the search engine marketing bidding method further includes:
and updating the historical search information and the trace information of the user regularly according to a preset data updating period, and performing model training again by using the updated historical search information and trace information of the user so as to update the prediction model.
Preferably, the process of processing the historical search information of the user and the trace information of the advertisement clicked by the user by using the machine learning technology in advance to obtain the training data includes:
performing word segmentation processing on the advertisement documents in the historical search information of the user by using a natural language processing technology in advance, and performing data dimension reduction processing on all the word segments obtained after the word segmentation processing by using a principal component analysis method to obtain a corresponding word segment set;
determining a characteristic value corresponding to each participle in the participle set according to the trace information and a preset characteristic value determination rule to obtain a corresponding characteristic value set;
and screening the characteristic value set to obtain N characteristic values, determining the weight corresponding to each characteristic value in the N characteristic values, and determining the N characteristic values and the corresponding participles and weights as the training data, wherein N is a positive integer.
Preferably, the process of screening the feature value set to obtain N feature values and determining the weight corresponding to each feature value in the N feature values includes:
performing logistic regression on each characteristic value in the characteristic value set separately to obtain an initial weight corresponding to each characteristic value;
screening the N characteristic values from the characteristic value set, wherein the initial weight corresponding to each characteristic value in the N characteristic values is not less than the initial weights corresponding to the rest characteristic values in the characteristic value set;
and performing logistic regression on the N characteristic values again to obtain the weight corresponding to each characteristic value in the N characteristic values.
Preferably, the preset feature value determination rule is as follows:
and if the trace information shows that any participle can bring effective click to a corresponding advertisement document, determining the characteristic value of the participle as 1, otherwise, determining the characteristic value of the participle as 0.
Preferably, the preset feature value determination rule is as follows:
if the trace information shows that any participle can bring effective click to a corresponding advertisement document, and the time for a user to browse the advertisement document is not less than the preset time, determining the characteristic value of the participle as 1;
if the trace information shows that any participle can bring effective click to a corresponding advertisement document, and the time for a user to browse the advertisement document is less than the preset time, determining the characteristic value of the participle to be 0.5;
and if the trace information shows that any participle cannot bring effective click to the corresponding advertisement document, determining the characteristic value of the participle as 0.
Preferably, the prediction model is:
wherein,fi(A) representing the ith feature value, w, in the ad document AiRepresenting the weight of the ith eigenvalue.
Preferably, in the process of calculating the bid price of the target keyword by using the cost value of the target keyword and the conversion rate, the corresponding calculation formula is as follows:
Pj＝Cj*CTR′；
wherein, PjRepresenting the corresponding bid of the target keyword in the j hour in any day; cjThe cost value corresponding to the target keyword in the j hour in any day time is shown, CTR' represents the conversion rate of the target keyword, and j ∈ {1, 2.
Preferably, CjThe calculation formula of (2) is as follows:
Cj＝b*αj；
where b represents the amount of spending the user is willing to incur for a valid click αjRepresents a time cost, wherein αjThe concrete expression is as follows:
wherein N isjRepresents the effective click volume within the j hour of any day, and T represents the effective click volume throughout the day.
The invention also discloses a search engine marketing bidding system based on machine learning, which comprises the following steps:
the prediction model construction module is used for processing historical search information of a user and trace information of advertisement clicking of the user by utilizing a machine learning technology in advance to obtain training data, and performing model training by utilizing the training data to obtain a prediction model for predicting the keyword conversion rate; wherein the user history search information includes a user search keyword and a searched advertisement document;
the conversion rate calculation module is used for calculating the conversion rate of the target keyword by utilizing the prediction model;
and the bid calculation module is used for calculating the bid of the target keyword by using the cost value of the target keyword and the conversion rate.
The search engine marketing bidding method comprises the following steps: processing historical search information of a user and trace information of advertisement clicking of the user by utilizing a machine learning technology in advance to obtain training data, and performing model training by utilizing the training data to obtain a prediction model for predicting the conversion rate of the keywords; the user historical search information comprises user search keywords and searched advertisement documents; calculating the conversion rate of the target keyword by using the prediction model; and calculating the bid of the target keyword by using the cost value and the conversion rate of the target keyword. Therefore, the training data for model training in the invention is obtained by processing the historical search information of the user and the trace information of the advertisement clicked by the user by using the machine learning technology, and the prediction model for predicting the keyword conversion rate can be obtained by performing model training on the training data. When the bid of the target keyword needs to be calculated, the conversion rate of the target keyword can be predicted by utilizing the prediction model, and then the bid of the target keyword is calculated by combining the cost of the target keyword, so that the bid of the target keyword is calculated, the influence of the conversion rate of the target keyword on the bid is considered when the bid of the target keyword is calculated, and the conversion rate of the keyword is an important index for measuring the advertisement promotion efficiency, so that the bid of the target keyword calculated by the method can be well matched with the advertisement promotion efficiency, namely, the matching degree between the bid of the keyword and the advertisement promotion efficiency is improved, and the advertisement investment return rate of an advertiser is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention 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 embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flowchart of a method for marketing and bidding for a search engine based on machine learning according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a search engine marketing bidding system based on machine learning according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, 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 invention.
The embodiment of the invention discloses a search engine marketing bidding method based on machine learning, which is shown in figure 1 and comprises the following steps:
step S11: processing historical search information of a user and trace information of advertisement clicking of the user by utilizing a machine learning technology in advance to obtain training data, and performing model training by utilizing the training data to obtain a prediction model for predicting the conversion rate of the keywords; wherein the user history search information includes the user search keyword and the searched advertisement document.
It should be noted that the user history search information includes search keywords input by the user and corresponding searched advertisement documents. The trace information of the user clicking the advertisement comprises a corresponding search keyword, a clicking event of the advertisement document and browsing duration of the advertisement document. The keyword conversion rate is an effective click rate of an advertisement document corresponding to the keyword.
Step S12: and calculating the conversion rate of the target keyword by using the prediction model.
Step S13: and calculating the bid of the target keyword by using the cost value and the conversion rate of the target keyword.
In addition, in order to make the prediction model adapt to the user searching habit and advertisement browsing habit which may change with time, the present invention may further include: and updating the historical search information of the user and the trace information of the advertisement clicked by the user regularly according to a preset data updating period, and performing model training again by using the updated historical search information and trace information of the user so as to update the prediction model. Namely, the current latest historical search information and trace information of the user are regularly collected, and then the model is retrained by using the information, so that the prediction model is continuously updated.
Therefore, in the embodiment of the invention, the training data for model training is obtained by processing the historical search information of the user and the trace information of the advertisement clicked by the user by using the machine learning technology, and the prediction model for predicting the keyword conversion rate can be obtained by performing model training on the training data. When the bid of the target keyword needs to be calculated, the conversion rate of the target keyword can be predicted by utilizing the prediction model, and then the bid of the target keyword is calculated by combining the cost of the target keyword, so that the bid of the target keyword is obtained by considering the influence of the conversion rate of the target keyword on the bid when the bid of the target keyword is calculated.
The embodiment of the invention discloses a specific search engine marketing bidding method based on machine learning, and compared with the previous embodiment, the embodiment further explains and optimizes the technical scheme. Specifically, the method comprises the following steps:
in step S11 of the previous embodiment, the process of processing the historical search information of the user and the trace information of the advertisement clicked by the user by using the machine learning technique in advance to obtain the training data specifically includes:
step S111: performing word segmentation processing on advertisement documents in the historical search information of the user by using a natural language processing technology in advance, and performing data dimension reduction processing on all word segments obtained after the word segmentation processing by using a principal component analysis method to obtain a corresponding word segment set;
step S112: determining a characteristic value corresponding to each participle in a participle set according to trace information of clicking the advertisement by a user and a preset characteristic value determination rule to obtain a corresponding characteristic value set;
step S113: and screening the characteristic value set to obtain N characteristic values, determining the weight corresponding to each characteristic value in the N characteristic values, and determining the N characteristic values and the corresponding participles and weights as training data, wherein N is a positive integer.
In the step S111, when performing word segmentation processing by using natural language processing technology, the same word segmentation scheme as that of chinese courtyard word segmentation may be adopted. In addition, during the word segmentation process, stop words can be filtered to delete certain fictional words and linguistic words like "the word", and the like, so that data noise is reduced.
The training data includes not only the N feature values, but also the word segmentation corresponding to each feature value of the N feature values and the weight corresponding to each word segmentation. The N may be a value manually set according to actual needs, and is not particularly limited herein.
Further, in step S113, the process of screening the feature value set to obtain N feature values, and determining the weight corresponding to each feature value in the N feature values includes:
step S1131: and performing logistic regression on each characteristic value in the characteristic value set separately to obtain an initial weight corresponding to each characteristic value.
Step S1132: and screening N characteristic values from the characteristic value set, wherein the initial weight corresponding to each characteristic value in the N characteristic values is not less than the initial weight corresponding to the rest characteristic values in the characteristic value set. That is, the N feature values are the first N largest feature values in the feature value set.
Step S1133: and performing logistic regression on the N characteristic values again to obtain the weight corresponding to each characteristic value in the N characteristic values.
It should be noted that, in step S1131, each feature value in the feature value set is subjected to a logistic regression separately, and in step S1133, the N feature values are subjected to logistic regression simultaneously.
In addition, the preset characteristic value determination rule can be set differently according to different actual requirements.
For example, the preset feature value determination rule may be: and if the trace information shows that any participle can bring effective click to the corresponding advertisement document, determining the characteristic value of the participle as 1, otherwise, determining the characteristic value of the participle as 0.
For another example, the preset feature value determination rule may be: if the trace information shows that any participle can bring effective click to a corresponding advertisement document, and the time for a user to browse the advertisement document is not less than the preset time, determining the characteristic value of the participle as 1; if the trace information shows that any participle can bring effective click to a corresponding advertisement document, and the time for a user to browse the advertisement document is less than the preset time, determining the characteristic value of the participle to be 0.5; and if the trace information shows that any participle cannot bring effective click to the corresponding advertisement document, determining the characteristic value of the participle as 0.
The prediction model in this embodiment is specifically:
wherein,fi(A) representing the ith feature value, w, in the ad document AiRepresenting the weight of the ith eigenvalue.
In step S13 of the previous embodiment, in the process of calculating the bid price of the target keyword by using the cost value and the conversion rate of the target keyword, the corresponding calculation formula is:
Pj＝CjCTR'; wherein,
Pjrepresents the corresponding bid of the target keyword, j ∈ {1, 2.., 24} in the j hour of any day;
Cjthe cost value corresponding to the target keyword in the j hour in any day time is shown;
CTR' represents the conversion rate of the target keyword.
Further, CjThe calculation formula of (2) is as follows: cj＝b*αj；
Where b represents the amount of spending the user is willing to incur for a valid click αjRepresents a time cost, wherein αjThe concrete expression is as follows:
wherein N isjRepresents the effective click volume within the j hour of any day, and T represents the effective click volume throughout the day.
Correspondingly, the embodiment of the invention also discloses a search engine marketing and bidding system based on machine learning, and as shown in fig. 2, the system includes:
the prediction model building module 21 is configured to pre-process the historical search information of the user and the trace information of the advertisement clicked by the user by using a machine learning technology to obtain training data, and perform model training by using the training data to obtain a prediction model for predicting the keyword conversion rate; the user historical search information comprises user search keywords and searched advertisement documents;
a conversion rate calculation module 22, configured to calculate a conversion rate of the target keyword using the prediction model;
and the bid calculation module 23 is configured to calculate a bid for the target keyword by using the cost value and the conversion rate of the target keyword.
For more specific working processes of the modules, reference may be made to relevant contents in the foregoing embodiments, and details are not repeated here.
Therefore, in the embodiment of the invention, the training data for model training is obtained by processing the historical search information of the user and the trace information of the advertisement clicked by the user by using the machine learning technology, and the prediction model for predicting the keyword conversion rate can be obtained by performing model training on the training data. When the bid of the target keyword needs to be calculated, the conversion rate of the target keyword can be predicted by utilizing the prediction model, and then the bid of the target keyword is calculated by combining the cost of the target keyword, so that the bid of the target keyword is obtained by considering the influence of the conversion rate of the target keyword on the bid when the bid of the target keyword is calculated.
Finally, it should also be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The search engine marketing bidding method and system based on machine learning provided by the invention are introduced in detail, a specific example is applied in the text to explain the principle and the implementation mode of the invention, and the description of the embodiment is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A search engine marketing bidding method based on machine learning is characterized by comprising the following steps:
processing historical search information of a user and trace information of advertisement clicking of the user by utilizing a machine learning technology in advance to obtain training data, and performing model training by utilizing the training data to obtain a prediction model for predicting the conversion rate of the keywords; wherein the user history search information includes a user search keyword and a searched advertisement document;
calculating the conversion rate of the target keyword by using the prediction model;
and calculating the bid of the target keyword by using the cost value of the target keyword and the conversion rate.
2. The machine-learning based search engine marketing bidding method of claim 1, further comprising:
and updating the historical search information and the trace information of the user regularly according to a preset data updating period, and performing model training again by using the updated historical search information and trace information of the user so as to update the prediction model.
3. The machine learning-based search engine marketing bidding method according to claim 1, wherein the process of processing historical search information of the user and trace information of advertisement clicking of the user by using a machine learning technology in advance to obtain training data comprises:
performing word segmentation processing on the advertisement documents in the historical search information of the user by using a natural language processing technology in advance, and performing data dimension reduction processing on all the word segments obtained after the word segmentation processing by using a principal component analysis method to obtain a corresponding word segment set;
determining a characteristic value corresponding to each participle in the participle set according to the trace information and a preset characteristic value determination rule to obtain a corresponding characteristic value set;
and screening the characteristic value set to obtain N characteristic values, determining the weight corresponding to each characteristic value in the N characteristic values, and determining the N characteristic values and the corresponding participles and weights as the training data, wherein N is a positive integer.
4. The machine-learning-based search engine marketing bidding method according to claim 3, wherein the process of screening the feature value set to obtain N feature values and determining the weight corresponding to each feature value of the N feature values comprises:
performing logistic regression on each characteristic value in the characteristic value set separately to obtain an initial weight corresponding to each characteristic value;
screening the N characteristic values from the characteristic value set, wherein the initial weight corresponding to each characteristic value in the N characteristic values is not less than the initial weights corresponding to the rest characteristic values in the characteristic value set;
and performing logistic regression on the N characteristic values again to obtain the weight corresponding to each characteristic value in the N characteristic values.
5. The machine-learning-based search engine marketing bidding method according to claim 3, wherein the preset feature value determination rule is:
and if the trace information shows that any participle can bring effective click to a corresponding advertisement document, determining the characteristic value of the participle as 1, otherwise, determining the characteristic value of the participle as 0.
6. The machine-learning-based search engine marketing bidding method according to claim 3, wherein the preset feature value determination rule is:
if the trace information shows that any participle can bring effective click to a corresponding advertisement document, and the time for a user to browse the advertisement document is not less than the preset time, determining the characteristic value of the participle as 1;
if the trace information shows that any participle can bring effective click to a corresponding advertisement document, and the time for a user to browse the advertisement document is less than the preset time, determining the characteristic value of the participle to be 0.5;
and if the trace information shows that any participle cannot bring effective click to the corresponding advertisement document, determining the characteristic value of the participle as 0.
7. The machine-learning based search engine marketing bidding method of claim 3, wherein the predictive model is:
wherein,fi(A) representing the ith feature value, w, in the ad document AiRepresenting the weight of the ith eigenvalue.
8. The machine-learning-based search engine marketing bidding method according to any one of claims 1 to 7, wherein in the process of calculating the bid for the target keyword by using the cost value of the target keyword and the conversion rate, the corresponding calculation formula is as follows:
Pj＝Cj*CTR′；
wherein, PjRepresenting the corresponding bid of the target keyword in the j hour in any day; cjThe cost value corresponding to the target keyword in the j hour in any day time is shown, CTR' represents the conversion rate of the target keyword, and j ∈ {1, 2.
9. The machine-learning based search engine marketing bidding method of claim 8, wherein C isjThe calculation formula of (2) is as follows:
Cj＝b*αj；
where b represents the amount of spending the user is willing to incur for a valid click αjRepresents a time cost, wherein αjIn particular toExpressed as:
wherein N isjRepresents the effective click volume within the j hour of any day, and T represents the effective click volume throughout the day.
10. A search engine marketing bidding system based on machine learning, comprising:
the prediction model construction module is used for processing historical search information of a user and trace information of advertisement clicking of the user by utilizing a machine learning technology in advance to obtain training data, and performing model training by utilizing the training data to obtain a prediction model for predicting the keyword conversion rate; wherein the user history search information includes a user search keyword and a searched advertisement document;
the conversion rate calculation module is used for calculating the conversion rate of the target keyword by utilizing the prediction model;
and the bid calculation module is used for calculating the bid of the target keyword by using the cost value of the target keyword and the conversion rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610240342.XA CN105912686A (en) | 2016-04-18 | 2016-04-18 | Search engine marketing bid method and system based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610240342.XA CN105912686A (en) | 2016-04-18 | 2016-04-18 | Search engine marketing bid method and system based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105912686A true CN105912686A (en) | 2016-08-31 |
Family
ID=56747259
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610240342.XA Pending CN105912686A (en) | 2016-04-18 | 2016-04-18 | Search engine marketing bid method and system based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105912686A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106649605A (en) * | 2016-11-28 | 2017-05-10 | 百度在线网络技术（北京）有限公司 | Triggering way and device of promoting key words |
| CN106844178A (en) * | 2017-01-22 | 2017-06-13 | 腾云天宇科技（北京）有限公司 | Prediction is presented method, computing device, server and the system of information transferring rate |
| CN108288171A (en) * | 2017-12-13 | 2018-07-17 | 努比亚技术有限公司 | Advertisement insertion, server and computer readable storage medium |
| CN108667875A (en) * | 2017-03-31 | 2018-10-16 | 北京京东尚科信息技术有限公司 | information updating method and device |
| CN108874794A (en) * | 2017-05-08 | 2018-11-23 | 广东原昇信息科技有限公司 | A kind of real-time SEM keyword changing effect analysis system and its analysis method |
| CN109272356A (en) * | 2018-09-19 | 2019-01-25 | 电子科技大学 | Optimal Bidding Strategies method based on AdWords |
| CN109583952A (en) * | 2018-11-28 | 2019-04-05 | 深圳前海微众银行股份有限公司 | Advertising Copy processing method, device, equipment and computer readable storage medium |
| CN110110187A (en) * | 2018-01-23 | 2019-08-09 | 优估（上海）信息科技有限公司 | A kind of price competing method of search engine, apparatus and system |
| CN111125574A (en) * | 2018-10-31 | 2020-05-08 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating information |
| CN112445968A (en) * | 2019-09-03 | 2021-03-05 | 百度(中国)有限公司 | Information pushing method, device, equipment and computer readable storage medium |
| CN112990986A (en) * | 2021-04-29 | 2021-06-18 | 广州宜推网络科技有限公司 | Advertisement service information publishing system based on block chain |
| CN114415890A (en) * | 2022-01-20 | 2022-04-29 | 北京大学第三医院（北京大学第三临床医学院） | Interface customization method of doctor evaluation system and teacher-side platform |
| TWI793170B (en) * | 2017-10-20 | 2023-02-21 | 美商雅虎廣告技術有限責任公司 | System, devices, and method for automated bidding using deep neural language models |
| CN119417532A (en) * | 2024-10-14 | 2025-02-11 | 音信云商服(武汉)科技有限公司 | A digital advertising delivery method, system and device based on big data analysis |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050137939A1 (en) * | 2003-12-19 | 2005-06-23 | Palo Alto Research Center Incorporated | Server-based keyword advertisement management |
| CN102479190A (en) * | 2010-11-22 | 2012-05-30 | 阿里巴巴集团控股有限公司 | Estimation value prediction method and device for search keywords |
| CN104252680A (en) * | 2013-06-28 | 2014-12-31 | 麦奇数位股份有限公司 | Keyword automatic pricing method and search engine marketing system |
| CN104731818A (en) * | 2013-12-24 | 2015-06-24 | 苏州开眼数据技术有限公司 | Keyword optimization method and keyword optimization device |
| CN105431875A (en) * | 2013-08-05 | 2016-03-23 | 雅虎公司 | Search engine marketing optimizer |
-
2016
- 2016-04-18 CN CN201610240342.XA patent/CN105912686A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050137939A1 (en) * | 2003-12-19 | 2005-06-23 | Palo Alto Research Center Incorporated | Server-based keyword advertisement management |
| CN102479190A (en) * | 2010-11-22 | 2012-05-30 | 阿里巴巴集团控股有限公司 | Estimation value prediction method and device for search keywords |
| CN104252680A (en) * | 2013-06-28 | 2014-12-31 | 麦奇数位股份有限公司 | Keyword automatic pricing method and search engine marketing system |
| CN105431875A (en) * | 2013-08-05 | 2016-03-23 | 雅虎公司 | Search engine marketing optimizer |
| CN104731818A (en) * | 2013-12-24 | 2015-06-24 | 苏州开眼数据技术有限公司 | Keyword optimization method and keyword optimization device |
Non-Patent Citations (2)
| Title |
|---|
| MATTHEW RICHARDSON等: ""Predicting clicks:estimating the click-through rate for new ads"", 《PROCEEDINGS OF THE 16TH INTERNATIONAL CONFERENCE ON WORLD WIDE WEB》 * |
| 张博: ""基于机器学习的关键词竞价系统的研究"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106649605A (en) * | 2016-11-28 | 2017-05-10 | 百度在线网络技术（北京）有限公司 | Triggering way and device of promoting key words |
| CN106844178A (en) * | 2017-01-22 | 2017-06-13 | 腾云天宇科技（北京）有限公司 | Prediction is presented method, computing device, server and the system of information transferring rate |
| CN106844178B (en) * | 2017-01-22 | 2019-11-15 | 腾云天宇科技（北京）有限公司 | Prediction is presented the method for information transferring rate, calculates equipment, server and system |
| CN108667875B (en) * | 2017-03-31 | 2021-03-30 | 北京京东尚科信息技术有限公司 | Information updating method and device |
| CN108667875A (en) * | 2017-03-31 | 2018-10-16 | 北京京东尚科信息技术有限公司 | information updating method and device |
| CN108874794A (en) * | 2017-05-08 | 2018-11-23 | 广东原昇信息科技有限公司 | A kind of real-time SEM keyword changing effect analysis system and its analysis method |
| TWI793170B (en) * | 2017-10-20 | 2023-02-21 | 美商雅虎廣告技術有限責任公司 | System, devices, and method for automated bidding using deep neural language models |
| CN108288171A (en) * | 2017-12-13 | 2018-07-17 | 努比亚技术有限公司 | Advertisement insertion, server and computer readable storage medium |
| CN110110187B (en) * | 2018-01-23 | 2021-07-16 | 优轩(北京)信息科技有限公司 | Bidding method, device and system of search engine |
| CN110110187A (en) * | 2018-01-23 | 2019-08-09 | 优估（上海）信息科技有限公司 | A kind of price competing method of search engine, apparatus and system |
| CN109272356A (en) * | 2018-09-19 | 2019-01-25 | 电子科技大学 | Optimal Bidding Strategies method based on AdWords |
| CN111125574A (en) * | 2018-10-31 | 2020-05-08 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating information |
| CN111125574B (en) * | 2018-10-31 | 2023-04-28 | 北京字节跳动网络技术有限公司 | Method and device for generating information |
| CN109583952A (en) * | 2018-11-28 | 2019-04-05 | 深圳前海微众银行股份有限公司 | Advertising Copy processing method, device, equipment and computer readable storage medium |
| CN112445968A (en) * | 2019-09-03 | 2021-03-05 | 百度(中国)有限公司 | Information pushing method, device, equipment and computer readable storage medium |
| CN112445968B (en) * | 2019-09-03 | 2023-04-21 | 百度(中国)有限公司 | Information pushing method, device, equipment and computer readable storage medium |
| CN112990986B (en) * | 2021-04-29 | 2021-07-16 | 广州宜推网络科技有限公司 | Advertisement service information publishing system based on block chain |
| CN112990986A (en) * | 2021-04-29 | 2021-06-18 | 广州宜推网络科技有限公司 | Advertisement service information publishing system based on block chain |
| CN114415890A (en) * | 2022-01-20 | 2022-04-29 | 北京大学第三医院（北京大学第三临床医学院） | Interface customization method of doctor evaluation system and teacher-side platform |
| CN119417532A (en) * | 2024-10-14 | 2025-02-11 | 音信云商服(武汉)科技有限公司 | A digital advertising delivery method, system and device based on big data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105912686A (en) | Search engine marketing bid method and system based on machine learning | |
| US12321983B2 (en) | Adjust ads auction using predicted user response to an in-ad survey | |
| Fan et al. | Product sales forecasting using online reviews and historical sales data: A method combining the Bass model and sentiment analysis | |
| US20210366006A1 (en) | Ranking of business object | |
| US11288709B2 (en) | Training and utilizing multi-phase learning models to provide digital content to client devices in a real-time digital bidding environment | |
| Miralles-Pechuán et al. | A methodology based on Deep Learning for advert value calculation in CPM, CPC and CPA networks | |
| CN107464141B (en) | Method and device for information popularization, electronic equipment and computer readable medium | |
| US20100250335A1 (en) | System and method using text features for click prediction of sponsored search advertisements | |
| US20110047025A1 (en) | Immediacy targeting in online advertising | |
| AU2017203306A1 (en) | Ad-words optimization based on performance across multiple channels | |
| US20080140519A1 (en) | Advertising based on simplified input expansion | |
| US20110054999A1 (en) | System and method for predicting user navigation within sponsored search advertisements | |
| CN111400613A (en) | Article recommendation method, device, medium and computer equipment | |
| US20180285748A1 (en) | Performance metric prediction for delivery of electronic media content items | |
| US11941073B2 (en) | Generating and implementing keyword clusters | |
| US11367117B1 (en) | Artificial intelligence system for generating network-accessible recommendations with explanatory metadata | |
| US20160104074A1 (en) | Recommending Bidded Terms | |
| US20160171560A1 (en) | Linguistic Personalization of Messages for Targeted Campaigns | |
| US20190080352A1 (en) | Segment Extension Based on Lookalike Selection | |
| Gaikwad et al. | E-commerce recommendation system using improved probabilistic model | |
| US20230316106A1 (en) | Method and apparatus for training content recommendation model, device, and storage medium | |
| CN115222433A (en) | Information recommendation method and device and storage medium | |
| Yang et al. | Large scale CVR prediction through dynamic transfer learning of global and local features | |
| Rainy | AI-driven marketing analytics for retail strategy: a systematic review of data-backed campaign optimization | |
| CN103530794A (en) | Content-based bidding in online advertising |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160831 |