CN118035044A - Recommendation accuracy evaluation method for big data recommendation algorithm - Google Patents
Recommendation accuracy evaluation method for big data recommendation algorithm Download PDFInfo
- Publication number
- CN118035044A CN118035044A CN202410431092.2A CN202410431092A CN118035044A CN 118035044 A CN118035044 A CN 118035044A CN 202410431092 A CN202410431092 A CN 202410431092A CN 118035044 A CN118035044 A CN 118035044A
- Authority
- CN
- China
- Prior art keywords
- recommendation
- user
- data
- product
- accuracy
- 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
- 238000011156 evaluation Methods 0.000 title claims abstract description 38
- 238000002372 labelling Methods 0.000 claims abstract description 10
- 238000004364 calculation method Methods 0.000 claims description 17
- 238000000034 method Methods 0.000 claims description 4
- 238000004806 packaging method and process Methods 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 description 4
- 239000000853 adhesive Substances 0.000 description 1
- 230000001070 adhesive effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a recommendation accuracy evaluation method of a big data recommendation algorithm in the technical field of product recommendation, which comprises the following steps: step S1, acquiring a data set of shopping big data and a big data recommendation algorithm to be evaluated; s2, calculating user data in the data set by utilizing a big data recommendation algorithm to obtain a product recommendation list of each user, and labeling the product recommendation list; step S3, respectively calculating user satisfaction, prediction accuracy, coverage rate, confidence level and diversity degree based on the data set and the product recommendation list; and S4, carrying out weighted summation on the user satisfaction, the prediction accuracy, the coverage rate, the confidence level and the diversity degree to obtain the recommendation accuracy. The invention has the advantages that: the recommendation accuracy of the big data recommendation algorithm is evaluated, and the product recommendation accuracy is greatly guaranteed.
Description
Technical Field
The invention relates to the technical field of product recommendation, in particular to a recommendation accuracy evaluation method of a big data recommendation algorithm.
Background
With the progress of technology, people have become more and more free from the network, and shopping, driving, game, work, social activities and the like are performed manually on the network, so that a large amount of user data, namely, big data, is generated. The big data contain huge value, for example, the user portrait can be obtained by analyzing the big data of shopping, and then corresponding product recommendation is carried out on the user, so that the accuracy of product recommendation is ensured, for example, the recommendation of products such as telephone fee packages, commodity brands, commodity models and the like is carried out, the user is prevented from receiving uninteresting product recommendation as much as possible, and a large number of big data recommendation algorithms are generated.
However, the conventional big data recommendation algorithm only analyzes shopping big data based on specific dimensions, and further performs product recommendation, for example, product recommendation based on dimensions such as consumption times, browsing time length, clicking times, etc., and the recommendation accuracy of the big data recommendation algorithm is often poor when the big data recommendation algorithm is actually used. Therefore, how to provide a recommendation accuracy evaluation method for a big data recommendation algorithm, so as to evaluate the recommendation accuracy of the big data recommendation algorithm, so as to ensure the product recommendation accuracy, becomes a technical problem to be solved urgently.
Disclosure of Invention
The invention aims to solve the technical problem of providing a recommendation accuracy evaluation method for a big data recommendation algorithm, which is used for evaluating the recommendation accuracy of the big data recommendation algorithm so as to ensure the recommendation accuracy of products.
The invention is realized in the following way: a recommendation accuracy evaluation method of a big data recommendation algorithm comprises the following steps:
Step S1, acquiring a data set of shopping big data and a big data recommendation algorithm to be evaluated;
S2, calculating user data in a data set by using the big data recommendation algorithm to obtain a product recommendation list of each user, and marking the product recommendation list;
Step S3, calculating user satisfaction, prediction accuracy, coverage rate, confidence level and diversity degree based on the data set and the product recommendation list respectively;
S4, carrying out weighted summation on the user satisfaction degree, the prediction accuracy degree, the coverage rate, the confidence degree and the diversity degree to obtain recommendation accuracy degree;
And S5, displaying the recommendation accuracy in real time, packaging the data set, the big data recommendation algorithm and the recommendation accuracy into an evaluation data packet, and storing and backing up the evaluation data packet.
Further, in the step S1, the data set includes user data related to a user and product data related to a product;
the user data at least comprises user identification, gender, age, occupation, purchase time, browsing history, purchase times, shopping evaluation, purchase time interval and purchase frequency;
The product data includes at least a product category, a brand, and a model.
Further, in the step S2, labeling the product recommendation list specifically includes:
labeling the product recommendation list, wherein the labeling comprises adhesion, predictive scoring and true scoring.
Further, in the step S3, the calculation formula of the user satisfaction is:；
Wherein CSD represents user satisfaction; u represents a set of users in the dataset; u represents a user; r (u) represents a product recommendation list for user u; t (u) represents the set of products with adhesion for user u in the dataset.
Further, in the step S3, the calculation formula of the prediction accuracy is:
；
wherein HR represents hit rate, i.e. prediction accuracy; u represents a set of users in the dataset; u represents a user; representing an indication function; r (u) represents a product recommendation list for user u; t (u) represents the set of products with adhesion for user u in the dataset.
Further, in the step S3, the coverage rate is calculated according to the formula:；
Wherein Coverage represents Coverage; u represents a set of users in the dataset; u represents a user; r (u) represents a product recommendation list for user u; t (u) represents the set of products with adhesion for user u in the dataset.
Further, in the step S3, the confidence coefficient calculating process is as follows:
Setting a confidence interval R i, and respectively calculating the times A in of the predictive scores of the user on the products in the confidence interval R i and the times A out outside the confidence interval R i;
confidence is calculated based on the number a in and the number a out: ；；
Wherein, Representing the confidence level; i represents a product; /(I)Representing a product set; /(I)Representing the sub-confidence of the ith product.
Further, in the step S3, the calculation formula of the diversity degree is:；；
wherein HD represents hamming distance, i.e. degree of diversity; u and t both represent users; u represents a set of users in the dataset; representing the difference degree of different user product recommendation lists; /(I) The product recommendation list of the user u and the user t is shown, and the number of the same products is shown; l represents the length of the product recommendation list.
Further, the step S5 specifically includes:
displaying the recommendation accuracy in real time through a display screen, encrypting the recommendation accuracy through a national encryption algorithm, and pushing the encrypted recommendation accuracy to a pre-associated mobile terminal in real time;
Hash calculation is carried out on the data set, the big data recommendation algorithm and the recommendation accuracy to obtain a hash value, the data set, the big data recommendation algorithm, the recommendation accuracy and the hash value are packaged into an evaluation data packet, the evaluation data packet is encrypted into first encrypted data through a DES encryption algorithm, the first encrypted data is divided into two sections based on a preset proportion, the second encrypted data is obtained in sequence before and after exchange, and the second encrypted data is stored and then backed up to a block chain.
The invention has the advantages that:
1. Calculating user data in the data set by using the big data recommendation algorithm through acquiring the data set of shopping big data and the big data recommendation algorithm to be evaluated, obtaining a product recommendation list of each user, and marking the product recommendation list; then, based on the data set and the product recommendation list, respectively calculating user satisfaction, prediction accuracy, coverage rate, confidence coefficient and diversity degree, and carrying out weighted summation on the user satisfaction, the prediction accuracy, the coverage rate, the confidence coefficient and the diversity degree to obtain recommendation accuracy; the recommendation accuracy of the product recommendation list is evaluated based on five dimensions of user satisfaction, prediction accuracy, coverage rate, confidence and diversity degree, so that the big data recommendation algorithm for generating the product recommendation list is evaluated, and finally the recommendation accuracy of the big data recommendation algorithm is evaluated, and the product recommendation accuracy is greatly guaranteed.
2. The recommendation accuracy is encrypted through the national encryption algorithm and then pushed to the pre-associated mobile terminal in real time, so that the recommendation accuracy is prevented from being stolen by plaintext in the transmission process, and the transmission safety of the recommendation accuracy is greatly improved.
3. Hash value is obtained by carrying out hash calculation on the data set, the big data recommendation algorithm and the recommendation accuracy, the data set, the big data recommendation algorithm, the recommendation accuracy and the hash value are packed into an evaluation data packet, and whether the data set, the big data recommendation algorithm and the recommendation accuracy are tampered or not can be checked quickly through the hash value; encrypting the evaluation data packet into first encrypted data through a DES encryption algorithm, so as to avoid the interception of the evaluation data packet by a plaintext; dividing the first encrypted data into two sections through a preset proportion, and exchanging the front-back sequence to obtain second encrypted data, wherein if the divided proportion is not known, decryption is directly carried out to obtain a messy code; the second encrypted data is backed up to the blockchain for decentralization storage, so that the second encrypted data is prevented from being tampered, at least four sets of security measures (hash calculation, DES encryption algorithm, pre-set proportion exchange of the front and back sequences and the blockchain) are adopted before and after, and the security of evaluating the storage backup of the data packet is greatly improved.
Drawings
The invention will be further described with reference to examples of embodiments with reference to the accompanying drawings.
FIG. 1 is a flow chart of a big data recommendation algorithm recommendation accuracy evaluation method of the present invention.
Detailed Description
The technical scheme in the embodiment of the application has the following overall thought: and evaluating the recommendation accuracy of the product recommendation list based on five dimensions of user satisfaction, prediction accuracy, coverage rate, confidence and diversity degree, and further evaluating a big data recommendation algorithm for generating the product recommendation list so as to evaluate the recommendation accuracy of the big data recommendation algorithm and ensure the product recommendation accuracy.
Referring to fig. 1, a preferred embodiment of a recommendation accuracy evaluation method of a big data recommendation algorithm of the present invention includes the following steps:
Step S1, acquiring a data set of shopping big data and a big data recommendation algorithm to be evaluated;
S2, calculating user data in a data set by using the big data recommendation algorithm to obtain a product recommendation list of each user, and marking the product recommendation list;
Step S3, calculating user satisfaction, prediction accuracy, coverage rate, confidence level and diversity degree based on the data set and the product recommendation list respectively;
The user satisfaction degree is that the proportion of adhesive products is contained in a product recommendation list; the adhesiveness, i.e., user adhesiveness customer stickiness, refers to the probability that a user will generate, after experiencing a product, a period of time for reuse, a period of time for interval, a frequency, and a desire to consume; the coverage rate is the proportion of products which are interested by the user and can be recommended by the big data recommendation algorithm in the product set which is interested by the user; the confidence level is the confidence level of the user on the prediction result of the big data recommendation algorithm; the degree of diversity is the diversity of the data in the data set;
S4, carrying out weighted summation on the user satisfaction degree, the prediction accuracy degree, the coverage rate, the confidence degree and the diversity degree to obtain recommendation accuracy degree;
And S5, displaying the recommendation accuracy in real time, packaging the data set, the big data recommendation algorithm and the recommendation accuracy into an evaluation data packet, and storing and backing up the evaluation data packet.
In the step S1, the data set includes user data related to a user and product data related to a product;
the user data at least comprises user identification, gender, age, occupation, purchase time, browsing history, purchase times, shopping evaluation, purchase time interval and purchase frequency;
The product data includes at least a product category, a brand, and a model.
In the step S2, labeling the product recommendation list specifically includes:
Labeling the product recommendation list, wherein the labeling comprises adhesiveness, predictive scores and true scores;
The adhesion is classified into adhesion and non-adhesion based on an adhesion index and a preset index interval; the adhesion indicators include, but are not limited to, length of reuse, time interval of purchase, frequency of purchase, probability of producing a consumer desire.
In the step S3, the calculation formula of the user satisfaction is:；
Wherein CSD represents user satisfaction; u represents a set of users in the dataset; u represents a user; r (u) represents a product recommendation list for user u; t (u) represents the set of products with adhesion for user u in the dataset.
In the step S3, the calculation formula of the prediction accuracy is:
；
wherein HR represents hit rate, i.e. prediction accuracy; u represents a set of users in the dataset; u represents a user; representing an indication function; r (u) represents a product recommendation list for user u; t (u) represents the set of products with adhesion for user u in the dataset.
In practice, the prediction accuracy can also be measured by the root mean square error RMSE and the mean absolute error MAE:；/>；
wherein U represents a user set in the dataset; u represents a user; i represents a product; Representing a product set; /(I) Representing the true score of user u on product i in the dataset; /(I)Representing the predictive scores generated by the big data recommendation algorithm for the user-sample pairs in the dataset.
In the step S3, the coverage rate is calculated according to the following formula:；
Wherein Coverage represents Coverage; u represents a set of users in the dataset; u represents a user; r (u) represents a product recommendation list for user u; t (u) represents the set of products with adhesion for user u in the dataset.
In the step S3, the confidence coefficient calculating process is as follows:
Setting a confidence interval R i, and respectively calculating the times A in of the predictive scores of the user on the products in the confidence interval R i and the times A out outside the confidence interval R i;
confidence is calculated based on the number a in and the number a out: ；；
Wherein, Representing the confidence level; i represents a product; /(I)Representing a product set; /(I)Representing the sub-confidence of the ith product.
Different users have different scores for different products, a certain confidence level is selected (for 90% confidence level, the corresponding Z value is 1.64, for 95% confidence level, the corresponding Z value is 1.96, the Z value represents the confidence level), and a confidence interval should exist for each product, which is expressed as that the confidence interval contains the mean value of the true scores under the selected confidence level. The calculation formula of the confidence interval R i is as follows:；/>；/>；
Wherein, Representing that the confidence interval contains a true average sample score; /(I)Representing the standard deviation of the samples; /(I)Representing the number of users in the dataset who have true scores for product i; /(I)Representing the true score of the user-product pair (u, i) in the dataset.
In the step S3, the calculation formula of the diversity degree is as follows:；/>；
wherein HD represents hamming distance, i.e. degree of diversity; u and t both represent users; u represents a set of users in the dataset; representing the difference degree of different user product recommendation lists; /(I) The product recommendation list of the user u and the user t is shown, and the number of the same products is shown; l represents the length of the product recommendation list.
In specific implementations, the degree of diversity can also be measured by shannon entropy Entropy:；/>；
Wherein U represents a user set in the dataset; u represents a user; Representing a son shannon entropy value corresponding to the user u; /(I) Representing shannon entropy, i.e. the degree of diversity; c represents a product category; c represents a product category set; p (c) represents the proportion of the product of product class c in R (u); r (u) represents the product recommendation list for user u.
The step S5 specifically comprises the following steps:
displaying the recommendation accuracy in real time through a display screen, encrypting the recommendation accuracy through a national encryption algorithm, and pushing the encrypted recommendation accuracy to a pre-associated mobile terminal in real time;
Hash calculation is carried out on the data set, the big data recommendation algorithm and the recommendation accuracy to obtain a hash value, the data set, the big data recommendation algorithm, the recommendation accuracy and the hash value are packaged into an evaluation data packet, the evaluation data packet is encrypted into first encrypted data through a DES encryption algorithm, the first encrypted data is divided into two sections based on a preset proportion, the second encrypted data is obtained in sequence before and after exchange, and the second encrypted data is stored and then backed up to a block chain.
In summary, the invention has the advantages that:
1. Calculating user data in the data set by using the big data recommendation algorithm through acquiring the data set of shopping big data and the big data recommendation algorithm to be evaluated, obtaining a product recommendation list of each user, and marking the product recommendation list; then, based on the data set and the product recommendation list, respectively calculating user satisfaction, prediction accuracy, coverage rate, confidence coefficient and diversity degree, and carrying out weighted summation on the user satisfaction, the prediction accuracy, the coverage rate, the confidence coefficient and the diversity degree to obtain recommendation accuracy; the recommendation accuracy of the product recommendation list is evaluated based on five dimensions of user satisfaction, prediction accuracy, coverage rate, confidence and diversity degree, so that the big data recommendation algorithm for generating the product recommendation list is evaluated, and finally the recommendation accuracy of the big data recommendation algorithm is evaluated, and the product recommendation accuracy is greatly guaranteed.
2. The recommendation accuracy is encrypted through the national encryption algorithm and then pushed to the pre-associated mobile terminal in real time, so that the recommendation accuracy is prevented from being stolen by plaintext in the transmission process, and the transmission safety of the recommendation accuracy is greatly improved.
3. Hash value is obtained by carrying out hash calculation on the data set, the big data recommendation algorithm and the recommendation accuracy, the data set, the big data recommendation algorithm, the recommendation accuracy and the hash value are packed into an evaluation data packet, and whether the data set, the big data recommendation algorithm and the recommendation accuracy are tampered or not can be checked quickly through the hash value; encrypting the evaluation data packet into first encrypted data through a DES encryption algorithm, so as to avoid the interception of the evaluation data packet by a plaintext; dividing the first encrypted data into two sections through a preset proportion, and exchanging the front-back sequence to obtain second encrypted data, wherein if the divided proportion is not known, decryption is directly carried out to obtain a messy code; the second encrypted data is backed up to the blockchain for decentralization storage, so that the second encrypted data is prevented from being tampered, at least four sets of security measures (hash calculation, DES encryption algorithm, pre-set proportion exchange of the front and back sequences and the blockchain) are adopted before and after, and the security of evaluating the storage backup of the data packet is greatly improved.
While specific embodiments of the invention have been described above, it will be appreciated by those skilled in the art that the specific embodiments described are illustrative only and not intended to limit the scope of the invention, and that equivalent modifications and variations of the invention in light of the spirit of the invention will be covered by the claims of the present invention.
Claims (8)
1. A big data recommendation algorithm recommendation accuracy evaluation method is characterized in that: the method comprises the following steps:
Step S1, acquiring a data set of shopping big data and a big data recommendation algorithm to be evaluated;
S2, calculating user data in a data set by using the big data recommendation algorithm to obtain a product recommendation list of each user, and marking the product recommendation list;
Step S3, calculating user satisfaction, prediction accuracy, coverage rate, confidence level and diversity degree based on the data set and the product recommendation list respectively;
the calculation formula of the prediction accuracy is as follows: ；
wherein HR represents hit rate, i.e. prediction accuracy; u represents a set of users in the dataset; u represents a user; Representing an indication function; r (u) represents a product recommendation list for user u; t (u) represents the product set with the adhesiveness of user u in the dataset;
S4, carrying out weighted summation on the user satisfaction degree, the prediction accuracy degree, the coverage rate, the confidence degree and the diversity degree to obtain recommendation accuracy degree;
And S5, displaying the recommendation accuracy in real time, packaging the data set, the big data recommendation algorithm and the recommendation accuracy into an evaluation data packet, and storing and backing up the evaluation data packet.
2. The big data recommendation algorithm recommendation accuracy evaluation method according to claim 1, wherein: in the step S1, the data set includes user data related to a user and product data related to a product;
the user data at least comprises user identification, gender, age, occupation, purchase time, browsing history, purchase times, shopping evaluation, purchase time interval and purchase frequency;
The product data includes at least a product category, a brand, and a model.
3. The big data recommendation algorithm recommendation accuracy evaluation method according to claim 1, wherein: in the step S2, labeling the product recommendation list specifically includes:
labeling the product recommendation list, wherein the labeling comprises adhesion, predictive scoring and true scoring.
4. The big data recommendation algorithm recommendation accuracy evaluation method according to claim 1, wherein: in the step S3, the calculation formula of the user satisfaction is: ; wherein CSD represents user satisfaction; u represents a set of users in the dataset; u represents a user; r (u) represents a product recommendation list for user u; t (u) represents the set of products with adhesion for user u in the dataset.
5. The big data recommendation algorithm recommendation accuracy evaluation method according to claim 1, wherein: in the step S3, the coverage rate is calculated according to the following formula:；
Wherein Coverage represents Coverage; u represents a set of users in the dataset; u represents a user; r (u) represents a product recommendation list for user u; t (u) represents the set of products with adhesion for user u in the dataset.
6. The big data recommendation algorithm recommendation accuracy evaluation method according to claim 1, wherein: in the step S3, the confidence coefficient calculating process is as follows:
Setting a confidence interval R i, and respectively calculating the times A in of the predictive scores of the user on the products in the confidence interval R i and the times A out outside the confidence interval R i;
confidence is calculated based on the number a in and the number a out: ；；
Wherein, Representing the confidence level; i represents a product; /(I)Representing a product set; /(I)Representing the sub-confidence of the ith product.
7. The big data recommendation algorithm recommendation accuracy evaluation method according to claim 1, wherein: in the step S3, the calculation formula of the diversity degree is as follows:；/>；
wherein HD represents hamming distance, i.e. degree of diversity; u and t both represent users; u represents a set of users in the dataset; representing the difference degree of different user product recommendation lists; /(I) The product recommendation list of the user u and the user t is shown, and the number of the same products is shown; l represents the length of the product recommendation list.
8. The big data recommendation algorithm recommendation accuracy evaluation method according to claim 1, wherein: the step S5 specifically comprises the following steps:
displaying the recommendation accuracy in real time through a display screen, encrypting the recommendation accuracy through a national encryption algorithm, and pushing the encrypted recommendation accuracy to a pre-associated mobile terminal in real time;
Hash calculation is carried out on the data set, the big data recommendation algorithm and the recommendation accuracy to obtain a hash value, the data set, the big data recommendation algorithm, the recommendation accuracy and the hash value are packaged into an evaluation data packet, the evaluation data packet is encrypted into first encrypted data through a DES encryption algorithm, the first encrypted data is divided into two sections based on a preset proportion, the second encrypted data is obtained in sequence before and after exchange, and the second encrypted data is stored and then backed up to a block chain.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410431092.2A CN118035044A (en) | 2024-04-11 | 2024-04-11 | Recommendation accuracy evaluation method for big data recommendation algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410431092.2A CN118035044A (en) | 2024-04-11 | 2024-04-11 | Recommendation accuracy evaluation method for big data recommendation algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118035044A true CN118035044A (en) | 2024-05-14 |
Family
ID=90989636
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410431092.2A Pending CN118035044A (en) | 2024-04-11 | 2024-04-11 | Recommendation accuracy evaluation method for big data recommendation algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118035044A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114493361A (en) * | 2022-02-21 | 2022-05-13 | 中科晶锐(长沙)科技有限公司 | Effectiveness evaluation method and device for commodity recommendation algorithm |
| CN115115435A (en) * | 2022-08-02 | 2022-09-27 | 北京工业大学 | E-commerce recommendation algorithm based on diversity |
| CN115834201A (en) * | 2022-11-23 | 2023-03-21 | 梅赛德斯-奔驰集团股份公司 | Data encryption method, data decryption method and data processing method for data storage system |
-
2024
- 2024-04-11 CN CN202410431092.2A patent/CN118035044A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114493361A (en) * | 2022-02-21 | 2022-05-13 | 中科晶锐(长沙)科技有限公司 | Effectiveness evaluation method and device for commodity recommendation algorithm |
| CN115115435A (en) * | 2022-08-02 | 2022-09-27 | 北京工业大学 | E-commerce recommendation algorithm based on diversity |
| CN115834201A (en) * | 2022-11-23 | 2023-03-21 | 梅赛德斯-奔驰集团股份公司 | Data encryption method, data decryption method and data processing method for data storage system |
Non-Patent Citations (2)
| Title |
|---|
| 小莹莹: "读书笔记 |《推荐系统实践》- 个性化推荐系统总结", pages 1 - 23, Retrieved from the Internet <URL:https://cloud.tencent.com/developer/article/1104303> * |
| 朱郁筱: "推荐系统评价指标综述", 电 子 科 技 大 学 学 报, vol. 41, no. 2, 31 March 2012 (2012-03-31), pages 1 - 13 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109523341B (en) | Anonymous cross-domain recommendation method based on block chain technology | |
| US10282755B2 (en) | Systems, methods, and media for mobile advertising conversion attribution | |
| CN109688469B (en) | Advertisement display method and device | |
| CN105657003B (en) | Information processing method and server | |
| US8620746B2 (en) | Scoring quality of traffic to network sites | |
| US20100114954A1 (en) | Realtime popularity prediction for events and queries | |
| US20160225014A1 (en) | System and method for building a targeted audience for an online advertising campaign | |
| US20150161661A1 (en) | Preservation of scores of the quality of traffic to network sites across clients and over time | |
| CN106997549A (en) | The method for pushing and system of a kind of advertising message | |
| CN109840782A (en) | Clicking rate prediction technique, device, server and storage medium | |
| CN104992348B (en) | A kind of method and apparatus of information displaying | |
| CN112465594B (en) | Live interaction method, device, equipment and storage medium | |
| CN110363604A (en) | Page generation method and device | |
| CN109711917B (en) | Information pushing method and device | |
| EP3469538A1 (en) | System and method for incorporating branded creative content into messaging services | |
| CN110392155A (en) | It has been shown that, processing method, device and the equipment of notification message | |
| CN112015986A (en) | Data pushing method and device, electronic equipment and computer readable storage medium | |
| CN106960033B (en) | Method and device for labeling information stream | |
| CN118035044A (en) | Recommendation accuracy evaluation method for big data recommendation algorithm | |
| JPWO2021240826A5 (en) | ||
| CN116843256A (en) | International trade logistics tracking method | |
| CN115357766B (en) | Data matching method and related device | |
| CN115880037A (en) | Commodity recommendation method based on multi-project planning integration analysis | |
| CN109919705A (en) | Shops's labeling method, device, equipment and readable storage medium storing program for executing | |
| CN114187029A (en) | Method and system for verifying commodity price parameters |
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 |