WO2019019396A1 - 推送结果预测方法、装置、计算机设备和存储介质 - Google Patents
推送结果预测方法、装置、计算机设备和存储介质 Download PDFInfo
- Publication number
- WO2019019396A1 WO2019019396A1 PCT/CN2017/104871 CN2017104871W WO2019019396A1 WO 2019019396 A1 WO2019019396 A1 WO 2019019396A1 CN 2017104871 W CN2017104871 W CN 2017104871W WO 2019019396 A1 WO2019019396 A1 WO 2019019396A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- user account
- account set
- click
- predicted
- information
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
-
- 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
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/535—Tracking the activity of the user
Definitions
- the present application relates to the field of information processing technologies, and in particular, to a push result prediction method, apparatus, computer device, and storage medium.
- the information is pushed according to the preset push parameters, and the information is pushed according to the preset target user account.
- some target user accounts are unavailable, and the information cannot be pushed to the target user account, which reduces the success rate of information push.
- a push result prediction method is provided.
- a method for predicting push results including:
- a push result prediction device includes:
- a push task acquisition module configured to obtain an information push task
- a user account extraction module configured to extract a target user account in the information push task, and obtain a target user account set
- the historical information obtaining module is configured to obtain historical online information corresponding to each user account in the target user account set;
- the online account selection module is configured to select an online user account from the set of target user accounts according to the obtained historical online information, and obtain a predicted arrival user account set;
- a tag information obtaining module configured to acquire a user interest tag and historical click information corresponding to each user account in the predicted user account set
- Clicking an account selection module configured to extract a predicted click user account set from the predicted arrival user account set according to the obtained user interest tag and historical click information
- the prediction result generating module is configured to generate a push prediction result according to the predicted arrival user account set and the predicted click user account set.
- a computer device comprising a memory and a processor, wherein the memory stores computer executable instructions that, when executed by the processor, cause the processor to perform the following steps:
- One or more storage media storing computer executable instructions that, when executed by one or more processors, cause the one or more processors to perform the following steps:
- 1 is an application environment diagram of a method for predicting a push result in an embodiment
- FIG. 2 is a structural block diagram of a server in a push result prediction system in an embodiment
- FIG. 3 is a schematic flow chart of a method for predicting a push result in an embodiment
- FIG. 4 is a schematic flowchart of a step of obtaining a predicted arrival user account set in an embodiment
- FIG. 5 is a schematic flowchart of a step of extracting a predicted click user account set in an embodiment
- FIG. 6 is a schematic flowchart of a step of extracting a user account according to a click prediction value in an embodiment
- Figure 7 is a block diagram showing the structure of a push result prediction apparatus in an embodiment
- FIG. 8 is a structural block diagram of an online account selection module in an embodiment
- FIG. 9 is a structural block diagram of a click account selection module in an embodiment
- Figure 10 is a block diagram showing the structure of a push result prediction apparatus in another embodiment.
- FIG. 1 is an application environment diagram of a method for predicting a push result in an embodiment.
- the push result prediction method is applied to a push result prediction system.
- the push result prediction system includes a terminal 110 and a server 120, wherein the terminal 110 is connected to the server 120 through a network.
- the terminal 110 may be a fixed terminal or a mobile terminal, and the fixed terminal may specifically be at least one of a printer, a scanner, and a monitor, and the mobile terminal may specifically be at least one of a tablet computer, a smart phone, a personal data assistant, and a digital camera.
- Server 120 can be a single server or a cluster of servers with multiple servers.
- the server 120 includes a processor, a non-volatile storage medium, an internal memory, and a network interface connected by a system bus.
- the processor of server 120 is used to provide computing and control capabilities to support the operation of the entire server 120, the memory for storing data, code instructions, etc., and the network interface for network communication with terminal 110.
- At least one computer executable instruction is stored on the memory, and the computer executable instructions are executable by the processor to implement the push result prediction method applicable to the server 120 provided in the embodiment of the present application.
- the memory may include a non-volatile storage medium such as a magnetic disk, an optical disk, or a read-only memory (ROM).
- a non-volatile storage medium such as a magnetic disk, an optical disk, or a read-only memory (ROM).
- the memory includes a non-volatile storage medium and an internal memory; the non-volatile storage medium stores an operating system, a database, and computer-executable instructions, and the database stores historical online information corresponding to each user account, History click information and user interest tags, the computer executable instructions executable by the processor to implement the push result prediction method described above; the internal memory provides cached operation for operating systems and computer executable instructions in the non-volatile storage medium surroundings.
- FIG. 2 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the server to which the solution of the present application is applied.
- the specific server may include a ratio. More or fewer components are shown in Figure 2, or some components are combined, or have different component arrangements.
- a method for predicting a push result is provided.
- This embodiment is exemplified by the method applied to the server 120 in FIG. 1, and the method specifically includes the following content:
- the terminal 110 acquires an operator account input by the push operator, generates a push prediction page request according to the acquired operator account, and sends a push prediction page request to the server 120.
- the server 120 receives the push prediction page request sent by the terminal 110, parses the push prediction page request, and analyzes and extracts the operator account number in the push prediction page request, and verifies whether the extracted operator account has the right to access the push prediction page. After verifying that the extracted operator account has the right to access the push prediction page, the server 120 acquires the push prediction page data, and returns the obtained push prediction page data to the terminal 110 corresponding to the push prediction page request.
- the terminal 110 receives the push prediction page data returned by the server 120, and displays the push prediction page based on the push prediction page data.
- the terminal 110 detects that the push operator touches the push prediction page.
- the pushed push result prediction command acquires the information push task input by the push operator on the push prediction page according to the push result prediction command, and transmits the acquired information push task to the server 120.
- the server 120 receives the information push task sent by the terminal 110.
- the information push task includes a target user account, and the target user account is a user account corresponding to the target user.
- the server 120 parses the information push task, and analyzes and extracts the target user account in the information push task, so that the extracted target user account constitutes the target user account set.
- the historical online information corresponding to each user account is stored in the database of the server 120.
- Historical online information includes online time information and user accounts.
- the server 120 extracts the user account in the target user account set, and queries historical online information corresponding to the extracted user account in the historical online information in the database.
- the server 120 detects whether the user account corresponding to the historical online information is online according to the historical online information corresponding to each user account.
- the server 120 detects the online user account in the target user account set, extracts the online user account, and constructs a predicted arrival user account set according to the extracted user account.
- the server 120 detects an offline user account in the target user account set, deletes the offline user account from the target user account set, and deletes the target user account set after the offline user account as a prediction. Reach the user account collection.
- the user interest tag and the historical click information corresponding to each user account are stored in the database in the server 120.
- the user interest tag is a user's favorite identifier corresponding to the user account.
- the user interest tag corresponding to the user account of the user who likes the soccer ball is a favorite football.
- the history click information includes the click time information and the user account of the user clicking the push message.
- the server 120 predicts each of the predicted user account sets.
- the user account extracts user interest tags and historical click information corresponding to each user account from the database.
- the server 120 obtains the information to be pushed corresponding to the information push task, and calculates the degree of interest according to the user interest tag corresponding to each user account in the predicted user account set and the acquired information to be pushed.
- the server 120 counts the historical click rate corresponding to each user account according to the historical click information corresponding to each user account predicted to reach the user account set.
- the server 120 extracts, from the predicted arrival user account set, a user account whose interest degree is higher than the preset interest degree and whose historical click rate is higher than the preset click rate threshold, and constructs a predicted click user account set according to the extracted user account.
- the server 120 generates a push prediction result corresponding to the information push task based on the predicted arrival user account set and the predicted click user account set.
- the server 120 counts the total number of user accounts in the target user account set to obtain the total amount of information push; the server 120 statistically predicts the total number of user accounts in the set of user accounts to obtain the predicted total amount of information arrival; Click the total number of user accounts in the user account collection to get the total amount of information click prediction.
- the server 120 obtains the information predicted arrival rate by dividing the statistical information arrival prediction total amount by the statistical information push total amount, and divides the statistical information click prediction total amount by the statistical information click prediction total amount to obtain the information prediction click rate. .
- the server 120 generates a push prediction result corresponding to the information push target based on the statistically predicted information prediction rate and the information predicted click rate.
- the method further includes: extracting a predicted arrival user account set in the push prediction result; and generating a target user account adjustment suggestion corresponding to the information push task according to the extracted predicted arrival user account set.
- the server 120 transmits the generated push prediction result to the terminal 110 corresponding to the information push task, causing the terminal 110 to display the push prediction result in the push prediction page.
- the push operator clicks the account adjustment suggestion generation button in the push prediction page through the input device of the terminal 110.
- the terminal 110 detects that the account adjustment suggestion generation button is clicked, triggering the account adjustment suggestion
- the instruction is generated, and the account adjustment suggestion generation instruction is sent to the server 120.
- the server 120 After receiving the account adjustment suggestion generation instruction sent by the terminal 110, the server 120 extracts the predicted arrival user account set in the push prediction result, and generates a target user account adjustment suggestion corresponding to the information push task according to the predicted user account in the user account set. The server 120 sends the generated target user account adjustment proposal to the terminal 110, so that the terminal 110 displays the account adjustment suggestion on the push prediction page.
- the terminal 110 detects that the push operator clicks the account adjustment confirmation button in the push prediction page through the input device, and the terminal 110 extracts the user account in the account adjustment proposal to obtain the predicted arrival user account set, and predicts the arrival of the user account set as the target of the information push task. User account collection.
- the target user account in the information push task is extracted, and the target user account set is obtained.
- the accuracy of predicting the arrival of the user account set and the predicted click user account set is ensured, thereby ensuring the generated push prediction result.
- the target user account in the information push task can be adjusted according to the push prediction result, thereby ensuring that the arrival amount and the click amount corresponding to the information push task can be improved, thereby improving the push success rate of the information push task.
- S308 specifically includes the step of acquiring a predicted arrival user account set, and the step specifically includes the following:
- the user account marked as invalid in the obtained historical online information is deleted from the target user account set, and the user account marked as invalid is deleted, and a valid user account set is obtained.
- the server 120 collects, according to historical online information corresponding to each user account, a user account that does not have historical online information within a preset time period, and marks a user account that does not have historical online information within a preset time period as Invalid user account.
- the server 120 queries the target user account set for the user account marked as invalid, deletes the queried user account from the target user account set, and forms a valid user account set according to the remaining user accounts in the target user account set.
- the preset time period can be the most recent month, two months or three months.
- the server 120 extracts valid users from the acquired historical online information.
- the account collects historical online information corresponding to each user account in China, and counts the number of online times corresponding to each user account in the preset time period according to the extracted historical online information.
- the server 120 compares the online number of the preset time period corresponding to each user account in the valid user account set with the preset number of times, and determines the user account whose online number is higher than the preset number of times in the preset time period. Extracting the determined user account from the set of valid user accounts, and obtaining the predicted arrival user account set.
- the user account marked as invalid is deleted from the target user account set to obtain a valid user account set, and then the predicted arrival user account set is extracted from the valid user account set according to the historical online information.
- the utilization of historical online information is improved, and the accuracy of selecting the predicted arrival user account set is also improved.
- S312 specifically includes the step of extracting a predicted click user account set, and the step specifically includes the following content:
- the server 120 parses the information push task by parsing the message to be pushed in the extracted information task.
- the message to be pushed is a message to be pushed to the user corresponding to the user account in the target user account set.
- the server 120 analyzes the push information by using the keyword extraction software, and extracts the message keyword from the information to be pushed by analyzing.
- the database of servers 120 stores message keywords corresponding to the messages to be pushed. After obtaining the message to be pushed, the server 120 queries the database for the message keyword corresponding to the obtained to-be-push message.
- the server 120 calculates the similarity between the user interest tag corresponding to each user account and the extracted message keyword, and the calculated similarity is respectively corresponding to the user account. Degree of interest.
- the historical click rate corresponding to each user account in the user account set is predicted and predicted.
- the server 120 statistically predicts the number of push message receptions and the number of push message clicks of each user account in the user account set according to the historical click information, and calculates the number of push messages received and the number of push message clicks.
- the historical click rate corresponding to the user account is obtained, that is, the server 120 divides the number of push message clicks by the number of push message receptions to obtain a historical click rate.
- the server 120 compares the degree of interest corresponding to each user account with a preset degree of interest, and compares the historical click rate corresponding to each user account with a preset click rate.
- the server 120 compares the user accounts whose interest degree is higher than the preset interest degree and the historical click rate is higher than the preset click rate from the predicted arrival user set, and constructs a predicted click user account set according to the extracted user account.
- the interest degree corresponding to the user account is calculated according to the user interest tag and the message keyword corresponding to the user account
- the historical click rate of the user account is calculated according to the historical click information
- the user account is extracted according to the degree of interest and the historical click rate.
- S510 specifically includes the step of extracting a user account according to the click prediction value, and the step specifically includes the following content:
- the server 120 obtains the weight corresponding to each of the interest degree and the historical click rate, according to the interest degree corresponding to each user account, the historical click rate, the weight corresponding to the degree of interest, and the right corresponding to the historical click rate.
- the value is weighted and calculated to obtain a click prediction value corresponding to each user account.
- the server 120 will predict the arrival of each user account pair in the set of user accounts.
- the predicted click value is compared with the preset threshold, and the user account whose click predicted value exceeds the preset threshold is selected from the predicted arrival user account set, and the predicted click user account set is formed according to the selected user account.
- the click prediction value corresponding to each user account is calculated according to the calculated interest degree and the historical click rate, and the predicted click user value is extracted from the predicted arrival user set by the calculated click prediction value, thereby greatly improving the prediction. Clicking on the extraction accuracy in the user account improves the prediction accuracy of the prediction results.
- a push result prediction apparatus 700 is extracted, and the apparatus specifically includes the following: a push task acquisition module 702, a user account extraction module 704, a history information acquisition module 706, and an online account selection module. 708.
- the push task obtaining module 702 is configured to obtain an information push task.
- the user account extraction module 704 is configured to extract a target user account in the information push task, and obtain a target user account set.
- the history information obtaining module 706 is configured to obtain historical online information corresponding to each user account in the target user account set.
- the online account selection module 708 is configured to select an online user account from the target user account set according to the obtained historical online information, and obtain a predicted arrival user account set.
- the tag information obtaining module 710 is configured to obtain a user interest tag and historical click information corresponding to each user account in the predicted user account set.
- the account selection module 712 is configured to extract a predicted click user account set from the predicted arrival user account set according to the obtained user interest tag and historical click information.
- the prediction result generating module 714 is configured to generate a push prediction result according to the predicted arrival user account set and the predicted click user account set.
- the target user account in the information push task is extracted, and the target user account set is obtained.
- the accuracy of predicting the arrival of the user account set and the predicted click user account set is ensured, thereby ensuring the generated push prediction result.
- the accuracy. And can root According to the push prediction result, the target user account in the information push task is adjusted, thereby ensuring that the amount of arrival and the amount of clicks corresponding to the information push task can be improved, thereby improving the push success rate of the information push task.
- the online account selection module 708 specifically includes the following: an effective account obtaining module 708a, an online number counting module 708b, and an online number comparison module 708c.
- the valid account obtaining module 708a is configured to delete the user account marked as invalid in the obtained historical online information from the target user account set to obtain a valid user account set.
- the online number statistics module 708b is configured to count the number of online times in the preset time period corresponding to each user account in the valid user account set according to the obtained historical online information.
- the online number comparison module 708c is configured to select, from the valid user account set, a user account whose online number is higher than the preset number of times in the preset time period, and obtain a predicted arrival user account set.
- the user account marked as invalid is deleted from the target user account set to obtain a valid user account set, and then the predicted arrival user account set is extracted from the valid user account set according to the historical online information.
- the utilization of historical online information is improved, and the accuracy of selecting the predicted arrival user account set is also improved.
- the click account selection module 712 specifically includes the following contents: a push message acquisition module 712a, a keyword extraction module 712b, a degree of interest calculation module 712c, a click rate statistics module 712d, and a click set extraction. Module 712e.
- the push message obtaining module 712a is configured to obtain a to-be-pushed message corresponding to the information push task.
- the keyword extraction module 712b is configured to extract a message keyword from the acquired message to be pushed.
- the interest degree calculation module 712c is configured to calculate, according to the acquired user interest tag and the extracted message keyword, the degree of interest corresponding to each user account in the predicted arrival user account set.
- the click rate statistics module 712d is configured to perform statistical prediction on the historical click rate corresponding to each user account in the user account set according to the obtained historical click information.
- the click collection extraction module 712e is configured to extract a predicted click user account set from the predicted arrival user account set according to the calculated interest degree and the statistical historical click rate.
- the click set extraction module 712e is also used to calculate the interest based on Degree and statistical history click rate, calculate the predicted click value corresponding to each user account in the user account set; from the predicted arrival user account set, select the user account whose click prediction value exceeds the preset threshold, and obtain the predicted click user account. set.
- the interest degree corresponding to the user account is calculated according to the user interest tag and the message keyword corresponding to the user account
- the historical click rate of the user account is calculated according to the historical click information
- the user account is extracted according to the degree of interest and the historical click rate.
- the push result prediction apparatus 700 specifically includes an adjustment suggestion generation module 716.
- the adjustment suggestion generating module 716 extracts the predicted arrival user account set in the push prediction result, and generates a target user account adjustment suggestion corresponding to the information push task according to the extracted predicted arrival user account set.
- the target user account adjustment proposal is generated according to the predicted arrival user account set in the push prediction result, and the user performing the information push task in the adjusted target user account set is adjusted according to the target user account, thereby ensuring that the message arrival amount can be improved.
- Each of the above-described push result prediction devices may be implemented in whole or in part by software, hardware, and a combination thereof.
- the network interface may be an Ethernet or a wireless network card.
- Each of the above modules may be embedded in a hardware form or independent of a processor in the server, or may be stored in a memory of the server in a software form, so that the processor calls to perform operations corresponding to the above modules.
- the processor can be a central processing unit (CPU), a microprocessor, a microcontroller, or the like.
- a computer device comprising a memory and a processor, wherein the memory stores computer executable instructions, the computer executable instructions being executed by the processor, such that the processor performs the step of: acquiring an information push task Extracting the target user account in the information push task, obtaining the target user account set; obtaining the historical online information corresponding to each user account in the target user account set; and selecting the online user account from the target user account set according to the obtained historical online information.
- the online user account is selected from the target user account set according to the obtained historical online information, and the predicted arrival user account set is obtained, including: the user account marked as invalid in the obtained historical online information, Deleting the user account marked as invalid from the target user account set to obtain a valid user account set; and collecting the online number of the preset time period corresponding to each user account in the valid user account set according to the obtained historical online information; In the user account set, the user account whose online number is higher than the preset number of times in the preset time period is selected, and the predicted arrival user account set is obtained.
- extracting the predicted click user account set from the predicted arrival user account set according to the obtained user interest tag and the historical click information including: obtaining the to-be-push message corresponding to the information push task; and obtaining the to-be-pushed message Extracting a message keyword from the message; calculating, according to the obtained user interest tag and the extracted message keyword, the degree of interest corresponding to each user account in the predicted arrival user account set; and statistically predicting the arrival according to the obtained historical click information
- the historical click rate corresponding to each user account in the user account set; and the predicted click user account set is extracted from the predicted arrival user account set according to the calculated interest degree and the statistical historical click rate.
- the predicted click user account set is extracted from the predicted arrival user account set according to the calculated interest degree and the statistical historical click rate, including: the calculated interest degree and the statistical history click. Rate, calculate the predicted click value corresponding to each user account in the set of user accounts; from the predicted arrival user account set, select a user account whose click prediction value exceeds a preset threshold, and obtain a predicted click user account set.
- the processor after generating the push prediction result according to the predicted arrival user account set and the predicted click user account set, performs the following steps: extracting the predicted arrival user account set in the push prediction result; and reaching the user according to the extracted prediction.
- the account set generates a target user account adjustment suggestion corresponding to the information push task.
- the target user account in the information push task is extracted, and the target user account set is obtained.
- the accuracy of predicting the arrival of the user account set and the predicted click user account set is ensured, thereby ensuring the generated push prediction result.
- the target user account in the information push task can be adjusted according to the push prediction result, thereby ensuring that the amount of arrival and the amount of clicks corresponding to the information push task can be improved, thereby improving the push of the information push task. power.
- One or more storage media storing computer-executable instructions that, when executed by one or more processors, cause one or more processors to perform the steps of: obtaining an information push task; extracting information in a push task
- the target user account obtains the target user account set; obtains the historical online information corresponding to each user account in the target user account set; and selects the online user account from the target user account set according to the obtained historical online information, and obtains the predicted arrival user account set.
- the user account set and the predicted click user account set generate a push prediction result.
- the online user account is selected from the target user account set according to the obtained historical online information, and the predicted arrival user account set is obtained, including: the user account marked as invalid in the obtained historical online information, Deleting the user account marked as invalid from the target user account set to obtain a valid user account set; and collecting the online number of the preset time period corresponding to each user account in the valid user account set according to the obtained historical online information; In the user account set, the user account whose online number is higher than the preset number of times in the preset time period is selected, and the predicted arrival user account set is obtained.
- extracting the predicted click user account set from the predicted arrival user account set according to the obtained user interest tag and the historical click information including: obtaining the to-be-push message corresponding to the information push task; and obtaining the to-be-pushed message Extracting a message keyword from the message; calculating, according to the obtained user interest tag and the extracted message keyword, the degree of interest corresponding to each user account in the predicted arrival user account set; and statistically predicting the arrival according to the obtained historical click information
- the historical click rate corresponding to each user account in the user account set; and the predicted click user account set is extracted from the predicted arrival user account set according to the calculated interest degree and the statistical historical click rate.
- the predicted click user account set is extracted from the predicted arrival user account set according to the calculated interest degree and the statistical historical click rate, including: the calculated interest degree and the statistical history click. Rate, calculate the predicted click value corresponding to each user account in the user account set; from the predicted arrival user account set, select the click predicted value to exceed the pre- Set a threshold user account to get a predicted click user account set.
- the processor after generating the push prediction result according to the predicted arrival user account set and the predicted click user account set, the processor further performs the steps of: extracting the predicted arrival user account set in the push prediction result; and arriving according to the extracted prediction.
- the user account set generates a target user account adjustment suggestion corresponding to the information push task.
- the target user account in the information push task is extracted, and the target user account set is obtained.
- the accuracy of predicting the arrival of the user account set and the predicted click user account set is ensured, thereby ensuring the generated push prediction result.
- the target user account in the information push task can be adjusted according to the push prediction result, thereby ensuring that the arrival amount and the click amount corresponding to the information push task can be improved, thereby improving the push success rate of the information push task.
- all or part of the processes in the foregoing embodiment may be completed by instructing related hardware by computer executable instructions, which may be stored in a computer readable storage medium.
- the program may be stored in a storage medium of the computer system and executed by at least one processor in the computer system to implement a flow including an embodiment of the methods as described above.
- the storage medium includes, but is not limited to, a magnetic disk, a USB flash drive, an optical disk, a read-only memory (ROM), and the like.
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Information Transfer Between Computers (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Description
Claims (20)
- 一种推送结果预测方法,包括:获取信息推送任务;提取所述信息推送任务中的目标用户账号,得到目标用户账号集合;获取所述目标用户账号集合中各用户账号对应的历史在线信息;根据获取到的历史在线信息,从所述目标用户账号集合中选取在线用户账号,得到预测到达用户账号集合;获取所述预测到达用户账号集合中各用户账号对应的用户兴趣标签和历史点击信息;根据获取到的用户兴趣标签和历史点击信息,从所述预测到达用户账号集合中提取预测点击用户账号集合;及根据所述预测到达用户账号集合和所述预测点击用户账号集合,生成推送预测结果。
- 根据权利要求1所述的方法,其特征在于,所述根据获取到的历史在线信息,从所述目标用户账号集合中选取在线用户账号,得到预测到达用户账号集合,包括:将获取到的历史在线信息中被标记为无效的用户账号,从所述目标用户账号集合中删除,得到有效用户账号集合;根据所述获取到的历史在线信息,统计所述有效用户账号集合中各用户账号对应的预设时间段内在线次数;及从所述有效用户账号集合中,选取所述预设时间段内在线次数高于预设次数的用户账号,得到预测到达用户账号集合。
- 根据权利要求1所述的方法,其特征在于,所述根据获取到的用户兴趣标签和历史点击信息,从所述预测到达用户账号集合中提取预测点击用户账号集合,包括:获取所述信息推送任务对应的待推送消息;从获取到的待推送消息中提取消息关键词;根据获取到的用户兴趣标签和提取到的消息关键词,分别计算所述预测到达用户账号集合中各用户账号对应的感兴趣度;根据获取到的历史点击信息,统计所述预测到达用户账号集合中各用户账号对应的历史点击率;及根据计算得到的感兴趣度和统计到的历史点击率,从所述预测到达用户账号集合中提取预测点击用户账号集合。
- 根据权利要求3所述的方法,其特征在于,所述根据计算得到的感兴趣度和统计到的历史点击率,从所述预测到达用户账号集合中提取预测点击用户账号集合,包括:根据计算得到的感兴趣度和统计到的历史点击率,计算所述预测到达用户账号集合中各用户账号对应的点击预测值;及从所述预测到达用户账号集合中,选取点击预测值超过预设阈值的用户账号,得到预测点击用户账号集合。
- 根据权利要求1所述的方法,其特征在于,所述根据所述预测到达用户账号集合和所述预测点击用户账号集合,生成推送预测结果之后,所述方法还包括:提取所述推送预测结果中的预测到达用户账号集合;及根据提取到的预测到达用户账号集合,生成与所述信息推送任务对应的目标用户账号调整建议。
- 一种推送结果预测装置,包括:推送任务获取模块,用于获取信息推送任务;用户账号提取模块,用于提取所述信息推送任务中的目标用户账号,得到目标用户账号集合;历史信息获取模块,用于获取所述目标用户账号集合中各用户账号对应的历史在线信息;在线账号选取模块,用于根据获取到的历史在线信息,从所述目标用户 账号集合中选取在线用户账号,得到预测到达用户账号集合;标签信息获取模块,用于获取所述预测到达用户账号集合中各用户账号对应的用户兴趣标签和历史点击信息;点击账号选取模块,用于根据获取到的用户兴趣标签和历史点击信息,从所述预测到达用户账号集合中提取预测点击用户账号集合;及预测结果生成模块,用于根据所述预测到达用户账号集合和所述预测点击用户账号集合,生成推送预测结果。
- 根据权利要求6所述的装置,其特征在于,所述在线账号选取模块包括:有效账号获得模块,用于将获取到的历史在线信息中被标记为无效的用户账号,从所述目标用户账号集合中删除,得到有效用户账号集合;在线次数统计模块,用于根据所述获取到的历史在线信息,统计所述有效用户账号集合中各用户账号对应的预设时间段内在线次数;及在线次数比较模块,用于从所述有效用户账号集合中,选取所述预设时间段内在线次数高于预设次数的用户账号,得到预测到达用户账号集合。
- 根据权利要求6所述的方法,其特征在于,所述点击账号选取模块包括:推送消息获取模块,用于获取所述信息推送任务对应的待推送消息;关键词提取模块,用于从获取到的待推送消息中提取消息关键词;感兴趣度计算模块,用于根据获取到的用户兴趣标签和提取到的消息关键词,分别计算所述预测到达用户账号集合中各用户账号对应的感兴趣度;点击率统计模块,用于根据获取到的历史点击信息,统计所述预测到达用户账号集合中各用户账号对应的历史点击率;及点击集合提取模块,用于根据计算得到的感兴趣度和统计到的历史点击率,从所述预测到达用户账号集合中提取预测点击用户账号集合。
- 根据权利要求8所述的装置,其特征在于,所述点击集合提取模块还用于根据计算得到的感兴趣度和统计到的历史点击率,计算所述预测到达用 户账号集合中各用户账号对应的点击预测值;从所述预测到达用户账号集合中,选取点击预测值超过预设阈值的用户账号,得到预测点击用户账号集合。
- 根据权利要求6所述的装置,其特征在于,所述装置还包括:调整建议生成模块,用于提取所述推送预测结果中的预测到达用户账号集合;根据提取到的预测到达用户账号集合,生成与所述信息推送任务对应的目标用户账号调整建议。
- 一种计算机设备,包括存储器和处理器,所述存储器中存储有计算机可执行指令,所述计算机可执行指令被所述处理器执行时,使得所述处理器执行以下步骤:获取信息推送任务;提取所述信息推送任务中的目标用户账号,得到目标用户账号集合;获取所述目标用户账号集合中各用户账号对应的历史在线信息;根据获取到的历史在线信息,从所述目标用户账号集合中选取在线用户账号,得到预测到达用户账号集合;获取所述预测到达用户账号集合中各用户账号对应的用户兴趣标签和历史点击信息;根据获取到的用户兴趣标签和历史点击信息,从所述预测到达用户账号集合中提取预测点击用户账号集合;及根据所述预测到达用户账号集合和所述预测点击用户账号集合,生成推送预测结果。
- 根据权利要求11所述的计算机设备,其特征在于,所述根据获取到的历史在线信息,从所述目标用户账号集合中选取在线用户账号,得到预测到达用户账号集合,包括:将获取到的历史在线信息中被标记为无效的用户账号,从所述目标用户账号集合中删除,得到有效用户账号集合;根据所述获取到的历史在线信息,统计所述有效用户账号集合中各用户 账号对应的预设时间段内在线次数;及从所述有效用户账号集合中,选取所述预设时间段内在线次数高于预设次数的用户账号,得到预测到达用户账号集合。
- 根据权利要求11所述的计算机设备,其特征在于,所述根据获取到的用户兴趣标签和历史点击信息,从所述预测到达用户账号集合中提取预测点击用户账号集合,包括:获取所述信息推送任务对应的待推送消息;从获取到的待推送消息中提取消息关键词;根据获取到的用户兴趣标签和提取到的消息关键词,分别计算所述预测到达用户账号集合中各用户账号对应的感兴趣度;根据获取到的历史点击信息,统计所述预测到达用户账号集合中各用户账号对应的历史点击率;及根据计算得到的感兴趣度和统计到的历史点击率,从所述预测到达用户账号集合中提取预测点击用户账号集合。
- 根据权利要求13所述的计算机设备,其特征在于,所述根据计算得到的感兴趣度和统计到的历史点击率,从所述预测到达用户账号集合中提取预测点击用户账号集合,包括:根据计算得到的感兴趣度和统计到的历史点击率,计算所述预测到达用户账号集合中各用户账号对应的点击预测值;及从所述预测到达用户账号集合中,选取点击预测值超过预设阈值的用户账号,得到预测点击用户账号集合。
- 根据权利要求11所述的计算机设备,其特征在于,所述根据所述预测到达用户账号集合和所述预测点击用户账号集合,生成推送预测结果之后,所述处理器还执行以下步骤:提取所述推送预测结果中的预测到达用户账号集合;及根据提取到的预测到达用户账号集合,生成与所述信息推送任务对应的目标用户账号调整建议。
- 一个或多个存储有计算机可执行指令的存储介质,所述计算机可执行指令被一个或多个处理器执行时,使得所述一个或多个处理器执行以下步骤:获取信息推送任务;提取所述信息推送任务中的目标用户账号,得到目标用户账号集合;获取所述目标用户账号集合中各用户账号对应的历史在线信息;根据获取到的历史在线信息,从所述目标用户账号集合中选取在线用户账号,得到预测到达用户账号集合;获取所述预测到达用户账号集合中各用户账号对应的用户兴趣标签和历史点击信息;根据获取到的用户兴趣标签和历史点击信息,从所述预测到达用户账号集合中提取预测点击用户账号集合;及根据所述预测到达用户账号集合和所述预测点击用户账号集合,生成推送预测结果。
- 根据权利要求16所述的存储介质,其特征在于,所述根据获取到的历史在线信息,从所述目标用户账号集合中选取在线用户账号,得到预测到达用户账号集合,包括:将获取到的历史在线信息中被标记为无效的用户账号,从所述目标用户账号集合中删除,得到有效用户账号集合;根据所述获取到的历史在线信息,统计所述有效用户账号集合中各用户账号对应的预设时间段内在线次数;及从所述有效用户账号集合中,选取所述预设时间段内在线次数高于预设次数的用户账号,得到预测到达用户账号集合。
- 根据权利要求16所述的存储介质,其特征在于,所述根据获取到的用户兴趣标签和历史点击信息,从所述预测到达用户账号集合中提取预测点击用户账号集合,包括:获取所述信息推送任务对应的待推送消息;从获取到的待推送消息中提取消息关键词;根据获取到的用户兴趣标签和提取到的消息关键词,分别计算所述预测到达用户账号集合中各用户账号对应的感兴趣度;根据获取到的历史点击信息,统计所述预测到达用户账号集合中各用户账号对应的历史点击率;及根据计算得到的感兴趣度和统计到的历史点击率,从所述预测到达用户账号集合中提取预测点击用户账号集合。
- 根据权利要求18所述的存储介质,其特征在于,所述根据计算得到的感兴趣度和统计到的历史点击率,从所述预测到达用户账号集合中提取预测点击用户账号集合,包括:根据计算得到的感兴趣度和统计到的历史点击率,计算所述预测到达用户账号集合中各用户账号对应的点击预测值;及从所述预测到达用户账号集合中,选取点击预测值超过预设阈值的用户账号,得到预测点击用户账号集合。
- 根据权利要求16所述的存储介质,其特征在于,所述根据所述预测到达用户账号集合和所述预测点击用户账号集合,生成推送预测结果之后,所述处理器还执行以下步骤:提取所述推送预测结果中的预测到达用户账号集合;及根据提取到的预测到达用户账号集合,生成与所述信息推送任务对应的目标用户账号调整建议。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710608094.4 | 2017-07-24 | ||
CN201710608094.4A CN107666506B (zh) | 2017-07-24 | 2017-07-24 | 推送结果预测方法、装置、计算机设备和存储介质 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2019019396A1 true WO2019019396A1 (zh) | 2019-01-31 |
Family
ID=61122492
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2017/104871 WO2019019396A1 (zh) | 2017-07-24 | 2017-09-30 | 推送结果预测方法、装置、计算机设备和存储介质 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN107666506B (zh) |
WO (1) | WO2019019396A1 (zh) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111311015A (zh) * | 2020-02-28 | 2020-06-19 | 深圳前海微众银行股份有限公司 | 个性化点击率预测方法、设备及可读存储介质 |
CN111460319A (zh) * | 2020-03-31 | 2020-07-28 | 深圳前海微众银行股份有限公司 | 消息播放优化方法、装置、设备及可读存储介质 |
CN112733012A (zh) * | 2020-12-30 | 2021-04-30 | 杭州网易再顾科技有限公司 | 一种执行用户触达的方法、装置、设备和介质 |
CN113254759A (zh) * | 2020-02-12 | 2021-08-13 | 北京沃东天骏信息技术有限公司 | 点击率预测方法、装置、介质及电子设备 |
CN113726839A (zh) * | 2021-06-17 | 2021-11-30 | 武汉理工数字传播工程有限公司 | 一种虚拟资源发送方法、装置、设备及存储介质 |
CN114742479A (zh) * | 2022-06-10 | 2022-07-12 | 深圳竹云科技股份有限公司 | 账号识别方法、装置、服务器及存储介质 |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109218411B (zh) * | 2018-08-22 | 2022-10-18 | 中国平安人寿保险股份有限公司 | 数据处理方法及装置、计算机可读存储介质、电子设备 |
CN110213216B (zh) * | 2018-08-22 | 2024-02-09 | 腾讯科技(北京)有限公司 | 确定不同账号集合中相同账号的方法、装置及存储介质 |
CN109271420B (zh) * | 2018-09-03 | 2023-07-11 | 深圳平安医疗健康科技服务有限公司 | 信息推送方法、装置、计算机设备和存储介质 |
CN109558542B (zh) * | 2018-12-04 | 2020-11-27 | 北京达佳互联信息技术有限公司 | 一种信息质量评估方法、信息推送方法及装置 |
CN110399550B (zh) * | 2019-03-14 | 2023-08-15 | 腾讯科技(深圳)有限公司 | 一种信息推荐方法及装置 |
CN110674395B (zh) * | 2019-08-23 | 2022-07-05 | 深圳市新系区块链技术有限公司 | 信息推送方法、装置及设备 |
CN111079006B (zh) * | 2019-12-09 | 2024-05-10 | 腾讯科技(深圳)有限公司 | 一种消息推送方法、装置、电子设备及介质 |
CN111368552B (zh) * | 2020-02-26 | 2023-09-26 | 北京市公安局 | 一种面向特定领域的网络用户群组划分方法及装置 |
CN111753201A (zh) * | 2020-06-23 | 2020-10-09 | 厦门美柚股份有限公司 | 信息推送方法、装置、终端及介质 |
CN113704616B (zh) * | 2021-08-30 | 2023-10-27 | 康键信息技术(深圳)有限公司 | 信息推送方法、装置、电子设备及可读存储介质 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102592235A (zh) * | 2011-12-28 | 2012-07-18 | 北京品友互动信息技术有限公司 | 一种互联网广告投放*** |
CN103150663A (zh) * | 2013-02-18 | 2013-06-12 | 亿赞普(北京)科技有限公司 | 一种网络投放数据投放的方法和装置 |
CN105227429A (zh) * | 2014-06-25 | 2016-01-06 | 腾讯科技(深圳)有限公司 | 一种信息推送方法和装置 |
WO2016180138A1 (zh) * | 2015-05-11 | 2016-11-17 | 腾讯科技(深圳)有限公司 | 推广信息投放有效性确定方法、监测服务器及终端 |
CN106649774A (zh) * | 2016-12-27 | 2017-05-10 | 北京百度网讯科技有限公司 | 基于人工智能的对象推送方法及装置 |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101123520B (zh) * | 2006-08-07 | 2012-07-04 | 腾讯科技(深圳)有限公司 | 一种定向投放广告的方法和广告*** |
US20090006188A1 (en) * | 2007-06-26 | 2009-01-01 | Microsoft Corporation | Associating an activity with an online advertisement |
US8738443B2 (en) * | 2009-05-18 | 2014-05-27 | George M. Davison | System and method for advertising using pushed video |
CN105897847A (zh) * | 2015-12-15 | 2016-08-24 | 乐视网信息技术(北京)股份有限公司 | 一种信息推送方法及信息推送装置 |
CN106790570A (zh) * | 2016-12-27 | 2017-05-31 | 山东开创云软件有限公司 | 一种消费者行为分析管理***及其分析方法 |
-
2017
- 2017-07-24 CN CN201710608094.4A patent/CN107666506B/zh active Active
- 2017-09-30 WO PCT/CN2017/104871 patent/WO2019019396A1/zh active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102592235A (zh) * | 2011-12-28 | 2012-07-18 | 北京品友互动信息技术有限公司 | 一种互联网广告投放*** |
CN103150663A (zh) * | 2013-02-18 | 2013-06-12 | 亿赞普(北京)科技有限公司 | 一种网络投放数据投放的方法和装置 |
CN105227429A (zh) * | 2014-06-25 | 2016-01-06 | 腾讯科技(深圳)有限公司 | 一种信息推送方法和装置 |
WO2016180138A1 (zh) * | 2015-05-11 | 2016-11-17 | 腾讯科技(深圳)有限公司 | 推广信息投放有效性确定方法、监测服务器及终端 |
CN106649774A (zh) * | 2016-12-27 | 2017-05-10 | 北京百度网讯科技有限公司 | 基于人工智能的对象推送方法及装置 |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113254759A (zh) * | 2020-02-12 | 2021-08-13 | 北京沃东天骏信息技术有限公司 | 点击率预测方法、装置、介质及电子设备 |
CN111311015A (zh) * | 2020-02-28 | 2020-06-19 | 深圳前海微众银行股份有限公司 | 个性化点击率预测方法、设备及可读存储介质 |
CN111460319A (zh) * | 2020-03-31 | 2020-07-28 | 深圳前海微众银行股份有限公司 | 消息播放优化方法、装置、设备及可读存储介质 |
CN112733012A (zh) * | 2020-12-30 | 2021-04-30 | 杭州网易再顾科技有限公司 | 一种执行用户触达的方法、装置、设备和介质 |
CN112733012B (zh) * | 2020-12-30 | 2023-06-02 | 杭州网易再顾科技有限公司 | 一种执行用户触达的方法、装置、设备和介质 |
CN113726839A (zh) * | 2021-06-17 | 2021-11-30 | 武汉理工数字传播工程有限公司 | 一种虚拟资源发送方法、装置、设备及存储介质 |
CN113726839B (zh) * | 2021-06-17 | 2023-05-26 | 武汉理工数字传播工程有限公司 | 一种虚拟资源发送方法、装置、设备及存储介质 |
CN114742479A (zh) * | 2022-06-10 | 2022-07-12 | 深圳竹云科技股份有限公司 | 账号识别方法、装置、服务器及存储介质 |
CN114742479B (zh) * | 2022-06-10 | 2022-09-06 | 深圳竹云科技股份有限公司 | 账号识别方法、装置、服务器及存储介质 |
Also Published As
Publication number | Publication date |
---|---|
CN107666506A (zh) | 2018-02-06 |
CN107666506B (zh) | 2019-11-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2019019396A1 (zh) | 推送结果预测方法、装置、计算机设备和存储介质 | |
CN111339327B (zh) | 作品推荐方法及装置、服务器和存储介质 | |
EP3399434A1 (en) | Short link processing method, device and server | |
WO2015188733A1 (en) | Group user level association method and system | |
JP6986187B2 (ja) | 人物識別方法、装置、電子デバイス、記憶媒体、及びプログラム | |
CN107305611B (zh) | 恶意账号对应的模型建立方法和装置、恶意账号识别的方法和装置 | |
WO2019055506A1 (en) | PREDICTING NON-OBSERVABLE PARAMETERS FOR DIGITAL COMPONENTS | |
US10313746B2 (en) | Server, client and video processing method | |
JP6309539B2 (ja) | 音声入力を実現する方法および装置 | |
CN110413867B (zh) | 用于内容推荐的方法及*** | |
WO2019019652A1 (zh) | 推送信息的处理方法、装置、计算机设备和存储介质 | |
WO2019019642A1 (zh) | 应用信息推送方法、装置、计算机设备和存储介质 | |
CN111586126A (zh) | 小程序预下载方法、装置、设备及存储介质 | |
US10085046B2 (en) | Video transmission method, gateway device, and video transmission system | |
US9866454B2 (en) | Generating anonymous data from web data | |
WO2021190470A1 (zh) | 电子设备定位的方法、装置、服务器、***以及存储介质 | |
WO2019011295A1 (zh) | 负一屏内容推送及请求方法、装置、服务器及终端 | |
CN108401005B (zh) | 一种表情推荐方法和装置 | |
CN109885729B (zh) | 一种显示数据的方法、装置及*** | |
WO2019210585A1 (zh) | 消息推送方法、计算机可读存储介质、终端设备及装置 | |
CN106570003B (zh) | 数据推送方法及装置 | |
CN103365892A (zh) | 一种用于对多个联系对象进行处理的方法与设备 | |
US10924365B2 (en) | Method and system for generating directed graphs | |
CN110457089B (zh) | 数据采集方法、装置、计算机可读存储介质和计算机设备 | |
CN106294457B (zh) | 网络信息推送方法及装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17919137 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 17919137 Country of ref document: EP Kind code of ref document: A1 |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205N DATED 17/03/2020) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 17919137 Country of ref document: EP Kind code of ref document: A1 |