WO2019019387A1 - 信息推送建议生成方法、装置、计算机设备和存储介质 - Google Patents

信息推送建议生成方法、装置、计算机设备和存储介质 Download PDF

Info

Publication number
WO2019019387A1
WO2019019387A1 PCT/CN2017/104771 CN2017104771W WO2019019387A1 WO 2019019387 A1 WO2019019387 A1 WO 2019019387A1 CN 2017104771 W CN2017104771 W CN 2017104771W WO 2019019387 A1 WO2019019387 A1 WO 2019019387A1
Authority
WO
WIPO (PCT)
Prior art keywords
push
historical
task
suggestion
tasks
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.)
Ceased
Application number
PCT/CN2017/104771
Other languages
English (en)
French (fr)
Inventor
王强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
OneConnect Financial Technology Co Ltd Shanghai
Original Assignee
OneConnect Financial Technology Co Ltd Shanghai
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by OneConnect Financial Technology Co Ltd Shanghai filed Critical OneConnect Financial Technology Co Ltd Shanghai
Publication of WO2019019387A1 publication Critical patent/WO2019019387A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

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 OR CALCULATING; 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

Definitions

  • the present application relates to the field of information processing technologies, and in particular, to a method, an apparatus, a computer device, and a storage medium for generating information push suggestions.
  • the information is pushed according to the preset push parameters.
  • the user receives too much push information, and the user has annoyed the push information, resulting in a more information push effect. low.
  • an information push suggestion generation method is provided.
  • a method for generating information push suggestions including:
  • the selected historical push task is analyzed, and a push suggestion corresponding to the push suggestion analysis request is generated.
  • An information push suggestion generating device comprising:
  • condition extraction module configured to extract a recommended analysis condition in the push suggestion analysis request
  • the task query module is configured to query, in the historical push task of the record, a historical push task that meets the extracted recommended analysis condition;
  • a score obtaining module configured to obtain a push effect score corresponding to each historical push task that is queried
  • a task selection module configured to select a preset number of historical push tasks from the historical push tasks according to the obtained push effect scores
  • the suggestion generating module is configured to analyze the selected historical push task, and generate a push suggestion corresponding to the push suggestion analysis request.
  • 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:
  • the selected historical push task is analyzed, and a push suggestion corresponding to the push suggestion analysis request is generated.
  • 1 is an application environment diagram of a method for generating information push suggestions in an embodiment
  • FIG. 2 is a structural block diagram of a server in an information push suggestion generation system in an embodiment
  • FIG. 3 is a schematic flowchart of a method for generating information push suggestions in an embodiment
  • FIG. 4 is a schematic flow chart of steps of generating a push suggestion according to a historical push task in an embodiment
  • FIG. 5 is a schematic flow chart of steps for generating a push time period suggestion in an embodiment
  • FIG. 6 is a flow chart showing the steps of calculating a push effect score in one embodiment
  • FIG. 7 is a structural block diagram of an information push suggestion generating apparatus in an embodiment
  • FIG. 8 is a structural block diagram of a suggestion generation module in an embodiment
  • Figure 9 is a block diagram showing the structure of an information push suggestion generating apparatus in another embodiment.
  • FIG. 1 is an application environment diagram of a method for generating information push suggestions in an embodiment.
  • the information push suggestion generation method is applied to an information push suggestion generation system.
  • the information push suggestion generation 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. .
  • FIG. 2 is a schematic diagram showing the internal structure of the server 120 in the information push suggestion generation system of FIG. 1 in an embodiment.
  • 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 instruction can be executed by the processor to implement the information push suggestion generation 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 memory includes a non-volatile storage medium and an internal memory; the non-volatile storage medium stores an operating system, computer-executable instructions, and a database executable by the processor to implement
  • the above-mentioned information push recommendation generation method stores the push task data in the database; the internal memory provides a cache running environment for the operating system and computer executable instructions in the non-volatile storage medium.
  • the information push suggestion generation method can also be applied to the terminal 110.
  • FIG. 2 is only a schematic 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 More or fewer components are shown in Figure 2, or some components are combined, or have different component arrangements.
  • a method for generating an information push suggestion 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 operator of the push, generates a proposal analysis page request according to the obtained operator account, and sends a request analysis page request to the server 120.
  • the server 120 receives the request analysis page request sent by the terminal 110, parses the request analysis page request, and parses the operator account in the request analysis page request to verify whether the extracted operator account has access to the suggestion analysis page. After verifying that the extracted operator account has the right to access the suggestion analysis page, the server 120 obtains the suggestion analysis page data, and returns the obtained suggestion analysis page data to the terminal 110 corresponding to the suggestion analysis page request.
  • the terminal 110 receives the suggested analysis page data returned by the server 120, and displays a suggestion analysis page according to the suggested analysis page data.
  • the terminal 110 acquires a push suggestion analysis request triggered by the push operator on the suggestion analysis page, and sends the obtained push suggestion analysis request to the server 120.
  • the push suggestion analysis request is used to trigger the server 120 to generate a push suggestion request.
  • the server 120 parses the push suggestion analysis request, and extracts the suggestion analysis condition from the push suggestion analysis request by parsing.
  • the recommended analysis condition may specifically include at least one of a push time, an information type, and a user type.
  • a plurality of historical push tasks are recorded in the server 120, and the historical push tasks are push tasks that have been executed.
  • Historical push tasks include push time, type of information, and At least one of the user types.
  • the server 120 queries the historical push tasks that meet the extracted suggestion analysis conditions among the plurality of historical push tasks recorded.
  • the extracted analysis condition includes the information type, and the information type is the advertisement information type.
  • the historical push task corresponding to the query advertisement information type in the recorded historical push task of the server 120 is the historical push task that meets the extracted suggestion analysis condition.
  • the server 120 further stores a push effect score corresponding to each recorded historical push task.
  • the push effect score is a score reflecting the push effect of the historical push task. The higher the push effect score, the better the push effect.
  • the server 120 queries the historical push task that meets the extracted recommended analysis condition, and obtains the push effect score corresponding to each of the searched historical push tasks.
  • the server 120 calculates the average push score of the searched historical push task according to the obtained push effect score, and selects the historical push task whose push effect score is higher than the push effect average score. .
  • S312 Analyze the selected historical push task, and generate a push suggestion corresponding to the push suggestion analysis request.
  • the server 120 extracts at least one of the information type, the user type, and the push time in the historical push task, and generates and pushes suggestions according to the extracted information type, user type, and time. Analyze the push recommendations for the request.
  • the push suggestion may include at least one of a recommended push information type, a recommended push time period, and a suggested push user type.
  • S312 may specifically include the following content: obtaining selected historical pushes The user types corresponding to the tasks are respectively classified; the selected historical push tasks are classified according to the user type; and the push effect scores corresponding to the history push tasks of the user types are determined, and the user type selection suggestions corresponding to the push suggestion analysis requests are determined.
  • the server 120 obtains the user types corresponding to the selected historical push tasks, and classifies the selected historical push tasks according to the acquired user types to obtain historical push tasks for each user type.
  • the server 120 acquires the push effect score corresponding to the historical push task of each user type, and calculates the average push score corresponding to each user type according to the obtained push effect score.
  • the server 120 averages the push effects corresponding to each user type and each user type to generate a user type selection suggestion.
  • the server 120 selects the user type with the highest average push effect from each user type, and generates a user type selection suggestion according to the selected user type.
  • the recommended and analyzed condition in the push recommendation analysis request is extracted, and the historical push task that meets the extracted recommended analysis condition is queried in the recorded historical push task, and the push effect score corresponding to each historical push task according to the query is extracted.
  • the preset number of historical push tasks are selected to ensure that the selected historical push task matches the push suggestion analysis request, and the accuracy of the selected historical push task is improved.
  • the selected historical push task is analyzed, and the push suggestion corresponding to the push suggestion analysis request is generated by analysis, the rationality and accuracy of the obtained push suggestion are determined, and the information push according to the generated push suggestion is improved to improve the push effect of the information push.
  • S306 further includes a step of generating a push recommendation according to the historical push task, the step specifically including the following:
  • the server 120 after the server 120 does not query the historical push task that meets the extracted recommended analysis condition, the server 120 obtains the push effect score corresponding to each historical push task of the record.
  • the server 120 compares the recorded push effect scores corresponding to each historical push task, and sorts the recorded historical push tasks according to the size of the push effect score by comparing.
  • the server 120 may specifically sort the recorded historical push tasks in descending order of the push effect scores.
  • the server 120 can also sort the recorded historical push tasks in descending order of the promotion effect scores.
  • the server 120 selects the previous historical push tasks from the sorted historical push tasks, and selects the presets. The number of historical push tasks.
  • the server 120 sorts the recorded historical push tasks according to the push effect scores in descending order, selects the sorted historical push tasks from the sorted historical push tasks, and selects the preset number. Historical push task.
  • S408 Analyze the selected historical push task, and generate a push suggestion corresponding to the push suggestion analysis request.
  • the server 120 analyzes the selected historical push task, extracts the corresponding information type, push time, and user type in each selected historical push task, and statistically extracts the extracted information type, push time, and user type corresponding storage.
  • the stored information type, push time, and user type determine the optimal push time period and the most appropriate user type for each information type.
  • the server 120 generates a push recommendation corresponding to the push suggestion analysis request according to the optimal push time period and the most appropriate user type corresponding to each information type.
  • the push suggestion may also include a corresponding sample of the push information copy.
  • the server 120 transmits the generated push advice to the terminal 110.
  • a historical push task with a higher push effect score is selected from the recorded historical push task, and the selected historical push task is analyzed to generate a push. Suggest. All the historical push tasks recorded are used as the basic task data, and the push effect scores are selected from the recorded historical push tasks according to the push effect scores, and the accuracy of the push suggestions is also ensured.
  • the server 120 parses the selected historical push tasks, and extracts the push time in each selected historical push task by parsing. .
  • the server 120 stores the recorded push logs of each historical push task, and the push log includes the push time corresponding to each historical push task. After selecting the historical push task from the recorded historical push task, the server 120 queries the push log corresponding to each historical push task, and extracts the push time corresponding to the historical push task from the queryed push log.
  • the server 120 stores a preset push time period, and after obtaining the push time corresponding to each selected historical push task, determining a push time period to which the push time corresponding to each history push task belongs, according to the determination.
  • the push time period classifies the selected historical push tasks, and obtains historical push tasks corresponding to each push time period by classification.
  • S506 Determine, according to the push effect score corresponding to the historical push task of each push time period after the classification, the push time segment recommendation corresponding to the push suggestion analysis request.
  • the server 120 acquires the push effect scores of the historical push tasks for the historical push tasks corresponding to the push time segments, and calculates the push of the historical push tasks corresponding to the push time segments according to the push effect scores of the historical push tasks. The average score of the effect.
  • the server 120 compares the average of the push effects of the historical push tasks corresponding to each push time period, and compares and selects the push time periods with the higher average scores of the preset number of push effects.
  • the server 120 generates a push time segment proposal corresponding to the push suggestion analysis request according to the selected push time period.
  • the server 120 compares the average of the push effects of the historical push tasks corresponding to each push time period, compares and obtains the push time period corresponding to the highest push effect average score, and generates a push time period according to the obtained push time period. Suggest.
  • the push time period suggestion may further include a sample of the push information copy of the historical push task corresponding to the acquired push time.
  • the selected historical push task is classified according to the push time period to which the push time of each historical push task belongs, and the obtained historical push task corresponding to each push time period is historically pushed for each push time period.
  • the push effect score of the task is analyzed to determine the push time.
  • the inter-segment recommendation ensures the accuracy of the push time period recommendation, and ensures that the push of the information according to the push time period can improve the push effect.
  • the information push suggesting method further includes the step of calculating a push effect score, and the step specifically includes the following:
  • the server 120 extracts the task identifier in the executed information push task, and stores the extracted task identifier in association with the executed information push task to obtain a recorded history push task.
  • the server 120 when the server 120 performs the information push task, the server 120 pushes the information to be pushed in the information push task to the terminal corresponding to each user account.
  • the terminal 110 returns the push feedback information to the server 120.
  • the push feedback information includes task identification, push arrival information, push click information, and push browsing information.
  • the server 120 parses the push feedback information, and extracts the task identifier, the push arrival information, the push click information, and the push browse information in the push feedback information by parsing.
  • the server 120 acquires the total amount of pushes of the historical push task corresponding to the extracted task identifier, and pushes the historical push task corresponding to the extracted push arrival information, the push browsing information, and the obtained push total statistical task identifier according to the extracted push arrival information. Performance data.
  • the push effect data includes push arrival rate, push click rate, arrival rate within 1 hour, click rate within 1 hour, and push browse rate.
  • the server 120 counts the push arrival rate and the arrival rate within one hour according to the push arrival information and the push total amount, and pushes the click rate and the one-hour click rate according to the push click information and the push total amount, and pushes according to the push browsing information and the push total amount. Browsing rate.
  • the server 120 obtains the weight corresponding to the push effect data, performs weighting calculation according to the statistically pushed weight effect data of each historical push task and the weight value corresponding to the obtained push effect data, and obtains the historical push task.
  • Push effect score the weight corresponding to the push effect data
  • the server 120 respectively obtains the weights corresponding to the push arrival rate, the push click rate, the one hour arrival rate, the one hour click rate, and the push browse rate in the push effect data, according to the obtained weights and
  • the pushed push effect data corresponding to each historical push task is weighted and calculated, and the calculated values are standardized to obtain a push effect score corresponding to each historical push task.
  • the historical push record is recorded, and the push effect data of each historical push task is counted according to the pushed push feedback information corresponding to the obtained historical push task, and the push effect score of each historical push task is calculated according to the statistical push effect data.
  • the feedback information By pushing the feedback information to statistically push the effect data, the authenticity of the push effect data is verified, thereby ensuring the accuracy of the push effect score calculated according to the push effect data.
  • an information push suggestion generating apparatus 700 is provided.
  • the apparatus specifically includes the following contents: a request obtaining module 702, a condition extracting module 704, a task querying module 706, a score obtaining module 708, and a task.
  • Module 710 and suggestion generation module 712 are selected.
  • the request obtaining module 702 is configured to obtain a push suggestion analysis request.
  • the condition extraction module 704 is configured to extract the recommended analysis condition in the push suggestion analysis request.
  • the task querying module 706 is configured to query, in the recorded historical push task, a historical push task that meets the extracted recommended analysis condition.
  • the score obtaining module 708 is configured to obtain a push effect score corresponding to each of the searched historical push tasks.
  • the task selection module 710 is configured to select a preset number of historical push tasks from each historical push task according to the obtained push effect score.
  • the suggestion generating module 712 is configured to analyze the selected historical push task, and generate a push suggestion corresponding to the push suggestion analysis request.
  • the suggestion generating module 712 is further configured to obtain a user type corresponding to each selected historical push task; and classify each selected historical push task according to the user type; According to the push effect score corresponding to the historical push task of each user type, the user type selection suggestion corresponding to the push suggestion analysis request is determined.
  • the score acquisition module 708 is further configured to: if the historical push task that meets the extracted suggestion analysis condition is not queried, obtain a push effect score corresponding to each of the recorded historical push tasks.
  • the task selection module 710 is further configured to sort the recorded historical push tasks according to the obtained push effect scores in descending order; and select a preset number of previous historical push tasks from the sorted historical push tasks.
  • a historical push task with a higher push effect score is selected from the recorded historical push task, and the selected historical push task is analyzed to generate a push. Suggest. All the historical push tasks recorded are used as the basic task data, and the push effect scores are selected from the recorded historical push tasks according to the push effect scores, and the accuracy of the push suggestions is also ensured.
  • the suggestion generation module 712 specifically includes the following: a push time acquisition module 712a, a push task assignment module 712b, and a push time suggestion module 712c.
  • the push time obtaining module 712a is configured to obtain a push time corresponding to each selected historical push task.
  • the push task assignment module 712b is configured to classify the selected historical push tasks according to the push time period to which the push time belongs.
  • the push time suggesting module 712c is configured to determine a push time period recommendation corresponding to the push suggestion analysis request according to the push effect score corresponding to the historical push task of each push time period after the classification.
  • the push time period to which the push time of each historical push task is selected belongs.
  • the selected historical push task is classified, and the obtained historical push task corresponding to each push time period is analyzed, and the push effect score of the historical push task corresponding to each push time period is analyzed, and the push time period suggestion is determined to ensure the push time period suggestion.
  • the accuracy ensures that the push of information according to the push time period can improve the push effect.
  • the information push suggestion generating apparatus 700 specifically includes the following: a task recording module 714 , an information acquiring module 716 , a data statistics module 718 , and a data computing module 720 .
  • the task record module 714 is configured to record the executed information push task as a historical push task
  • the information obtaining module 716 is configured to obtain the push feedback information corresponding to the recorded historical push task.
  • the data statistics module 718 is configured to statistically record the push effect data corresponding to the historical push task according to the obtained push feedback information
  • the data calculation module 720 is configured to calculate a push effect score of the recorded historical push task according to the statistical push effect data.
  • the historical push record is recorded, and the push effect data of each historical push task is counted according to the pushed push feedback information corresponding to the obtained historical push task, and the push effect score of each historical push task is calculated according to the statistical push effect data.
  • the feedback information By pushing the feedback information to statistically push the effect data, the authenticity of the push effect data is verified, thereby ensuring the accuracy of the push effect score calculated according to the push effect data.
  • Each of the above-described information push suggestion generating means 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 apparatus comprising a memory and a processor, wherein the memory stores computer executable instructions, the computer executable instructions being executed by the processor, causing the processor to perform the step of: obtaining a push suggestion analysis request Extracting the recommended analysis condition in the push suggestion analysis request; in the historical push task of the record, querying the historical push task that meets the extracted recommended analysis condition; obtaining the push effect score corresponding to each of the searched historical push tasks; The push effect score is obtained, and a preset number of historical push tasks are selected from each historical push task; the selected historical push task is analyzed, and a push suggestion corresponding to the push suggestion analysis request is generated.
  • the processor further performs the following steps: if the historical push task that meets the extracted recommended analysis condition is not queried And obtaining the corresponding push effect scores of each historical push task of the record; sorting the historical push tasks of the records according to the obtained push effect scores from high to low; and selecting presets from the sorted historical push tasks The quantity is sorted by the previous historical push task; the selected historical push task is analyzed, and a push suggestion corresponding to the push suggestion analysis request is generated.
  • the selected historical push task is analyzed, and the push suggestion corresponding to the push suggestion analysis request is generated, including: obtaining the push time corresponding to each selected historical push task; and selecting according to the push time segment to which the push time belongs Each historical push task is classified; and according to the push effect score corresponding to the historical push task of each push time period after the classification, the push time segment suggestion corresponding to the push suggestion analysis request is determined.
  • the selected historical push task is analyzed, and the push suggestion corresponding to the push suggestion analysis request is generated, including: obtaining the user type corresponding to each selected historical push task; and selecting each historical push according to the user type.
  • the task is classified; according to the push effect score corresponding to the historical push task of each user type, the user type selection suggestion corresponding to the push suggestion analysis request is determined.
  • the processor further performs the steps of: recording the executed information push task as a historical push task; acquiring the pushed push feedback information corresponding to the recorded historical push task; and performing historical push task according to the obtained push feedback information. Corresponding push effect data; calculating the push effect score of the recorded historical push task according to the statistical push effect data.
  • the recommended and analyzed condition in the push recommendation analysis request is extracted, and the historical push task that meets the extracted recommended analysis condition is queried in the recorded historical push task, and the push effect score corresponding to each historical push task according to the query is extracted.
  • the preset number of historical push tasks are selected to ensure that the selected historical push task matches the push suggestion analysis request, and the accuracy of the selected historical push task is improved. Analyze the selected historical push task, analyze and generate the push suggestion corresponding to the push suggestion analysis request, determine the rationality and accuracy of the obtained push suggestion, and ensure the root Pushing information according to the generated push suggestions improves the push effect of information push.
  • 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 a push suggestion analysis request; extracting push recommendations The recommended analysis condition in the request is analyzed; in the historical push task of the record, the historical push task that meets the extracted recommended analysis condition is queried; the push effect score corresponding to each historical push task obtained by the query is obtained; and the obtained push effect is obtained according to the obtained The score is selected from a historical push task by a preset number of historical push tasks; the selected historical push task is analyzed, and a push suggestion corresponding to the push suggestion analysis request is generated.
  • the processor further performs the following steps: if the historical push task that meets the extracted recommended analysis condition is not queried And obtaining the corresponding push effect scores of each historical push task of the record; sorting the historical push tasks of the records according to the obtained push effect scores from high to low; and selecting presets from the sorted historical push tasks The quantity is sorted by the previous historical push task; the selected historical push task is analyzed, and a push suggestion corresponding to the push suggestion analysis request is generated.
  • the selected historical push task is analyzed, and the push suggestion corresponding to the push suggestion analysis request is generated, including: obtaining the push time corresponding to each selected historical push task; and selecting according to the push time segment to which the push time belongs Each historical push task is classified; and according to the push effect score corresponding to the historical push task of each push time period after the classification, the push time segment suggestion corresponding to the push suggestion analysis request is determined.
  • the selected historical push task is analyzed, and the push suggestion corresponding to the push suggestion analysis request is generated, including: obtaining the user type corresponding to each selected historical push task; and selecting each historical push according to the user type.
  • the task is classified; according to the push effect score corresponding to the historical push task of each user type, the user type selection suggestion corresponding to the push suggestion analysis request is determined.
  • the processor further performs the steps of: recording the executed information push task as a historical push task; acquiring the pushed push feedback information corresponding to the recorded historical push task; and performing historical push task according to the obtained push feedback information.
  • Corresponding push effect data according to statistics The push effect data calculates the push effect score of the historical push task recorded.
  • the recommended and analyzed condition in the push recommendation analysis request is extracted, and the historical push task that meets the extracted recommended analysis condition is queried in the recorded historical push task, and the push effect score corresponding to each historical push task according to the query is extracted.
  • the preset number of historical push tasks are selected to ensure that the selected historical push task matches the push suggestion analysis request, and the accuracy of the selected historical push task is improved.
  • the selected historical push task is analyzed, and the push suggestion corresponding to the push suggestion analysis request is generated by analysis, the rationality and accuracy of the obtained push suggestion are determined, and the information push according to the generated push suggestion is improved to improve the push effect of the information push.
  • 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)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Debugging And Monitoring (AREA)
  • Medical Treatment And Welfare Office Work (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种信息推送建议生成方法，所述方法包括：获取推送建议分析请求；提取所述推送建议分析请求中的建议分析条件；在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务；获取查询到的各历史推送任务分别对应的推送效果分数；根据获取到的推送效果分数，从所述各历史推送任务中选取预设数量的历史推送任务；对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议。

Description

信息推送建议生成方法、装置、计算机设备和存储介质
本申请要求于2017年7月24日提交中国专利局，申请号为2017106077068，发明名称为“信息推送建议生成方法、装置、计算机设备和存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及信息处理技术领域，特别是涉及一种信息推送建议生成方法、装置、计算机设备和存储介质。
背景技术
随着互联网技术的迅猛发展，各种信息可以通过互联网向用户推送，以便用户可以及时获取到相关的信息，随着消息推送需求的不断增加，通过互联网进行信息推送的任务量也增加，用户收到推送信息也随之增加。
传统的推送过程，都是根据预设推送参数对信息进行推送，随着信息推送量的增加，用户接收到的太多的推送信息，用户对推送信息反而产生了厌烦情绪，导致信息推送效果较低。
发明内容
根据本申请的各种实施例，提供一种信息推送建议生成方法、装置、计算机设备和存储介质。
一种信息推送建议生成方法，包括：
获取推送建议分析请求；
提取所述推送建议分析请求中的建议分析条件；
在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务；
获取查询到的各历史推送任务分别对应的推送效果分数；
根据获取到的推送效果分数，从所述各历史推送任务中选取预设数量的 历史推送任务；及
对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议。
一种信息推送建议生成装置，所述装置包括：
请求获取模块，用于获取推送建议分析请求；
条件提取模块，用于提取所述推送建议分析请求中的建议分析条件；
任务查询模块，用于在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务；
分数获取模块，用于获取查询到的各历史推送任务分别对应的推送效果分数；
任务选取模块，用于根据获取到的推送效果分数，从所述各历史推送任务中选取预设数量的历史推送任务；及
建议生成模块，用于对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议。
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：
获取推送建议分析请求；
提取所述推送建议分析请求中的建议分析条件；
在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务；
获取查询到的各历史推送任务分别对应的推送效果分数；
根据获取到的推送效果分数，从所述各历史推送任务中选取预设数量的历史推送任务；及
对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的 推送建议。
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：
获取推送建议分析请求；
提取所述推送建议分析请求中的建议分析条件；
在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务；
获取查询到的各历史推送任务分别对应的推送效果分数；
根据获取到的推送效果分数，从所述各历史推送任务中选取预设数量的历史推送任务；及
对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征、目的和优点将从说明书、附图以及权利要求书变得明显。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它的附图。
图1为一个实施例中信息推送建议生成方法的应用环境图；
图2为一个实施例中信息推送建议生成系统中的服务器的结构框图；
图3为一个实施例中信息推送建议生成方法的流程示意图；
图4为一个实施例中根据历史推送任务生成推送建议的步骤的流程示意图；
图5为一个实施例中生成推送时间段建议的步骤的流程示意图；
图6为一个实施例中计算推送效果分数的步骤的流程示意图；
图7为一个实施例中信息推送建议生成装置的结构框图；
图8为一个实施例中建议生成模块的结构框图；
图9为另一个实施例中信息推送建议生成装置的结构框图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
图1为一个实施例中信息推送建议生成方法的应用环境图。参照图1，该信息推送建议生成方法应用于信息推送建议生成系统。信息推送建议生成系统包括终端110和服务器120，其中终端110通过网络与服务器120连接。终端110可以是固定终端或移动终端，固定终端具体可以是打印机、扫描仪和监控器中的至少一种，移动终端具体可以是平板电脑、智能手机、个人数据助理和数码相机中的至少一种。
图2为一个实施例中图1信息推送建议生成系统中的服务器120的内部结构示意图。如图2所示，该服务器120包括通过系统总线连接的处理器、非易失性存储介质、内存储器和网络接口。服务器120的处理器用于提供计算和控制能力，支撑整个服务器120的运行，存储器用于存储数据、代码指令等，网络接口用于与终端110进行网络通信。存储器上存储有至少一个计算机可执行指令，该计算机可执行指令可被处理器执行，以实现本申请实施例中提供的适用于服务器120的信息推送建议生成方法。存储器可包括磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)等非易失性存储介质。例如，在一个实施例中，存储器包括非易失性存储介质及内存储器；非易失性存储介质存储有操作系统、计算机可执行指令和数据库，该计算机可执行指令可被处理器执行以实现上述的信信息推送建议生成方法，数据库中存储着推送任务数据；内存储器为非易失性存储介质中的操作系统及计算机可执行指令提供高速缓存的运行环境。信息推送建议生成方法还可以适用于终端 110。
本领域技术人员可以理解，图2中示出的结构，仅仅是与本申请方案相关的部分结构的示意图，并不构成对本申请方案所应用于其上的服务器的限定，具体的服务器可以包括比图2中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
如图3所示，在一个实施例中，提供一种信息推送建议生成方法，本实施例以该方法应用于图1中的服务器120来举例说明，该方法具体包括以下内容：
S302，获取推送建议分析请求。
在一个实施例中，终端110获取推送运营人员输入的运营人员账号，根据获取到的运营人员账号生成建议分析页面请求，将建议分析页面请求发送至服务器120。
服务器120接收终端110发送的建议分析页面请求，对建议分析页面请求进行解析，通过解析提取建议分析页面请求中的运营人员账号，验证提取到的运营人员账号是否有访问建议分析页面的权限。服务器120在验证提取到的运营人员账号有访问建议分析页面的权限后，获取建议分析页面数据，将获取到的建议分析页面数据返回至建议分析页面请求对应的终端110。
终端110接收到服务器120返回的建议分析页面数据，根据建议分析页面数据显示建议分析页面。终端110获取推送运营人员在建议分析页面触发的推送建议分析请求，将获取的推送建议分析请求发送至服务器120。推送建议分析请求用于触发服务器120生成推送建议的请求。
S304，提取推送建议分析请求中的建议分析条件。
在一个实施例中，服务器120在接收到推送建议分析请求后，对推送建议分析请求进行解析，通过解析从推送建议分析请求中提取建议分析条件。建议分析条件具体可以包括推送时间、信息类型和用户类型中的至少一种。
S306，在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务。
在一个实施例中，服务器120中记录有多个历史推送任务，历史推送任务为已经执行完毕的推送任务。历史推送任务中包括推送时间、信息类型和 用户类型中的至少一种。服务器120在记录的多个历史推送任务中查询符合提取到的建议分析条件的历史推送任务。
举例说明，提取到的建议分析条件中包括信息类型，信息类型为广告信息类型。服务器120在记录的历史推送任务中的查询广告信息类型对应的历史推送任务，查询到的历史推送任务为符合提取到的建议分析条件的历史推送任务。
S308，获取查询到的各历史推送任务分别对应的推送效果分数。
在一个实施例中，服务器120中还存储着记录的各历史推送任务分别对应的推送效果分数。推送效果分数为反映历史推送任务的推送效果的分数，推送效果分数越高，表示推送效果越好。服务器120在查询到符合提取到的建议分析条件的历史推送任务，获取查询到的各历史推送任务分别对应的推送效果分数。
S310，根据获取到的推送效果分数，从各历史推送任务中选取预设数量的历史推送任务。
在一个实施例中，服务器120在获取到的各历史推送任务的推送效果分数，将各历史推送任务的推送效果分数进行比较，选取推送效果分数最高的历史推送任务，再从剩下的历史推送任务中选取推送效果分数最高的历史推送任务，直至选取的历史推送任务的数量达到预设数量为止。
在一个实施例中，服务器120在获取到推送效果分数后，根据获取到的推送效果分数计算查询到的历史推送任务的推送效果平均分，选取推送效果分数高于推送效果平均分的历史推送任务。
S312，对选取的历史推送任务进行分析，生成与推送建议分析请求对应的推送建议。
在一个实施例中，服务器120在选取历史推送任务后，提取历史推送任务中的信息类型、用户类型和推送时间中的至少一种，根据提取到的信息类型、用户类型和时间生成与推送建议分析请求对应的推送建议。推送建议可以包括建议推送信息类型、建议推送时间段和建议推送用户类型中的至少一种。
在一个实施例中，S312具体可以包括以下内容：获取选取的各历史推送 任务分别对应的用户类型；按照用户类型对选取的各历史推送任务进行分类；根据各用户类型的历史推送任务所对应的推送效果分数，确定推送建议分析请求对应的用户类型选取建议。
在一个实施例中，服务器120获取选取的各历史推送任务分别对应的用户类型，根据获取的用户类型对选取的各历史推送任务进行分类，得到各用户类型的历史推送任务。服务器120获取各用户类型的历史推送任务所对应的推送效果分数，根据获取到的推送效果分数计算各用户类型对应的推送效果平均分。服务器120将各用户类型和各用户类型对应的推送效果平均分生成用户类型选取建议。
在一个实施例中，服务器120将各用户类型对应的推送效果平均分进行比较，通过比较选取预设数量的推送效果平均分较高的用户类型，根据选取的用户类型和选取的用户类型对应的推送效果平均分用户类型选取建议。
在另一个实施例中，服务器120从各用户类型中选取推送效果平均分最高的用户类型，根据选取的用户类型生成用户类型选取建议。
本实施例中，提取推送建议分析请求中的及建议分析条件，在记录的历史推送任务中查询符合提取到的建议分析条件的历史推送任务，根据查询到的各历史推送任务对应的推送效果分数，选取预设数量的历史推送任务，保证选取的历史推送任务与推送建议分析请求相匹配，提高了选取的历史推送任务的准确性。对选取的历史推送任务进行分析，通过分析生成与推送建议分析请求对应的推送建议，确定得到的推送建议的合理性和准确性，保证根据生成的推送建议进行信息推送提高信息推送的推送效果。
如图4所示，在一个实施例中，S306之后还包括根据历史推送任务生成推送建议的步骤，该步骤具体包括以下内容：
S402，若未查询到符合提取到的建议分析条件的历史推送任务，则获取记录的各历史推送任务分别对应的推送效果分数。
在一个实施例中，服务器120在记录的历史推送任务中未查询到符合提取到的建议分析条件的历史推送任务后，则获取记录的各历史推送任务分别对应的推送效果分数。
S404，按照获取到的推送效果分数由高到低的顺序对记录的各历史推送 任务进行排序。
在一个实施例中，服务器120将记录的各历史推送任务对应的推送效果分数进行比较，通过比较根据推送效果分数的大小对记录的各历史推送任务排序。服务器120具体可以按照推送效果分数由高到低的顺序对记录的各历史推送任务进行排序。服务器120还可以按照推行效果分数由低到高的顺序对记录的各历史推送任务进行排序。
S406，从排序后的历史推送任务中选取预设数量排序在前的历史推送任务。
在一个实施例中，服务器120在对记录的各历史推送任务按照推送效果分数由高到低的顺序排序后，从排序后的历史推动任务中选取排序在前的历史推送任务，且选取预设数量的历史推送任务。
在一个实施例中，服务器120对记录的各历史推送任务按照推送效果分数由低到高的顺序排序后，从排序后的历史推送任务中选取排序在后的历史推送任务，且选取预设数量的历史推送任务。
S408，对选取的历史推送任务进行分析，生成与推送建议分析请求对应的推送建议。
在一个实施例中，服务器120对选取的历史推送任务进行分析，提取选取的各历史推送任务中对应的信息类型、推送时间和用户类型，统计提取到的信息类型、推送时间和用户类型对应存储，存储的信息类型、推送时间和用户类型确定各信息类型对应的最佳推送时间段和最合适的用户类型。服务器120根据各信息类型对应的最佳推送时间段和最合适的用户类型生成与推送建议分析请求对应的推送建议。推送建议中还可以包括对应的推送信息文案样例。服务器120将生成的推送建议发送至终端110。
本实施例中，在未查询到符合提取到的建议分析条件的历史推送任务时，从记录的历史推送任务选取推送效果分数较高的历史推送任务，对选取的历史推送任务进行分析，生成推送建议。以记录的所有历史推送任务作为基础任务数据，根据推送效果分数从记录的历史推送任务中选取推送效果分数较高历史推送任务，同样保证了推送建议的准确性。
如图5所示，在一个实施例中，S312具体还包括生成推送时间段建议的 步骤，该步骤具体包括以下内容：
S502，获取选取的各历史推送任务分别对应的推送时间。
在一个实施例中，服务器120在从记录的历史推送任务中选取预设数量的历史推送任务之后，对选取到的各历史推送任务进行解析，通过解析提取选取的各历史推送任务中的推送时间。
在一个实施例中，服务器120中存储着记录的各历史推送任务的推送日志，推送日志中包括各历史推送任务对应的推送时间。服务器120从记录的历史推送任务中选取历史推送任务后，查询选取到的各历史推送任务对应的推送日志，从查询到的推送日志中提取历史推送任务对应的推送时间。
S504，按照推送时间所属推送时间段对选取的各历史推送任务进行分类。
在一个实施例中，服务器120中存储着预设的推送时间段，在获取到选取的各历史推送任务对应的推送时间后，确定各历史推送任务对应的推送时间所属的推送时间段，根据确定的推送时间段对选取到的历史推送任务进行分类，通过分类得到各推送时间段对应的历史推送任务。
S506，根据分类后各推送时间段的历史推送任务所对应的推送效果分数，确定推送建议分析请求对应的推送时间段建议。
在一个实施例中，服务器120对于各推送时间段对应的历史推送任务，获取各历史推送任务的推送效果分数，根据各历史推送任务的推送效果分数计算各推送时间段对应的历史推送任务的推送效果平均分。服务器120将各推送时间段对应的历史推送任务的推送效果平均分进行比较，通过比较选取预设数量的推送效果平均分较高的推送时间段。服务器120根据选取的推送时间段生成推送建议分析请求对应的推送时间段建议。
在一个实施例中，服务器120将各推送时间段对应的历史推送任务的推送效果平均分进行比较，通过比较获取最高推送效果平均分对应的推送时间段，根据获取的推送时间段生成推送时间段建议。推送时间段建议还可以包括获取的推送时间对应的历史推送任务的推送信息文案样例。
本实施例中，根据选取的各历史推送任务的推送时间所属的推送时间段对选取的历史推送任务进行分类，得到的各推送时间段对应的历史推送任务，对各推送时间段对应的历史推送任务的推送效果分数进行分析，确定推送时 间段建议，确保了推送时间段建议的准确性，保证根据推送时间段进行信息推送可以提高推送效果。
如图6所示，在一个实施例中，信息推送建议生成方法具体还包括计算推送效果分数的步骤，该步骤具体包括以下内容：
S602，记录已执行的信息推送任务为历史推送任务。
在一个实施例中，服务器120在执行完信息推送任务后，将已执行的信息推送任务作为历史推送任务进行记录，得到记录的历史推送任务。
在一个实施例中，服务器120提取已执行的信息推送任务中的任务标识，将提取到的任务标识与已执行的信息推送任务对应存储，得到记录的历史推送任务。
S604，获取记录的历史推送任务对应的推送反馈信息。
在一个实施例中，服务器120在执行信息推送任务时，将信息推送任务中的待推送信息推送至各用户账号分别对应的终端。终端110向服务器120返回推送反馈信息。推送反馈信息中包括任务标识、推送到达信息、推送点击信息和推送浏览信息。
S606，根据获取到的推送反馈信息统计记录的历史推送任务对应的推送效果数据。
在一个实施例中，服务器120对推送反馈信息进行解析，通过解析提取推送反馈信息中的任务标识、推送到达信息、推送点击信息和推送浏览信息。服务器120获取提取到的任务标识对应的历史推送任务的推送总量，根据提取到的推送到达信息、推送点击信息、推送浏览信息和获取到的推送总量统计任务标识对应的历史推送任务的推送效果数据。
在一个实施例中，推送效果数据包括推送到达率、推送点击率、1小时内到达率、1小时内点击率和推送浏览率。服务器120根据推送到达信息和推送总量统计推送到达率和1小时内到达率，根据推送点击信息和推送总量统计推送点击率和1小时内点击率，根据推送浏览信息和推送总量统计推送浏览率。
S608，根据统计的推送效果数据计算记录的历史推送任务的推送效果分数。
在一个实施例中，服务器120获取推送效果数据对应的权值，根据统计到的各历史推送任务的推送效果数据和获取到的推送效果数据对应的权值进行加权计算，得到各历史推送任务的推送效果分数。
在一个实施例中，服务器120分别获取推送效果数据中推送到达率、推送点击率、1小时内到达率、1小时内点击率和推送浏览率分别对应的权值，根据获取到的权值和统计到的各历史推送任务对应的推送效果数据进行加权和计算，将计算得到的值进行标准化处理得到各历史推送任务对应的推送效果分数。
本实施例中，记录历史推送记录，根据获取到的历史推送任务对应的推送反馈信息统计各历史推送任务的推送效果数据，根据统计到的推送效果数据计算各历史推送任务的推送效果分数。通过推送反馈信息统计推送效果数据，包证推送效果数据的真实性，从而保证根据推送效果数据计算得到的推送效果分数的准确性。
如图7所示，在一个实施例中，提供一种信息推送建议生成装置700，该装置具体包括以下内容：请求获取模块702、条件提取模块704、任务查询模块706、分数获取模块708、任务选取模块710和建议生成模块712。
请求获取模块702，用于获取推送建议分析请求。
条件提取模块704，用于提取推送建议分析请求中的建议分析条件。
任务查询模块706，用于在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务。
分数获取模块708，用于获取查询到的各历史推送任务分别对应的推送效果分数。
任务选取模块710，用于根据获取到的推送效果分数，从各历史推送任务中选取预设数量的历史推送任务。
建议生成模块712，用于对选取的历史推送任务进行分析，生成与推送建议分析请求对应的推送建议。
在一个实施例中，建议生成模块712还用于获取选取的各历史推送任务分别对应的用户类型；按照用户类型对选取的各历史推送任务进行分类；根 据各用户类型的历史推送任务所对应的推送效果分数，确定推送建议分析请求对应的用户类型选取建议。
本实施例中，提取推送建议分析请求中的及建议分析条件，在记录的历史推送任务中查询符合提取到的建议分析条件的历史推送任务，根据查询到的各历史推送任务对应的推送效果分数，选取预设数量的历史推送任务，保证选取的历史推送任务与推送建议分析请求相匹配，提高了选取的历史推送任务的准确性。对选取的历史推送任务进行分析，通过分析生成与推送建议分析请求对应的推送建议，确定得到的推送建议的合理性和准确性，保证根据生成的推送建议进行信息推送提高信息推送的推送效果。
在一个实施例中，分数获取模块708还用于若未查询到符合提取到的建议分析条件的历史推送任务，则获取记录的各历史推送任务分别对应的推送效果分数。
任务选取模块710还用于按照获取到的推送效果分数由高到低的顺序对记录的各历史推送任务进行排序；从排序后的历史推送任务中选取预设数量排序在前的历史推送任务。
本实施例中，在未查询到符合提取到的建议分析条件的历史推送任务时，从记录的历史推送任务选取推送效果分数较高的历史推送任务，对选取的历史推送任务进行分析，生成推送建议。以记录的所有历史推送任务作为基础任务数据，根据推送效果分数从记录的历史推送任务中选取推送效果分数较高历史推送任务，同样保证了推送建议的准确性。
如图8所示，在一个实施例中，建议生成模块712具体包括以下内容：推送时间获取模块712a、推送任务分配模块712b和推送时间建议模块712c。
推送时间获取模块712a，用于获取选取的各历史推送任务分别对应的推送时间。
推送任务分配模块712b，用于按照推送时间所属推送时间段对选取的各历史推送任务进行分类。
推送时间建议模块712c，用于根据分类后各推送时间段的历史推送任务所对应的推送效果分数，确定推送建议分析请求对应的推送时间段建议。
本实施例中，根据选取的各历史推送任务的推送时间所属的推送时间段 对选取的历史推送任务进行分类，得到的各推送时间段对应的历史推送任务，对各推送时间段对应的历史推送任务的推送效果分数进行分析，确定推送时间段建议，确保了推送时间段建议的准确性，保证根据推送时间段进行信息推送可以提高推送效果。
如图9所示，在一个实施例中，信息推送建议生成装置700具体还包括以下内容：任务记录模块714、信息获取模块716、数据统计模块718和数据计算模块720。
任务记录模块714，用于记录已执行的信息推送任务为历史推送任务；
信息获取模块716，用于获取记录的历史推送任务对应的推送反馈信息；
数据统计模块718，用于根据获取到的推送反馈信息统计记录的历史推送任务对应的推送效果数据；
数据计算模块720，用于根据统计的推送效果数据计算记录的历史推送任务的推送效果分数。
本实施例中，记录历史推送记录，根据获取到的历史推送任务对应的推送反馈信息统计各历史推送任务的推送效果数据，根据统计到的推送效果数据计算各历史推送任务的推送效果分数。通过推送反馈信息统计推送效果数据，包证推送效果数据的真实性，从而保证根据推送效果数据计算得到的推送效果分数的准确性。
上述信息推送建议生成装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。其中，网络接口可以是以太网或无线网卡等。上述各模块可以硬件形式内嵌于或独立于服务器中的处理器中，也可以以软件形式存储于服务器的存储器中，以便于处理器调用执行以上各个模块对应的操作。该处理器可以为中央处理单元(CPU)、微处理器、单片机等。
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：获取推送建议分析请求；提取推送建议分析请求中的建议分析条件；在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务；获取查询到的各历史推送任务分别对应的推送效果分数；根据获取 到的推送效果分数，从各历史推送任务中选取预设数量的历史推送任务；对选取的历史推送任务进行分析，生成与推送建议分析请求对应的推送建议。
在一个实施例中，在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务之后，处理器还执行以下步骤：若未查询到符合提取到的建议分析条件的历史推送任务，则获取记录的各历史推送任务分别对应的推送效果分数；按照获取到的推送效果分数由高到低的顺序对记录的各历史推送任务进行排序；从排序后的历史推送任务中选取预设数量排序在前的历史推送任务；对选取的历史推送任务进行分析，生成与推送建议分析请求对应的推送建议。
在一个实施例中，对选取的历史推送任务进行分析，生成与推送建议分析请求对应的推送建议，包括：获取选取的各历史推送任务分别对应的推送时间；按照推送时间所属推送时间段对选取的各历史推送任务进行分类；根据分类后各推送时间段的历史推送任务所对应的推送效果分数，确定推送建议分析请求对应的推送时间段建议。
在一个实施例中，对选取的历史推送任务进行分析，生成与推送建议分析请求对应的推送建议，包括：获取选取的各历史推送任务分别对应的用户类型；按照用户类型对选取的各历史推送任务进行分类；根据各用户类型的历史推送任务所对应的推送效果分数，确定推送建议分析请求对应的用户类型选取建议。
在一个实施例中，处理器还执行以下步骤：记录已执行的信息推送任务为历史推送任务；获取记录的历史推送任务对应的推送反馈信息；根据获取到的推送反馈信息统计记录的历史推送任务对应的推送效果数据；根据统计的推送效果数据计算记录的历史推送任务的推送效果分数。
本实施例中，提取推送建议分析请求中的及建议分析条件，在记录的历史推送任务中查询符合提取到的建议分析条件的历史推送任务，根据查询到的各历史推送任务对应的推送效果分数，选取预设数量的历史推送任务，保证选取的历史推送任务与推送建议分析请求相匹配，提高了选取的历史推送任务的准确性。对选取的历史推送任务进行分析，通过分析生成与推送建议分析请求对应的推送建议，确定得到的推送建议的合理性和准确性，保证根 据生成的推送建议进行信息推送提高信息推送的推送效果。
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，使得一个或多个处理器执行以下步骤：获取推送建议分析请求；提取推送建议分析请求中的建议分析条件；在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务；获取查询到的各历史推送任务分别对应的推送效果分数；根据获取到的推送效果分数，从各历史推送任务中选取预设数量的历史推送任务；对选取的历史推送任务进行分析，生成与推送建议分析请求对应的推送建议。
在一个实施例中，在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务之后，处理器还执行以下步骤：若未查询到符合提取到的建议分析条件的历史推送任务，则获取记录的各历史推送任务分别对应的推送效果分数；按照获取到的推送效果分数由高到低的顺序对记录的各历史推送任务进行排序；从排序后的历史推送任务中选取预设数量排序在前的历史推送任务；对选取的历史推送任务进行分析，生成与推送建议分析请求对应的推送建议。
在一个实施例中，对选取的历史推送任务进行分析，生成与推送建议分析请求对应的推送建议，包括：获取选取的各历史推送任务分别对应的推送时间；按照推送时间所属推送时间段对选取的各历史推送任务进行分类；根据分类后各推送时间段的历史推送任务所对应的推送效果分数，确定推送建议分析请求对应的推送时间段建议。
在一个实施例中，对选取的历史推送任务进行分析，生成与推送建议分析请求对应的推送建议，包括：获取选取的各历史推送任务分别对应的用户类型；按照用户类型对选取的各历史推送任务进行分类；根据各用户类型的历史推送任务所对应的推送效果分数，确定推送建议分析请求对应的用户类型选取建议。
在一个实施例中，处理器还执行以下步骤：记录已执行的信息推送任务为历史推送任务；获取记录的历史推送任务对应的推送反馈信息；根据获取到的推送反馈信息统计记录的历史推送任务对应的推送效果数据；根据统计 的推送效果数据计算记录的历史推送任务的推送效果分数。
本实施例中，提取推送建议分析请求中的及建议分析条件，在记录的历史推送任务中查询符合提取到的建议分析条件的历史推送任务，根据查询到的各历史推送任务对应的推送效果分数，选取预设数量的历史推送任务，保证选取的历史推送任务与推送建议分析请求相匹配，提高了选取的历史推送任务的准确性。对选取的历史推送任务进行分析，通过分析生成与推送建议分析请求对应的推送建议，确定得到的推送建议的合理性和准确性，保证根据生成的推送建议进行信息推送提高信息推送的推送效果。
根据本实施例的一个示例，上述实施例方法中的全部或部分流程，可以通过计算机可执行指令来指令相关的硬件来完成，所述计算机可执行指令可存储于一计算机可读取存储介质中，如本申请实施例中，该程序可存储于计算机系统的存储介质中，并被该计算机系统中的至少一个处理器执行，以实现包括如上述各方法的实施例的流程。该存储介质包括但不限于磁碟、优盘、光盘、只读存储记忆体(Read-Only Memory，ROM)等。
以上所述实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。

Claims (20)

  1. 一种信息推送建议生成方法，包括：
    获取推送建议分析请求；
    提取所述推送建议分析请求中的建议分析条件；
    在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务；
    获取查询到的各历史推送任务分别对应的推送效果分数；
    根据获取到的推送效果分数，从所述各历史推送任务中选取预设数量的历史推送任务；及
    对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议。
  2. 根据权利要求1所述的方法，其特征在于，所述在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务之后，所述方法还包括：
    若未查询到符合所述提取到的建议分析条件的历史推送任务，则获取记录的各历史推送任务分别对应的推送效果分数；
    按照获取到的推送效果分数由高到低的顺序对所述记录的各历史推送任务进行排序；
    从排序后的历史推送任务中选取预设数量排序在前的历史推送任务；及
    执行所述对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议的步骤。
  3. 根据权利要求2所述的方法，其特征在于，所述对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议，包括：
    获取选取的各历史推送任务分别对应的推送时间；
    按照推送时间所属推送时间段对所述选取的各历史推送任务进行分类；及
    根据分类后各推送时间段的历史推送任务所对应的推送效果分数，确定 所述推送建议分析请求对应的推送时间段建议。
  4. 根据权利要求2所述的方法，其特征在于，所述对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议，包括：
    获取选取的各历史推送任务分别对应的用户类型；
    按照用户类型对所述选取的各历史推送任务进行分类；及
    根据各用户类型的历史推送任务所对应的推送效果分数，确定所述推送建议分析请求对应的用户类型选取建议。
  5. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    记录已执行的信息推送任务为历史推送任务；
    获取记录的历史推送任务对应的推送反馈信息；
    根据获取到的推送反馈信息统计所述记录的历史推送任务对应的推送效果数据；及
    根据统计的推送效果数据计算所述记录的历史推送任务的推送效果分数。
  6. 一种信息推送建议生成装置，包括：
    请求获取模块，用于获取推送建议分析请求；
    条件提取模块，用于提取所述推送建议分析请求中的建议分析条件；
    任务查询模块，用于在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务；
    分数获取模块，用于获取查询到的各历史推送任务分别对应的推送效果分数；
    任务选取模块，用于根据获取到的推送效果分数，从所述各历史推送任务中选取预设数量的历史推送任务；及
    建议生成模块，用于对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议。
  7. 根据权利要求6所述的装置，其特征在于，所述分数获取模块还用于若未查询到符合所述提取到的建议分析条件的历史推送任务，则获取记录的 各历史推送任务分别对应的推送效果分数；及
    所述任务选取模块还用于按照获取到的推送效果分数由高到低的顺序对所述记录的各历史推送任务进行排序；从排序后的历史推送任务中选取预设数量排序在前的历史推送任务。
  8. 根据权利要求7所述的装置，其特征在于，所述建议生成模块还用于获取选取的各历史推送任务分别对应的推送时间；按照推送时间所属推送时间段对所述选取的各历史推送任务进行分类；根据分类后各推送时间段的历史推送任务所对应的推送效果分数，确定所述推送建议分析请求对应的推送时间段建议。
  9. 根据权利要求7所述的装置，其特征在于，所述建议生成模块包括：
    推送时间获取模块，用于获取选取的各历史推送任务分别对应的推送时间；
    推送任务分配模块，用于按照推送时间所属推送时间段对所述选取的各历史推送任务进行分类；及
    推送时间建议模块，用于根据分类后各推送时间段的历史推送任务所对应的推送效果分数，确定所述推送建议分析请求对应的推送时间段建议。
  10. 根据权利要求6所述的装置，其特征在于，所述装置还包括：
    任务记录模块，用于记录已执行的信息推送任务为历史推送任务；
    信息获取模块，用于获取记录的历史推送任务对应的推送反馈信息；
    数据统计模块，用于根据获取到的推送反馈信息统计所述记录的历史推送任务对应的推送效果数据；及
    数据计算模块，用于根据统计的推送效果数据计算所述记录的历史推送任务的推送效果分数。
  11. 一种计算机设备，所述计算机设备包括存储器，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：
    获取推送建议分析请求；
    提取所述推送建议分析请求中的建议分析条件；
    在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务；
    获取查询到的各历史推送任务分别对应的推送效果分数；
    根据获取到的推送效果分数，从所述各历史推送任务中选取预设数量的历史推送任务；及
    对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议。
  12. 根据权利要求11所述的计算机设备，其特征在于，所述在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务之后，所述处理器还执行以下步骤：
    若未查询到符合所述提取到的建议分析条件的历史推送任务，则获取记录的各历史推送任务分别对应的推送效果分数；
    按照获取到的推送效果分数由高到低的顺序对所述记录的各历史推送任务进行排序；
    从排序后的历史推送任务中选取预设数量排序在前的历史推送任务；及
    执行所述对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议的步骤。
  13. 根据权利要求12所述的计算机设备，其特征在于，所述对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议，包括：
    获取选取的各历史推送任务分别对应的推送时间；
    按照推送时间所属推送时间段对所述选取的各历史推送任务进行分类；及
    根据分类后各推送时间段的历史推送任务所对应的推送效果分数，确定所述推送建议分析请求对应的推送时间段建议。
  14. 根据权利要求12所述的计算机设备，其特征在于，所述对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议，包括：
    获取选取的各历史推送任务分别对应的用户类型；
    按照用户类型对所述选取的各历史推送任务进行分类；及
    根据各用户类型的历史推送任务所对应的推送效果分数，确定所述推送建议分析请求对应的用户类型选取建议。
  15. 根据权利要求11所述的方法，其特征在于，所述处理器还执行以下步骤：
    记录已执行的信息推送任务为历史推送任务；
    获取记录的历史推送任务对应的推送反馈信息；
    根据获取到的推送反馈信息统计所述记录的历史推送任务对应的推送效果数据；及
    根据统计的推送效果数据计算所述记录的历史推送任务的推送效果分数。
  16. 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：
    获取推送建议分析请求；
    提取所述推送建议分析请求中的建议分析条件；
    在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务；
    获取查询到的各历史推送任务分别对应的推送效果分数；
    根据获取到的推送效果分数，从所述各历史推送任务中选取预设数量的历史推送任务；及
    对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议。
  17. 根据权利要求16所述的存储介质，其特征在于，所述在记录的历史推送任务中，查询符合提取到的建议分析条件的历史推送任务之后，所述处理器还执行以下步骤：
    若未查询到符合所述提取到的建议分析条件的历史推送任务，则获取记录的各历史推送任务分别对应的推送效果分数；
    按照获取到的推送效果分数由高到低的顺序对所述记录的各历史推送任务进行排序；
    从排序后的历史推送任务中选取预设数量排序在前的历史推送任务；及
    执行所述对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议的步骤。
  18. 根据权利要求17所述的计算机设备，其特征在于，所述对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议，包括：
    获取选取的各历史推送任务分别对应的推送时间；
    按照推送时间所属推送时间段对所述选取的各历史推送任务进行分类；及
    根据分类后各推送时间段的历史推送任务所对应的推送效果分数，确定所述推送建议分析请求对应的推送时间段建议。
  19. 根据权利要求17所述的存储介质，其特征在于，所述对选取的历史推送任务进行分析，生成与所述推送建议分析请求对应的推送建议，包括：
    获取选取的各历史推送任务分别对应的用户类型；
    按照用户类型对所述选取的各历史推送任务进行分类；及
    根据各用户类型的历史推送任务所对应的推送效果分数，确定所述推送建议分析请求对应的用户类型选取建议。
  20. 根据权利要求16所述的存储介质，其特征在于，所述处理器还执行以下步骤：
    记录已执行的信息推送任务为历史推送任务；
    获取记录的历史推送任务对应的推送反馈信息；
    根据获取到的推送反馈信息统计所述记录的历史推送任务对应的推送效果数据；及
    根据统计的推送效果数据计算所述记录的历史推送任务的推送效果分数。
PCT/CN2017/104771 2017-07-24 2017-09-30 信息推送建议生成方法、装置、计算机设备和存储介质 Ceased WO2019019387A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710607706.8A CN107682388B (zh) 2017-07-24 2017-07-24 信息推送建议生成方法、装置、计算机设备和存储介质
CN201710607706.8 2017-07-24

Publications (1)

Publication Number Publication Date
WO2019019387A1 true WO2019019387A1 (zh) 2019-01-31

Family

ID=61133660

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/104771 Ceased WO2019019387A1 (zh) 2017-07-24 2017-09-30 信息推送建议生成方法、装置、计算机设备和存储介质

Country Status (2)

Country Link
CN (1) CN107682388B (zh)
WO (1) WO2019019387A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112801685A (zh) * 2020-09-10 2021-05-14 腾讯科技（深圳）有限公司 信息推送方法、装置、计算机设备及存储介质
CN113268668A (zh) * 2021-05-31 2021-08-17 广州简知信息科技有限公司 一种基于大数据精准运营的推送管理系统
CN115484306A (zh) * 2022-09-15 2022-12-16 益智(广州)信息技术有限公司 一种基于互联网及大数据的在线教育系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103235808A (zh) * 2013-04-22 2013-08-07 亿赞普(北京)科技有限公司 一种互联网信息推送的方法和装置
US8768379B2 (en) * 2007-04-08 2014-07-01 Enhanced Geographic Llc Systems and methods to recommend businesses to a user of a wireless device based on a location history associated with the user
CN104090912A (zh) * 2014-06-10 2014-10-08 腾讯科技（深圳）有限公司 信息推送方法及装置
CN104239450A (zh) * 2014-09-01 2014-12-24 百度在线网络技术（北京）有限公司 搜索推荐方法和装置
CN106095867A (zh) * 2016-06-03 2016-11-09 北京奇虎科技有限公司 一种基于行业分析的图书推荐方法及装置

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10033581B2 (en) * 2015-12-08 2018-07-24 Yokogawa Electric Corporation Method for retrieval of device-type managers
CN106708938A (zh) * 2016-11-18 2017-05-24 北京大米科技有限公司 用于辅助推荐的方法及装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8768379B2 (en) * 2007-04-08 2014-07-01 Enhanced Geographic Llc Systems and methods to recommend businesses to a user of a wireless device based on a location history associated with the user
CN103235808A (zh) * 2013-04-22 2013-08-07 亿赞普(北京)科技有限公司 一种互联网信息推送的方法和装置
CN104090912A (zh) * 2014-06-10 2014-10-08 腾讯科技（深圳）有限公司 信息推送方法及装置
CN104239450A (zh) * 2014-09-01 2014-12-24 百度在线网络技术（北京）有限公司 搜索推荐方法和装置
CN106095867A (zh) * 2016-06-03 2016-11-09 北京奇虎科技有限公司 一种基于行业分析的图书推荐方法及装置

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112801685A (zh) * 2020-09-10 2021-05-14 腾讯科技（深圳）有限公司 信息推送方法、装置、计算机设备及存储介质
CN113268668A (zh) * 2021-05-31 2021-08-17 广州简知信息科技有限公司 一种基于大数据精准运营的推送管理系统
CN115484306A (zh) * 2022-09-15 2022-12-16 益智(广州)信息技术有限公司 一种基于互联网及大数据的在线教育系统
CN115484306B (zh) * 2022-09-15 2024-01-02 优课达教育科技(北京)有限公司 一种基于互联网及大数据的在线教育系统

Also Published As

Publication number Publication date
CN107682388A (zh) 2018-02-09
CN107682388B (zh) 2020-01-24

Similar Documents

Publication Publication Date Title
CN104298679B (zh) 应用业务推荐方法及装置
US11526799B2 (en) Identification and application of hyperparameters for machine learning
WO2021047326A1 (zh) 信息推荐方法、装置、计算机设备和存储介质
KR101999471B1 (ko) 정보 추천 방법 및 장치
US20160285672A1 (en) Method and system for processing network media information
CN109241084B (zh) 数据的查询方法、终端设备及介质
US9355250B2 (en) Method and system for rapidly scanning files
CN115025497B (zh) 异常数据检测方法、装置、电子设备及存储介质
CN104408640B (zh) 应用软件推荐方法及装置
US10915534B2 (en) Extreme value computation
CN113268403B (zh) 时间序列的分析预测方法、装置、设备及存储介质
US9582586B2 (en) Massive rule-based classification engine
WO2019019387A1 (zh) 信息推送建议生成方法、装置、计算机设备和存储介质
WO2020253369A1 (zh) 生成兴趣标签的方法、装置、计算机设备和存储介质
CN106998336B (zh) 渠道中的用户检测方法和装置
CN120470183B (zh) 一种基于协同过滤的多业务场景推荐建模方法及系统
CN114971643B (zh) 异常交易识别方法、装置、设备及存储介质
CN109597941B (zh) 排序方法及装置、电子设备和存储介质
CN110634018A (zh) 一种流失用户的特征刻画方法、识别方法及相关装置
CN114662005A (zh) 用户行为轨迹的消息推送方法、装置、设备及存储介质
CN113204525B (zh) 一种基于分布式存储系统的文档预览管理系统
CN111859063A (zh) 一种用于监控互联网中转载文章信息的控制方法及装置
CN114463044A (zh) 广告计划调整方法、装置、计算机设备及存储介质
CN115168509A (zh) 风控数据的处理方法及装置、存储介质、计算机设备
CN115048999A (zh) 标签优化方法、装置、电子设备和可读介质

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

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 1205A DATED 28.05.2020)

122 Ep: pct application non-entry in european phase

Ref document number: 17919461

Country of ref document: EP

Kind code of ref document: A1