WO2017071251A1 - 信息推送方法和装置 - Google Patents

信息推送方法和装置 Download PDF

Info

Publication number
WO2017071251A1
WO2017071251A1 PCT/CN2016/086214 CN2016086214W WO2017071251A1 WO 2017071251 A1 WO2017071251 A1 WO 2017071251A1 CN 2016086214 W CN2016086214 W CN 2016086214W WO 2017071251 A1 WO2017071251 A1 WO 2017071251A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
scene model
site
user
push
Prior art date
Application number
PCT/CN2016/086214
Other languages
English (en)
French (fr)
Inventor
徐云峰
赵继承
陈炜于
Original Assignee
百度在线网络技术（北京）有限公司
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 百度在线网络技术（北京）有限公司 filed Critical 百度在线网络技术（北京）有限公司
Publication of WO2017071251A1 publication Critical patent/WO2017071251A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/951—Indexing; Web crawling techniques

Definitions

  • the present application relates to the field of computer technologies, and in particular, to the field of Internet technologies, and in particular, to an information push method and an information push device.
  • Information Push also known as “webcasting” is a technology that reduces information overload by pushing the information the user needs on the Internet through certain technical standards or protocols. Information push technology can reduce the time it takes for users to search on the network by actively pushing information to users.
  • the existing information push is generally implemented by the following steps: first, collecting historical data of users and sites; secondly, performing statistical analysis, cluster analysis, and machine learning on the collected data, and selecting site content that the user is interested in or associated with. Finally, the content of the selected site is pushed to the user.
  • This kind of information push method is complicated to operate, especially when facing new users or new sites, the information push effect is poor due to the lack of historical data; and the user's interest or preference is also changing with time, site size or content information. There will also be changes, resulting in the problem of poor scalability of this information push method.
  • the purpose of the present application is to propose an improved information push method and information push device to solve the technical problems mentioned in the above background art.
  • the application provides an information pushing method, where the method includes:
  • Obtaining a current scene model wherein the scene model includes: the user is within a predetermined time period Operation behavior information and/or browsing behavior information at at least one site, and user characteristics of the user and site features of the site; matching the current scene model with at least one preset scene model to obtain and a preset scene model in which the similarity of the current scene model is greater than a preset threshold; selecting at least one push information from the plurality of information to be pushed and pushing based on the information pushing rule associated with the acquired preset scene model .
  • the operational behavior information includes search behavior information and/or click behavior information.
  • the site feature comprises at least one of the following: a site type, a site size, and an average website access depth.
  • the matching the current scene model with the at least one preset scene model comprises: based on one or more of operation behavior information, browsing behavior information, user features, and site features, The current scene model is matched with the at least one preset scene model.
  • the information push rule includes at least one of the following: a calculation rule, a sort rule, and an adjustment rule.
  • the present application provides an information pushing apparatus, where the apparatus includes: an acquiring unit, configured to acquire a current scene model, where the scene model includes: operation behavior information of the user at the at least one site within a predetermined time period and/or Or browsing behavior information, and user characteristics of the user and site features of the site; a matching unit, configured to match the current scene model with at least one preset scene model, and obtain the current scene model a preset scene model whose similarity is greater than a preset threshold; a pushing unit, configured to select at least one push information from a plurality of information to be pushed and push based on an information pushing rule associated with the acquired preset scene model .
  • the operational behavior information includes search behavior information and/or click behavior information.
  • the site feature comprises at least one of the following: a site type, a site size, and an average website access depth.
  • the matching unit is specifically configured to: compare the current scene model with the at least one preset based on one or more of operation behavior information, browsing behavior information, user features, and site features. Matching the scene model to obtain the current A preset scene model in which the similarity of the scene model is greater than a preset threshold.
  • the information push rule includes at least one of the following: a calculation rule, a sort rule, and an adjustment rule.
  • the information pushing method and device obtaineds a current scene model, compares the current scene model with a preset scene model, and acquires a preset with a similarity to the current scene model that is greater than a preset threshold.
  • the scene model finally selects the push information based on the acquired information push rule associated with the preset scene model and pushes the information to the user.
  • the method solves the problem that the traditional information push method is complicated in operation and poor in expandability.
  • FIG. 1 is an exemplary system architecture diagram to which the present application can be applied;
  • FIG. 2 is a flow chart of one embodiment of an information push method according to the present application.
  • FIG. 3 is a schematic diagram of an application scenario of an information pushing method according to the present application.
  • FIG. 5 is a flowchart of an implementation manner of acquiring push information based on an information push rule in an information push method according to the present application
  • FIG. 6 is a schematic structural diagram of an embodiment of an information pushing apparatus according to the present application.
  • FIG. 7 is a schematic structural diagram of a computer system suitable for implementing a terminal device or a server of an embodiment of the present application.
  • FIG. 1 illustrates an exemplary system architecture 100 of an embodiment in which an information push method or information push device of the present application may be applied.
  • system architecture 100 can include terminal devices 101, 102, 103, network 104, and server 105.
  • the network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105.
  • Network 104 may include various types of connections, such as wired, wireless communication links, fiber optic cables, and the like.
  • the user can interact with the server 105 over the network 104 using the terminal devices 101, 102, 103 to receive or transmit messages and the like.
  • Various client applications such as news applications, online shopping applications, novel applications, music applications, email clients, social device software, etc., can be installed on the terminal devices 101, 102, and 103, and the user can Various applications for searching, clicking, browsing, etc.
  • the terminal devices 101, 102, 103 may be various electronic devices, including but not limited to smart phones, tablets, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III, motion picture expert compression standard audio layer 3) MP4 (Moving Picture Experts Group Audio Layer IV) player, laptop portable computer and desktop computer, etc.
  • MP3 players Moving Picture Experts Group Audio Layer III, motion picture expert compression standard audio layer 3
  • MP4 Moving Picture Experts Group Audio Layer IV
  • the server 105 may be a server that provides various services, such as a back-end web server that provides support for news applications on the terminal devices 101, 102, 103, online shopping applications, and the like.
  • the back-end web server may analyze and process the request data of the news application, the online shopping application, and the like, and feed back the processing result to the terminal device.
  • the information pushing method provided by the embodiment of the present application is generally performed by the server 105. Accordingly, the information pushing device is generally disposed in the server 105.
  • terminal devices, networks, and servers in Figure 1 is merely illustrative. Depending on the implementation needs, there can be any number of terminal devices, networks, and servers.
  • the above information pushing method includes the following steps:
  • Step 201 Acquire a current scene model.
  • the electronic device used by the user can acquire the current scene model of the user locally or remotely.
  • the scene model refers to a model that characterizes the user's network usage scenario by means of some parameters.
  • the parameter includes at least information of operational behavior information and/or browsing behavior information of the user at the at least one site within a predetermined time period, and user characteristics of the user and site characteristics of the above-mentioned site.
  • users can access different sites using applications installed on the terminal.
  • the electronic device used by the user may acquire at least one of the following information: the user is in the above The operational behavior information of the site, the browsing behavior information of the user at the above site, the user characteristics of the user, and the site characteristics of the above site.
  • wireless connection methods include, but are not limited to, 3G/4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods now known or developed in the future.
  • the predetermined period of time may be specified by the electronic device according to a setting instruction of the user.
  • the user may be provided with a setting interface to set the predetermined time period to "12:00-13:00" in a day; or the predetermined time period may also be set by default, such as "last 30 minutes.”
  • the foregoing operational behavior information includes search behavior information and/or click behavior information.
  • the search behavior information may include search keyword behavior information.
  • the search keyword behavior information may include a search term used by the user in searching, for example, the search terms “mobile phone” and “automobile product”.
  • the click behavior information may include information included in the user's click location or click object.
  • the above-mentioned click object may be each information item displayed directly on the website page or displayed as a search result, which may include, for example, but not limited to, a picture, a video, a text, a web link, and the like.
  • the foregoing site features may include a site type, a site size, an average website access depth, and the like.
  • the site types include, but are not limited to, news sites, fiction sites, online shopping sites, music sites, etc.; site size includes but is not limited to the number of products in the site, the number of users, etc.; and the average site access depth in the site features It refers to the average value of the number of website pages that the user browses during the browsing of the site.
  • the foregoing user feature includes a user line.
  • the user behavior feature refers to a web page viewed by the user, a keyword searched by the user, a product purchased by the user, and a new/old user who is a search/browsing site, and the user's natural characteristics include the age of the user, Gender, education and other characteristics.
  • Step 202 Match the current scene model with the at least one preset scene model, and obtain a preset scene model whose similarity with the current scene model is greater than a preset threshold.
  • the preset scene model may be obtained by performing statistical analysis on one or more of historical operation behavior information, historical browsing behavior information, user characteristics, and site characteristics of each site.
  • the obtained preset scene model may be a model for characterizing a network usage scenario in which a user first accesses a music class site from a news site by means of some parameters.
  • the above parameters may include information such as operational behavior information and/or browsing behavior information of the user at the news site and the music site, user characteristics of the user, site characteristics of the above news site and music site, and the like.
  • the preset scene model may also be a model for characterizing a network usage scenario in which an old user of an e-commerce site sequentially accesses a plurality of different classification pages by means of some parameters.
  • the above parameters may include information such as operation behavior information and/or browsing behavior information of the user at the e-commerce site, user characteristics of the user, site characteristics of the e-commerce site, and the like.
  • the electronic device may match the current scene model with the preset scene model by using various means to obtain the similarity with the current scene model.
  • the similarity between the current scene model and the preset scene model may be determined by calculating the similarity between the information included in the current scene model and the information included in the preset scene model.
  • the information included in the scenario model may be browsing behavior information or site features in the scenario model.
  • a preset scene model matching the current scene model is selected according to the similarity obtained above.
  • the feedback push effect corresponding to the preset scene model may be continuously acquired. Based on the similarity and the push effect of each of the preset scene models, the preset preset scene model is obtained.
  • Step 203 Push a rule based on information associated with the acquired preset scene model, At least one push information is selected from a plurality of pieces of information to be pushed and pushed.
  • the information push rule associated with the preset scene model acquired in step 202 may be obtained, and the information push rule is used to select multiple pieces of information to be pushed corresponding to the current scene model.
  • the information pushing rule may be a screening rule, and further may be a time-based filtering rule, where the time information of the information to be pushed may be sorted according to the time sequence, and the top or bottom ranked first.
  • the post-pending information is used as push information.
  • the electronic device used by the user pushes the selected at least one push information to the user.
  • the above-described push information may be presented to the user on the screen of the electronic device; or the push information may also be presented in the form of a voice broadcast.
  • FIG. 3 is a schematic diagram of an application scenario of the information pushing method according to the embodiment.
  • the user's current network usage scenario is shown by reference numeral 301 in FIG. 3.
  • the user first logs in to the e-commerce site and enters the search term "mobile phone” to browse the mobile phone product. After that, the user continues to input the search term "car”. Supplies, and clicked on the "Car Pillow” image that appears on the search results page.
  • parameter information of the current scene model based on the network usage scenario of the current user: related operation behavior information and/or browsing behavior information of the user at the e-commerce site (such as the user's search, browsing, click behavior information, etc.)
  • the user characteristics of the user such as whether it is a member of the e-commerce site, gender, age, etc.
  • site characteristics of the e-commerce site such as the size of the e-commerce site and the average access depth of the website, etc.
  • the obtained user-related search, click and browse information, and the user feature and the site feature are matched with the preset scene model to obtain a preset scene model whose similarity is greater than a preset threshold, as indicated by reference numeral 302.
  • at least one push information may be selected from the plurality of information to be pushed based on the associated information push rule based on the acquired preset scene model, and the selected push information may be pushed to The above users.
  • the method provided by the foregoing embodiment of the present application matches the operation behavior information, the browsing behavior information, the user feature, the site feature of the current scenario model, the operation behavior information of the preset scenario model, the browsing behavior information, the user feature, and the site feature. Obtaining an information push rule of a preset scene model with a similarity greater than a threshold, and selecting push information based on the information push rule, the method solves the complicated operation of the traditional information push method Poor scalability.
  • Step 401 Acquire a current scene model.
  • the electronic device used by the user can acquire the current scene model through a wired connection or a wireless connection.
  • the scenario model includes parameter information: operational behavior information and/or browsing behavior information of the user at the at least one site within a predetermined time period, and user characteristics of the user and site characteristics of the above-mentioned site.
  • the user operation behavior information may include search behavior information and/or click behavior information.
  • Step 402 Match the current scene model with at least one preset scene model based on one or more of the operation behavior information, the browsing behavior information, the user feature, and the site feature.
  • the electronic device used by the user acquires one or more of the operation behavior information, the browsing behavior information, the user feature, and the site feature in the current scene model, and respectively respectively and the preset scene model.
  • the middle operation behavior information, browsing behavior information, user characteristics, and site characteristics are matched.
  • a preset scene model whose similarity with the current scene model is greater than a preset threshold is obtained based on a similarity algorithm or the like.
  • Step 403 Acquire an information pushing rule associated with the acquired preset scene model.
  • the information pushing rule may include at least one of the following: a calculation rule, a sorting rule, and an adjustment rule.
  • the calculation rule may be a rule for calculating the associated data of the current scene model according to the algorithm corresponding to the preset preset scene model.
  • the above algorithm may be a recommendation algorithm such as an association rule algorithm or a collaborative filtering algorithm.
  • the above data includes user behavior data associated with the scene model and content data in the information to be pushed, and the like.
  • the user behavior data such as the webpage information browsed by the user history, the searched keyword information, the published microblog information, the blog information published by the user, and the commodity information purchased by the user may be collected by the electronic device used by the user.
  • the sorting rule may be a rule for sorting the push information according to the optimization target of the preset preset scene model acquired above.
  • the optimization goal includes at least one of the following: click rate, content relevance, timeliness, average access depth, and average access duration.
  • the click rate of the push information is predicted based on the prediction model used to predict the click rate, and the prediction results are sorted in descending order.
  • the adjustment rule may be a rule for adjusting the push information according to the business rule of the preset scene model obtained above.
  • the adjustment rules include at least one of the following: blacklist filtering, novelty, and creativity. For example, when the service rule is blacklist filtering, the push information is filtered based on the blacklist set by the site, and the filtered information to be pushed is obtained.
  • Step 404 Select at least one push information from the plurality of to-be-push information based on the information push rule.
  • the electronic device used by the user selects at least one push information from the plurality of to-be-pushed information by using the information push rule in step 403.
  • Step 405 pushing at least one push information to the user.
  • the electronic device used by the user pushes the selected at least one push information obtained in step 404 to the user.
  • the above-described push information may be presented to the user on the screen of the electronic device; or the push information may also be presented in the form of a voice broadcast.
  • the flow 400 of the information push method in the present embodiment highlights one of the operational behavior information, the browsing behavior information, the user feature, and the site feature, as compared with the embodiment corresponding to FIG. a plurality of steps 402 of matching the current scene model with the at least one preset scene model and a step 403 of acquiring an information push rule associated with the acquired preset scene model.
  • step 404 "selecting at least one push information from a plurality of to-be-pushed information based on the information push rule" may be implemented by the following steps:
  • Step 501 Process a plurality of to-be-pushed targets according to a calculation rule.
  • the electronic device used by the user acquires the calculation rule associated with the acquired preset scene model to process the plurality of to-be-pushed information.
  • the related data of the current scene model is calculated by using the acquired algorithm corresponding to the preset scene model (for example, an association rule algorithm, a collaborative filtering algorithm, etc.), and finally the calculation result is obtained from the plurality of to-be-pushed information.
  • the data includes at least user behavior data in the scene model and content data of the information to be pushed.
  • Step 502 Determine whether the acquired preset scene model has a corresponding optimization target.
  • the electronic device used by the user determines whether the acquired preset scene model has a corresponding optimization target. If there is no corresponding optimization target, then go to step 507 to obtain the push information, wherein the calculation result obtained above is the push information; if there is a corresponding optimization target, go to step 503.
  • Step 503 Sort the calculation results according to the sorting rule.
  • the electronic device used by the user sorts the obtained calculation result based on the optimized target corresponding to the acquired preset scene model.
  • the optimization goal includes at least one of the following: click rate, content relevance, timeliness, average access depth, average access duration, and the like.
  • the click rate of the information to be pushed in the calculation result is predicted according to the prediction model for predicting the click rate, and the prediction results are sorted in descending order.
  • Step 504 Determine whether the acquired preset scene model has a corresponding business rule.
  • the electronic device used by the user determines whether the acquired preset scenario model has a corresponding business rule. If there is no corresponding business rule, go to step 508 to obtain the push information, wherein the obtained sort result is the push information; if there is a corresponding business rule, go to step 505.
  • Step 505 Adjust the sorting result according to the adjustment rule.
  • the electronic device used by the user adjusts the obtained ranking result based on the acquired business rule corresponding to the preset scenario model, and obtains the adjustment result.
  • the above adjustments include sorting, filtering, screening, and the like. For example, when the service rule is blacklisted, the information to be pushed in the foregoing sorting result is filtered based on the blacklist set by the site, and the filtered result is obtained.
  • step 506 the push information is obtained.
  • the adjustment result obtained in the above step 505 is the push information, and the electronic device used by the user can push the push information to the user.
  • the above-mentioned push information may be presented to the user on the screen of the electronic device; or the push information may also be presented in the form of a voice broadcast.
  • the implementation manner provided by the foregoing embodiment of the present application obtains a calculation rule and a sort by matching the operation behavior information, the browsing behavior information, the user feature, and the site feature of the scenario model. Rules, adjustment rules, and the like push rules, and select push information from a plurality of to-be-push information according to the obtained information push rule. Therefore, the implementation manner can obtain the push information of the current scene model through a preset scene model, thereby further improving information push efficiency and accuracy.
  • the present application provides an embodiment of an information pushing apparatus, and the apparatus embodiment corresponds to the method embodiment shown in FIG. Used in a variety of electronic devices.
  • the information pushing apparatus 600 described in this embodiment includes an obtaining unit 601, a matching unit 602, and a pushing unit 603.
  • the acquiring unit 601 is configured to acquire a current scene model, where the scenario model includes: operation behavior information and/or browsing behavior information of the user at the at least one site within a predetermined time period, and user characteristics of the user and site characteristics of the foregoing site.
  • the matching unit 602 is configured to match the current scene model with the at least one preset scene model, and acquire a preset scene model whose similarity with the current scene model is greater than a preset threshold; the pushing unit 603 is configured to be based on the acquired The information push rule associated with the preset scene model selects at least one push information from a plurality of information to be pushed and pushes.
  • the acquiring unit 601 of the information pushing device 600 may acquire the user's operation behavior information and/or browsing behavior information from the terminal where the site is located through a wired connection manner or a wireless connection manner, and use the user characteristics of the above-mentioned household and the above Site characteristics of the site.
  • the matching unit 602 may calculate the current scene model and the preset based on one or more of the operation behavior information, the browsing behavior information, the user feature, and the site feature.
  • the similarity value of the scene model is obtained, so that the preset scene model whose similarity value with the current scene model is greater than the preset threshold is obtained for use by the pushing unit 603.
  • the operation behavior information includes search behavior information and/or click behavior information; the site characteristics include at least one of the following: a site type, a site size, and an average website access depth.
  • the pushing unit 603 can obtain the information pushing rule associated with the preset scene model obtained above, and select at least one push information to be pushed to the user from the plurality of information to be pushed corresponding to the current scene model based on the information pushing rule.
  • the information push rule includes at least one of the following: a calculation rule, a sort rule, and an adjustment rule.
  • the above information pushing device 600 also includes some of its His well-known structures, such as processors, memories, etc., are not shown in FIG. 6 in order to unnecessarily obscure the embodiments of the present disclosure.
  • FIG. 7 a block diagram of a computer system 700 suitable for use in implementing a terminal device or server of an embodiment of the present application is shown.
  • computer system 700 includes a central processing unit (CPU) 701 that can be loaded into a program in random access memory (RAM) 703 according to a program stored in read only memory (ROM) 702 or from storage portion 708. And perform various appropriate actions and processes.
  • RAM random access memory
  • ROM read only memory
  • RAM 703 various programs and data required for the operation of the system 700 are also stored.
  • the CPU 701, the ROM 702, and the RAM 703 are connected to each other through a bus 704.
  • An input/output (I/O) interface 705 is also coupled to bus 704.
  • the following components are connected to the I/O interface 705: an input portion 707 including a keyboard, a mouse, etc.; an output portion 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), and the like, and a speaker; a storage portion 708 including a hard disk or the like And a communication portion 709 including a network interface card such as a LAN card, a modem, or the like.
  • the communication section 709 performs communication processing via a network such as the Internet.
  • Driver 710 is also connected to I/O interface 705 as needed.
  • a removable medium 711 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory or the like, is mounted on the drive 710 as needed so that a computer program read therefrom is installed into the storage portion 708 as needed.
  • an embodiment of the present disclosure includes a computer program product comprising a computer program tangibly embodied on a machine readable medium, the computer program comprising program code for executing the method illustrated in the flowchart.
  • the computer program can be downloaded and installed from the network via communication portion 709, and/or installed from removable media 711.
  • each block of the flowchart or block diagrams can represent a module, a program segment, or a portion of code that includes one or more logic for implementing the specified.
  • Functional executable instructions can also be noted that in some alternative implementations, the functions noted in the blocks may also be presented in a different order than that illustrated in the drawings. Health. For example, two successively represented blocks may in fact be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending upon the functionality involved.
  • each block of the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts can be implemented in a dedicated hardware-based system that performs the specified function or operation. Or it can be implemented by a combination of dedicated hardware and computer instructions.
  • the units involved in the embodiments of the present application may be implemented by software or by hardware.
  • the described unit may also be provided in the processor, for example, as a processor including an acquisition unit, a matching unit, and a push unit.
  • the names of these units do not constitute a limitation on the unit itself under certain circumstances.
  • the acquisition unit may also be described as “a unit that acquires the current scene model”.
  • the present application further provides a non-volatile computer storage medium, which may be a non-volatile computer storage medium included in the apparatus described in the foregoing embodiments; It may be a non-volatile computer storage medium that exists alone and is not assembled into the terminal.
  • the non-volatile computer storage medium stores one or more programs, when the one or more programs are executed by a device, causing the device to: acquire a current scene model, wherein the scene model includes: the user is at a predetermined time Operation behavior information and/or browsing behavior information of at least one site in the segment, and user characteristics of the user and site features of the site; matching the current scene model with at least one preset scene model to obtain and a preset scene model in which the similarity of the current scene model is greater than a preset threshold; selecting at least one push information from the plurality of information to be pushed and performing based on the information pushing rule associated with the acquired preset scene model Push.
  • the scene model includes: the user is at a predetermined time Operation behavior information and/or browsing behavior information of at least one site in the segment, and user characteristics of the user and site features of the site; matching the current scene model with at least one preset scene model to obtain and a preset scene model in which the similarity of the current scene model is greater than a preset threshold;

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

一种信息推送方法和装置。所述方法的一具体实施方式包括：获取当前的场景模型（201），其中场景模型包括：用户在预定时间段内在至少一个站点的操作行为信息和/或浏览行为信息，以及所述用户的用户特征和所述站点的站点特征；将所述当前的场景模型与至少一个预设场景模型进行匹配，获取与所述当前的场景模型的相似度大于预设阈值的预设场景模型（202）；基于与所获取的预设场景模型相关联的信息推送规则，从多个待推送的信息中选择至少一个推送信息并进行推送（203）。该实施方式解决了信息推送方法操作复杂和扩展性差的问题。

Description

信息推送方法和装置
相关申请的交叉引用
本申请要求于2015年10月28日提交的中国专利申请号为“201510710944.2”的优先权，其全部内容作为整体并入本申请中。
技术领域
本申请涉及计算机技术领域，具体涉及互联网技术领域，尤其涉及信息推送方法及信息推送装置。
背景技术
信息推送，又称为“网络广播”，是通过一定的技术标准或协议，在互联网上通过推送用户需要的信息来减少信息过载的一项技术。信息推送技术通过主动推送信息给用户，可以减少用户在网络上搜索所花的时间。
现有的信息推送一般是通过如下步骤实现的：首先，搜集用户和站点的历史数据；其次，对搜集的数据进行统计分析、聚类分析以及机器学习，选择用户感兴趣或相关联的站点内容；最后，将上述选择的站点内容推送给用户。这种信息推送方法操作复杂，尤其是在面对新用户或新站点时，由于缺少历史数据使得信息推送效果比较差；而且，用户的兴趣或偏好也是随时间而不断变化，站点规模或内容信息也会发生变化，导致这种信息推送方法还存在扩展性差的问题。
发明内容
本申请的目的在于提出一种改进的信息推送方法和信息推送装置，来解决以上背景技术部分提到的技术问题。
第一方面，本申请提供了一种信息推送方法，所述方法包括：
获取当前的场景模型，其中场景模型包括：用户在预定时间段内 在至少一个站点的操作行为信息和/或浏览行为信息，以及所述用户的用户特征和所述站点的站点特征；将所述当前的场景模型与至少一个预设场景模型进行匹配，获取与所述当前的场景模型的相似度大于预设阈值的预设场景模型；基于与所获取的预设场景模型相关联的信息推送规则，从多个待推送的信息中选择至少一个推送信息并进行推送。
在一些实施例中，所述操作行为信息包括搜索行为信息和/或点击行为信息。
在一些实施例中，所述站点特征包括以下至少一项：站点类型、站点规模、平均网站访问深度。
在一些实施例中，所述将所述当前的场景模型与至少一个预设场景模型进行匹配，包括：基于操作行为信息、浏览行为信息、用户特征和站点特征中的一项或多项，将所述当前的场景模型与所述至少一个预设场景模型进行匹配。
在一些实施例中，所述信息推送规则包括以下至少一项：计算规则、排序规则、调整规则。
第二方面，本申请提供了一种信息推送装置，所述装置包括：获取单元，用于获取当前的场景模型，其中场景模型包括：用户在预定时间段内在至少一个站点的操作行为信息和/或浏览行为信息，以及所述用户的用户特征和所述站点的站点特征；匹配单元，用于将所述当前的场景模型与至少一个预设场景模型进行匹配，获取与所述当前的场景模型的相似度大于预设阈值的预设场景模型；推送单元，用于基于与所获取的预设场景模型相关联的信息推送规则，从多个待推送的信息中选择至少一个推送信息并进行推送。
在一些实施例中，所述操作行为信息包括搜索行为信息和/或点击行为信息。
在一些实施例中，所述站点特征包括以下至少一项：站点类型、站点规模、平均网站访问深度。
在一些实施例中，所述匹配单元具体用于：基于操作行为信息、浏览行为信息、用户特征和站点特征中的一项或多项，将所述当前的场景模型与所述至少一个预设场景模型进行匹配，获取与所述当前的 场景模型的相似度大于预设阈值的预设场景模型。
在一些实施例中，所述信息推送规则包括以下至少一项：计算规则、排序规则、调整规则。
本申请提供的信息推送方法和装置，通过获取当前的场景模型，将所述当前的场景模型与预设场景模型相对比，获取与所述当前的场景模型的相似度大于预设阈值的预设场景模型，最后基于所获取的预设场景模型相关联的信息推送规则选择推送信息并向用户推送该信息，该方法解决了传统信息推送方法操作复杂和扩展性差的问题。
附图说明
通过阅读参照以下附图所作的对非限制性实施例所作的详细描述，本申请的其它特征、目的和优点将会变得更明显：
图1是本申请可以应用于其中的示例性系统架构图；
图2是根据本申请的信息推送方法的一个实施例的流程图；
图3是根据本申请的信息推送方法的一个应用场景的示意图；
图4是根据本申请的信息推送方法的又一个实施例的流程图；
图5是根据本申请的信息推送方法中，基于信息推送规则获取推送信息的一种实现方式的流程图；
图6是根据本申请的信息推送装置的一个实施例的结构示意图；
图7是适于用来实现本申请实施例的终端设备或服务器的计算机系统的结构示意图。
具体实施方式
下面结合附图和实施例对本申请作进一步的详细说明。可以理解的是，此处所描述的具体实施例仅仅用于解释相关发明，而非对该发明的限定。另外还需要说明的是，为了便于描述，附图中仅示出了与有关发明相关的部分。
需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互组合。下面将参考附图并结合实施例来详细说明本申请。
图1示出了可以应用本申请的信息推送方法或信息推送装置的实施例的示例性系统架构100。
如图1所示，系统架构100可以包括终端设备101、102、103，网络104和服务器105。网络104用以在终端设备101、102、103和服务器105之间提供通信链路的介质。网络104可以包括各种连接类型，例如有线、无线通信链路或者光纤电缆等等。
用户可以使用终端设备101、102、103通过网络104与服务器105交互，以接收或发送消息等。终端设备101、102、103上可以安装有各种客户端应用，例如新闻类应用、网购类应用、小说类应用、音乐类应用、邮箱客户端、社交装置软件等，用户可以对终端设备上的各种应用进行搜索、点击、浏览等操作。
终端设备101、102、103可以是各种电子设备，包括但不限于智能手机、平板电脑、电子书阅读器、MP3播放器(Moving Picture Experts Group Audio Layer III，动态影像专家压缩标准音频层面3)、MP4(Moving Picture Experts Group Audio Layer IV，动态影像专家压缩标准音频层面4)播放器、膝上型便携计算机和台式计算机等等。
服务器105可以是提供各种服务的服务器，例如对终端设备101、102、103上的新闻类应用、网购类应用等提供支持的后台网站服务器。后台网站服务器可以对接收到上述新闻类应用、网购类应用等的请求数据进行分析等处理，并将处理结果反馈给终端设备。
需要说明的是，本申请实施例所提供的信息推送方法一般由服务器105执行，相应地，信息推送装置一般设置于服务器105中。
应该理解，图1中的终端设备、网络和服务器的数目仅仅是示意性的。根据实现需要，可以具有任意数目的终端设备、网络和服务器。
继续参考图2，其示出了根据本申请的信息推送方法的一个实施例的流程200。上述的信息推送方法，包括以下步骤：
步骤201，获取当前的场景模型。
在本实施例中，用户所使用的电子设备(例如图1所示的服务器)可以从本地或者远程地获取用户当前的场景模型。这里，场景模型是指借助于一些参数对用户的网络使用场景进行表征的模型。其中，上 述参数至少包括如下信息：用户在预定时间段内在至少一个站点的操作行为信息和/或浏览行为信息，以及上述用户的用户特征和上述站点的站点特征。
通常，用户可以利用终端上安装的应用访问不同的站点。这里，当用户通过有线连接方式或者无线连接方式从安装了新闻类应用、网购类应用等的终端设备访问相应的站点时，用户所使用的电子设备可以获取如下信息的至少一项：用户在上述站点的操作行为信息、用户在上述站点的浏览行为信息、该用户的用户特征、上述站点的站点特征。
需要指出的是，上述无线连接方式包括但不限于3G/4G连接、WiFi连接、蓝牙连接、WiMAX连接、Zigbee连接、UWB(ultra wideband)连接、以及其它现在已知或将来开发的无线连接方式。
在本实施例中，上述预定时间段可以由电子设备根据用户的设置指令而指定。例如，可以给用户提供设置界面，以将预定时间段设置为一天中的“12:00-13:00”；或者上述预定时间段也可以被缺省设置，例如“最近三十分钟”。
在本实施例的一些可选的实现方式中，上述操作行为信息包括搜索行为信息和/或点击行为信息。在这里，搜索行为信息可以包括搜索关键词行为信息。其中，搜索关键词行为信息可以包括用户在搜索时所使用的搜索词，例如，搜索词“手机”、“汽车用品”。点击行为信息可以包括用户的点击位置或点击对象等包含的信息。具体地，上述点击对象可以是站点页面上直接显示的或者作为搜索结果显示的各个信息项，其例如可以包括但不限于，图片、视频、文字、网址链接等。
在本实施例的一些可选的实现方式中，上述站点特征可以包括站点类型、站点规模、平均网站访问深度等。在这里，站点类型包括但不限于新闻类站点、小说类站点、网购类站点、音乐类站点等；站点规模包括但不限于站点中产品数量、用户数量等；而站点特征中的平均网站访问深度是指用户在浏览该站点的过程中浏览的网站页数的平均值，这些条件项为本领域技术人员所熟知，在此不作赘述。
在本实施例的一些可选的实现方式中，上述用户特征包括用户行 为特征以及用户自然特征。在这里，用户行为特征是指用户浏览过的网页、用户搜索过的关键词、用户购买的商品、以及该用户是搜索/浏览站点的新/老用户等特征，用户自然特征包括用户的年龄、性别、学历等特征。
步骤202，将当前的场景模型与至少一个预设场景模型进行匹配，获取与当前的场景模型的相似度大于预设阈值的预设场景模型。
在本实施例中，预设场景模型可以通过对各用户的历史操作行为信息、历史浏览行为信息、用户特征以及各站点的站点特征中的一项或多项进行统计分析获得。例如，得到的预设场景模型可以是借助于一些参数对用户从新闻站点第一次访问某音乐类站点这一网络使用场景进行表征的模型。这里，上述参数可以包括如下信息：用户在新闻类站点和音乐类站点的操作行为信息和/或浏览行为信息、该用户的用户特征、上述新闻类站点和音乐类站点的站点特征等。或者预设场景模型还可以是借助于一些参数对电商类站点的老用户依次访问多个不同分类页这一网络使用场景进行表征的模型。这里，上述参数可以包括如下信息：用户在电商类站点的操作行为信息和/或浏览行为信息、该用户的用户特征、该电商类站点的站点特征等。
在本实施例中，在上述电子设备获取用户当前的场景模型之后，其可以利用各种手段将上述当前的场景模型与上述预设场景模型进行匹配，获取与所述当前的场景模型的相似度大于预设阈值的预设场景模型。具体的，可以通过计算当前的场景模型所包含信息与预设场景模型所包含信息的相似度来确定当前的场景模型与预设场景模型的相似度。其中，上述场景模型所包含信息可以是该场景模型中的浏览行为信息或站点特征等。最后，根据上述获取的相似度选择与当前的场景模型匹配的预设场景模型。
在本实施例的一些可选的实现方式中，在获取与当前的场景模型的相似度值大于预设阈值的预设场景模型之后，还可以继续获取预设场景模型对应的反馈推送效果。基于上述每一个预设场景模型的相似度和推送效果进行综合考虑，获取需要的预设场景模型。
步骤203，基于与所获取的预设场景模型相关联的信息推送规则， 从多个待推送的信息中选择至少一个推送信息并进行推送。
在本实施例中，可以根据在步骤202中所获取的预设场景模型，获取与之相关联的信息推送规则，利用上述信息推送规则从当前的场景模型对应的多个待推送的信息中选择可以推送给用户的推送信息。例如，上述信息推送规则可以是筛选规则，进一步的可以是基于时间的筛选规则，这里可以根据待推送信息的时间信息，将其按照时间的先后顺序进行排序，选出排名最靠前或最靠后的待推送信息作为推送信息。
在本实施例中，用户所使用的电子设备将选择的至少一个推送信息推送给用户。作为示例，可以在电子设备的屏幕上向用户呈现上述推送信息；或者也可以用语音播报的形式来呈现上述推送信息。
继续参见图3，图3是根据本实施例的信息推送方法的应用场景的一个示意图。用户当前的网络使用场景如图3中的标号301所示，用户首先登陆了电商类站点，并输入了搜索词“手机”，浏览了手机产品，之后，上述用户继续输入了搜索词“汽车用品”，并且点击了搜索结果页面中出现的“汽车抱枕”图片。基于上述当前用户的网络使用场景得到当前的场景模型的如下参数信息：上述用户在电商类站点的相关操作行为信息和/或浏览行为信息(如上述用户的搜索、浏览、点击行为信息等)，该用户的用户特征(如是否为上述电商类站点的会员，以及性别、年龄等)，以及该电商类站点的站点特征(如上述电商站点的规模和网站平均访问深度等)。之后，用获得的用户相关的搜索、点击和浏览信息，以及用户特征和站点特征与预设的场景模型相匹配，获得相似度大于预设阈值的预设场景模型，如标号302所示。最后，如标号303所示，可以用上述基于与所获取的预设场景模型的相关联信息推送规则，从多个待推送的信息中选择至少一个推送信息，并将所选择的推送信息推送给上述用户。
本申请的上述实施例提供的方法通过将当前的场景模型的操作行为信息、浏览行为信息、用户特征、站点特征与预设场景模型的操作行为信息、浏览行为信息、用户特征、站点特征相匹配，获得相似度大于阈值的预设场景模型的相关联的信息推送规则，并基于上述信息推送规则选择推送信息，该方法解决了传统信息推送方法操作复杂和 扩展性差的问题。
进一步参考图4，其示出了信息推送方法的又一个实施例的流程400。该信息推送方法的流程400，包括以下步骤：步骤401，获取当前的场景模型。
在本实施例中，用户所使用的电子设备(例如图1所示的服务器)可以通过有线连接方式或者无线连接方式获取当前的场景模型。在这里，场景模型包括如下参数信息：用户在预定时间段内在至少一个站点的操作行为信息和/或浏览行为信息，以及上述用户的用户特征以及上述站点的站点特征。其中，上述用户操作行为信息可以包括搜索行为信息和/或点击行为信息。
步骤402，基于操作行为信息、浏览行为信息、用户特征和站点特征中的一项或多项，将当前的场景模型与至少一个预设场景模型进行匹配。
在本实施例中，用户所使用的电子设备获取上述当前的场景模型中的操作行为信息、浏览行为信息、用户特征、站点特征中的一项或多项，并将其分别与预设场景模型中操作行为信息、浏览行为信息、用户特征、站点特征相匹配。最后，基于相似度算法等获取与上述当前的场景模型的相似度大于预设阈值的预设场景模型。
步骤403，获取与所获取的预设场景模型相关联的信息推送规则。
在本实施例中，信息推送规则可以包括以下至少一项：计算规则、排序规则、调整规则。这里，计算规则可以是根据上述所获取的预设场景模型对应的算法对当前的场景模型的相关联数据进行计算的规则。上述算法可以是关联规则算法、协同过滤算法等推荐算法。上述数据包括与场景模型相关联的用户行为数据以及待推送信息中的内容数据等。这里可以通过用户所使用的电子设备收集用户历史浏览过的网页信息、搜索过的关键词信息、发表的微博信息、用户发表的博客(blog)信息以及用户购买的商品信息等用户行为数据。排序规则可以是根据上述获取的预设场景模型的优化目标对待推送信息进行排序的规则。其中，优化目标包括以下至少一项：点击率、内容相关性、时效性、平均访问深度、平均访问时长。例如，当上述优化目标为点 击率时，则根据用于预测点击率的预测模型对待推送信息的点击率进行预测，并将预测结果按照从大到小的顺序进行排序。调整规则可以是根据上述获取的预设场景模型的业务规则对待推送信息进行调整的规则。其中，调整规则包括以下至少一项：黑名单过滤、新颖性、创造性。例如，当上述业务规则为黑名单过滤时，则基于站点设置的黑名单对待推送信息进行过滤，得到过滤后的待推送信息。
步骤404，基于信息推送规则从多个待推送信息中选择至少一个推送信息。
在本实施例中，用户所使用的电子设备通过步骤403中的信息推送规则，从多个待推送信息中选择至少一个推送信息。
步骤405，向用户推送至少一个推送信息。
在本实施例中，用户所使用的电子设备将步骤404中得到的所选择的至少一个推送信息推送给用户。作为示例，可以在电子设备的屏幕上向用户呈现上述推送信息；或者也可以用语音播报的形式来呈现上述推送信息。
从图4中可以看出，与图2对应的实施例相比，本实施例中的信息推送方法的流程400突出了基于操作行为信息、浏览行为信息、用户特征和站点特征中的一项或多项，将上述当前的场景模型与至少一个预设场景模型进行匹配的步骤402和获取与所获取的预设场景模型相关联的信息推送规则的步骤403。
在本实施例的一些可选的实现方式中，如图5所示，步骤404“基于信息推送规则从多个待推送信息中选择至少一个推送信息”可以通过如下步骤来实现：
步骤501，根据计算规则处理多个待推送目标。
在本实现方式中，用户所使用的电子设备获取与所获取的预设场景模型相关联的计算规则处理上述多个待推送信息。具体的，其是通过所获取的预设场景模型对应的算法(例如关联规则算法、协同过滤算法等)对当前的场景模型的相关联数据进行计算，最后从多个待推送信息中获取计算结果。其中，上述数据至少包括场景模型中的用户行为数据以及待推送信息的内容数据。
步骤502，判断所获取的预设场景模型是否有对应的优化目标。
在本实现方式中，用户所使用的电子设备判断上述所获取的预设场景模型是否有对应的优化目标。如果没有对应的优化目标，则转到步骤507，获取推送信息，其中上述获得的计算结果即为推送信息；如果有对应的优化目标，则转到步骤503。
步骤503，根据排序规则对计算结果进行排序。
在本实现方式中，用户所使用的电子设备基于所获取的预设场景模型对应的优化目标对上述获得的计算结果进行排序。其中，优化目标包括以下至少一项：点击率、内容相关性、时效性、平均访问深度、平均访问时长等。例如，当上述优化目标为点击率时，则根据用于预测点击率的预测模型对计算结果中的待推送信息的点击率进行预测，并将预测结果按照从大到小的顺序进行排序。
步骤504，判断所获取的预设场景模型是否有对应的业务规则。
在本实现方式中，用户所使用的电子设备判断上述所获取的预设场景模型是否有对应的业务规则。如果没有对应的业务规则，则转到步骤508，获取推送信息，其中上述获取的排序结果即为推送信息；如果有对应的业务规则，则转到步骤505。
步骤505，根据调整规则对排序结果进行调整。
在本实现方式中，用户所使用的电子设备基于所获取的预设场景模型对应的业务规则对上述获得的排序结果进行调整，并获取调整结果。其中，上述调整包括排序、过滤、筛选等。例如，当上述业务规则为黑名单过滤时，则基于站点设置的黑名单对上述排序结果中的待推送信息进行过滤，得到过滤后的结果。
步骤506，获取推送信息。
在本实现方式中，上述步骤505中得到的调整结果即为推送信息，用户所使用的电子设备可以将上述推送信息推送给用户。这里，可以在电子设备的屏幕上向用户呈现上述推送信息；或者也可以用语音播报的形式来呈现上述推送信息。
本申请的上述实施例提供的实现方式通过场景模型的操作行为信息、浏览行为信息、用户特征和站点特征的匹配获得计算规则、排序 规则、调整规则等信息推送规则，并根据所获得的信息推送规则从多个待推送信息中选择推送信息。由此，本实现方式可以通过预设的场景模型获取当前的场景模型的推送信息，从而进一步提高信息推送效率和准确性。
进一步参考图6，作为对上述各图所示方法的实现，本申请提供了一种信息推送装置的一个实施例，该装置实施例与图2所示的方法实施例相对应，该装置具体可以应用于各种电子设备中。
如图6所示，本实施例所述的信息推送装置600包括：获取单元601、匹配单元602和推送单元603。其中，获取单元601用于获取当前的场景模型，其中场景模型包括：用户在预定时间段内在至少一个站点的操作行为信息和/或浏览行为信息，以及上述用户的用户特征和上述站点的站点特征；匹配单元602用于将当前的场景模型与至少一个预设场景模型进行匹配，获取与当前的场景模型的相似度大于预设阈值的预设场景模型；推送单元603用于基于与所获取的预设场景模型相关联的信息推送规则，从多个待推送的信息中选择至少一个推送信息并进行推送。在本实施例中，信息推送装置600的获取单元601可以通过有线连接方式或者无线连接方式从站点所在的终端获取用户的操作行为信息和/或浏览行为信息，以及用上述户的用户特征和上述站点的站点特征。
在本实施例中，获取单元601获取用户当前的场景模型之后，匹配单元602可以基于操作行为信息、浏览行为信息、用户特征和站点特征中的一项或多项计算当前的场景模型与预设场景模型的相似度值，从而获得与当前的场景模型的相似度值大于预设阈值的预设场景模型供推送单元603使用。其中，操作行为信息包括搜索行为信息和/或点击行为信息；站点特征包括以下至少一项：站点类型、站点规模、平均网站访问深度。推送单元603可以获得与上述获得的预设场景模型相关联的信息推送规则，并基于该信息推送规则从当前的场景模型对应的多个待推送的信息中选择至少一个推送信息推送给用户。其中，信息推送规则包括以下至少一项：计算规则、排序规则、调整规则。
本领域技术人员可以理解，上述信息推送装置600还包括一些其 他公知结构，例如处理器、存储器等，为了不必要地模糊本公开的实施例，这些公知的结构在图6中未示出。
下面参考图7，其示出了适于用来实现本申请实施例的终端设备或服务器的计算机系统700的结构示意图。
如图7所示，计算机系统700包括中央处理单元(CPU)701，其可以根据存储在只读存储器(ROM)702中的程序或者从存储部分708加载到随机访问存储器(RAM)703中的程序而执行各种适当的动作和处理。在RAM 703中，还存储有系统700操作所需的各种程序和数据。CPU 701、ROM 702以及RAM 703通过总线704彼此相连。输入/输出(I/O)接口705也连接至总线704。
以下部件连接至I/O接口705：包括键盘、鼠标等的输入部分707；包括诸如阴极射线管(CRT)、液晶显示器(LCD)等以及扬声器等的输出部分707；包括硬盘等的存储部分708；以及包括诸如LAN卡、调制解调器等的网络接口卡的通信部分709。通信部分709经由诸如因特网的网络执行通信处理。驱动器710也根据需要连接至I/O接口705。可拆卸介质711，诸如磁盘、光盘、磁光盘、半导体存储器等等，根据需要安装在驱动器710上，以便于从其上读出的计算机程序根据需要被安装入存储部分708。
特别地，根据本公开的实施例，上文参考流程图描述的过程可以被实现为计算机软件程序。例如，本公开的实施例包括一种计算机程序产品，其包括有形地包含在机器可读介质上的计算机程序，所述计算机程序包含用于执行流程图所示的方法的程序代码。在这样的实施例中，该计算机程序可以通过通信部分709从网络上被下载和安装，和/或从可拆卸介质711被安装。
附图中的流程图和框图，图示了按照本申请各种实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分，所述模块、程序段、或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发 生。例如，两个接连地表示的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或操作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
描述于本申请实施例中所涉及到的单元可以通过软件的方式实现，也可以通过硬件的方式来实现。所描述的单元也可以设置在处理器中，例如，可以描述为：一种处理器包括获取单元、匹配单元和推送单元。其中，这些单元的名称在某种情况下并不构成对该单元本身的限定，例如，获取单元还可以被描述为“获取当前的场景模型的单元”。
作为另一方面，本申请还提供了一种非易失性计算机存储介质，该非易失性计算机存储介质可以是上述实施例中所述装置中所包含的非易失性计算机存储介质；也可以是单独存在，未装配入终端中的非易失性计算机存储介质。上述非易失性计算机存储介质存储有一个或者多个程序，当所述一个或者多个程序被一个设备执行时，使得所述设备：获取当前的场景模型，其中场景模型包括：用户在预定时间段内在至少一个站点的操作行为信息和/或浏览行为信息，以及所述用户的用户特征和所述站点的站点特征；将所述当前的场景模型与至少一个预设场景模型进行匹配，获取与所述当前的场景模型的相似度大于预设阈值的预设场景模型；基于与所获取的预设场景模型相关联的信息推送规则，从多个待推送的信息中选择至少一个推送信息并进行推送。
以上描述仅为本申请的较佳实施例以及对所运用技术原理的说明。本领域技术人员应当理解，本申请中所涉及的发明范围，并不限于上述技术特征的特定组合而成的技术方案，同时也应涵盖在不脱离所述发明构思的情况下，由上述技术特征或其等同特征进行任意组合而形成的其它技术方案。例如上述特征与本申请中公开的(但不限于)具有类似功能的技术特征进行互相替换而形成的技术方案。

Claims (12)

  1. 一种信息推送方法，其特征在于，包括：
    获取当前的场景模型，其中场景模型包括：用户在预定时间段内在至少一个站点的操作行为信息和/或浏览行为信息，以及所述用户的用户特征和所述站点的站点特征；
    将所述当前的场景模型与至少一个预设场景模型进行匹配，获取与所述当前的场景模型的相似度大于预设阈值的预设场景模型；
    基于与所获取的预设场景模型相关联的信息推送规则，从多个待推送的信息中选择至少一个推送信息并进行推送。
  2. 根据权利要求1所述的信息推送方法，其特征在于，所述操作行为信息包括搜索行为信息和/或点击行为信息。
  3. 根据权利要求1所述的信息推送方法，其特征在于，所述站点特征包括以下至少一项：站点类型、站点规模、平均网站访问深度。
  4. 根据权利要求1-3所述的信息推送方法，其特征在于，所述将所述当前的场景模型与至少一个预设场景模型进行匹配，包括：
    基于操作行为信息、浏览行为信息、用户特征和站点特征中的一项或多项，将所述当前的场景模型与所述至少一个预设场景模型进行匹配。
  5. 根据权利要求1所述的信息推送方法，其特征在于，所述信息推送规则包括以下至少一项：计算规则、排序规则、调整规则。
  6. 一种信息推送装置，其特征在于，包括：
    获取单元，用于获取当前的场景模型，其中场景模型包括：用户在预定时间段内在至少一个站点的操作行为信息和/或浏览行为信息，以及所述用户的用户特征和所述站点的站点特征；
    匹配单元，用于将所述当前的场景模型与至少一个预设场景模型进行匹配，获取与所述当前的场景模型的相似度大于预设阈值的预设场景模型；
    推送单元，用于基于与所获取的预设场景模型相关联的信息推送规则，从多个待推送的信息中选择至少一个推送信息并进行推送。
  7. 根据权利要求6所述的信息推送装置，其特征在于，所述操作行为信息包括搜索行为信息和/或点击行为信息。
  8. 根据权利要求6所述的信息推送装置，其特征在于，所述站点特征包括以下至少一项：站点类型、站点规模、平均网站访问深度。
  9. 根据权利要求6-9所述的信息推送装置，其特征在于，所述匹配单元具体用于：
    基于操作行为信息、浏览行为信息、用户特征和站点特征中的一项或多项，将所述当前的场景模型与所述至少一个预设场景模型进行匹配，获取与所述当前的场景模型的相似度大于预设阈值的预设场景模型。
  10. 根据权利要求6所述的信息推送装置，其特征在于，所述信息推送规则包括以下至少一项：计算规则、排序规则、调整规则。
  11. 一种设备，包括：
    处理器；和
    存储器，
    所述存储器中存储有能够被所述处理器执行的计算机可读指令，在所述计算机可读指令被执行时，所述处理器执行信息推送方法，所述方法包括：
    获取当前的场景模型，其中场景模型包括：用户在预定时间段内在至少一个站点的操作行为信息和/或浏览行为信息，以及所述用户的用户特征和所述站点的站点特征；
    将所述当前的场景模型与至少一个预设场景模型进行匹配，获取与所述当前的场景模型的相似度大于预设阈值的预设场景模型；
    基于与所获取的预设场景模型相关联的信息推送规则，从多个待推送的信息中选择至少一个推送信息并进行推送。
  12. 一种非易失性计算机存储介质，所述计算机存储介质存储有能够被处理器执行的计算机可读指令，当所述计算机可读指令被处理器执行时，所述处理器执行信息推送方法，所述方法包括：
    获取当前的场景模型，其中场景模型包括：用户在预定时间段内在至少一个站点的操作行为信息和/或浏览行为信息，以及所述用户的用户特征和所述站点的站点特征；
    将所述当前的场景模型与至少一个预设场景模型进行匹配，获取与所述当前的场景模型的相似度大于预设阈值的预设场景模型；
    基于与所获取的预设场景模型相关联的信息推送规则，从多个待推送的信息中选择至少一个推送信息并进行推送。
PCT/CN2016/086214 2015-10-28 2016-06-17 信息推送方法和装置 WO2017071251A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510710944.2 2015-10-28
CN201510710944.2A CN105320766B (zh) 2015-10-28 2015-10-28 信息推送方法和装置

Publications (1)

Publication Number Publication Date
WO2017071251A1 true WO2017071251A1 (zh) 2017-05-04

Family

ID=55248152

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/086214 WO2017071251A1 (zh) 2015-10-28 2016-06-17 信息推送方法和装置

Country Status (2)

Country Link
CN (1) CN105320766B (zh)
WO (1) WO2017071251A1 (zh)

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110427453A (zh) * 2019-05-31 2019-11-08 平安科技（深圳）有限公司 数据的相似度计算方法、装置、计算机设备及存储介质
CN110910201A (zh) * 2019-10-18 2020-03-24 中国平安人寿保险股份有限公司 信息推荐的控制方法、装置、计算机设备及存储介质
CN111191115A (zh) * 2019-12-10 2020-05-22 未鲲(上海)科技服务有限公司 基于用户数据的信息推送方法及装置
CN111556116A (zh) * 2020-04-22 2020-08-18 深圳市元征科技股份有限公司 一种培训媒体推送方法、培训媒体的显示方法及相关装置
CN111600952A (zh) * 2020-05-15 2020-08-28 华人运通(上海)云计算科技有限公司 场景推送方法、执行方法、装置、终端、服务器和系统
CN111831903A (zh) * 2020-06-15 2020-10-27 北京嘀嘀无限科技发展有限公司 服务站点推荐方法、装置、设备及存储介质
CN111859156A (zh) * 2020-08-04 2020-10-30 上海风秩科技有限公司 发布人群的确定方法、装置、可读存储介质及电子设备
CN111898019A (zh) * 2019-05-06 2020-11-06 北京达佳互联信息技术有限公司 信息推送方法及装置
CN112231594A (zh) * 2019-06-27 2021-01-15 北京百度网讯科技有限公司 信息处理方法和装置
CN112541123A (zh) * 2020-12-24 2021-03-23 北京百度网讯科技有限公司 地图图区推荐方法、装置、设备及存储介质
CN114326423A (zh) * 2021-12-21 2022-04-12 美的集团股份有限公司 一种场景化控制方案的自动生成方法及场景化控制设备
CN114398558A (zh) * 2022-01-19 2022-04-26 北京百度网讯科技有限公司 信息推荐方法、装置、电子设备和存储介质
CN115580649A (zh) * 2022-08-31 2023-01-06 嘉兴学院 一种基于用户网络行为的智能信息推送方法
CN115665699A (zh) * 2022-12-27 2023-01-31 博信通信股份有限公司 一种多场景的信号覆盖优化方法、装置、设备及介质

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105320766B (zh) * 2015-10-28 2019-04-19 百度在线网络技术（北京）有限公司 信息推送方法和装置
CN107436893A (zh) * 2016-05-26 2017-12-05 北京搜狗科技发展有限公司 一种网页推荐方法和装置
CN106294778B (zh) * 2016-08-11 2019-09-10 北京小米移动软件有限公司 信息推送方法和装置
CN107733775A (zh) * 2016-08-12 2018-02-23 北京京东尚科信息技术有限公司 信息推送方法和装置
CN106548364B (zh) * 2016-09-27 2020-12-11 腾讯科技（北京）有限公司 信息发送方法及装置
CN106572390B (zh) * 2016-11-09 2020-03-13 咪咕音乐有限公司 一种音视频推荐方法及设备
CN107341172B (zh) * 2017-05-12 2020-06-19 阿里巴巴（中国）有限公司 视频收益计算建模装置与方法及视频推荐装置与方法
CN107147724A (zh) * 2017-05-23 2017-09-08 深圳市金立通信设备有限公司 一种消息推送方法、服务器及计算机可读存储介质
CN110049079A (zh) * 2018-01-16 2019-07-23 阿里巴巴集团控股有限公司 信息推送及模型训练方法、装置、设备及存储介质
CN108470034B (zh) * 2018-02-01 2019-09-20 百度在线网络技术（北京）有限公司 一种智能设备服务提供方法及系统
CN116112549A (zh) * 2023-01-13 2023-05-12 岚图汽车科技有限公司 一种车辆场景推荐方法及相关设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101520878A (zh) * 2009-04-03 2009-09-02 华为技术有限公司 向用户进行广告推送的方法、装置和系统
CN103064863A (zh) * 2011-10-24 2013-04-24 北京百度网讯科技有限公司 一种提供推荐信息的方法与设备
CN103399860A (zh) * 2013-07-04 2013-11-20 北京百纳威尔科技有限公司 内容显示方法和装置
CN103428075A (zh) * 2013-08-20 2013-12-04 贝壳网际（北京）安全技术有限公司 一种信息推送方法及装置
CN104008184A (zh) * 2014-06-10 2014-08-27 百度在线网络技术（北京）有限公司 信息的推送方法和装置
CN105320766A (zh) * 2015-10-28 2016-02-10 百度在线网络技术（北京）有限公司 信息推送方法和装置

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010128198A1 (en) * 2009-05-08 2010-11-11 Zokem Oy System and method for behavioural and contextual data analytics
CN104679743B (zh) * 2013-11-26 2019-10-29 阿里巴巴集团控股有限公司 一种确定用户的偏好模式的方法及装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101520878A (zh) * 2009-04-03 2009-09-02 华为技术有限公司 向用户进行广告推送的方法、装置和系统
CN103064863A (zh) * 2011-10-24 2013-04-24 北京百度网讯科技有限公司 一种提供推荐信息的方法与设备
CN103399860A (zh) * 2013-07-04 2013-11-20 北京百纳威尔科技有限公司 内容显示方法和装置
CN103428075A (zh) * 2013-08-20 2013-12-04 贝壳网际（北京）安全技术有限公司 一种信息推送方法及装置
CN104008184A (zh) * 2014-06-10 2014-08-27 百度在线网络技术（北京）有限公司 信息的推送方法和装置
CN105320766A (zh) * 2015-10-28 2016-02-10 百度在线网络技术（北京）有限公司 信息推送方法和装置

Cited By (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111898019A (zh) * 2019-05-06 2020-11-06 北京达佳互联信息技术有限公司 信息推送方法及装置
CN111898019B (zh) * 2019-05-06 2024-04-16 北京达佳互联信息技术有限公司 信息推送方法及装置
CN110427453B (zh) * 2019-05-31 2024-03-19 平安科技（深圳）有限公司 数据的相似度计算方法、装置、计算机设备及存储介质
CN110427453A (zh) * 2019-05-31 2019-11-08 平安科技（深圳）有限公司 数据的相似度计算方法、装置、计算机设备及存储介质
CN112231594B (zh) * 2019-06-27 2024-04-12 北京百度网讯科技有限公司 信息处理方法和装置
CN112231594A (zh) * 2019-06-27 2021-01-15 北京百度网讯科技有限公司 信息处理方法和装置
CN110910201A (zh) * 2019-10-18 2020-03-24 中国平安人寿保险股份有限公司 信息推荐的控制方法、装置、计算机设备及存储介质
CN111191115B (zh) * 2019-12-10 2022-10-11 未鲲(上海)科技服务有限公司 基于用户数据的信息推送方法及装置
CN111191115A (zh) * 2019-12-10 2020-05-22 未鲲(上海)科技服务有限公司 基于用户数据的信息推送方法及装置
CN111556116B (zh) * 2020-04-22 2023-05-30 深圳市元征科技股份有限公司 一种培训媒体推送方法、培训媒体的显示方法及相关装置
CN111556116A (zh) * 2020-04-22 2020-08-18 深圳市元征科技股份有限公司 一种培训媒体推送方法、培训媒体的显示方法及相关装置
CN111600952A (zh) * 2020-05-15 2020-08-28 华人运通(上海)云计算科技有限公司 场景推送方法、执行方法、装置、终端、服务器和系统
CN111831903A (zh) * 2020-06-15 2020-10-27 北京嘀嘀无限科技发展有限公司 服务站点推荐方法、装置、设备及存储介质
CN111859156A (zh) * 2020-08-04 2020-10-30 上海风秩科技有限公司 发布人群的确定方法、装置、可读存储介质及电子设备
CN111859156B (zh) * 2020-08-04 2024-02-02 上海秒针网络科技有限公司 发布人群的确定方法、装置、可读存储介质及电子设备
CN112541123A (zh) * 2020-12-24 2021-03-23 北京百度网讯科技有限公司 地图图区推荐方法、装置、设备及存储介质
CN114326423A (zh) * 2021-12-21 2022-04-12 美的集团股份有限公司 一种场景化控制方案的自动生成方法及场景化控制设备
CN114398558B (zh) * 2022-01-19 2023-05-23 北京百度网讯科技有限公司 信息推荐方法、装置、电子设备和存储介质
CN114398558A (zh) * 2022-01-19 2022-04-26 北京百度网讯科技有限公司 信息推荐方法、装置、电子设备和存储介质
CN115580649A (zh) * 2022-08-31 2023-01-06 嘉兴学院 一种基于用户网络行为的智能信息推送方法
CN115580649B (zh) * 2022-08-31 2024-05-03 嘉兴学院 一种基于用户网络行为的智能信息推送方法
CN115665699B (zh) * 2022-12-27 2023-03-28 博信通信股份有限公司 一种多场景的信号覆盖优化方法、装置、设备及介质
CN115665699A (zh) * 2022-12-27 2023-01-31 博信通信股份有限公司 一种多场景的信号覆盖优化方法、装置、设备及介质

Also Published As

Publication number Publication date
CN105320766A (zh) 2016-02-10
CN105320766B (zh) 2019-04-19

Similar Documents

Publication Publication Date Title
WO2017071251A1 (zh) 信息推送方法和装置
JP6967612B2 (ja) 情報検索方法、装置及びシステム
US11494666B2 (en) Systems and methods for online content recommendation
RU2636702C1 (ru) Способ и устройство для выбора сетевого ресурса в качестве источника содержимого для системы рекомендаций
CN109460514B (zh) 用于推送信息的方法和装置
US10430481B2 (en) Method and apparatus for generating a content recommendation in a recommendation system
CN110162693B (zh) 一种信息推荐的方法以及服务器
CN109086439B (zh) 信息推荐方法及装置
WO2017035970A1 (zh) 信息推送的方法和装置
US9078036B2 (en) Mechanism to evaluate the geographic popularity of geographically-located user-generated content items
US8589434B2 (en) Recommendations based on topic clusters
JP2020509449A (ja) 警告するための方法と装置
TW201911080A (zh) 搜索方法、搜索伺服器和搜索系統
CN114329201B (zh) 深度学习模型的训练方法、内容推荐方法和装置
CN105431834A (zh) 基于用户历史创建内容共享平台的个性化和连续播放列表
WO2009108576A2 (en) Prioritizing media assets for publication
CN106776707A (zh) 信息推送的方法和装置
CN110209931B (zh) 媒体内容的推送方法和装置、存储介质、电子装置
JP2024503774A (ja) 融合パラメータの特定方法及び装置、情報推奨方法及び装置、パラメータ測定モデルのトレーニング方法及び装置、電子機器、記憶媒体、並びにコンピュータプログラム
US20170148056A1 (en) Information processing device, control method, and program
RU2714594C1 (ru) Способ и система определения параметра релевантность для элементов содержимого
CN106354856B (zh) 基于人工智能的深度神经网络强化搜索方法和装置
CN110866040B (zh) 用户画像生成方法、装置和系统
EP3267386A1 (en) Method and apparatus for generating a content recommendation in a recommendation system
US20210117465A1 (en) Generic card feature extraction based on card rendering as an image

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: 16858694

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: 16858694

Country of ref document: EP

Kind code of ref document: A1