WO2021195922A1 - 一种当前页面信息刷新方法和系统 - Google Patents

一种当前页面信息刷新方法和系统 Download PDF

Info

Publication number
WO2021195922A1
WO2021195922A1 PCT/CN2020/082306 CN2020082306W WO2021195922A1 WO 2021195922 A1 WO2021195922 A1 WO 2021195922A1 CN 2020082306 W CN2020082306 W CN 2020082306W WO 2021195922 A1 WO2021195922 A1 WO 2021195922A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
user
currently displayed
information data
interest
Prior art date
Application number
PCT/CN2020/082306
Other languages
English (en)
French (fr)
Inventor
姚成杰
Original Assignee
浙江核新同花顺网络信息股份有限公司
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 浙江核新同花顺网络信息股份有限公司 filed Critical 浙江核新同花顺网络信息股份有限公司
Priority to CN202080005701.8A priority Critical patent/CN112912915A/zh
Priority to PCT/CN2020/082306 priority patent/WO2021195922A1/zh
Priority to US17/295,055 priority patent/US11971934B2/en
Publication of WO2021195922A1 publication Critical patent/WO2021195922A1/zh
Priority to US18/396,847 priority patent/US20240126823A1/en

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
    • 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
    • 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
    • G06F16/957—Browsing optimisation, e.g. caching or content distillation
    • G06F16/9574—Browsing optimisation, e.g. caching or content distillation of access to content, e.g. by caching
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/017—Gesture based interaction, e.g. based on a set of recognized hand gestures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising

Definitions

  • This application relates to the field of content recommendation, and in particular to a method and system for refreshing current page information.
  • One of the embodiments of the present application provides a method for refreshing current page information.
  • the method includes obtaining the currently displayed information data in response to receiving a user's information refresh request; determining the information that may be of interest to the user based on the information refresh request and the currently displayed information data; and showing the user the possibility Interested information.
  • the manner of the user's information refresh request includes gesture password, continuous click operation, key click operation, pause touch screen operation, continuous shaking operation, voice input operation, face recognition, expression recognition or iris recognition.
  • the information refresh request is a reverse information refresh request.
  • the information refresh request includes a reverse threshold, and the reverse threshold is used to characterize the degree of association between the refreshed information and the currently displayed information data.
  • the types of the reversal threshold include at least similar reversals, different reversals, or heterogeneous minority reversals.
  • the determining the information that may be of interest to the user based on the information refresh request and the currently displayed information data includes: determining the currently displayed information data type based on the currently displayed information data; The information refresh request and the currently displayed information data type perform reverse processing on the currently displayed information data type to obtain the type of information that may be of interest to the user; determine based on the type of information that may be of interest to the user Information that users may be interested in.
  • the determining the currently displayed information data type based on the currently displayed information data includes: using a machine learning model to process the currently displayed information data to determine the currently displayed information data type.
  • the machine learning model includes a classification model; the machine learning model is obtained by the following method: obtaining training samples; the training samples include historically displayed information data and historically displayed information data types; wherein The historically displayed information data type flag is used as the reference information data type; based on the training sample, an initial model is trained to obtain the machine learning model.
  • the reverse processing of the currently displayed information data type based on the currently displayed information data type to obtain the information type that may be of interest to the user includes: based on the currently displayed information data Type, select at least one data type among the types of the reverse threshold value, and use it as the information type that the user may be interested in.
  • the determining the information that may be of interest to the user based on the information refresh request and the currently displayed information data includes: in response to receiving the user's information refresh request, obtaining the user's historical browsing information; The currently displayed information data and the user’s historical browsing information determine the type of information data that is not of interest; based on the information refresh request and the type of information data that are not of interest, perform a check on the type of information data that is not of interest Reverse processing to obtain the type of information that may be of interest to the user; based on the type of information that may be of interest to the user, determine the type of information that may be of interest to the user.
  • the presenting the information that may be of interest to the user includes: presenting at least a part of the information that may be of interest on the user terminal according to a feed stream.
  • the system includes: at least one memory for storing computer instructions; at least one processor in communication with the memory, wherein when the at least one processor executes the computer instructions, the at least one processor causes the System execution: In response to receiving the user's information refresh request, obtain the currently displayed information data; based on the information refresh request and the currently displayed information data, determine the information that may be of interest to the user; show the user the possibility Interest information.
  • the information refresh request is a reverse information refresh request.
  • the at least one processor causes the system to further execute: based on the currently displayed information Data, determine the currently displayed information data type; based on the information refresh request and the currently displayed information data type, perform reverse processing on the currently displayed information data type to obtain information types that may be of interest to the user; based on The type of information that the user may be interested in determines the information that the user may be interested in.
  • the at least one processor in order to determine the currently displayed information data type based on the currently displayed information data, causes the system to further execute: use a machine learning model to process the currently displayed information data, Determine the data type of the information currently displayed.
  • the at least one processor in order to reversely process the currently displayed information data type based on the currently displayed information data type to obtain information types that may be of interest to the user, the at least one processor enables the system Further execution: based on the currently displayed information data type, at least one data type among the types of the reverse threshold value is selected and used as the information type that the user may be interested in.
  • the at least one processor causes the system to further execute: in response to receiving user information Refresh request to obtain the user’s historical browsing information; based on the currently displayed information data and the user’s historical browsing information, determine the type of information data that is not of interest; based on the information refresh request and the type of information data that are not of interest , Performing reverse processing on the uninteresting information data types to obtain the information types that the user may be interested in; based on the information types that the user may be interested in, determine the information that the user may be interested in.
  • the at least one processor causes the system to further execute: display at least of the information that may be of interest on the user terminal according to the feed stream. Part.
  • One of the embodiments of the present application provides a current page information refresh system.
  • the system includes: an obtaining module, which is used to obtain the currently displayed information data in response to receiving a user’s information refresh request; and a determining module, which is used to determine the user’s possibility based on the information refresh request and the currently displayed information data Interested information; the display module is used to show the information that may be of interest to the user.
  • the storage medium stores computer instructions, and when the computer reads the computer instructions in the storage medium, the computer executes the method described in any embodiment of the present application.
  • Fig. 1 is a schematic diagram of an application scenario of a current page information refresh system according to some embodiments of the present application
  • Figure 2 is a block diagram of a current page information refresh system according to some embodiments of the present application.
  • Fig. 3 is an exemplary flowchart of a method for refreshing current page information according to some embodiments of the present application
  • Fig. 4 is an exemplary flowchart of a method for determining information that a user may be interested in according to some embodiments of the present application
  • Fig. 5 is an exemplary flowchart of a method for determining information that a user may be interested in according to another embodiment of the present application.
  • Fig. 6 is an exemplary flowchart of a machine learning model training method according to some embodiments of the present application.
  • system is a method for distinguishing different components, elements, parts, parts, or assemblies of different levels.
  • the words can be replaced by other expressions.
  • Fig. 1 is a schematic diagram of an application scenario of a current page information refresh system according to some embodiments of the present application.
  • the current page information refreshing system 100 can refresh the display information of the current page according to the needs of the user, so that the user can learn more other types of information.
  • the current page information refresh system 100 may be a service platform for the Internet or other networks.
  • the current page information refresh system 100 may be an online service platform that provides users with information or video information.
  • the current page information refreshing system 100 can be applied to online shopping services, such as buying clothes, books, daily necessities, and so on.
  • the current page information refreshing system 100 can also be applied to the field of travel (eg, tourism) services.
  • the current page information refresh system 100 may include, but is not limited to, a server 110, a user terminal 120, a storage device 130, an information source 140, and a network 150.
  • the server 110 may be used to process information and/or data related to a service request, for example, to process a user's information refresh request. Specifically, the server 110 may receive an information refresh request from the user terminal 120, and process the information refresh request to send the user terminal 120 information that may be of interest to the user.
  • the server 110 may be a single server or a group of servers. The server group may be centralized or distributed (for example, the server 110 may be a distributed system). In some embodiments, the server 110 may be local or remote. For example, the server 110 may access information and/or data stored in the storage device 130 and the user terminal 120 through the network 150.
  • the server 110 may be directly connected to the storage device 130 and the user terminal 120 to access the stored information and/or data.
  • the server 110 may be implemented on a cloud platform.
  • the cloud platform may include private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, inter-cloud, multiple clouds, etc., or any combination of the foregoing examples.
  • the server 110 may include a processing engine 112.
  • the processing engine 112 may process data and/or information related to the current page information refresh request to perform one or more functions described in this application. For example, the processing engine 112 can receive the information refresh request sent by the user terminal 120, and obtain the currently displayed information data, and then can determine the information that may be of interest to the user based on the information refresh request and the currently displayed information data, and finally can show the user the possibility Interested information.
  • the processing engine 112 may include one or more processing engines (for example, a single-chip processing engine or a multi-chip processor).
  • the processing engine 112 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), an application specific instruction set processor (ASIP), an image processing unit (GPU), a physical operation processing unit (PPU), and digital signal processing.
  • CPU central processing unit
  • ASIC application specific integrated circuit
  • ASIP application specific instruction set processor
  • GPU graphics processing unit
  • PPU physical operation processing unit
  • DSP Controller
  • FPGA Field Programmable Gate Array
  • PLD Programmable Logic Device
  • Controller Microcontroller Unit, Reduced Instruction Set Computer (RISC), Microprocessor, etc. or any combination of the above.
  • the user terminal 120 may be a person, tool, or other entity directly related to the information refresh request.
  • the user can be an information refresh requester.
  • “user” and “user terminal” can be used interchangeably.
  • the user terminal 120 may include a mobile device 120-1, a tablet computer 120-2, a notebook computer 120-3, a desktop computer 120-4, etc., or any combination thereof.
  • the mobile device 120-1 may include a smart home device, a wearable device, a smart mobile device, a virtual reality device, an augmented reality device, etc., or any combination thereof.
  • smart home devices may include smart lighting devices, smart electrical appliance control devices, smart monitoring devices, smart TVs, smart cameras, walkie-talkies, etc., or any combination thereof.
  • the wearable device may include smart bracelets, smart footwear, smart glasses, smart helmets, smart watches, smart wearers, smart backpacks, smart accessories, etc., or any combination thereof.
  • smart mobile devices may include smart phones, personal digital assistants (PDAs), gaming devices, navigation devices, point of sale (POS), etc., or any combination thereof.
  • the virtual reality device and/or augmented reality device may include a virtual reality helmet, virtual reality glasses, virtual reality goggles, augmented virtual reality helmets, augmented reality glasses, augmented reality goggles, etc., or any combination thereof.
  • the virtual reality device and/or the augmented reality device may include Google Glass, Oculus Rift, HoloLens, Gear VR, or the like.
  • the storage device 130 may store data and/or instructions related to the user's information refresh request. In some embodiments, the storage device 130 may store the currently displayed information data. In some embodiments, the storage device 130 may store historical browsing information of the user. In some embodiments, the storage device 130 may store data and/or instructions used by the server 110 to execute or use to complete the exemplary methods described in this application. In some embodiments, the storage device 130 may include mass memory, removable memory, volatile read-write memory, read-only memory (ROM), etc., or any combination thereof. Exemplary mass storage devices may include magnetic disks, optical disks, solid state disks, and the like. Exemplary removable storage may include flash drives, floppy disks, optical disks, memory cards, compact disks, magnetic tapes, and the like.
  • An exemplary volatile read-only memory may include random access memory (RAM).
  • RAM may include dynamic RAM (DRAM), double rate synchronous dynamic RAM (DDR SDRAM), static RAM (SRAM), thyristor RAM (T-RAM), zero capacitance RAM (Z-RAM), and the like.
  • exemplary ROMs may include mask ROM (MROM), programmable ROM (PROM), erasable programmable ROM (PEROM), electronically erasable programmable ROM (EEPROM), compact disk ROM (CD-ROM), and digital General-purpose disk ROM, etc.
  • the storage device 130 may be implemented on a cloud platform.
  • the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc., or any combination thereof.
  • the storage device 130 may be connected to the network 150 to communicate with one or more components (for example, the server 110, the user terminal 120) in the current page information refresh system 100.
  • One or more components in the current page information refresh system 100 can access data or instructions stored in the storage device 130 through the network 150.
  • the storage device 130 may directly connect or communicate with one or more components (for example, the server 110 and the user terminal 120) in the current page information refreshing system 100.
  • the storage device 130 may be part of the server 110.
  • the network 150 may facilitate the exchange of information and/or data.
  • one or more components in the current page information refresh system 100 may send to/from other components in the current page information refresh system 100 via the network 150 And/or receive information and/or data.
  • the server 110 may obtain/acquire a service request (for example, an information refresh request) from the user terminal 120 through the network 150.
  • the network 150 may be any form of wired or wireless network or any combination thereof.
  • the network 150 may include a cable network, a wired network, an optical fiber network, a telecommunication network, an internal network, the Internet, a local area network (LAN), a wide area network (WAN), a wireless local area network (WLAN), a metropolitan area network (MAN), Wide Area Network (WAN), Public Switched Telephone Network (PSTN), Bluetooth Network, Zigbee Network, Near Field Communication (NFC) Network, Global System for Mobile Communications (GSM) Network, Code Division Multiple Access (CDMA) Network, Time Division Multiple Access ( TDMA) network, general packet radio service (GPRS) network, enhanced data rate GSM evolution (EDGE) network, wideband code division multiple access (WCDMA) network, high-speed downlink packet access (HSDPA) network, long-term evolution (LTE) Network, user datagram protocol (UDP) network, transmission control protocol/Internet protocol (TCP/IP) network, short message service (SMS) network, wireless application protocol (WAP) network, ultra-wideband (UWB) network, infrared
  • the current page information refresh system 100 may include one or more network access points.
  • the current page information refreshing system 100 may include wired or wireless network access points, such as base stations and/or wireless access points 150-1, 150-2, ..., and one or more components of the current page information refreshing system 100 may pass through It is connected to the network 150 to exchange data and/or information.
  • the information source 140 may generally refer to all information sources except the information provided by the user terminal 120.
  • the information source 140 may include, but is not limited to, various information sources that can provide information, such as shopping websites, portal websites, stock exchanges, microblogs, blogs, personal websites, and libraries.
  • the information source 140 may be implemented in a single central server, multiple servers connected through a communication link, or multiple personal devices.
  • the personal device can generate content (for example, referred to as "user-generated content"), such as uploading text, sound, image, video, etc. to the cloud server, so that the cloud server can be combined with Multiple personal devices connected to it form an information source.
  • the information source 140 may provide some relevant information, including but not limited to securities news, market analysis, social hotspots, financial opinions, market analysis, industry research reports, company announcements, investment opportunities, funds, commodities, and Hong Kong stocks.
  • relevant information including but not limited to securities news, market analysis, social hotspots, financial opinions, market analysis, industry research reports, company announcements, investment opportunities, funds, commodities, and Hong Kong stocks.
  • US stocks etc.
  • Fig. 2 is a block diagram of a current page information refresh system according to some embodiments of the present application.
  • the current page information refresh system may include an acquisition module 210, a determination module 220, a display module 230, and a machine learning model training module 240.
  • the obtaining module 210 may be used to obtain the currently displayed information data in response to receiving the user's information refresh request.
  • the manner of the user's information refresh request may include gesture password, continuous click operation, key click operation, pause touch screen operation, continuous shaking operation, voice input operation, face recognition, expression recognition, or iris recognition.
  • the information refresh request may be a reverse information refresh request.
  • the information refresh request may include a reverse threshold, and the reverse threshold is used to characterize the degree of association between the refreshed information and the currently displayed information data.
  • the types of reversal thresholds may include at least similar reversals, different reversals, or heterogeneous minority reversals. For a detailed description of obtaining the currently displayed information data, please refer to FIG. 3, which will not be repeated here.
  • the determining module 220 may be used to determine information that may be of interest to the user based on the information refresh request and the currently displayed information data.
  • the currently displayed information data type can be determined based on the currently displayed information data, and then based on the information refresh request and the currently displayed information data type, reverse processing can be performed on the currently displayed information data type, so that the user may be interested Therefore, based on the type of information that the user may be interested in, the information that the user may be interested in can be determined.
  • the information that may be of interest to the user please refer to the content of FIG. 4, which will not be repeated here.
  • information that may be of interest to the user may also be determined based on the information refresh request, the currently displayed information data, and the user's historical browsing information. Specifically, in response to receiving a user's information refresh request, the user's historical browsing information can be obtained, and the type of information data that is not of interest can be determined based on the currently displayed information data and the user's historical browsing information, and then based on the information refresh request And the uninteresting information data type, reverse processing the uninteresting information data type to obtain the information type that the user may be interested in, so that the information that the user may be interested in can be determined based on the information type that the user may be interested in. For a detailed description of determining the information that may be of interest to the user, refer to the content of FIG. 5, which will not be repeated here.
  • the display module 230 may be used to display information that may be of interest to the user. Specifically, at least a part of the information that may be of interest can be displayed on the user terminal according to the feed stream mode. For a detailed description of displaying information that may be of interest to the user, refer to FIG. 3, which is not repeated here.
  • the machine learning model training module 240 may be used to train the initial model to obtain the machine learning model.
  • training samples can be obtained.
  • the training samples can include historically displayed information data and historically displayed information data types; historically displayed information data types can be marked as reference information data types; then based on the training samples, the initial model can be obtained by training Machine learning model.
  • the initial model can be obtained by training Machine learning model.
  • system and its modules shown in FIG. 2 can be implemented in various ways.
  • the system and its modules may be implemented by hardware, software, or a combination of software and hardware.
  • the hardware part can be implemented using dedicated logic;
  • the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware.
  • processor control codes for example on a carrier medium such as a disk, CD or DVD-ROM, such as a read-only memory (firmware Such codes are provided on a programmable memory or a data carrier such as an optical or electronic signal carrier.
  • the system and its modules of this application can not only be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc. It may also be implemented by software executed by various types of processors, or may be implemented by a combination of the above hardware circuit and software (for example, firmware).
  • the acquisition module 210, the determination module 220, the display module 230, and the machine learning model training module 240 may be different modules in one system, or one module may implement the above two or more modules.
  • the determining module 220 and the machine learning model training module 240 may be two modules, or one module can simultaneously determine information that may be of interest to the user and model training functions.
  • each module may share a storage module, and each module may also have its own storage module. Such deformations are all within the protection scope of this application.
  • Fig. 3 is an exemplary flowchart of a method for refreshing current page information according to some embodiments of the present application.
  • Step 310 In response to receiving the user's information refresh request, obtain the currently displayed information data.
  • step 310 may be implemented by the acquisition module 210.
  • the user may include a user who uses or browses the interface of the current application (or first-party), and may also be a user who uses or browses a third-party application through the current application interface.
  • Third-party applications are other applications relative to the current application. For example, if the Tencent Video website is the current application, and a shopping advertisement on JD.com appears on the page of the Tencent Video website, JD.com is a third-party application. Users can jump to the shopping page of JD.com by clicking on the shopping ad on JD on the Tencent Video website page.
  • the information refresh request may include an instruction for the user to request a re-recommendation of the currently browsed resource. For example, after a user has browsed on the Douyin Short Video APP interface for a period of time, and hopes that the Douyin Short Video APP recommends some new videos, they can send a refresh request by pulling down the page.
  • the manner in which the user sends the information refresh request may include gesture password, continuous tap operation, key tap operation, pause touch screen operation, continuous shaking operation, voice input operation, face recognition, expression recognition, or iris recognition.
  • Gesture passwords can include drawing ⁇ , drawing ⁇ , or drawing Z on the screen.
  • the gesture password can also include tapping the password input control on the screen, and entering a number password or specific characters in the pop-up password input box.
  • the continuous tap operation may include multiple (e.g., 2, 3, or 4) taps on the screen within a short time (e.g., 3 seconds). Clicking the button operation can set a function button on the current application interface, and the information refresh request can be completed by clicking the function button.
  • Pausing the touch screen operation can include touching the screen continuously for a period of time (eg, 3 seconds, 4 seconds, or 5 seconds).
  • the continuous shaking operation may be based on a certain intensity (for example, 3 seconds, 4 seconds, or 5 seconds) to keep the user terminal device in a motion state (for example, shaking the mobile phone strongly for 5 seconds).
  • the voice input operation may include a specific voice content text or instruction contained in the user's voice segment, for example, a content text or instruction containing "information refresh".
  • Face recognition may be to take an image or video stream of a human face through a camera on a user terminal device, and perform identity recognition based on facial feature information to verify whether an information refresh request is sent. For example, if the face verification is passed, the information refresh request instruction is sent; otherwise, the information refresh request instruction is not sent.
  • the facial expression recognition can be to take an image or a video stream of a human face through a camera on the user terminal device, and separate a specific facial expression state to verify whether to send an information refresh request.
  • the facial expression status is "smiling”
  • the verification is passed, and the information refresh request instruction is sent; if the facial expression state is other states (such as frowning, crying, angry), the verification is not passed, and the information refresh request instruction is not sent.
  • Iris recognition may be to photograph the user's iris through a camera on the user terminal device to verify whether to send an information refresh request. For example, if the iris verification is passed, the information refresh request instruction is sent; otherwise, the information refresh request instruction is not sent.
  • the currently displayed information data may be the information data of the first-party application currently displayed on the user terminal, or may be the information data of the third-party application currently displayed on the user terminal.
  • the information refresh request may be a request to refresh the information displayed on the current user terminal page.
  • the information refresh request may be an information refresh request in the same direction.
  • the information refresh request in the same direction may include an instruction for the user to request to re-recommend other information of the same type as the information displayed on the current user terminal page.
  • the information refresh request may also be a reverse information refresh request.
  • the information reverse refresh request may include an instruction that the user requests to re-recommend other information of a different type from the information displayed on the current user terminal page.
  • the manner in which the user sends the information refresh request in the same direction or the information reverse refresh request may be set in advance by the user, or may be set by default in the system.
  • the information refresh request may include a reverse threshold, and the reverse threshold may be used to characterize the degree of relevance between the refreshed information and the currently displayed information data, that is, the level or degree of reverse.
  • the reverse threshold may include 0-10 levels, and may also be replaced by other symbols describing the intensity of the levels (for example, A-K levels), which is not limited in this description. Specifically, if the reverse threshold is set to be higher, the recommended content can be made to have a higher degree of reverse, and the refreshed information has a lower correlation with the currently displayed information data; if the reverse threshold is set to be low, you can If the recommended content is partially reversed, the refreshed information has a higher degree of relevance to the currently displayed information data.
  • a user uses a car APP
  • the price range of the car that the user has viewed or followed for a long time is between 100,000 and 200,000; when the user sets the reverse threshold to level 0, the user sends a reverse refresh request for information.
  • the price of APP recommended models is between 100,000 and 200,000; when the user sets the reverse threshold to level 10, after the user sends a reverse information refresh request, the car APP recommends prices other than 100,000 to 200,000 Car model:
  • the automotive APP recommends some models of 100,000 to 200,000, and some models of other price ranges.
  • the types of reversal thresholds may include at least similar reversals, different types of reversals, or heterogeneous minority reversals.
  • the reverse of the same kind can be the reverse of the small category in the same category, for example, the small category in the big category is basketball in sports, and the reverse of the same kind can be the latest in sports such as table tennis, badminton, volleyball, etc. News.
  • Different types of reverses can be reverses that do not belong to the same category. For example, a certain category is sports, and different types of reverses can be entertainment, military, finance, tourism, history, etc.
  • Heterogeneous niche reverses may be categories (including large and/or small categories) with low attention and few attention groups, for example, astronomy, mathematical conjecture, religious research, curling projects, etc.
  • the type of the reversal threshold may be set to at least one of the same type of reversal, different types of reversals, and heterogeneous minority reversals.
  • the way of setting the reverse threshold may include radar chart, percentage, level, and so on.
  • the radar chart may include one or more types of reverse thresholds, and by setting the proportions of the types of different reverse thresholds, the degree of reverse refresh of the reverse refresh can be adjusted.
  • the radar chart includes 4 categories: similar non-reverse, similar reverse, different reverse, and heterogeneous minority reverse.
  • the same non-reverse can be the same as the type of information currently displayed.
  • the proportions of the categories are: 0%, 10%, 50%, 40%.
  • the recommended content will not contain the same content as the currently displayed information data type, but will include 10% of the same reverse content and 50% different Class reverse and 40% heterogeneous niche reverse.
  • the recommended content will all recommend heterogeneous and niche reverse content.
  • the radar chart may also include only three categories: similar reverses, different types of reverses, and heterogeneous niche reverses.
  • the radar chart may also be in other forms, which are not limited in this application.
  • the proportions of different categories the recommended content can be reversed in different types and degrees.
  • the degree of reverse refresh can be adjusted, and the percentages or levels correspond to the proportions of each category in different reverse threshold types.
  • the percentage of the reverse threshold is set to 80% (or level H)
  • the recommended content may include 50% different types of reverses and 50% heterogeneous niche reverses.
  • the percentage or level can be preset by the system or determined by the background server through a corresponding algorithm, which is not limited in this application.
  • the reverse threshold can be set in the function setting box that pops up on the application page after the user sends the information refresh request; it can also be set on the function control of the application page before the user sends the information refresh request. Settings; it can also be set as the system default.
  • the user's information reverse refresh request may be the user's information reverse refresh request for the recommended content of the current application.
  • the user's information reverse refresh request may also be the user's information reverse refresh request for the recommended content of the third-party application.
  • the third-party application can obtain the user's reverse refresh request for the third-party application's information on the current application.
  • the third-party application may obtain the user's reverse refresh request for the third-party application's information based on the plug-in of the third-party application.
  • the plug-ins of third-party applications can only run on the first-party application platform specified by the program (may support multiple platforms at the same time), and cannot run separately from the designated first-party application platform.
  • Taobao requires advertisements on application platforms such as Toutiao, Douyin, and Weibo
  • Taobao plug-ins for third-party applications can support first-party application platforms such as Toutiao, Douyin, and Weibo to enter Taobao for shopping.
  • the plug-in of the third-party application is a tool for the third-party application to interact with the current application.
  • the plug-in of the third-party application may include obtaining the user's information of the third-party application in the current application, reverse refresh request, and obtaining the currently displayed third-party application
  • the information data of the user, the information to determine the third-party application that the user may be interested in, and the information of the third-party application that the user may be interested in are displayed to the user.
  • the acquisition module 210 can be operated by the user on the user terminal (such as: gesture password, continuous click operation, click button operation, pause touch screen operation, continuous shaking operation, voice input operation, face recognition, expression Recognition or iris recognition) to obtain the user’s information refresh request.
  • the user can use the application more conveniently and quickly, and bring a better user experience to the user.
  • Step 320 Determine information that may be of interest to the user based on the information refresh request and the currently displayed information data.
  • step 320 may be implemented by the determining module 220.
  • the information that the user may be interested in may include information content that the user desires to browse. In some embodiments, the information that the user may be interested in may be the information content of the first-party application that the user desires to browse, or the information content of the third-party application that the user desires to browse.
  • information that may be of interest to the user may be determined based on the information refresh request and the currently displayed information data.
  • the currently displayed information data type can be determined based on the currently displayed information data, and then based on the information refresh request and the currently displayed information data type, reverse processing can be performed on the currently displayed information data type, so that the user may be interested
  • reverse processing can be performed on the currently displayed information data type, so that the user may be interested
  • determine the information that the user may be interested in determine the information that the user may be interested in.
  • information that may be of interest to the user may also be determined based on the information refresh request, the currently displayed information data, and the user's historical browsing information. Specifically, in response to receiving the user's information refresh request, the user's historical browsing information can be obtained, and then based on the currently displayed information data and the user's historical browsing information, the type of information data that is not of interest can be determined, so that the information refresh request can be based And uninteresting information data types, reverse processing the uninteresting information data types to obtain the information types that the user may be interested in, and then based on the information types that the user may be interested in, determine the information that the user may be interested in. For a detailed description of determining the information that may be of interest to the user, refer to FIG. 5, which will not be repeated here.
  • the third-party application plug-in may determine information that may be of interest to the user based on the currently displayed information data of the current application. Specifically, the third-party application plug-in can map the data type of the current application (first-party application) to the data type of the third-party application, and establish a relationship between at least one data type in the current application and at least one data type in the third-party application.
  • the currently displayed information data type of the current application can be determined according to the currently displayed information data of the current application and the data type in the current application (first party), and then the third-party application can be determined according to the correspondence between the above data types
  • the corresponding information data type on the third-party application reverse processing is performed on the corresponding information data type on the third-party application, so as to determine the type of information that may be of interest to the user on the third-party application, and then determine the information that may be of interest to the user on the third-party application.
  • the content of the information that may be of interest to the user determined on the third-party application is similar to the description in FIG. 4, and will not be repeated here.
  • the correspondence between the at least one data type in the current application and the at least one data type in the third-party application may be one-to-one correspondence, one-to-many, many-to-one, or many-to-many, which is not limited in this application.
  • first-party application For the method of obtaining the data type of the current application (first-party application), refer to step 420, which will not be repeated here.
  • the third-party application plug-in may also determine the user based on the currently displayed information data of the current application and the historical browsing information of the user of the current application. Information that may be of interest. Specifically, the third-party application plug-in can map the data type of the current application (first-party application) to the data type of the third-party application, and establish a relationship between at least one data type in the current application and at least one data type in the third-party application.
  • the information data type of the current application that is not of interest can be determined, and then according to the above
  • the correspondence of data types determines the corresponding uninterested information data types on third-party applications, and reversely processes the uninterested information data types on third-party applications, thereby determining the types of information that users may be interested in on third-party applications, and then Identify information that may be of interest to users on third-party applications.
  • the content of the information that may be of interest to the user determined on the third-party application is similar to the description in FIG. 5, and will not be repeated here.
  • the determining module 220 may determine information that may be of interest to the user based on the information refresh request and the currently displayed information data.
  • Step 330 Show the user information that may be of interest.
  • step 330 may be implemented by the display module 230.
  • the information that may be of interest may be all data information generated within a period of time (eg, 1 week, 3 days, 1 day, 12 hours, or 1 hour) before the current time in the current application or a third-party application.
  • at least a part of the information that may be of interest may be displayed on the user terminal page according to the feed stream.
  • Feed stream is a way of presenting content to users and continuously updating. Based on the information refresh request and the currently displayed information data, it can determine the information that may be of interest to the user, and refresh the page to display the information that may be of interest to the user.
  • the feed stream may include three modes: Push, Pull, and Hybrid.
  • the push mode may be that after the user generates content, the server pushes the determined information that the user may be interested in to some other users, which is suitable for applications with a relatively uniform number of user relationships and an upper limit, such as Moments of Friends;
  • the pull mode is when an information refresh request is issued, the page presents updated data according to certain rules, such as update time, popularity, editorial recommendations, etc., suitable for applications with a small number of users and low daily activity;
  • push-pull combined mode can include online push , Offline pull (e.g., after the Weibo big V publishes a dynamic, it will only be pushed to fans online at the same time, and offline fans will pull the dynamic after it goes online); or regular push or offline pull (e.g., after the Weibo big V publishes a dynamic, with The form of resident process is pushed to fans' attention) two kinds.
  • the feed stream may continuously update information that may be of interest to the user through user operations (eg, pull-up or pull-down operations).
  • the potentially interesting information displayed in the feed stream can be sorted by Timeline (time order), for example, sorted in the order of posting time, with the first posted first seen first, and then posted second listed at the top.
  • the potentially interesting information displayed by the feed stream may be ranked by rank (non-time factor), for example, by popularity, the information that may be of interest to the user is ranked by popularity, and the most popular is recommended first.
  • obtaining the currently displayed information data in step 310 may be obtained when the user logs in on the current application platform.
  • receiving the user's information refresh request it is determined that the user may be interested in information, and it is not necessarily limited to receiving the user's information. Obtained after refresh request.
  • Fig. 4 is an exemplary flowchart of a method for determining information that a user may be interested in according to some embodiments of the present application. As shown in FIG. 4, the method 400 for determining information that a user may be interested in may include:
  • Step 410 Determine the currently displayed information data type based on the currently displayed information data. In some embodiments, this step 410 may be performed by the determining module 220.
  • the currently displayed information data type may be the data type of the currently displayed information data.
  • the data type may be a data tag of massive data in the current page information refresh system database (for example, the storage device 130).
  • the data label can be obtained by labeling the massive data based on an algorithm.
  • the data type may include a large category label and a small category label.
  • the data type can be composed of two or more types.
  • the currently displayed information data type may include at least one or more data types.
  • a machine learning model may be used to determine the type of information data currently displayed. Specifically, the currently displayed information data can be input into the machine learning model for processing, and the currently displayed information data type can be output.
  • the machine learning model may include a classification model.
  • a classification model For example, decision trees, Bayesian classification, random forests, support vector machines, neural networks and other models.
  • the decision tree model may include, but is not limited to, Classification and Regression Tree (CART), Iterative Dichotomiser 3 (ID3), C4.5 algorithm, Random Forest (Random Forest), Chisquared Automatic Interaction Detection (CHAID), Multivariate Adaptive Regression Splines (MARS), Gradient Boosting Machine (GBM), etc., or any combination thereof.
  • CART Classification and Regression Tree
  • ID3 Iterative Dichotomiser 3
  • CHAI Chisquared Automatic Interaction Detection
  • MAM Multivariate Adaptive Regression Splines
  • GBM Gradient Boosting Machine
  • the determining module 220 may access the machine learning model stored in the storage device 130 through the network 150, and determine the type of information data currently displayed based on the currently displayed information data.
  • Step 420 Based on the information refresh request and the currently displayed information data type, reverse processing the currently displayed information data type to obtain information types that may be of interest to the user. In some embodiments, this step 420 may be performed by the determining module 220.
  • the type of information that the user may be interested in may include the data type of the information content that the user desires to browse.
  • at least one data type among the types of the reverse threshold may be selected as the information type that the user may be interested in.
  • the data type can be obtained, and then at least one data type other than the currently displayed information data type (that is, in the reverse threshold type) from the data type can be selected as the information type that may be of interest to the user .
  • the types of reversal thresholds can include at least similar reversals, different types of reversals, or heterogeneous niche reversals.
  • Each of the foregoing reverse threshold types may include one or more data types.
  • the type of the reverse threshold and the currently displayed information data type may constitute a complete set of data types.
  • the data type may be obtained by classifying data marked with data tags based on a classification algorithm.
  • the classification algorithm as a supervised machine learning method, is to classify the type of data that has been labeled, and the number of categories is fixed.
  • the classification algorithm may include a decision tree algorithm, a K-Nearest Neighbor (KNN) algorithm, a Bayes algorithm, a support vector machine algorithm, and the like.
  • the data type may also be obtained by clustering massive data based on a clustering algorithm.
  • the clustering algorithm can also be used as a preprocessing step of the classification algorithm in the data mining algorithm.
  • the clustering algorithm is an unsupervised machine learning method that does not require manual labeling and pre-training of the classifier.
  • the categories are automatically generated during the clustering process, and the category data is uncertain.
  • data with a data type of sub-labels can be clustered into data with large labels based on a clustering algorithm.
  • part of the data has sub-labels such as football, UEFA Published League, star Messi, Barcelona Football Club, recent events, etc.
  • the other part of the data has sub-category tags such as basketball, NBA, star Curry, Golden State Warriors, regular season, etc.
  • the two parts of data can be clustered into big-category tags based on similarity or deep learning algorithms.
  • the big-category tags can For sports.
  • the clustering algorithm may include K-Means clustering algorithm, mean shift clustering algorithm, density-based clustering algorithm (Density-Based Spatial Clustering of Applications with Noise, DBSCAN), using Gaussian Hybrid model (GMM) expectation maximization (EM) clustering algorithm, agglomerated hierarchical clustering algorithm, graph community detection (Graph Community Detection) clustering algorithm.
  • K-Means clustering algorithm mean shift clustering algorithm
  • density-based clustering algorithm Density-Based Spatial Clustering of Applications with Noise, DBSCAN
  • GMM Gaussian Hybrid model
  • EM expectation maximization
  • agglomerated hierarchical clustering algorithm graph community detection (Graph Community Detection) clustering algorithm.
  • the data type may be acquired in real time or in advance, and the real-time acquisition may be acquired based on a classification algorithm or a clustering algorithm. In some embodiments, the data type may be obtained based on factors such as time, popularity, scenario, and collaborative recommendation. For example, the data type of data updated in the past week, the data type of the latest/hottest data, and Obtain the data type of the data produced by the data producer that the platform cooperates with.
  • Step 430 Determine information that may be of interest to the user based on the type of information that may be of interest to the user. In some embodiments, this step 430 may be performed by the determining module 220.
  • Each information type can have one or more corresponding information data.
  • the information that may be of interest to the user may be determined based on the type of information that may be of interest to the user.
  • the combination of data types, the information that users may be interested in can be the corresponding Information data of a combination of two data types. Since there are one or more information data corresponding to each information type, the amount of information that the user may be interested in is relatively large, and it will not be listed here.
  • the type of data and the type of information that the user may be interested in in the process 400 are not limited to the number listed, and may also be other numbers.
  • performing reverse processing in step 420 to obtain information types that may be of interest to the user can also be replaced with: selecting at least one data type in the complete set of data types other than the currently displayed information data type, and using it as the user possibility The type of information of interest.
  • Fig. 5 is an exemplary flowchart of a method for determining information that may be of interest to a user according to another embodiment of the present application. As shown in FIG. 5, the method 500 for determining information that may be of interest to a user may include:
  • Step 510 In response to receiving the user's information refresh request, obtain the user's historical browsing information. In some embodiments, this step 510 may be performed by the obtaining module 210.
  • the user's historical browsing information may include multi-dimensional information such as the browsing content, browsing time, and browsing frequency of the current application that the user browses before the current moment (for example, one month, one week, three days, or one day).
  • Browsing information can include pictures, text, video, audio, and so on.
  • the user's historical browsing information may include, but is not limited to: posting, following, favorites, comments, like content and/or time and other information.
  • the user's historical browsing information may include cloud-stored historical browsing data or locally stored cookie data.
  • the historical browsing data stored in the cloud may be the historical browsing data of the user stored in the cloud storage (for example, the storage device 130).
  • the locally stored cookie data may include small text files stored in the local client (eg, the user terminal 120).
  • the cookie data can include the user's personal information and historical browsing data.
  • the obtaining module 210 may obtain the user's historical browsing information in the cloud storage or the local client in response to receiving the user's information refresh request.
  • the user's portrait By analyzing the user's historical browsing information, the user's portrait can be portrayed, and data support can also be provided for various operating projects. For example, the number of clicks on a product in a certain shop of a shopping website far exceeds that of other commodities, and the merchants of the shop can be guided to increase the production or inventory of the product.
  • Step 520 Determine the type of information data that is not of interest based on the currently displayed information data and the user's historical browsing information. In some embodiments, this step 520 may be performed by the determining module 220.
  • the type of information data that is not of interest may be the type of information data that the user does not want to browse. In some embodiments, the type of information data that is not of interest may be determined based on the currently displayed information data and the user's historical browsing information, respectively. The type of information data that is not of interest may include the type of information data currently displayed and the type of historical preference of the user. For more details on determining the type of information data currently displayed based on the currently displayed information data, please refer to step 410, which is not repeated here. In some embodiments, the user's historical preference type can also be determined based on the user's historical browsing information.
  • the historical browsing data stored in the cloud may be processed to obtain the user's historical preference type.
  • a machine learning model can be used to process the user's historical browsing information to determine the user's historical preference type.
  • the machine learning model may be the same as the machine learning model in step 410, and will not be repeated here. For more details of the training method of the machine learning model, refer to FIG. 6 and its description, which will not be repeated here.
  • the category attribute in the locally stored cookie data can be extracted as the user's historical preference type.
  • the locally stored cookie data can be obtained, and the information refresh request can be extracted.
  • the data type (or called the category attribute) of the cookie data is the user's historical preference type.
  • the locally stored cookie data can be an encrypted hash code, and the server can decrypt the hash code, and then read the data type as the user's historical preference type.
  • the information data type that is not of interest can be determined.
  • the determining module 220 may determine the type of information data that is not of interest based on the currently displayed information data and the user's historical browsing information.
  • Step 530 Based on the information refresh request and the uninteresting information data types, reverse processing the uninteresting information data types to obtain information types that may be of interest to the user. In some embodiments, this step 530 may be performed by the determining module 220.
  • At least one data type among the types of the reverse threshold may be selected as the type of information that the user may be interested in.
  • the data type can be obtained, and then at least one data type other than the information data type that is not of interest (ie, the type of reverse threshold) among the data types can be selected as the information type that the user may be interested in.
  • the type of reverse threshold and the type of uninteresting information data may form a complete set of data types.
  • the reverse processing of the information data type that is not of interest is similar to the reverse processing of the currently displayed information data type. For more information about the reverse processing, please refer to step 420, which will not be repeated here.
  • the determining module 220 may perform reverse processing on the uninteresting information data types based on the information refresh request and the uninteresting information data types to obtain information types that may be of interest to the user.
  • Step 540 Determine information that may be of interest to the user based on the type of information that may be of interest to the user. In some embodiments, this step 540 may be performed by the display module 230.
  • step 540 please refer to the detailed description of step 430, which is not repeated here.
  • performing the reverse processing in step 530 to obtain the information types that the user may be interested in can also be replaced with: selecting at least one data type from the complete set of data types except for the information data types that are not of interest, and use it as the user's possible feelings. The type of information of interest.
  • Fig. 6 is an exemplary flowchart of a machine learning model training method according to some embodiments of the present application.
  • the machine learning model training method 600 may be executed by the machine learning model training module 240.
  • Step 610 Obtain training samples.
  • the training samples may include a certain amount of historically displayed information data and historically displayed information data types, and the training samples are used to train the machine learning model.
  • the historically displayed information data may include historically displayed information data on the user terminal.
  • the information data type displayed in the history may be a data type corresponding to the information data displayed in the history.
  • step 610 may also include preprocessing the acquired training samples to make them meet training requirements.
  • the preprocessing may include format conversion, normalization, identification, and so on.
  • the machine learning model training module 240 may also mark the acquired training samples.
  • the historically displayed information data type can be marked as the reference information data type. For example, in a certain training sample, it is known that the type of information data displayed in the history is "sports", then the training sample can be marked as "sports".
  • the information data type of the training sample can be obtained through a questionnaire survey. For example, a certain amount of historically displayed information data can be selected in advance, and the corresponding information data type can be obtained through manual questionnaire survey.
  • the labeling process of training samples can be performed manually or by computer programs.
  • the training sample may also be divided into a training set and a verification set.
  • the training samples can be divided according to a certain ratio.
  • the division ratio can be 80% for the training set and 20% for the validation set.
  • the machine learning model training module 240 may access the information and/or data stored in the storage device 130 through the network 150 to obtain training samples. In some embodiments, the machine learning model training module 240 may obtain training samples through an interface. In some embodiments, the machine learning model training module 240 may also obtain training samples in other ways, which is not limited in this application.
  • Step 620 based on the training samples, train an initial model to obtain a machine learning model.
  • the initial model may include a classification model.
  • a classification model For example, decision trees, Bayesian classification, random forests, support vector machines, neural networks and other models.
  • the decision tree model may include, but is not limited to, Classification and Regression Tree (CART), Iterative Dichotomiser 3 (ID3), C4.5 algorithm, Random Forest (Random Forest), Chisquared Automatic Interaction Detection (CHAID), Multivariate Adaptive Regression Splines (MARS), Gradient Boosting Machine (GBM), etc., or any combination thereof.
  • the training of the initial model may include: 1) Dividing the sample data into a training set, a verification set, and a test set.
  • the sample data can be divided randomly according to a certain proportion.
  • the training set accounts for 85%
  • the verification set accounts for 10%
  • the test set accounts for 5%.
  • the training process meets certain conditions, for example, the number of training reaches the upper limit of the predefined number of iterations, or the value of the loss function is less than the predetermined value, so The model training process can be stopped, and the trained machine learning model can be obtained.
  • the comparison result may include a match and a mismatch between the output result and the tag identification.
  • the matching may mean that the difference between the output information data type and the reference information data type mark is within 2%, otherwise it is regarded as a mismatch.
  • step 5 If the comparison result meets the verification requirements (you can set it according to actual needs, for example, you can set the output information data type after training for more than 95% of the sample data in the verification set to match the reference information data type mark), then go to step 5). Test, otherwise it is deemed as substandard (for example, the accuracy of the output information data type is low).
  • the parameters of the trained model can be adjusted, and based on the adjusted model, step 2) is executed again. 5) Input the sample data in the test set into the trained machine learning model for calculation, and obtain the output result.
  • step 6) Compare the output result of the sample data in the test set in step 5) with the identification of the corresponding sample data to determine whether the training result meets the requirements (can be set according to actual needs, for example, more than 98% of the sample data in the test set can be set).
  • the output result obtained by the trained model matches the corresponding label identification, the training result is considered to meet the requirements, and the training result is denied that the training result does not meet the requirements). If the training result does not meet the requirements, re-prepare the sample data or re-divide the training set, validation set, and test set, and continue training until the model test is passed.
  • the training set, the verification set, and the test set can be divided according to other methods or proportions, some of these steps can be omitted, and other steps can be added.
  • the currently displayed information data and the currently displayed information data type may also be used as training sample data to train the machine learning model, and iteratively update the machine learning model. For example, after the currently displayed information data type is determined, the currently displayed information data type and the currently displayed information data type are used as training samples to update the machine learning model. When the user uses the current application or a third-party application again, determine The accuracy of the currently displayed information data type will be improved.
  • the machine learning model training module 240 may access the information and/or data stored in the storage device 130 through the network 150 to train the initial model to obtain the machine learning model based on the training samples.
  • step 620 of the process 600 can be further subdivided into steps such as step 620 model training, step 630 model verification, and step 640 model testing.
  • the division ratio may be 90% for the training set, 7% for the verification set, and 3% for the test set.
  • the possible beneficial effects brought by the embodiments of the present application include but are not limited to: (1) The information that may be of interest to the user is determined through the currently displayed information data, and data content that is different from the currently displayed information data type can be recommended to the user, which broadens The user's browsing field of vision; (2) By setting a simple information refresh request input method on the user terminal, the interactive operation between the user and the user terminal interface can be simplified, and the user experience can be improved. It should be noted that different embodiments may have different beneficial effects. In different embodiments, the possible beneficial effects may be any one or a combination of the above, or any other beneficial effects that may be obtained.
  • this application uses specific words to describe the embodiments of the application.
  • “one embodiment”, “an embodiment”, and/or “some embodiments” mean a certain feature, structure, or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that “one embodiment” or “one embodiment” or “an alternative embodiment” mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. .
  • some features, structures, or characteristics in one or more embodiments of the present application can be appropriately combined.
  • the computer storage medium may contain a propagated data signal containing a computer program code, for example on a baseband or as part of a carrier wave.
  • the propagated signal may have multiple manifestations, including electromagnetic forms, optical forms, etc., or a suitable combination.
  • the computer storage medium may be any computer readable medium other than the computer readable storage medium, and the medium may be connected to an instruction execution system, device, or device to realize communication, propagation, or transmission of the program for use.
  • the program code located on the computer storage medium can be transmitted through any suitable medium, including radio, cable, fiber optic cable, RF, or similar medium, or any combination of the above medium.
  • the computer program codes required for the operation of each part of this application can be written in any one or more programming languages, including object-oriented programming languages such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python Etc., conventional programming languages such as C language, Visual Basic, Fortran 2003, Perl, COBOL 2002, PHP, ABAP, dynamic programming languages such as Python, Ruby and Groovy, or other programming languages.
  • the program code can be run entirely on the user's computer, or run as an independent software package on the user's computer, or partly run on the user's computer and partly run on a remote computer, or run entirely on the remote computer or server.
  • the remote computer can be connected to the user's computer through any form of network, such as a local area network (LAN) or a wide area network (WAN), or connected to an external computer (for example, via the Internet), or in a cloud computing environment, or as a service Use software as a service (SaaS).
  • LAN local area network
  • WAN wide area network
  • SaaS service Use software as a service
  • numbers describing the number of ingredients and attributes are used. It should be understood that such numbers used in the description of the embodiments use the modifier "about”, “approximately” or “substantially” in some examples. Retouch. Unless otherwise stated, “approximately”, “approximately” or “substantially” indicates that the number is allowed to vary by ⁇ 20%.
  • the numerical parameters used in the specification and claims are approximate values, and the approximate values can be changed according to the required characteristics of individual embodiments. In some embodiments, the numerical parameter should consider the prescribed effective digits and adopt the method of general digit retention. Although the numerical ranges and parameters used to confirm the breadth of the ranges in some embodiments of the present application are approximate values, in specific embodiments, the setting of such numerical values is as accurate as possible within the feasible range.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Business, Economics & Management (AREA)
  • Human Computer Interaction (AREA)
  • Development Economics (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

本申请实施例公开了一种当前页面信息刷新方法和系统。该方法可以包括：响应于接收到用户的信息刷新请求，获取当前显示的信息数据；基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息；向用户展示所述可能感兴趣的信息。本申请通过当前显示的信息数据确定用户可能感兴趣的信息，并向用户推荐与当前显示的信息数据类型不同的数据内容，拓宽了用户的浏览视野，提高了用户的体验度。

Description

一种当前页面信息刷新方法和系统 技术领域
本申请涉及内容推荐领域，特别涉及一种当前页面信息刷新方法和系统。
背景技术
随着互联网技术的迅速发展，用户打开视频、资讯、音乐、购物等网站时，都会收到该网站推荐的内容。对于网站推荐的内容，用户可能并不感兴趣，而想要了解更多不同类型的新信息。因此，有必要提供一种当前页面信息刷新方法和系统。
发明内容
本申请实施例之一提供一种当前页面信息刷新方法。所述方法包括响应于接收到用户的信息刷新请求，获取当前显示的信息数据；基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息；向用户展示所述可能感兴趣的信息。
在一些实施例中，所述用户的信息刷新请求的方式包括手势密码、连续点击操作、点击按键操作、停顿触摸屏幕操作、连续晃动操作、语音输入操作、人脸识别、表情识别或虹膜识别。
在一些实施例中，所述信息刷新请求为信息反向刷新请求。
在一些实施例中，所述信息刷新请求包括反向阈值，所述反向阈值用于表征刷新后的信息与所述当前显示的信息数据的关联度。
在一些实施例中，所述反向阈值的类型至少包括同类反向、不同类反向或异类小众反向。
在一些实施例中，所述基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息包括：基于所述当前显示的信息数据，确定当前显示的信息数据类型；基于所述信息刷新请求和所述当前显示的信息数据类型，对所述当前显示的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型；基于所述用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。
在一些实施例中，所述基于所述当前显示的信息数据，确定当前显示的信息数据类型包括：利用机器学习模型处理所述当前显示的信息数据，确定当前显示的信息数 据类型。
在一些实施例中，所述机器学习模型包括分类模型；所述机器学习模型通过以下方法获得：获取训练样本；所述训练样本包括历史显示的信息数据以及历史显示的信息数据类型；其中，将所述历史显示的信息数据类型标记作为参考信息数据类型；基于所述训练样本，训练初始模型得到所述机器学习模型。
在一些实施例中，所述基于所述当前显示的信息数据类型，对所述当前显示的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型包括：基于所述当前显示的信息数据类型，选取反向阈值的类型中至少一种数据类型，将其作为所述用户可能感兴趣的信息类型。
在一些实施例中，所述基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息包括：响应于接收到用户的信息刷新请求，获取用户的历史浏览信息；基于所述当前显示的信息数据和所述用户的历史浏览信息，确定不感兴趣的信息数据类型；基于所述信息刷新请求和所述不感兴趣的信息数据类型，对所述不感兴趣的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型；基于所述用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。
在一些实施例中，所述向用户展示所述可能感兴趣的信息包括：根据Feed流的方式在用户终端上展示所述可能感兴趣的信息中至少一部分。
本申请实施例之一提供一种当前页面信息刷新系统。所述系统包括：用于存储计算机指令的至少一个存储器；与所述存储器通讯的至少一个处理器，其中当所述至少一个处理器执行所述计算机指令时，所述至少一个处理器使所述系统执行：响应于接收到用户的信息刷新请求，获取当前显示的信息数据；基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息；向用户展示所述可能感兴趣的信息。
在一些实施例中，所述信息刷新请求为信息反向刷新请求。
在一些实施例中，为基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息，所述至少一个处理器使所述系统进一步执行：基于所述当前显示的信息数据，确定当前显示的信息数据类型；基于所述信息刷新请求和所述当前显示的信息数据类型，对所述当前显示的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型；基于所述用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。
在一些实施例中，为基于所述当前显示的信息数据，确定当前显示的信息数据类型，所述至少一个处理器使所述系统进一步执行：利用机器学习模型处理所述当前显 示的信息数据，确定当前显示的信息数据类型。
在一些实施例中，为基于所述当前显示的信息数据类型，对所述当前显示的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型，所述至少一个处理器使所述系统进一步执行：基于所述当前显示的信息数据类型，选取反向阈值的类型中至少一种数据类型，将其作为所述用户可能感兴趣的信息类型。
在一些实施例中，为基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息，所述至少一个处理器使所述系统进一步执行：响应于接收到用户的信息刷新请求，获取用户的历史浏览信息；基于所述当前显示的信息数据和所述用户的历史浏览信息，确定不感兴趣的信息数据类型；基于所述信息刷新请求和所述不感兴趣的信息数据类型，对所述不感兴趣的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型；基于所述用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。
在一些实施例中，为向用户展示所述可能感兴趣的信息，所述至少一个处理器使所述系统进一步执行：根据Feed流的方式在用户终端上展示所述可能感兴趣的信息中至少一部分。
本申请实施例之一提供一种当前页面信息刷新系统。所述系统包括：获取模块，用于响应于接收到用户的信息刷新请求，获取当前显示的信息数据；确定模块，用于基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息；展示模块，用于向用户展示所述可能感兴趣的信息。
根据本申请的另一方面，涉及一种计算机可读存储介质。所述存储介质存储计算机指令，当计算机读取所述存储介质中的所述计算机指令后，所述计算机执行如本申请任一实施例所述的方法。
附图说明
本申请将以示例性实施例的方式进一步说明，这些示例性实施例将通过附图进行详细描述。这些实施例并非限制性的，在这些实施例中，相同的编号表示相同的结构，其中：
图1是根据本申请一些实施例所示的当前页面信息刷新系统的应用场景示意图；
图2是根据本申请一些实施例所示的当前页面信息刷新系统的模块图；
图3是根据本申请一些实施例所示的当前页面信息刷新方法的示例性流程图；
图4是根据本申请一些实施例所示的确定用户可能感兴趣信息的方法的示例性 流程图；
图5是根据本申请又一实施例所示的确定用户可能感兴趣信息的方法的示例性流程图；以及
图6是根据本申请一些实施例所示的机器学习模型训练方法的示例性流程图。
具体实施方式
为了更清楚地说明本申请实施例的技术方案，下面将对实施例描述中所需要使用的附图作简单的介绍。显而易见地，下面描述中的附图仅仅是本申请的一些示例或实施例，对于本领域的普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图将本申请应用于其它类似情景。除非从语言环境中显而易见或另做说明，图中相同标号代表相同结构或操作。
应当理解，本文使用的“系统”、“装置”、“单元”和/或“模组”是用于区分不同级别的不同组件、元件、部件、部分或装配的一种方法。然而，如果其他词语可实现相同的目的，则可通过其他表达来替换所述词语。
如本申请和权利要求书中所示，除非上下文明确提示例外情形，“一”、“一个”、“一种”和/或“该”等词并非特指单数，也可包括复数。一般说来，术语“包括”与“包含”仅提示包括已明确标识的步骤和元素，而这些步骤和元素不构成一个排它性的罗列，方法或者设备也可能包含其它的步骤或元素。
本申请中使用了流程图用来说明根据本申请的实施例的系统所执行的操作。应当理解的是，前面或后面操作不一定按照顺序来精确地执行。相反，可以按照倒序或同时处理各个步骤。同时，也可以将其他操作添加到这些过程中，或从这些过程移除某一步或数步操作。
图1是根据本申请一些实施例所示的当前页面信息刷新系统的应用场景示意图。
当前页面信息刷新系统100可以根据用户的需求刷新当前页面的显示信息，便于用户了解更多其他类型的信息。当前页面信息刷新系统100可以是用于互联网或者其它网络的服务平台。例如，当前页面信息刷新系统100可以是为用户提供资讯信息或视频信息的线上服务平台。在一些实施例中，当前页面信息刷新系统100可以应用于网购服务，例如购买衣物、书、生活用品等。在一些实施例中，当前页面信息刷新系统100还可以应用于出行(如，旅游)服务领域。当前页面信息刷新系统100可以包括但不限于服务器110、用户终端120、存储设备130、信息源140和网络150。
在一些实施例中，服务器110可以用于处理与服务请求有关的信息和/或数据，例如，用于处理用户的信息刷新请求。具体的，服务器110可以从用户终端120接收信息刷新请求，并处理该信息刷新请求以向用户终端120发送用户可能感兴趣的信息。在一些实施例中，服务器110可以是单个的服务器或者服务器群组。所述服务器群可以是集中式的或分布式的(例如，服务器110可以是分布式的系统)。在一些实施例中，服务器110可以是本地的或远程的。例如，服务器110可以通过网络150访问存储在存储设备130、用户终端120中的信息和/或数据。再例如，服务器110可以直接连接到存储设备130、用户终端120以访问存储的信息和/或数据。在一些实施例中，服务器110可以在一个云平台上实现。仅作为示例，所述云平台可以包括私有云、公共云、混合云、社区云、分布云、云之间、多重云等或上述举例的任意组合。
在一些实施例中，服务器110可以包括处理引擎112。处理引擎112可处理与当前页面信息刷新请求有关的数据和/或信息以执行一个或多个本申请中描述的功能。例如，处理引擎112可以接收用户终端120发送的信息刷新请求，并获取当前显示的信息数据，然后可以基于信息刷新请求和当前显示的信息数据确定用户可能感兴趣的信息，最后可以向用户展示可能感兴趣的信息。在一些实施例中，处理引擎112可以包括一个或以上处理引擎(例如，单芯片处理引擎或多芯片处理器)。仅作为示例，处理引擎112可以包括中央处理单元(CPU)、专用集成电路(ASIC)、专用指令集处理器(ASIP)、图像处理单元(GPU)、物理运算处理单元(PPU)、数字信号处理器(DSP)、现场可编程门阵列(FPGA)、可编程逻辑装置(PLD)、控制器、微控制器单元、精简指令集计算机(RISC)、微处理器等或以上任意组合。
在一些实施例中，用户终端120可以是与信息刷新请求直接相关的个人、工具或其他实体。用户可以是信息刷新请求者。在本申请中，“用户”、“用户终端”可以互换使用。在一些实施例中，用户终端120可以包括移动设备120-1、平板电脑120-2、笔记本电脑120-3、以及台式电脑120-4等或其任意组合。在一些实施例中，移动设备120-1可以包括智能家居设备、可穿戴设备、智能移动设备、虚拟现实设备、增强现实设备等或其任意组合。在一些实施例中，智能家居设备可以包括智能照明设备、智能电器控制设备、智能监控设备、智能电视、智能摄像机、对讲机等或其任意组合。在一些实施例中，可穿戴设备可以包括智能手镯、智能鞋袜、智能眼镜、智能头盔、智能手表、智能穿着、智能背包、智能配件等或其任意组合。在一些实施例中，智能移动设备可以包括智能电话、个人数字助理(PDA)、游戏设备、导航设备、销售点(POS)等或其任 意组合。在一些实施例中，虚拟现实设备和/或增强现实设备可以包括虚拟现实头盔、虚拟现实眼镜、虚拟现实眼罩、增强型虚拟现实头盔、增强现实眼镜、增强现实眼罩等或其任意组合。例如，虚拟现实设备和/或增强现实设备可以包括Google Glass、Oculus Rift、HoloLens或Gear VR等。
存储设备130可以存储与用户的信息刷新请求相关的数据和/或指令。在一些实施例中，存储设备130可以存储当前显示的信息数据。在一些实施例中，存储设备130可以存储用户的历史浏览信息。在一些实施例中，存储设备130可以存储服务器110用于执行或使用来完成本申请中描述的示例性方法的数据和/或指令。在一些实施例中，存储设备130可以包括大容量存储器、可移动存储器、易失性读写存储器、只读存储器(ROM)等或其任意组合。示例性的大容量储存器可以包括磁盘、光盘、固态磁盘等。示例性可移动存储器可以包括闪存驱动器、软盘、光盘、存储卡、压缩盘、磁带等。示例性的挥发性只读存储器可以包括随机存取内存(RAM)。示例性的RAM可包括动态RAM(DRAM)、双倍速率同步动态RAM(DDR SDRAM)、静态RAM(SRAM)、闸流体RAM(T-RAM)和零电容RAM(Z-RAM)等。示例性的ROM可以包括掩模ROM(MROM)、可编程ROM(PROM)、可擦除可编程ROM(PEROM)、电子可擦除可编程ROM(EEPROM)、光盘ROM(CD-ROM)和数字通用磁盘ROM等。在一些实施例中，所述存储设备130可以在云平台上实现。仅作为示例，所述云平台可以包括私有云、公共云、混合云、社区云、分布云、内部云、多层云等或其任意组合。
在一些实施例中，存储设备130可以连接到网络150以与当前页面信息刷新系统100中的一个或以上组件(例如，服务器110、用户终端120)通信。当前页面信息刷新系统100中的一个或以上组件可以通过网络150访问存储设备130中存储的数据或指令。在一些实施例中，存储设备130可以与当前页面信息刷新系统100中的一个或以上组件(例如，服务器110、用户终端120)直接连接或通信。在一些实施例中，存储设备130可以是服务器110的一部分。
网络150可以促进信息和/或数据的交换。在一些实施例中，当前页面信息刷新系统100中的一个或以上组件(例如，服务器110、用户终端120和存储设备130)可以通过网络150向/从当前页面信息刷新系统100中的其他组件发送和/或接收信息和/或数据。例如，服务器110可以通过网络150从用户终端120获得/获取服务请求(如，信息刷新请求)。在一些实施例中，网络150可以为任意形式的有线或无线网络或其任意组合。仅作为示例，网络150可以包括缆线网络、有线网络、光纤网络、远程通信网 络、内部网络、互联网、局域网(LAN)、广域网(WAN)、无线局域网(WLAN)、城域网(MAN)、广域网(WAN)、公共交换电话网络(PSTN)、蓝牙网络、紫蜂网络、近场通讯(NFC)网络、全球移动通讯系统(GSM)网络、码分多址(CDMA)网络、时分多址(TDMA)网络、通用分组无线服务(GPRS)网络、增强数据速率GSM演进(EDGE)网络、宽带码分多址接入(WCDMA)网络、高速下行分组接入(HSDPA)网络、长期演进(LTE)网络、用户数据报协议(UDP)网络、传输控制协议/互联网协议(TCP/IP)网络、短讯息服务(SMS)网络、无线应用协议(WAP)网络、超宽带(UWB)网络、红外线等或其任意组合。在一些实施例中，当前页面信息刷新系统100可以包括一个或以上网络接入点。例如，当前页面信息刷新系统100可以包括有线或无线网络接入点，例如基站和/或无线接入点150-1、150-2、…，当前页面信息刷新系统100的一个或以上组件可以通过其连接到网络150以交换数据和/或信息。
在一些实施例中，信息源140可以泛指除用户终端120提供的信息以外的所有信息源。信息源140可以包括但不限于购物网站、门户网站、证券交易所、微博、博客、个人网站、图书馆等可以提供信息的各种信息源。信息源140可以在单个中央服务器、通过通信链路连接的多个服务器或多个个人设备中实现。当信息源140在多个个人设备中实现时，个人设备可以生成内容(例如，被称为“用户生成内容”)，例如向云端服务器上传文字、声音、图像、视频等，从而使云端服务器连同与其连接的多个个人设备一起组成信息源。在一些实施例中，信息源140可以提供一些相关信息，包括但不限于证券要闻、大盘分析、社会热点、财经观点、行情分析、行业研报、公司公告、投资机会、基金、大宗商品、港股、美股等中的一种或几种的组合。
图2是根据本申请一些实施例所示的当前页面信息刷新系统的模块图。
如图2所示，该当前页面信息刷新系统可以包括获取模块210、确定模块220、展示模块230和机器学习模型训练模块240。
获取模块210可以用于响应于接收到用户的信息刷新请求，获取当前显示的信息数据。在一些实施例中，用户的信息刷新请求的方式可以包括手势密码、连续点击操作、点击按键操作、停顿触摸屏幕操作、连续晃动操作、语音输入操作、人脸识别、表情识别或虹膜识别。在一些实施例中，信息刷新请求可以为信息反向刷新请求。在一些实施例中，信息刷新请求可以包括反向阈值，反向阈值用于表征刷新后的信息与当前显示的信息数据的关联度。在一些实施例中，反向阈值的类型至少可以包括同类 反向、不同类反向或异类小众反向。关于获取当前显示的信息数据的详细描述可以参见图3，在此不再赘述。
确定模块220可以用于基于信息刷新请求和当前显示的信息数据，确定用户可能感兴趣的信息。具体的，可以基于当前显示的信息数据，确定当前显示的信息数据类型，然后可以基于信息刷新请求和当前显示的信息数据类型，对当前显示的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型，从而可以基于用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。关于确定用户可能感兴趣的信息的详细描述可以参见图4的内容，在此不再赘述。
在一些实施例中，还可以基于信息刷新请求、当前显示的信息数据和用户的历史浏览信息，确定用户可能感兴趣的信息。具体的，可以响应于接收到用户的信息刷新请求，获取用户的历史浏览信息，还可以基于当前显示的信息数据和用户的历史浏览信息，确定不感兴趣的信息数据类型，然后可以基于信息刷新请求和不感兴趣的信息数据类型，对不感兴趣的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型，从而可以基于用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。关于确定用户可能感兴趣的信息的详细描述可以参见图5的内容，在此不再赘述。
展示模块230可以用于向用户展示可能感兴趣的信息。具体的，可以根据Feed流的方式在用户终端上展示可能感兴趣的信息中至少一部分。关于向用户展示可能感兴趣的信息的详细描述可以参见图3，在此不再赘述。
机器学习模型训练模块240可以用于训练初始模型得到机器学习模型。具体的，可以获取训练样本，训练样本可以包括历史显示的信息数据以及历史显示的信息数据类型；可以将历史显示的信息数据类型标记作为参考信息数据类型；然后可以基于训练样本，训练初始模型得到机器学习模型。关于训练初始模型得到机器学习模型的详细描述可以参见图6，在此不再赘述。
应当理解，图2所示的系统及其模块可以利用各种方式来实现。例如，在一些实施例中，系统及其模块可以通过硬件、软件或者软件和硬件的结合来实现。其中，硬件部分可以利用专用逻辑来实现；软件部分则可以存储在存储器中，由适当的指令执行系统，例如微处理器或者专用设计硬件来执行。本领域技术人员可以理解上述的方法和系统可以使用计算机可执行指令和/或包含在处理器控制代码中来实现，例如在诸如磁盘、CD或DVD-ROM的载体介质、诸如只读存储器(固件)的可编程的存储器或者诸如光学或电子信号载体的数据载体上提供了这样的代码。本申请的系统及其模块不仅可以有 诸如超大规模集成电路或门阵列、诸如逻辑芯片、晶体管等的半导体、或者诸如现场可编程门阵列、可编程逻辑设备等的可编程硬件设备的硬件电路实现，也可以用例如由各种类型的处理器所执行的软件实现，还可以由上述硬件电路和软件的结合(例如，固件)来实现。
需要注意的是，以上对于当前页面信息刷新系统及其模块的描述，仅为描述方便，并不能把本申请限制在所举实施例范围之内。可以理解，对于本领域的技术人员来说，在了解该系统的原理后，可能在不背离这一原理的情况下，对各个模块进行任意组合，或者构成子系统与其他模块连接。例如，在一些实施例中，获取模块210、确定模块220、展示模块230和机器学习模型训练模块240可以是一个系统中的不同模块，也可以是一个模块实现上述的两个或两个以上模块的功能。又例如，确定模块220以及机器学习模型训练模块240可以是两个模块，也可以是一个模块同时具有确定用户可能感兴趣的信息和模型训练功能。再例如，各个模块可以共用一个存储模块，各个模块也可以分别具有各自的存储模块。诸如此类的变形，均在本申请的保护范围之内。
图3是根据本申请一些实施例所示的当前页面信息刷新方法的示例性流程图。
步骤310，响应于接收到用户的信息刷新请求，获取当前显示的信息数据。在一些实施例中，步骤310可以由获取模块210实现。
在一些实施例中，用户可以包括使用或浏览当前应用(或称第一方)界面的用户，也可以为通过当前应用界面来使用或浏览第三方应用的用户。第三方应用是相对于当前应用而言的其他应用。例如，腾讯视频网站为当前应用，在腾讯视频网站页面出现了京东的购物广告，则京东为第三方应用，用户通过点击腾讯视频网站页面中京东的购物广告可以跳转至京东的购物页面。
在一些实施例中，信息刷新请求可以包括用户请求重新推荐当前浏览资源的指令。例如，用户在抖音短视频APP界面浏览了一段时间后，希望抖音短视频APP推荐一些新的视频，可以通过下拉页面来发出刷新请求。在一些实施例中，用户发送信息刷新请求的方式可以包括手势密码、连续点击操作、点击按键操作、停顿触摸屏幕操作、连续晃动操作、语音输入操作、人脸识别、表情识别或虹膜识别。手势密码可以包括在屏幕上画×、画√或画Z型等。手势密码还可以包括点击屏幕上密码输入控件，在弹出的密码输入框中输入数字密码或特定字符。连续点击操作可以包括在较短时间(如，3秒)内完成多次(如，2次、3次或4次)点击屏幕操作。点击按键操作可以为在当前应用界面上设置功能按键，通过点击该功能按键即可完成信息刷新请求。停顿触摸屏幕操作 可以包括在一段时间(如，3秒、4秒或5秒)内连续触摸屏幕。连续晃动操作可以是基于一定强度持续(如，3秒、4秒或5秒)使用户终端设备处于运动状态(如，强烈摇晃手机5秒)。语音输入操作可以包括用户的语音片段中包含特定的语音内容文本或指令，例如包含“信息刷新”的内容文本或指令。人脸识别可以是通过用户终端设备上的摄像装置拍摄人脸的图像或视频流，并基于脸部特征信息进行身份识别，以验证是否发送信息刷新请求。例如，人脸验证通过，则发送信息刷新请求的指令；反之，则不发送信息刷新请求的指令。表情识别可以是通过用户终端设备上的摄像装置拍摄人脸的图像或视频流，并分离出特定的表情状态，以验证是否发送信息刷新请求。例如，表情状态为“微笑”，则验证通过，发送信息刷新请求的指令；表情状态为其他状态(如，皱眉、哭泣、生气)，则验证不通过，不发送信息刷新请求的指令。虹膜识别可以是通过用户终端设备上的摄像装置拍摄用户的虹膜，以验证是否发送信息刷新请求。例如，虹膜验证通过，则发送信息刷新请求的指令；反之，则不发送信息刷新请求的指令。
在一些实施例中，当前显示的信息数据可以为用户终端上当前显示的第一方应用的信息数据，也可以为用户终端上当前显示的第三方应用的信息数据。
信息刷新请求可以为对当前用户终端页面显示的信息进行刷新的请求。在一些实施例中，信息刷新请求可以为信息同向刷新请求。信息同向刷新请求可以包括用户请求重新推荐与当前用户终端页面显示的信息类型相同的其他信息的指令。在一些实施例中，信息刷新请求还可以为信息反向刷新请求。信息反向刷新请求可以包括用户请求重新推荐与当前用户终端页面显示的信息类型不同的其他信息的指令。在一些实施例中，用户发送信息同向刷新请求或信息反向刷新请求的方式可以为用户提前设置，也可以为系统默认设置。
在一些实施例中，信息刷新请求可以包括反向阈值，反向阈值可以用于表征刷新后的信息与当前显示的信息数据的关联度，即反向的级别或程度。在一些实施例中，反向阈值可以包括0-10级，也可以用其它描述级别强度的符号(如，A-K级)代替，本说明对此不作限制。具体的，若设置反向阈值较高时，可以使推荐内容的反向程度较高，则刷新后的信息与当前显示的信息数据的关联度较低；若设置反向阈值较低时，可以使推荐内容部分反向，则刷新后的信息与当前显示的信息数据的关联度较高。例如，某用户使用汽车类APP，用户长期查看或关注的车价格区间在10-20万之间；当用户设置反向阈值为0级时，该用户发送信息反向刷新请求后，该汽车类APP推荐车型价格都在10-20万之间；当用户设置反向阈值为10级时，该用户发送信息反向刷新请求后，该 汽车类APP推荐的是除10-20万价格之外的车型；当用户设置反向阈值为1-9级时，该用户发送信息反向刷新请求后，该汽车类APP推荐一部分10-20万的车型，也推荐一部分其它价位区间的车型。
在一些实施例中，反向阈值的类型至少可以包括同类反向、不同类反向或异类小众反向。同类反向可以是在同属于一个大类里的小类反向，例如，大类里的小类别为体育运动中的篮球，同类反向可以为体育运动中的乒乓球、羽毛球、排球等最新资讯。不同类反向可以为不属于同一大类的反向，例如，某一类别为体育运动，不同类反向可以为娱乐、军事、财经、旅游、历史等。异类小众反向可以是关注度低且关注群体极少的类别(包括大类和/或小类)，例如，天文、数学猜想、宗教研究、冰壶项目等。在一些实施例中，可以设置反向阈值的类型为同类反向、不同类反向和异类小众反向中至少一个。在一些实施例中，反向阈值的设置方式可以包括雷达图、百分比、等级等。在一些实施例中，雷达图可以包括一个或多个反向阈值的类型，通过设置不同反向阈值的类型所占比例，可以调整反向刷新的反向程度。例如，雷达图上包括4个类别：同类不反向、同类反向、不同类反向和异类小众反向，同类不反向可以为与当前显示的信息内容类型相同，通过设置上述4个类别所占比例为：0％、10％、50％、40％，推荐内容中将不会包含与当前显示的信息数据类型相同的内容，但会包括10％同类反向的内容、50％不同类反向和40％异类小众反向。又例如，还可以设置同类不反向所占比例为0％、同类反向所占比例为0％、不同类反向所占比例为0％和异类小众反向所占比例为100％，推荐内容中将会全部推荐异类小众反向的内容。可选地，雷达图上也可以仅包括3个类别：同类反向、不同类反向和异类小众反向，雷达图还可以为其他形式，本申请对此不作限制。通过设置不同类别所占比例，可以使推荐内容进行不同类型和不同程度的反向。又例如，通过设置不同的百分比或等级，可以调整反向刷新的程度，该百分比或等级对应不同反向阈值类型中每一种类别所占的比例。仅作为示例，设置反向阈值的百分比为80％(或等级H)时，推荐内容中可以包含50％不同类反向和50％异类小众反向。该百分比或等级可以为系统预先设定或后台服务器通过相应算法来确定，本申请对此不作限定。在一些实施例中，反向阈值的设置可以是在用户发送信息刷新请求后，应用页面弹出的功能设置框中进行设置；也可以为用户发送信息刷新请求前，在应用页面的功能控件上进行设置；还可以为系统默认设置。
在一些实施例中，用户的信息反向刷新请求可以为用户对当前应用的推荐内容的信息反向刷新请求。
在另一些实施例中，用户的信息反向刷新请求还可以为用户对第三方应用的推荐内容的信息反向刷新请求。第三方应用可以获取用户在当前应用上对第三方应用的信息反向刷新请求。在一些实施例中，第三方应用获取用户对第三方应用的信息反向刷新请求可以是基于第三方应用的插件进行的。第三方应用的插件只能在其程序规定的第一方应用平台(可能同时支持多个平台)上运行，而不能脱离指定的第一方应用平台单独运行。例如，淘宝需要在今日头条、抖音、微博等应用平台投放的广告，第三方应用的淘宝插件可以支持通过今日头条、抖音以及微博等第一方应用平台进入淘宝以进行购物。第三方应用的插件是第三方应用同当前应用进行数据交互的工具，例如，第三方应用的插件可以包括获取当前应用中用户对第三方应用的信息反向刷新请求、获取当前显示的第三方应用的信息数据、确定用户可能感兴趣的第三方应用的信息、向用户展示可能感兴趣的第三方应用的信息等用途。
在一些实施例中，获取模块210可以通过用户在用户终端上的操作(如：手势密码、连续点击操作、点击按键操作、停顿触摸屏幕操作、连续晃动操作、语音输入操作、人脸识别、表情识别或虹膜识别)来获取用户的信息刷新请求。通过设置较简单的信息刷新请求操作，可以使用户在使用应用时更加方便、快捷，给用户带来更好的使用体验。
步骤320，基于信息刷新请求和当前显示的信息数据，确定用户可能感兴趣的信息。在一些实施例中，步骤320可以由确定模块220实现。
在一些实施例中，用户可能感兴趣的信息可以包括用户期望浏览的信息内容。在一些实施例中，用户可能感兴趣的信息可以为用户期望浏览的第一方应用的信息内容，也可以为用户期望浏览的第三方应用的信息内容。
在一些实施例中，可以基于信息刷新请求和当前显示的信息数据，确定用户可能感兴趣的信息。具体的，可以基于当前显示的信息数据，确定当前显示的信息数据类型，然后可以基于信息刷新请求和当前显示的信息数据类型，对当前显示的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型，从而基于用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。关于确定用户可能感兴趣的信息的详细描述可以参见图4，在此不再赘述。
在一些实施例中，还可以基于信息刷新请求、当前显示的信息数据和用户的历史浏览信息，确定用户可能感兴趣的信息。具体的，可以响应于接收到用户的信息刷新请求，获取用户的历史浏览信息，然后可以基于当前显示的信息数据和用户的历史浏览 信息，确定不感兴趣的信息数据类型，从而可以基于信息刷新请求和不感兴趣的信息数据类型，对不感兴趣的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型，进而可以基于用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。关于确定用户可能感兴趣的信息的详细描述可以参见图5，在此不再赘述。
在一些实施例中，若用户的信息刷新请求是对第三方应用的信息刷新请求时，第三方应用插件可以根据当前应用的当前显示的信息数据，确定用户可能感兴趣的信息。具体的，第三方应用插件可以将当前应用(第一方应用)的数据类型映射到第三方应用的数据类型上，建立当前应用中至少一种数据类型与第三方应用中至少一种数据类型的对应关系；然后可以根据当前应用的当前显示的信息数据和当前应用(第一方)中的数据类型，确定当前应用的当前显示的信息数据类型，然后根据上述数据类型的对应关系确定第三方应用上对应的信息数据类型，对第三方应用上对应的信息数据类型进行反向处理，从而确定第三方应用上用户可能感兴趣的信息类型，进而确定第三方应用上用户可能感兴趣的信息。在第三方应用上确定用户可能感兴趣的信息的内容与图4的描述类似，在此不作赘述。可选的，当前应用中至少一种数据类型与第三方应用中至少一种数据类型的对应关系可以是一一对应、一对多、多对一或者多对多，本申请对此不作限制。当前应用(第一方应用)的数据类型的获取方法可以参见步骤420，在此不作赘述。
在一些实施例中，若用户的信息刷新请求是对第三方应用的信息刷新请求时，第三方应用插件还可以根据当前应用的当前显示的信息数据、当前应用的用户的历史浏览信息，确定用户可能感兴趣的信息。具体的，第三方应用插件可以将当前应用(第一方应用)的数据类型映射到第三方应用的数据类型上，建立当前应用中至少一种数据类型与第三方应用中至少一种数据类型的对应关系；然后可以根据当前应用的当前显示的信息数据、当前应用的用户的历史浏览信息和当前应用(第一方)中的数据类型，确定当前应用的不感兴趣的信息数据类型，然后根据上述数据类型的对应关系确定第三方应用上对应的不感兴趣的信息数据类型，对第三方应用上不感兴趣的信息数据类型进行反向处理，从而确定第三方应用上用户可能感兴趣的信息类型，进而确定第三方应用上用户可能感兴趣的信息。在第三方应用上确定用户可能感兴趣的信息的内容与图5的描述类似，在此不作赘述。
在一些实施例中，确定模块220可以基于信息刷新请求和当前显示的信息数据，确定用户可能感兴趣的信息。
步骤330，向用户展示可能感兴趣的信息。在一些实施例中，步骤330可以由 展示模块230实现。
在一些实施例中，可能感兴趣的信息可以为当前应用或第三方应用中当前时刻前一段时间(如，1周、3天、1天、12小时或1小时)内生成的所有数据信息。在一些实施例中，可以根据Feed流的方式在用户终端页面上展示该可能感兴趣的信息中至少一部分。Feed流是一种呈现内容给用户并持续更新的方式，可以基于信息刷新请求和当前显示的信息数据确定用户可能感兴趣的信息，并刷新页面展示该用户可能感兴趣的信息。在一些实施例中，Feed流可以包括推(Push)、拉(Pull)和推拉结合(Hybrid)三种模式。在一些实施例中，推模式可以是用户生成内容后，服务端将确定的用户可能感兴趣的信息推送给部分其他用户，适用于用户关系数比较均匀，且有上限的应用，如朋友圈；拉模式是发出信息刷新请求的指令时，页面按一定规则呈现更新数据，例如按更新时间、热度、编辑推荐等，适用于用户数较少且日活量低的应用；推拉结合模式可以包括在线推、离线拉(如，微博大V发布动态后，有限推送给同时在线的粉丝，离线粉丝上线后再拉取该动态)；或是定时推、离线拉(如，微博大V发布动态后，以常驻进程的形式推送到粉丝关注)两种。在一些实施例中，Feed流可以通过用户的操作(如，上拉或者下拉操作)不断更新用户可能感兴趣的信息。在一些实施例中，Feed流展示的可能感兴趣的信息可以通过Timeline(时间顺序)排序，例如，按发布的时间顺序排序，先发布的先看到、后发布的排列在最顶端。在一些实施例中，Feed流展示的可能感兴趣的信息可以通过Rank(非时间因子)排序，例如，按热度排序，将用户可能感兴趣的信息按照热度级别排序，最热门的优先推荐。
应当注意的是，上述有关流程300的描述仅仅是为了示例和说明，而不限定本申请的适用范围。对于本领域技术人员来说，在本申请的指导下可以对流程300进行各种修正和改变。然而，这些修正和改变仍在本申请的范围之内。例如，步骤310中获取当前显示的信息数据可以是用户在当前应用平台登录时就已经获取，接收到用户的信息刷新请求时，确定用户可能感兴趣的信息，不必局限于在接收到用户的信息刷新请求后获取。
图4是根据本申请一些实施例所示的确定用户可能感兴趣信息的方法的示例性流程图。如图4所示，该确定用户可能感兴趣信息的方法400可以包括：
步骤410，基于当前显示的信息数据，确定当前显示的信息数据类型。在一些实施例中，该步骤410可以由确定模块220执行。
在一些实施例中，当前显示的信息数据类型可以为当前显示的信息数据的数据 类型。在一些实施例中，数据类型可以为当前页面信息刷新系统数据库(如，存储设备130)中的海量数据的数据标签。该数据标签可以为基于算法对该海量数据进行标记得到。在一些实施例中，数据类型可以包括大类标签和小类标签。在一些实施例中，数据类型可以由两种或多种类型组成。当前显示的信息数据类型可以至少包括一种或多种数据类型。
在一些实施例中，可以利用机器学习模型确定当前显示的信息数据类型。具体的，可以将当前显示的信息数据输入机器学习模型中进行处理，输出当前显示的信息数据类型。
在一些实施例中，机器学习模型可以包括分类模型。例如，决策树、贝叶斯分类、随机森林、支持向量机、神经网络等模型。在一些实施例中，决策树模型可以包括但不限于分类及回归树(Classification And Regression Tree，CART)、迭代二叉树三代(Iterative Dichotomiser 3，ID3)、C4.5算法、随机森林(Random Forest)、卡方自动交互检测(Chisquared Automatic Interaction Detection，CHAID)、多元自适应回归样条(Multivariate Adaptive Regression Splines，MARS)以及梯度推进机(Gradient Boosting Machine，GBM)等或其任意组合。该机器学习模型可以由初始模型训练得到。关于初始模型及其训练方法的更多内容可以参见图6及其描述，在此不作赘述。
在一些实施例中，确定模块220可以通过网络150访问存储于存储设备130中的机器学习模型，并基于当前显示的信息数据，确定当前显示的信息数据类型。
步骤420，基于信息刷新请求和当前显示的信息数据类型，对当前显示的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型。在一些实施例中，该步骤420可以由确定模块220执行。
在一些实施例中，用户可能感兴趣的信息类型可以包括用户期望浏览的信息内容的数据类型。在一些实施例中，可以基于当前显示的信息数据类型，选取反向阈值的类型中至少一种数据类型，将其作为用户可能感兴趣的信息类型。具体的，可以获取数据类型，然后可以选取该数据类型中除当前显示的信息数据类型外(即，反向阈值的类型中)的至少一种数据类型，将其作为用户可能感兴趣的信息类型。
反向阈值的类型至少可以包括同类反向、不同类反向或异类小众反向。上述每一种反向阈值的类型中可以分别包含一种或多种数据类型。在一些实施例中，反向阈值的类型和当前显示的信息数据类型可以组成数据类型的全集合。
在一些实施例中，数据类型可以为基于分类算法对标记有数据标签的数据进行分类得到。分类算法作为一种监督机器学习方法，是对已经标注的数据类型的分类，类别数目固定。在一些实施例中，分类算法可以包括决策树算法、K最近邻(k-Nearest Neighbor，KNN)算法、贝叶斯算法、支持向量机算法等。在一些实施例中，数据类型还可以为基于聚类算法对海量数据进行聚类得到。在一些实施例中，聚类算法也可以作为数据挖掘算法中分类算法的一个预处理步骤。聚类算法是一种无监督的机器学习方法，不需要人工标注和预先训练分类器，类别在聚类过程中自动生成，类别数据不确定。优选地，数据类型为小类标签的数据可以基于聚类算法聚类成大类标签的数据，例如，一部分数据带有足球、欧冠、球星梅西、巴塞罗那足球俱乐部、近期赛事等小类标签，另一部分数据带有篮球、NBA、球星库里、金州勇士队、常规赛等小类标签，两部分数据可以基于相似度或深度学习算法聚类成为大类标签的数据，该大类标签可以为体育。在一些实施例中，聚类算法可以包括K均值(K-Means)聚类算法、均值漂移聚类算法、基于密度的聚类算法(Density-Based Spatial Clustering of Applications with Noise，DBSCAN)、使用高斯混合模型(GMM)的期望最大化(EM)聚类算法、凝聚层次聚类算法、图团体检测(Graph Community Detection)聚类算法。
在一些实施例中，数据类型可以为实时获取，也可以为预先获取，实时获取可以为基于分类算法或聚类算法得到。在一些实施例中，数据类型可以为基于时间、热门度、场景、合作推荐等因素来获取，例如，可以获取近一周更新的数据的数据类型、获取最新/最热门的数据的数据类型、以及获取平台合作的数据生产者产生的数据的数据类型。
下面通过具体示例对得到用户可能感兴趣的信息类型进行阐述：
例如，数据类型可以有大类M＝{m1、m2、m3}、N＝{n1、n2、n3}、O＝{o1、o2、o3}、P＝{p1、p2、p3}，其中m1、…、p1等表示大类中的小类，若当前显示的信息数据类型为m1和n2，用户可能感兴趣的信息类型可以为{m2、m3、n1、n3、O＝{o1、o2、o3}、P＝{p1、p2、p3}}中一种或多种不同的数据类型组合，一种或多种不同的数据类型组合可以组成多个用户可能感兴趣的信息类型的数据库，则用户可能感兴趣的信息类型的数据库中可以包含与用户可能感兴趣的信息类型{m2}、{m3}、…、{m2、m3、n1、n3、O＝{o1、o2、o3}、P＝{p1、p2、p3}}对应的
Figure PCTCN2020082306-appb-000001
Figure PCTCN2020082306-appb-000002
种数据类型的组合。
步骤430，基于所述用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。 在一些实施例中，该步骤430可以由确定模块220执行。
每种信息类型可以有一个或多个对应的信息数据。在一些实施例中，可以基于用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。仍以上述例子为例进行说明，用户可能感兴趣的信息类型的数据库中可以包含与用户可能感兴趣的信息类型{m2}、{m3}、…、{m2、m3、n1、n3、O＝{o1、o2、o3}、P＝{p1、p2、p3}}对应的
Figure PCTCN2020082306-appb-000003
Figure PCTCN2020082306-appb-000004
种数据类型的组合，用户可能感兴趣的信息可以为对应
Figure PCTCN2020082306-appb-000005
种数据类型的组合的信息数据。由于每种信息类型对应的信息数据有一个或多个，因此用户可能感兴趣的信息的数量较大，在此不作列举。
应当注意的是，上述有关流程400的描述仅仅是为了示例和说明，而不限定本申请的适用范围。对于本领域技术人员来说，在本申请的指导下可以对流程400进行各种修正和改变。然而，这些修正和改变仍在本申请的范围之内。例如，流程400中数据类型和用户可能感兴趣的信息类型不限于所列举的数量，还可以为其他数量。又例如，步骤420中进行反向处理得到用户可能感兴趣的信息类型还可以替换为：选取数据类型的全集合中除当前显示的信息数据类型外的至少一种数据类型，将其作为用户可能感兴趣的信息类型。
图5是根据本申请又一实施例所示的确定用户可能感兴趣信息的方法的示例性流程图。如图5所示，该确定用户可能感兴趣信息的方法500可以包括：
步骤510，响应于接收到用户的信息刷新请求，获取用户的历史浏览信息。在一些实施例中，该步骤510可以由获取模块210执行。
在一些实施例中，用户的历史浏览信息可以包括在当前时刻之前(如，一个月、一周、三天或一天)用户浏览当前应用的浏览内容、浏览时间、浏览频次等多维度信息。浏览信息可以包括图片、文本、视频、音频等。在另一些实施例中，用户的历史浏览信息可以包括但不限于：发帖、关注、收藏、评论、点赞的内容和/或时间等信息。
在一些实施例中，用户的历史浏览信息可以包括云端存储的历史浏览数据或本地存储的cookie数据。云端存储的历史浏览数据可以是存储在云端存储器(如，存储设备130)中的用户的历史浏览数据。本地存储的cookie数据可以包括存储在本地客户端(如，用户终端120)中的小文本文件。cookie数据可以包含用户的个人信息和历史浏览数据。
在一些实施例中，获取模块210可以响应于接收到用户的信息刷新请求，在云端存储器或本地客户端中获取用户的历史浏览信息。
通过分析用户的历史浏览信息可以刻画用户画像，还可以为各个运营项目提供数据支持。例如，某购物网站的某店铺中一件商品的点击量远远超过其它商品，可以指导该店铺的商家提高该商品的生产量或库存量。
步骤520，基于当前显示的信息数据和用户的历史浏览信息，确定不感兴趣的信息数据类型。在一些实施例中，该步骤520可以由确定模块220执行。
在一些实施例中，不感兴趣的信息数据类型可以为用户不愿意浏览的信息数据类型。在一些实施例中，可以分别基于当前显示的信息数据和用户的历史浏览信息，确定不感兴趣的信息数据类型。不感兴趣的信息数据类型可以包括当前显示的信息数据类型和用户的历史偏好类型。关于基于当前显示的信息数据，确定当前显示的信息数据类型的更多内容可以参见步骤410，在此不作赘述。在一些实施例中，还可以基于用户的历史浏览信息，确定用户的历史偏好类型。
在一些实施例中，若用户的历史浏览信息为云端存储的历史浏览数据，可以对云端存储的历史浏览数据进行处理得到用户的历史偏好类型。具体的，可以利用机器学习模型处理用户的历史浏览信息，确定用户的历史偏好类型。该机器学习模型可以与步骤410中的机器学习模型相同，在此不作赘述。该机器学习模型的训练方法的更多内容可以参见图6及其描述，在此不作赘述。
在一些实施例中，若用户的历史浏览信息为本地存储的cookie数据，可以提取本地存储的cookie数据中的类别属性作为用户的历史偏好类型。具体的，可以获取本地存储的cookie数据，并可以提取信息刷新请求将该cookie数据的数据类型(或称为类别属性)，即为用户的历史偏好类型。本地存储的cookie数据可以是经加密的hash码，服务器可以对hash码进行解密，然后读取其中的数据类型作为用户的历史偏好类型。例如，服务器解密后的cookie数据为document.cookie＝“userID＝828；userName＝hulk；class＝篮球”，该cookie数据的数据类型为“篮球”，因此用户的历史偏好类型为“篮球”。
根据当前显示的信息数据类型和用户的历史偏好类型，可以确定不感兴趣的信息数据类型。
在一些实施例中，确定模块220可以基于当前显示的信息数据和用户的历史浏览信息，确定不感兴趣的信息数据类型。
步骤530，基于信息刷新请求和不感兴趣的信息数据类型，对所述不感兴趣的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型。在一些实施例中，该步骤530可以由确定模块220执行。
在一些实施例中，可以基于不感兴趣的信息数据类型，选取反向阈值的类型中至少一种数据类型，将其作为用户可能感兴趣的信息类型。具体的，可以获取数据类型，然后可以选取该数据类型中除不感兴趣的信息数据类型外(即，反向阈值的类型中)的至少一种数据类型，将其作为用户可能感兴趣的信息类型。在一些实施例中，反向阈值的类型和不感兴趣的信息数据类型可以组成数据类型的全集合。对不感兴趣的信息数据类型进行反向处理与对当前显示的信息数据类型进行反向处理类似，关于反向处理的更多内容可以参见步骤420，在此不做赘述。
在一些实施例中，确定模块220可以基于信息刷新请求和不感兴趣的信息数据类型，对不感兴趣的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型。
步骤540，基于用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。在一些实施例中，该步骤540可以由展示模块230执行。
关于步骤540的更多内容可以参见步骤430的详细描述，在此不作赘述。
应当注意的是，上述有关流程500的描述仅仅是为了示例和说明，而不限定本申请的适用范围。对于本领域技术人员来说，在本申请的指导下可以对流程500进行各种修正和改变。然而，这些修正和改变仍在本申请的范围之内。例如，步骤530中进行反向处理得到用户可能感兴趣的信息类型还可以替换为：选取数据类型的全集合中除不感兴趣的信息数据类型外的至少一种数据类型，将其作为用户可能感兴趣的信息类型。
图6是根据本申请一些实施例所示的机器学习模型训练方法的示例性流程图。在一些实施例中，该机器学习模型训练方法600可以由机器学习模型训练模块240执行。
步骤610，获取训练样本。
在一些实施例中，训练样本可以包括一定数量的历史显示的信息数据以及历史显示的信息数据类型，训练样本用于训练机器学习模型。历史显示的信息数据可以包括历史上显示在用户终端上的信息数据。历史显示的信息数据类型可以为历史显示的信息数据对应的数据类型。
在一些实施例中，步骤610还可以包括对获取的训练样本进行预处理，使其符合训练的要求。所述预处理可以包括格式转换、归一化、标识等。
在一些实施例中，机器学习模型训练模块240还可以对获取的训练样本进行标记。具体的，可以将历史显示的信息数据类型标记为参考信息数据类型。例如，在某一训练样本中，已知历史显示的信息数据类型为“体育”，则可以将该训练样本标记为“体育”。在一些实施例中，训练样本的信息数据类型可以通过问卷调查的获取。例如，可以提前选取一定数量的历史显示的信息数据，通过人工问卷调查的方式得到对应的信息数据类型。在一些实施例中，训练样本的标记过程可以通过人工或计算机程序进行。
在一些实施例中，还可以将训练样本进行划分，划分为训练集和验证集。具体的，可以对训练样本按一定的比例进行划分。例如，划分比例可以是训练集80％、验证集20％。
在一些实施例中，机器学习模型训练模块240可以通过网络150访问存储于存储设备130中信息和/或资料以获取训练样本。在一些实施例中，机器学习模型训练模块240可以通过接口获取训练样本。在一些实施例中，机器学习模型训练模块240还可以通过其他方式获取训练样本，本申请对此不作限制。
步骤620，基于训练样本，训练初始模型得到机器学习模型。
在一些实施例中，所述初始模型可以包括分类模型。例如，决策树、贝叶斯分类、随机森林、支持向量机、神经网络等模型。在一些实施例中，决策树模型可以包括但不限于分类及回归树(Classification And Regression Tree，CART)、迭代二叉树三代(Iterative Dichotomiser 3，ID3)、C4.5算法、随机森林(Random Forest)、卡方自动交互检测(Chisquared Automatic Interaction Detection，CHAID)、多元自适应回归样条(Multivariate Adaptive Regression Splines，MARS)以及梯度推进机(Gradient Boosting Machine，GBM)等或其任意组合。
在一些实施例中，所述初始模型的训练可以包括：1)将样本数据划分为训练集、验证集、测试集。可以对样本数据按一定的比例随机地进行划分。优选地，训练集占比85％、验证集占比10％、测试集占比5％。2)将训练集中的样本数据输入待训练的初始模型中进行训练，当训练过程满足一定条件时，例如，训练次数达到预定义的迭代次数上限值、或损失函数的值小于预定值，所述模型训练过程可以停止，并获取训练后的机器学习模型。3)将验证集中的样本数据输入上述训练后的机器学习模型中进行计算，获得信息数据类型输出结果。4)对比验证集中样本数据在3)中的输出结果与相应样本数据的标识(如，参考信息数据类型)，获取对比结果。在一些实施例中，所述对比结果可以包括输出结果与标签标识匹配以及不匹配。所述匹配可以指输出的信息数据类型与 参考信息数据类型标记差距在2％以内，否则视为不匹配。若对比结果满足验证要求(可以根据实际需要自行设置，如，可设定验证集中95％以上样本数据经训练后的输出信息数据类型与参考信息数据类型标记匹配)，则转入步骤5)进行测试，否则认定为不达要求(例如，输出信息数据类型的准确率低)。经过训练后的模型的参数可以被调整，并基于调整后的模型，再次执行步骤2)。5)将测试集中的样本数据输入训练后的机器学习模型进行计算，获得输出结果。6)对比测试集中样本数据在步骤5)中的输出结果与相应样本数据的标识，判断训练结果是否达到要求(可根据实际需要自行设定，如，可设定测试集中98％以上的样本数据经训练后的模型得到的输出结果与相应标签标识匹配，则认为训练结果达到要求，否认认为训练结果未达到要求)。如果训练结果未达到要求，则重新准备样本数据或者重新划分训练集、验证集、测试集，进行继续训练，直至通过模型测试。
对上述说明的步骤和实施方法，可以进行各种变化，比如可以按其他方法或比例划分训练集、验证集和测试集，可以忽略其中某些步骤，可以增加其他步骤等。
在一些实施例中，还可以将当前显示的信息数据和当前显示的信息数据类型作为训练样本数据训练机器学习模型，对机器学习模型进行迭代更新。例如，在确定当前显示的信息数据类型之后将当前显示的信息数据类型和当前显示的信息数据类型作为训练样本对机器学习模型进行更新后，当该用户再次使用当前应用或第三方应用时，确定当前显示的信息数据类型的准确性将提升。
在一些实施例中，机器学习模型训练模块240可以通过网络150访问存储于存储设备130中的信息和/或资料，以基于训练样本，训练初始模型得到机器学习模型。
应当注意的是，上述有关流程600的描述仅仅是为了示例和说明，而不限定本申请的适用范围。对于本领域技术人员来说，在本申请的指导下可以对流程600进行各种修正和改变。然而，这些修正和改变仍在本申请的范围之内。例如，流程600的步骤620可以进一步细分为步骤620模型训练、步骤630模型验证、步骤640模型测试等步骤。又例如，划分比例可以是训练集90％、验证集7％、测试集3％。
本申请实施例可能带来的有益效果包括但不限于：(1)通过当前显示的信息数据确定用户可能感兴趣的信息，可以向用户推荐与当前显示的信息数据类型不同的数据内容，拓宽了用户的浏览视野；(2)通过在用户终端设置简单的信息刷新请求输入方式，可以简化用户与用户终端界面的交互操作，提高了用户的体验度。需要说明的是，不同实施例可能产生的有益效果不同，在不同的实施例里，可能产生的有益效果可以是 以上任意一种或几种的组合，也可以是其他任何可能获得的有益效果。
上文已对基本概念做了描述，显然，对于本领域技术人员来说，上述详细披露仅仅作为示例，而并不构成对本申请的限定。虽然此处并没有明确说明，本领域技术人员可能会对本申请进行各种修改、改进和修正。该类修改、改进和修正在本申请中被建议，所以该类修改、改进、修正仍属于本申请示范实施例的精神和范围。
同时，本申请使用了特定词语来描述本申请的实施例。如“一个实施例”、“一实施例”、和/或“一些实施例”意指与本申请至少一个实施例相关的某一特征、结构或特点。因此，应强调并注意的是，本说明书中在不同位置两次或多次提及的“一实施例”或“一个实施例”或“一个替代性实施例”并不一定是指同一实施例。此外，本申请的一个或多个实施例中的某些特征、结构或特点可以进行适当的组合。
此外，本领域技术人员可以理解，本申请的各方面可以通过若干具有可专利性的种类或情况进行说明和描述，包括任何新的和有用的工序、机器、产品或物质的组合，或对他们的任何新的和有用的改进。相应地，本申请的各个方面可以完全由硬件执行、可以完全由软件(包括固件、常驻软件、微码等)执行、也可以由硬件和软件组合执行。以上硬件或软件均可被称为“数据块”、“模块”、“引擎”、“单元”、“组件”或“系统”。此外，本申请的各方面可能表现为位于一个或多个计算机可读介质中的计算机产品，该产品包括计算机可读程序编码。
计算机存储介质可能包含一个内含有计算机程序编码的传播数据信号，例如在基带上或作为载波的一部分。该传播信号可能有多种表现形式，包括电磁形式、光形式等，或合适的组合形式。计算机存储介质可以是除计算机可读存储介质之外的任何计算机可读介质，该介质可以通过连接至一个指令执行系统、装置或设备以实现通讯、传播或传输供使用的程序。位于计算机存储介质上的程序编码可以通过任何合适的介质进行传播，包括无线电、电缆、光纤电缆、RF、或类似介质，或任何上述介质的组合。
本申请各部分操作所需的计算机程序编码可以用任意一种或多种程序语言编写，包括面向对象编程语言如Java、Scala、Smalltalk、Eiffel、JADE、Emerald、C++、C#、VB.NET、Python等，常规程序化编程语言如C语言、Visual Basic、Fortran 2003、Perl、COBOL 2002、PHP、ABAP，动态编程语言如Python、Ruby和Groovy，或其他编程语言等。该程序编码可以完全在用户计算机上运行、或作为独立的软件包在用户计算机上运行、或部分在用户计算机上运行部分在远程计算机运行、或完全在远程计算机或服务器上运行。在后种情况下，远程计算机可以通过任何网络形式与用户计算机连接，比 如局域网(LAN)或广域网(WAN)，或连接至外部计算机(例如通过因特网)，或在云计算环境中，或作为服务使用如软件即服务(SaaS)。
此外，除非权利要求中明确说明，本申请所述处理元素和序列的顺序、数字字母的使用、或其他名称的使用，并非用于限定本申请流程和方法的顺序。尽管上述披露中通过各种示例讨论了一些目前认为有用的发明实施例，但应当理解的是，该类细节仅起到说明的目的，附加的权利要求并不仅限于披露的实施例，相反，权利要求旨在覆盖所有符合本申请实施例实质和范围的修正和等价组合。例如，虽然以上所描述的系统组件可以通过硬件设备实现，但是也可以只通过软件的解决方案得以实现，如在现有的服务器或移动设备上安装所描述的系统。
同理，应当注意的是，为了简化本申请披露的表述，从而帮助对一个或多个发明实施例的理解，前文对本申请实施例的描述中，有时会将多种特征归并至一个实施例、附图或对其的描述中。但是，这种披露方法并不意味着本申请对象所需要的特征比权利要求中提及的特征多。实际上，实施例的特征要少于上述披露的单个实施例的全部特征。
一些实施例中使用了描述成分、属性数量的数字，应当理解的是，此类用于实施例描述的数字，在一些示例中使用了修饰词“大约”、“近似”或“大体上”来修饰。除非另外说明，“大约”、“近似”或“大体上”表明所述数字允许有±20％的变化。相应地，在一些实施例中，说明书和权利要求中使用的数值参数均为近似值，该近似值根据个别实施例所需特点可以发生改变。在一些实施例中，数值参数应考虑规定的有效数位并采用一般位数保留的方法。尽管本申请一些实施例中用于确认其范围广度的数值域和参数为近似值，在具体实施例中，此类数值的设定在可行范围内尽可能精确。
针对本申请引用的每个专利、专利申请、专利申请公开物和其他材料，如文章、书籍、说明书、出版物、文档等，特此将其全部内容并入本申请作为参考。与本申请内容不一致或产生冲突的申请历史文件除外，对本申请权利要求最广范围有限制的文件(当前或之后附加于本申请中的)也除外。需要说明的是，如果本申请附属材料中的描述、定义、和/或术语的使用与本申请所述内容有不一致或冲突的地方，以本申请的描述、定义和/或术语的使用为准。
最后，应当理解的是，本申请中所述实施例仅用以说明本申请实施例的原则。其他的变形也可能属于本申请的范围。因此，作为示例而非限制，本申请实施例的替代配置可视为与本申请的教导一致。相应地，本申请的实施例不仅限于本申请明确介绍和描述的实施例。

Claims (20)

  1. 一种当前页面信息刷新方法，其特征在于，包括：
    响应于接收到用户的信息刷新请求，获取当前显示的信息数据；
    基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息；
    向用户展示所述可能感兴趣的信息。
  2. 如权利要求1所述的方法，其特征在于，所述用户的信息刷新请求的方式包括手势密码、连续点击操作、点击按键操作、停顿触摸屏幕操作、连续晃动操作、语音输入操作、人脸识别、表情识别或虹膜识别。
  3. 如权利要求1所述的方法，其特征在于，所述信息刷新请求为信息反向刷新请求。
  4. 如权利要求3所述的方法，其特征在于，所述信息刷新请求包括反向阈值，所述反向阈值用于表征刷新后的信息与所述当前显示的信息数据的关联度。
  5. 如权利要求4所述的方法，其特征在于，所述反向阈值的类型至少包括同类反向、不同类反向或异类小众反向。
  6. 如权利要求3所述的方法，其特征在于，所述基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息包括：
    基于所述当前显示的信息数据，确定当前显示的信息数据类型；
    基于所述信息刷新请求和所述当前显示的信息数据类型，对所述当前显示的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型；
    基于所述用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。
  7. 如权利要求6所述的方法，其特征在于，所述基于所述当前显示的信息数据，确定当前显示的信息数据类型包括：
    利用机器学习模型处理所述当前显示的信息数据，确定当前显示的信息数据类型。
  8. 如权利要求7所述的方法，其特征在于，所述机器学习模型包括分类模型；
    所述机器学习模型通过以下方法获得：
    获取训练样本；所述训练样本包括历史显示的信息数据以及历史显示的信息数据类型；其中，将所述历史显示的信息数据类型标记作为参考信息数据类型；
    基于所述训练样本，训练初始模型得到所述机器学习模型。
  9. 如权利要求6所述的方法，其特征在于，所述基于所述当前显示的信息数据类型，对所述当前显示的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型包括：
    基于所述当前显示的信息数据类型，选取反向阈值的类型中至少一种数据类型，将其作为所述用户可能感兴趣的信息类型。
  10. 如权利要求3所述的方法，其特征在于，所述基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息包括：
    响应于接收到用户的信息刷新请求，获取用户的历史浏览信息；
    基于所述当前显示的信息数据和所述用户的历史浏览信息，确定不感兴趣的信息数据类型；
    基于所述信息刷新请求和所述不感兴趣的信息数据类型，对所述不感兴趣的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型；
    基于所述用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。
  11. 如权利要求1所述的方法，其特征在于，所述向用户展示所述可能感兴趣的信息包括：
    根据Feed流的方式在用户终端上展示所述可能感兴趣的信息中至少一部分。
  12. 一种当前页面信息刷新系统，其特征在于，所述系统包括：
    用于存储计算机指令的至少一个存储器；
    与所述存储器通讯的至少一个处理器，其中当所述至少一个处理器执行所述计算机指令时，所述至少一个处理器使所述系统执行：
    响应于接收到用户的信息刷新请求，获取当前显示的信息数据；基于所述信息刷 新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息；
    向用户展示所述可能感兴趣的信息。
  13. 如权利要求12所述的系统，其特征在于，所述信息刷新请求为信息反向刷新请求。
  14. 如权利要求13所述的系统，其特征在于，为基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息，所述至少一个处理器使所述系统进一步执行：
    基于所述当前显示的信息数据，确定当前显示的信息数据类型；
    基于所述信息刷新请求和所述当前显示的信息数据类型，对所述当前显示的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型；
    基于所述用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。
  15. 如权利要求14所述的系统，其特征在于，为基于所述当前显示的信息数据，确定当前显示的信息数据类型，所述至少一个处理器使所述系统进一步执行：
    利用机器学习模型处理所述当前显示的信息数据，确定当前显示的信息数据类型。
  16. 如权利要求14所述的系统，其特征在于，为基于所述当前显示的信息数据类型，对所述当前显示的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型，所述至少一个处理器使所述系统进一步执行：
    基于所述当前显示的信息数据类型，选取反向阈值的类型中至少一种数据类型，将其作为所述用户可能感兴趣的信息类型。
  17. 如权利要求13所述的系统，其特征在于，为基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息，所述至少一个处理器使所述系统进一步执行：
    响应于接收到用户的信息刷新请求，获取用户的历史浏览信息；
    基于所述当前显示的信息数据和所述用户的历史浏览信息，确定不感兴趣的信息数据类型；
    基于所述信息刷新请求和所述不感兴趣的信息数据类型，对所述不感兴趣的信息数据类型进行反向处理，得到用户可能感兴趣的信息类型；
    基于所述用户可能感兴趣的信息类型，确定用户可能感兴趣的信息。
  18. 如权利要求12所述的系统，其特征在于，为向用户展示所述可能感兴趣的信息，所述至少一个处理器使所述系统进一步执行：
    根据Feed流的方式在用户终端上展示所述可能感兴趣的信息中至少一部分。
  19. 一种当前页面信息刷新系统，其特征在于，所述系统包括：
    获取模块，用于响应于接收到用户的信息刷新请求，获取当前显示的信息数据；
    确定模块，用于基于所述信息刷新请求和所述当前显示的信息数据，确定用户可能感兴趣的信息；
    展示模块，用于向用户展示所述可能感兴趣的信息。
  20. 一种计算机可读存储介质，所述存储介质存储计算机指令，当计算机读取所述存储介质中的所述计算机指令后，所述计算机执行如权利要求1～11中任一项所述的方法。
PCT/CN2020/082306 2020-03-31 2020-03-31 一种当前页面信息刷新方法和系统 WO2021195922A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
CN202080005701.8A CN112912915A (zh) 2020-03-31 2020-03-31 一种当前页面信息刷新方法和系统
PCT/CN2020/082306 WO2021195922A1 (zh) 2020-03-31 2020-03-31 一种当前页面信息刷新方法和系统
US17/295,055 US11971934B2 (en) 2020-03-31 2020-03-31 Methods and systems for refreshing current page information
US18/396,847 US20240126823A1 (en) 2020-03-31 2023-12-27 Methods and systems for refreshing current page information

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/082306 WO2021195922A1 (zh) 2020-03-31 2020-03-31 一种当前页面信息刷新方法和系统

Related Child Applications (2)

Application Number Title Priority Date Filing Date
US17/295,055 A-371-Of-International US11971934B2 (en) 2020-03-31 2020-03-31 Methods and systems for refreshing current page information
US18/396,847 Continuation US20240126823A1 (en) 2020-03-31 2023-12-27 Methods and systems for refreshing current page information

Publications (1)

Publication Number Publication Date
WO2021195922A1 true WO2021195922A1 (zh) 2021-10-07

Family

ID=76112914

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/082306 WO2021195922A1 (zh) 2020-03-31 2020-03-31 一种当前页面信息刷新方法和系统

Country Status (3)

Country Link
US (2) US11971934B2 (zh)
CN (1) CN112912915A (zh)
WO (1) WO2021195922A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114880562A (zh) * 2022-05-10 2022-08-09 北京百度网讯科技有限公司 用于推荐信息的方法和装置
CN117591747B (zh) * 2024-01-11 2024-05-07 浙江同花顺智能科技有限公司 一种信息生成式推荐方法、装置、电子设备及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102830922A (zh) * 2012-08-07 2012-12-19 晶赞广告（上海）有限公司 一种广告点击效果的热点数据可视化方法
CN105260458A (zh) * 2015-10-15 2016-01-20 海信集团有限公司 一种用于显示装置的视频推荐方法及显示装置
US20160196582A1 (en) * 2015-01-02 2016-07-07 Verizon Patent And Licensing Inc. Subscriber location audience insights for enterprise networks
CN107220850A (zh) * 2017-05-25 2017-09-29 努比亚技术有限公司 一种广告的推送方法、终端及计算机可读存储介质
CN109741134A (zh) * 2018-12-28 2019-05-10 出门问问信息科技有限公司 信息推送方法、装置、电子设备及计算机可读存储介质
CN111080378A (zh) * 2020-01-22 2020-04-28 成都捷顺宝信息科技有限公司 一种基于人脸识别的广告精准投放系统及方法

Family Cites Families (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6317722B1 (en) * 1998-09-18 2001-11-13 Amazon.Com, Inc. Use of electronic shopping carts to generate personal recommendations
US8949899B2 (en) * 2005-03-04 2015-02-03 Sharp Laboratories Of America, Inc. Collaborative recommendation system
CN102332006B (zh) 2011-08-03 2016-08-03 百度在线网络技术（北京）有限公司 一种信息推送控制方法及装置
US20130091013A1 (en) * 2011-10-07 2013-04-11 Microsoft Corporation Presenting Targeted Social Advertisements
US20150213487A1 (en) * 2013-02-03 2015-07-30 Arize Nwosu Methods and systems for advertising using scrollable refresh trigger
US20140250177A1 (en) 2013-03-01 2014-09-04 Google Inc. Recommending content based on proxy-based preference indications
CN103886090B (zh) * 2014-03-31 2018-01-02 北京搜狗科技发展有限公司 基于用户喜好的内容推荐方法及装置
CN111060377B (zh) 2014-04-28 2023-01-03 深圳迈瑞生物医疗电子股份有限公司 样本篮传送系统及方法
US10332185B2 (en) 2014-05-22 2019-06-25 Google Llc Using status of sign-on to online services for content item recommendations
CN104239587B (zh) 2014-10-17 2017-09-12 北京字节跳动网络技术有限公司 新闻列表刷新的方法及装置
CN104809154B (zh) 2015-03-19 2019-03-08 百度在线网络技术（北京）有限公司 用于资讯推荐的方法及装置
CN105630868B (zh) * 2015-12-15 2019-05-31 北京奇虎科技有限公司 一种向用户推荐内容的方法及系统
CN105574182A (zh) * 2015-12-22 2016-05-11 北京搜狗科技发展有限公司 一种新闻推荐方法和装置、一种用于新闻推荐的装置
CN105740468B (zh) 2016-03-07 2019-10-18 达而观信息科技（上海）有限公司 一种结合内容发布方信息的个性化推荐方法及系统
WO2018040069A1 (zh) * 2016-09-02 2018-03-08 浙江核新同花顺网络信息股份有限公司 信息推荐系统及方法
CN107944033B (zh) 2017-12-13 2022-02-18 北京百度网讯科技有限公司 关联话题推荐方法和装置
CN108647349A (zh) * 2018-05-15 2018-10-12 优视科技有限公司 一种内容推荐方法、装置和终端设备
CN109376536B (zh) 2018-08-21 2023-11-14 中国平安人寿保险股份有限公司 获取Cookie的方法、装置、计算机设备以及存储介质
CN109284488B (zh) 2018-09-06 2021-11-19 郑州云海信息技术有限公司 基于本地存储修改前端表格列数据的方法、装置及介质
CN109819284B (zh) 2019-02-18 2022-11-15 平安科技（深圳）有限公司 一种短视频推荐方法、装置、计算机设备及存储介质
CN109992715B (zh) * 2019-03-28 2021-08-03 网易传媒科技（北京）有限公司 信息展示方法、装置、介质和计算设备
CN110321477B (zh) 2019-05-24 2022-09-09 平安科技（深圳）有限公司 信息推荐方法、装置、终端及存储介质
CN110378732B (zh) * 2019-07-18 2023-01-06 腾讯科技（深圳）有限公司 信息显示方法、信息关联方法、装置、设备及存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102830922A (zh) * 2012-08-07 2012-12-19 晶赞广告（上海）有限公司 一种广告点击效果的热点数据可视化方法
US20160196582A1 (en) * 2015-01-02 2016-07-07 Verizon Patent And Licensing Inc. Subscriber location audience insights for enterprise networks
CN105260458A (zh) * 2015-10-15 2016-01-20 海信集团有限公司 一种用于显示装置的视频推荐方法及显示装置
CN107220850A (zh) * 2017-05-25 2017-09-29 努比亚技术有限公司 一种广告的推送方法、终端及计算机可读存储介质
CN109741134A (zh) * 2018-12-28 2019-05-10 出门问问信息科技有限公司 信息推送方法、装置、电子设备及计算机可读存储介质
CN111080378A (zh) * 2020-01-22 2020-04-28 成都捷顺宝信息科技有限公司 一种基于人脸识别的广告精准投放系统及方法

Also Published As

Publication number Publication date
US11971934B2 (en) 2024-04-30
US20240126823A1 (en) 2024-04-18
CN112912915A (zh) 2021-06-04
US20220309114A1 (en) 2022-09-29

Similar Documents

Publication Publication Date Title
US9959551B1 (en) Customer-level cross-channel message planner
US11290413B2 (en) Trend detection for content targeting using an information distribution system
US9348898B2 (en) Recommendation system with dual collaborative filter usage matrix
US9183282B2 (en) Methods and systems for inferring user attributes in a social networking system
US10031738B2 (en) Providing application recommendations
US9436919B2 (en) System and method of tuning item classification
US20240126823A1 (en) Methods and systems for refreshing current page information
US11188992B2 (en) Inferring appropriate courses for recommendation based on member characteristics
WO2018121700A1 (zh) 基于已安装应用来推荐应用信息的方法、装置、终端设备及存储介质
US11049167B1 (en) Clustering interactions for user missions
US9767417B1 (en) Category predictions for user behavior
US9767204B1 (en) Category predictions identifying a search frequency
US20150287092A1 (en) Social networking consumer product organization and presentation application
US20180349977A1 (en) Determination of unique items based on generating descriptive vectors of users
US10949480B2 (en) Personalized per-member model in feed
US20190163780A1 (en) Generalized linear mixed models for improving search
US10474670B1 (en) Category predictions with browse node probabilities
US11151661B2 (en) Feed actor optimization
US20150278907A1 (en) User Inactivity Aware Recommendation System
EP4298560A1 (en) Automated machine learning to generate recommendations for websites or applications
US20200175084A1 (en) Incorporating contextual information in large-scale personalized follow recommendations
US11514115B2 (en) Feed optimization
US10387934B1 (en) Method medium and system for category prediction for a changed shopping mission
US10341272B2 (en) Personality reply for digital content
US20200118038A1 (en) Techniques for improving downstream utility in making follow recommendations

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: 20928507

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: 20928507

Country of ref document: EP

Kind code of ref document: A1