WO2016041517A1 - 智能提醒方法、系统和装置 - Google Patents

智能提醒方法、系统和装置 Download PDF

Info

Publication number
WO2016041517A1
WO2016041517A1 PCT/CN2015/089909 CN2015089909W WO2016041517A1 WO 2016041517 A1 WO2016041517 A1 WO 2016041517A1 CN 2015089909 W CN2015089909 W CN 2015089909W WO 2016041517 A1 WO2016041517 A1 WO 2016041517A1
Authority
WO
WIPO (PCT)
Prior art keywords
message
reminder
type
feature data
push
Prior art date
Application number
PCT/CN2015/089909
Other languages
English (en)
French (fr)
Inventor
张瞰
刘宏军
曾令伟
袁传顺
孟雷
张俊豪
王翀
张骄阳
Original Assignee
上海触乐信息科技有限公司
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 上海触乐信息科技有限公司 filed Critical 上海触乐信息科技有限公司
Priority to US15/512,824 priority Critical patent/US20170286912A1/en
Priority to EP15841242.9A priority patent/EP3211925A4/en
Publication of WO2016041517A1 publication Critical patent/WO2016041517A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/12—Messaging; Mailboxes; Announcements
    • 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
    • G06Q10/00—Administration; Management
    • G06Q10/10—Office automation; Time management
    • G06Q10/109—Time management, e.g. calendars, reminders, meetings or time accounting
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/21—Monitoring or handling of messages
    • H04L51/224—Monitoring or handling of messages providing notification on incoming messages, e.g. pushed notifications of received messages
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/21—Monitoring or handling of messages
    • H04L51/226—Delivery according to priorities
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/12—Messaging; Mailboxes; Announcements
    • H04W4/14—Short messaging services, e.g. short message services [SMS] or unstructured supplementary service data [USSD]

Definitions

  • the invention relates to the field of human-computer interaction of electronic devices, in particular to the technical field of electronic device intelligent reminding, in particular to a smart reminding method and system.
  • different merchants may use different software to make notifications or reminders. For example, some merchants use SMS messages, some merchants use Facebook or WeChat accounts, and some merchants use emails. For users, there is no unified mechanism for many different merchants. Different forms of push messages are managed. Not to mention the establishment of a unified reminder mechanism based on these different forms of push messages.
  • the technical problem to be solved by the present invention is that a reminder message is obtained from a push message received by the electronic device, and a smart reminder is implemented.
  • the present invention provides a smart reminding method, including: acquiring a push message received by an electronic device terminal, wherein the push message includes a reminder message, and the reminder message includes classification associated feature data; The classifying association feature data, filtering the push message, obtaining a reminder message, and classifying the reminder message into different types, wherein each type of the reminder message is associated with at least one action; according to the reminder The different types of messages that perform the actions associated with the type of alert message.
  • the present invention also provides a smart reminding system, comprising: a message receiving device, And a detecting device, configured to: according to the classified association feature data in the received push message, detect the push message, obtain the reminder message, and calibrate the reminder message to be different a type, wherein each type of reminder message is associated with at least one action; and a reminder device is configured to alert the user according to the reminder message.
  • a smart reminding system comprising: a message receiving device, And a detecting device, configured to: according to the classified association feature data in the received push message, detect the push message, obtain the reminder message, and calibrate the reminder message to be different a type, wherein each type of reminder message is associated with at least one action; and a reminder device is configured to alert the user according to the reminder message.
  • the present invention fully considers the characteristics of the reminder message, processes the push message according to the classified association feature data, obtains the reminder message therefrom, and further divides the reminder message into different types, and according to the input of the user, Perform corresponding actions for different types to implement smart reminders for push messages.
  • FIG. 1 is a block diagram of a smart reminder system according to an embodiment of the present invention.
  • FIG. 2 is a schematic structural diagram of a smart reminder system according to an embodiment of the present invention.
  • FIG. 3 is a schematic structural diagram of an exemplary data of a reminder message stored in a local database according to an embodiment of the present invention.
  • FIG. 4 and FIG. 5 are schematic diagrams showing different structures of a processor according to an embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of a processor according to another embodiment of the present invention.
  • FIG. 7 is a schematic diagram of the operation of a specific embodiment of the sorting device shown in FIG. 6.
  • FIG. 8 is a schematic structural diagram of an intelligent reminding system according to another embodiment of the present invention.
  • FIG. 9 and FIG. 10 are schematic diagrams of reminding a device to remind a user on different interfaces according to an embodiment of the present invention.
  • FIG. 11 is a schematic structural diagram of a smart reminder system according to still another embodiment of the present invention.
  • FIG. 12 to FIG. 14 are schematic diagrams showing operations of a smart reminder system according to still another embodiment of the present invention.
  • FIG. 15 is a schematic flowchart diagram of a smart reminding method according to an embodiment of the present invention.
  • the smart reminder system provided by the present invention can be configured to be based on the push received by the electronic device Sending a message, extracting the classified association feature data, detecting the push message, using the message meeting the reminder condition as a reminder message, thereby alerting the user, and classifying the reminder message into different types according to the classified association feature data, when detecting When the user selects the reminder message, the action associated with the type is performed according to the type of the alert message.
  • FIG. 1 shows a block diagram of an exemplary smart reminder system 100.
  • the smart reminder system 100 can include an electronic device 110.
  • the electronic device 110 may be a communication electronic device having a message receiving function, such as a mobile phone, a smart phone, or a PDA; or may be another electronic device that can receive a message associated with a set account through a network, such as a tablet, a camera, Wear electronic devices, car navigation devices, and electronic interactive terminals installed in public places such as stations or schools.
  • a message receiving function such as a mobile phone, a smart phone, or a PDA
  • a network such as a tablet, a camera, Wear electronic devices, car navigation devices, and electronic interactive terminals installed in public places such as stations or schools.
  • the electronic device 110 can access the telecommunication network through technologies such as CDMA, 2G, 3G, 4G, etc., and receive the push message 101 sent through the telecommunication network; or through broadband, such as ADSL, VDSL, optical fiber, wireless, cable television, satellite, etc. Or access the Internet through narrowband, such as telephone dial-up access, GPRS, 2G, 3G, etc., and receive push messages 101 associated with certain set accounts.
  • technologies such as CDMA, 2G, 3G, 4G, etc.
  • broadband such as ADSL, VDSL, optical fiber, wireless, cable television, satellite, etc.
  • narrowband such as telephone dial-up access, GPRS, 2G, 3G, etc.
  • the push message 101 mentioned herein may be a mobile phone short message.
  • the advertisement or notification message of the merchant is transmitted to the short message processing center (SMSC) via the base station, and further pushed to the user's mobile phone through the GMS network or the short message gateway.
  • the user can receive and open the short message through the mobile phone, and see the content of the short message through the mobile phone screen.
  • the push message 101 may also be a message associated with the set account, such as a message sent to a specific email address, iTune account, facebook account, WeChat account, QQ account or other account, the message is pushed to the designated electronic device or The account is specified so that the user can receive it by specifying the electronic device or via a designated account.
  • the detecting device 120 or the cloud detecting device 130 extracts the classified association feature data from the push message 101 received by the electronic device 110, detects the push message 101, excludes the non-alert message, obtains the reminder message, and calibrates the reminder messages. For different types, the user is alerted by the reminder device 140, wherein each type of reminder message is associated with at least one action.
  • the detecting device 120 parses all the short messages obtained by the mobile phone, or may also parse the set proportion of the mobile phone short messages, for example, 1%-30%, or may also filter the matching settings.
  • Regular mobile phone text messages are parsed, for example, only text messages from non-phonebook contacts are parsed or only text content contained in the message content is parsed, such as "amount” or "yuan” or "balance", or only parsing from settings
  • the mobile phone number of the number for example, the short message from the sender number 100861.
  • the detecting device 120 extracts the classified association feature data that can identify different types of short messages, acquires the reminder message, and calibrates it to a different type.
  • the detecting device 120 determines, according to whether the mobile phone short message to be parsed includes the set classification association feature data, to implement calibration of the type of the mobile phone short message.
  • the classification association feature data may include a character string formed by a keyword in the short message content or a shorthand word thereof, or may be a short message sender number, or a short message base station service center code, or a combination of the foregoing feature data.
  • the detecting device 120 detects that the short message content has the classification related feature data, for example, a character string including the keyword “yuan”, “balance”, “insufficient”, etc., it is marked as a payment reminding short message, or detects that the short message content is included.
  • the keyword “Express”, “include”, “outbound”, “order number” or a synonym of the keyword is used, the short message is marked as a courier reminder message.
  • the detecting device 120 extracts the short message sender number from the short message, for example, when the short message sender number is 1-800-604-9961, the mobile phone number of the Bank of America is identified by identifying the short message sender number. The message is then categorized as a bank reminder message.
  • the short message base station service center code may be used as the classification association feature data, and the detecting device 120 detects the push message according to the classified association feature data, obtains the reminder message, and performs type calibration on the reminder message.
  • the detecting device 120 may further perform calibration on the classification of the mobile phone short message according to the ratio of the classified association feature data corresponding to each category included in the mobile phone short message to be parsed.
  • the detecting device 120 may further perform semantic analysis on the short message content, or use a regular expression to match the short message content, and determine whether there is content that is the same or similar to the classified type related feature data of the set type. And mark the message as the type of setting.
  • the detecting device 120 further detects the user's input signal, and when the user's selection of the reminder message is detected, performs an action 150 associated therewith based on the type of the alert message. For example, when the user is reminded by the reminder device 140 and the user is selected to select the payment SMS, the action associated with the payment type is acquired and executed, for example, the link of the payment interface is opened, so that the user can directly perform the premium refill.
  • detection device 120 or cloud detection device 130 may be an executable program that can be read and executed by a computer processor.
  • the detection device 120 can be installed in the electronic device 110 and receive the push message 101 by listening to the message received by the electronic device 110.
  • the cloud detecting device 130 may be installed in the cloud server to acquire the push message 101 received via the electronic device 110 through network transmission.
  • the smart reminder system 100 can include both the local detecting device 120 and the cloud detecting device 130, and the local detecting device 120 and the cloud detecting device 130 are sequentially or simultaneously acquired by the electronic device 110 according to system settings or according to network status. Push messages are detected.
  • the detected reminder message is further sent to the reminding device 140, and the reminding device 140 reminds the user by using a form of video, audio, or the like.
  • the reminder device 140 can include one or more of displaying text or graphics to the user. a display screen or other display device or a driver for driving the display device to display a reminder message to the user; or a sound device such as a speaker or a driver for driving the sound device, and playing the reminder message in the form of voice It may also include devices for performing other reminder functions, such as devices that implement reminders by indicators, vibrations, and the like.
  • FIG. 2 shows an exemplary structural block diagram of the smart reminder system 100.
  • the smart reminder system 100 can include a receiving device 210, a processor 220, a memory 240, and a reminder device 140.
  • the processor 220 may be a central processing unit (“CPU”) or a graphics processing unit (“GPU”). Specifically, the processor 220 may further include one or more printed circuit boards or microprocessor module chips, and the computer The sequence of program instructions is to perform various methods that will be explained in more detail below.
  • the processor 220 is configured to receive a push message from the receiving device 210, filter the push message according to the classified association feature data therein, and obtain the reminder message in the push message stored in the memory 240, and The alert message is categorized into different types, and the selection of the alert message is monitored; when the select alert message is detected, an action associated with the alert message type is retrieved from the memory 240 to perform the associated action.
  • Memory 240 can include one or more of random access memory (“RAM”) and read only memory (“ROM”). Computer program instructions can be accessed and read from a ROM or any other suitable memory location and loaded into RAM for execution by processor 220.
  • memory 240 can store one or more software applications. Software applications stored in memory 240 may include operating systems for both general computer systems and devices for software control.
  • memory 240 can store only a portion of the entire software application or storage software application that can be executed by processor 220.
  • the memory 240 can store smart reminder software executable by the processor 220 and execute the smart reminder method.
  • memory 240 can also store one or more types of primary data, user data, application data, and program code.
  • the memory 240 can store the local database 330.
  • local database 330 includes one or more alert messages.
  • FIG. 3 illustrates an exemplary reminder message including one or more data fields that store information describing categories and associated actions indicated by the alert message, such as a category of the category to which the alert message belongs.
  • Correlation feature data such as a keyword, a sender number, a base station service center code, a message content, and the like, for example, category information associated with the reminder message, such as a category name, a category description, and the like, and, for example, action information associated with the reminder message, such as Associated actions, action data, etc.
  • the database can be used to broadly include any data format for storing data.
  • receiving device 210 and reminder device 140 can be coupled to processor 220 via appropriate interface circuitry.
  • receiving device 210 can be a short message receiving device, or a mail receiving device, or other communication software receiving device.
  • the smart reminder system 100 can further include a communication interface 231 that can provide a communication connection such that the smart reminder system 100 can exchange information with certain external devices.
  • communication interface 231 can include a network interface (not shown) configured to transmit and receive information from cloud service 230.
  • the cloud service 230 can be implemented as a web service, a cloud storage service, or the like on the Internet.
  • processor 220 further includes a filtering device 310 and a calibration device 320, in accordance with certain embodiments of the present invention.
  • the receiving device 210 obtains the push message, and sends the push message to the filtering device 310.
  • the filtering device 310 filters the push message to obtain the reminder message
  • the calibration device 320 calibrates the reminder message according to the classified association feature data.
  • the reminder messages are divided into different types, each of which is associated with at least one of the set actions in database 330.
  • the reminder message processed via the filtering device 310 and the calibration device 320 is sent to the reminder device 140 and prompted by the reminder device 140.
  • the receiving device 210 receives N push messages within a first time threshold and sends all of the push messages to the filtering device 310.
  • the filtering device 310 filters all the push messages by using the set filtering manner, and obtains the reminder message therefrom.
  • the filtering device 310 may filter the push message according to the classified association feature data. For example, the filtering device 310 may filter the push message according to the short message sender number, exclude the push message from the contact, and use the push message from the non-contact as the reminder message; or the filtering device 310 may also be based on the key in the short message content.
  • the word or keyword combination filters the push message. For example, the filtering device 310 may use the short message of the short message that sets the keyword “balance”, “amount”, “insufficient” or set the keyword as the reminder message.
  • the filtering device 310 may also filter the push message according to the SMS base station service center code. For example, the filtering device 310 may exclude some push messages from the pseudo base station, thereby avoiding alerting the user by using some fraudulent text messages as reminder messages.
  • the filtering device 310 can also perform multiple filtering using multiple classification associated feature data.
  • the filtering device 310 can also filter the push message by using a set ratio. For example, the filtering device 310 can classify the associated feature data filtering for 1%-30% of the push messages.
  • the calibration device 320 For the alert message filtered by the filtering device 310, the calibration device 320 further classifies the alert message according to the classified association feature data, and calibrates the alert message to a different type.
  • Filter device 310, calibration device 320, and database 330 may be local, such as at the same client as receiving device 210 or reminder device 140. According to other embodiments, referring to FIG. 5, the filtering device 310, the calibration device 320, and the database 330 may be located in the cloud, and the transmission of data and signals is performed between the receiving device 210 or the reminding device 140 through a communication interface.
  • the processor 220 further includes a sorting device 410 and a structured device 420; wherein the sorting device 410 receives push messages from the receiving device 210 and filters and sorts them according to the classified association feature data. . Then, the classification device 410 sends the acquired reminder message and its corresponding type to the structuring device 420, and forms a reminder message with a setting structure via the structuring device 420, wherein the setting structure includes the type corresponding to the reminder message. . The structuring device 420 will further send and send a reminder message of the setting structure to the reminder device 140 and to the database 330 for saving.
  • the classification device 410 may further include: a first classifier that filters and detects the push message, and when detecting that the push message includes the set classification association feature data, the push message is marked as A reminder message belonging to a type associated with the set classification associated feature data.
  • the first classifier filters and detects the received push message, and when the push message includes "purchased”, "?”, “?", “?”, “railway”, " When the flight
  • the wildcard "?” is used to identify one or more characters, and in other embodiments, other identifiers may be used as wildcards.
  • the first classifier indexes the push messages as reminder messages belonging to the "ticket” category; when the first classifier detects that the push message includes a category association such as "seller", “buyer”, “order”, “group purchase”, “voucher”, etc.
  • the push message is marked as a reminder message belonging to the “shopping” category; when the first classifier detects that the push message includes the classified association feature data such as “single number”, “express” and “delivery”, the first classifier These push messages are categorized as reminder messages belonging to the "Express” category; when the first classifier detects from "10086?" or “10001?" or “10011?” and the message content contains "Amount” "Insufficient” "? When the class is associated with the feature data, the push messages are categorized as alert messages belonging to the "operator” category.
  • the first type of classifier filters the push messages one by one for each type. For example, the first classifier first determines whether the received short message belongs to the “operator” category reminder message; when it does not belong, it continues to determine whether it belongs to other types until the type corresponding to the short message is obtained. In one embodiment, when the push message does not belong to any of the alert types, the push message is a non-alert message.
  • the first classifier may further comprise an extraction module and a determination module.
  • the extraction module extracts the push message according to the type of the classified association feature data, and the determination module sequentially performs the determination. For example, the extraction module first extracts the short message sender number of all short messages, and the determining module determines whether the short message sender number conforms to the “operator” type, that is, includes “10086?” or “10001?” or “10011?”; If the sender's number does not meet the "operator" type of short message, the extraction module extracts the SMS base station service center code or the short message content, and judges through the judgment module until the type corresponding to the short message is obtained. In another embodiment, the first classifier Make judgments one by one.
  • the extraction module extracts the short message sender number, the short message base station service center code, and the short message content corresponding to the short message, and then the determining module determines, according to the extracted content, which type the short message belongs to.
  • the determining module may directly perform the determining according to a certain category association feature data or a certain category association feature data, for example, when the determining module detects that the short message content of the short message includes “railway” or “ When the "flying" keyword is used, the judging module judges that the short message is a "ticketing" type reminder message.
  • the determining module may determine according to the combination of the plurality of classified association feature data, for example, the determining module includes only the “amount” and the “insufficient” in the short message content of the short message detected, and the short message thereof When the sender number is "10086", the short message is judged as the "operator” short message.
  • the classification device 420 may further include: a second classifier that extracts the push association message by extracting the classification association feature data in the received push message and calculating the probability that the push message belongs to each different type respectively Scaled to the corresponding type.
  • the second classifier extracts the classification-related feature data x i of the i-th push message as ⁇ x 1i , x 2i , . . . , x ni ⁇ , and then, due to type A, type B. . .
  • the type n corresponds to the weight A, the weight B, ..., the weight n; respectively, and the weight A is ⁇ w A0 , w A1 , ..., w An ⁇ , and the weight B is ⁇ w B0 , w B1 ,...,w Bn ⁇ ,..., the weight n is ⁇ w n0 , w n1 , . . . , w nn ⁇ , and the second classifier separately calculates the probability that the push message belongs to each type.
  • the probability that the second classifier obtains that the push message belongs to type A is
  • the probability that the second classifier obtains that the push message belongs to type B is
  • the second classifier may also employ a Bayes classification algorithm, a neural network (NN) algorithm, a convolutional neural network (CNN) algorithm, a support vector machine (SVM) classification algorithm, or other classification algorithm.
  • NN neural network
  • CNN convolutional neural network
  • SVM support vector machine
  • a Bayes classification algorithm calculates the probability that the push message belongs to type n. For example, when there are many types, a naive Bayesian classification algorithm can be employed to increase the reliability of the data.
  • the classification device 410 can include a semantic analyzer that structurally reviews the contextual nature of the push message to determine the type to which the push message belongs.
  • the classification device 410 may further include a matching module that matches the push message with a regular expression and determines whether the push message conforms to the filtering logic of the regular expression.
  • the smart reminder system 100 further includes an update device 340.
  • the update device 340 is adapted to update the corresponding setting parameters of the sorting device 410. Specifically, for example, for the first classifier, the update device 340 updates the filter parameters of the classifier, where the filter parameters include whether the first classifier filters by type or one by one according to a short message. The first classifier extracts a certain type of classified association feature data of all the short messages for filtering or filters all the classified related feature data of each short message. For another example, for the second classifier, the update device 340 updates the weights corresponding to the respective types in the classifier.
  • the update device 340 may also update the action corresponding to the alert message.
  • the reminder message and its corresponding classification are obtained, the reminder message and its corresponding category are sent to the reminder device 140.
  • the reminder messages and their corresponding categories may also be formed into a set structure via the structuring device 420 before being sent to the reminder device 140, wherein the setting structure includes the The type of reminder message.
  • the mobile device balance of the push message "Your number is 138xxxxxxxx” from the number "10086” by the sorting device 410 is 5.76 yuan, please recharge in time to ensure that the mobile phone is unblocked, thank you for "processing", and obtaining the push message belongs to "operation a "type” reminder message, and the structured device 420, based on the short message content, forms a set structure, for example, json structure data:
  • the alert message is structured by the structuring device 420 to facilitate receipt and processing of the data by other receiving programs that are not the alert message, thereby forming a unified alert mark.
  • the structured device 420 may structure the reminding message according to the display manner adopted by the reminding device 140, for example, a program for displaying the reminding message.
  • the reminder message can be read by the displayed program and displayed by the display device 140.
  • the reminder device 140 sets different reminder marks according to different types to which the reminder message belongs, thereby reminding the user.
  • the reminder mark may be a type name, a type description, and the like corresponding to the reminder message.
  • the reminding device 140 can call the interface of other existing applications to make reminders, such as a short message or a notepad, or redraw the reminding interface to remind.
  • the reminder device 140 displays a short message receiving interface, and sets a reminder mark in a vicinity of the short message to implement reminding the user.
  • the reminder device 140 may summarize all reminder messages and alert the user in their newly drawn reminder interface. In the redrawn reminder interface, the reminder device 140 may display the reminder message and the reminder mark, or may only display the reminder mark.
  • the processor 220 further includes a monitoring execution device 520 that further includes a display device 510.
  • display device 510 displays the alert message in a visual form through the screen.
  • the display device 510 may also be a playback device or a vibration device or the like to alert the user in audio or other form.
  • the monitoring execution device 520 detects further input by the user, acquires an action corresponding to the alert message selected by the user, and performs the action.
  • the monitoring execution device 520 can include an input monitoring device 521, an action acquisition device 522, and an action execution device 523.
  • the input monitoring device 521 detects whether the user has an input, for example, when detecting that the user touches or tickes or selects another reminder message on the screen by using an input device such as a finger or a stylus, the user is recorded.
  • the input signal is transmitted to the action acquisition device 522 and the recorded user input signal.
  • the action obtaining device 522 determines the reminder message selected by the user according to the received user input signal, for example, the coordinates touched by the user or the pixel touched by the user, and acquires an action corresponding to the reminder message.
  • the action execution device 523 performs an action acquired via the action acquisition device 522, such as opening an associated link, picture, text, or the like.
  • the user's input signal is sent to the action acquisition device 522.
  • the action obtaining device 522 determines the input signal of the user, acquires the reminder message selected by the user as a “courier” type reminder message, and acquires an action corresponding to the reminder message and action data from the database 330.
  • the action corresponding to the "Express" type reminder message is "Open" Express Check Inquiring the 'link', and the associated action data is the courier number
  • the action acquisition device 522 sends the relevant link and action data to the action execution device 523.
  • the action execution device 523 further opens the link according to the action data for query.
  • the query result interface is displayed to the user.
  • the action obtaining device 522 determines the user input signal monitored by the input monitoring device 521, it is determined that the reminder message selected by the user is a "shopping" type reminder message, and is acquired from the database 330.
  • the action corresponding to the reminder message and the action data For example, if the action corresponding to the “shopping” type reminder message is “opening the link of the purchased item”, and the associated action data is the group purchase ticket picture, the action obtaining device 522 sends the related picture and action data to the action executing device. 523.
  • the action execution device 523 further opens the picture file according to the action data, and displays the corresponding group purchase ticket picture to the user.
  • the action execution device 523 can also include further selecting a user input.
  • the action obtaining device 522 acquires the reminder message selected by the user as a “ticket” type reminder message
  • the action corresponding to the reminder message and the action data are acquired from the database 330.
  • the action corresponding to the "ticket” type reminder message is "open the link of the nearby ticket picking point” "open the link of the 'destination'” "open the link of the pick-up”
  • the associated action data is "Current location 'Shanghai Xuhui District” ''Destination' Beijing''
  • the action acquisition device 522 transmits the relevant link and action data to the action execution device 523.
  • the action execution device 523 displays these three actions to the user and detects the user's input signal.
  • the action execution device 523 detects that the user's input signal is that "destination” is selected, the related link of "Beijing” is opened according to the associated action data "destination 'Beijing'" and "link to open 'destination'” And display the results interface to the user.
  • a smart reminding method including: Step S1, acquiring a received push message, where the push message includes a reminder message, where the reminder message is included Include classification association feature data; step S2, filtering the push message according to the classification association feature data, acquiring an alert message, and calibrating the alert message to different types, wherein each type of the alert message is At least one action is associated; step S3, performing an action associated with the type of the alert message according to different types of the alert message.
  • the pushed short message can be obtained by monitoring the system short message.
  • the push message is filtered to obtain a reminder message, and then the filtered reminder message is calibrated, and the alert message is sent according to the classified association feature data.
  • Scaled to a different type For example, for a short message, the short message is categorized into different types according to the sender number of the short message, the content of the short message, the code of the base station service center, and the like.
  • the push message is classified according to the classification association feature data, and the push message that does not belong to any of the set types is marked as a non-alert message.
  • the push message is detected, and when the push message includes an indication setting When the classification of the type is associated with the feature data, the push message is calibrated to the reminder message of the set type.
  • the push message is categorized as a reminder message of a type with the highest probability by calculating a probability that the push message belongs to a different type.
  • the push message is calibrated to the set type of reminder message based on semantic analysis of the content of the push message.
  • a regular expression is used to match the content of the push message to determine whether there is content that is the same or similar to the category-related feature data of the set type.
  • step S2 further comprises an updating step of updating the correspondence between the classification association feature data and the calibration type, or updating the action associated with the calibration type.
  • the action associated with its type After getting the alert message and its associated classification, perform the action associated with its type. Specifically, after the reminder message is presented to the user, the user's input is further detected, and the selected reminder message is judged according to the user's input; the action associated with the selected reminder message and the action data are acquired, and the associated action is performed.
  • the method further includes: structuring the reminder message, and reminding the user in a set order and a reminding manner according to different types of the reminder message.
  • the present invention fully considers the characteristics of the reminder message, processes the push message according to the classified association feature data, obtains the reminder message therefrom, and further divides the reminder message into different types, and according to the input of the user, Perform corresponding actions for different types to implement smart reminders for push messages.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Economics (AREA)
  • Data Mining & Analysis (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

一种智能提醒方法和系统，其中，该智能提醒系统包括：消息接收设备（110），用于接收与设定账号关联的推送消息（101）；检测设备（120），用于根据所接收的推送消息（101）中的分类关联特征数据，对推送消息（101）进行检测，获取提醒消息，并将所述提醒消息标定为不同的类型，其中每一类型的提醒消息与至少一种动作相关联；提醒设备（140），用于根据所述提醒消息对用户进行提醒。该方法能够在基于推送消息（101）中的分类关联特征数据，有效地从推送消息中获取提醒消息并对提醒消息进行分类，以及基于该提醒消息的类型执行关联的动作，实现智能提醒，提高了用户体验。

Description

智能提醒方法、系统和装置 技术领域
本发明涉及电子设备人机交互领域，特别涉及电子设备智能提醒技术领域，尤其是智能提醒方法和系统。
背景技术
近年来，随着智能手机的广泛普及，商户通过手机短信或社交账号推送信息的方式对客户进行通知也越来越常见。在这些推送消息中，不乏存在着一些用户想要了解，或确实对用户起到提醒作用的消息，例如快递通知消息、火车票订购确认消息、团购券购买凭证消息、话费提醒消息等，然而由于大量垃圾信息相伴而来，常常使得用户手机或通讯终端的电子设备被大量非提醒性的广告或其它垃圾短信淹没，从而使得用户往往需要花大量的时间阅读众多似是而非的推送消息，并从中甄别出真正有用的消息，不仅降低了效率，而且常常会造成重要消息的遗漏以至于引起工作或生活的不便。
此外，不同的商户可能通过不同的软件进行通知或提醒，例如有些商户通过手机短信，有些商户通过facebook或微信账号，有些商户通过邮件，对于用户而言，缺乏统一的机制对来自诸多不同商户的不同形式的推送消息进行管理。更不必说根据这些不同形式的推送消息建立统一的提醒机制。
为了解决上述技术问题，有必要提供一种应用于电子设备上更为智能的提醒方法、系统和装置。
发明内容
本发明要解决的技术问题是：从电子设备接收到的推送消息中获取提醒消息，并实现智能提醒。
根据本发明的一个方面，本发明提供了一种智能提醒方法，包括：获取电子设备终端接收到的推送消息，其中所述推送消息包含提醒消息，所述提醒消息中包含分类关联特征数据；根据所述分类关联特征数据，对所述推送消息进行过滤，获取提醒消息，并且将所述提醒消息标定为不同的类型，其中每一种提醒消息的类型与至少一种动作相关联；根据该提醒消息的不同类型，执行与该提醒消息的类型相关联的动作。
根据本发明的另一个方面，本发明还提供了一种智能提醒系统，包括：消息接收设备， 用于接收与设定账号关联的推送消息；检测设备，用于根据所接收的推送消息中的分类关联特征数据，对推送消息进行检测，获取提醒消息，并将所述提醒消息标定为不同的类型，其中每一类型的提醒消息与至少一种动作相关联；提醒设备，用于根据所述提醒消息对用户进行提醒。
相较于现有技术，本发明充分考虑了提醒消息的特点，根据分类关联特征数据对推送消息进行处理，从中获取提醒消息，并进一步将提醒消息划分为不同的类型，并根据用户的输入，针对不同类型执行对应的动作，从而实现对推送消息的智能提醒。
附图说明
通过阅读参照以下附图对非限制性实施例所作的详细描述，本发明的其它特征、目的和优点将会变得更明显：
图1为本发明一种实施方式中智能提醒系统的框图示意图。
图2为本发明一种实施方式中智能提醒系统的结构图示意图。
图3为本发明一种实施方式中本地数据库中所存储的提醒消息示例性数据结构示意图。
图4和图5为本发明具体实施方式中处理器的不同结构示意图。
图6为本发明另一种实施方式中处理器的结构示意图。
图7为如图6所示分类设备一种具体实施方式的操作示意图。
图8为本发明另一种实施方式中智能提醒系统的结构示意图。
图9和图10为本发明具体实施方式中提醒设备在不同界面对用户进行提醒的示意图。
图11为本发明又一种实施方式中智能提醒系统的结构示意图。
图12至图14为本发明又一种实施方式中智能提醒系统的操作示意图。
图15为本发明一种实施方式中智能提醒方法的流程示意图。
具体实施方式
以下将参考附图对本发明智能提醒方法和系统示例实施方式进行更为全面的描述。附图中相同的附图标记将用来指示相同或相似的部件。尽管以下描述了本发明的若干示例性实施例和特征，但是在不背离本发明的发明思路的情况下，对本发明进行的修改、调整以及其它替换实现，例如，对附图所示部件进行等同替换、添加或修改，或通过替换、重新排序或添加步骤，不应造成对本发明的限制。本发明的适当范围应由所附权利要求所界定。
根据某些实施例，本发明所提供的智能提醒系统可被配置成基于电子设备所接收到的推 送消息，通过提取分类关联特征数据，对推送消息进行检测，将符合提醒条件的消息作为提醒消息，从而对用户进行提醒，并且根据分类关联特征数据将提醒消息划分为不同的类型，当检测到用户对提醒消息的选择时，根据所述提醒消息的类型执行与其类型关联的动作。
图1示出示例性智能提醒系统100的框图。根据某些实施例，该智能提醒系统100可包括电子设备110。该电子设备110可以是具有消息接收功能的通信电子设备，例如移动电话、智能电话、或者PDA；也可以是其它可通过联网接收与设定账号关联的消息的电子设备，例如平板电脑、相机、穿戴电子设备、车载导航设备、设置在车站或学校等公共场所的电子交互终端。
电子设备110可通过CDMA、2G、3G、4G等技术接入电信网络，接收通过电信网络发送的推送消息101；也可通过宽带，例如ADSL、VDSL、光纤、无线、有线电视、卫星等方式，或通过窄带，例如电话拨号接入、GPRS、2G、3G等方式接入互联网，接收与某些设定账号关联的推送消息101。
其中，本文中所提及的推送消息101可以是手机短信，例如，商家的广告或通知消息经由基站传输至短信息处理中心(SMSC)，进一步通过GMS网络或短信网关推送至用户的手机上，使得用户能够通过手机接收并打开该短信，以及通过手机屏幕看到该短信的内容。该推送消息101也可以是与设定账号关联的消息，例如发送至某特定电子邮件地址、iTune账号、facebook账号、微信账号、QQ账号或其它账号的消息，该消息被推送至指定电子设备或指定账号，使得用户能够通过指定电子设备或经由指定账号对其进行接收。
接着，检测设备120或者云端检测设备130从电子设备110所接收到的推送消息101中提取分类关联特征数据，对推送消息101进行检测，排除非提醒消息，获得提醒消息，并将这些提醒消息标定为不同的类型，通过提醒设备140对用户进行提醒，其中每一类型的提醒消息与至少一种动作相关联。
根据本发明的某些实施例，检测设备120对手机所获得的所有短信进行解析，或者也可以对设定比例的手机短信进行解析，例如1％-30％，或者也可以对符合设定过滤规则的手机短信进行解析，例如，仅解析来自非电话簿联络人的手机短信或仅解析短信内容中包含设定内容，例如“金额”或“元”或“余额”，或仅解析来自设定号码的手机短信，例如来自发送方号码为100861的短信。
根据某些实施例，对于待解析的手机短信，检测设备120提取可标识不同类型短信的分类关联特征数据，获取提醒消息，并将其标定为不同的类型。
在一种实施方式中，检测设备120根据检测待解析的手机短信中是否包含设定的分类关联特征数据，以实现对该手机短信所属类型的标定。其中，分类关联特征数据可包含由短信内容中的关键字或其近义字词构成的字符串，也可为短信发送方号码，也可为短信基站服务中心代码，或者上述特征数据的组合。
例如，检测设备120检测到短信内容中具有分类关联特征数据，例如包含关键字“元”“余额”“不足”等的字符串时，将其标定为缴费提醒短信，或者检测到短信内容中包含与关键字“快递”“包括”“出库”“订单号”或其近义词的字符串时，将这样的短信标定为快递提醒短信。
又例如，检测设备120从短信中提取短信发送方号码，例如获取短信发送方号码为1-800-604-9961时，通过识别该短信发送方号码为美国银行(Bank of America)的客服号码，进而将该短信标定为银行提醒短信。
又例如，还可将短信基站服务中心代码作为分类关联特征数据，检测设备120根据该分类关联特征数据对推送消息进行检测，获得提醒消息，并对提醒消息进行类型标定。
在另一种实施方式中，检测设备120还可根据检测待解析的手机短信中所包含的各个分类对应的分类关联特征数据的比率，以实现对该手机短信所属分类的标定。
在又一种实施方式中，检测设备120还可对短信内容进行语义分析，或采用正则表达式对短信内容中进行匹配，判断是否存在与设定类型的分类关联特征数据相同或相近的内容，并将该短信标定为该设定类型。
检测设备120还会进一步检测用户的输入信号，当检测到用户对提醒消息的选择时，根据提醒消息的类型执行与之关联的动作150。例如，当通过提醒设备140对用户进行提醒，并检测到用户对缴费短信进行选择时，获取并执行与缴费类型关联的动作，例如打开缴费界面的链接，便于用户直接进行缴费充值。
根据某些实施例，检测设备120或云端检测设备130可为能够被计算机处理器读取并运行的可执行程序。例如，检测设备120可安装在电子设备110中，并通过监听电子设备110的消息接收，获取推送消息101。又例如，云端检测设备130可安装在云端服务器中，通过网络传输获取经由电子设备110接收的推送消息101。根据另一些实施例，智能提醒系统100可既包含本地检测设备120以及云端检测设备130，根据系统设置或根据网络状态，使得本地检测设备120和云端检测设备130依次或同时对电子设备110所获取的推送消息进行检测。
检测到的提醒消息会进一步发送至提醒设备140，提醒设备140将提醒消息通过视频、音频等形式，对用户进行提醒。例如提醒设备140可包括向用户显示文本或图形的一个或多 个显示屏或其它显示设备或者对显示设备进行驱动的驱动程序，将提醒消息显示给用户；也可包括喇叭等声音设备或者对声音设备进行驱动的驱动程序，用语音的形式将提醒消息播放出来；还可包括用于执行其它提醒功能的设备，例如通过指示灯、振动等实现提醒的设备。
图2示出了智能提醒系统100示例性结构框图。根据某些实施例，该智能提醒系统100可包括接收设备210、处理器220、存储器240、提醒设备140。
其中，处理器220可以是中央处理单元(“CPU”)或图形处理单元(“GPU”)，具体来说，处理器220还可包括一个或多个印刷电路板或微处理模块芯片，运行计算机程序指令序列以执行将在下文中更详细解释的各种方法。在某些实施例中，处理器220可配置为接收来自接收设备210的推送消息，根据其中的分类关联特征数据，对推送消息进行过滤，获取推送消息中的提醒消息存储在存储器240中，并将提醒消息标定为不同的类型，以及监测对提醒消息的选择；当检测到选择提醒消息时，从存储器240中获取与该提醒消息类型关联的动作，执行该关联动作。
存储器240可包括随机存取存储器(“RAM”)和只读存储器(“ROM”)中的一种或多种。计算机程序指令可从ROM或任何其它合适的存储器位置访问和读取，并且被加载到RAM中以供处理器220执行。例如，存储器240可存储一个或多个软件应用。存储在存储器240中的软件应用可包括用于普通计算机系统以及用于软件控制的设备的操作系统。此外，存储器240可存储整个软件应用或者存储软件应用中的可由处理器220执行的仅仅一部分。例如，存储器240可存储可由处理器220执行的智能提醒软件并且执行该智能提醒方法。
在某些实施例中，存储器240也可存储主数据、用户数据、应用数据和程序代码中一种或多种类型。例如，存储器240可存储本地数据库330。在某些实施例中，本地数据库330包括一个或多个提醒消息。例如，图3示出包括一个或多个数据字段的示例性提醒消息，该一个或多个数据字段存储描述该提醒消息所指示的类别及关联动作的相关信息，例如该提醒消息所属类别的分类关联特征数据，如关键词、发送方号码、基站服务中心代码、消息内容等，又例如该提醒消息关联的类别信息，如类别名称、类别描述等，又例如该提醒消息关联的动作信息，如关联的动作、动作数据等。其中，所述数据库可用于宽泛地包括用于存储数据的任何数据格式。
在某些实施例中，接收设备210和提醒设备140可通过适当的接口电路耦合至处理器220。在某些实施例中，接收设备210可以是短信接收设备、或邮件接收设备、或其它通讯软件接收设备。
根据某些实施方式，智能提醒系统100还可进一步包括通信接口231，通信接口231可提供通信连接，使得智能提醒系统100可与某些外部设备交换信息。根据一个实施例，通信接口231可包括网络接口(未示出)，该网络接口被配置成传送和接收来自云端服务230的信息。根据某些实施例，云端服务230可被实现为因特网上的web服务、云端存储服务等。
参考图4，根据本发明某些实施方式，处理器220进一步包括过滤设备310和标定设备320。其中，接收设备210获取推送消息，并将推送消息发送至过滤设备310；接着，过滤设备310对推送消息进行过滤，获得提醒消息，并且，标定设备320根据分类关联特征数据对提醒消息进行标定，将提醒消息划分为不同的类型，其中每一种类型在数据库330中与至少一种设定的动作相关联。经由过滤设备310和标定设备320处理过的提醒消息被发送至提醒设备140，并通过提醒设备140进行提示。
在一种具体实施方式中，接收设备210在第一时间阈值内接收到N条推送消息，并将这些推送消息全部发送至过滤设备310。
接着，过滤设备310采用设定的过滤方式对全部推送消息进行过滤，从中获取提醒消息。
其中，过滤设备310可根据分类关联特征数据对推送消息进行过滤。例如，过滤设备310可根据短信发送方号码对推送消息进行过滤，排除来自联系人的推送消息，将来自非联系人的推送消息作为提醒消息；或者，过滤设备310也可根据短信内容中的关键字或关键字组合对推送消息进行过滤，例如，过滤设备310可将短信内容中包含设定关键字“余额”“金额”“不足”或设定关键字的近义字词的短信作为提醒消息；或者，过滤设备310还可根据短信基站服务中心代码对推送消息进行过滤。例如过滤设备310可排除一些来自伪基站的推送消息，从而避免将一些诈骗短信作为提醒消息对用户进行提醒。此外，过滤设备310还可采用多个分类关联特征数据进行多次过滤。
在另一种实施方式中，过滤设备310还可采用设定比例对所述推送消息进行过滤。例如，过滤设备310可对1％-30％的推送消息进行分类关联特征数据过滤。
对于经由过滤设备310过滤的提醒消息，标定设备320进一步根据分类关联特征数据对其进行分类标定，将提醒消息标定为不同的类型。
过滤设备310、标定设备320以及数据库330可位于本地，例如与接收设备210或者提醒设备140位于同一客户端。根据另一些实施方式，参考图5，过滤设备310、标定设备320以及数据库330可位于云端，其与接收设备210或者提醒设备140之间通过通信接口进行数据和信号的传输。
根据某些实施方式，参考图6，处理器220进一步包括分类设备410和结构化设备420；其中，分类设备410接收来自接收设备210的推送消息，并根据分类关联特征数据对其进行过滤和分类。接着，分类设备410将获取的提醒消息以及其对应的类型送至结构化设备420，经由结构化设备420形成具有设定结构的提醒消息，其中，该设定结构中包含该提醒消息对应的类型。结构化设备420进一步将并将该设定结构的提醒消息发送至提醒设备140，以及发送至数据库330进行保存。
在一种实施方式中，分类设备410进一步可包括：第一分类器，对推送消息进行过滤和检测，当检测到所述推送消息包含设定的分类关联特征数据时，将该推送消息标定为属于与该设定的分类关联特征数据相关联的类型的提醒消息。
例如，参考图7，第一分类器对接收到的推送消息进行过滤和检测，当推送消息包含“已购”、“？次”、“？车”、“？开”、“铁路”、“航班|机票|行程”、“起飞”、“登机”等分类关联特征数据时。其中通配符“？”用于标识一个或多个字符，在不同实施方式中，也可采用其它标识符作为通配符。第一分类器将这些推送消息标定为属于“票务”类别的提醒消息；当第一分类器检测到推送消息包含“卖家”、“买家”“订单”“团购”“？券”等分类关联特征数据时，将这些推送消息标定为属于“购物”类别的提醒消息；当第一分类器检测到推送消息包含“单号”“快递”“派送”等分类关联特征数据时，第一分类器将这些推送消息标定为属于“快递”类别的提醒消息；当第一分类器检测到来自“10086？”或者“10001？”或“10011？”且消息内容中包含“金额”“不足”“？元”等分类关联特征数据时，将这些推送消息标定为属于“运营商”类别的提醒消息。
根据某些实施方式，第一类分类器按照每一个类型逐个对推送消息进行过滤。例如，第一类分类器首先判断接收到的短信是否属于“运营商”类别提醒消息；当不属于时，再继续判断是否属于其它类型，直至获得该短信对应的类型。在一种实施方式中，当推送消息不属于任何一种提醒类型时，该推送消息为非提醒消息。
根据某些实施方式，第一分类器还可进一步包括提取模块和判断模块。在一种实施方式中，提取模块分别按照分类关联特征数据的种类从推送消息中进行提取，由判断模块逐次进行判断。例如，提取模块先提取所有短信的短信发送方号码，判断模块判断短信发送方号码是否符合“运营商”类型，即包含“10086？”或者“10001？”或“10011？”；接着，对于短信发送方号码不符合“运营商”类型的短信，提取模块再提取其短信基站服务中心代码或短信内容，经由判断模块进行判断，直至获得与该短信对应的类型。在另一种实施方式中，第一分类器 逐条短信进行判断。例如，提取模块提取该短信对应的短信发送方号码、短信基站服务中心代码、短信内容，接着，判断模块根据所提取的内容判断该短信属于哪一种类型。其中，在某些实施例中，判断模块可根据某一种分类关联特征数据或者某一个分类关联特征数据直接进行判断，例如，当判断模块检测到该短信的短信内容中包含“铁路”或“航班”关键词时，则判断模块判断该短信为“票务”类型提醒消息。在某些实施例中，判断模块可根据多个分类关联特征数据的组合进行判断，例如，判断模块只有在检测到该短信的短信内容中既包含“金额”也包含“不足”，并且其短信发送方号码为“10086”时，才将该短信判断为“运营商”短信。
根据某些实施方式，分类设备420还可包括：第二分类器，通过提取所接收到的推送消息中的分类关联特征数据，并且计算该推送消息分别属于各个不同类型的概率，将该推送消息标定为对应的类型。
例如，第二分类器提取第i条推送消息的分类关联特征数据xi为{x1i，x2i，...，xni}，接着，由于类型A、类型B。。。类型n分别对应于权值A、权值B、...、权值n；并且，权值A为{wA0，wA1，...，wAn}，权值B为{wB0，wB1，...，wBn}，...，权值n为{wn0，wn1，...，wnn}，第二分类器分别计算该推送消息属于各个类型的概率。
具体来说，第二分类器获取该推送消息属于类型A的概率为
Figure PCTCN2015089909-appb-000001
其中，fA(xi)＝wA0+wA1·x1i+wA2·x2i+...+wAn·xni。
类似的，第二分类器获取该推送消息属于类型B的概率为
Figure PCTCN2015089909-appb-000002
其中，fB(xi)＝wB0+wB1·x1i+wB2·x2i+...+wBn·xni。
直至第二分类器获取该推送消息属于类型n的概率为
Figure PCTCN2015089909-appb-000003
其中，fn(xi)＝wn0+wn1·x1i+wn2·x2i+...+wnn·xni。
根据某些实施方式，第二分类器还可采用贝叶斯(Bayes)分类算法、神经网络(NN)算法、卷积神经网络(CNN)算法、支持向量机(SVM)分类算法或其它分类算法计算该推送消息属于类型n的概率。例如，当类型较多时，可采用朴素贝叶斯分类算法，以增加数据的可靠性。
接着，第二分类器根据该推送消息属于不同类型的概率，将所述推送消息标定为概率最大的类型。也就是说，当PA＝max{PA、PB、...、Pn}时，该推送消息属于类型A。
在另一种实施方式中，分类设备410可包括语义分析器，对推送消息从结构上进行上下文相关性质的审查，判断推送消息所属的类型。在又一种实施方式中，分类设备410还可包括匹配模块，采用正则表达式对推送消息进行匹配，判断推送消息是否符合该正则表达式的过滤逻辑。
根据某些实施方式，参考图8，智能提醒系统100进一步包括更新设备340。其中，该更新设备340适于对分类设备410相应的设置参数进行更新。具体来说，例如，对于第一分类器，更新设备340更新该分类器的过滤参数，所述过滤参数包括，所述第一分类器是按类型进行逐个过滤还是按照短信进行逐条过滤，所述第一分类器是提取所有短信中的某一种类分类关联特征数据进行过滤还是对每一条短信的所有分类关联特征数据进行过滤等。又例如，对于第二分类器，更新设备340更新该分类器中各个类型对应的权值。
在另一种实施方式中，更新设备340还可对提醒消息对应的动作进行更新。
当获取提醒消息及其对应的分类后，提醒消息和其对应的分类被发送至提醒设备140。
根据某些实施方式，提醒消息和其对应的分类在被发送至提醒设备140之前，还可经由结构化设备420，将这些提醒消息形成设定的结构，其中，所述设定结构中包含该提醒消息对应的类型。
例如，经分类设备410对来自号码“10086”的推送消息“您号码为138xxxxxxxx”的手机余额为5.76元，请及时充值，以保证手机畅通，谢谢配合”的处理，获取该推送消息属于“运营商”类型提醒消息，并且结构化设备420，根据该短信内容，形成设定的结构，例如，json结构数据：
Figure PCTCN2015089909-appb-000004
Figure PCTCN2015089909-appb-000005
通过结构化设备420将提醒消息结构化，便于通过其它非该提醒消息的接收程序对该数据进行接收和处理，从而形成统一的提醒标记。
在一种实施方式中，当提醒设备140采用显示的方式进行提醒时，结构化设备420可根据提醒设备140所采用的显示方式，例如显示该提醒消息的程序，对所述提醒消息进行结构化，使得所述提醒消息能够由显示的程序进行读取并通过显示设备140进行显示。
接着，提醒设备140接收到结构化的提醒消息后，按照提醒消息所属的不同类型，设置不同的提醒标记，从而对用户进行提醒。其中，所述提醒标记可为提醒消息对应的类型名称、类型描述等。
其中，提醒设备140可调用其它已有应用程序的界面进行提醒，例如短信、记事本，也可重新绘制提醒界面进行提醒。参考图9，根据某些实施方式，提醒设备140显示短信接收界面，并在短信的邻近位置设置提醒标记，实现对用户进行提醒。参考图10，根据某些实施方式，提醒设备140可汇总所有的提醒消息，在其新绘制的提醒界面，对用户进行提醒。在重新绘制的提醒界面中，提醒设备140可显示提醒消息以及提醒标记，也可仅显示提醒标记。
参考图11，处理器220进一步包括监测执行设备520，提醒设备140进一步包括显示设备510。在某些实施方式中，显示设备510将提醒消息以视觉形式通过屏幕显示出来。在其它的实施方式中，显示设备510还可以是播放设备或振动设备等，以音频或其它形式对用户进行提醒。
根据某些实施方式，监测执行设备520检测用户的进一步输入，获取与用户所选择的提醒消息对应的动作并执行该动作。在某些实施方式中，监测执行设备520可包括输入监测设备521、动作获取设备522和动作执行设备523。其中，输入监测设备521对用户是否存在输入进行检测，例如当检测到用户通过手指、点触笔等输入装置在屏幕上进行点触或勾选或其它选择提醒消息的输入操作时，记录用户的输入信号并将所记录的用户输入信号传输至动作获取设备522。动作获取设备522根据接收到的用户输入信号，例如用户点触的坐标或用户点触的像素，判断用户所选择的提醒消息，并获取与该提醒消息对应的动作。接着，动作执行设备523执行经由动作获取设备522所获取的动作，例如打开关联的链接、图片、文本等。
根据某些实施方式，参考图12，当输入监测设备521监测到用户对某提醒消息进行了选择时，将用户的输入信号发送至动作获取设备522。动作获取设备522对用户的输入信号进行判断，获取用户选择的提醒消息为“快递”类提醒消息，并且从数据库330中获取与该提醒消息对应的动作以及动作数据。例如，与该“快递”类型提醒消息对应的动作为“打开’快递查 询’的链接”，且关联的动作数据为快递单号，则动作获取设备522将相关的链接和动作数据发送至动作执行设备523。动作执行设备523进一步根据该动作数据打开该链接进行查询，并将查询结果界面显示给用户。
根据某些实施方式，参考图13，当动作获取设备522根据输入监测设备521监测到的用户输入信号，判断出用户选择的提醒消息为“购物”类提醒消息，并且从数据库330中获取与该提醒消息对应的动作以及动作数据。例如，与该“购物”类型提醒消息对应的动作为“打开所购物品的链接”，且关联的动作数据为团购券图片，则动作获取设备522将相关的图片和动作数据发送至动作执行设备523。动作执行设备523进一步根据该动作数据打开该图片文件，将对应的团购券图片显示给用户。
根据某些实施方式，动作执行设备523还可包括进一步对用户输入进行选择。参考图14，当动作获取设备522获取到用户选择的提醒消息为“票务”类提醒消息，并且从数据库330中获取与该提醒消息对应的动作以及动作数据。例如，与该“票务”类型提醒消息对应的动作为“打开’附近取票点’的链接”“打开‘目的地’的链接”“打开‘接机’的链接”，且关联的动作数据为“当前所在地‘上海徐汇区”’“目的地‘北京”’，则动作获取设备522将相关的链接和动作数据发送至动作执行设备523。动作执行设备523将这三个动作显示给用户，并检测用户的输入信号。当动作执行设备523检测到用户的输入信号是选择了“目的地”时，根据关联的动作数据“目的地‘北京”’以及“打开‘目的地’的链接”，打开“北京”的相关链接，并将结果界面显示给用户。
参考图15，根据本发明的某些实施方式，提供了一种智能提醒方法，包括：步骤S1，获取所接收到的推送消息，其中，所述推送消息中包含提醒消息，所述提醒消息中包含分类关联特征数据；步骤S2，根据所述分类关联特征数据，对所述推送消息进行过滤，获取提醒消息，并且将所述提醒消息标定为不同的类型，其中每一种提醒消息的类型与至少一种动作相关联；步骤S3，根据所述提醒消息的不同类型，执行与所述提醒消息的类型相关联的动作。
根据某些实施方式，在安卓系统中，可通过对系统短信进行监听获取推送的短信。
当获取推送消息后，根据分类关联特征数据在一种实施方式中，对所述推送消息进行过滤，获取提醒消息，接着，对过滤后的提醒消息进行标定，根据分类关联特征数据，将提醒消息标定为不同的类型。例如，对于短信而言，根据短信发送方号码、短信内容、短信基站服务中心代码等，将短信标定为不同的类型。在另一种实施方式中，根据分类关联特征数据，对所述推送消息进行分类，并且将不属于任何设定类型的推送消息标定为非提醒消息。
具体来说，在一种实施方式中，对推送消息中进行检测，当所述推送消息包括指示设定 类型的所述分类关联特征数据时，将所述推送消息标定为该设定类型的提醒消息。在另一种实施方式中，基于混合类型的所述分类关联特征数据，通过计算所述推送消息属于不同类型的概率，将所述推送消息标定为属于概率最大的类型的提醒消息。在又一种实施方式中，根据对推送消息的内容进行语义分析，将所述推送消息标定为该设定类型的提醒消息。在又一种实施方式中，采用正则表达式对推送消息的内容中进行匹配，判断是否存在与设定类型的分类关联特征数据相同或相近的内容。
根据某些实施方式，步骤S2还进一步包括更新步骤，对所述分类关联特征数据与标定类型之间的对应关系进行更新，或者对所述标定类型关联的动作进行更新。
获取提醒消息以及其关联的分类之后，执行与其类型关联的动作。具体来说，将提醒消息提示给用户后，进一步检测用户的输入，并根据用户的输入判断所选择的提醒消息；获取与所选提醒消息关联的动作以及动作数据，并执行该关联动作。
在一种实施方式中，获取提醒消息之后还包括：对该提醒消息进行结构化，并按照该提醒消息的不同类型，以设定的顺序和提醒方式对用户进进行提醒。
相较于现有技术，本发明充分考虑了提醒消息的特点，根据分类关联特征数据对推送消息进行处理，从中获取提醒消息，并进一步将提醒消息划分为不同的类型，并根据用户的输入，针对不同类型执行对应的动作，从而实现对推送消息的智能提醒。
以上对本发明的具体实施例进行了描述。需要理解的是，本发明并不局限于上述特定实施方式，本领域技术人员可以在权利要求的范围内做出各种变形或修改，这并不影响本发明的实质内容。

Claims (16)

  1. 一种智能提醒方法，包括：
    获取电子设备终端接收到的推送消息，其中所述推送消息包含提醒消息，所述提醒消息中包含分类关联特征数据；
    根据所述分类关联特征数据，对所述推送消息进行过滤，获取提醒消息，并且将所述提醒消息标定为不同的类型，其中每一种提醒消息的类型与至少一种动作相关联；
    根据所述提醒消息的不同类型，执行与所述提醒消息的类型相关联的动作。
  2. 如权利要求1所述的智能提醒方法，其特征在于，所述执行与所述提醒消息的类型相关联的动作包括：检测是否存在对所述提醒消息的触发信号；当检测到存在触发信号时，获取与所述提醒消息的类型关联的动作；执行所述关联的动作。
  3. 如权利要求2所述的智能提醒方法，其特征在于，所述执行与提醒消息的类型关联的动作还包括：根据所述触发信号选择对应的关联动作。
  4. 如权利要求1或2所述的智能提醒方法，其特征在于，所述执行与提醒消息的类型相关联的动作包括：对所述提醒消息进行结构化，并按照所述提醒消息的不同类型，以设定的顺序和提醒方式对用户进行提醒。
  5. 如权利要求1所述的智能提醒方法，其特征在于，所述将提醒消息标定为不同的类型包括：对所述推送消息进行检测，当所述推送消息包括指示设定类型的所述分类关联特征数据时，将所述推送消息标定为该设定类型的提醒消息。
  6. 如权利要求1所述的智能提醒方法，其特征在于，所述将提醒消息标定为不同的类型包括：基于混合类型的所述分类关联特征数据，通过计算所述推送消息属于不同类型的概率，将所述推送消息标定为属于概率最大的类型的提醒消息。
  7. 如权利要求1所述的智能提醒方法，其特征在于，所述将提醒消息标定为不同的类型进一步还包括：执行更新步骤，对所述分类关联特征数据与标定类型之间的对应关系进行更新，或者对所述标定类型关联的动作进行更新。
  8. 如权利要求1或4所述的智能提醒方法，其特征在于，所述分类关联特征数据包括：消息文本、消息发送方信息、短信基站服务中心代码。
  9. 如权利要求1所述的智能提醒方法，其特征在于，所述推送消息为手机短信。
  10. 一种智能提醒系统，其特征在于，包括：
    消息接收设备，用于接收与设定账号关联的推送消息；
    检测设备，用于根据所接收的推送消息中的分类关联特征数据，对推送消息进行检测，获取提醒消息，并将所述提醒消息标定为不同的类型，其中每一类型的提醒消息与至少一种动作相关联；
    提醒设备，用于根据所述提醒消息对用户进行提醒。
  11. 如权利要求10所述的智能提醒系统，其特征在于，所述检测设备进一步包括：分类设备，适于根据所述分类关联特征数据，对所述推送消息进行检测和分类。
  12. 如权利要求11所述的智能提醒系统，其特征在于，所述分类设备包括第一分类器或第二分类器或其组合，其中，所述第一分类器对所述推送消息进行检测，当检测到所述推送消息包含设定的分类关联特征数据时，将该推送消息标定为与该设定的分类关联特征数据相关联的类型的提醒消息；所述第二分类器提取所述推送消息中的分类关联特征数据，并且通过计算所述推送消息分别属于各个不同类型的概率，将所述推送消息标定为概率最大的类型的提醒消息。
  13. 如权利要求11所述的智能提醒系统，其特征在于，所述检测设备进一步包括更新设备，适于对所述分类关联特征数据与标定类型之间的对应关系进行更新，或者对所述标定类型关联的动作进行更新。
  14. 如权利要求11所述的智能提醒系统，其特征在于，所述检测设备进一步包括：结构化设备，适于对所述提醒消息进行结构化，形成设定结构的数据形式，其中，所述设定结构包含所述提醒消息对应的类型。
  15. 如权利要求所述的智能提醒系统，其特征在于，所述检测设备进一步包括：监测执行设备，适于监测用户的进一步输入，获取与用户所选择的提醒消息对应的动作并执行该动作。
  16. 如权利要求所述的智能提醒系统，其特征在于，所述检测设备进一步包括：过滤设备，适于根据分类关联特征数据，对推送消息进行过滤，获得提醒消息；标定设备，适于根据分类关联特征数据对所述提醒消息进行标定，将所述提醒消息划分为不同的类型，其中每一种类型与至少一种设定的动作相关联。
PCT/CN2015/089909 2014-09-18 2015-09-18 智能提醒方法、系统和装置 WO2016041517A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US15/512,824 US20170286912A1 (en) 2014-09-18 2015-09-18 Intelligent reminder methods, systems and apparatuses
EP15841242.9A EP3211925A4 (en) 2014-09-18 2015-09-18 Smart reminding method, system and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410477689.7 2014-09-18
CN201410477689.7A CN104507165B (zh) 2014-09-18 2014-09-18 智能提醒方法、系统和装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/463,188 Continuation-In-Part US20170301488A1 (en) 2014-09-18 2017-03-20 Key Switch, Key Unit and Computer Keyboard

Publications (1)

Publication Number Publication Date
WO2016041517A1 true WO2016041517A1 (zh) 2016-03-24

Family

ID=52948864

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/089909 WO2016041517A1 (zh) 2014-09-18 2015-09-18 智能提醒方法、系统和装置

Country Status (4)

Country Link
US (1) US20170286912A1 (zh)
EP (1) EP3211925A4 (zh)
CN (1) CN104507165B (zh)
WO (1) WO2016041517A1 (zh)

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104507165B (zh) * 2014-09-18 2019-06-28 上海触乐信息科技有限公司 智能提醒方法、系统和装置
CN105578431B (zh) * 2016-01-20 2019-09-06 北京智驾互联信息服务有限公司 提醒消息发送装置及方法
US11025779B1 (en) 2016-04-22 2021-06-01 Wells Fargo Bank, N.A. Automated payment reminders
CN106211106A (zh) * 2016-06-29 2016-12-07 北京小米移动软件有限公司 基于短信的信息推荐方法和装置
CN105933908A (zh) * 2016-07-01 2016-09-07 北京奇虎科技有限公司 信息识别处理方法和装置、移动终端、服务器及系统
CN109005517B (zh) * 2017-06-06 2022-05-27 阿里巴巴集团控股有限公司 活动提醒方法、活动提醒消息生成方法及装置
CN107172303A (zh) * 2017-07-07 2017-09-15 珠海市小源科技有限公司 短信分类方法及其装置
CN109429190A (zh) * 2017-08-25 2019-03-05 珠海格力电器股份有限公司 一种短信分类显示方法及装置
US10896620B2 (en) * 2018-06-07 2021-01-19 Microsoft Technology Licensing, Llc Programmable interface for automated learning refreshers
CN108805535A (zh) * 2018-06-29 2018-11-13 广州科鑫信息科技有限公司 一种综合智慧办公接入信息分类管理方法
CN109067864B (zh) * 2018-07-25 2021-03-05 网易（杭州）网络有限公司 通知消息推送方法、装置及电子设备
CN109660603A (zh) * 2018-11-27 2019-04-19 浙江万朋教育科技股份有限公司 一种实现即时或定时控制远程设备数据展示的方法
CN112291419A (zh) * 2019-07-12 2021-01-29 西安中兴新软件有限责任公司 终端消息提醒显示的方法、装置、终端及计算机可读存储介质
CN111079879A (zh) * 2019-12-31 2020-04-28 华高数字科技有限公司 一种垃圾分类的处理回收方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101631155A (zh) * 2008-07-15 2010-01-20 上海乐金广电电子有限公司 短信息分类提示的方法
EP2299710A1 (en) * 2009-09-22 2011-03-23 TELEFONAKTIEBOLAGET LM ERICSSON (publ) Differentiating IPTV notifications
CN102256012A (zh) * 2011-06-28 2011-11-23 中兴通讯股份有限公司 事件提醒设置的装置及方法
CN103167434A (zh) * 2013-03-04 2013-06-19 珠海金山办公软件有限公司 一种信息的多强度通知方法、系统及装置
CN103929537A (zh) * 2014-04-03 2014-07-16 北京深思数盾科技有限公司 基于不同级别信息的实时提醒方法
CN104184887A (zh) * 2014-07-29 2014-12-03 小米科技有限责任公司 消息提醒方法、装置及终端设备
CN104507165A (zh) * 2014-09-18 2015-04-08 上海触乐信息科技有限公司 智能提醒方法、系统和装置

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8024415B2 (en) * 2001-03-16 2011-09-20 Microsoft Corporation Priorities generation and management
CN1529524A (zh) * 2003-09-26 2004-09-15 移动通讯中的短消息分类方法
US20130275899A1 (en) * 2010-01-18 2013-10-17 Apple Inc. Application Gateway for Providing Different User Interfaces for Limited Distraction and Non-Limited Distraction Contexts
US8868661B2 (en) * 2008-10-08 2014-10-21 Verizon Patent And Licensing Inc. Message management based on metadata
US8793254B2 (en) * 2011-08-18 2014-07-29 Nicholas H. Evancich Methods and apparatus for classifying content
CN102572108A (zh) * 2012-01-31 2012-07-11 盘丝无限(北京)科技有限公司 一种优化手机消息服务的方法和系统
US8977304B2 (en) * 2012-04-27 2015-03-10 Sony Corporation Systems and methods for prioritizing messages on a mobile device
CN103533152B (zh) * 2013-04-10 2016-08-03 Tcl集团股份有限公司 一种移动终端的短信处理方法及系统
CN103491245A (zh) * 2013-09-06 2014-01-01 天脉聚源(北京)传媒科技有限公司 一种存储短信的方法、装置和终端设备
US20150186494A1 (en) * 2013-12-31 2015-07-02 Google Inc. Systems and methods for message categorization management
US20160086116A1 (en) * 2014-07-27 2016-03-24 Supriya Rao Method and system of an automatically managed calendar and contextual task list

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101631155A (zh) * 2008-07-15 2010-01-20 上海乐金广电电子有限公司 短信息分类提示的方法
EP2299710A1 (en) * 2009-09-22 2011-03-23 TELEFONAKTIEBOLAGET LM ERICSSON (publ) Differentiating IPTV notifications
CN102256012A (zh) * 2011-06-28 2011-11-23 中兴通讯股份有限公司 事件提醒设置的装置及方法
CN103167434A (zh) * 2013-03-04 2013-06-19 珠海金山办公软件有限公司 一种信息的多强度通知方法、系统及装置
CN103929537A (zh) * 2014-04-03 2014-07-16 北京深思数盾科技有限公司 基于不同级别信息的实时提醒方法
CN104184887A (zh) * 2014-07-29 2014-12-03 小米科技有限责任公司 消息提醒方法、装置及终端设备
CN104507165A (zh) * 2014-09-18 2015-04-08 上海触乐信息科技有限公司 智能提醒方法、系统和装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3211925A4 *

Also Published As

Publication number Publication date
EP3211925A4 (en) 2018-04-25
CN104507165A (zh) 2015-04-08
EP3211925A1 (en) 2017-08-30
CN104507165B (zh) 2019-06-28
US20170286912A1 (en) 2017-10-05

Similar Documents

Publication Publication Date Title
WO2016041517A1 (zh) 智能提醒方法、系统和装置
US11729131B2 (en) Systems and methods for displaying unseen labels in a clustering in-box environment
CN109769065B (zh) 消息显示方法、装置、移动终端及存储介质
US20200358735A1 (en) System and method for entity detection in machine generated messages
CN106557330B (zh) 移动终端系统通知栏通知信息处理方法、装置和移动终端
CN106302094B (zh) 消息推送方法及装置
US10567325B2 (en) System and method for email message following from a user's inbox
US20170222960A1 (en) Spam processing with continuous model training
CN110781286A (zh) 用于生成推荐消息的序列到序列的分类模型
US10674321B2 (en) Lost device detection using geospatial location data
CN109690479A (zh) 一种关联通知消息的方法、装置及移动终端
US20180248830A1 (en) Methods for generating message notifications
KR102460293B1 (ko) 거래 피드백의 처리
AU2017235911A1 (en) Item recognition
KR20230007264A (ko) 이메일 서비스 제공 방법 및 장치
WO2017016384A1 (zh) 短信处理方法、信息处理方法及装置、移动终端、存储介质
KR102381436B1 (ko) 웹과 관련된 데이터에서 아이템을 검출하기 위한 전자 장치 및 방법
CN112200617B (zh) 业务提供方法、装置和电子设备
CN108710659A (zh) 一种信息分类方法和终端设备
CN109739998B (zh) 一种信息分类方法及装置
CA2938042C (en) Selecting a communication mode
CN109451447A (zh) 一种鉴别垃圾信息的方法、装置、存储介质和设备
US20230056075A1 (en) Random forest predictive spam detection
CN111726448A (zh) 号码标记方法、装置、计算机存储介质及电子设备
CN111260508A (zh) 一种数据处理方法、装置、设备及存储介质

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: 15841242

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 15512824

Country of ref document: US

REEP Request for entry into the european phase

Ref document number: 2015841242

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2015841242

Country of ref document: EP