WO2014036854A1 - 云推送的方法、系统和装置 - Google Patents
云推送的方法、系统和装置 Download PDFInfo
- Publication number
- WO2014036854A1 WO2014036854A1 PCT/CN2013/078872 CN2013078872W WO2014036854A1 WO 2014036854 A1 WO2014036854 A1 WO 2014036854A1 CN 2013078872 W CN2013078872 W CN 2013078872W WO 2014036854 A1 WO2014036854 A1 WO 2014036854A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- user
- mobile terminal
- time
- push
- usage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/2866—Architectures; Arrangements
- H04L67/30—Profiles
- H04L67/303—Terminal profiles
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/535—Tracking the activity of the user
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
Definitions
- the present invention relates to the field of computer science and technology, and in particular, to a method, system and device for cloud push. Background technique
- the push service passively accepts the request to actively send the message, and the server actively sends the message to the client, and the client chooses to receive.
- the existing technologies for implementing push services mainly include the following two methods:
- a technical solution for real-time push services Regardless of the actual usage of the user on the mobile terminal, when the information arrives at the server, the server immediately pushes the information to the user.
- This method is suitable for types of push information with high timeliness, such as weather forecast, traffic conditions, etc. Users expect to receive update information at the first time.
- Personalized push services generally focus on content personalization, collecting information and preferences of users in mobile terminals.
- the server pushes for technology-type information, such as advertisements for technology products. It is also possible to accumulate a large amount of original data by collecting information of the user using the social network or the browsing information of the user using the browser, and cluster the information out of the user's preference or content of interest, for example, implementing personalized recommendation like advertisement push.
- Push service can also be carried out according to the geographical location, and the discount information of the surrounding merchants is pushed according to the geographical location where the user moves the device. In general, however, such push services are based on the user's personal information, identifying the user's individual needs and optimizing the push content.
- the scene of the mobile device user's use of the device also directly affects the operation taken by the user when receiving the push information. For example, when the user is lying on the sofa at home and browsing the news on the mobile phone, if there is push information, the possibility of the user viewing the message is more than normal. In the work, the probability of being on the road is great, so the effect of the push service is better.
- the push information needs to push the content that the user is interested in at the right time and in the scene.
- the prior art does not provide a good solution to this demand.
- the object of the present invention is to at least solve one of the above technical drawbacks.
- the first object of the present invention is to provide a cloud push method, which analyzes the time and scene of a user using a mobile terminal, provides a personalized customization service for the user, and pushes information for the user at an appropriate time and place, thereby improving The possibility that the information was clicked improved the user experience.
- a second object of the present invention is to provide a cloud push device.
- a third object of the present invention is to propose another method of cloud pushing.
- a fourth object of the present invention is to provide a mobile terminal.
- a fifth object of the present invention is to provide a cloud push system.
- an embodiment of the first aspect of the present invention discloses a method for cloud push, comprising the steps of: receiving a usage time and/or a usage scenario of a user recorded by a mobile terminal using the mobile terminal, and the mobile terminal Attribute tag; calculating a preferred push time for the user based on the usage time and/or usage scenario; obtaining a push to the user based on the attribute tag.
- a preferred push time is obtained, and the push time is personalized for the user according to the record of each user, and is optimized during the push time.
- the information pushed by the user takes into account the user's usage scenarios and usage habits. Try to push at the time when the user may read the information, thereby increasing the probability that the information is clicked and reading, and avoiding the real-time push may be pushed to the user when the user is inconvenient to read. The inconvenience caused has improved the user experience.
- the attribute tag of the mobile terminal includes the physical identity of the mobile terminal and/or the identity information of the user.
- the usage scenario includes location information of the mobile terminal, posture information when the user uses the mobile terminal, motion information when the user uses the mobile terminal, and the One or more of the light intensity information of the environment in which the mobile terminal is located.
- calculating a preferred push time for the user according to the usage time and/or a usage scenario specifically: calculating, according to the usage time and/or using a scenario, according to an optimization time slice algorithm The preferred push time of the user.
- the optimization time slice algorithm includes the following steps: obtaining usage time of the mobile terminal by the user in daily use, and marking and sorting the usage time; The length of time of the use time; calculating a weight value for each use time according to the length of time/or use scenario calculation.
- the method further includes: obtaining a call record of the user; if the ratio of the user's talk time to the use time is greater than a preset value during the use time, the use time is The corresponding weight value is set to zero.
- the method further includes: summarizing weight values of the usage time of the user in N days, to calculate a preferred push time for the user, where N is greater than
- An embodiment of the second aspect of the present invention discloses a cloud push device, including: a receiving module, configured to receive, by a mobile terminal, a usage time and/or a usage scenario of a user using the mobile terminal and attributes of the mobile terminal a calculation module, configured to calculate a preferred push time for the user according to the usage time and/or a usage scenario; a push module, configured to obtain, according to the attribute tag, push content pushed to the user, and The push content is pushed to the mobile terminal during a preferred push time.
- a preferred push time is obtained, and the push time is personalized for the user according to the record of each user, and is optimized during the push time.
- the information pushed by the user takes into account the user's usage scenarios and usage habits. Try to push at the time when the user may read the information, thereby increasing the probability that the information is clicked and reading, and avoiding the real-time push may be pushed to the user when the user is inconvenient to read. The inconvenience caused has improved the user experience.
- the attribute tag of the mobile terminal includes the physical identity of the mobile terminal and/or the identity information of the user.
- the usage scenario includes location information of the mobile terminal, posture information when the user uses the mobile terminal, motion information when the user uses the mobile terminal, and light intensity information of an environment in which the mobile terminal is located One or more of them.
- the computing module calculates a preferred push time for the user based on the usage time and/or usage scenario in accordance with an optimized time slice algorithm.
- the computing module obtains that the user uses the device every day. Determining the usage time of the mobile terminal, marking and sorting the usage time, calculating a length of time each time the usage time, and calculating a weight for each usage time according to the length of time/or using a scene calculation value.
- the calculating module is further configured to obtain a call record of the user, and if a ratio of the user's talk time to the use time is greater than a preset value in use time, The weight value corresponding to the usage time is set to zero.
- the calculation module aggregates the weight values of the usage time of the user in the N days to calculate a preferred push time for the user, where N is an integer greater than 1.
- An embodiment of the third aspect of the present invention discloses a method for cloud push, comprising the steps of: collecting usage time and/or usage scenario of a user using a mobile terminal; using the usage time and/or usage scenario and the mobile terminal
- the attribute tag is sent to the cloud server, so that the cloud server calculates a preferred push time for the user according to the usage time and/or a usage scenario, and obtains push content according to the attribute tag; and a receiving module, configured to receive the The push content pushed by the cloud server to the mobile terminal during the preferred push time.
- the usage time and/or usage scenario of the user is collected and transmitted to the cloud server, so that the cloud server can obtain the user's usage information, thereby making a judgment on the user's situation and proposing a more accurate service.
- An embodiment of the fourth aspect of the present invention discloses a mobile terminal, including: an acquisition module, configured to collect usage time and/or usage scenarios of a user; and a sending module, configured to use the usage time and/or usage scenario and move The attribute tag of the terminal is sent to the cloud server, so that the cloud server calculates a preferred push time for the user according to the usage time and/or a usage scenario, and obtains push content according to the attribute tag; and a receiving module, configured to receive The push content pushed by the cloud server to the mobile terminal during the preferred push time.
- the mobile terminal collects and sends the usage time and/or the usage scenario of the user to the cloud server, so that the cloud server can obtain the usage information of the user, thereby
- the situation makes a judgment and proposes a more precise service.
- the embodiment of the fifth aspect of the present invention discloses a cloud push system, comprising: a cloud push device disclosed in the embodiment of the second aspect of the present invention and a mobile terminal disclosed in the embodiment of the fourth aspect of the present invention .
- a preferred push time is obtained, and the push time is customized for the user according to the record of each user, and is presented to the user in the preferred push time.
- the pushed information takes into account the user's usage scenarios and usage habits. Try to push at the time when the user may read the information, thereby increasing the probability that the information is clicked and reading, and avoiding the real-time push may be pushed to the user when the user is inconvenient to read. The inconvenience has improved the user experience.
- FIG. 1 is a flowchart of a method for cloud push according to an embodiment of the present invention
- FIGS. 2 and 3 are schematic diagrams showing a state of use of a user according to an embodiment of the present invention.
- FIG. 4 is a schematic diagram of data transmission according to an embodiment of the present invention.
- FIG. 5 is a bar graph of user usage time according to an embodiment of the present invention.
- FIG. 6 is a schematic structural diagram of an apparatus for cloud push according to an embodiment of the present invention.
- FIG. 7 is a flow chart of a method for cloud push according to an embodiment of the present invention.
- FIG. 8 is a schematic structural diagram of a mobile terminal according to an embodiment of the present invention.
- FIG. 9 is a schematic structural diagram of a system for cloud push according to an embodiment of the present invention.
- the embodiments of the present invention are described in detail below, and the examples of the embodiments are illustrated in the drawings, wherein the same or similar reference numerals are used to refer to the same or similar elements or elements having the same or similar functions.
- the embodiments described below with reference to the drawings are intended to be illustrative of the invention and are not to be construed as limiting.
- Step S110 Receive a usage time and/or a use scenario of a user using the mobile terminal recorded by the mobile terminal and an attribute tag of the mobile terminal.
- the attribute tag of the mobile terminal includes the physical identity of the mobile terminal and/or the identity information of the user.
- the usage scenario includes one or more of location information of the mobile terminal, posture information when the user uses the mobile terminal, motion information when the user uses the mobile terminal, and light intensity information of the environment in which the mobile terminal is located.
- each time the user uses the mobile terminal information is collected and the information parameters defining the user's use of the mobile terminal are as follows:
- the device unique attribute tag (a) The device unique attribute tag (UID) consists of two parts:
- the physical standard of the mobile terminal is only : 3 ⁇ 4.
- the IMEI International Mobile Equipment Identity number
- IMSI International Mobile Subscriber Identification Number
- the user account is the user logo of the user on a service system platform, specifically the email address or the account of the website, and the mobile terminal may be used by multiple people, each person Log in with a different account.
- the user's identity information is optional. For those who do not set up login on the mobile terminal, the default user identity information can be used. The user is determined by the physical attributes of the mobile terminal and/or the identity information of the user, and the data record associated with it is used.
- Start using time point Identify each lighting of the screen as the starting point of a new user using the mobile terminal, and record the time point, marked as StartTime.
- End of use time point Each time the screen is turned off as the end point of the user's use of the mobile terminal once, the time point is recorded, and is marked as EndTime.
- Attitude information when the user uses the mobile terminal and the motion letter when the user uses the mobile terminal Information According to the sensor device on the mobile terminal, the position of the user when using the device is recognized, such as “standing”, “lying” and “sitting", etc., the position may also include other situations, techniques in the field The personnel can easily expand according to the needs of the user.
- the value of the motion information when the user uses the mobile terminal is, for example, "stationary” and “moving”, and the motion information can also include other values, for example, finely dividing the moving speed. Combine both the body position and the movement information and mark it as BodyStatus, for example: “Standing (still)", “Standing (moving)", “Lying (still)”, “Lying (moving)”, etc. Wait.
- the posture of the user when using the mobile terminal can be sensed by an acceleration sensor on the mobile terminal, such as an accelerometer and a gyroscope, to sense the azimuth, rotation angle, and tilt angle on the horizontal plane. It is possible to detect that the mobile terminal is in a state of positive vertical, vertical vertical, left horizontal, right horizontal, vertical, and downward.
- the direction sensor is used to determine the position of the user, for example, whether the user is standing or lying. When the user uses the device, the orientation angle of the device is different in different postures of the station and the lying.
- the direction sensor can distinguish between the two cases. As shown in Fig. 2 and Fig. 3, the judgment of the user's posture "standing” and “lying” is realized by the data returned by the direction sensor. Other postures can also be judged when the mobile terminal is in the vertical, vertical, left horizontal, right horizontal, vertical, and downward directions when the user uses the mobile terminal.
- the acceleration sensor can also be used to determine the motion state of the mobile terminal. Acceleration is a spatial vector. By measuring the components on its three axes, you can understand the motion state of the object and guess whether the user is still or moving. For push messages, when the user is at rest, such as lying on a sofa, they are generally more willing to view the new push message. Conversely, in the mobile state, users on the bus may find it inconvenient. However, the above examples are for illustrative purposes only, and are not absolute, such as on a subway, or when the user is more comfortable, it is easier to click on the push message. Because of the uncertainty of BodyStatus, the statistically optimized time slice algorithm is also used to optimize the push effect.
- Location information of the mobile terminal The location information when the user uses the mobile terminal is obtained by base station positioning or GPS (Global Position System) positioning, and the user's location can be inferred based on the time of the user using the mobile terminal. Marked as Place, the values are, for example, "fixed place", “indeterminate place”, etc.
- fixed place means that users often use mobile terminals in certain fixed geographical locations. For example, users may often use mobile terminals at home, in offices, and in restaurants. When a user uses a mobile terminal in a fixed location, they are generally more willing to receive new messages. "Uncertain place” means other locations than fixed places, such as when the user is on the move, or when going out, there is often no more More time and effort to browse push information.
- the method for judging the fixed location of the user is: recording the position coordinates of the mobile terminal at the same time in recording the user's use of the mobile terminal, which may be performed by base station positioning or GPS positioning. Then, according to the change of coordinates, it is identified whether the user is in a certain fixed position. For example, when the user uses the mobile phone at home, the coordinate value obtained each time should be very close, can be regarded as the same position, after collecting a large amount of position information, and then The recurring position coordinates are extracted based on the statistical rule, and these positions are used as a fixed place for the user.
- the specific value can be divided according to the light intensity, for example, the average is divided into four stages, each of which corresponds to “very bright”, “normally bright”, “dark” " , ⁇ “ “ , can also be divided according to the actual situation, for example, ⁇ "" and ⁇ "" occupies a large range of light intensity, etc., it should be understood that in the actual implementation process, according to the actual demand according to the light Multiple values.
- the ambient light sensor on the mobile terminal collects the light conditions of the location where the mobile terminal is located. The intensity of external light will have an impact on the user's use of the mobile terminal. For example, in the hot sun, the user generally does not click on the received push message, and the user may use the mobile terminal without lights at night, and the push received at this time. The message is more likely to be clicked by the user.
- FIG. 4 is an illustration of data transmission of the method.
- the user's mobile device that is, the mobile terminal, is sent to the cloud server in a batch process after the information is collected. It should be understood that it can also be sent to other servers or devices, and the batch data sent is marked as UserTable.
- the cloud server After receiving the data uploaded by the mobile terminal, stores the data in the database table corresponding to each user, and forms a mapping relationship by using the unique attribute flag of the user equipment.
- the data for each user in the table is independent, so each user's push strategy is also independent, thus enabling personalization.
- Table 1 Step S120 Calculate the preferred push time for the user according to the usage time and/or the usage scenario.
- the preferred push time for the user may be calculated based on the time of use and/or the usage scenario in accordance with the optimized time slice algorithm.
- the optimized time slice algorithm includes the following steps: Step S121: Obtain the usage time of the mobile terminal used by the user in daily, and mark and sort the usage time.
- Step S122 Calculate the length of time of each use time.
- Step S123 Calculate the weight value of each use time according to the length of time / or use the scene calculation.
- the optimized time slice algorithm is described below with a specific example. It can be understood that this example is for illustrative purposes only, and the embodiment of the present invention is not limited thereto.
- the data of the user using the mobile terminal is obtained from the database according to the unique attribute tag (UID) corresponding to the user using the mobile terminal.
- UID unique attribute tag
- TimeRecord contains the fields defined in Table 2:
- the value of the weight value ranges from 0 to 1. The larger the value, the more meaningful the usage time is and the better it is to perform the push.
- the default value of the weight value is k, and the growth and reduction are 0.1. Different field values in TimeRecord will affect the increase or decrease of the weight value, which is calculated as follows:
- the weight value corresponding to the use time is set to zero. For example, if the user's time all calls, it means that the usage time is not suitable for pushing information to the user.
- the light is generally bright, and the corresponding weight value is increased, which is brighter, and the corresponding weight value is lowered.
- each usage time can be represented as a block in FIG. 5.
- the horizontal axis of the graph represents time, and the vertical axis represents weight values.
- the area of each square represents the importance of the time of use. It is more suitable for information push during this usage time.
- the TimeRecord weight value indicates that the user is in a certain state during the usage time. For example, when the user uses the mobile terminal, the posture information is "lying” and the weight of "standing” is higher, because the user is generally physically and mentally lying while lying down. A more comfortable state, the probability of receiving and clicking on push messages is large.
- the usage time in FIG. 5 is expressed in a two-dimensional space in the form of a hash, and the area with the most overlap is counted, that is, the user's use time has been formed to form a certain rule, and the possibility that the user continues to use the mobile terminal during the use time is predicted. Higher.
- the program takes out all the data of Monday, and then hashes it into the above two-bit table, and calculates the area where the usage time overlaps and the area with the high time distribution density.
- the time range [Tn-1, ⁇ ] the sum of all time-use areas in this interval (the overlapping areas need to be repeatedly accumulated to calculate the area), denoted as ⁇ ( ⁇ -1, ⁇ ), in the total time axis [0,24
- the calculation is performed in steps of m minutes (for example, the value of m can be 5 minutes, 10 minutes, 20 minutes, 30 minutes, etc.), for example, the value is 5 minutes first, and the time axis is calculated for each step increment.
- the sum of the use time areas of each of the scattered intervals is then sorted from high to low for the use time area of each of the intervals, and a set of preferred push times can be obtained at a given step size.
- the step size m is then modified to iterate to obtain a set of preferred push times for the next step, for example 10 minutes.
- a set of multiple preferred push times can be obtained under all step value conditions, which can be understood as a time of use suitable for pushing information to the user, and a set of minimum values of the step size can be prioritized.
- the sorted usage time is stored in the database to form a user-specific preferred push schedule, as shown in Table 3.
- the mobile terminal can select the push timing based on the information in this table. Due to the user The status of the seven days of the week may be different, so it is necessary to calculate the mobile terminal usage for different days for the user to more accurately understand the user and perform different push policies according to different dates.
- the optimization time-of-use algorithm is used to complete the calculation of the user's original data. The result is: A record table that reflects the user's preferred push time, which records the user's 7-day-a-week, which time period within 24 hours is more likely to view the push information.
- a user uses a mobile phone in a fixed place between 8 and 9 o'clock every Saturday, but the frequency of screen clicks is lower than the frequency of clicks required by the average game.
- the user is likely to be browsing the web. Priority can be given to pushing messages to users during this time period.
- Step S130 Obtain the push content pushed to the user according to the attribute tag, and push the push content to the mobile terminal at the preferred push time.
- the push service scheduler can be launched to select the appropriate time to push information for the corresponding user based on the time period range in the preferred push time set.
- Each user saves a preferred push time table, from which the decision is made to schedule the execution of the push.
- the preferred push schedule content for each user is generally different, so different users will receive push messages at different points in time.
- the weight value of the corresponding usage time which is the weight value in TimeRecord. If the user does not view the message at the first time, the weight value is not increased. If the user does not view this message, the weight value of the corresponding usage time is lowered.
- the cloud push method collects the time situation of the user using the mobile terminal, and estimates the use scenario by using the time and location information of the mobile terminal, and calculates an optimal time period for pushing the information to the user through an algorithm, which is different.
- Users provide personalized customization services.
- Information Push service tries to avoid disturbing users, and chooses to arrive at a more comfortable and appropriate time for users, thereby improving user acceptance and click-through rate and improving user experience.
- the present invention also proposes a cloud push device.
- the cloud push device includes: a receiving module 11, a calculating module 12, and a pushing module 13.
- the receiving module 11 is configured to receive a usage time of the user using the mobile terminal and/or use the scene and a property tag of the mobile terminal recorded by the mobile terminal.
- the calculation module 12 is configured to calculate a preferred push time for the user based on the time of use and/or the usage scenario.
- the push module 13 is configured to obtain the push content pushed to the user based on the attribute tag, and push the push content to the mobile terminal within the preferred push time.
- the attribute tag of the mobile terminal includes a physical identifier of the mobile terminal and/or identity information of the user.
- the usage scenario includes one or more of location information of the mobile terminal, gesture information when the user uses the mobile terminal, a motion message when the user uses the mobile terminal, and light intensity information of the environment in which the mobile terminal is located.
- the user receives the usage time and/or usage scenario of the mobile terminal and the attribute tag of the mobile terminal, and collects and defines the user for each time the user uses the information collected by the mobile terminal.
- the information parameters for using the mobile terminal are as follows:
- the device unique attribute tag (a) The device unique attribute tag (UID) consists of two parts:
- the user account is the user logo of the user on a service system platform, specifically the email address or the account of the website, and the mobile terminal may be used by multiple people, each person Log in with a different account.
- the user's identity information is optional. For those who do not set up login on the mobile terminal, the default user identity information can be used. The user is determined by the physical attributes of the mobile terminal and/or the identity information of the user, and the data record associated with it is used.
- End of use time point Each time the screen is turned off as the end point of the user's use of the mobile terminal once, the time point is recorded, and is marked as EndTime.
- posture information when the user uses the mobile terminal and motion information when the user uses the mobile terminal According to the sensor device on the mobile terminal, the posture of the user when using the device, such as “standing”, “lying” and “Sit” and the like, the positional situation may also include other situations, and those skilled in the art can easily expand according to the needs of the user, and the motion information values when the user uses the mobile terminal are, for example, "stationary” and “mobile", exercise.
- the information may also include other values, such as a fine division of the speed of movement, and the like. Combine both the body position and the movement information and mark it as BodyStatus, for example: “Standing (still)", “Standing (moving)", “Lying (still)”,
- the posture can be sensed by an acceleration sensor on the mobile terminal, such as an accelerometer and a gyroscope (induction of the azimuth, rotation angle and tilt angle on the horizontal plane), so that the mobile terminal can be detected to be positive.
- an acceleration sensor on the mobile terminal such as an accelerometer and a gyroscope (induction of the azimuth, rotation angle and tilt angle on the horizontal plane), so that the mobile terminal can be detected to be positive.
- the direction sensor is used to determine the position of the user, for example, whether the user is standing or lying.
- the orientation angle of the device is different in different postures of the station and the lying.
- the direction sensor can distinguish between the two situations. As shown in Fig. 2 and Fig. 3, the data returned by the direction sensor realizes the judgment of the user's posture "standing" and "lying".
- Other postures can also be judged when the mobile terminal is in the vertical, vertical, left horizontal, right horizontal, vertical, and downward directions when the user uses the mobile terminal.
- the acceleration sensor can also be used to determine the motion state of the mobile terminal. Acceleration is a space vector. By measuring the components on its three axes, you can understand the motion state of the object and guess whether the user is still or moving. For push messages, when the user is at rest, such as lying on a sofa, they are generally more willing to view the new push message. Conversely, in the mobile state, users on the bus may find it inconvenient. However, the above examples are for illustrative purposes only, and are not absolute, such as on a subway, or when the user is more comfortable, it is easier to click on the push message. Because BodyStatus is uncertain, the statistics-based optimization time slice algorithm is also used to optimize the push effect.
- Location information of the mobile terminal The location information when the user uses the mobile terminal is obtained by base station positioning or GPS positioning, and combined with the time of the user using the mobile terminal, the location of the user can be inferred, marked as Place, and the value is, for example, " Fixed places", “indeterminate places”, etc.
- fixed place means that the user often uses the mobile terminal in certain fixed geographical locations, such as the user may often use the mobile terminal at home, office, or restaurant.
- fixed location means that other locations than the fixed location, such as when the user is on the move or out, often do not have more time and effort to view the push information.
- the method for judging the fixed location of the user is: recording the position coordinates of the mobile terminal at the same time in recording the user's use of the mobile terminal, which may be performed by base station positioning or GPS positioning. Then, according to the change of coordinates, it is identified whether the user is in a certain fixed position. For example, when the user uses the mobile phone at home, the coordinate value obtained each time should be very close, can be regarded as the same position, after collecting a large amount of position information, and then The recurring position coordinates are extracted based on the statistical rule, and these positions are used as a fixed place for the user.
- the specific value can be divided according to the light intensity, for example, the average is divided into four stages, each of which corresponds to “very bright”, “normally bright”, “dark” " , ⁇ “ “ , can also be divided according to the actual situation, for example, ⁇ "" and ⁇ "" occupies a large range of light intensity, etc., it should be understood that in the actual implementation process, according to the actual demand according to the light Multiple values.
- the ambient light sensor on the mobile terminal collects the light conditions of the location where the mobile terminal is located. The intensity of external light will have an impact on the user's use of the mobile terminal. For example, in the hot sun, the user generally does not click on the received push message, and the user may use the mobile terminal without lights at night, and the push received at this time. The message is more likely to be clicked by the user.
- each time information collection can be represented as one line record, and the receiving module 11 receives each piece of information recorded by the mobile terminal.
- the calculation module 12 calculates a preferred push time for the user based on the time of use and/or the usage scenario in accordance with the optimized time slice algorithm.
- the calculation module 12 obtains the usage time of the user using the mobile terminal in daily, marks and sorts the usage time, calculates the length of time of each use time, and calculates each use according to the length of time/or using the scene calculation.
- the weight value of time is the weight value of time.
- the data of the user using the mobile terminal is obtained from the database according to the unique attribute tag corresponding to the user using the mobile terminal.
- TimeRecord contains the fields defined in Table 2.
- the value of the weight value ranges from 0 to 1. The larger the value, the more meaningful the usage time is and the better it is to perform the push.
- the default value of the weight value is k, and the growth and reduction are 0.1. Different field values in TimeRecord will affect the increase or decrease of the weight value, which is calculated as follows:
- the weight value corresponding to the use time is set to zero. For example, if the user's time all calls, it means that the usage time is not suitable for pushing information to the user.
- R4 user's attitude information when using the mobile terminal When the value is "lying”, the corresponding weight value is added, and “Standing (moving)" lowers the corresponding weight value.
- the light is generally bright, and the corresponding weight value is increased, which is brighter, and the corresponding weight value is lowered.
- each usage time can be represented as a block in FIG. 5.
- the horizontal axis of the graph represents time, and the vertical axis represents weight values.
- the area of each square represents the importance of the time of use. It is more suitable for information push during this usage time.
- the TimeRecord weight value indicates that the user is in a certain state during the usage time. For example, when the user uses the mobile terminal, the posture information is "lying” and the weight of "standing” is higher, because the user is generally physically and mentally lying while lying down. A more comfortable state, the probability of receiving and clicking on push messages is large.
- the usage time in FIG. 5 is expressed in a two-dimensional space in the form of a hash, and the area with the most overlap is counted, that is, the user is considered to use time. A certain law has been formed, and it is highly probable that the user will continue to use the mobile terminal during this usage time.
- the program takes out all the data of Monday, and then hashes it into the above two-bit table, and calculates the area where the usage time overlaps and the area with the high time distribution density.
- the time range [ ⁇ - ⁇ , ⁇ ] the sum of all the time-use areas in this interval (the overlapping areas need to be repeatedly accumulated to calculate the area), denoted as ⁇ ( ⁇ -1, ⁇ ), in the total time axis [0,24
- the calculation is performed in steps of m minutes (for example, the value of m can be 5 minutes, 10 minutes, 20 minutes, 30 minutes, etc.), for example, the value is 5 minutes first, and the time axis is calculated for each step increment.
- the sum of the use time areas of each of the scattered intervals is then sorted from high to low for the use time area of each of the intervals, and a set of preferred push times can be obtained at a given step size.
- the step size m is then modified to iterate to obtain a set of preferred push times for the next step, for example 10 minutes.
- a set of multiple preferred push times can be obtained under all step value conditions, which can be understood as a time of use suitable for pushing information to the user, and a set of minimum values of the step size can be prioritized.
- the sorted usage time is stored in the database to form a user-specific preferred push time table, as shown in Table 3.
- the mobile terminal can select the push timing based on the information in this table. Since the user may have different status in seven days of the week, it is necessary to calculate the usage of the mobile terminal for different days for the user to more accurately understand the user and perform different push policies according to different dates.
- the optimal use time algorithm completes the calculation of the user's original data. The result is:
- the response user's preferred push time record table records which time periods within 24 hours of the user are more likely to view the push information within 24 hours per day.
- a user uses a mobile phone in a fixed place between 8 and 9 o'clock every Saturday, but the frequency of screen clicks is lower than the frequency of clicks required by the average game.
- the user is likely to be browsing the web. Priority can be given to pushing messages to users during this time period.
- the push module 13 may initiate a push service scheduler to select an appropriate timing to push information for the corresponding user based on the range of time periods in the preferred push time set.
- a push service scheduler to select an appropriate timing to push information for the corresponding user based on the range of time periods in the preferred push time set.
- Each user saves the preferred push schedule, according to which table determines when to schedule execution push give away.
- the preferred push schedule content for each user is generally different, so different users will receive push messages at different points in time.
- the push is considered to be a direct hit, and the weight value of the corresponding use time, that is, the weight value in TimeRecord, needs to be increased. If the user does not view the message at the first time, the weight value is not increased. If the user does not view this message, the weight value of the corresponding usage time is lowered.
- the actual situation feedback can further understand the user's usage habits and optimize the push time.
- a preferred push time is obtained, and the push time is personalized for the user according to the record of each user, and is optimized during the push time.
- the information pushed by the user takes into account the user's usage scenarios and usage habits. Try to push at the time when the user may read the information, thereby increasing the probability that the information is clicked and reading, and avoiding the real-time push may be pushed to the user when the user is inconvenient to read. The inconvenience caused has improved the user experience.
- the present invention also proposes another method of cloud push.
- a method of cloud push according to an embodiment of the present invention is described below with reference to FIG. 7, including the following steps:
- S210 Collect usage time and/or usage scenario of the user using the mobile terminal.
- the usage scenario includes one or more of location information of the mobile terminal, posture information when the user uses the mobile terminal, motion information when the user uses the mobile terminal, and light intensity information of the environment in which the mobile terminal is located.
- S220 Send the usage time and/or the usage scenario and the attribute tag of the mobile terminal to the cloud server, so that the cloud server calculates the preferred push time for the user according to the usage time and/or the usage scenario and obtains the push content according to the attribute tag.
- the attribute tag of the mobile terminal includes the physical identity of the mobile terminal and/or the identity information of the user.
- S230 Receive the push content that the cloud server pushes to the mobile terminal during the preferred push time.
- information is collected and the information parameters defining the user's use of the mobile terminal are as follows:
- the device unique attribute tag (a) The device unique attribute tag (UID) consists of two parts:
- the user account is the user logo of the user on a service system platform, specifically the email address or the account of the website, and the mobile terminal may be used by multiple people, each person Log in with a different account.
- the identity information of the user is optional.
- the login that is not set on the mobile terminal the default user identity information may be used.
- the user is determined by the physical attributes of the mobile terminal and/or the identity information of the user, and uses the data records associated therewith.
- Start using time point Identify each lighting of the screen as the starting point of a new user using the mobile terminal, and record the time point, marked as StartTime.
- End of use time point Each time the screen is turned off as the end point of the user's use of the mobile terminal once, the time point is recorded, and is marked as EndTime.
- the posture information when the user uses the mobile terminal and the motion information when the user uses the mobile terminal According to the sensor device on the mobile terminal, the posture of the user when using the device, such as “standing", “lying” and “Sit” and the like, the positional situation may also include other situations, and those skilled in the art can easily expand according to the needs of the user, and the motion information values when the user uses the mobile terminal are, for example, "stationary” and “mobile", exercise.
- the information may also include other values, such as a fine division of the speed of movement, and the like. Combine both the body position and the movement information and mark it as BodyStatus, for example: “Standing (still)", “Standing (moving)”, “Lying (still)”, “Lying (moving)”, etc. Wait.
- the posture of the user when using the mobile terminal can be sensed by an acceleration sensor on the mobile terminal, such as an accelerometer and a gyroscope, to sense the azimuth, rotation angle, and tilt angle on the horizontal plane. It is possible to detect that the mobile terminal is in a state of positive vertical, vertical vertical, left horizontal, right horizontal, vertical, and downward.
- the direction sensor is used to determine the position of the user, for example, whether the user is standing or lying. When the user uses the device, the orientation angle of the device is different in different postures of the station and the lying.
- the direction sensor can distinguish between the two cases. As shown in Fig. 2 and Fig. 3, the judgment of the user's posture "standing” and “lying” is realized by the data returned by the direction sensor. Other postures can also be judged when the mobile terminal is in the vertical, vertical, left horizontal, right horizontal, vertical, and downward directions when the user uses the mobile terminal.
- the acceleration sensor can also be used to determine the motion state of the mobile terminal. Acceleration is a space vector. By measuring the components on its three axes, you can understand the motion state of the object and guess whether the user is still or moving. For push messages, when the user is at rest, such as lying on a sofa, they are generally more willing to view the new push message. Conversely, in the moving state Next, users on the bus may find it inconvenient. However, the above examples are for illustrative purposes only, and are not absolute, such as on a subway, or when the user is more comfortable, it is easier to click on the push message. Because BodyStatus is uncertain, the statistics-based optimization time slice algorithm is also used to optimize the push effect.
- Location information of the mobile terminal The location information when the user uses the mobile terminal is obtained by base station positioning or GPS (Global Position System) positioning, and the user's location can be inferred based on the time of the user using the mobile terminal. Marked as Place, the values are, for example, "fixed place", “indeterminate place”, etc.
- fixed place means that the user often uses the mobile terminal in certain fixed geographical locations, such as the user may often use the mobile terminal at home, office, or restaurant.
- fixed location means that other locations than the fixed location, such as when the user is on the move or out, often do not have more time and effort to view the push information.
- the method for determining the fixed location of the user is: recording the position coordinates of the mobile terminal at the same time in recording the user's use of the mobile terminal, and the method may be through base station positioning or GPS positioning. Then, according to the change of coordinates, it is identified whether the user is in a certain fixed position. For example, when the user uses the mobile phone at home, the coordinate value obtained each time should be very close, can be regarded as the same position, after collecting a large amount of position information, and then The recurring position coordinates are extracted based on the statistical rule, and these positions are used as a fixed place for the user.
- the specific value can be divided according to the light intensity, for example, the average is divided into four stages, each of which corresponds to “very bright”, “normally bright”, “dark” " , ⁇ “ “ , can also be divided according to the actual situation, for example, ⁇ "" and ⁇ "" occupies a large range of light intensity, etc., it should be understood that in the actual implementation process, according to the actual demand according to the light Multiple values.
- the ambient light sensor on the mobile terminal collects the light conditions of the location where the mobile terminal is located. The intensity of external light will have an impact on the user's use of the mobile terminal. For example, in the hot sun, the user generally does not click on the received push message, and the user may use the mobile terminal without lights at night, and the push received at this time. The message is more likely to be clicked by the user.
- FIG. 4 is an illustration of data transmission of the method.
- the user's mobile device that is, the mobile terminal, after the information is collected, is sent to the cloud server in a batch process, and it should be understood that it can also be sent to the mobile terminal.
- the bulk data sent by his server or device is marked as UserTable.
- the cloud server After receiving the data uploaded by the mobile terminal, stores the data in the database table corresponding to each user, and forms a mapping relationship by using the unique attribute flag of the user equipment.
- the data for each user in the table is independent, so each user's push strategy is also independent, thus enabling personalization.
- the usage time and/or usage scenario of the user is collected and transmitted to the cloud server, so that the cloud server can obtain the user's usage information, thereby making a judgment on the user's situation and proposing a more accurate service.
- the present invention also proposes a mobile terminal.
- the mobile terminal includes: an acquisition module 21, a transmitting module 22, and a receiving module 23.
- the acquisition module 21 is configured to collect usage time and/or usage scenarios of the user.
- the sending module 22 is configured to send the usage time and/or the usage scenario and the attribute tag of the mobile terminal to the cloud server, so that the cloud server calculates the preferred push time for the user according to the usage time and/or the usage scenario, and obtains the push content according to the attribute tag.
- the receiving module 23 is configured to receive the push content pushed by the cloud server to the mobile terminal during the preferred push time.
- the attribute tag of the mobile terminal includes a physical identifier of the mobile terminal and/or identity information of the user.
- the usage scenario includes one or more of location information of the mobile terminal, gesture information when the user uses the mobile terminal, a motion message when the user uses the mobile terminal, and light intensity information of the environment in which the mobile terminal is located.
- the information parameters specifically collected and defined by the user using the mobile terminal are as follows:
- the device unique attribute tag (a) The device unique attribute tag (UID) consists of two parts:
- the user account is the user logo of the user on a service system platform, specifically the email address or the account of the website, and the mobile terminal may be used by multiple people, each person Log in with a different account.
- the user's identity information is optional. For those who do not set up login on the mobile terminal, the default user identity information can be used. The user is determined by the physical attributes of the mobile terminal and/or the identity information of the user, and the data record associated with it is used.
- Start using time point Identify each lighting of the screen as the starting point of a new user using the mobile terminal, and record the time point, marked as StartTime.
- End of use time point recognizes the closing of the screen each time as the user uses the mobile terminal once End point, record the time point, marked as EndTime.
- posture information when the user uses the mobile terminal and motion information when the user uses the mobile terminal According to the sensor device on the mobile terminal, the posture of the user when using the device, such as “standing”, “lying” and “Sit” and the like, the positional situation may also include other situations, and those skilled in the art can easily expand according to the needs of the user, and the motion information values when the user uses the mobile terminal are, for example, "stationary” and “mobile", exercise.
- the information may also include other values, such as a fine division of the speed of movement, and the like. Combine both the body position and the movement information and mark it as BodyStatus, for example: “Standing (still)", “Standing (moving)", “Lying (still)”,
- the posture can be sensed by an acceleration sensor on the mobile terminal, such as an accelerometer and a gyroscope (induction of the azimuth, rotation angle and tilt angle on the horizontal plane), so that the mobile terminal can be detected to be positive.
- an acceleration sensor on the mobile terminal such as an accelerometer and a gyroscope (induction of the azimuth, rotation angle and tilt angle on the horizontal plane), so that the mobile terminal can be detected to be positive.
- the direction sensor is used to determine the position of the user, for example, whether the user is standing or lying.
- the orientation angle of the device is different in different postures of the station and the lying.
- the direction sensor can distinguish between the two situations. As shown in Fig. 2 and Fig. 3, the data returned by the direction sensor realizes the judgment of the user's posture "standing" and "lying".
- Other postures can also be judged when the mobile terminal is in the vertical, vertical, left horizontal, right horizontal, vertical, and downward directions when the user uses the mobile terminal.
- the acceleration sensor can also be used to determine the motion state of the mobile terminal. Acceleration is a spatial vector. By measuring the components on its three axes, you can understand the motion state of the object and guess whether the user is still or moving. For push messages, when the user is at rest, such as lying on a sofa, they are generally more willing to view the new push message. Conversely, in the mobile state, users on the bus may find it inconvenient. However, the above examples are for illustrative purposes only, and are not absolute, such as on a subway, or when the user is more comfortable, it is easier to click on the push message. Because of the uncertainty of BodyStatus, the statistically optimized time slice algorithm is also used to optimize the push effect.
- Location information of the mobile terminal The location information when the user uses the mobile terminal is obtained by base station positioning or GPS positioning, and the time of the user using the mobile terminal can be inferred.
- "fixed place” means that users often use mobile terminals in certain fixed geographical locations. For example, users may often use mobile terminals at home, in offices, and in restaurants. When a user uses a mobile terminal in a fixed location, they are generally more willing to receive new messages.
- Uncertain location means that other locations than the fixed location, such as when the user is on the move or out, often have less time and effort to view the push information.
- the method for determining the fixed location of the user is: recording the position coordinates of the mobile terminal at the same time in recording the user's use of the mobile terminal, and the method may be through base station positioning or GPS positioning. Then, according to the change of coordinates, it is identified whether the user is in a certain fixed position. For example, when the user uses the mobile phone at home, the coordinate value obtained each time should be very close, can be regarded as the same position, after collecting a large amount of position information, and then The recurring position coordinates are extracted based on the statistical rule, and these positions are used as a fixed place for the user.
- Light intensity information of the environment in which the mobile terminal is located According to the sensor device on the mobile terminal, the external light condition when the device is used is identified, and is marked as Light, and the values include, for example, "very bright” and “normally bright”. “dark”, dark”, etc., the specific value can be divided according to the light intensity, for example, the average is divided into four stages, each stage corresponding to "very bright", “normally bright”, “dark”, dark” It can also be divided according to the actual situation. For example, the range of light intensity occupied by "bright” and “dark” is relatively large. It should be understood that, in actual implementation, the light is divided into multiple values according to actual needs.
- the ambient light sensor on the mobile terminal collects the light conditions of the location where the mobile terminal is located.
- the intensity of external light will have an impact on the user's use of the mobile terminal. For example, in the hot sun, the user generally does not click on the received push message, and the user may use the mobile terminal without lights at night, and the push received at this time. The message is more likely to be clicked by the user.
- the collection of each piece of information of the acquisition module 21 can be represented as one line of record.
- the mobile terminal collects and sends the user's usage time and/or usage scenario to the cloud server, so that the cloud server can obtain the user's usage information, thereby making a judgment on the user's situation and proposing a more accurate service. .
- a cloud push system includes a cloud push device 110 according to any of the above-described embodiments of the present invention, and the mobile device described in any of the above embodiments. Terminal 120.
- the cloud push system collects a situation in which a user uses a mobile terminal, and estimates a usage scenario by using a time, a location, and the like information of the mobile terminal, and calculates an optimal time period for pushing information to the user through an algorithm, for different users.
- Personalized customization services are provided.
- Information push The service information push service tries to avoid disturbing the user as much as possible, and tries to arrive at a more comfortable and appropriate time point for the user, thereby improving the user's acceptance of the information and the click rate, and improving the user experience.
- the description of the terms “one embodiment”, “some embodiments”, “example”, “specific example”, or “some examples” and the like means specific features described in connection with the embodiment or example.
- a structure, material or feature is included in at least one embodiment or example of the invention.
- the schematic representation of the above terms does not necessarily mean the same embodiment or example.
- the particular features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples.
- the term “plurality” means two or more.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Telephonic Communication Services (AREA)
- Telephone Function (AREA)
- Mobile Radio Communication Systems (AREA)
Description
云推送的方法、 系统和装置
本发明涉及计算机科学技术领域, 特别涉及一种云推送的方法、 系统和 装置。 背景技术
传统互联网访问一般都是用户进行请求、服务器返回结果, 而推送服务 变被动接受请求为主动发送信息, 由服务器主动向客户发送信息, 客户选择 接收。
现有实现推送服务的技术主要包括如下两种方式:
实时推送服务的技术方案。 不考虑移动终端上用户的实际使用情况, 当 信息到达服务器时, 服务器即刻将信息推送给用户。 这种方式适用于时效性 要求较高的推送信息类型, 例如天气预报、 交通路况等信息, 用户期望在第 一时间收到更新信息。
个性化推送服务的技术方案。个性化推送服务一般集中于内容个性化方 面, 收集移动终端中用户偏好的信息并进行发送。 对于喜欢科技新闻类型的 用户, 服务器为其推送科技类型的信息, 例如科技产品的广告。 还可以通过 收集用户使用社交网络的信息或者用户使用浏览器的浏览信息,积累大量的 原始数据, 并将信息聚类出用户的偏好或感兴趣的内容, 例如实现像广告推 送这样的个性化推荐。 也可以根据地理位置进行推送服务, 根据用户移动设 备所处的地理位置, 推送周边商家的打折促销信息。 但是总体看来, 这类推 送服务都是根据用户的个人信息, 识别用户的个性化的需求, 从而对推送内 容进行优化。
但是, 在如何从众多用户信息中进行采集, 以及如何有效使用这些信息 的问题上, 目前的技术方案都没有充分考虑到移动设备的独特性质, 例如设
备移动性和用户使用手机时间段的碎片化。 用户使用手机的时间是碎片化 的, 并且从总体看来有一定规律, 例如用户可能早上起来会用 10分钟看微 博, 在上班路上可能会看 15分钟新闻, 晚上可能会花 30分钟深度阅读博客 等。现有的推送服务方案都没有充分利用这些特点来对推送服务的产品设计 进行改进, 使得推送的消息失去了在第一时间点获取用户关注的机会。 而一 旦失去第一时间点, 将会增大用户忽略推送信息的可能性, 特别是当用户的 信息中心中积累了太多新消息提示标记的情况下。
移动设备用户使用设备的场景也直接影响用户在接收到推送信息时采 取的操作, 比如当用户在家躺在沙发用手机浏览新闻时, 如果有推送信息过 来, 用户查看该消息的可能性会比平常在工作中, 在路上的时候概率大, 因 此推送服务的效果也就更好。
现有的移动终端, 比如 iPhone, Android手机, 大部分都有统一的消息 中心, 所有应用相关的消息提醒、 新消息等都会存储在消息中心。 在积累了 较多新消息时, 用户往往没有时间和精力逐一点击查看所有的新消息提示。 为了改善消息提醒杂乱而引起的视觉上的不适,用户可能会选择一次清除全 部新消息提醒, 此时服务器端推送的信息可能根本就不会被用户点击。
综上所述, 推送信息需要在合适的时间和场景下, 推送用户感兴趣的内 容。 而现有技术并没有对此需求提出很好的解决方案。
本发明的目的旨在至少解决上述技术缺陷之一。
为此, 本发明的第一个目的在于提出一种云推送的方法, 该方法分析用 户使用移动终端的时间和场景, 为用户提供个性化定制服务, 在合适的时间 地点为用户推送信息, 提高了信息被点击的可能性, 改善了用户体验。
本发明的第二个目的在于提出一种云推送的装置。
本发明的第三个目的在于提出另一种云推送的方法。
本发明的第四个目的在于提出一种移动终端。
本发明的第五个目的在于提出一种云推送的系统。
为达到上述目的, 本发明第一方面的实施例公开了一种云推送的方法, 包括以下步骤: 接收移动终端记录的用户使用所述移动终端的使用时间和 / 或使用场景以及所述移动终端的属性标记; 根据所述使用时间和 /或使用场 景计算针对所述用户的优选推送时间;根据所述属性标记获得向所述用户推 端。
在本发明的实施例中, 通过记录并分析用户的使用时间和 /或使用场景, 得出优选推送时间,优选推送时间根据每个用户的记录不同而为用户个性化 定制,在优选推送时间内向用户推送的信息考虑了用户的使用场景和使用习 惯, 尽量选择在用户可能阅读信息的时间进行推送, 从而提高信息被点击阅 读的几率, 避免了实时推送可能在用户不方便阅读时推送给用户带来的不 便, 改善了用户体验。
在本发明的一个实施例中,所述移动终端的属性标记包括上述移动终端 的物理标识和 /或所述用户的身份信息。
在本发明的一个实施例中, 所述使用场景包括所述移动终端的位置信 息、 所述用户使用所述移动终端时的姿态信息、 所述用户使用所述移动终端 时的运动信息和所述移动终端所处环境的光线强度信息中的一种或多种。
在本发明的一个实施例中, 根据所述使用时间和 /或使用场景计算针对 所述用户的优选推送时间, 具体包括: 依据最优化时间片算法根据所述使用 时间和 /或使用场景计算针对所述用户的优选推送时间。
在本发明的一个实施例中, 所述最优化时间片算法包括以下步骤: 获得 所述用户在每天中使用所述移动终端的使用时间,并对所述使用时间进行标 记和排序; 计算每次所述使用时间的时间长度; 根据所述时间长度 /或使用 场景计算计算每次所述使用时间的权重值。
在本发明的一个实施例中, 还包括: 获得所述用户的通话记录; 如果在 使用时间中, 所述用户的通话时间与所述使用时间之比大于预设值, 则将所 述使用时间对应的权重值置为零。
在本发明的一个实施例中, 还包括: 对所述用户在 N 天中的使用时间 的权重值进行汇总, 以计算针对所述用户的优选推送时间, 其中, N为大于
1的整数。
本发明第二方面的实施例公开了一种云推送的装置, 包括: 接收模块, 用于接收移动终端记录的用户使用所述移动终端的使用时间和 /或使用场景 以及所述移动终端的属性标记; 计算模块, 用于根据所述使用时间和 /或使 用场景计算针对所述用户的优选推送时间; 推送模块, 用于根据所述属性标 记获得向所述用户推送的推送内容,并在所述优选推送时间内将所述推送内 容推送至所述移动终端。
在本发明的实施例中, 通过记录并分析用户的使用时间和 /或使用场景, 得出优选推送时间,优选推送时间根据每个用户的记录不同而为用户个性化 定制,在优选推送时间内向用户推送的信息考虑了用户的使用场景和使用习 惯, 尽量选择在用户可能阅读信息的时间进行推送, 从而提高信息被点击阅 读的几率, 避免了实时推送可能在用户不方便阅读时推送给用户带来的不 便, 改善了用户体验。
在本发明的一个实施例中,所述移动终端的属性标记包括上述移动终端 的物理标识和 /或所述用户的身份信息。
所述使用场景包括所述移动终端的位置信息、所述用户使用所述移动终 端时的姿态信息、所述用户使用所述移动终端时的运动信息和所述移动终端 所处环境的光线强度信息中的一种或多种。
在本发明的一个实施例中,所述计算模块依据最优化时间片算法根据所 述使用时间和 /或使用场景计算针对所述用户的优选推送时间。
在本发明的一个实施例中,所述计算模块获得所述用户在每天中使用所
述移动终端的使用时间, 并对所述使用时间进行标记和排序, 并计算每次所 述使用时间的时间长度, 以及根据所述时间长度 /或使用场景计算计算每次 所述使用时间的权重值。
在本发明的一个实施例中,所述计算模块还用于获得所述用户的通话记 录, 且如果在使用时间中, 所述用户的通话时间与所述使用时间之比大于预 设值, 则将所述使用时间对应的权重值置为零。
在本发明的一个实施例中, 所述计算模块对所述用户在 N天中的使用 时间的权重值进行汇总, 以计算针对所述用户的优选推送时间, 其中, N为 大于 1的整数。
本发明第三方面的实施例公开了一种云推送的方法, 包括以下步骤: 采 集用户使用移动终端的使用时间和 /或使用场景; 将所述使用时间和 /或使用 场景以及所述移动终端的属性标记发送至云端服务器,以使所述云端服务器 根据所述使用时间和 /或使用场景计算针对所述用户的优选推送时间以及根 据所述属性标记获得推送内容; 接收模块, 用于接收所述云端服务器在所述 优选推送时间内向所述移动终端推送的所述推送内容。
在本发明的实施例中, 采集和发送用户的使用时间和 /或使用场景至云 端服务器, 使得云端服务器能够得到用户的使用信息, 从而对用户的情况做 出判断而提出更精确的服务。
本发明第四方面的实施例公开了一种移动终端, 包括: 采集模块, 用于 采集用户的使用时间和 /或使用场景; 发送模块, 用于将所述使用时间和 /或 使用场景以及移动终端的属性标记发送至云端服务器,以使所述云端服务器 根据所述使用时间和 /或使用场景计算针对所述用户的优选推送时间以及根 据所述属性标记获得推送内容; 接收模块, 用于接收所述云端服务器在所述 优选推送时间内向所述移动终端推送的推送内容。
在本发明的实施例中, 移动终端采集和发送用户的使用时间和 /或使用 场景至云端服务器, 使得云端服务器能够得到用户的使用信息, 从而对用户
的情况做出判断而提出更精确的服务。
本发明第五方面的实施例公开了一种云推送的系统, 包括: 本发明第二 方面的实施例公开的一种云推送的装置和本发明第四方面的实施例公开的 一种移动终端。
在本发明的实施例中, 通过分析计算用户的使用时间和 /或使用场景, 得出优选推送时间,优选推送时间根据每个用户的记录不同而为用户个性化 定制,在优选推送时间内向用户推送的信息考虑了用户的使用场景和使用习 惯, 尽量选择在用户可能阅读信息的时间进行推送, 从而提高信息被点击阅 读的几率, 避免了实时推送可能在用户不方便阅读时推送给用户带来的不 便, 改善了用户体验。
本发明附加的方面和优点将在下面的描述中部分给出,部分将从下面的 描述中变得明显, 或通过本发明的实践了解到。
本发明上述的和 /或附加的方面和优点从下面结合附图对实施例的描述 中将变得明显和容易理解, 其中:
图 1为根据本发明实施例的云推送的方法的流程图;
图 2和 3为根据本发明实施例的用户使用状态示意图;
图 4为根据本发明实施例的数据发送的示意图;
图 5为根据本发明实施例的用户使用时间条形图;
图 6为根据本发明实施例的云推送的装置的结构示意图;
图 7为根据本发明实施例的云推送的方法的流程图;
图 8为根据本发明实施例的移动终端的结构示意图;
图 9为根据本发明实施例的云推送的系统的结构示意图。
下面详细描述本发明的实施例, 所述实施例的示例在附图中示出, 其中 自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能 的元件。 下面通过参考附图描述的实施例是示例性的, 仅用于解释本发明, 而不能解释为对本发明的限制。
下面参考图 1描述根据本发明实施例的云推送的方法, 包括以下步骤: 步骤 S 110: 接收移动终端记录的用户使用移动终端的使用时间和 /或使 用场景以及移动终端的属性标记。
其中, 移动终端的属性标记包括上述移动终端的物理标识和 /或用户的 身份信息。使用场景包括移动终端的位置信息、 用户使用移动终端时的姿态 信息、用户使用移动终端时的运动信息和移动终端所处环境的光线强度信息 中的一种或多种。
在本发明的一个实施例中, 每次用户使用移动终端时,收集信息并定义 用户使用移动终端的信息参数如下:
( a )设备唯一属性标记 (UID)包括两部分内容:
(一) 、 移动终端的物理标只: :¾。手机的 IMEI ( International Mobile Equipment Identity number, 国际移动设备识另 ll码 )号码和 IMSI(International Mobile Subscriber Identification Number, 国际移动用户识别号码)号码等。
(二) 、 用户的身份信息: 例如用户账号, 用户账号是用户在某个服务 系统平台上的用户标志, 具体可以是 Email邮箱号或者网站的账号, 移动终 端可能会有多人使用, 每个人用不同的账号登陆。
用户的身份信息是可选的, 对于没有在移动终端上设置登陆的, 可以采 用默认用户身份信息的形式。 通过移动终端的物理属性和 /或用户的身份信 息来确定的用户, 并使用与之相关联的数据记录。
(b)开始使用时间点:将屏幕的每一次点亮识别为一次全新的用户使用移 动终端的起点, 记录下时间点, 标记为 StartTime。
(c)结束使用时间点:将屏幕每次的关闭识别为用户一次使用移动终端的 结束点, 记录时间点, 标记为 EndTime。
(d)通话时间记录: 获取用户在 [StartTime, EndTime] 之间内的通话时间 记录, 将用户打电话的行为单独识别, 标记为 CallTime。
(e)屏幕点击记录: 在 [StartTime, EndTime] 之间, 记录用户每次点击屏 幕的操作时间点, 标记为 ClickList。
(f) 用户使用移动终端时的姿态信息和用户使用移动终端时的运动信
息: 根据移动终端上的传感器装置, 识别出用户使用设备时的体位情况, 例 如 "站着" 、 "躺着" 和 "坐着" 等等, 体位情况还可以包括其他情况, 本 领域的技术人员很容易根据用户的需求进行扩展,用户使用移动终端时的运 动信息取值例如 "静止" 和 "移动中" , 运动信息还可以包括其他的取值, 例如, 对移动速度进行细致划分等。将体位情况和移动信息两者结合后标记 为 BodyStatus,例如: "站着 (静止) " , "站着 (移动中 ) " , "躺着 (静 止) " , "躺着 (移动中) " 等等。
用户使用移动终端时的姿态可以通过移动终端上的加速度传感器,如加 速计 (Accelerometer)和陀螺仪(Gyroscope )等传感器装置实现方向感应(感 应水平面上的方位角、旋转角和倾斜角),从而可以检测移动终端处于正竖、 倒竖、 左横、 右横, 仰、俯等状态。 利用方向感应器来判断用户的体位情况, 例如, 用户是处于站着还是躺着的姿态, 一般用户在使用设备时, 在站和躺 的不同姿态下, 设备的方位角度是有不同的, 通过方向感应器可以区别这两 种情况, 如图 2和图 3所示, 通过方向传感器返回的数据, 实现对用户姿态 "站着" 和 "躺着" 的判断。 其他的体位情况也可以跟用户使用移动终端时 移动终端处于正竖、 倒竖、 左横、 右横, 仰、 俯等状态进行判断。
加速度传感器还可以用来判断移动终端的运动状态。 加速度是空间矢 量, 通过测得其三个坐标轴上的分量, 可以了解物体的运动状态, 从而推测 用户在静止状态还是移动状态。对于推送消息来说,当用户处于静止状态下, 如躺在沙发上的时候, 一般更愿意查看新到的推送消息。 反之, 在移动状态 下, 如公交车上用户可能就会觉得不方便。 然而上述举例仅出于示例目的, 并非绝对情况, 如在地铁上, 或在用户在比较舒适的情况下, 会更容易点击 推送消息。 因为 BodyStatus 有不确定性, 因此下面还采用基于统计的最优 化时间片算法, 优化推送的效果。
(g) 移动终端的位置信息: 通过基站定位或者 GPS ( Global Position System, 全球定位系统)定位获取用户使用移动终端时的位置信息, 结合用 户使用移动终端的时间情况, 可以推断出用户的场所, 标记为 Place, 取值 例如为 "固定场所" , "不确定场所" 等。
其中, "固定场所"表示的是用户经常在某些固定的地理位置使用移动 终端, 如用户可能常在家、 办公室、 餐馆使用移动终端。 当用户在固定场所 使用移动终端时, 一般会更愿意接收新消息。 "不确定场所" 表示除了固定 场所之外的其他位置, 例如用户在移动中, 或者外出的情况下, 往往没有更
多的时间和精力浏览推送信息。
判断用户固定场所的方法为:在记录用户每次使用移动终端的过程同时 记录移动终端所处的位置坐标, 可以通过基站定位或 GPS 定位的方式。 然 后根据坐标的变化来识别用户是否在某一固定的位置,如用户在家使用手机 的时候, 每次得到的坐标值应该是非常接近, 可以认为是相同位置, 在搜集 的大量位置信息后, 然后基于统计的规律提取出反复出现的位置坐标, 将这 些位置作为用户的固定场所。
(h) 移动终端所处环境的光线强度信息:根据移动终端上的传感器装置, 识别出设备使用时外界的光线情况,标记为 Light, 取值例如包括: "很亮",
"一般亮" , "暗" , 艮暗" 等, 具体地取值可根据光线强度进行划分, 例如平均划分为四个阶段,每个阶段分别对应 "很亮", "一般亮", "暗" , 艮暗" , 还可以根据实际情况进行划分, 例如, 艮亮" 和 艮暗" 占的 光线强度范围较大等等, 应当理解, 在实际的实施过程中, 根据光线按照实 际需求划分成多个取值。移动终端上的环境光传感器采集移动终端所处场所 的光线情况。外部光线的强弱对用户使用移动终端会产生影响,如在烈日下, 用户一般不会点击收到的推送消息,而用户可能在晚上没有灯光的情况下使 用移动终端, 此时收到的推送消息更可能被用户点击阅读,
如下表 1表示, 每一次信息的收集可以表示为一行记录。 如图 4所示, 图 4是对本方法数据发送的说明。 用户的移动设备即移动终端, 完成信息收 集后, 以批处理的方式发送给推送至云端服务器, 应当理解还可以发送至其 他服务器或设备, 发送的批量数据标记为 UserTable。 云端服务器收到移动 终端上传的数据后, 把数据存入每个用户对应的数据库表中, 通过用户设备 唯一属性标志形成映射关系。表中每个用户的数据是独立的, 因此每个用户 的推送策略也是独立的, 从而实现个性化定制。
表 1 步骤 S120:根据使用时间和 /或使用场景计算针对用户的优选推送时间。 在本发明的一个实施例中, 可依据最优化时间片算法根据使用时间和 / 或使用场景计算针对用户的优选推送时间。
具体地, 最优化时间片算法包括以下步骤:
步骤 S121 : 获得用户在每天中使用移动终端的使用时间, 并对使用时 间进行标记和排序。
步骤 S122: 计算每次使用时间的时间长度。
步骤 S 123:根据时间长度 /或使用场景计算计算每次使用时间的权重值。 下面以一个具体的例子对最优化时间片算法进行说明, 可以理解的是, 此例仅出于示例目的, 本发明实施例不限于此。
( 1 )根据用户使用移动终端对应的唯一属性标记 (UID) , 从数据库中获 取用户使用移动终端的数据。
( 2 ) 将每天的数据按时间顺序进行排序, 计算每次使用时间的时间长 度并划分出用户每天不同的使用时间, 将使用时间以及相关信息标记为 TimeRecord, TimeRecord包含表 2定义的字段:
表 2
( 3 )根据权重值的计算规则, 计算 TimeRecord的使用时间的权重值。 例中计算每次使用时间的权重值计算规则如下:
权重值的取值范围是为 0到 1 , 值越大表示使用时间越有意义, 更适合 执行推送。 权重值的缺省值是 k, 增长和降低幅度为 0.1 , TimeRecord中不 同的字段值都会影响权重值的增减, 具体如下方式计算:
R1 时间长度越长, 增加相应的权重值。
R2 通话时间越长, 降^ 相应的权重值。 如果在使用时间中, 用户的通 话时间与使用时间之比大于预设值, 则将使用时间对应的权重值置为零。 例 如如果用户的时间全部进行通话, 表示此使用时间内不适合推送信息给用 户。
R3 屏幕点击频率越高, 降低相应的权重值。 例如用户在玩游戏时需要 频繁的点击屏幕, 此使用时间内不适合推送信息给用户。
R4 用户使用移动终端时的姿态信息: 取值 "躺着" 时, 增加相应的权 重值, "站着 (移动中) " 降低相应的权重值。
R5 移动终端所处环境的光线强度信息: 光线一般亮, 增加相应的权重 值, 较亮, 降低相应的权重值。
如图 5所示, 每个使用时间可以表示为图 5中的一个方块, 图的横轴代 表时间, 纵轴代表权重值, 每个方块的面积代表这个使用时间的重要程度, 面积越大表示在这个使用时间内更适合进行信息推送。
TimeRecord权重值代表用户在该使用时间内处于某种状态, 例如, 用 户使用移动终端时的姿态信息为 "躺着" 的比 "站着" 的权重高, 原因是用 户躺着的时候身心一般处于更舒适的状态, 接收并点击推送消息的概率大。
( 4 )对用户在 N天中的使用时间的权重值进行汇总, 以计算针对用户 的优选推送时间, 其中, N为大于 1的整数。 例如可以取 N=30 , 对 30天内 用户使用移动终端的使用时间段列表数据进行统计。将图 5中使用时间以散 列的形式表示在二维空间中, 统计出重叠最多的区域, 即认为用户使用时间 已形成一定的规律,预测用户继续在此使用时间内使用移动终端的可能性较 高。
具体的,对于一周中不同的天数, 比如周一,程序取出所有周一的数据, 然后散列在上面的二位空间表中,通过计算得到使用时间重叠的区域和使用 时间分布密度高的区域。
计算方法可以有多种, 下面描述一种可能的实现方式:
选取时间范围 [Tn- 1 , Τη] , 在此区间内所有使用时间面积总和(重叠的 区域需要重复累积计算面积) 记作∑(η-1,η) , 在总的时间轴 [0,24]小时内, 按步长 m分钟(比如 m取值可以为 5分钟、 10分钟、 20分钟、 30分钟等) 进行计算, 例如先取值 5分钟, 对于每一次步长的增加, 计算时间轴上每个 散落区间的使用时间面积和,然后对每个区间的使用时间面积从高到低进行 排序, 可获得在给定步长下, 获得优选推送时间的集合。 然后修改步长 m 进行迭代, 获得在下一个步长, 例如 10分钟的条件下的优选推送时间的集 合。 依次类推, 可以获得在所有步长值条件下的多个优选推送时间的集合, 这些集合可以理解为适合推送信息给用户的使用时间,并且可以优先考虑步 长取最小值的集合。
( 5 ) 将排序后的使用时间存入数据库, 形成用户特定的优选推送时间 表, 如表 3所示。 移动终端可以根据此表中的信息选择推送时机。 由于用户
一周的七天中所处的状态可能不同,因此需要为用户计算不同天的移动终端 使用情况, 以更加准确的了解用户, 按不同日期执行不同的推送策略。 最优 化使用时间算法完成对用户原始数据的计算, 结果为: 反应用户优选推送时 间的记录表, 记录了用户每周 7天, 每天 24小时内的哪些时间段更可能浏 览推送信息。 例如, 某个用户在每个周六的晚上 8点到 9点之间在固定场所 使用手机, 但屏幕点击的频率又低于一般游戏需要的点击频率, 则用户很可 能是在浏览网页, 因此可以优先考虑在这个时间段内给用户推送消息。
表 3 在计算的过程中, 还需要考虑国家法定假期因素的干扰。 对于落在假期 内的使用时间, 可能按照之前收集的用户信息计算出优选推送时间并不合 适。 例如假期的周一, 用户不需要上班, 因此用户使用手机的时间模型可能 就会变化, 可能更接近周六日的使用模型。
步骤 S130: 根据属性标记获得向用户推送的推送内容, 并在优选推送 时间将推送内容推送至移动终端。
例如, 可以启动推送服务调度程序, 根据优选推送时间集合中的时间段 范围, 选择适当时机为相应的用户推送信息。每个用户都保存优选推送时间 表, 根据此表决定在什么时间调度执行推送。每个用户的优选推送时间表内 容一般是不同的, 因此不同的用户将在不同的时间点接收到推送消息。
此外, 在用户收到推送消息的第一时间点, 如果用户点击了消息, 则认
为这个推送为直接命中,需要增加相应的使用时间的权重值,即 TimeRecord 中的权重值。 如果用户不是在第一时间查看了消息, 则不增加权重值。 如果 用户没有查看这个消息, 则降低相应的使用时间的权重值。 通过实际情况的 反馈能进一步了解用户的使用习惯, 从而对推送时间进行优化。
根据本发明实施例的云推送的方法收集用户使用移动终端的时间情况, 通过用户使用移动终端的时间、位置等信息推测使用场景并通过算法计算出 推送信息给用户的最优时间段, 为不同用户提供了个性化的定制服务。信息 推送服务尽量避免打扰用户, 选择在用户更舒服更合适的时间点到达, 从而 提高用户对信息的接受度和点击率, 改善了用户体验。
为了实现上述实施例, 本发明还提出一种云推送的装置。
下面参考图 6描述根据本发明实施例的云推送的装置。 如图 6所示, 云 推送的装置包括: 接收模块 11、 计算模块 12和推送模块 13。 具体地, 接收 模块 11用于接收移动终端记录的用户使用移动终端的使用时间和 /或使用场 景以及移动终端的属性标记。 计算模块 12用于根据使用时间和 /或使用场景 计算针对用户的优选推送时间。 推送模块 13用于根据属性标记获得向用户 推送的推送内容, 并在优选推送时间内将推送内容推送至移动终端。
其中, 移动终端的属性标记包括移动终端的物理标识和 /或用户的身份 信息。使用场景包括移动终端的位置信息、用户使用移动终端时的姿态信息、 用户使用移动终端时的运动信, 和移动终端所处环境的光线强度信息中的 一种或多种。
在本发明的一个实施例中, 接收模块 11接收的用户使用移动终端的使 用时间和 /或使用场景以及移动终端的属性标记, 为每次用户使用移动终端 时收集的信息, 具体收集并定义用户使用移动终端的信息参数如下:
( a )设备唯一属性标记 (UID)包括两部分内容:
(一) 、 移动终端的物理标识: 如手机的 IMEI号码和 IMSI号码等。
(二) 、 用户的身份信息: 例如用户账号, 用户账号是用户在某个服务 系统平台上的用户标志, 具体可以是 Email邮箱号或者网站的账号, 移动终 端可能会有多人使用, 每个人用不同的账号登陆。
用户的身份信息是可选的, 对于没有在移动终端上设置登陆的, 可以采 用默认用户身份信息的形式。 通过移动终端的物理属性和 /或用户的身份信 息来确定的用户, 并使用与之相关联的数据记录。
(b)开始使用时间点:将屏幕的每一次点亮识别为一次全新的用户使用移
动终端的起点, 记录下时间点, 标记为 StartTime。
(c)结束使用时间点:将屏幕每次的关闭识别为用户一次使用移动终端的 结束点, 记录时间点, 标记为 EndTime。
(d)通话时间记录: 获取用户在 [StartTime, EndTime] 之间内的通话时间 记录, 将用户打电话的行为单独识别, 标记为 CallTime。
(e)屏幕点击记录: 在 [StartTime, EndTime] 之间, 记录用户每次点击屏 幕的操作时间点, 标记为 ClickList。
(f)用户使用移动终端时的姿态信息和用户使用移动终端时的运动信息: 根据移动终端上的传感器装置,识别出用户使用设备时的体位情况,例如 "站 着" 、 "躺着" 和 "坐着" 等等, 体位情况还可以包括其他情况, 本领域的 技术人员很容易根据用户的需求进行扩展,用户使用移动终端时的运动信息 取值例如 "静止" 和 "移动中" , 运动信息还可以包括其他的取值, 例如, 对移动速度进行细致划分等。 将体位情况和移动信息两者结合后标记为 BodyStatus,例如: "站着(静止 ) " , "站着(移动中)" , "躺着(静止 ) " ,
"躺着 (移动中) " 等等。
用户使用移动终端时的姿态可以通过移动终端上的加速度传感器,如加 速计和陀螺仪等传感器装置实现方向感应(感应水平面上的方位角、旋转角 和倾斜角) , 从而可以检测移动终端处于正竖、 倒竖、 左横、 右横, 仰、 俯 等状态。 利用方向感应器来判断用户的体位情况, 例如, 用户是处于站着还 是躺着的姿态, 一般用户在使用设备时, 在站和躺的不同姿态下, 设备的方 位角度是有不同的, 通过方向感应器可以区别这两种情况, 如图 2 和图 3 所示, 通过方向传感器返回的数据, 实现对用户姿态 "站着" 和 "躺着" 的 判断。 其他的体位情况也可以跟用户使用移动终端时移动终端处于正竖、倒 竖、 左横、 右横, 仰、 俯等状态进行判断。
加速度传感器还可以用来判断移动终端的运动状态。 加速度是空间矢 量, 通过测得其三个坐标轴上的分量, 可以了解物体的运动状态, 从而推测 用户在静止状态还是移动状态。对于推送消息来说,当用户处于静止状态下, 如躺在沙发上的时候, 一般更愿意查看新到的推送消息。 反之, 在移动状态 下, 如公交车上用户可能就会觉得不方便。 然而上述举例仅出于示例目的, 并非绝对情况, 如在地铁上, 或在用户在比较舒适的情况下, 会更容易点击 推送消息。 因为 BodyStatus 有不确定性, 因此下面还采用基于统计的最优 化时间片算法, 优化推送的效果。
(g)移动终端的位置信息: 通过基站定位或者 GPS定位获取用户使用移 动终端时的位置信息, 结合用户使用移动终端的时间情况, 可以推断出用户 的场所, 标记为 Place, 取值例如为 "固定场所" , "不确定场所" 等。
其中, "固定场所"表示的是用户经常在某些固定的地理位置使用移动 终端, 如用户可能常在家、 办公室、 餐馆使用移动终端。 当用户在固定场所 使用移动终端时, 一般会更愿意接收新消息。 "不确定场所" 表示除了固定 场所之外的其他位置, 例如用户在移动中, 或者外出的情况下, 往往没有更 多的时间和精力浏览推送信息。
判断用户固定场所的方法为:在记录用户每次使用移动终端的过程同时 记录移动终端所处的位置坐标, 可以通过基站定位或 GPS 定位的方式。 然 后根据坐标的变化来识别用户是否在某一固定的位置,如用户在家使用手机 的时候, 每次得到的坐标值应该是非常接近, 可以认为是相同位置, 在搜集 的大量位置信息后, 然后基于统计的规律提取出反复出现的位置坐标, 将这 些位置作为用户的固定场所。
(h)移动终端所处环境的光线强度信息: 根据移动终端上的传感器装置, 识别出设备使用时外界的光线情况,标记为 Light, 取值例如包括: "很亮",
"一般亮" , "暗" , 艮暗" 等, 具体地取值可根据光线强度进行划分, 例如平均划分为四个阶段,每个阶段分别对应 "很亮", "一般亮", "暗" , 艮暗" , 还可以根据实际情况进行划分, 例如, 艮亮" 和 艮暗" 占的 光线强度范围较大等等, 应当理解, 在实际的实施过程中, 根据光线按照实 际需求划分成多个取值。移动终端上的环境光传感器采集移动终端所处场所 的光线情况。外部光线的强弱对用户使用移动终端会产生影响,如在烈日下, 用户一般不会点击收到的推送消息,而用户可能在晚上没有灯光的情况下使 用移动终端, 此时收到的推送消息更可能被用户点击阅读,
如表 1表示, 每一次信息的收集可以表示为一行记录, 接收模块 11接 收到移动终端记录的每条信息。
在本发明的一个实施例中, 计算模块 12依据最优化时间片算法根据使 用时间和 /或使用场景计算针对用户的优选推送时间。
具体地, 计算模块 12获得用户在每天中使用移动终端的使用时间, 并 对使用时间进行标记和排序, 并计算每次使用时间的时间长度, 以及根据时 间长度 /或使用场景计算计算每次使用时间的权重值。
下面以一个具体的例子对最优化时间片算法进行说明, 可以理解的是,
此例仅出于示例目的, 本发明实施例不限于此。
( 1 ) 根据用户使用移动终端对应的唯一属性标记, 从数据库中获取用 户使用移动终端的数据。
( 2 ) 将每天的数据按时间顺序进行排序, 计算每次使用时间的时间长 度并划分出用户每天不同的使用时间, 将使用时间以及相关信息标记为
TimeRecord, TimeRecord包含表 2定义的字段。
( 3 )根据权重值的计算规则, 计算 TimeRecord的使用时间的权重值。 例中计算每次使用时间的权重值计算规则如下:
权重值的取值范围是为 0到 1 , 值越大表示使用时间越有意义, 更适合 执行推送。 权重值的缺省值是 k, 增长和降低幅度为 0.1 , TimeRecord中不 同的字段值都会影响权重值的增减, 具体如下方式计算:
R1 时间长度越长, 增加相应的权重值。
R2 通话时间越长, 降^ 相应的权重值。 如果在使用时间中, 用户的通 话时间与使用时间之比大于预设值, 则将使用时间对应的权重值置为零。 例 如如果用户的时间全部进行通话, 表示此使用时间内不适合推送信息给用 户。
R3 屏幕点击频率越高, 降低相应的权重值。 例如用户在玩游戏时需要 频繁的点击屏幕, 此使用时间内不适合推送信息给用户。
R4 用户使用移动终端时的姿态信息: 取值 "躺着" 时, 增加相应的权 重值, "站着 (移动中) " 降低相应的权重值。
R5 移动终端所处环境的光线强度信息: 光线一般亮, 增加相应的权重 值, 较亮, 降低相应的权重值。
如图 5所示, 每个使用时间可以表示为图 5中的一个方块, 图的横轴代 表时间, 纵轴代表权重值, 每个方块的面积代表这个使用时间的重要程度, 面积越大表示在这个使用时间内更适合进行信息推送。
TimeRecord权重值代表用户在该使用时间内处于某种状态, 例如, 用 户使用移动终端时的姿态信息为 "躺着" 的比 "站着" 的权重高, 原因是用 户躺着的时候身心一般处于更舒适的状态, 接收并点击推送消息的概率大。
( 4 )对用户在 N天中的使用时间的权重值进行汇总, 以计算针对用户 的优选推送时间, 其中, N为大于 1的整数。 例如可以取 N=30 , 对 30天内 用户使用移动终端的使用时间段列表数据进行统计。将图 5中使用时间以散 列的形式表示在二维空间中, 统计出重叠最多的区域, 即认为用户使用时间
已形成一定的规律,预测用户继续在此使用时间内使用移动终端的可能性较 高。
具体的,对于一周中不同的天数, 比如周一,程序取出所有周一的数据, 然后散列在上面的二位空间表中,通过计算得到使用时间重叠的区域和使用 时间分布密度高的区域。
计算方法可以有多种, 下面描述一种可能的实现方式:
选取时间范围 [Τη- Ι, Τη] , 在此区间内所有使用时间面积总和(重叠的 区域需要重复累积计算面积) 记作∑(η-1,η) , 在总的时间轴 [0,24]小时内, 按步长 m分钟(比如 m取值可以为 5分钟、 10分钟、 20分钟、 30分钟等) 进行计算, 例如先取值 5分钟, 对于每一次步长的增加, 计算时间轴上每个 散落区间的使用时间面积和,然后对每个区间的使用时间面积从高到低进行 排序, 可获得在给定步长下, 获得优选推送时间的集合。 然后修改步长 m 进行迭代, 获得在下一个步长, 例如 10分钟的条件下的优选推送时间的集 合。 依次类推, 可以获得在所有步长值条件下的多个优选推送时间的集合, 这些集合可以理解为适合推送信息给用户的使用时间,并且可以优先考虑步 长取最小值的集合。
( 5 ) 将排序后的使用时间存入数据库, 形成用户特定的优选推送时间 表, 如表 3所示。 移动终端可以根据此表中的信息选择推送时机。 由于用户 一周的七天中所处的状态可能不同,因此需要为用户计算不同天的移动终端 使用情况, 以更加准确的了解用户, 按不同日期执行不同的推送策略。 最优 化使用时间算法完成对用户原始数据的计算, 结果为: 反应用户优选推送时 间的记录表, 记录了用户每周 7天, 每天 24小时内的哪些时间段更可能浏 览推送信息。 例如, 某个用户在每个周六的晚上 8点到 9点之间在固定场所 使用手机, 但屏幕点击的频率又低于一般游戏需要的点击频率, 则用户很可 能是在浏览网页, 因此可以优先考虑在这个时间段内给用户推送消息。
在计算的过程中, 还需要考虑国家法定假期因素的干扰。 对于落在假期 内的使用时间, 可能按照之前收集的用户信息计算出优选推送时间并不合 适。 例如假期的周一, 用户不需要上班, 因此用户使用手机的时间模型可能 就会变化, 可能更接近周六日的使用模型。
在本发明的一个实施例中, 推送模块 13可以启动推送服务调度程序, 根据优选推送时间集合中的时间段范围,选择适当时机为相应的用户推送信 息。每个用户都保存优选推送时间表, 根据此表决定在什么时间调度执行推
送。每个用户的优选推送时间表内容一般是不同的, 因此不同的用户将在不 同的时间点接收到推送消息。
此外, 在用户收到推送消息的第一时间点, 如果用户点击了消息, 则认 为这个推送为直接命中,需要增加相应的使用时间的权重值,即 TimeRecord 中的权重值。 如果用户不是在第一时间查看了消息, 则不增加权重值。 如果 用户没有查看这个消息, 则降低相应的使用时间的权重值。 通过实际情况的 反馈能进一步了解用户的使用习惯, 从而对推送时间进行优化。
在本发明的实施例中, 通过记录并分析用户的使用时间和 /或使用场景, 得出优选推送时间,优选推送时间根据每个用户的记录不同而为用户个性化 定制,在优选推送时间内向用户推送的信息考虑了用户的使用场景和使用习 惯, 尽量选择在用户可能阅读信息的时间进行推送, 从而提高信息被点击阅 读的几率, 避免了实时推送可能在用户不方便阅读时推送给用户带来的不 便, 改善了用户体验。
本发明还提出另一种云推送的方法。下面参考图 7描述根据本发明实施 例的云推送的方法, 包括以下步骤:
S210: 采集用户使用移动终端的使用时间和 /或使用场景。
其中, 使用场景包括移动终端的位置信息、 用户使用移动终端时的姿态 信息、用户使用移动终端时的运动信息和移动终端所处环境的光线强度信息 中的一种或多种。
S220: 将使用时间和 /或使用场景以及移动终端的属性标记发送至云端 服务器, 以使云端服务器根据使用时间和 /或使用场景计算针对用户的优选 推送时间以及根据属性标记获得推送内容。
其中, 移动终端的属性标记包括上述移动终端的物理标识和 /或用户的 身份信息。
S230: 接收云端服务器在优选推送时间内向移动终端推送的推送内容。 在本发明的一个实施例中, 每次用户使用移动终端时,收集信息并定义 用户使用移动终端的信息参数如下:
( a )设备唯一属性标记 (UID)包括两部分内容:
(一) 、 移动终端的物理标识: 如手机的 IMEI号码和 IMSI号码等。
(二) 、 用户的身份信息: 例如用户账号, 用户账号是用户在某个服务 系统平台上的用户标志, 具体可以是 Email邮箱号或者网站的账号, 移动终 端可能会有多人使用, 每个人用不同的账号登陆。
用户的身份信息是可选的, 对于没有在移动终端上设置登陆的, 可以采 用默认用户身份信息的形式。 通过移动终端的物理属性和 /或用户的身份信 息来确定的用户, 并使用与之相关联的数据记录。
(b)开始使用时间点:将屏幕的每一次点亮识别为一次全新的用户使用移 动终端的起点, 记录下时间点, 标记为 StartTime。
(c)结束使用时间点:将屏幕每次的关闭识别为用户一次使用移动终端的 结束点, 记录时间点, 标记为 EndTime。
(d)通话时间记录: 获取用户在 [StartTime, EndTime] 之间内的通话时间 记录, 将用户打电话的行为单独识别, 标记为 CallTime。
(e)屏幕点击记录: 在 [StartTime, EndTime] 之间, 记录用户每次点击屏 幕的操作时间点, 标记为 ClickList。
(f) 用户使用移动终端时的姿态信息和用户使用移动终端时的运动信 息: 根据移动终端上的传感器装置, 识别出用户使用设备时的体位情况, 例 如 "站着" 、 "躺着" 和 "坐着" 等等, 体位情况还可以包括其他情况, 本 领域的技术人员很容易根据用户的需求进行扩展,用户使用移动终端时的运 动信息取值例如 "静止" 和 "移动中" , 运动信息还可以包括其他的取值, 例如, 对移动速度进行细致划分等。 将体位情况和移动信息两者结合后标记 为 BodyStatus,例如: "站着 (静止) " , "站着 (移动中 ) " , "躺着 (静 止) " , "躺着 (移动中) " 等等。
用户使用移动终端时的姿态可以通过移动终端上的加速度传感器,如加 速计 (Accelerometer)和陀螺仪(Gyroscope )等传感器装置实现方向感应(感 应水平面上的方位角、旋转角和倾斜角),从而可以检测移动终端处于正竖、 倒竖、 左横、 右横, 仰、俯等状态。 利用方向感应器来判断用户的体位情况, 例如, 用户是处于站着还是躺着的姿态, 一般用户在使用设备时, 在站和躺 的不同姿态下, 设备的方位角度是有不同的, 通过方向感应器可以区别这两 种情况, 如图 2和图 3所示, 通过方向传感器返回的数据, 实现对用户姿态 "站着" 和 "躺着" 的判断。 其他的体位情况也可以跟用户使用移动终端时 移动终端处于正竖、 倒竖、 左横、 右横, 仰、 俯等状态进行判断。
加速度传感器还可以用来判断移动终端的运动状态。 加速度是空间矢 量, 通过测得其三个坐标轴上的分量, 可以了解物体的运动状态, 从而推测 用户在静止状态还是移动状态。对于推送消息来说,当用户处于静止状态下, 如躺在沙发上的时候, 一般更愿意查看新到的推送消息。 反之, 在移动状态
下, 如公交车上用户可能就会觉得不方便。 然而上述举例仅出于示例目的, 并非绝对情况, 如在地铁上, 或在用户在比较舒适的情况下, 会更容易点击 推送消息。 因为 BodyStatus 有不确定性, 因此下面还采用基于统计的最优 化时间片算法, 优化推送的效果。
(g) 移动终端的位置信息: 通过基站定位或者 GPS ( Global Position System, 全球定位系统)定位获取用户使用移动终端时的位置信息, 结合用 户使用移动终端的时间情况, 可以推断出用户的场所, 标记为 Place, 取值 例如为 "固定场所" , "不确定场所" 等。
其中, "固定场所"表示的是用户经常在某些固定的地理位置使用移动 终端, 如用户可能常在家、 办公室、 餐馆使用移动终端。 当用户在固定场所 使用移动终端时, 一般会更愿意接收新消息。 "不确定场所" 表示除了固定 场所之外的其他位置, 例如用户在移动中, 或者外出的情况下, 往往没有更 多的时间和精力浏览推送信息。
判断用户固定场所的方法为:在记录用户每次使用移动终端的过程同时 记录移动终端所处的位置坐标, 可以通过基站定位或 GPS定位的方式。 然 后根据坐标的变化来识别用户是否在某一固定的位置,如用户在家使用手机 的时候, 每次得到的坐标值应该是非常接近, 可以认为是相同位置, 在搜集 的大量位置信息后, 然后基于统计的规律提取出反复出现的位置坐标, 将这 些位置作为用户的固定场所。
(h) 移动终端所处环境的光线强度信息:根据移动终端上的传感器装置, 识别出设备使用时外界的光线情况,标记为 Light, 取值例如包括: "很亮",
"一般亮" , "暗" , 艮暗" 等, 具体地取值可根据光线强度进行划分, 例如平均划分为四个阶段,每个阶段分别对应 "很亮", "一般亮", "暗" , 艮暗" , 还可以根据实际情况进行划分, 例如, 艮亮" 和 艮暗" 占的 光线强度范围较大等等, 应当理解, 在实际的实施过程中, 根据光线按照实 际需求划分成多个取值。移动终端上的环境光传感器采集移动终端所处场所 的光线情况。外部光线的强弱对用户使用移动终端会产生影响,如在烈日下, 用户一般不会点击收到的推送消息,而用户可能在晚上没有灯光的情况下使 用移动终端, 此时收到的推送消息更可能被用户点击阅读,
如下表 1表示, 每一次信息的收集可以表示为一行记录。 如图 4所示, 图 4是对本方法数据发送的说明。 用户的移动设备即移动终端, 完成信息收 集后, 以批处理的方式发送给推送至云端服务器, 应当理解还可以发送至其
他服务器或设备, 发送的批量数据标记为 UserTable。 云端服务器收到移动 终端上传的数据后, 把数据存入每个用户对应的数据库表中, 通过用户设备 唯一属性标志形成映射关系。表中每个用户的数据是独立的, 因此每个用户 的推送策略也是独立的, 从而实现个性化定制。
在本发明的实施例中, 采集和发送用户的使用时间和 /或使用场景至云 端服务器, 使得云端服务器能够得到用户的使用信息, 从而对用户的情况做 出判断而提出更精确的服务。
为了实现上述实施例, 本发明还提出一种移动终端。
下面参考图 8描述根据本发明实施例的移动终端。 如图 8所示, 移动终 端包括: 采集模块 21、 发送模块 22和接收模块 23。
具体地, 采集模块 21用于采集用户的使用时间和 /或使用场景。 发送模 块 22用于将使用时间和 /或使用场景以及移动终端的属性标记发送至云端服 务器, 以使云端服务器根据使用时间和 /或使用场景计算针对用户的优选推 送时间以及根据属性标记获得推送内容。 接收模块 23用于接收云端服务器 在优选推送时间内向移动终端推送的推送内容。
其中, 移动终端的属性标记包括移动终端的物理标识和 /或用户的身份 信息。使用场景包括移动终端的位置信息、用户使用移动终端时的姿态信息、 用户使用移动终端时的运动信, 和移动终端所处环境的光线强度信息中的 一种或多种。
在本发明的一个实施例中, 每次用户使用移动终端时收集的信息, 具体 收集并定义用户使用移动终端的信息参数如下:
( a )设备唯一属性标记 (UID)包括两部分内容:
(一) 、 移动终端的物理标识: 如手机的 IMEI号码和 IMSI号码等。
(二) 、 用户的身份信息: 例如用户账号, 用户账号是用户在某个服务 系统平台上的用户标志, 具体可以是 Email邮箱号或者网站的账号, 移动终 端可能会有多人使用, 每个人用不同的账号登陆。
用户的身份信息是可选的, 对于没有在移动终端上设置登陆的, 可以采 用默认用户身份信息的形式。 通过移动终端的物理属性和 /或用户的身份信 息来确定的用户, 并使用与之相关联的数据记录。
(b)开始使用时间点:将屏幕的每一次点亮识别为一次全新的用户使用移 动终端的起点, 记录下时间点, 标记为 StartTime。
(c)结束使用时间点:将屏幕每次的关闭识别为用户一次使用移动终端的
结束点, 记录时间点, 标记为 EndTime。
(d)通话时间记录: 获取用户在 [StartTime, EndTime] 之间内的通话时间 记录, 将用户打电话的行为单独识别, 标记为 CallTime。
(e)屏幕点击记录: 在 [StartTime, EndTime] 之间, 记录用户每次点击屏 幕的操作时间点, 标记为 ClickList。
(f)用户使用移动终端时的姿态信息和用户使用移动终端时的运动信息: 根据移动终端上的传感器装置,识别出用户使用设备时的体位情况,例如 "站 着" 、 "躺着" 和 "坐着" 等等, 体位情况还可以包括其他情况, 本领域的 技术人员很容易根据用户的需求进行扩展,用户使用移动终端时的运动信息 取值例如 "静止" 和 "移动中" , 运动信息还可以包括其他的取值, 例如, 对移动速度进行细致划分等。 将体位情况和移动信息两者结合后标记为 BodyStatus,例如: "站着(静止 ) " , "站着(移动中)" , "躺着(静止 ) " ,
"躺着 (移动中) " 等等。
用户使用移动终端时的姿态可以通过移动终端上的加速度传感器,如加 速计和陀螺仪等传感器装置实现方向感应(感应水平面上的方位角、旋转角 和倾斜角) , 从而可以检测移动终端处于正竖、 倒竖、 左横、 右横, 仰、 俯 等状态。 利用方向感应器来判断用户的体位情况, 例如, 用户是处于站着还 是躺着的姿态, 一般用户在使用设备时, 在站和躺的不同姿态下, 设备的方 位角度是有不同的, 通过方向感应器可以区别这两种情况, 如图 2 和图 3 所示, 通过方向传感器返回的数据, 实现对用户姿态 "站着" 和 "躺着" 的 判断。 其他的体位情况也可以跟用户使用移动终端时移动终端处于正竖、倒 竖、 左横、 右横, 仰、 俯等状态进行判断。
加速度传感器还可以用来判断移动终端的运动状态。 加速度是空间矢 量, 通过测得其三个坐标轴上的分量, 可以了解物体的运动状态, 从而推测 用户在静止状态还是移动状态。对于推送消息来说,当用户处于静止状态下, 如躺在沙发上的时候, 一般更愿意查看新到的推送消息。 反之, 在移动状态 下, 如公交车上用户可能就会觉得不方便。 然而上述举例仅出于示例目的, 并非绝对情况, 如在地铁上, 或在用户在比较舒适的情况下, 会更容易点击 推送消息。 因为 BodyStatus 有不确定性, 因此下面还采用基于统计的最优 化时间片算法, 优化推送的效果。
(g)移动终端的位置信息: 通过基站定位或者 GPS定位获取用户使用移 动终端时的位置信息, 结合用户使用移动终端的时间情况, 可以推断出用户
的场所, 标记为 Place, 取值例如为 "固定场所" , "不确定场所" 等。 其中, "固定场所"表示的是用户经常在某些固定的地理位置使用移动 终端, 如用户可能常在家、 办公室、 餐馆使用移动终端。 当用户在固定场所 使用移动终端时, 一般会更愿意接收新消息。 "不确定场所" 表示除了固定 场所之外的其他位置, 例如用户在移动中, 或者外出的情况下, 往往没有更 多的时间和精力浏览推送信息。
判断用户固定场所的方法为:在记录用户每次使用移动终端的过程同时 记录移动终端所处的位置坐标, 可以通过基站定位或 GPS定位的方式。 然 后根据坐标的变化来识别用户是否在某一固定的位置,如用户在家使用手机 的时候, 每次得到的坐标值应该是非常接近, 可以认为是相同位置, 在搜集 的大量位置信息后, 然后基于统计的规律提取出反复出现的位置坐标, 将这 些位置作为用户的固定场所。
(h)移动终端所处环境的光线强度信息: 根据移动终端上的传感器装置, 识别出设备使用时外界的光线情况,标记为 Light, 取值例如包括: "很亮", "一般亮" , "暗" , 艮暗" 等, 具体地取值可根据光线强度进行划分, 例如平均划分为四个阶段,每个阶段分别对应 "很亮", "一般亮", "暗" , 艮暗" , 还可以根据实际情况进行划分, 例如, 艮亮" 和 艮暗" 占的 光线强度范围较大等等, 应当理解, 在实际的实施过程中, 根据光线按照实 际需求划分成多个取值。移动终端上的环境光传感器采集移动终端所处场所 的光线情况。外部光线的强弱对用户使用移动终端会产生影响,如在烈日下, 用户一般不会点击收到的推送消息,而用户可能在晚上没有灯光的情况下使 用移动终端, 此时收到的推送消息更可能被用户点击阅读,
如表 1表示, 采集模块 21的每一次信息的收集可以表示为一行记录。 在本发明的实施例中, 移动终端采集和发送用户的使用时间和 /或使用 场景至云端服务器, 使得云端服务器能够得到用户的使用信息, 从而对用户 的情况做出判断而提出更精确的服务。
下面参考图 9描述根据本发明实施例的云推送的系统, 如图 9所示, 包 括本发明上述任一项实施例所述的云推送的装置 110 和上述任一项实施例 所述的移动终端 120。
根据本发明实施例的云推送的系统收集用户使用移动终端的情况,通过 用户使用移动终端的时间、位置等信息推测使用场景并通过算法计算出推送 信息给用户的最优时间段, 为不同用户提供了个性化的定制服务。信息推送
服务信息推送服务尽量避免打扰用户,尽量选择在用户更舒服更合适的时间 点到达, 从而提高用户对信息的接受度和点击率, 改善了用户体验。
在本说明书的描述中, 参考术语 "一个实施例" 、 "一些实施例" 、 "示 例" 、 "具体示例" 、 或 "一些示例" 等的描述意指结合该实施例或示例描 述的具体特征、 结构、材料或者特点包含于本发明的至少一个实施例或示例 中。 在本说明书中, 对上述术语的示意性表述不一定指的是相同的实施例或 示例。 而且, 描述的具体特征、 结构、 材料或者特点可以在任何的一个或多 个实施例或示例中以合适的方式结合。 在本发明中, 术语 "多个" 是指两个 或两个以上。
尽管已经示出和描述了本发明的实施例,对于本领域的普通技术人员而 言,可以理解在不脱离本发明的原理和精神的情况下可以对这些实施例进行 多种变化、修改、替换和变型,本发明的范围由所附权利要求及其等同限定。
Claims
1、 一种云推送的方法, 其特征在于, 包括以下步骤:
接收移动终端记录的用户使用所述移动终端的使用时间和 /或使用场景 以及所述移动终端的属性标记;
根据所述使用时间和 /或使用场景计算针对所述用户的优选推送时间; 根据所述属性标记获得向所述用户推送的推送内容,并在所述优选推送 时间内将所述推送内容推送至所述移动终端。
2、 如权利要求 1所述的云推送的方法, 其特征在于, 所述移动终端的 属性标记包括所述移动终端的物理标识和 /或所述用户的身份信息。
3、 如权利要求 1或 2所述的云推送的方法, 其特征在于, 所述使用场 景包括所述移动终端的位置信息、 所述用户使用所述移动终端时的姿态信 息、所述用户使用所述移动终端时的运动信息和所述移动终端所处环境的光 线强度信息中的一种或多种。
4、 如权利要求 1-3任一项所述的云推送的方法, 其特征在于, 根据所 述使用时间和 /或使用场景计算针对所述用户的优选推送时间, 具体包括: 依据最优化时间片算法根据所述使用时间和 /或使用场景计算针对所述 用户的优选推送时间。
5、 如权利要求 4所述的云推送的方法, 其特征在于, 所述最优化时间 片算法包括以下步骤:
获得所述用户在每天中使用所述移动终端的使用时间,并对所述使用时 间进行标记和排序;
计算每次所述使用时间的时间长度;
根据所述时间长度 /或使用场景计算计算每次所述使用时间的权重值。
6、 如权利要求 5所述的云推送的方法, 其特征在于, 还包括: 获得所述用户的通话记录;
如果在使用时间中,所述用户的通话时间与所述使用时间之比大于预设
值, 则将所述使用时间对应的权重值置为零。
7、 如权利要求 5或 6所述的云推送的方法, 其特征在于, 还包括: 对所述用户在 N天中的使用时间的权重值进行汇总, 以计算针对所述 用户的优选推送时间, 其中, N为大于 1的整数。
8、 一种云推送的装置, 其特征在于, 包括:
接收模块,用于接收移动终端记录的用户使用所述移动终端的使用时间 和 /或使用场景以及所述移动终端的属性标记;
计算模块, 用于根据所述使用时间和 /或使用场景计算针对所述用户的 优选推送时间;
推送模块, 用于根据所述属性标记获得向所述用户推送的推送内容, 并
9、 如权利要求 8所述的云推送的装置, 其特征在于, 所述移动终端的 属性标记包括所述移动终端的物理标识和 /或所述用户的身份信息。
10、 如权利要求 8或 9所述的云推送的装置, 其特征在于, 所述使用场 景包括所述移动终端的位置信息、 所述用户使用所述移动终端时的姿态信 息、所述用户使用所述移动终端时的运动信息和所述移动终端所处环境的光 线强度信息中的一种或多种。
11、 如权利要求 8-10任一项所述的云推送的装置, 其特征在于, 所述 计算模块依据最优化时间片算法根据所述使用时间和 /或使用场景计算针对 所述用户的优选推送时间。
12、 如权利要求 11所述的云推送的装置, 其特征在于, 所述计算模块 获得所述用户在每天中使用所述移动终端的使用时间,并对所述使用时间进 行标记和排序, 并计算每次所述使用时间的时间长度, 以及根据所述时间长 度 /或使用场景计算计算每次所述使用时间的权重值。
13、 如权利要求 12所述的云推送的装置, 其特征在于, 所述计算模块 还用于获得所述用户的通话记录, 且如果在使用时间中, 所述用户的通话时
间与所述使用时间之比大于预设值, 则将所述使用时间对应的权重值置为
14、 如权利要求 12或 13所述的云推送的装置, 其特征在于, 所述计算 模块对所述用户在 N天中的使用时间的权重值进行汇总, 以计算针对所述 用户的优选推送时间, 其中, N为大于 1的整数。
15、 一种云推送的方法, 其特征在于, 包括以下步骤:
采集用户使用移动终端的使用时间和 /或使用场景;
将所述使用时间和 /或使用场景以及所述移动终端的属性标记发送至云 端服务器, 以使所述云端服务器根据所述使用时间和 /或使用场景计算针对 所述用户的优选推送时间以及根据所述属性标记获得推送内容; 述推送内容。
16、 如权利要求 15所述的云推送的方法, 其特征在于, 所述移动终端 的属性标记包括所述移动终端的物理标识和 /或所述用户的身份信息。
17、 如权利要求 15所述的云推送的方法, 其特征在于, 所述使用场景 包括所述移动终端的位置信息、 所述用户使用所述移动终端时的姿态信息、 所述用户使用所述移动终端时的运动信息和所述移动终端所处环境的光线 强度信息中的一种或多种。
18、 一种移动终端, 其特征在于, 包括:
采集模块, 用于采集用户的使用时间和 /或使用场景;
发送模块, 用于将所述使用时间和 /或使用场景以及移动终端的属性标 记发送至云端服务器, 以使所述云端服务器根据所述使用时间和 /或使用场 景计算针对所述用户的优选推送时间以及根据所述属性标记获得推送内容; 终端推送的推送内容。
19、 如权利要求 18所述的移动终端, 其特征在于, 所述移动终端的属
性标记包括所述移动终端的物理标识和 /或所述用户的身份信息。
20、 如权利要求 18所述的移动终端, 其特征在于, 所述使用场景包括 所述移动终端的位置信息、 所述用户使用所述移动终端时的姿态信息、 所述 用户使用所述移动终端时的运动信息和所述移动终端所处环境的光线强度 信息中的一种或多种。
21、 一种云推送的系统, 其特征在于, 包括权利要求 8-14任一项所述 云端服务器和 18-20任一项所述的移动终端。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210328616.2A CN102882936B (zh) | 2012-09-06 | 2012-09-06 | 云推送的方法、系统和装置 |
| CN201210328616.2 | 2012-09-06 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2014036854A1 true WO2014036854A1 (zh) | 2014-03-13 |
Family
ID=47484078
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2013/078872 Ceased WO2014036854A1 (zh) | 2012-09-06 | 2013-07-05 | 云推送的方法、系统和装置 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN102882936B (zh) |
| WO (1) | WO2014036854A1 (zh) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105282014A (zh) * | 2015-11-17 | 2016-01-27 | 中国建设银行股份有限公司 | 一种信息推送方法和装置 |
| CN108259588A (zh) * | 2018-01-09 | 2018-07-06 | 上海创图网络科技股份有限公司 | 一种基于大数据的文化云平台的推送方法及装置 |
| WO2019019891A1 (en) * | 2017-07-28 | 2019-01-31 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | METHOD, MOBILE TERMINAL, AND NON-TRANSITORY COMPUTER-READABLE STORAGE MEDIUM USED TO ADJUST THE SCAN FREQUENCY OF A TOUCH SCREEN |
| US10764379B2 (en) | 2013-09-06 | 2020-09-01 | Nokia Technologies Oy | Method, apparatus and system for enhancing a user's engagement with a service |
Families Citing this family (47)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102882936B (zh) * | 2012-09-06 | 2015-11-25 | 百度在线网络技术(北京)有限公司 | 云推送的方法、系统和装置 |
| CN104052714B (zh) * | 2013-03-12 | 2019-02-26 | 腾讯科技(深圳)有限公司 | 多媒体信息的推送方法及服务器 |
| CN104168174A (zh) * | 2013-05-16 | 2014-11-26 | 阿里巴巴集团控股有限公司 | 一种传输信息的方法及装置 |
| TWI533265B (zh) * | 2013-11-11 | 2016-05-11 | Chunghwa Telecom Co Ltd | Method of presenting a recommendation message on a map |
| US10467641B2 (en) | 2013-12-20 | 2019-11-05 | Ncr Corporation | Mobile proximity-based analytics |
| CN104750464B (zh) * | 2013-12-25 | 2019-09-10 | 中国移动通信集团公司 | 一种用户状态感知、管理方法及设备 |
| CN104754112A (zh) * | 2013-12-31 | 2015-07-01 | 中兴通讯股份有限公司 | 用户信息获取方法及移动终端 |
| JP2015162162A (ja) * | 2014-02-28 | 2015-09-07 | 東芝テック株式会社 | 情報配信装置、情報配信プログラム |
| CN104980327B (zh) * | 2014-04-03 | 2019-05-03 | 腾讯科技(深圳)有限公司 | 一种消息推送方法及装置 |
| CN104065565B (zh) * | 2014-06-12 | 2017-04-05 | 北京奇虎科技有限公司 | 推送消息的方法、服务器、客户端装置和系统 |
| CN104063462A (zh) * | 2014-06-25 | 2014-09-24 | 北京智谷睿拓技术服务有限公司 | 信息通信方法、装置和系统 |
| CN104063457A (zh) * | 2014-06-25 | 2014-09-24 | 北京智谷睿拓技术服务有限公司 | 信息通信方法和系统及终端 |
| CN104065729B (zh) * | 2014-06-27 | 2017-12-05 | 广州猎豹网络科技有限公司 | 信息推送方法及装置 |
| CN104980480B (zh) * | 2014-09-15 | 2018-09-07 | 腾讯科技(北京)有限公司 | 数据信息分享方法及装置 |
| GB2544662A (en) * | 2014-09-23 | 2017-05-24 | Google Inc | Notifying users of relevant content |
| CN104363302A (zh) * | 2014-11-27 | 2015-02-18 | 北京奇艺世纪科技有限公司 | 一种离线下载文件的方法及装置 |
| CN105791352B (zh) * | 2014-12-24 | 2020-05-29 | 深圳市腾讯计算机系统有限公司 | 应用的消息推送方法和系统 |
| CN104836724B (zh) * | 2015-03-31 | 2018-02-09 | 北京奇艺世纪科技有限公司 | 内容更新提醒方法及装置 |
| CN106156462A (zh) * | 2015-04-01 | 2016-11-23 | 时云医疗科技(上海)有限公司 | 一种基于体征数据的行为信息推送方法及系统 |
| CN104852842A (zh) * | 2015-04-29 | 2015-08-19 | 小米科技有限责任公司 | 信息播报的方法及装置 |
| CN104994062B (zh) * | 2015-05-18 | 2019-04-30 | 百度在线网络技术(北京)有限公司 | 一种信息推送方法及装置 |
| CN104793980B (zh) * | 2015-05-19 | 2019-01-25 | 北京奇虎科技有限公司 | 应用程序更新通知方法及其装置 |
| CN106294410A (zh) * | 2015-05-22 | 2017-01-04 | 苏宁云商集团股份有限公司 | 一种个性化信息推送时间的确定方法及确定系统 |
| CN105099870B (zh) * | 2015-05-28 | 2019-02-15 | 努比亚技术有限公司 | 一种消息推送方法及装置 |
| CN106257886B (zh) | 2015-06-17 | 2020-06-23 | 腾讯科技(深圳)有限公司 | 一种信息处理方法、装置、终端及服务器 |
| CN105095465B (zh) * | 2015-07-31 | 2020-08-07 | 北京金山安全软件有限公司 | 一种信息推荐方法、系统及装置 |
| CN106407239A (zh) * | 2015-08-03 | 2017-02-15 | 阿里巴巴集团控股有限公司 | 用于推荐及辅助推荐信息的方法及装置 |
| CN105653580A (zh) * | 2015-12-18 | 2016-06-08 | 北京奇虎科技有限公司 | 特征信息确定、判定方法及装置以及其应用方法和系统 |
| CN105610929A (zh) * | 2015-12-24 | 2016-05-25 | 北京奇虎科技有限公司 | 一种个性化的数据推送方法和装置 |
| US20190253519A1 (en) * | 2016-06-23 | 2019-08-15 | Koninklijke Philips N.V. | Method, apparatus and machine readable medium for measuring user availability or receptiveness to notifications |
| CN106096059A (zh) * | 2016-07-11 | 2016-11-09 | 上海亿保健康管理有限公司 | 一种基于场景推送信息的方法 |
| CN106557292A (zh) * | 2016-11-11 | 2017-04-05 | 珠海市魅族科技有限公司 | 信息显示方法及装置 |
| CN108093121B (zh) * | 2016-11-23 | 2020-09-11 | 阿里巴巴集团控股有限公司 | 一种终端设备的提醒控制方法、装置以及电子设备 |
| CN108199948A (zh) * | 2016-12-08 | 2018-06-22 | 北京京东尚科信息技术有限公司 | 消息推送方法及装置 |
| CN106487664B (zh) * | 2016-12-08 | 2020-07-03 | 宇龙计算机通信科技(深圳)有限公司 | 一种信息推荐的方法、装置及移动终端 |
| CN106779844B (zh) * | 2016-12-15 | 2020-12-29 | 北京金山安全软件有限公司 | 一种获取增值业务信息的方法、装置及电子设备 |
| CN106815361A (zh) * | 2017-01-24 | 2017-06-09 | 维沃移动通信有限公司 | 一种浏览器页面显示方法及移动终端 |
| CN108123999A (zh) * | 2017-11-28 | 2018-06-05 | 维沃移动通信有限公司 | 一种消息推送方法及移动终端 |
| CN108234290A (zh) * | 2017-12-26 | 2018-06-29 | 珠海市魅族科技有限公司 | 消息推送方法及装置、计算机装置和计算机可读存储介质 |
| CN109978575B (zh) * | 2017-12-27 | 2021-06-04 | 中国移动通信集团广东有限公司 | 一种挖掘用户流量经营场景的方法及装置 |
| CN108769157B (zh) * | 2018-05-16 | 2021-03-19 | 北京奇虎科技有限公司 | 消息弹窗的展示方法、装置、计算设备及计算机存储介质 |
| CN109118069A (zh) * | 2018-07-27 | 2019-01-01 | 重庆工程职业技术学院 | 一种工程进度及造价管理系统的管理方法 |
| CN109598576A (zh) | 2018-10-25 | 2019-04-09 | 阿里巴巴集团控股有限公司 | 服务推荐方法、装置及设备 |
| CN109714399A (zh) * | 2018-12-11 | 2019-05-03 | 浙江口碑网络技术有限公司 | 通知消息的推送方法及装置、存储介质、电子装置 |
| WO2020252639A1 (zh) * | 2019-06-17 | 2020-12-24 | 深圳市欢太科技有限公司 | 内容推送方法及相关产品 |
| CN110493306A (zh) * | 2019-07-05 | 2019-11-22 | 口碑(上海)信息技术有限公司 | 信息推送方法及客户端、服务端、系统 |
| CN110351371A (zh) * | 2019-07-15 | 2019-10-18 | 星联云服科技有限公司 | 一种在云存储系统中进行数据推送的方法及系统 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102300163A (zh) * | 2011-09-22 | 2011-12-28 | 宇龙计算机通信科技(深圳)有限公司 | 信息推送方法、移动终端和系统 |
| CN102486786A (zh) * | 2010-12-02 | 2012-06-06 | 腾讯科技(深圳)有限公司 | 一种向移动终端投放信息的方法及信息投放装置 |
| CN102546777A (zh) * | 2011-12-27 | 2012-07-04 | 广东宇天科技有限公司 | 移动终端桌面信息推送系统及其方法 |
| CN102882936A (zh) * | 2012-09-06 | 2013-01-16 | 百度在线网络技术(北京)有限公司 | 云推送的方法、系统和装置 |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7817983B2 (en) * | 2005-03-14 | 2010-10-19 | Qualcomm Incorporated | Method and apparatus for monitoring usage patterns of a wireless device |
| CN1972468B (zh) * | 2006-10-26 | 2010-05-12 | Ut斯达康通讯有限公司 | 一种信息推送装置和方法 |
| CN101217569A (zh) * | 2008-01-15 | 2008-07-09 | 杭州华三通信技术有限公司 | 一种页面的推送方法、系统和装置 |
| CN101600167A (zh) * | 2008-06-06 | 2009-12-09 | 瞬联软件科技(北京)有限公司 | 面向移动应用的信息自适应交互系统及其实现方法 |
| CN101751448B (zh) * | 2009-07-22 | 2012-06-27 | 中国科学院自动化研究所 | 一种基于情景信息的个性化资源信息的推荐方法 |
| CN101620625B (zh) * | 2009-07-30 | 2012-10-31 | 腾讯科技(深圳)有限公司 | 一种搜索关键词排序方法、装置和搜索引擎 |
-
2012
- 2012-09-06 CN CN201210328616.2A patent/CN102882936B/zh active Active
-
2013
- 2013-07-05 WO PCT/CN2013/078872 patent/WO2014036854A1/zh not_active Ceased
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102486786A (zh) * | 2010-12-02 | 2012-06-06 | 腾讯科技(深圳)有限公司 | 一种向移动终端投放信息的方法及信息投放装置 |
| CN102300163A (zh) * | 2011-09-22 | 2011-12-28 | 宇龙计算机通信科技(深圳)有限公司 | 信息推送方法、移动终端和系统 |
| CN102546777A (zh) * | 2011-12-27 | 2012-07-04 | 广东宇天科技有限公司 | 移动终端桌面信息推送系统及其方法 |
| CN102882936A (zh) * | 2012-09-06 | 2013-01-16 | 百度在线网络技术(北京)有限公司 | 云推送的方法、系统和装置 |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10764379B2 (en) | 2013-09-06 | 2020-09-01 | Nokia Technologies Oy | Method, apparatus and system for enhancing a user's engagement with a service |
| CN105282014A (zh) * | 2015-11-17 | 2016-01-27 | 中国建设银行股份有限公司 | 一种信息推送方法和装置 |
| WO2019019891A1 (en) * | 2017-07-28 | 2019-01-31 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | METHOD, MOBILE TERMINAL, AND NON-TRANSITORY COMPUTER-READABLE STORAGE MEDIUM USED TO ADJUST THE SCAN FREQUENCY OF A TOUCH SCREEN |
| US10761642B2 (en) | 2017-07-28 | 2020-09-01 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | Method, mobile terminal and non-transitory computer-readable storage medium for adjusting scanning frequency of touch screen |
| CN108259588A (zh) * | 2018-01-09 | 2018-07-06 | 上海创图网络科技股份有限公司 | 一种基于大数据的文化云平台的推送方法及装置 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102882936A (zh) | 2013-01-16 |
| CN102882936B (zh) | 2015-11-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102882936B (zh) | 云推送的方法、系统和装置 | |
| TWI428564B (zh) | 個人化距離之判定及顯示系統與方法 | |
| CN107273437B (zh) | 提供与用户可能访问的地点相关的信息的方法和系统 | |
| US10230803B2 (en) | System and method for improved mapping and routing | |
| US12598151B2 (en) | Location based content system for mobile applications | |
| US20200204643A1 (en) | User profile generation method and terminal | |
| CN106600314B (zh) | 一种广告推送系统 | |
| US20200118191A1 (en) | Apparatus and method for recommending place | |
| US20100082427A1 (en) | System and Method for Context Enhanced Ad Creation | |
| WO2016191515A1 (en) | Personalized information from venues of interest | |
| AU2014275203A1 (en) | Modeling significant locations | |
| EP2329450A2 (en) | System and method for context enhanced messaging | |
| US20180032585A1 (en) | Systems and methods for context-based contact suggestion | |
| US20220078135A1 (en) | Signal upload optimization | |
| US20200272676A1 (en) | Characterizing a place by features of a user visit | |
| JP6396686B2 (ja) | 行動判定装置、行動判定方法及びプログラム | |
| JP2014071540A (ja) | 情報配信装置、プログラム及び方法 | |
| CN105706409B (zh) | 用于增强用户对于服务的参与度的方法、设备及系统 | |
| JP6378994B2 (ja) | 配信装置及び配信方法 |
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: 13836058 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: 13836058 Country of ref document: EP Kind code of ref document: A1 |


