WO2017076004A1 - 预测预定时刻的用户位置的方法和装置 - Google Patents

预测预定时刻的用户位置的方法和装置 Download PDF

Info

Publication number
WO2017076004A1
WO2017076004A1 PCT/CN2016/086215 CN2016086215W WO2017076004A1 WO 2017076004 A1 WO2017076004 A1 WO 2017076004A1 CN 2016086215 W CN2016086215 W CN 2016086215W WO 2017076004 A1 WO2017076004 A1 WO 2017076004A1
Authority
WO
WIPO (PCT)
Prior art keywords
candidate
candidate stay
weight
determining
point
Prior art date
Application number
PCT/CN2016/086215
Other languages
English (en)
French (fr)
Inventor
何佳倍
吴海山
武政伟
韩艳
Original Assignee
百度在线网络技术(北京)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 百度在线网络技术(北京)有限公司 filed Critical 百度在线网络技术(北京)有限公司
Publication of WO2017076004A1 publication Critical patent/WO2017076004A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W4/00Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02Services making use of location information
    • H04W4/029Location-based management or tracking services
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S5/00Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
    • G01S5/02Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
    • G01S5/10Position of receiver fixed by co-ordinating a plurality of position lines defined by path-difference measurements, e.g. omega or decca systems
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W4/00Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02Services making use of location information
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W4/00Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/12Messaging; Mailboxes; Announcements
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W64/00Locating users or terminals or network equipment for network management purposes, e.g. mobility management
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W8/00Network data management
    • H04W8/22Processing or transfer of terminal data, e.g. status or physical capabilities

Definitions

  • the present application relates to the field of computer technologies, and in particular, to the field of Internet technologies, and in particular, to a method and apparatus for predicting a user location at a predetermined time.
  • Information Push also known as “webcasting” is a technology that reduces information overload by pushing the information the user needs on the Internet through certain technical standards or protocols. Information push technology can reduce the time it takes for users to search on the network by actively pushing information to users.
  • the pushed content is often associated with the geographic location where the user receiving the push information is located.
  • the generation of the push information often takes a certain amount of time. Therefore, if the geographical location of the user at a certain moment in the future can be accurately predicted, the accuracy and pertinence of the information push will be greatly improved, and the push information can be more effectively utilized by the user who receives the information.
  • the geographic location of a user at a particular moment is usually predicted by the regularity of the geographic location of the user. For example, the user is usually at the company at 10 am, then the user is predicted to be at the company at 10 am on a certain day.
  • the influence of the current position of the user on the predicted position is not considered. For example, when the user is at the company at 10 am, he is at home at 9 am. And if you predict that the user is 9 o'clock in the zoo, then the possibility of 10 o'clock in the company will be greatly reduced.
  • the purpose of the present application is to propose an improved method and apparatus for predicting the position of a user at a predetermined time to solve the technical problems mentioned in the background section above.
  • the present application provides a method for predicting a location of a user at a predetermined time, including: acquiring current location information and current time information of a user; and determining, in the candidate stay point set, based on the current location information and the current time information. a first weight of each candidate stay point at a predetermined time; determining a second weight of each candidate stay point in the candidate stay point set based on the predetermined time; and selecting a stay point based on the first weight value and the second weight value A user location corresponding to a predetermined time is determined in the set.
  • determining the first weight of each candidate stay point in the candidate stay point set at the predetermined moment based on the current location information and the current time information includes: acquiring a historical stay point of the user as a candidate stay point; determining a transition probability of the first candidate stay point in the candidate stay point set to the second candidate stay point; and determining, according to the transition probability, a first weight of each candidate stay point at a predetermined time; wherein, the first candidate stay point and the first The two candidate stay points are any candidate stay points in the candidate stay point set, and the transition probability is a probability that the path with the first candidate stay point as the starting point and the second candidate stay point as the end point is generated within the predetermined time interval.
  • determining the first weight of each candidate stay point at a predetermined moment based on the transition probability comprises: determining an N ⁇ N-order transition matrix S based on the transition probability, where N is a candidate stay in the candidate stay point set The number of points; based on the transfer matrix S, determining a first weight P 1 of each candidate stay point at a predetermined time; wherein:
  • t 2 is a predetermined time interval
  • t 1 is a predetermined time
  • t 0 is a current time
  • determining the second weight of each candidate stay point in the candidate stay point set at the predetermined moment comprises: acquiring a historical stay point of the user as a candidate stay point; acquiring historical time information corresponding to each candidate stay point; And determining, according to each candidate stay point and historical time information corresponding to each candidate stay point, a second weight of each candidate stay point at a predetermined time.
  • determining, according to the predetermined moment, the second weight of each candidate stay point in the set of candidate stay points comprises: determining a stay probability of the user at each candidate stay point within a plurality of preset historical time intervals; Based on the dwell probability, a K ⁇ N-order time matrix T is generated, where K is the number of historical time intervals, N is the number of candidate stay points, and from the time matrix, the N-dimensional column vector T i corresponding to the predetermined time is determined, Wherein, 1 ⁇ i ⁇ K, each element in the column vector is a second weight of each candidate stay point at a predetermined time.
  • each historical time interval has the same duration.
  • K is an even number; when 1 ⁇ i ⁇ K/2, each element in T i is the dwell probability of each candidate stay point in the ith historical time interval of the working day; When +1 ⁇ i ⁇ K, each element in T i is the staying probability of each candidate stay point in the iK/2th historical time interval of the holiday.
  • K 48.
  • determining, according to the first weight and the second weight, the user location corresponding to the predetermined moment in the set of candidate stay points comprises: determining, according to T i ⁇ P 1 , each candidate stay point corresponding to the predetermined moment a prediction weight; and a candidate stay point having the largest predicted weight among each candidate stay point as a user position corresponding to the predetermined time.
  • the present application provides an apparatus for predicting a user location at a predetermined time, including: an obtaining module configured to acquire current location information and current time information of a user; and a first weight determining module configured to be based on a current Determining, by the location information and the current time information, a first weight of each candidate stay point in the candidate stay point set at a predetermined time; the second weight determining module configured to determine each candidate stay point in the candidate stay point set a second weight at a predetermined time; and a position prediction module configured to determine a user location corresponding to the predetermined time in the set of candidate stay points based on the first weight and the second weight.
  • the first weight determining module is further configured to: acquire a historical stay point of the user as a candidate stay point; and determine a transition probability of the first candidate stay point in the candidate stay point set to transfer to the second candidate stay point And determining, according to the transition probability, a first weight of each candidate stay point at a predetermined moment; wherein, the first candidate stay point and the second candidate stay point are any candidate stay points in the candidate stay point set, and the transition probability is predetermined During the time interval, a probability of a path starting from the first candidate stay point and ending with the second candidate stay point is generated.
  • the first weight determining module when determining the first weight of each candidate stay point at a predetermined time based on the transition probability, is further configured to: determine an N ⁇ N-order transition matrix S based on the transition probability, where N is the number of candidate stay points in the set of candidate stay points; and based on the transition matrix S, determining a first weight P 1 of each candidate stay point at a predetermined time; wherein:
  • t 2 is a predetermined time interval
  • t 1 is a predetermined time
  • t 0 is a current time
  • the second weight determining module is further configured to: acquire a historical stay point of the user as a candidate stay point; acquire historical time information corresponding to each candidate stay point; and based on each candidate stay point and each candidate stay point Corresponding historical time information, determining a second weight of each candidate stay point at a predetermined time.
  • the second weight determination module is further configured to: when determining, according to each candidate stay point and the historical time information corresponding to each candidate stay point, the second weight of each candidate stay point at the predetermined time: Determining the staying probability of the user at each candidate stay point in a plurality of preset historical time intervals; generating a K ⁇ N-order time matrix T based on the staying probability, where K is the number of historical time intervals, and N is the candidate stay point And the N-dimensional column vector T i corresponding to the predetermined time is determined from the time matrix, where 1 ⁇ i ⁇ K, and each element in the column vector is the second weight of each candidate stay point at a predetermined time.
  • each historical time interval has the same duration.
  • K is an even number; when 1 ⁇ i ⁇ K/2, each element in T i is the dwell probability of each candidate stay point in the ith historical time interval of the working day; When +1 ⁇ i ⁇ K, each element in T i is the staying probability of each candidate stay point in the iK/2th historical time interval of the holiday.
  • K 48.
  • the location prediction module is further configured to: determine a prediction weight of each candidate stay point corresponding to the predetermined time instant based on T i ⁇ P 1 ; and select the candidate with the largest prediction weight among each candidate stay point The stay point serves as a user position corresponding to the predetermined time.
  • the method and apparatus for predicting a user location at a predetermined time provided by the present application, determining a first weight of each candidate stay point in a candidate stay point set at a predetermined time based on current location information of the user, and determining based on the current time information
  • the second weight of each candidate stay point in the set of candidate stay points at a predetermined time thereby predicting the user position at a certain predetermined time in the future, so that the prediction result is not only related to the predetermined time but also to the user.
  • Pre-position correlation can improve the accuracy of user location prediction.
  • the method and apparatus for predicting the user position at the predetermined time of the present application are used for position prediction, thereby improving the accuracy of information push and Targeted.
  • FIG. 1 is an exemplary system architecture diagram to which the present application can be applied;
  • FIG. 2 is a flow diagram of one embodiment of a method of predicting a user location at a predetermined time in accordance with the present application
  • FIG. 3 is a flowchart of an optional implementation manner of determining, according to current location information, a first weight of each candidate stay point in a candidate stay point set at a predetermined time in FIG. 2;
  • FIG. 4 is a flowchart of an optional implementation manner of determining, according to current time information, a second weight of each candidate stay point in a candidate stay point set at a predetermined time in FIG. 2;
  • FIG. 5 is a schematic structural diagram of an embodiment of an apparatus for predicting a user position at a predetermined time according to the present application
  • FIG. 6 is a schematic structural diagram of a computer system suitable for implementing a terminal device or a server of an embodiment of the present application.
  • FIG. 1 illustrates an exemplary system architecture 100 of an embodiment of a web page generation method or web page generation apparatus to which the present application may be applied.
  • system architecture 100 can include terminal devices 101, 102, 103, network 104, and server 105.
  • the network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105.
  • Network 104 may include various types of connections, such as wired, wireless communication links, fiber optic cables, and the like.
  • the user can interact with the server 105 over the network 104 using the terminal devices 101, 102, 103 to receive or transmit messages and the like.
  • Various communication client applications such as a web browser application, a shopping application, a search application, an instant communication tool, a mailbox client, a social platform software, and the like, may be installed on the terminal devices 101, 102, and 103.
  • the terminal devices 101, 102, 103 may be various electronic devices having a display screen and having the ability to acquire their own geographic location information, including but not limited to smartphones, tablets, e-book readers, MP3 players (Moving Picture Experts) Group Audio Layer III, motion picture expert compresses standard audio layer 3), MP4 (Moving Picture Experts Group Audio Layer IV) player, laptop portable computer and desktop computer, and so on.
  • smartphones including but not limited to smartphones, tablets, e-book readers, MP3 players (Moving Picture Experts) Group Audio Layer III, motion picture expert compresses standard audio layer 3), MP4 (Moving Picture Experts Group Audio Layer IV) player, laptop portable computer and desktop computer, and so on.
  • the server 105 may be a server that provides various services, such as a server that generates push information based on the predicted location of the terminal device at a certain predetermined time in the future based on the current geographic location of the terminal device 101, 102, 103, and based on the predicted location. .
  • the server 105 can perform processing such as analyzing the received current geographic location and the like, and feed back the processing result (for example, push information) to the terminal device.
  • the method for predicting the user location at a predetermined time is generally performed by the server 105. Accordingly, the device for predicting the user location at the predetermined time is generally disposed in the server 105.
  • the server 105 can directly interact with the terminal devices 101, 102, and 103 through the network 104 to obtain the terminal devices 101, 102, and 103.
  • the current geographic location of the corresponding user may acquire data such as the current geographic location of the user corresponding to the terminal devices 101, 102, 103 from other servers that can interact with the terminal devices 101, 102, 103.
  • terminal devices, networks, and servers in Figure 1 is merely illustrative. Depending on the implementation needs, there can be any number of terminal devices, networks, and servers.
  • the method for predicting a user location at a predetermined time includes the following steps:
  • Step 210 Acquire current location information and current time information of the user.
  • the electronic device for example, the server 105 shown in FIG. 1 on which the method for predicting the location of the user at the predetermined time can be used from the terminal device or the user can be used by the user through a wired connection or a wireless connection.
  • Other electronic devices that use the terminal device for data interaction acquire the current location information and current time information of the user.
  • the above wireless connection manner may include but is not limited to 3G/4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods that are now known or developed in the future. .
  • the electronic device on which the method for predicting the location of the user at the predetermined time can receive the current location information actively uploaded by the terminal device used by the user, and the time when the current location information is received is used as the Current time information corresponding to the current location information.
  • the electronic device on which the method for predicting the location of the user at a predetermined time is run may request the terminal device used by the user to acquire current location information of the terminal device, based on the terminal device's request for the request. Allow to get the current location information of the terminal device.
  • the positioning module on the terminal device can be turned on based on the permission of the terminal device to obtain the request for the current location information to obtain the current location information of the terminal device.
  • the electronic device on which the method of predicting the location of the user at the predetermined time is operated may use the time at which the current location information is acquired as the current time information corresponding to the current location information.
  • the electronic device on which the method for predicting the location of the user at a predetermined time is run may also be based on the terminal device and the electronic device or the terminal device and a server capable of data interaction with the electronic device.
  • the interaction content (for example, a search keyword) determines the current location information of the terminal device and the current time information corresponding to the current location information.
  • Step 220 Determine, according to the current location information and the current time information, a first weight of each candidate stay point in the candidate stay point set at a predetermined time.
  • a plurality of candidate stay points may be included in the set of candidate stay points.
  • each candidate stay point in the set of candidate stay points may be, for example, a location where the user has reached and the dwell time exceeds a predetermined time threshold (eg, 1 hour).
  • a predetermined time threshold eg, 1 hour
  • each candidate stay point in the candidate stay point set may also be some "hot" location. For example, a well-known tourist attraction, business center, etc. in the city where the user is located.
  • the first weight can be understood, for example, as: in the case that the user is currently in a certain location, the user is in each candidate stay point in the candidate stay point set at a certain predetermined time in the future at a time interval from the current time. "Possibility”.
  • Step 230 Determine a second weight of each candidate stay point in the candidate stay point set based on the predetermined time.
  • the candidate stay point set of this step may have the same candidate stay point as the candidate stay point set in step 220.
  • the second weight can be understood, for example, as the "probability" of each candidate stay point in the set of candidate stay points at the predetermined time.
  • Step 240 Determine, according to the first weight value and the second weight value, a user location corresponding to the predetermined time in the candidate stay point set.
  • the first weight is related to the time difference between the current position of the user and the future predetermined time at which the position prediction needs to be performed, and the second weight and the position prediction are required.
  • the scheduled time is relevant. That is to say, when performing the position prediction at the predetermined time, the influence of the current position of the user on the prediction result is considered, and the time difference between the current time and the future predetermined time at which the position prediction needs to be performed is considered. The impact of the results.
  • the method for predicting the user position at the predetermined time in this embodiment has a higher prediction accuracy.
  • the step 220 determines, according to the current location information and the current time information, the first weight of each candidate stay point in the candidate stay point set at the predetermined time, and may adopt the process shown in FIG. 3 300 to achieve.
  • step 310 the user's historical stay point is acquired as a candidate stay point.
  • the historical stay point can be analyzed, for example, by the user. Move the trajectory to get.
  • a location where the user has reached and the dwell time exceeds a predetermined time threshold may be used as the historical stop point for the user.
  • a location where the user has reached a certain location more than a predetermined number of thresholds may also be used as the historical stop point of the user.
  • a transition probability of the first candidate stay point in the set of candidate stay points to the second candidate stay point is determined.
  • the first candidate stay point and the second candidate stay point are any candidate stay points in the candidate stay point set, and the transition probability may be within a predetermined time interval, generating a first candidate stay point as a starting point, and a second candidate The probability that the stop point is the path of the end point.
  • the candidate stay point set includes three candidate stay points of A, B, and C, and the predetermined time interval is 1 hour. Then, within one hour, the probability that the user takes A as the starting point and the path ending with B as the starting point can be used as the transition probability from A to B. That is to say, the probability that the current position of the user is at A and the position after 1 hour is B is the transition probability of A to B.
  • a transition probability between candidate stay points can be counted based on a first-order Markov model.
  • the candidate stay point set includes three candidate stay points A, B, and C as an example.
  • the number of times that the user takes A as the starting point and B as the end point is b
  • the number of times the user takes A as the starting point and C as the ending point is c
  • the value of b/(b+c) can be used as the value.
  • the transition probability from A to B, and the value of c/(b+c) is taken as the transition probability from A to C.
  • step 330 based on the transition probability, a first weight of each candidate stay point at a predetermined time is determined.
  • the candidate stay point set includes three candidate stay points A, B, and C as an example. If the interval between the predetermined time and the current time is a predetermined time interval as described above, then The transition probability from A to B is taken as the first weight between point A and point B.
  • the first weight of each candidate stay point can be, for example, Expressed in the form of a matrix.
  • determining, according to the transition probability of step 330, the first weight of each candidate stay point at a predetermined moment may be implemented, for example, by:
  • an N ⁇ N-order transition matrix S is determined, where N is the number of candidate stay points in the set of candidate stay points.
  • the value of each element s ij in the transition matrix S may be, for example, a transition probability from the candidate stay point i to the candidate stay point j, where N, i, j are positive integers, and i, j ⁇ N.
  • t 2 is a predetermined time interval
  • t 1 is a predetermined time
  • t 0 is a current time.
  • determining a second weight of each candidate stay point in the set of candidate stay points based on the predetermined time of step 230 may be implemented, for example, by the process 400 shown in FIG. 4 .
  • step 410 the user's historical stay point is obtained as a candidate stay point.
  • a historical stop point for the user may be obtained in a manner similar to step 310 in FIG.
  • step 420 historical time information corresponding to each candidate stay point is obtained.
  • the user once appeared at the zoo at 10 am and stayed for a while, then "10 am” can be used as a historical time information for the "park" candidate stop.
  • a second weight of each candidate stay point at a predetermined time is determined based on each candidate stay point and historical time information corresponding to each candidate stay point.
  • the second weight of each candidate stay point at a predetermined moment may be understood as the "probability" of the user at each of the candidate stay points at the predetermined moment.
  • determining, according to each candidate stay point and the historical time information corresponding to each candidate stay point, the second weight of each candidate stay point at a predetermined time may be implemented by, for example, the following manner :
  • 24 hours of a day may be divided into multiple historical time intervals, for example, 0 to 1 point is a historical time interval, and 1 to 2 is a historical time interval, ..., 23 Point ⁇ 24 points is a historical time interval. The probability of the user staying at each candidate stay point is counted separately in these historical time intervals.
  • a K ⁇ N-order time matrix T is generated, where K is the number of historical time intervals and N is the number of candidate stay points.
  • each element t ij represents the probability that the user appears at the jth candidate stay point in the ith historical time interval.
  • an N-dimensional column vector T i corresponding to the predetermined time is determined, where 1 ⁇ i ⁇ K, and each element in the column vector is a second weight of each candidate stay point at a predetermined time.
  • the predetermined time falls within the ith time interval in the time matrix T, and in these alternative implementations, each element in the column vector T i is taken as the predetermined time a second weight corresponding to each candidate stay point.
  • the second weight corresponding to each candidate stay point in T i is also the stay probability of the candidate stay points in the ith historical time interval.
  • each historical time interval may have the same duration.
  • a historical time interval can be divided equally into multiple (eg, 6, 12, or 24, etc.) 24 hours a day.
  • each historical time interval may also have a different duration. For example, based on data mining, it is known that during a certain period of the day (for example, from 1 am to 5 pm), the user's position rarely changes, and at another time of the day (for example, from 12:00 to 14:00) The user's location may change more frequently. Then, in these optional implementations, for example, 1 to 5 am may be used as a historical time interval, and 12 to 14 noon may be further divided into multiple (for example, 2 or 4) historical time. Interval.
  • the probability that a user appears at each candidate stay point on weekdays may be significantly different from the probability that it appears at each candidate stay point during a holiday. For example, on weekdays, users have a higher probability of appearing in the company between 10:00 am and 11:00 am, while during holidays, users appear in the company's time between 10:00 and 11:00. The probability may be significantly lower than the probability that the user will appear in the company within the same time interval of the working day.
  • the staying probabilities of the candidate stay points corresponding to the respective historical time intervals of the working day are respectively acquired, and the stays of the candidate stay points corresponding to the respective historical time intervals of the holiday are respectively acquired.
  • Probability to form a time matrix T For example, K can be set to an even number.
  • each element in the column vector T i may be the dwell probability of each candidate stay point in the ith historical time interval of the working day.
  • K/2+1 ⁇ i ⁇ K each element in the column vector T i may be the staying probability of each candidate stay point in the iK/2th historical time interval of the holiday.
  • K 48.
  • 24 hours a day of the working day can be equally divided into 24 historical time intervals
  • the daily 24 hours of the holiday can be equally divided into 24 historical time. Interval.
  • the column vector T i in the time matrix T may represent a staying probability of each candidate stay point within a certain historical time interval of the working day, or a candidate stay point within a certain historical time interval representing the holiday. Probability of stay.
  • determining, by the first weight value and the second weight value, the user location corresponding to the predetermined time in the candidate stay point set may be, for example, The following way to achieve:
  • the predicted weight of each candidate stay point corresponding to the predetermined time is determined based on T i ⁇ P 1 .
  • T i is an N-dimensional column vector
  • N is the number of candidate stay points
  • P 1 is an N ⁇ N-order matrix. Therefore, after matrix multiplication of T i and P 1 , the result is still an N-dimensional column vector.
  • the candidate stay point having the largest predicted weight is used as the user position corresponding to the predetermined time.
  • the predicted user position of the predetermined time can be made not only related to the predetermined time but also related to the current position of the user, thereby improving the accuracy of the user position prediction. Sex.
  • the method and apparatus for predicting the user position at the predetermined time in the embodiment of the present application perform position prediction, thereby improving the accuracy of the information push. Sexual and pertinent.
  • the present application provides an embodiment 500 of an apparatus for predicting a user location at a predetermined time, the apparatus embodiment being in accordance with the method embodiment illustrated in FIG.
  • the device can be specifically applied to various electronic devices.
  • the apparatus for predicting the user location at a predetermined time in this embodiment may include an obtaining module 510, a first weight determining module 520, a second weight determining module 530, and a position predicting module 540.
  • the obtaining module 510 is configured to acquire current location information and current time information of the user.
  • the first weight determining module 520 is configured to determine, according to the current location information and the current time information, a first weight of each candidate stay point in the candidate stay point set at a predetermined time.
  • the second weight determination module 530 is configurable to determine a second weight of each candidate stay point in the set of candidate stay points based on the predetermined time instant.
  • the location prediction module 540 is configurable to determine a user location corresponding to the predetermined moment in the set of candidate stay points based on the first weight and the second weight.
  • the first weight determination module 520 can be further configured. And a method for: acquiring a historical stay point of the user as a candidate stay point; determining a transition probability of the first candidate stay point in the candidate stay point set to transfer to the second candidate stay point; and determining, according to the transition probability, each candidate stay point at a predetermined moment The first weight.
  • the first candidate stay point and the second candidate stay point may be any candidate stay points in the candidate stay point set, and the transition probability may be within a predetermined time interval, and the first candidate stay point is generated as a starting point, and the second The probability that the candidate stay point is the path of the end point.
  • the first weight determining module 520 may further configure, when determining the first weight of each candidate stay point at a predetermined moment based on the transition probability, to determine N ⁇ based on the transition probability.
  • An N-th order transition matrix S wherein N is the number of candidate stay points in the set of candidate stay points; and based on the transition matrix S, determining a first weight P 1 of each candidate stay point at a predetermined time; wherein:
  • t 2 is a predetermined time interval
  • t 1 is a predetermined time
  • t 0 is a current time.
  • the second weight determining module 530 may be further configured to: acquire a historical stay point of the user as a candidate stay point; acquire historical time information corresponding to each candidate stay point; and base each candidate The stay point and the historical time information corresponding to each candidate stay point determine a second weight of each candidate stay point at a predetermined time.
  • the second weight determining module 530 determines, when each candidate stay point has a second weight at a predetermined moment, based on each candidate stay point and historical time information corresponding to each candidate stay point, The method may be further configured to: determine a stay probability of the user at each candidate stay point in a plurality of preset historical time intervals; and generate a K ⁇ N order time matrix T based on the stay probability, where K is the number of historical time intervals N is the number of candidate stay points; and from the time matrix, the N-dimensional column vector T i corresponding to the predetermined time is determined, where 1 ⁇ i ⁇ K, and each element in the column vector is a candidate stay point at a predetermined time The second weight.
  • each historical time interval may have the same duration.
  • K may be an even number; when 1 ⁇ i ⁇ K/2, each element in T i is a staying probability of each candidate stay point in the ith historical time interval of the working day; When K/2+1 ⁇ i ⁇ K, each element in T i is the staying probability of each candidate stay point in the iK/2th historical time interval of the holiday.
  • K 48.
  • the location prediction module 540 is further configured to: determine a prediction weight of each candidate stay point corresponding to the predetermined moment based on T i ⁇ P 1 ; and have each of the candidate stay points The candidate stay point of the maximum predicted weight is taken as the user position corresponding to the predetermined time.
  • the web page generating apparatus 500 described above also includes other well-known structures such as a processor, a memory, etc., which are not shown in FIG. 5 in order to unnecessarily obscure the embodiments of the present disclosure.
  • FIG. 6 a block diagram of a computer system 600 suitable for use in implementing a terminal device or server of an embodiment of the present application is shown.
  • computer system 600 includes a central processing unit (CPU) 601 that can be loaded into a program in random access memory (RAM) 603 according to a program stored in read only memory (ROM) 602 or from storage portion 608. And perform various appropriate actions and processes.
  • RAM random access memory
  • ROM read only memory
  • RAM random access memory
  • various programs and data required for the operation of the system 600 are also stored.
  • the CPU 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604.
  • An input/output (I/O) interface 605 is also coupled to bus 604.
  • the following components are connected to the I/O interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), and the like, and a storage portion 608 including a hard disk or the like. And a communication portion 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the Internet.
  • Driver 610 is also coupled to I/O interface 605 as needed.
  • a removable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory or the like, is mounted on the drive 610 as needed so that a computer program read therefrom is installed into the storage portion 608 as needed.
  • an embodiment of the present disclosure includes a computer program product comprising a computer program tangibly embodied on a machine readable medium, the computer program comprising program code for executing the method illustrated in the flowchart.
  • the computer program can be downloaded and installed from the network via the communication portion 609, and/or installed from the removable medium 611.
  • each block of the flowchart or block diagrams can represent a module, a program segment, or a portion of code that includes one or more logic for implementing the specified.
  • Functional executable instructions can also occur in a different order than that illustrated in the drawings. For example, two successively represented blocks may in fact be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending upon the functionality involved.
  • each block of the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts can be implemented in a dedicated hardware-based system that performs the specified function or operation. Or it can be implemented by a combination of dedicated hardware and computer instructions.
  • the units involved in the embodiments of the present application may be implemented by software or by hardware.
  • the described unit may also be disposed in the processor.
  • a processor includes an acquisition module, a first weight determination module, a second weight determination module, and a position prediction module.
  • the name of these units does not constitute a limitation on the unit itself in some cases.
  • the acquisition module may also be described as “a module that acquires current location information and current time information of the user”.
  • the present application further provides a non-volatile computer storage medium, which may be a non-volatile computer storage medium included in the apparatus described in the foregoing embodiments; It may be a non-volatile computer storage medium that exists alone and is not assembled into the terminal.
  • the non-volatile computer storage medium stores one or more programs, when the one or more programs are executed by a device, causing the device to: acquire current location information and current time information of the user; based on current location information And determining, by the current time information, a first weight of each candidate stay point in the candidate stay point set at a predetermined time; determining, according to the predetermined time, a second weight of each candidate stay point in the candidate stay point set; and based on the first The weight and the second weight determine a user location corresponding to the predetermined time in the set of candidate stay points.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Navigation (AREA)

Abstract

本申请公开了一种预测预定时刻的用户位置的方法和装置。所述方法的一具体实施方式包括:获取用户的当前位置信息和当前时刻信息;基于当前位置信息和当前时刻信息,确定候选停留点集合中的各候选停留点在预定时刻的第一权值;基于预定时刻,确定候选停留点集合中的各候选停留点的第二权值;以及基于第一权值和第二权值,在候选停留点集合中确定与预定时刻对应的用户位置。该实施方式实现位置预测的准确性。

Description

预测预定时刻的用户位置的方法和装置
相关申请的交叉引用
本申请要求于2015年11月06日提交的中国专利申请号为“201510753326.6”的优先权,其全部内容作为整体并入本申请中。
技术领域
本申请涉及计算机技术领域,具体涉及互联网技术领域,尤其涉及预测预定时刻的用户位置的方法和装置。
背景技术
信息推送,又称为“网络广播”,是通过一定的技术标准或协议,在互联网上通过推送用户需要的信息来减少信息过载的一项技术。信息推送技术通过主动推送信息给用户,可以减少用户在网络上搜索所花的时间。
在信息推送时,推送的内容往往与接收推送信息的用户所处的地理位置相关联,此外,推送信息的生成往往需要花费一定的时间。因此,如果能准确地预测出未来某一时刻用户所处的地理位置,则将大大提高信息推送的准确性和针对性,进而使得推送信息更有效地被接收到该信息的用户所利用。
在现有技术中,通常通过用户所处地理位置在时间上的规律性来预测用户在某一特定时刻的地理位置。例如,用户通常上午10点在公司,那么预测用户在某一日的上午10点时,也在公司。
然而,现有的位置预测方案中,未考虑用户的当前位置对预测位置的影响。例如,用户平时上午10点在公司的时候,上午9点都在家里。而如果预测时,发现用户9点在动物园,那么10点在公司的可能性将大大降低。
发明内容
本申请的目的在于提出一种改进的预测预定时刻的用户位置的方法和装置,来解决以上背景技术部分提到的技术问题。
第一方面,本申请提供了一种预测预定时刻的用户位置的方法,包括:获取用户的当前位置信息和当前时刻信息;基于当前位置信息和所述当前时刻信息,确定候选停留点集合中的各候选停留点在预定时刻的第一权值;基于预定时刻,确定候选停留点集合中的各候选停留点的第二权值;以及基于第一权值和第二权值,在候选停留点集合中确定与预定时刻对应的用户位置。
在一些实施例中,基于当前位置信息和所述当前时刻信息,确定候选停留点集合中的各候选停留点在预定时刻的第一权值包括:获取用户的历史停留点作为候选停留点;确定候选停留点集合中的第一候选停留点向第二候选停留点转移的转移概率;以及基于转移概率,确定各候选停留点在预定时刻的第一权值;其中,第一候选停留点和第二候选停留点均为候选停留点集合中的任意候选停留点,转移概率为预定时间间隔内,生成以第一候选停留点为起点,以第二候选停留点为终点的路径的概率。
在一些实施例中,基于转移概率,确定各候选停留点在预定时刻的第一权值包括:基于转移概率,确定N×N阶转移矩阵S,其中,N为候选停留点集合中的候选停留点的数量;基于转移矩阵S,确定各候选停留点在预定时刻的第一权值P1;其中:
Figure PCTCN2016086215-appb-000001
或者,
Figure PCTCN2016086215-appb-000002
t2为预定时间间隔,t1为预定时刻,t0为当前时刻。
在一些实施例中,确定候选停留点集合中的各候选停留点在预定时刻的第二权值包括:获取用户的历史停留点作为候选停留点;获取与各候选停留点对应的历史时间信息;以及基于各候选停留点和与各候选停留点对应的历史时间信息,确定各候选停留点在预定时刻的第二权值。
在一些实施例中,基于预定时刻,确定候选停留点集合中的各候选停留点的第二权值包括:确定用户在多个预设的历史时间区间内, 处于各候选停留点的停留概率;基于停留概率,生成K×N阶时间矩阵T,其中,K为历史时间区间的数量,N为候选停留点的数量;以及从时间矩阵中,确定与预定时刻对应的N维列向量Ti,其中,1≤i≤K,列向量中的各元素为各候选停留点在预定时刻的第二权值。
在一些实施例中,各历史时间区间具有相同的时长。
在一些实施例中,K为偶数;当1≤i≤K/2时,Ti中的各元素为工作日的第i个历史时间区间内,各候选停留点的停留概率;当K/2+1≤i≤K时,Ti中的各元素为节假日的第i-K/2个历史时间区间内,各候选停留点的停留概率。
在一些实施例中,K=48。
在一些实施例中,基于第一权值和第二权值,在候选停留点集合中确定与预定时刻对应的用户位置包括:基于Ti×P1确定与预定时刻对应的各候选停留点的预测权值;以及将各候选停留点中,具有最大预测权值的候选停留点作为与预定时刻对应的用户位置。
第二方面,本申请提供了一种预测预定时刻的用户位置的装置,包括:获取模块,配置用于获取用户的当前位置信息和当前时刻信息;第一权值确定模块,配置用于基于当前位置信息和所述当前时刻信息,确定候选停留点集合中的各候选停留点在预定时刻的第一权值;第二权值确定模块,配置用于确定候选停留点集合中的各候选停留点在预定时刻的第二权值;以及位置预测模块,配置用于基于第一权值和第二权值,在候选停留点集合中确定与预定时刻对应的用户位置。
在一些实施例中,第一权值确定模块进一步配置用于:获取用户的历史停留点作为候选停留点;确定候选停留点集合中的第一候选停留点向第二候选停留点转移的转移概率;以及基于转移概率,确定各候选停留点在预定时刻的第一权值;其中,第一候选停留点和第二候选停留点均为候选停留点集合中的任意候选停留点,转移概率为预定时间间隔内,生成以第一候选停留点为起点,以第二候选停留点为终点的路径的概率。
根据本申请的第一权值确定模块在基于转移概率,确定各候选停留点在预定时刻的第一权值时,进一步配置用于:基于转移概率,确 定N×N阶转移矩阵S,其中,N为候选停留点集合中的候选停留点的数量;以及基于转移矩阵S,确定各候选停留点在预定时刻的第一权值P1;其中:
Figure PCTCN2016086215-appb-000003
或者,
Figure PCTCN2016086215-appb-000004
t2为预定时间间隔,t1为预定时刻,t0为当前时刻。
根据本申请的第二权值确定模块进一步配置用于:获取用户的历史停留点作为候选停留点;获取与各候选停留点对应的历史时间信息;以及基于各候选停留点和与各候选停留点对应的历史时间信息,确定各候选停留点在预定时刻的第二权值。
在一些实施例中,第二权值确定模块在基于各候选停留点和与各候选停留点对应的历史时间信息,确定各候选停留点在预定时刻的第二权值时,进一步配置用于:确定用户在多个预设的历史时间区间内,处于各候选停留点的停留概率;基于停留概率,生成K×N阶时间矩阵T,其中,K为历史时间区间的数量,N为候选停留点的数量;以及从时间矩阵中,确定与预定时刻对应的N维列向量Ti,其中,1≤i≤K,列向量中的各元素为各候选停留点在预定时刻的第二权值。
在一些实施例中,各历史时间区间具有相同的时长。
在一些实施例中,K为偶数;当1≤i≤K/2时,Ti中的各元素为工作日的第i个历史时间区间内,各候选停留点的停留概率;当K/2+1≤i≤K时,Ti中的各元素为节假日的第i-K/2个历史时间区间内,各候选停留点的停留概率。
在一些实施例中,K=48。
在一些实施例中,位置预测模块进一步配置用于:基于Ti×P1确定与预定时刻对应的各候选停留点的预测权值;以及将各候选停留点中,具有最大预测权值的候选停留点作为与预定时刻对应的用户位置。
本申请提供的预测预定时刻的用户位置的方法和装置,通过基于用户的当前位置信息来确定候选停留点集合中的各候选停留点在预定时刻的第一权值,并基于当前时刻信息来确定候选停留点集合中的各候选停留点在预定时刻的第二权值,进而对未来的某个预定时刻的用户位置进行预测,使得预测结果不仅与预定时刻相关,还与用户的当 前位置相关,可提高用户位置预测的准确性。
在一些应用场景中,当利用用户在未来的某个预定时刻的位置来进行信息推送时,采用本申请的预测预定时刻的用户位置的方法和装置进行位置预测,可以提高信息推送的准确性和针对性。
附图说明
通过阅读参照以下附图所作的对非限制性实施例所作的详细描述,本申请的其它特征、目的和优点将会变得更明显:
图1是本申请可以应用于其中的示例性系统架构图;
图2是根据本申请的预测预定时刻的用户位置的方法的一个实施例的流程图;
图3是图2中的基于当前位置信息,确定候选停留点集合中的各候选停留点在预定时刻的第一权值的一种可选的实现方式的流程图;
图4是图2中的基于当前时刻信息,确定候选停留点集合中的各候选停留点在预定时刻的第二权值的一种可选的实现方式的流程图;
图5是根据本申请的预测预定时刻的用户位置的装置的一个实施例的结构示意图;
图6是适于用来实现本申请实施例的终端设备或服务器的计算机系统的结构示意图。
具体实施方式
下面结合附图和实施例对本申请作进一步的详细说明。可以理解的是,此处所描述的具体实施例仅仅用于解释相关发明,而非对该发明的限定。另外还需要说明的是,为了便于描述,附图中仅示出了与有关发明相关的部分。
需要说明的是,在不冲突的情况下,本申请中的实施例及实施例中的特征可以相互组合。下面将参考附图并结合实施例来详细说明本申请。
图1示出了可以应用本申请的网页生成方法或网页生成装置的实施例的示例性系统架构100。
如图1所示,系统架构100可以包括终端设备101、102、103,网络104和服务器105。网络104用以在终端设备101、102、103和服务器105之间提供通信链路的介质。网络104可以包括各种连接类型,例如有线、无线通信链路或者光纤电缆等等。
用户可以使用终端设备101、102、103通过网络104与服务器105交互,以接收或发送消息等。终端设备101、102、103上可以安装有各种通讯客户端应用,例如网页浏览器应用、购物类应用、搜索类应用、即时通信工具、邮箱客户端、社交平台软件等。
终端设备101、102、103可以是具有显示屏并且具有获取自身的地理位置信息的能力的各种电子设备,包括但不限于智能手机、平板电脑、电子书阅读器、MP3播放器(Moving Picture Experts Group Audio Layer III,动态影像专家压缩标准音频层面3)、MP4(Moving Picture Experts Group Audio Layer IV,动态影像专家压缩标准音频层面4)播放器、膝上型便携计算机和台式计算机等等。
服务器105可以是提供各种服务的服务器,例如基于终端设备101、102、103的当前地理位置,基于预测的终端设备在未来的某一预定时刻的位置,并基于预测的位置生成推送信息的服务器。该服务器105可以对接收到的当前地理位置等数据进行分析等处理,并将处理结果(例如推送信息)反馈给终端设备。
需要说明的是,本申请实施例所提供的预测预定时刻的用户位置的方法一般由服务器105执行,相应地,预测预定时刻的用户位置的装置一般设置于服务器105中。
此外,图1所示的系统架构仅仅是示意性的,在实际的应用场景中,服务器105可以直接通过网络104来与终端设备101、102、103交互,以获取与终端设备101、102、103对应的用户的当前地理位置。或者,服务器105还可以从其它的可与终端设备101、102、103进行交互的服务器获取与终端设备101、102、103对应的用户的当前地理位置等数据。
应该理解,图1中的终端设备、网络和服务器的数目仅仅是示意性的。根据实现需要,可以具有任意数目的终端设备、网络和服务器。
继续参考图2,其示出了根据本申请的预测预定时刻的用户位置的方法的一个实施例的流程200。所述的预测预定时刻的用户位置的方法,包括以下步骤:
步骤210,获取用户的当前位置信息和当前时刻信息。
在本实施例中,预测预定时刻的用户位置的方法运行于其上的电子设备(例如图1所示的服务器105)可以通过有线连接方式或者无线连接方式从用户使用的终端设备或者可与用户使用的终端设备进行数据交互的其它电子设备获取用户的当前位置信息和当前时刻信息。需要指出的是,上述无线连接方式可以包括但不限于3G/4G连接、WiFi连接、蓝牙连接、WiMAX连接、Zigbee连接、UWB(ultra wideband)连接、以及其他现在已知或将来开发的无线连接方式。
在一些可选的实现方式中,预测预定时刻的用户位置的方法运行于其上的电子设备可以接收用户使用的终端设备主动上传的当前位置信息,并以接收到当前位置信息的时刻作为与该当前位置信息对应的当前时刻信息。
或者,在另一些可选的实现方式中,预测预定时刻的用户位置的方法运行于其上的电子设备可以向用户使用的终端设备请求获取终端设备的当前位置信息,基于终端设备对该请求的允许,来获取终端设备的当前位置信息。例如,可基于终端设备对获取其当前位置信息的请求的允许,开启终端设备上的定位模块,以得到该终端设备的当前位置信息。类似地,预测预定时刻的用户位置的方法运行于其上的电子设备可将获取到当前位置信息的时刻作为与该当前位置信息对应的当前时刻信息。
或者,在另一些可选的实现方式中,预测预定时刻的用户位置的方法运行于其上的电子设备还可以基于终端设备与该电子设备或者终端设备和能够与该电子设备进行数据交互的服务器之间的交互内容(例如,搜索关键词),来确定终端设备的当前的位置信息和与该当前位置信息对应的当前时刻信息。
步骤220,基于当前位置信息和当前时刻信息,确定候选停留点集合中的各候选停留点在预定时刻的第一权值。
在一些可选的实现方式中,候选停留点集合中可以包括多个候选停留点。在一些可选的实现方式的一些应用场景中,候选停留点集合中的各候选停留点例如可以是用户曾经到达并且停留时间超过一预定时间阈值(例如1小时)的地点。或者,在这一些可选的实现方式的另一些应用场景中,候选停留点集合中的各候选停留点还可以是一些“热门”的地点。例如,用户所在城市的知名旅游景点、商业中心等等。
在这里,第一权值例如可以理解为:在用户当前处于某一地点的情况下,用户在与当前时刻间隔一段时间的未来的某一预定时刻,处于候选停留点集合中的各候选停留点的“可能性”。
步骤230,基于预定时刻,确定候选停留点集合中的各候选停留点的第二权值。
在一些可选的实现方式中,本步骤的候选停留点集合可以与步骤220中的候选停留点集合具有相同的候选停留点。此外,与步骤220中的第一权值类似地,第二权值例如可以理解为:用户在预定时刻处于候选停留点集合中的各候选停留点的“可能性”。
步骤240,基于第一权值和第二权值,在候选停留点集合中确定与预定时刻对应的用户位置。
在本实施例的预测预定时刻的用户位置的方法中,第一权值与用户的当前位置以及需要进行位置预测的未来的预定时刻之间的时间差相关,而第二权值与需要进行位置预测的预定时刻相关。也即是说,在进行预定时刻的位置预测时,既考虑了用户所处的当前位置对预测结果的影响,也考虑了当前时刻和需要进行位置预测的未来的预定时刻之间的时间差对预测结果的影响。与现有的位置预测方案相比,本实施例的预测预定时刻的用户位置的方法预测准确度更高。
在一些可选的实现方式中,步骤220的基于当前位置信息和当前时刻信息,确定候选停留点集合中的各候选停留点在预定时刻的第一权值,可以采用如图3所示的流程300来实现。
具体而言,在步骤310中,获取用户的历史停留点作为候选停留点。
在一些可选的实现方式中,历史停留点例如可以通过分析用户的 移动轨迹来获得。在这些可选的实现方式的一些应用场景中,如上所述,可以将用户曾经到达并且停留时间超过一预定时间阈值的地点作为用户的历史停留点。或者,在这些可选的实现方式的另一些应用场景中,还可以将用户曾经到达某一地点的次数超过一预定的次数阈值的地点作为该用户的历史停留点。
在步骤320中,确定候选停留点集合中的第一候选停留点向第二候选停留点转移的转移概率。在这里,第一候选停留点和第二候选停留点均为候选停留点集合中的任意候选停留点,转移概率可以为预定时间间隔内,生成以第一候选停留点为起点,以第二候选停留点为终点的路径的概率。
例如,假设候选停留点集合中包含A、B、C三个候选停留点,预定时间间隔为1小时。那么,在1小时之内,生成用户以A为起点,以B为终点的路径的概率,即可作为从A到B的转移概率。也即是说,用户当前位置在A且1小时之后的位置在B的概率,即为A到B的转移概率。在一些可选的实现方式中,例如可以基于一阶马尔科夫模型来统计各候选停留点之间的转移概率。
在一些应用场景中,同样以候选停留点集合包含A、B、C三个候选停留点为例。假设用户以A为起点以B为终点的路径发生的次数为b,而用户以A为起点以C为终点的路径发生的次数为c,那么,可以将b/(b+c)的数值作为从A到B的转移概率,而将c/(b+c)的数值作为从A到C的转移概率。
在步骤330中,基于转移概率,确定各候选停留点在预定时刻的第一权值。
在一些可选的实现方式中,同样以候选停留点集合包含A、B、C三个候选停留点为例,假设预定时刻与当前时刻之间间隔为如上所述的预定时间间隔,那么,可以将从A到B的转移概率作为A点和B点之间的第一权值。
由于候选停留点集合中的任意两个候选停留点之间,以及任意一个候选停留点和其自身之间均具有一转移概率,因此,在一些可选的实现方式中,候选停留点集合中的各候选停留点的第一权值例如可以 用矩阵的形式来表达。
在一些可选的实现方式中,步骤330的基于转移概率,确定各候选停留点在预定时刻的第一权值例如可以通过如下的方式来实现:
首先,基于转移概率,确定N×N阶转移矩阵S,其中,N为候选停留点集合中的候选停留点的数量。转移矩阵S中的每一个元素sij的数值,例如可以是从候选停留点i向候选停留点j转移的转移概率,在这里,N、i、j为正整数,且i、j≤N。
接着,基于转移矩阵S,确定各候选停留点在预定时刻的第一权值P1;其中:
Figure PCTCN2016086215-appb-000005
或者,
Figure PCTCN2016086215-appb-000006
在上述公式(1)及公式(2)中,t2为预定时间间隔,t1为预定时刻,t0为当前时刻。
也即是说,若基于上述公式(1)来确定P1,则将t1与t0之差与t2相除后向下取整(假设得到的数值为k),再将转移矩阵作k次矩阵乘法,最终得到第一权值的矩阵表达P1
类似地,若基于上述公式(2)来确定P1,则将t1与t0之差与t2相除后向上取整(假设得到的数值为m),再将转移矩阵作m次矩阵乘法,最终得到第一权值的矩阵表达P1
在这些可选的实现方式中,显然,矩阵P1与转移矩阵S的阶数相同,也为N×N阶。
在一些可选的实现方式中,步骤230的基于预定时刻,确定候选停留点集合中的各候选停留点的第二权值,例如可以采用如图4所示的流程400来实现。
具体而言,在步骤410中,获取用户的历史停留点作为候选停留点。
在一些可选的实现方式中,例如,可以采用与图3中的步骤310类似的方式来获得用户的历史停留点。
在步骤420中,获取与各候选停留点对应的历史时间信息。
例如,在一些可选的实现方式中,用户曾经在上午10点出现在动物园,并停留了一段时间,那么“上午10点”即可作为“动物园”这一候选停留点的一个历史时间信息。
在步骤430中,基于各候选停留点和与各候选停留点对应的历史时间信息,确定各候选停留点在预定时刻的第二权值。
在一些可选的实现方式中,各候选停留点在预定时刻的第二权值可以理解为,用户在该预定时刻,出现在各个候选停留点的“可能性”。
在一些可选的实现方式中,步骤430的基于各候选停留点和与各候选停留点对应的历史时间信息,确定各候选停留点在预定时刻的第二权值例如可以通过如下的方式来实现:
首先,确定用户在多个预设的历史时间区间内,处于各候选停留点的停留概率。
例如在一些可选的实现方式中,可以将一天的24小时划分为多个历史时间区间,例如0点~1点为一个历史时间区间,1点~2点为一个历史时间区间,…,23点~24点为一个历史时间区间。分别统计在这些历史时间区间内,用户处于各个候选停留点的停留概率。
接着,基于停留概率,生成K×N阶时间矩阵T,其中,K为历史时间区间的数量,N为候选停留点的数量。
在时间矩阵T中,每一个元素tij表示在第i个历史时间区间,用户出现在第j个候选停留点的概率。在这里,显然有1≤i≤K,1≤j≤N。
接着,从时间矩阵中,确定与预定时刻对应的N维列向量Ti,其中,1≤i≤K,列向量中的各元素为各候选停留点在预定时刻的第二权值。
例如,在一些可选的实现方式中,预定时刻落入时间矩阵T中的第i个时间区间,那么在这些可选的实现方式中,将列向量Ti中的各个元素作为在该预定时刻,与各候选停留点对应的第二权值。在这些可选的实现方式中,Ti中与各候选停留点对应的第二权值也就是这些候选停留点在第i个历史时间区间的停留概率。
在一些可选的实现方式中,各历史时间区间可以具有相同的时长。 例如,可以将一天24小时平均地划分为多个(例如,6个、12个或24个,等等)历史时间区间。
或者,在另一些可选的实现方式中,各历史时间区间也可以具有不同的时长。例如,基于数据挖掘,获知在一天中的某一时间段(例如凌晨1点~5点),用户的位置鲜有改变,而在一天中的另一时间段(例如中午12点~14点),用户的位置可能发生较为频繁的改变。那么,在这些可选的实现方式中,例如,可以将凌晨1点~5点作为一个历史时间区间,而将中午12点~14点进一步划分为多个(例如2个或4个)历史时间区间。
在一些应用场景中,用户在工作日时,其出现在各候选停留点的概率可能显著地不同于在节假日时,其出现在各候选停留点的概率。例如,在工作日时,用户在上午10点~11点这一时间区间内出现在公司的概率较高,而在节假日时,用户在上午10点~11点这一时间区间内出现在公司的概率可能显著地低于工作日的同一时间区间内,用户出现在公司的概率。
为了解决上述问题,在一些可选的实现方式中,分别获取对应于工作日的各历史时间区间的各候选停留点的停留概率,以及对应于节假日的各历史时间区间的各候选停留点的停留概率,以形成时间矩阵T。例如,可以设K为偶数。当1≤i≤K/2时,列向量Ti中的各元素可以为工作日的第i个历史时间区间内,各候选停留点的停留概率。而当K/2+1≤i≤K时,列向量Ti中的各元素可以为节假日的第i-K/2个历史时间区间内,各候选停留点的停留概率。
在一些可选的实现方式中,K=48。在这些可选的实现方式的一些应用场景中,例如,可以将工作日的一日24小时平均地分为24个历史时间区间,并将节假日的一日24小时平均地分为24个历史时间区间。在这些应用场景中,时间矩阵T中的列向量Ti可能代表工作日的某一历史时间区间内,各候选停留点的停留概率,或者代表节假日的某一历史时间区间内,各候选停留点的停留概率。
在一些可选的实现方式中,步骤240的基于第一权值和第二权值,在候选停留点集合中确定与预定时刻对应的用户位置例如可以通过如 下的方式来实现:
首先,基于Ti×P1确定与预定时刻对应的各候选停留点的预测权值。
由于Ti为N维列向量,N为候选停留点的数量,而P1为N×N阶矩阵,因此,将Ti与P1进行矩阵乘法之后,得到的结果仍然为N维列向量。
接着,将各候选停留点中,具有最大预测权值的候选停留点作为与预定时刻对应的用户位置。
采用如上所述的实施例的预测预定时刻的用户位置的方法,可以使得预测得到的预定时刻的用户位置不仅与该预定时刻相关,还与用户的当前位置相关,从而可提高用户位置预测的准确性。
在一些应用场景中,当利用用户在未来的某个预定时刻的位置来进行信息推送时,采用本申请实施例的预测预定时刻的用户位置的方法和装置进行位置预测,可以提高信息推送的准确性和针对性。
进一步参考图5,作为对上述各图所示方法的实现,本申请提供了一种预测预定时刻的用户位置的装置的一个实施例500,该装置实施例与图2所示的方法实施例相对应,该装置具体可以应用于各种电子设备中。
如图5所示,本实施例的预测预定时刻的用户位置的装置可包括获取模块510、第一权值确定模块520、第二权值确定模块530和位置预测模块540。
其中,获取模块510可配置用于获取用户的当前位置信息和当前时刻信息。
第一权值确定模块520可配置用于基于当前位置信息和当前时刻信息,确定候选停留点集合中的各候选停留点在预定时刻的第一权值。
第二权值确定模块530可配置用于基于预定时刻,确定候选停留点集合中的各候选停留点的第二权值。
位置预测模块540可配置用于基于第一权值和第二权值,在候选停留点集合中确定与预定时刻对应的用户位置。
在一些可选的实现方式中,第一权值确定模块520可进一步配置 用于:获取用户的历史停留点作为候选停留点;确定候选停留点集合中的第一候选停留点向第二候选停留点转移的转移概率;以及基于转移概率,确定各候选停留点在预定时刻的第一权值。在这里,第一候选停留点和第二候选停留点均可为候选停留点集合中的任意候选停留点,转移概率可以为预定时间间隔内,生成以第一候选停留点为起点,以第二候选停留点为终点的路径的概率。
在一些可选的实现方式中,第一权值确定模块520在基于转移概率,确定各候选停留点在预定时刻的第一权值时,还可进一步配置用于:基于转移概率,确定N×N阶转移矩阵S,其中,N为候选停留点集合中的候选停留点的数量;以及基于转移矩阵S,确定各候选停留点在预定时刻的第一权值P1;其中:
Figure PCTCN2016086215-appb-000007
或者,
Figure PCTCN2016086215-appb-000008
t2为预定时间间隔,t1为预定时刻,t0为当前时刻。
在一些可选的实现方式中,第二权值确定模块530还可进一步配置用于:获取用户的历史停留点作为候选停留点;获取与各候选停留点对应的历史时间信息;以及基于各候选停留点和与各候选停留点对应的历史时间信息,确定各候选停留点在预定时刻的第二权值。
在一些可选的实现方式中,第二权值确定模块530在基于各候选停留点和与各候选停留点对应的历史时间信息,确定各候选停留点在预定时刻的第二权值时,还可进一步配置用于:确定用户在多个预设的历史时间区间内,处于各候选停留点的停留概率;基于停留概率,生成K×N阶时间矩阵T,其中,K为历史时间区间的数量,N为候选停留点的数量;以及从时间矩阵中,确定与预定时刻对应的N维列向量Ti,其中,1≤i≤K,列向量中的各元素为各候选停留点在预定时刻的第二权值。
在一些可选的实现方式中,各历史时间区间可以具有相同的时长。
在一些可选的实现方式中,K可以为偶数;当1≤i≤K/2时,Ti中的各元素为工作日的第i个历史时间区间内,各候选停留点的停留概率;当K/2+1≤i≤K时,Ti中的各元素为节假日的第i-K/2个历史时间 区间内,各候选停留点的停留概率。
在一些可选的实现方式中,K=48。
在一些可选的实现方式中,位置预测模块540还可进一步配置用于:基于Ti×P1确定与预定时刻对应的各候选停留点的预测权值;以及将各候选停留点中,具有最大预测权值的候选停留点作为与预定时刻对应的用户位置。
本领域技术人员可以理解,上述网页生成装置500还包括一些其他公知结构,例如处理器、存储器等,为了不必要地模糊本公开的实施例,这些公知的结构在图5中未示出。
下面参考图6,其示出了适于用来实现本申请实施例的终端设备或服务器的计算机系统600的结构示意图。
如图6所示,计算机系统600包括中央处理单元(CPU)601,其可以根据存储在只读存储器(ROM)602中的程序或者从存储部分608加载到随机访问存储器(RAM)603中的程序而执行各种适当的动作和处理。在RAM 603中,还存储有系统600操作所需的各种程序和数据。CPU 601、ROM 602以及RAM 603通过总线604彼此相连。输入/输出(I/O)接口605也连接至总线604。
以下部件连接至I/O接口605:包括键盘、鼠标等的输入部分606;包括诸如阴极射线管(CRT)、液晶显示器(LCD)等以及扬声器等的输出部分607;包括硬盘等的存储部分608;以及包括诸如LAN卡、调制解调器等的网络接口卡的通信部分609。通信部分609经由诸如因特网的网络执行通信处理。驱动器610也根据需要连接至I/O接口605。可拆卸介质611,诸如磁盘、光盘、磁光盘、半导体存储器等等,根据需要安装在驱动器610上,以便于从其上读出的计算机程序根据需要被安装入存储部分608。
特别地,根据本公开的实施例,上文参考流程图描述的过程可以被实现为计算机软件程序。例如,本公开的实施例包括一种计算机程序产品,其包括有形地包含在机器可读介质上的计算机程序,所述计算机程序包含用于执行流程图所示的方法的程序代码。在这样的实施 例中,该计算机程序可以通过通信部分609从网络上被下载和安装,和/或从可拆卸介质611被安装。
附图中的流程图和框图,图示了按照本申请各种实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上,流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分,所述模块、程序段、或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意,在有些作为替换的实现中,方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如,两个接连地表示的方框实际上可以基本并行地执行,它们有时也可以按相反的顺序执行,这依所涉及的功能而定。也要注意的是,框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合,可以用执行规定的功能或操作的专用的基于硬件的系统来实现,或者可以用专用硬件与计算机指令的组合来实现。
描述于本申请实施例中所涉及到的单元可以通过软件的方式实现,也可以通过硬件的方式来实现。所描述的单元也可以设置在处理器中,例如,可以描述为:一种处理器包括获取模块、第一权值确定模块、第二权值确定模块和位置预测模块。其中,这些单元的名称在某种情况下并不构成对该单元本身的限定,例如,获取模块还可以被描述为“获取用户的当前位置信息和当前时刻信息的模块”。
作为另一方面,本申请还提供了一种非易失性计算机存储介质,该非易失性计算机存储介质可以是上述实施例中所述装置中所包含的非易失性计算机存储介质;也可以是单独存在,未装配入终端中的非易失性计算机存储介质。上述非易失性计算机存储介质存储有一个或者多个程序,当所述一个或者多个程序被一个设备执行时,使得所述设备:获取用户的当前位置信息和当前时刻信息;基于当前位置信息和当前时刻信息,确定候选停留点集合中的各候选停留点在预定时刻的第一权值;基于预定时刻,确定候选停留点集合中的各候选停留点的第二权值;以及基于第一权值和第二权值,在候选停留点集合中确定与预定时刻对应的用户位置。
以上描述仅为本申请的较佳实施例以及对所运用技术原理的说 明。本领域技术人员应当理解,本申请中所涉及的发明范围,并不限于上述技术特征的特定组合而成的技术方案,同时也应涵盖在不脱离所述发明构思的情况下,由上述技术特征或其等同特征进行任意组合而形成的其它技术方案。例如上述特征与本申请中公开的(但不限于)具有类似功能的技术特征进行互相替换而形成的技术方案。

Claims (20)

  1. 一种预测预定时刻的用户位置的方法,其特征在于,包括:
    获取用户的当前位置信息和当前时刻信息;
    基于所述当前位置信息和所述当前时刻信息,确定候选停留点集合中的各候选停留点在所述预定时刻的第一权值;
    基于所述预定时刻,确定所述候选停留点集合中的各所述候选停留点的第二权值;以及
    基于所述第一权值和所述第二权值,在所述候选停留点集合中确定与所述预定时刻对应的所述用户位置。
  2. 根据权利要求1所述的方法,其特征在于,所述基于所述当前位置信息和所述当前时刻信息,确定候选停留点集合中的各候选停留点在所述预定时刻的第一权值包括:
    获取所述用户的历史停留点作为候选停留点;
    确定所述候选停留点集合中的第一候选停留点向第二候选停留点转移的转移概率;以及
    基于所述转移概率,确定各所述候选停留点在所述预定时刻的第一权值;
    其中,所述第一候选停留点和所述第二候选停留点均为所述候选停留点集合中的任意候选停留点,所述转移概率为预定时间间隔内,生成以所述第一候选停留点为起点,以所述第二候选停留点为终点的路径的概率。
  3. 根据权利要求2所述的方法,其特征在于,所述基于所述转移概率,确定各所述候选停留点在所述预定时刻的第一权值包括:
    基于所述转移概率,确定N×N阶转移矩阵S,其中,N为所述候选停留点集合中的所述候选停留点的数量;
    基于所述转移矩阵S,确定各所述候选停留点在所述预定时刻的第一权值P1
    其中:
    Figure PCTCN2016086215-appb-100001
    或者,
    Figure PCTCN2016086215-appb-100002
    t2为所述预定时间间隔,t1为所述预定时刻,t0为当前时刻。
  4. 根据权利要求1-3任意一项所述的方法,其特征在于,基于所述预定时刻,确定所述候选停留点集合中的各所述候选停留点的第二权值包括:
    获取所述用户的历史停留点作为候选停留点;
    获取与各所述候选停留点对应的历史时间信息;以及
    基于各所述候选停留点和与各所述候选停留点对应的历史时间信息,确定各候选停留点在所述预定时刻的第二权值。
  5. 根据权利要求4所述的方法,其特征在于,所述基于各所述候选停留点和与各所述候选停留点对应的历史时间信息,确定各候选停留点在所述预定时刻的第二权值包括:
    确定所述用户在多个预设的历史时间区间内,处于各所述候选停留点的停留概率;
    基于所述停留概率,生成K×N阶时间矩阵T,其中,K为所述历史时间区间的数量,N为所述候选停留点的数量;以及
    从所述时间矩阵中,确定与所述预定时刻对应的N维列向量Ti,其中,1≤i≤K,所述列向量中的各元素为各所述候选停留点在所述预定时刻的第二权值。
  6. 根据权利要求5所述的方法,其特征在于:
    各所述历史时间区间具有相同的时长。
  7. 根据权利要求5所述的方法,其特征在于:
    K为偶数;
    当1≤i≤K/2时,Ti中的各元素为工作日的第i个历史时间区间内, 各所述候选停留点的停留概率;
    当K/2+1≤i≤K时,Ti中的各元素为节假日的第i-K/2个历史时间区间内,各所述候选停留点的停留概率。
  8. 根据权利要求7所述的方法,其特征在于:
    K=48。
  9. 根据权利要求5所述的方法,其特征在于,所述基于所述第一权值和所述第二权值,在所述候选停留点集合中确定与所述预定时刻对应的所述用户位置包括:
    基于Ti×P1确定与所述预定时刻对应的各所述候选停留点的预测权值;以及
    将各所述候选停留点中,具有最大预测权值的候选停留点作为与所述预定时刻对应的所述用户位置。
  10. 一种预测预定时刻的用户位置的装置,其特征在于,包括:
    获取模块,配置用于获取用户的当前位置信息和当前时刻信息;
    第一权值确定模块,配置用于基于所述当前位置信息和所述当前时刻信息,确定候选停留点集合中的各候选停留点在所述预定时刻的第一权值;
    第二权值确定模块,配置用于基于所述预定时刻,确定所述候选停留点集合中的各所述候选停留点的第二权值;以及
    位置预测模块,配置用于基于所述第一权值和所述第二权值,在所述候选停留点集合中确定与所述预定时刻对应的所述用户位置。
  11. 根据权利要求10所述的装置,其特征在于,所述第一权值确定模块进一步配置用于:
    获取所述用户的历史停留点作为候选停留点;
    确定所述候选停留点集合中的第一候选停留点向第二候选停留点转移的转移概率;以及
    基于所述转移概率,确定各所述候选停留点在所述预定时刻的第一权值;
    其中,所述第一候选停留点和所述第二候选停留点均为所述候选停留点集合中的任意候选停留点,所述转移概率为预定时间间隔内,生成以所述第一候选停留点为起点,以所述第二候选停留点为终点的路径的概率。
  12. 根据权利要求11所述的装置,其特征在于,所述第一权值确定模块在基于所述转移概率,确定各所述候选停留点在所述预定时刻的第一权值时,进一步配置用于:
    基于所述转移概率,确定N×N阶转移矩阵S,其中,N为所述候选停留点集合中的所述候选停留点的数量;以及
    基于所述转移矩阵S,确定各所述候选停留点在所述预定时刻的第一权值P1
    其中:
    Figure PCTCN2016086215-appb-100003
    或者,
    Figure PCTCN2016086215-appb-100004
    t2为所述预定时间间隔,t1为所述预定时刻,t0为当前时刻。
  13. 根据权利要求10-12任意一项所述的装置,其特征在于,所述第二权值确定模块进一步配置用于:
    获取所述用户的历史停留点作为候选停留点;
    获取与各所述候选停留点对应的历史时间信息;以及
    基于各所述候选停留点和与各所述候选停留点对应的历史时间信息,确定各候选停留点在所述预定时刻的第二权值。
  14. 根据权利要求13所述的装置,其特征在于,所述第二权值确定模块在基于各所述候选停留点和与各所述候选停留点对应的历史时间信息,确定各候选停留点在所述预定时刻的第二权值时,进一步配置用于:
    确定所述用户在多个预设的历史时间区间内,处于各所述候选停留点的停留概率;
    基于所述停留概率,生成K×N阶时间矩阵T,其中,K为所述历史时间区间的数量,N为所述候选停留点的数量;以及
    从所述时间矩阵中,确定与所述预定时刻对应的N维列向量Ti,其中,1≤i≤K,所述列向量中的各元素为各所述候选停留点在所述预定时刻的第二权值。
  15. 根据权利要求14所述的装置,其特征在于:
    各所述历史时间区间具有相同的时长。
  16. 根据权利要求14所述的装置,其特征在于:
    K为偶数;
    当1≤i≤K/2时,Ti中的各元素为工作日的第i个历史时间区间内,各所述候选停留点的停留概率;
    当K/2+1≤i≤K时,Ti中的各元素为节假日的第i-K/2个历史时间区间内,各所述候选停留点的停留概率。
  17. 根据权利要求16所述的装置,其特征在于:
    K=48。
  18. 根据权利要求14所述的装置,其特征在于,位置预测模块进一步配置用于:
    基于Ti×P1确定与所述预定时刻对应的各所述候选停留点的预测权值;以及
    将各所述候选停留点中,具有最大预测权值的候选停留点作为与所述预定时刻对应的所述用户位置。
  19. 一种设备,包括:
    处理器;和
    存储器,
    所述存储器中存储有能够被所述处理器执行的计算机可读指令,在所述计算机可读指令被执行时,所述处理器执行预测预定时刻的用户位置的方法,所述方法包括:
    获取用户的当前位置信息和当前时刻信息;
    基于所述当前位置信息和所述当前时刻信息,确定候选停留点集合中的各候选停留点在所述预定时刻的第一权值;
    基于所述预定时刻,确定所述候选停留点集合中的各所述候选停留点的第二权值;以及
    基于所述第一权值和所述第二权值,在所述候选停留点集合中确定与所述预定时刻对应的所述用户位置。
  20. 一种非易失性计算机存储介质,所述计算机存储介质存储有能够被处理器执行的计算机可读指令,当所述计算机可读指令被处理器执行时,所述处理器执行预测预定时刻的用户位置的方法,所述方法包括:
    获取用户的当前位置信息和当前时刻信息;
    基于所述当前位置信息和所述当前时刻信息,确定候选停留点集合中的各候选停留点在所述预定时刻的第一权值;
    基于所述预定时刻,确定所述候选停留点集合中的各所述候选停留点的第二权值;以及
    基于所述第一权值和所述第二权值,在所述候选停留点集合中确定与所述预定时刻对应的所述用户位置。
PCT/CN2016/086215 2015-11-06 2016-06-17 预测预定时刻的用户位置的方法和装置 WO2017076004A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510753326.6A CN105357638B (zh) 2015-11-06 2015-11-06 预测预定时刻的用户位置的方法和装置
CN201510753326.6 2015-11-06

Publications (1)

Publication Number Publication Date
WO2017076004A1 true WO2017076004A1 (zh) 2017-05-11

Family

ID=55333471

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/086215 WO2017076004A1 (zh) 2015-11-06 2016-06-17 预测预定时刻的用户位置的方法和装置

Country Status (2)

Country Link
CN (1) CN105357638B (zh)
WO (1) WO2017076004A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112616073A (zh) * 2020-11-17 2021-04-06 北京达佳互联信息技术有限公司 视频推送方法、装置、服务器和存储介质
CN113589338A (zh) * 2021-07-29 2021-11-02 成都乐动信息技术有限公司 检测用户在运动过程中的停留点的方法、装置及电子设备
CN115083147A (zh) * 2022-05-10 2022-09-20 广州汽车集团股份有限公司 目的地预测方法、装置以及车辆
CN116485047A (zh) * 2023-06-25 2023-07-25 广东省科学院广州地理研究所 获取候鸟迁徙中途停歇地的方法和装置

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105357638B (zh) * 2015-11-06 2019-10-22 百度在线网络技术(北京)有限公司 预测预定时刻的用户位置的方法和装置
CN110008414B (zh) * 2016-03-31 2021-07-27 百度在线网络技术(北京)有限公司 地理信息点的确定方法和装置
CN106101224B (zh) * 2016-06-12 2018-07-17 腾讯科技(深圳)有限公司 识别用户所在地理位置的类别的方法及装置
CN106202236A (zh) * 2016-06-28 2016-12-07 联想(北京)有限公司 一种用户位置预测方法及装置
CN108063802B (zh) * 2017-12-01 2020-07-28 南京邮电大学 基于边缘计算的用户位置动态性建模优化方法
CN108154329A (zh) * 2017-12-25 2018-06-12 维沃移动通信有限公司 一种自取件的提醒方法及装置
CN108256050B (zh) * 2018-01-12 2020-08-04 携程旅游网络技术(上海)有限公司 行程订单的关联服务推送方法、系统、设备和存储介质
CN109803228B (zh) * 2018-12-28 2020-10-27 中国联合网络通信集团有限公司 业务热点区域定位方法、装置、设备和可读介质
CN110781855B (zh) * 2019-11-04 2022-12-06 浙江大华技术股份有限公司 目标停留点的确定方法、装置、设备及存储装置
CN111222057B (zh) * 2019-11-20 2024-01-12 腾讯科技(深圳)有限公司 一种信息处理方法、装置及计算机可读存储介质
CN111006682B (zh) * 2019-12-31 2023-11-21 斑马网络技术有限公司 导航路线规划方法、装置、电子设备及存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104581622A (zh) * 2013-10-28 2015-04-29 华为技术有限公司 移动用户位置预测方法及设备
CN104796866A (zh) * 2015-05-06 2015-07-22 北京我联科技有限公司 室内定位方法和装置
CN104931041A (zh) * 2015-05-03 2015-09-23 西北工业大学 一种基于用户轨迹数据的地点序列预测方法
CN105357638A (zh) * 2015-11-06 2016-02-24 百度在线网络技术(北京)有限公司 预测预定时刻的用户位置的方法和装置

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2756183A1 (en) * 2009-03-30 2010-10-07 Aha Mobile, Inc. Predictive search with location-based application
US8639803B2 (en) * 2010-09-01 2014-01-28 Telefonaktiebolaget L M Ericsson (Publ) Systems and method for predicting the future location of an entity
EP3101561A1 (en) * 2012-06-22 2016-12-07 Google, Inc. Ranking nearby destinations based on visit likelihoods and predicting future visits to places from location history
CN103942229B (zh) * 2013-01-22 2017-05-03 日电(中国)有限公司 目的地预测设备和方法
CN104833365B (zh) * 2014-02-12 2017-12-08 华为技术有限公司 一种用户目的地点的预测方法及装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104581622A (zh) * 2013-10-28 2015-04-29 华为技术有限公司 移动用户位置预测方法及设备
CN104931041A (zh) * 2015-05-03 2015-09-23 西北工业大学 一种基于用户轨迹数据的地点序列预测方法
CN104796866A (zh) * 2015-05-06 2015-07-22 北京我联科技有限公司 室内定位方法和装置
CN105357638A (zh) * 2015-11-06 2016-02-24 百度在线网络技术(北京)有限公司 预测预定时刻的用户位置的方法和装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
YU , RUIYUN ET AL.: "Social-Aware Mobile User Location Prediction Algorithm in Participatory Sensing Systems", CHINESE JOURNAL OF COMPUTERS, vol. 38, no. 2, 28 February 2015 (2015-02-28), pages 377 - 380 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112616073A (zh) * 2020-11-17 2021-04-06 北京达佳互联信息技术有限公司 视频推送方法、装置、服务器和存储介质
CN112616073B (zh) * 2020-11-17 2022-09-09 北京达佳互联信息技术有限公司 视频推送方法、装置、服务器和存储介质
CN113589338A (zh) * 2021-07-29 2021-11-02 成都乐动信息技术有限公司 检测用户在运动过程中的停留点的方法、装置及电子设备
CN115083147A (zh) * 2022-05-10 2022-09-20 广州汽车集团股份有限公司 目的地预测方法、装置以及车辆
CN116485047A (zh) * 2023-06-25 2023-07-25 广东省科学院广州地理研究所 获取候鸟迁徙中途停歇地的方法和装置
CN116485047B (zh) * 2023-06-25 2023-12-26 广东省科学院广州地理研究所 获取候鸟迁徙中途停歇地的方法和装置

Also Published As

Publication number Publication date
CN105357638A (zh) 2016-02-24
CN105357638B (zh) 2019-10-22

Similar Documents

Publication Publication Date Title
WO2017076004A1 (zh) 预测预定时刻的用户位置的方法和装置
CN109460514B (zh) 用于推送信息的方法和装置
US11099730B2 (en) Map interface interaction
JP6967612B2 (ja) 情報検索方法、装置及びシステム
WO2017118345A1 (zh) 电子书籍的推送方法和装置
WO2017071251A1 (zh) 信息推送方法和装置
CN106471851B (zh) 基于学习模型的设备定位
US10904707B2 (en) Location-based service implementing method and apparatus
WO2013126244A1 (en) Content pre-fetching for computing devices
CN109672978B (zh) 一种无线热点扫描频率控制方法及装置
US10417274B2 (en) Method and apparatus for pushing information
CN107241215B (zh) 用户行为预测方法和装置
JP2012022530A (ja) 行動予測装置、方法及びプログラム
US20190392480A1 (en) Method and apparatus for guiding service flow
CN108810047B (zh) 用于确定信息推送准确率的方法、装置及服务器
CN110866040B (zh) 用户画像生成方法、装置和系统
CN106776707A (zh) 信息推送的方法和装置
CN110009101B (zh) 用于生成量化神经网络的方法和装置
CN109413212B (zh) 用于处理请求的方法和装置
CN105432038A (zh) 应用排名计算装置和使用信息收集装置
CN111898061B (zh) 搜索网络的方法、装置、电子设备和计算机可读介质
Zhai et al. Optimal task partition with delay requirement in mobile crowdsourcing
CN112218114A (zh) 视频缓存控制方法、装置和计算机可读存储介质
CN112883291B (zh) 目的地位置的推荐方法、装置和服务器
US20170249707A1 (en) Intelligent share queuing

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: 16861274

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: 16861274

Country of ref document: EP

Kind code of ref document: A1