Attribution method and system for outbound travel trajectory of user, and medium and program product

Abstract

Provided in the present application are an attribution method and system for an outbound travel trajectory of a user, and a computer-readable storage medium and a computer program product. In the method, on the basis of a user trip and gate information, the number of outbound times of a user at each gate of a gate set of a subway station is determined, and on this basis, an exit to which the user belongs is determined; and an outbound travel trajectory of the user is made to correspond to the exit to which the user belongs, so as to solve the technical problem of it being impossible to realize localization during subway travel and thus obtain the outbound travel trajectory of the user, thereby avoiding the limitation of existing realization that relies on buried points to determine a subway exit.

Description

User outbound travel trajectory attribution method and system, medium, program product TECHNICAL FIELD

[0001] The present application relates to the technical field of computer application, in particular to a user outbound travel trajectory attribution method and system, computer readable storage medium and computer program product. BACKGROUND

[0002] User travel, whether active in the subway station or active in the tunnel between subway stations, is difficult to obtain its own positioning in order to meet the needs of terminal applications.

[0003] The determination of the geographic location of the terminal application in the subway exit is often achieved by GPS (Global Positioning System, Global Positioning System) positioning, but it is limited by the poor or even no GPS signal in the subway, so the terminal application is difficult to determine the geographic location through the GPS signal in the process of subway exit.

[0004] In addition, in addition to being limited by the lack of GPS signal or poor GPS signal in the subway, the terminal application for obtaining positioning has reasons such as denied location permission, and cannot implement positioning for the user's outbound travel trajectory to determine the outbound port to which the user belongs. SUMMARY

[0005] One purpose of the present application is to solve the technical problem of being unable to achieve positioning in subway travel and thus obtain the user's outbound travel trajectory, and to accurately correspond the outbound port for the user's outbound travel trajectory.

[0006] According to one aspect of an embodiment of the present application, a user outbound travel trajectory attribution method is disclosed, the method comprising:

[0007] Obtaining the number of times the user exits each gate of the gate group at the subway station;

[0008] According to the distribution of the number of times the user exits each gate of the gate group at the subway station and the user portrait, predicting the outbound port to which the user belongs, the outbound port being a passable outbound port from the gate group;

[0009] Corresponding the user's travel trajectory to the outbound port to which the user belongs according to the outbound port to which the user belongs;

[0010] Aggregating the users passing through the outbound port through the travel trajectory and the user portrait, and assigning the users aggregated to the location mapped by the outbound port.

[0011] According to an aspect of the embodiment of the present application, the obtaining the number of times of the user's exiting from each gate of the group of gates of the subway station comprises:

[0012] obtaining the travel data of the user, the travel data at least describing the user's exiting from the exit station and the exit gate involved in the generated user travel;

[0013] obtaining the number of times of the user's exiting from each gate of the group of gates of the subway station from the gate information according to the user's exit station and the exit gate of the travel data of the user.

[0014] According to an aspect of the embodiment of the present application, the predicting the exit port to which the user belongs according to the distribution of the number of times of the user's exiting from each gate of the group of gates of the subway station and the user portrait comprises:

[0015] predicting according to the number of times of the user's exiting from each gate of the group of gates of the subway station and the user portrait to obtain the probability of the user's exiting direction, the exiting direction corresponding to at least one of the exit ports;

[0016] determining the exit port to which the user belongs according to the probability of the user's exiting direction.

[0017] According to an aspect of the embodiment of the present application, before the predicting the exit port to which the user belongs according to the distribution of the number of times of the user's exiting from each gate of the group of gates of the subway station and the user portrait, the method further comprises:

[0018] judging whether the number of times of the user's exiting from the group of gates of the subway station is less than a set threshold value, and if the number of times is less than the set threshold value, jumping to perform the prediction of the user's exiting direction.

[0019] According to an aspect of the embodiment of the present application, the method further comprises:

[0020] if the number of times of the user's exiting from the group of gates of the subway station is not less than the set threshold value, determining the exit port to which the user belongs by performing historical exit port bias statistics.

[0021] According to an aspect of the embodiment of the present application, the determining the exit port to which the user belongs by performing the historical exit port bias statistics comprises:

[0022] determining the gate to which the user is biased in the group of gates according to the distribution statistics of the number of times of the user's exiting from each gate of the group of gates of the subway station;

[0023] determining the exit port located on the same side as the gate as the exit port to which the user belongs according to the location of the gate.

[0024] According to an aspect of the embodiments of the present application, after the users passing through the exit are aggregated by the travel trajectory and the user portrait, the method further comprises:

[0025] The user portrait of the aggregated users is assigned to the position corresponding to the exit for delivery control on the position, and the position includes a media position and a store position.

[0026] According to an aspect of the embodiments of the present application, the present application discloses a user outbound travel trajectory attribution system, which comprises a memory, a processor and a computer program stored in the memory, and the processor executes the computer program to realize the steps of the method according to any one of the preceding embodiments.

[0027] According to an aspect of the embodiments of the present application, the present application discloses a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the steps of the method according to any one of the preceding embodiments.

[0028] According to an aspect of the embodiments of the present application, the present application discloses a computer program product, which comprises a computer program, and the computer program is executed by a processor to realize the steps of the method according to any one of the preceding embodiments.

[0029] The embodiments of the present application attribute the outbound travel trajectory of the user, can accurately correspond the outbound travel trajectory to the exit, are no longer limited by the signal problem of GPS positioning and the acquisition problem of position permission, specifically, the outbound times of the user at each gate of the gate group of the subway station are acquired first, then the outbound exit of the user is predicted according to the distribution of the outbound times of the user at each gate of the gate group of the subway station and the user portrait, the outbound exit is the outbound exit through which the user can pass from the gate group, the travel trajectory of the user is corresponded to the outbound exit to which the user belongs, finally, the users passing through the exit are aggregated by the travel trajectory and the user portrait, and then the aggregated users are assigned to the position mapped by the exit, the whole process does not need GPS positioning, and also does not need to obtain the position permission since the position is not acquired, and the outbound travel trajectory of the user is accurately corresponded to the exit by another way.

[0030] Other characteristics and advantages of the present application will become apparent from the following detailed description, or will be learned by practice of the present application.

[0031] It should be understood that the foregoing general description and the following detailed description are only exemplary, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0032] The foregoing and other objects, features and advantages of the present application will become more apparent from the following detailed description, which proceeds with reference to the accompanying drawings.

[0033] FIG. 1 shows a flow chart of a method for attributing a user outbound travel trajectory according to an embodiment of the present application.

[0034] FIG. 2 is a flow chart of a method for describing the step of obtaining the number of times a user passes through each gate of a group of gates in a subway station according to the corresponding embodiment of FIG. 1.

[0035] FIG. 3 is a flow chart of a method for describing the step of predicting the outbound port of a user according to the distribution of the number of times the user passes through each gate of a group of gates in a subway station and the user portrait according to the corresponding embodiment of FIG. 1.

[0036] FIG. 4 is a flow chart of a method for describing the step of determining the outbound port of a user attribute by performing historical outbound port bias statistics according to an embodiment.

[0037] FIG. 5 is a schematic diagram of a subway station scenario according to an embodiment.

[0038] FIG. 6 is a data flow diagram according to an embodiment. DETAILED DESCRIPTION

[0039] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations may, however, be implemented in many different forms and should not be construed as limited to the examples set forth herein; rather, these example implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of example implementations to those skilled in the art. Like reference numerals refer to like elements throughout the various figures and examples. The figures are not necessarily drawn to scale, and the size of some of the elements can have been exaggerated for the sake of illustration.

[0040] Furthermore, described features, structures, or characteristics can be combined in any suitable manner in one or more example implementations. In the following description, numerous specific details are provided to give a thorough understanding of example implementations. One skilled in the relevant art will recognize, however, that the techniques described herein can be practiced without one or more of the specific details, or with other methods, components, materials, and so forth. In other instances, well-known structures, materials, or operations are not shown or described in detail in order to avoid obscuring aspects of the example implementations.

[0041] Some of the block diagrams in the drawings are functional entities that do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0042] The positioning of the current location by the terminal application is mostly obtained by GPS positioning after obtaining the app (Application) location permission.

[0043] That is, the accurate positioning in the prior art relies on two conditions, i.e., obtaining the app location permission and stable GPS signal.

[0044] Once the app location permission is refused or the GPS signal is poor or even no GPS signal, the positioning of the current location cannot be realized.

[0045] Therefore, the subway travel is also limited by the two conditions to realize the tracking and positioning of the outbound travel trajectory, and the outbound port of the user cannot be known due to the failure of positioning, and the travel of the user cannot be corresponded to the outbound port.

[0046] In view of the failure of positioning in the subway travel, without relying on the GPS signal and without obtaining the location permission, the present application provides a method for attributing the outbound travel trajectory of the user.

[0047] Referring to FIG. 1, FIG. 1 shows a flowchart of a method for attributing the outbound travel trajectory of the user according to an embodiment of the present application. The method for attributing the outbound travel trajectory of the user provided by the present application comprises:

[0048] Step S110, obtaining the outbound times of the user at each gate of the gate group of the subway station;

[0049] Step S120, predicting the outbound port to which the user belongs according to the distribution of the outbound times of the user at each gate of the gate group of the subway station and the user portrait, the outbound port being the passable outbound port from the gate group;

[0050] Step S130, corresponding the travel trajectory of the user to the outbound port to which the user belongs;

[0051] Step S140, aggregating the users passing through the outbound port according to the travel trajectory and the user portrait, and giving the location of the outbound port mapped by the aggregated users.

[0052] The steps will be described in detail below.

[0053] Without positioning based on location permission by GPS in the subway station hall and tunnel section between stations, the exit of the user can also be obtained, and the user's journey and travel trajectory generated by subway travel can be accurately corresponded to the exit, greatly solving the problem that the location cannot be obtained during subway travel, especially the exit of subway travel, and enabling the terminal application function to be accurately and quickly realized based on location during subway travel.

[0054] The present application is oriented to the exit of the user, and the travel trajectory of the user is implemented to determine the exit to which the user belongs, so that the location in the direction corresponding to the exit will be the location requested by the terminal application to realize its own function for the user.

[0055] In step S110, the user will perform an exit travel trajectory attribution process for the subway station where the user exits in the user's journey, so as to accurately correspond to the exit of the user in the travel trajectory of the subway travel.

[0056] In the execution of step S110, the number of exits of each gate of the exit gate group is obtained for the user at the subway station where the user exits, i.e. the number of exits of each gate of the exit gate group of the user at the subway station.

[0057] The number of exits of each gate of the user is obtained from the origin-destination (OD) data of the user. For example, the OD data of the user indicates the user's journey when the user enters and exits the station, and the exit station and exit gate involved in the generated user journey.

[0058] For each user journey generated by the user, the exit of the user from a certain gate at a certain station can be obtained, so that the number of exits of each gate of the user at a station can be determined by counting all the user journeys of the user.

[0059] On this basis, combined with the existing gate information, the gate group where each gate is located and the line and city where the gate group is located are determined, and finally the number of exits of each gate of the gate group at the subway station of the user is obtained.

[0060] In an exemplary embodiment, the journey data exists in the form of a user journey table. Each user journey generated when the user enters and exits the station is stored in the constructed user journey table.

[0061] In the user journey table, the field information in the user journey includes date, journey identification (journey id), user identification (user id), city identification (city id), line serial number, entry time, entry station code, entry gate code, exit time, exit station code and exit gate code, etc., which are not listed one by one.

[0062] In an exemplary embodiment, the gate information is collected based on the station design, and is used to indicate the location of each gate. The gate information is in the form of a gate information table, and the field information in the table includes: gate code, city identification, line serial number, station code, gate group serial number in the station, gate serial number in the station, and device type.

[0063] Through the statistics of the trip data and the matching of the trip data and the gate information, the number of times of the user's exit at a gate and the location of the gate, in particular, the gate group where the gate is located, can be determined, so as to finally determine the number of times of the user's exit at each gate in the gate group in the subway station.

[0064] Through the execution of step S110, the subway exit at each gate in the gate group in each station is recorded and counted, so as to correspond the exit track of the user's exit at this station to the exit gate.

[0065] Please also refer to FIG. 2, which is a method flowchart for describing the step of acquiring the number of times of the user's exit at each gate in the gate group in the subway station, according to the corresponding embodiment of FIG. 1.

[0066] The step S110 provided by the embodiment of the present application for acquiring the number of times of the user's exit at each gate in the gate group in the subway station includes:

[0067] Step S111: acquiring the trip data of the user, the trip data at least describing the exit station and the exit gate involved in the generated user trip of the user in content;

[0068] Step S112: obtaining the number of times of the user's exit at each gate in the gate group in the subway station from the gate information according to the exit station and the exit gate of the user, based on the trip data of the user.

[0069] The two steps will be described in detail below.

[0070] The trip data of the user is acquired, and the user trip table is acquired as an example. The user trip and the exit gate code in the user trip are obtained from the user trip table.

[0071] For the number of times of the user's exit at a gate, the number of times of the user's exit at the corresponding gate is counted according to the exit gate code of each user trip, and the gate group where the gate is located is determined based on the gate information, so as to finally obtain the number of times of the user's exit at each gate in the gate group in the subway station.

[0072] The gate information is exemplarily the gate information table as mentioned above. The gate information table is constructed by collecting the gate information of each gate in each gate group in each subway station, so as to obtain the gate information table.

[0073] Further, the gate information table represents the position of the gate, and the gate code contained therein indicates the unique code of the gate in the city. Exemplarily, the gate code can be composed of the station code, the device type and the device number, and the device number is the in-station serial number of the gate in the gate information table.

[0074] For example, the station code is 0253, the device type is 01, and the device number is 07, then the gate number is 02530107.

[0075] As mentioned above, in addition to the gate code, the fields in the gate information table also include the city identifier, the line serial number, the station code, the in-station gate group serial number, the in-station gate serial number and the device type.

[0076] The city identifier is used to uniquely identify the city in which it is located, in other words, each city has a unique city identifier; the station code is used to uniquely identify the station in the city, so that each subway station has a unique code in the city; the in-station gate group serial number is used to identify each group of gates in the station, for example, if there are 4 rows of gates in the station, then the serial numbers 1-4 are correspondingly assigned; the in-station gate serial number is the device number, for example, if there are 10 gates in the station, then the device number can be the serial numbers 1-10.

[0077] Based on the gate information table and the user travel table, the user obtains the number of times of exiting at each gate in the gate group of the subway station, which describes the user's exiting behavior. The user who generates the user travel by taking the subway obtains the number of times of exiting at the exiting gate in the gate group of the subway station corresponding to the destination station in the user travel by executing step S110, and further corresponds the user travel to the exit port when the user arrives at the corresponding subway station based on the data, so as to realize the exit port positioning of the user in the exiting process.

[0078] Step S110 obtains the number of times of exiting of the user at each gate in the gate group of a subway station, and step S120 predicts the exit port to which the user belongs based on the distribution of the number of times of exiting and the corresponding user portrait, so as to determine the most possible exit port for the user travel of the user.

[0079] In step S120, the user portrait is used to describe the user state of the user, which can be composed of the user label constructed. Exemplarily, the user portrait exists in the form of a user label table, which includes a user identifier and a plurality of user labels corresponding to the user identifier.

[0080] The user identification uniquely identifies the user, and the user label includes but is not limited to gender, age, life stage, occupation, total number of rides in the past 7 / 14 / 30 / 90 days, total ride time in the past 7 / 14 / 30 / 90 days, average arrival time of the first user trip on weekdays in the past 7 / 14 / 30 / 90 days, average departure time of the last trip on weekdays in the past 7 / 14 / 30 / 90 days, average commuting distance per trip in the past 7 / 14 / 30 / 90 days, total number of rides, high-frequency stations, total number of commercial benefits received, total number of commercial benefits redeemed, total amount of commercial benefits redeemed, average single price of commercial benefits redeemed, total number of travel benefits received, number of each card, and frequency of using the card.

[0081] The distribution of the number of times the user exits at each gate of the gate group in the subway station describes the number of times the single user exits at gate 1 of the gate group, the number of times the single user exits at gate 2 of the gate group, …, and the number of times the single user exits at gate n_zj (n_zj represents the number of gates in the entire gate group) of the gate group.

[0082] The distribution of the number of times the user exits at each gate of the gate group in the subway station and the user portrait are used as features of the user to predict the exit port to which the user belongs.

[0083] It should be understood that for a user, the number of times the user exits at a single gate group in a certain time range, such as a single month, can be low, and is affected by various factors such as actual conditions. Therefore, using the distribution of the number of times the user exits at each gate of the gate group in the subway station and the user portrait as features of the user provides a data basis for predicting the exit port to which the user belongs, and also improves the richness and scalability of the features, which is beneficial to achieving accurate prediction based on existing data.

[0084] The user belongs to the exit port predicted by the user, which should be noted that the exit port obtained by the prediction is an exit port corresponding to the gate group, which is the most likely exit port that the user can pass through after exiting from the gate group.

[0085] A gate group corresponds to at least one exit port, and the exit port that the user is most likely to pass through after exiting from a gate of the gate group is the exit port to which the user is predicted to belong.

[0086] The prediction of the exit port to which the user belongs can obtain the probability of the user belonging to each exit port through a pre-trained classification model, and then determine the exit port to which the user belongs based on the probability.

[0087] For example, the pre-trained can be a binary classification model, which uses the distribution of the number of times the user exits at each gate of the gate group in the subway station and the user portrait as features, and according to the direction of the user after exiting, the corresponding direction value of the features is labeled. Each direction value uniquely corresponds to an exit port that can be passed through after exiting from the gate group.

[0088] For example, if the user passes left after exiting the gate group, the direction value is marked as 0; and correspondingly, if the user passes right after exiting the gate group, the direction value is marked as 1.

[0089] On this basis, the training data set is constructed , wherein each single record in the training data set is a feature of the record, which is used to describe the user state of the corresponding user.

[0090] Specifically, includes the distribution of the number of times that the user exits each gate of the gate group in the subway station and the user portrait, and the user portrait can be represented by multiple user labels. is the feature of user i at gate group j, is the m-dimensional real number space of the m-dimensional feature mapping of user i.

[0091] Exemplarily, =[date (YYYYMM), user label 1 of user i, user label 2 of user i, …, user label of user i at gate 1 of gate group j, the number of times that user i exits gate 2 of gate group j, …, the number of times that user i exits gate of gate group j], date (YYYYMM) represents the month, such as 202312, which indicates December 2023, and the date feature indicates the time range corresponding to other features. The user label refers to the th user label of user i, i.e. represents the number of users, represents the number of gate groups that user i faces when exiting the subway, represents the number of user labels, is the direction value marked for the feature in the single record of the corresponding training data set. represents the direction of the final exit passage of user i corresponding to the feature (especially the actual number of exits).

[0092] That is, the in the single record of the training data set is the marked direction value. For example, if user i exits gate group j to the left, then , and if to the right, then .

[0093] ​The model is trained through the constructed training data set, so as to obtain a model capable of accurately predicting the outbound port to which the user belongs, thereby providing accurate judgment of the outbound port to which the user belongs, and no longer needing to acquire the position, that is, no longer being limited by GPS positioning and acquisition of position permission.

[0094] Further, the training for outbound port prediction is performed on the gate group corresponding to only the unique outbound port. The construction of the training data set is performed on the gate group corresponding to the unique outbound port, features are constructed according to all records related to the gate group, and the constructed features are labeled with direction values according to the unique direction of the gate group.

[0095] Specifically, the outbound times of the user at each gate of the gate group in the subway station are obtained through the trip data and the gate information, that is, the user-gate group-outbound times data of the user are obtained, which can be in the form of a table, that is, a table maintained according to the fields of user, subway station, gate group, gate, and outbound times, and the user-gate group-outbound times data corresponding to only the unique outbound port are acquired from the table, the outbound times of the user at each gate of the gate group in the subway station are obtained from the user-gate group-outbound times data corresponding to only the unique outbound port, and features are constructed accordingly.

[0096] In addition, the corresponding user portrait is extracted for feature construction, and the constructed features are labeled with direction values to finally form a training data set.

[0097] The constructed training data set screens out the influence of other outbound ports, only considers the prediction of the outbound port to which the user belongs in the feature dimension, so that the subsequent prediction is not affected by other factors, and the accuracy and reliability of the prediction are greatly improved.

[0098] Exemplarily, the gate group entrance and exit distance relationship is also constructed, which is used to indicate the distance of each gate of the gate group in a station to the outbound port in each direction, so that after the most likely direction of the user after exiting the gate is predicted, the most close outbound port can be directly mapped according to the gate group entrance and exit distance relationship, and the outbound port is the outbound port to which the user belongs.

[0099] As described above, the model trained through the constructed training data set can be a binary classification model, and thus the obtained binary classification model is used to realize the prediction of the outbound port to which the user belongs. Specifically, the binary classification model is used to predict the most likely direction of the user after exiting the gate group according to the outbound times of the user at each gate of the gate group in the subway station and the user portrait, and the outbound port mapped by the direction is the outbound port to which the user belongs.

[0100] Further, the outbound times of the user at each gate of the gate group in the subway station and the user portrait are used to construct a prediction data set Unlike the training dataset, the prediction dataset is not limited to the gate group corresponding to a unique exit. In other words, regardless of whether the gate group corresponds to a unique exit, all of the user's data will be used to construct the prediction dataset, and then the exit to which the user belongs will be predicted by the features of the constructed prediction dataset.

[0101] Prediction dataset In the middle, each record This represents the characteristics of user i in gate group j, whose exit point has not yet been determined. As mentioned earlier, these characteristics describe the number of times user i exits each gate in gate group j and the user's tag. The list of included characteristics is similar to... similar.

[0102] Specifically, in order to obtain the applicable binary classification model First, the problem loss function for the t-th iteration is constructed as follows: ,in, It is the predicted value of the ij-th record in the t-th iteration. For the t-th tree, q represents tree The structure, w is a tree The weights of the middle leaves, and using Let represent the weight of the i-th leaf. It's about trees. The regularization term with complexity of is then At point By making a second-order approximation, we get: ,

[0103] in, ;

[0104] Therefore, the optimal solution to the problem is: , where q is a tree with a fixed structure, and γ and λ are the coefficients of the first-order and second-order regularization terms, respectively.

[0105] Based on the iterative convergence of the optimal solution in each round on the training dataset, a binary classification model is obtained. .

[0106] Based on the obtained binary classification model , will predict the dataset Each record Use binary classification models one by one The prediction is performed to obtain the characteristic bias of user i in gate group j, that is, the probability that user i belongs to the exit corresponding to gate group j. .

[0107] For example, if the gate group j corresponds to two outbound gates. In other words, the user i has two directions after the gate group j, each direction corresponds to an outbound gate, then the probability of the user i being inclined to another direction, i.e. belonging to another outbound gate, is In addition, if there are more than two outbound gates in one direction of the gate group j, i.e. the outbound gates in this direction are not unique, when the binary classification model predicts that the user i is inclined to the direction with more than two outbound gates after the gate group j, the outbound gate that the user i is most likely to pass through will be determined according to the flow of each outbound gate, which is the outbound gate to which the user i belongs.

[0108] Specifically, each outbound gate in one direction has its flow, which can be perceived by the passenger flow counter set by the outbound gate. The flow proportion of each outbound gate is determined according to the flow corresponding to each outbound gate, and then the outbound gate to which the user i belongs is determined according to the flow proportion.

[0109] Further explanation, for determining the outbound gate that the user i is most likely to pass through according to the flow of each outbound gate is a process of determining the outbound gate to which the user i belongs by taking a random number in proportion. The proportion referred to is the proportion of the flow proportion of each outbound gate.

[0110] After the binary classification model predicts the direction to which the user i is inclined after the gate group j, if there are multiple outbound gates in the direction to which the user i is inclined, the outbound gate to which the user i belongs will be determined based on the flow proportion of each outbound gate and the random number taken for the user i.

[0111] For multiple outbound gates in the direction to which the user i is inclined, the random number range mapped by each outbound gate is constructed according to the flow proportion of each outbound gate. For example, there are three outbound gates in the same direction, and the flow proportion of each outbound gate between the three outbound gates is 0.3, 0.2 and 0.5 respectively. Therefore, the random number range mapped by each outbound gate is constructed, i.e. the random number range constructed by the first outbound gate with a flow proportion of 0.3 is [0, 0.3), the random number range constructed by the second outbound gate with a flow proportion of 0.2 is [0.3, 0.3+0.2), and the random number range constructed by the third outbound gate with a flow proportion of 0.5 is [0.3+0.2, 0.3+0.2+0.5).

[0112] A random number between 0 and 1 is taken for the user i, and the outbound port attributed to the user i is determined according to the random number range into which the obtained random number falls. If the random number falls in the random number range [0, 0.3), the user journey of the user i corresponds to the first outbound port; if the random number falls in the random number range [0.3, 0.3+0.2), the user i is attributed to the second outbound port; if the random number falls in the random number range [0.3+0.2, 0.3+0.2+0.5), the user i is attributed to the third outbound port.

[0113] Please also refer to FIG. 3, which is a method flowchart illustrating the step of attributing the outbound port of a user according to the distribution of the number of times the user exits at each gate of a gate set of a subway station and the user portrait, according to the corresponding embodiment of FIG. 1.

[0114] The step S120 of attributing the outbound port of a user according to the distribution of the number of times the user exits at each gate of a gate set of a subway station and the user portrait provided by the embodiments of the present application comprises:

[0115] In step S121, the probability of the outbound direction of a user is obtained according to the number of times the user exits at each gate of a gate set of a subway station and the prediction based on the user portrait.

[0116] In step S122, the outbound port to which the user is attributed is determined according to the probability of the outbound direction of the user.

[0117] The two steps will be described in detail below.

[0118] In the execution of step S121, the features are constructed from the number of times the user exits at each gate of a gate set of a subway station and the user labels contained in the user portrait, and the prediction is performed through a pre-trained model, such as the binary classification model mentioned above, to obtain the probability of each outbound direction after the user exits at the gate set of the designated subway station.

[0119] The directions through which the user can pass after exiting at a gate set are more than one, for example, in many cases, the user will have one or two outbound directions after exiting at a gate set, and there are passable outbound ports in each outbound direction.

[0120] After obtaining the probability of the outbound direction of the user, the direction through which the user passes after exiting at the gate set can be determined from the probability, i.e., the outbound direction corresponding to the high probability.

[0121] The outbound port in the outbound direction corresponding to the high probability is determined. In an exemplary embodiment, in the case where there is only one outbound port in the outbound direction corresponding to the high probability, the outbound port closest to the gate set exit distance relationship is determined according to the outbound direction corresponding to the high probability, and this outbound port is taken as the outbound port to which the user is attributed.

[0122] In another exemplary embodiment, there are more than two outbound ports in the outbound direction corresponding to the high probability, for example, bifurcated into more than two outbound ports via the import and export channel in this outbound direction, at this time, the outbound port to which the user belongs will be determined from the more than two outbound ports according to the proportion of the flow of each outbound port and the random number obtained.

[0123] Therefore, it is no longer necessary to obtain the location of the user, that is, it is no longer necessary to obtain the location permission and then perform GPS positioning, and the user's user journey and user trajectory can be accurately corresponded to the predicted outbound port, and accurate prediction of the user's trajectory after the user exits is realized.

[0124] In another embodiment of the present application, before the step S120 of predicting the outbound direction to which the user belongs according to the distribution of the number of times the user exits each gate of the gate group of the subway station and the user portrait, the user outbound travel trajectory attribution method provided by the present application further comprises the following steps:

[0125] It is judged whether the number of times the user exits each gate of the gate group of the subway station is less than a set threshold value, and if it is less than the set threshold value, the step S120 of predicting the outbound direction to which the user belongs is executed.

[0126] If the number of times the user exits each gate of the gate group of the subway station is not less than the set threshold value, the outbound port to which the user belongs is determined by performing historical outbound port bias statistics.

[0127] Specifically, it should be first pointed out that if a gate group is frequently exited by the user, it indicates that the gate corresponding to the gate group in an outbound direction is the user's preferred gate. For such users, the prediction of the outbound direction to which the user belongs will not be performed, and the outbound port to which the user belongs can be determined only by the bias statistics of the historical outbound port, thereby reducing the calculation complexity and improving the execution efficiency.

[0128] The number of times the user exits the gate group of the subway station is used to judge whether the user frequently exits the current gate group of the subway station. The number of times the user exits the gate group of the subway station can be obtained based on the journey data and gate information of the user, further, can be obtained from the user gate group exit data constructed by the journey data and the gate information, of course, is not limited to this, but also can be obtained by statistics of the user's exit data on each gate of the gate group of the subway station, which is not limited here.

[0129] For example, the set threshold value can be 5, that is, if the number of times the user exits a gate group in a subway station is not less than 5 times within a set time range, it indicates that the outbound direction corresponding to the gate group and the outbound port in the outbound direction are consistent with the user's outbound travel, and thus the outbound port can be determined as the user's frequently used outbound port.

[0130] Further, for the outbound times not less than the set threshold, the gate to which the user is inclined is determined, that is, according to the outbound times distribution of the user at the gate group, the left half group or the right half group of the gate group to which the user is inclined is counted, and no matter the left half group gate or the right half group gate, it corresponds to an outbound direction, and the outbound direction of the user passing after outbound is determined by the inclined gate, and then the outbound port in the outbound direction is the outbound port of the user.

[0131] At this point, it should be noted that if the number of gates in the gate group is odd, the middle gate belongs to both the left half group gate and the right half group gate, therefore, the outbound of the user at the middle gate will be counted as the outbound times of the user at the left half group gate and the right half group gate.

[0132] If the outbound times of the user at the gate group of the subway station are less than the set threshold, it means that neither the subway station nor the gate group is frequently passed by the user, and therefore, more abundant data is needed to construct features, that is, the distribution of the outbound times of the user at each gate of the gate group of the subway station and the user portrait as described above, and then the prediction is implemented through the pre-constructed model.

[0133] Please also refer to FIG. 4, which shows a method flowchart of the step of determining the outbound port of the user attribute by performing the historical outbound port inclination statistics according to an embodiment.

[0134] The step of determining the outbound port of the user attribute by performing the historical outbound port inclination statistics provided by the embodiments of the present application comprises:

[0135] Step S301, determining the gate to which the user is inclined at the gate group of the subway station according to the distribution of the outbound times of the user at each gate of the gate group of the subway station;

[0136] Step S302, determining the outbound port of the user as the outbound port located on the same side as the gate.

[0137] For a gate group, the outbound times distribution of a user at each gate in the gate group is counted to determine whether the user is inclined to the left half group gate or the right half group gate.

[0138] The inclination of the user at the gate group determined according to the statistics can determine the outbound direction and the outbound port to which the outbound is inclined, and then the outbound port located on the same side is the commonly used outbound port of the user, and the user journey and travel trajectory of the user are corresponded to the outbound port.

[0139] Thus, after determining the outbound port to which the user belongs, the outbound port to which the user belongs is corresponded to the outbound port through the execution of step S130, and the user journey of the user in the subway is also mapped to the outbound port.

[0140] In other words, corresponding to the outbound port of the outbound trajectory, on the one hand, which users pass through the outbound port, and then the positions of these users corresponding to the outbound port are determined, that is, it is determined which users correspond to the positions on the outbound port, and the user portraits of these users are corresponded to the positions on the outbound port, such as media positions, shop positions, etc.

[0141] On the other hand, corresponding to the outbound port of the outbound trajectory, the trajectory of the tracked outbound trajectory in the station is provided, the limited GPS positioning and the required position permission are jumped out, and the user's in-station trajectory can also be obtained, thereby enhancing the in-station tracking performance.

[0142] With the outbound trajectory of the user and the user journey corresponding to the outbound port, the users are aggregated for the positions on the outbound port through the execution of step S140, so as to assign the aggregated users to the positions on the outbound port.

[0143] The user aggregation refers to determining the user group corresponding to the position on the outbound port, and describing and representing the aggregated user group through the user portrait of the aggregated user.

[0144] Corresponding the outbound trajectory to the outbound port to which the user belongs, the corresponding user journey can correspond to the outbound port, and the user of the user journey also corresponds to the outbound port, so that the users corresponding to the same outbound port are aggregated.

[0145] The position mapped by the outbound port refers to the position on the outbound port. For example, in terms of position type, the position includes but is not limited to media positions, shop positions, etc.; from the perspective of geographical position, the position mapped by the outbound port is any position in the direction of the outbound port, such as an in-station position in the direction of the outbound port, an outbound port passage in the direction of the outbound port, and an out-station position out of the station by the outbound port.

[0146] In data, the user exists in the form of user portrait and outbound trajectory, and the user aggregation is realized through the outbound trajectory and user portrait, so as to assign the user portraits of the aggregated users to the positions mapped by the outbound port, so as to accurately locate and describe the user characteristics of the position through the assigned user portraits, for example, assigning the user portraits of the aggregated users to the media positions, thereby constructing the corresponding media portraits.

[0147] Based on this, in one embodiment of the present application, after executing step S140, the user outbound trajectory attribution method provided by the present application further comprises:

[0148] For the position corresponding to the outbound port, the user portrait of the aggregated user is assigned to the position for the delivery control on the position, which includes media positions and store positions.

[0149] The delivery control on a position refers to the control of the content delivered on the position. The content referred to includes media content and goods, etc. Based on the user portrait of the aggregated user, the content delivered on the position and the delivery frequency, etc. are controlled.

[0150] In another exemplary embodiment, for the user outbound travel trajectory attribution method provided in the present application, a sliding time window is set, for example, the set sliding time window is 30 days. The sliding time window is used to control the outbound times of the user at each gate of the gate group of the subway station, and even the acquisition of the user portrait, so as to ensure the effectiveness of the obtained data, and thus enhance the accuracy of the position corresponding to the user's journey.

[0151] Correspondingly, the journey data of the user and the gate information are also updated with the addition of the user's journey and the change of the gate position. The outbound times of the user at each gate of the gate group of the subway station are obtained under the control of the sliding time window.

[0152] In an exemplary embodiment, a distance factor is also set, which is used to measure the distance between the elevator facility of the platform to the station hall and the gate position.

[0153] That is, for the gates at which the user has exited at the gate group of the subway station, it is determined whether the distance between the elevator facility and the gate position exceeds the distance factor. If it exceeds the distance factor, the determination of the outbound port to which the user belongs is realized through the execution process.

[0154] If the distance between the elevator facility and the gate position does not exceed the set distance factor, it is determined that the user chooses the gate closest to the elevator facility to exit, and the outbound port co-located with the gate is the outbound port to which the user belongs.

[0155] In this way, the outbound behavior of the user is screened, so as to shield the influence of the distance between the elevator facility and the gate.

[0156] The attribution process of the user outbound travel trajectory as described above will be described below in conjunction with a specific example.

[0157] Please also refer to FIG. 5, which is a subway station scene diagram according to an embodiment. In the subway station scene shown in FIG. 5, four gate groups divide the gate area of the station hall, and the users from the boarding platform to the station hall will exit through a gate group and pass through an outbound port.

[0158] For example, the gate group with serial numbers 1-6 corresponds to the A exit and B exit, i.e. for the outbound user, the left half group corresponds to the A exit and the right half group corresponds to the B exit.

[0159] Taking the outbound of a user at the gate group with serial numbers 1-6 as an example, please refer to FIG. 6 which is a data flow diagram according to an embodiment. For the gate group composed of the gates with serial numbers 1-6 shown in FIG. 5, the user gate group outbound data is obtained based on the user itinerary table and the gate information table, i.e. the user gate group card swiping table under the user card swiping outbound. The user gate group card swiping table indicates the number of times of card swiping of each gate of each gate group of the user in the subway station in the past n days, i.e. the number of times shown on the gates with serial numbers 1-6 in FIG. 5.

[0160] The number of times of gate card swiping is the number of times of outbound of the user at each gate of the gate group of the subway station.

[0161] Through the gate group with serial numbers 1-6, the outbound exit of the user is locked to the A exit and B exit, and the C exit and D exit are excluded.

[0162] According to the number of times of gate card swiping, the number of times of outbound of the user at the gate group of the subway station is determined to be 29 which is greater than the set threshold value 5, so the history outbound exit bias statistics is performed according to the number of times of gate card swiping of each gate, and then the outbound exit to which the user belongs is determined. Specifically, the number of times of gate card swiping of the gates with serial numbers 1-3 is 10, 8 and 7 respectively, and the number of times of gate card swiping of the gates with serial numbers 4-6 is 3, 0 and 1 respectively, so the user is biased to the half group of gates with serial numbers 1-3, and therefore the A exit to which the half group of gates with serial numbers 1-3 is biased is the commonly used outbound exit of the user at the subway station.

[0163] If the number of times of outbound of the user at the gate group of the subway station is less than the set threshold value, the data flow shown in FIG. 6 is constructed to implement prediction by the binary classification model. The binary classification model is trained by the constructed training data set, i.e. the binary classification model training record table.

[0164] Through the exemplary embodiments described above, the limitations and shortcomings of determining the attributed outbound exit based on the user location information obtained by app burying are avoided; specifically, the location permission is no longer needed and is limited to the GPS signal, the exposure and implementation of the front-end business are not affected, and the high population coverage is obtained.

[0165] Those skilled in the art can clearly understand, through the description of the above embodiments, that the example embodiments described herein can be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solutions according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a U disk, a mobile hard disk, etc.) or a network, and includes a number of instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to perform the method according to the embodiments of the present application.

[0166] In the example embodiments of the present application, a computer program medium is also provided, which stores computer readable instructions, when the computer readable instructions are executed by a processor of a computer, the computer executes the method described in the above method embodiment part.

[0167] According to one embodiment of the present application, a program product for implementing the method in the above method embodiment is also provided, which can adopt a portable compact disc read-only memory (CD-ROM) and include program codes, and can run on a terminal device, such as a personal computer. However, the program product of the present application is not limited to this, and in this document, the readable storage medium can be any tangible medium containing or storing a program, which can be used or combined with an instruction execution system, device or apparatus.

[0168] The program product can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium may, for example, be but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0169] The computer readable signal medium can include a data signal propagated in a baseband or as a part of a carrier wave, in which readable program codes are borne. Such a propagated data signal can take various forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The readable signal medium can also be any readable medium other than the readable storage medium, which can send, propagate or transmit a program for use by or in connection with an instruction execution system, device or apparatus.

[0170] The program code embodied on the computer readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0171] The program code, when executed by the processor, can cause the processing circuit to perform any of the features and / or operations described herein. The program code can be stored on a non-transitory computer-readable storage medium as an algorithm executable by the processor. In one example, components of the system can be configured to perform a method provided herein. Alternatively, or additionally, components can be configured to receive a computer- program product storing instructions that, when executed by the processor, cause the processor to perform a method provided herein.

[0172] It should be noted that although several modules or units of devices for performing actions are mentioned in the above detailed description, such a division is not mandatory. Indeed, according to an embodiment of the application, features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, features and functions of one module or unit described above can be further divided into several modules or units.

[0173] Furthermore, although individual steps of the methods in the present application are described in a particular order in the figures, this is not required or implied as to the order of the steps, nor is it required that all of the steps be performed to achieve the desired result. Additionally or alternatively, certain steps can be omitted, multiple steps can be combined into one step, one step can be broken into multiple steps, etc.

[0174] From the above description of embodiments, those skilled in the art will readily perceive that the example embodiments described herein can be practiced by various other methods than those expressly described above. It should also be noted that the example embodiments described above can be practiced using software configured to perform the methods according to the embodiments of the present application, and / or using hardware configured to perform the methods according to the embodiments of the present application. Accordingly, the technical solutions according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or a network, and includes a plurality of instructions to cause a computing device (such as a personal computer, a server, a mobile terminal, or a network device, etc.) to perform the methods according to the embodiments of the present application.

[0175] Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the application being indicated by the following claims.

Claims

1. A method for attributing a user's outbound travel trajectory, characterized in that, The method comprises: For the user who generates a user trip by taking the subway, the number of times of the user's exit from each gate of a gate group in a subway station is obtained, and the subway station is the destination station in the user trip; According to the distribution of the number of times of the user's exit from each gate of the gate group in the subway station and the user portrait, the most possible direction of the user after exiting from the gate group is predicted, and the exit gate mapped by the direction is taken as the exit gate to which the user belongs; According to the exit gate to which the user belongs, the travel trajectory of the user is correspondingly mapped to the exit gate; Through the travel trajectory and the user portrait, the users passing through the exit gate are aggregated, and the users aggregated are given a position mapped by the exit gate, and the position is a position requested by a terminal application for realizing its own function for a user; The prediction is realized by a pre-trained classification model, and the training data set of the classification model is constructed for the gate group corresponding to the unique exit gate.

2. The method of claim 1, wherein, The number of times of the user's exit from each gate of the gate group in the subway station is obtained, comprising: Obtaining the travel data of the user, and the travel data at least describes the exit station and the exit gate involved in the generated user trip in content; According to the travel data of the user, the number of times of the user's exit from each gate of the gate group in the subway station is obtained from the gate information according to the exit station and the exit gate of the user.

3. The method of claim 2, wherein, According to the distribution of the number of times of the user's exit from each gate of the gate group in the subway station and the user portrait, the most possible direction of the user after exiting from the gate group is predicted, and the exit gate mapped by the direction is taken as the exit gate to which the user belongs, comprising: According to the number of times of the user's exit from each gate of the gate group in the subway station and the user portrait, the probability of the user's exit direction is obtained, and the exit direction corresponds to at least one exit gate; According to the probability of the user's exit direction, the exit gate to which the user belongs is determined.

4. The method of claim 1, wherein, Before the exit gate to which the user belongs is predicted according to the distribution of the number of times of the user's exit from each gate of the gate group in the subway station and the user portrait, the method further comprises: If the number of times of the user's exit from the gate group in the subway station is less than a set threshold, the prediction of the exit direction to which the user belongs is executed.

5. The method of claim 4, wherein, The method further comprises: If the number of times of the user's exit from the gate group in the subway station is not less than a set threshold, the exit gate to which the user belongs is determined by executing historical exit gate bias statistics.

6. The method of claim 5, wherein, The exit gate to which the user belongs is determined by executing historical exit gate bias statistics, comprising: According to the distribution statistics of the number of times of the user's exit from each gate of the gate group in the subway station, the gate to which the user is biased in the gate group is determined; According to the position of the gate, the exit gate located on the same side is determined as the exit gate to which the user belongs.

7. The method of claim 1, wherein, After the users passing through the exit gate are aggregated through the travel trajectory and the user portrait, the method further comprises: For a location to which the outbound port corresponds, a user profile of the aggregated user is assigned to the location for delivery control on the location, the location including a media location and a merchant location.

8. An attribution system for user outbound travel trajectories, the system comprising a memory, a processor, and a computer program stored on the memory, wherein, The processor executes the computer program to implement the steps of the method of any one of claims 1 to 7.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 7.

10. A computer program product comprising a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 7. The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 7.

Images