WO2018059122A1 - 服务推荐方法、终端、服务器和存储介质 - Google Patents
服务推荐方法、终端、服务器和存储介质 Download PDFInfo
- Publication number
- WO2018059122A1 WO2018059122A1 PCT/CN2017/095946 CN2017095946W WO2018059122A1 WO 2018059122 A1 WO2018059122 A1 WO 2018059122A1 CN 2017095946 W CN2017095946 W CN 2017095946W WO 2018059122 A1 WO2018059122 A1 WO 2018059122A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- user
- cluster
- information
- service
- service recommendation
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Definitions
- the present application relates to the field of information processing technologies, and in particular, to a service recommendation method, a terminal, a server, and a storage medium.
- recommending data to users mainly adopts data recommendation methods based on user preferences.
- Data recommendations to extend the user's source of information For example, through the user's past product browsing record, product collection record or product purchase record, the user's preference can be analyzed, so that the product can be recommended to the user according to the preference.
- the recommended data needs to collect a large number of historical behavior samples of the user, so that the user's preferences can be analyzed comprehensively, and the historical behavior samples of the collected users are very limited, which leads to one-sided recommendation data, and it is difficult to ensure recommendation data. Richness.
- a service recommendation method a terminal, a server, and a storage medium are provided.
- a service recommendation method is implemented on a server, the server includes a memory and a processor, and the method includes:
- the user cluster clusters the corresponding user identifier according to each user recommendation manner in the personal service recommendation method set of the multi-user source;
- the cluster general service recommendation manner is generated according to the personal service recommendation manner corresponding to the user cluster in the personal service recommendation method set;
- the service information corresponding to the met trigger condition in the cluster general service recommendation mode is recommended.
- a service recommendation method is implemented in a terminal, where the terminal includes a memory and a processor, and the method includes:
- the server queries the user cluster to which the reported user identifier belongs, and obtains the cluster general service recommendation mode corresponding to the user cluster, where the user scenario information meets the cluster common
- the service information corresponding to the met trigger condition in the cluster general service recommendation mode is recommended;
- the user cluster according to the individual service recommendation manner in the personal service recommendation mode set of the multi-user source Corresponding user identifiers are obtained by clustering;
- the cluster general service recommendation manner is generated according to the personal service recommendation manner corresponding to the user cluster in the set of personal service recommendation manners;
- a server comprising a memory and a processor, the memory storing computer readable instructions, the computer readable instructions being executed by the processor such that the processor performs the following steps:
- the user cluster clusters the corresponding user identifier according to each user recommendation manner in the personal service recommendation method set of the multi-user source;
- the cluster general service recommendation manner is generated according to the personal service recommendation manner corresponding to the user cluster in the personal service recommendation method set;
- the service information corresponding to the met trigger condition in the cluster general service recommendation mode is recommended.
- a terminal comprising a memory and a processor, the memory storing computer readable instructions, the computer readable instructions being executed by the processor such that the processor performs the following steps:
- the server queries the user cluster to which the reported user identifier belongs, and obtains the cluster general service recommendation mode corresponding to the user cluster, where the user scenario information meets the cluster common
- the service information corresponding to the met trigger condition in the cluster general service recommendation mode is recommended;
- the user cluster according to the individual service recommendation manner in the personal service recommendation mode set of the multi-user source Corresponding user identifiers are obtained by clustering;
- the cluster general service recommendation manner is generated according to the personal service recommendation manner corresponding to the user cluster in the set of personal service recommendation manners;
- a non-transitory computer readable storage medium storing computer readable instructions that, when executed by a processor, cause the processor to perform the steps of the service recommendation method.
- FIG. 1 is an application environment diagram of a service recommendation system in an embodiment
- FIG. 2 is a schematic diagram showing the internal structure of a terminal in an embodiment
- FIG. 3 is a schematic diagram showing the internal structure of a server in an embodiment
- FIG. 4 is a schematic flow chart of a service recommendation method in an embodiment
- FIG. 5 is a schematic diagram of a process of clustering a user cluster, determining a cluster general service recommendation manner, and performing service recommendation in an embodiment
- FIG. 6 is a schematic flow chart of a service recommendation method in another embodiment
- FIG. 7 is a flow chart showing the steps of forming a personal service recommendation method in an embodiment
- FIG. 8 is a flow chart showing the steps of forming a personal service recommendation method in another embodiment
- FIG. 9 is a schematic flowchart of a step of forming a trigger condition according to user behavior scene information and forming service information according to user behavior information in an embodiment
- FIG. 10 is a schematic flowchart of a step of counting the number of user scene behavior records corresponding to the acquired user identifier and including user behavior scene information of the same type in a statistical time period in an embodiment
- FIG. 11 is a schematic flow chart of a service recommendation method in another embodiment
- FIG. 12 is a schematic diagram of a terminal display notification interface in an embodiment
- FIG. 13 is a schematic diagram of a terminal display notification interface in another embodiment
- FIG. 14 is a schematic diagram of a terminal display notification interface in still another embodiment
- 15 is a schematic flowchart of a step of triggering user feedback and transmitting to a server, so that the server adjusts weight according to user feedback;
- Figure 16 is a block diagram showing the structure of a server in an embodiment
- 17 is a structural block diagram of a server in another embodiment
- Figure 19 is a block diagram showing the structure of a terminal in another embodiment.
- FIG. 1 is a diagram of an application environment of a service recommendation system in an embodiment.
- the service recommendation system includes a terminal 110 and a server 120.
- the terminal 110 and the server 120 are connected through a network.
- the terminal 110 can be configured to report the user scene behavior record and the corresponding user identifier to the server 120.
- the server 120 is configured to obtain a user scene behavior record and a corresponding user identifier;
- the user scene behavior record includes user behavior information and corresponding user behavior scene information;
- the server 120 is configured to form a trigger condition according to the user behavior scene information, and form a service according to the user behavior information.
- the server 120 is configured to associate the formed trigger condition with the formed service information to form a personal service recommender corresponding to the acquired user identifier. formula.
- the server 120 may specifically count the number of user scene behavior records corresponding to the acquired user identifier and including the user behavior scene information of the same type in the statistical time period; when the number of statistics in the statistical time period is higher than a preset threshold, according to the same type
- the user behavior scenario information forms a trigger condition, and the service information is formed according to the user behavior information corresponding to the user behavior scenario information of the same type.
- the user behavior scene information includes user behavior time and/or user behavior geographic location.
- the server 120 may be configured to generate a fuzzy range of the user behavior scene information included in each user scene behavior record corresponding to the acquired user identifier, and cluster the user behavior scene information in each user scene behavior record according to the corresponding fuzzy range, and determine to include the same type.
- the user scene behavior record of the user behavior scenario information; the number of user scene behavior records including the same user behavior scenario information is counted in the statistical time period.
- the terminal 110 may report the user identifier and the corresponding user scene information to the server 120.
- the server 120 may be configured to obtain the user identifier and the corresponding user scenario information; query the user cluster to which the obtained user identifier belongs; and the user cluster aggregates the corresponding user identifier according to each user service recommendation manner in the personal service recommendation mode set of the multi-user source.
- the server 120 can be configured to obtain a cluster general service recommendation mode corresponding to the user cluster; the cluster general service recommendation mode is generated according to the personal service recommendation manner corresponding to the user cluster in the personal service recommendation mode set; the server 120 can be configured to: when the user scenario information is satisfied.
- the terminal 110 recommends the service information corresponding to the satisfied trigger condition in the cluster general service recommendation mode.
- the server 120 may specifically recommend the service information corresponding to the met trigger condition in the personal service recommendation mode when the user scenario information satisfies the trigger condition in the personal service recommendation manner corresponding to the acquired user identifier; When the scenario information does not meet the trigger condition in the personal service recommendation mode corresponding to the obtained user identifier, the user cluster to which the obtained user identifier belongs is queried.
- the server 120 may select a corresponding cluster general service recommendation manner according to the obtained weights corresponding to each cluster general service recommendation manner; the weight and the corresponding cluster general service recommendation manner are shared in the user cluster as the personal service recommendation manner.
- the number of users or the number of users is related; the service information corresponding to the triggered trigger condition in the recommended general service recommendation mode of the cluster is recommended.
- terminal 110 may report user feedback to the server 120 for the recommended service information.
- the server 120 may obtain user feedback for the recommended service information;
- the weight corresponding to the cluster general service recommendation method in which the recommended service information is adjusted is adjusted.
- FIG. 2 is a schematic diagram showing the internal structure of a terminal in an embodiment.
- the terminal includes a processor connected through a system bus, a non-volatile storage medium, an internal memory, a network interface, a display screen, and an input device.
- the non-volatile storage medium of the terminal stores an operating system, and can also store computer readable instructions.
- the processor can implement a service recommendation method suitable for the terminal.
- the processor of the terminal is used to provide computing and control capabilities to support the operation of the entire terminal.
- Computer readable instructions may be stored in the internal memory in the terminal, the computer readable instructions being executable by the processor to cause the processor to perform a service recommendation method.
- the network interface of the terminal is used for network communication with the server, such as reporting the user identifier and corresponding user scene information, receiving the recommended service information, and the like.
- the display screen of the terminal may be a liquid crystal display or an electronic ink display screen
- the input device of the terminal may be a touch layer covered on the display screen, or a button, a trackball or a touchpad provided on the terminal housing, or may be an external connection. Keyboard, trackpad or mouse.
- the terminal can be a personal computer, a mobile terminal or a wearable device, such as a mobile phone, a tablet or a personal digital assistant. It will be understood by those skilled in the art that the structure shown in FIG.
- FIG. 2 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the terminal to which the solution of the present application is applied.
- the specific terminal may include a ratio. More or fewer components are shown in Figure 2, or some components are combined, or have different component arrangements.
- FIG. 3 is a schematic diagram showing the internal structure of a server in an embodiment.
- the server includes a processor, a non-volatile storage medium, an internal memory, and a network interface connected by a system bus.
- the non-volatile storage medium of the server stores an operating system, and can also store computer readable instructions.
- the processor can implement a service recommendation method suitable for the server.
- the server's processor is used to provide computing and control capabilities that support the operation of the entire server.
- the computer's internal memory can store computer readable instructions that, when executed by the processor, cause the processor to perform a service recommendation method.
- the server's network interface is used to communicate with external terminals over a network connection.
- the server can be implemented with a stand-alone server or a server cluster consisting of multiple servers.
- a stand-alone server or a server cluster consisting of multiple servers.
- the specific server may include a ratio. More or more There are few parts, or some parts are combined, or have different part arrangements.
- the service recommendation method can be applied to the terminal 110 or the server 120. This embodiment is mainly applied to the server in FIG. 3 by using the method. Referring to FIG. 4, the service recommendation method in this embodiment specifically includes the following steps:
- the user identifier is used to uniquely identify the corresponding user.
- User information corresponding to the user scene is information indicating the scene in which the user is located.
- the user scene information may be user instant scene information.
- the user instant scene information is information indicating the immediate state of the scene in which the user is located. Instant refers to the current time or approximate current time.
- the user instant scene information may include a combination of one or more of user terminal instant time, user instant geographic location, or user immediate operation information.
- the user terminal instant time refers to the instant time of the terminal logged in with the user ID.
- the real time of the user terminal can take different forms as needed, such as a form indicated by a date and a time point of the day, or a form represented only by the time point of the day.
- the user terminal's instant time can be accurate to hours or minutes or seconds or milliseconds.
- the user's instant location is used to indicate the user's immediate location.
- the user's real-time location can be used in the location of the terminal, such as the longitude and latitude of the location of the terminal.
- the user's real-time geographic location can also adopt the information point (POI, full name of Point of Information) where the user or the terminal is located.
- POI full name of Point of Information
- the user's real-time operation information is information that records the user's immediate operation on the terminal, such as the application identifier of the application currently being used on the terminal, the function identifier of the function being used in the application that the terminal is currently being used, and the data generated by the current operation.
- the application identifier of the application currently being used on the terminal such as the application identifier of the application currently being used on the terminal, the function identifier of the function being used in the application that the terminal is currently being used, and the data generated by the current operation.
- the application identifier of the application currently being used on the terminal such as the application identifier of the application currently being used on the terminal, the function identifier of the function being used in the application that the terminal is currently being used, and the data generated by the current operation.
- S404 Query the user cluster to which the obtained user identifier belongs; the user cluster obtains the corresponding user identifier according to each user service recommendation manner in the personal service recommendation method set of the multi-user source.
- the service recommendation method refers to the data on which the service recommendation is based, and may also be referred to as a service recommendation policy.
- the personal service recommendation method is a personal service recommendation method.
- the personal service recommendation method corresponds to the user identifier, and indicates that the user recommendation is performed by using the personal service recommendation method corresponding to the user identifier.
- the personal service recommendation method set is a multi-user source, indicating that the set includes more than one user's personal service recommendation method.
- Personal service recommendation methods include trigger conditions and corresponding
- the service information indicates that the service information corresponding to the trigger condition is recommended to the terminal corresponding to the corresponding user identifier when the trigger condition is met.
- a user cluster is a collection of user IDs obtained by clustering. Clustering is the process of dividing objects into different subsets so that similar objects belong to the same subset.
- each person service recommendation manner in the multi-user source personal service recommendation method set corresponds to a user identifier, and the user identifiers are clustered according to the similarity of the personal service recommendation manners to obtain more than one user cluster.
- the personal service recommendation methods corresponding to the user IDs in the same user cluster have similarities.
- the server may specifically cluster the individual service recommendation manners in the personal service recommendation method set into a personal service recommendation mode subset, and then configure the user identifier corresponding to each individual service recommendation mode subset to form a corresponding user cluster.
- clustering may be adopted, such as K-means clustering algorithm, artificial neural network algorithm or Support Vector Machine (SVM).
- K-means clustering algorithm the personal service recommendation method in the personal service recommendation method set can be abstracted into a vector and then clustered.
- clustering can be performed as follows: (1) randomly select K cluster centers in the personal service recommendation method set; (2) traverse the personal service recommendation method set, and select the personal service recommendation method.
- the personal service recommendation method in the collection is divided into the nearest cluster center to get the corresponding cluster; (3) calculate the average value of each cluster and use it as a new cluster center; repeat (2) and (3) until it is satisfied.
- Iterative stop condition is, for example, the maximum number of iterations, or the last calculated cluster center is smaller than the preset value of the cluster center calculated earlier than the preset value.
- the corresponding service information is the take-out application service information; in another personal service recommendation mode, The trigger condition is the working time period and/or the working time period, and the corresponding service information is the online car application service information. If the number of user identifiers corresponding to the two personal service recommendation methods exceeds a predetermined value, the user identifiers corresponding to the two personal service recommendation methods may be clustered into one user cluster, indicating a white-collar user group.
- the cluster general service recommendation mode is generated according to the personal service recommendation manner corresponding to the user cluster in the personal service recommendation mode set.
- the general service recommendation method of the cluster is to push the service identifier of the corresponding user cluster.
- the personal service recommendation method corresponding to the user cluster in the personal service recommendation method set refers to the personal service recommendation mode in the personal service recommendation method set corresponding to the user identifier in the user cluster.
- the cluster general service recommendation method may specifically adopt all personal service recommendation methods corresponding to the user cluster in the personal service recommendation method set.
- the personal service recommendation mode corresponding to the user cluster in the personal service recommendation mode set the personal service recommendation mode in which the number of user identifiers in the corresponding user cluster is greater than the preset value is used as the cluster general service recommendation corresponding to the user cluster. the way.
- the cluster general service recommendation method may also be adjusted based on the personal service recommendation method corresponding to the user cluster in the personal service recommendation method set.
- the cluster general service recommendation method includes a trigger condition and corresponding service information.
- the terminal corresponding to the user identifier is recommended to the cluster general service recommendation mode and the trigger.
- the service information is information that provides a service to the user, and may be information or a link, and the link may be a web link, a connection to open a specified application, or a connection to open a specified function of a specified application.
- the foregoing white-collar user group if a large number of user identifiers (such as more than a predetermined number or a predetermined proportion of user identifiers), have the same or similar personal service recommendation methods, such as a trigger condition of Saturday or near Saturday, and
- the corresponding service information is a personal service recommendation method for the movie ticket purchase application service information.
- the personal service recommendation method may be used as a cluster general service recommendation mode of the white-collar user group.
- the user terminal corresponding to any user identifier in the white-collar user group belongs to Saturday or near Saturday, the corresponding terminal is identified to the user. Recommend movie ticket purchase application service information.
- the service recommendation method may query the user cluster to which the user identifier belongs, and the user cluster shall correspond to the user according to the individual service recommendation manner in the personal service recommendation method set of the multi-user source.
- the identifier is obtained by clustering, and then the user cluster represents a user group similar to the user indicated by the obtained user identifier in the personal service recommendation manner.
- the cluster general service recommendation method corresponding to the user cluster is based on the collection of personal service recommendation methods.
- the personal service recommendation manner corresponding to the user cluster is generated, so that the individual service recommendation manner can be used in the user cluster.
- the service information corresponding to the met trigger condition in the general service recommendation mode of the cluster is recommended to avoid the one-sided recommendation service information and ensure the richness of the recommended service information.
- the service recommendation method further includes: when the user scenario information meets a trigger condition in the personal service recommendation manner corresponding to the acquired user identifier, recommending the trigger in the personal service recommendation manner and the satisfaction The service information corresponding to the condition; when the user scenario information does not satisfy the trigger condition in the personal service recommendation mode corresponding to the acquired user identifier, step S404 is performed.
- the server may first determine whether the user scene information meets the trigger condition in the personal service recommendation manner corresponding to the user identifier. When the trigger condition in the personal service recommendation mode is met, the server may directly identify the service information corresponding to the satisfied trigger condition in the personal service recommendation mode of the corresponding terminal recommendation. The server may further determine whether the user scenario information meets the triggering condition in the cluster general service recommendation mode corresponding to the user cluster to which the user identifier belongs, when the triggering condition in the user service recommendation mode is not met.
- the server may identify the service information corresponding to the met trigger condition in the recommended general service recommendation mode of the terminal.
- the server may record the user scenario information as the user behavior scenario information. After recommending the service information to the terminal, the server may receive user feedback of the user terminal for the service information.
- the service recommendation is preferentially performed according to the personal service recommendation method, and the accuracy of the service recommendation may be preferentially ensured; if the service recommendation is not performed according to the cluster general service recommendation method when the service recommendation cannot be performed according to the service recommendation mode, the user may be recommended. Service information for potential needs to meet user needs as much as possible.
- the service recommendation method further includes the step of forming a personal service recommendation method.
- the step of forming a personal service recommendation manner specifically includes the following steps:
- the user scene behavior record includes user behavior information and corresponding user behavior scene information.
- the user scene behavior record is data that records the behavior of the user in a specific scenario.
- User behavior can be the behavior of the user operating the terminal.
- the user scene behavior record may be generated by the user scene information and user feedback reported by the user terminal, or may be directly reported to the server by the user terminal.
- the user scene behavior record may include user behavior scene information and corresponding user behavior information, and the user behavior scene information includes user behavior time and/or user behavior geographic location.
- the user behavior time can indicate the time when the user behavior occurs, and the user behavior geographic location can indicate the geographic location where the user behavior occurs.
- the user scene behavior record can include a user behavior track.
- the user behavior track can represent the order in which multiple user actions occur.
- a user scene behavior record can be recorded as: 2016-8-19, 19:32, Shenzhen Science and Technology Park XX cafe, XX cafe Member Application - Membership Card.
- the user scene behavior record can indicate that the user performed the user behavior time at 19:32 on August 19, 2016, and implemented the user in the XX cafe Member Application at the location of the user behavior of the Shenzhen Science and Technology Park XX cafe. Card features this user behavior.
- a user scene behavior record can be recorded as: Service A, Service B.
- the user scene behavior record indicates that the user has used service B after using service A.
- the user scene behavior record represents a user behavior track.
- the server may directly form the trigger condition for the user behavior scenario information, such as the user behavior time and/or the user behavior geographic location to form a trigger condition.
- the user behavior that occurs first in the user behavior track can be regarded as the user behavior scene information, and the user behavior that occurs later can be regarded as the user behavior information.
- the user behavior record records the user behavior track
- the user behavior occurring in the user behavior track may be triggered, and the user behavior occurring in the user behavior track forms service information corresponding to the trigger condition.
- S706 Correspond to the formed service information, and form a personal service recommendation manner corresponding to the acquired user identifier.
- the personal service recommendation mode is generated by using the user scene behavior record, and the service recommendation can be accurately performed based on the personal service recommendation mode, and the service recommendation accuracy is improved.
- step S704 specifically includes the following steps:
- Count in the statistical time period, the number of user scene behavior records corresponding to the acquired user identifier and including the same type of user behavior scene information.
- the statistical time period may be a preset number of days, specifically one week or one month, and an appropriate statistical time period may be selected according to the accuracy requirement of the service recommendation.
- the same kind of user behavior scene information is the same or similar user behavior scene information, for example, the user behavior time difference is within a preset range, or the user behavior geographical position is within a preset range.
- the same kind of user behavior scene information may also include the same user behavior track segment, and the user behavior track segment is a part of the user behavior track intercepted.
- the three user scene behavior records can be recorded as the same user scene behavior record statistics.
- the preset threshold can be set as needed.
- the server can form a trigger condition that can simultaneously satisfy the same kind of user behavior scene information.
- the server may form service information capable of covering user behavior information corresponding to user behavior scenario information of the same type.
- the personal service recommendation manner is formed according to a large number of user scene behavior records in the user scene behavior record including the user behavior scene information of the same type, so that the personal service recommendation manner can accurately reflect the user's service recommendation requirement. Regardless of the service recommendation based on the personal service recommendation method or the service recommendation based on the cluster general service recommendation method, certain accuracy can be guaranteed.
- the user behavior scenario information includes a user behavior time and/or a user behavior geographic location; referring to FIG. 10, step S902 includes the following steps:
- S1002 Generate a fuzzy range of user behavior scene information included in each user scene behavior record corresponding to the acquired user identifier.
- the server may obtain all user scene behavior records corresponding to the user identifier, and generate The user behavior time and/or the fuzzy range of the user behavior geographic location included in each user scene behavior record.
- the fuzzy range refers to a range obtained by expanding the range based on the user behavior scene information.
- the fuzzy range of the user's behavior time such as the scope extended by the user's behavior time, such as the user behavior time is 9:00, and the expansion range is 9:00 from 9:00 to the center. 10:00.
- the fuzzy range of the geographic location may be a range formed by expanding the range centered on the user's behavioral geographic location.
- S1004 The user behavior scene information in each user scene behavior record is clustered according to a corresponding fuzzy range, and the user scene behavior record including the same user behavior scene information is determined.
- the user behavior scene information is clustered according to the fuzzy range, and the user behavior scene information having the intersection or the intersection area larger than the preset area may be clustered into the same type, thereby determining the user including the same type of user behavior scene information.
- Scene behavior record is
- S1006 Counts the number of user scene behavior records including the same user behavior scenario information in the statistical time period.
- the user behavior record including the same user behavior scene information is determined by using the user behavior time and/or the fuzzy range of the user behavior geographic location, so that the user scene behavior record can be generated more effectively, and the user behavior scene information is avoided. Specifically, the number of user scene behavior records is too small.
- the step of recommending the service information corresponding to the met triggering condition in the cluster general service recommendation mode in step S408 specifically includes: selecting a corresponding cluster general service recommendation manner according to the weights corresponding to the acquired general service recommendation manners of each cluster.
- the weight and the corresponding cluster general service recommendation method are related to the number of users or the number of users shared in the user cluster as the personal service recommendation method; the service information corresponding to the satisfied trigger condition in the recommended general service recommendation mode of the cluster is recommended.
- the user cluster may correspond to a plurality of cluster general service recommendation manners, and each cluster general service recommendation manner has a weight, and when the service recommendation is performed, a cluster general service recommendation manner may be selected from multiple cluster general service recommendation manners according to the weight.
- the server may determine the probability of each general service recommendation mode according to the weight, thereby selecting the cluster general service recommendation mode according to the probability.
- the server can also directly select the cluster general service recommendation method with the largest weight to perform service recommendation. When the weights are the same, the server can randomly select from the cluster general service recommendation methods with the same weight.
- the weight of the cluster general service recommendation method corresponding to the weight may be positively related to the number of users or the number of users shared in the user cluster as the personal service recommendation method, and may of course be negative correlation.
- the white-collar user group includes User A, User B, User C, and User D
- the cluster general service recommendation methods corresponding to the white-collar user group are E and F.
- the general service recommendation mode E of the cluster is the personal service recommendation mode common to the user A, the user B, and the user C
- the cluster general service recommendation mode F is only the personal service recommendation mode of the user A and the user B
- the weight of the E is higher than F.
- the weight of the server will be prioritized based on the cluster general service recommendation method E with high weight.
- the service recommendation is selected according to the weight of the cluster general service recommendation mode, and the appropriate cluster general service recommendation mode is selected for service recommendation when there are multiple cluster general service recommendation modes, and the service recommendation can satisfy the service requirement of the user as much as possible. .
- the step of recommending the service information corresponding to the met trigger condition in the personal service recommendation manner includes: selecting a corresponding personal general service recommendation manner according to the weight corresponding to each individual general service recommendation manner; and recommending the selected personal universal service The service information corresponding to the satisfied trigger condition in the recommended mode.
- the weight may be related to the number of user scene behavior records that correspond to the acquired user identification and including the same type of user behavior scene information within the statistical time period.
- the service recommendation method further includes: obtaining user feedback for the recommended service information; and adjusting a weight corresponding to the cluster general service recommendation manner in which the recommended service information is located according to the user feedback.
- the user feedback may indicate acceptance of the service information or rejection of the service information.
- the server may increase the weight corresponding to the cluster general service recommendation mode in which the recommended service information is located.
- the server may lower the weight corresponding to the cluster general service recommendation mode in which the recommended service information is located.
- the weight corresponding to the cluster general service recommendation mode can be adjusted through user feedback, so that the service recommendation policy can be dynamically adjusted to meet the user's changing service requirements.
- FIG. 11 is a schematic flow chart of a service recommendation method in another embodiment. This embodiment is mainly illustrated by applying the method to the terminal in FIG. 2 described above. Referring to FIG. 11, the method specifically includes the following steps:
- S1102 Reporting the user identifier and the corresponding user scenario information to the server, so that the server queries the user cluster to which the reported user identifier belongs, and obtains the cluster general service recommendation manner corresponding to the user cluster.
- the service information corresponding to the met trigger condition in the cluster general service recommendation mode is recommended; the user cluster is served according to the individual service recommendation mode set of the multi-user source.
- the recommended method is to cluster the corresponding user identifiers; the cluster general service recommendation method is generated according to the personal service recommendation manner corresponding to the user cluster in the personal service recommendation method set.
- S1104 Receive service information recommended by the server.
- S1106 Display a notification interface, and display a corresponding service entry in the notification interface according to the service information.
- the notification interface may be generated by a notification channel provided by the terminal operating system, such as a drop-down notification bar or a top notification bar.
- the service information can include a link that the terminal can display as a service portal.
- the service entry is used to trigger a call to the corresponding service.
- the terminal may detect a user operation instruction for the service portal, and trigger a service pointed to by the service entry when detecting the user operation instruction, such as opening an application providing the service, or opening a specified function in the service, or linking to a specified webpage or the like.
- the service recommendation method may query the user cluster to which the user identifier belongs, and the user cluster shall correspond to the user according to the individual service recommendation manner in the personal service recommendation method set of the multi-user source.
- the identifier is obtained by clustering, and then the user cluster represents a user group similar to the user indicated by the obtained user identifier in the personal service recommendation manner.
- the cluster general service recommendation method corresponding to the user cluster is generated according to the personal service recommendation manner corresponding to the user cluster in the personal service recommendation method set, so that the individual service recommendation manner can be common in the user cluster.
- the service information corresponding to the met trigger condition in the general service recommendation mode of the cluster is recommended to avoid the one-sided recommendation service information and ensure the richness of the recommended service information.
- the step S1102 includes: reporting the user identifier and the corresponding user scene information to the server, so that the server recommends the personal service recommendation when the user scene information meets the trigger condition in the personal service recommendation manner corresponding to the reported user identifier.
- the service information corresponding to the triggered triggering condition in the mode when the user scenario information does not meet the triggering condition in the personal service recommendation manner corresponding to the reported user identifier, querying the user cluster to which the reported user identifier belongs, and acquiring the user cluster corresponding to the user cluster
- the cluster general service recommendation mode when the user scenario information meets the trigger condition in the cluster general service recommendation mode, the service information corresponding to the met trigger condition in the cluster general service recommendation mode is recommended.
- the terminal when the terminal uses the image editing application, the terminal reports the image editing application identifier and the user identifier to the server, and the server has a personal service recommendation manner indicating the user behavior track corresponding to the user identifier, and the service is provided.
- the triggering condition of the recommended mode is the image editing application identifier
- the image sharing application identifier corresponding to the triggering condition is obtained from the service recommendation mode to the terminal.
- the terminal displays a notification interface, and a service portal for entering the picture sharing application is displayed in the notification interface.
- the terminal may further receive service information that is sent by the server according to the instant time of the user terminal, and display a notification interface on the desktop of the terminal operating system, where the service portal is displayed.
- the service information may be recommended by the server according to the cluster general service recommendation manner of the personal service recommendation manner that is not corresponding to the user identifier, and after receiving the service information, the terminal may prompt the service information according to the individual not corresponding to the user identifier.
- the recommended general service recommendation method for the service recommendation method is recommended.
- the recommended service information is included in the cluster general service recommendation mode selected according to the weights corresponding to the acquired cluster general service recommendation manners.
- the method further includes the steps of: triggering user feedback and sending to the server, so that the server adjusts the weight according to the user feedback, specifically including the following steps:
- S1502 Display a service cancellation control in the notification interface according to the service information.
- S1506 When detecting a user operation instruction for the service portal or the service cancellation control, send user feedback to the server, so that the server adjusts the weight corresponding to the cluster general service recommendation manner in which the recommended service information is located according to the user feedback.
- the terminal when detecting the user operation instruction for the service portal, the terminal generates user feedback indicating that the service information is accepted and sends the message to the server; when the terminal detects the user operation instruction for the service portal or the service cancellation control, the terminal generates a service denial service.
- the user of the information is fed back to the server, so that the server adjusts the weight corresponding to the cluster general service recommendation method according to the user feedback according to the user feedback.
- the weight corresponding to the cluster general service recommendation mode can be adjusted by user feedback. This allows the service recommendation policy to be dynamically adjusted to meet the changing service needs of users.
- FIG. 16 is a block diagram showing the structure of the server 1600 in one embodiment.
- the internal structure of the server 1600 may correspond to the structure as shown in FIG. 3, and each of the following modules may be implemented in whole or in part by software, hardware, or a combination thereof.
- the server 1600 includes: a data acquisition module 1601, a cluster general service recommendation mode acquisition module 1602, and a service information recommendation module 1603.
- the data obtaining module 1601 is configured to obtain a user identifier and corresponding user scene information.
- the cluster general service recommendation mode obtaining module 1602 is configured to query the user cluster to which the obtained user identifier belongs; obtain the cluster general service recommendation mode corresponding to the user cluster; and the user cluster recommends each person in the personal service recommendation mode set according to the multi-user source.
- the method comprises: clustering the corresponding user identifiers; the cluster general service recommendation manner is generated according to the personal service recommendation manner corresponding to the user cluster in the personal service recommendation method set;
- the service information recommendation module 1603 is configured to recommend service information corresponding to the met trigger condition in the general service recommendation mode of the cluster when the user scenario information meets the trigger condition in the cluster general service recommendation mode.
- the service information recommendation module 1603 is further configured to: when the user scenario information meets a trigger condition in the personal service recommendation manner corresponding to the acquired user identifier, recommend a service corresponding to the met trigger condition in the personal service recommendation manner. information.
- the cluster general service recommendation manner obtaining module 1602 is further configured to query the user cluster to which the obtained user identifier belongs when the user scenario information does not meet the trigger condition in the personal service recommendation manner corresponding to the obtained user identifier.
- FIG. 17 is a block diagram showing the structure of a server 1600 in another embodiment.
- the server 1600 further includes: a personal service recommendation manner generating module 1604, configured to acquire a user scene behavior record and a corresponding user identifier; the user scene behavior record includes user behavior information and corresponding user behavior scene information; The information forming trigger condition is formed according to the user behavior information; the formed trigger condition is corresponding to the formed service information, and the personal service recommendation manner corresponding to the acquired user identifier is formed.
- the personal service recommendation method generation module 1604 is further configured to use the statistical time period. Counting the number of user scene behavior records corresponding to the obtained user identifiers and including the user behavior scene information of the same type; when the number of statistics in the statistical period is higher than the preset threshold, the trigger condition is formed according to the similar user behavior scene information, according to User behavior information corresponding to similar user behavior scenario information forms service information.
- the user behavior scenario information includes a user behavior time and/or a user behavior geographic location; the personal service recommendation manner generating module 1604 is further configured to generate a user behavior scenario included in each user scenario behavior record corresponding to the acquired user identifier.
- the fuzzy range of the information; the user behavior scene information in each user scene behavior record is clustered according to the corresponding fuzzy range, and the user scene behavior record including the same user behavior scene information is determined; the statistics include the same type of user behavior in the statistical time period. The number of user scene behavior records for scene information.
- the service information recommendation module 1603 is further configured to select a corresponding cluster general service recommendation manner according to the weights corresponding to the obtained cluster common service recommendation manners; the weight and the corresponding cluster general service recommendation manner are used as the personal service recommendation manner.
- the number of users or the number of users shared in the user cluster is related; the service information corresponding to the triggered trigger condition in the recommended general service recommendation mode of the cluster is recommended.
- the server 1600 further includes: a weight adjustment module 1605, configured to obtain user feedback for the recommended service information; and adjust a weight corresponding to the cluster general service recommendation manner in which the recommended service information is located according to the user feedback.
- a weight adjustment module 1605 configured to obtain user feedback for the recommended service information; and adjust a weight corresponding to the cluster general service recommendation manner in which the recommended service information is located according to the user feedback.
- the server 1600 can query the user cluster to which the user identifier belongs, and the user cluster associates the corresponding user identifier according to each user recommendation manner in the personal service recommendation method set of the multi-user source. Clustering is performed, so that the user cluster represents a user group similar to the user indicated by the obtained user identifier in the personal service recommendation manner.
- the cluster general service recommendation method corresponding to the user cluster is generated according to the personal service recommendation manner corresponding to the user cluster in the personal service recommendation method set, so that the individual service recommendation manner can be common in the user cluster.
- the service information corresponding to the met trigger condition in the general service recommendation mode of the cluster is recommended to avoid the one-sided recommendation service information and ensure the richness of the recommended service information.
- FIG. 18 is a block diagram showing the structure of a terminal 1800 in an embodiment.
- the internal structure of the terminal 1800 can correspond to In the structure shown in FIG. 2, each of the following modules may be implemented in whole or in part by software, hardware, or a combination thereof.
- the terminal 1800 includes: a reporting module 1801, a service information receiving module 1802, and a notification interface displaying module 1803.
- the reporting module 1801 is configured to report the user identifier and the corresponding user scenario information to the server, so that the server queries the user cluster to which the reported user identifier belongs, and obtains the cluster general service recommendation mode corresponding to the user cluster, and the user scenario information satisfies the cluster general service recommendation.
- the service information corresponding to the met trigger condition in the cluster general service recommendation mode is recommended; the user cluster aggregates the corresponding user identifier according to each person service recommendation manner in the personal service recommendation mode set of the multi-user source.
- the class general service recommendation method is generated according to the personal service recommendation method corresponding to the user cluster in the personal service recommendation method set.
- the service information receiving module 1802 is configured to receive service information recommended by the server.
- the notification interface display module 1803 is configured to display a notification interface, and display a corresponding service portal in the notification interface according to the service information.
- the reporting module 1801 is further configured to report the user identifier and the corresponding user scenario information to the server, so that the server recommends the user when the user scenario information meets a trigger condition in the personal service recommendation manner corresponding to the reported user identifier.
- the service information corresponding to the triggered triggering condition in the service recommendation mode when the user scenario information does not meet the triggering condition in the personal service recommendation manner corresponding to the reported user identifier, querying the user cluster to which the reported user identifier belongs, and acquiring the user cluster
- the corresponding general service recommendation mode of the cluster when the user scenario information meets the trigger condition in the general service recommendation mode of the cluster, the service information corresponding to the met trigger condition in the general service recommendation mode of the cluster is recommended.
- the recommended service information is included in the cluster general service recommendation mode selected according to the weights corresponding to the acquired cluster general service recommendation manners.
- the notification interface display module 1803 is further configured to display the service cancellation control in the notification interface according to the service information.
- the terminal 1800 further includes an application invoking module 1804 and a user feedback transmitting module 1805.
- the application invoking module 1804 is configured to enter an application pointed to by the service portal when a user operation instruction for the service portal is detected.
- the user feedback sending module 1805 is configured to send user feedback to the server when detecting a user operation instruction for the service portal or the service cancellation control, so that the server adjusts the cluster general service recommendation manner corresponding to the recommended service information according to the user feedback. Weights.
- the server 1600 can query the user cluster to which the user identifier belongs, and the user cluster associates the corresponding user identifier according to each user recommendation manner in the personal service recommendation method set of the multi-user source. Clustering is performed, so that the user cluster represents a user group similar to the user indicated by the obtained user identifier in the personal service recommendation manner.
- the cluster general service recommendation method corresponding to the user cluster is generated according to the personal service recommendation manner corresponding to the user cluster in the personal service recommendation method set, so that the individual service recommendation manner can be common in the user cluster.
- the service information corresponding to the met trigger condition in the general service recommendation mode of the cluster is recommended to avoid the one-sided recommendation service information and ensure the richness of the recommended service information.
- the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or the like.
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
一种服务推荐方法,包括:获取用户标识和相应的用户场景信息(S402);查询获取的用户标识所属的用户集群;所述用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到(S404);获取所述用户集群对应的集群通用服务推荐方式;所述集群通用服务推荐方式根据所述个人服务推荐方式集合中与所述用户集群对应的个人服务推荐方式生成(S406);当所述用户场景信息满足所述集群通用服务推荐方式中的触发条件时,推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息(S408)。
Description
相关申请的交叉引用
本申请要求于2016年9月29日提交中国专利局,申请号为2016108656520,发明名称为“服务推荐方法和装置”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
本申请涉及信息处理技术领域,特别是涉及一种服务推荐方法、终端、服务器和存储介质。
目前,向用户推荐数据主要采用基于用户喜好的数据推荐方式,此种方式需要采集用户自身大量的历史行为样本,从大量的历史行为样本中分析出用户可能的喜好,从而按照分析出的喜好进行数据推荐,以扩展用户的信息来源渠道。比如,通过用户过往的商品浏览记录、商品收藏记录或者商品购买记录,可以分析出用户的偏好,从而后续可按照该偏好向用户推荐商品。
然而,目前推荐数据需要采集用户自身大量的历史行为样本,才能够较为全面的分析出用户喜好,而通常采集到的用户的历史行为样本是非常有限的,导致推荐数据片面化,难以保证推荐数据的丰富度。
发明内容
根据本申请的各种实施例,提供一种服务推荐方法、终端、服务器和存储介质。
一种服务推荐方法,执行于服务器,所述服务器包括存储器和处理器,所述方法包括:
获取用户标识和相应的用户场景信息;
查询获取的用户标识所属的用户集群;所述用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到;
获取所述用户集群对应的集群通用服务推荐方式;所述集群通用服务推荐方式根据所述个人服务推荐方式集合中与所述用户集群对应的个人服务推荐方式生成;及
当所述用户场景信息满足所述集群通用服务推荐方式中的触发条件时,推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息。
一种服务推荐方法,执行于终端,所述终端包括存储器和处理器,所述方法包括:
向服务器上报用户标识和相应的用户场景信息,使得所述服务器查询上报的用户标识所属的用户集群,获取所述用户集群对应的集群通用服务推荐方式,当所述用户场景信息满足所述集群通用服务推荐方式中的触发条件时,推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息;所述用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到;所述集群通用服务推荐方式根据所述个人服务推荐方式集合中与所述用户集群对应的个人服务推荐方式生成;
接收所述服务器推荐的所述服务信息;
展示通知界面,根据所述服务信息在所述通知界面中展示相应的服务入口。
一种服务器,包括存储器和处理器,所述存储器中储存有计算机可读指令,所述计算机可读指令被所述处理器执行时,使得所述处理器执行以下步骤:
获取用户标识和相应的用户场景信息;
查询获取的用户标识所属的用户集群;所述用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到;
获取所述用户集群对应的集群通用服务推荐方式;所述集群通用服务推荐方式根据所述个人服务推荐方式集合中与所述用户集群对应的个人服务推荐方式生成;及
当所述用户场景信息满足所述集群通用服务推荐方式中的触发条件时,推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息。
一种终端,包括存储器和处理器,所述存储器中储存有计算机可读指令,所述计算机可读指令被所述处理器执行时,使得所述处理器执行以下步骤:
向服务器上报用户标识和相应的用户场景信息,使得所述服务器查询上报的用户标识所属的用户集群,获取所述用户集群对应的集群通用服务推荐方式,当所述用户场景信息满足所述集群通用服务推荐方式中的触发条件时,推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息;所述用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到;所述集群通用服务推荐方式根据所述个人服务推荐方式集合中与所述用户集群对应的个人服务推荐方式生成;
接收所述服务器推荐的所述服务信息;
展示通知界面,根据所述服务信息在所述通知界面中展示相应的服务入口。
一种非易失性的计算机可读存储介质,存储有计算机可读指令,所述计算机可读指令被处理器执行时,使得所述处理器执行所述服务推荐方法的步骤。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征、目的和优点将从说明书、附图以及权利要求书变得明显。
为了更清楚地说明本申请实施例中的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其它的附图。
图1为一个实施例中服务推荐***的应用环境图;
图2为一个实施例中终端的内部结构示意图;
图3为一个实施例中服务器的内部结构示意图;
图4为一个实施例中服务推荐方法的流程示意图;
图5为一个实施例中聚类得到用户集群、确定集群通用服务推荐方式并进行服务推荐的过程的示意图;
图6为另一个实施例中服务推荐方法的流程示意图;
图7为一个实施例中形成个人服务推荐方式的步骤的流程示意图;
图8为另一个实施例中形成个人服务推荐方式的步骤的流程示意图;
图9为一个实施例中根据用户行为场景信息形成触发条件,根据用户行为信息形成服务信息的步骤的流程示意图;
图10为一个实施例中统计时间段内统计对应于获取的用户标识且包括同类的用户行为场景信息的用户场景行为记录的数量的步骤的流程示意图;
图11为另一个实施例中服务推荐方法的流程示意图;
图12为一个实施例中终端展示通知界面的示意图;
图13为另一个实施例中终端展示通知界面的示意图;
图14为再一个实施例中终端展示通知界面的示意图;
图15为一个实施例中触发用户反馈并向服务器发送,使得服务器根据用户反馈调整权重的步骤的流程示意图;
图16为一个实施例中服务器的结构框图;
图17为另一个实施例中服务器的结构框图;
图18为一个实施例中终端的结构框图;
图19为另一个实施例中终端的结构框图。
为了使本申请的目的、技术方案及优点更加清楚明白,以下结合附图及实施例,对本申请进行进一步详细说明。应当理解,此处所描述的具体实施例仅仅用以解释本申请,并不用于限定本申请。
图1为一个实施例中服务推荐***的应用环境图。参照图1,该服务推荐***包括终端110和服务器120。终端110和服务器120通过网络连接。终端110可用于向服务器120上报用户场景行为记录以及相应的用户标识。服务器120用于获取用户场景行为记录以及相应的用户标识;用户场景行为记录包括用户行为信息和相应的用户行为场景信息;服务器120用于根据用户行为场景信息形成触发条件,根据用户行为信息形成服务信息;服务器120用于将形成的触发条件与形成的服务信息对应,形成与获取的用户标识对应的个人服务推荐方
式。服务器120具体可在统计时间段内统计对应于获取的用户标识且包括同类的用户行为场景信息的用户场景行为记录的数量;当统计时间段内统计的数量高于预设阈值时,根据同类的用户行为场景信息形成触发条件,根据与同类的用户行为场景信息对应的用户行为信息形成服务信息。
在一个实施例中,用户行为场景信息包括用户行为时间和/或用户行为地理位置。服务器120可用于生成与获取的用户标识对应的各用户场景行为记录所包括用户行为场景信息的模糊范围;将各用户场景行为记录中的用户行为场景信息按照相应的模糊范围聚类,确定包括同类的用户行为场景信息的用户场景行为记录;在统计时间段内统计包括同类的用户行为场景信息的用户场景行为记录的数量。
在一个实施例中,终端110具体可向服务器120上报用户标识和相应的用户场景信息。服务器120可用于获取用户标识和相应的用户场景信息;查询获取的用户标识所属的用户集群;用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到;服务器120可用于获取用户集群对应的集群通用服务推荐方式;集群通用服务推荐方式根据个人服务推荐方式集合中与用户集群对应的个人服务推荐方式生成;服务器120可用于当用户场景信息满足集群通用服务推荐方式中的触发条件时,向终端110推荐集群通用服务推荐方式中与满足的触发条件对应的服务信息。
在一个实施例中,服务器120具体可当用户场景信息满足与获取的用户标识对应的个人服务推荐方式中的触发条件时,推荐个人服务推荐方式中与满足的触发条件对应的服务信息;当用户场景信息不满足与获取的用户标识对应的个人服务推荐方式中的触发条件时,查询获取的用户标识所属的用户集群。
在一个实施例中,服务器120可按照获取的各集群通用服务推荐方式对应的权重选择相应的集群通用服务推荐方式;权重与对应的集群通用服务推荐方式作为个人服务推荐方式在用户集群中被共用的用户数或用户数占比相关;推荐选择的集群通用服务推荐方式中与满足的触发条件对应的服务信息。
在一个实施例中,终端110可向服务器120上报针对被推荐的服务信息的用户反馈。服务器120可获取针对被推荐的服务信息的用户反馈;根据用户反
馈调整被推荐的服务信息所在的集群通用服务推荐方式对应的权重。
图2为一个实施例中终端的内部结构示意图。参照图2,该终端包括通过***总线连接的处理器、非易失性存储介质、内存储器、网络接口、显示屏和输入装置。其中,终端的非易失性存储介质存储有操作***,还可存储有计算机可读指令,该计算机可读指令被处理器执行时,可使得处理器实现一种适用于终端的服务推荐方法。终端的处理器用于提供计算和控制能力,支撑整个终端的运行。终端中的内存储器中可存储有计算机可读指令,该计算机可读指令被处理器执行时,可使得处理器执行一种服务推荐方法。终端的网络接口用于与服务器进行网络通信,如上报用户标识和相应的用户场景信息、接收被推荐的服务信息等。终端的显示屏可以是液晶显示屏或者电子墨水显示屏,终端的输入装置可以是显示屏上覆盖的触摸层,也可以是终端外壳上设置的按键、轨迹球或触控板,还可以是外接的键盘、触控板或鼠标等。该终端可以是个人计算机、移动终端或者穿戴式设备,移动终端比如手机、平板电脑或者个人数字助理等。本领域技术人员可以理解,图2中示出的结构,仅仅是与本申请方案相关的部分结构的框图,并不构成对本申请方案所应用于其上的终端的限定,具体的终端可以包括比图2中所示更多或更少的部件,或者组合某些部件,或者具有不同的部件布置。
图3为一个实施例中服务器的内部结构示意图。如图3所示,该服务器包括通过***总线连接的处理器、非易失性存储介质、内存储器和网络接口。其中,该服务器的非易失性存储介质存储有操作***,还可存储有计算机可读指令,该计算机可读指令被处理器执行时,可使得处理器实现一种适用于服务器的服务推荐方法。该服务器的处理器用于提供计算和控制能力,支撑整个服务器的运行。该服务器的内存储器中可储存有计算机可读指令,该计算机可读指令被处理器执行时,可使得处理器执行一种服务推荐方法。该服务器的网络接口用于据以与外部的终端通过网络连接通信。服务器可以用独立的服务器或者是多个服务器组成的服务器集群来实现。本领域技术人员可以理解,图3中示出的结构,仅仅是与本申请方案相关的部分结构的框图,并不构成对本申请方案所应用于其上的服务器的限定,具体的服务器可以包括比图中所示更多或更
少的部件,或者组合某些部件,或者具有不同的部件布置。
图4为一个实施例中服务推荐方法的流程示意图。该服务推荐方法可应用于终端110或者服务器120,本实施例主要以该方法应用于上述图3中的服务器来举例说明。参照图4,本实施例的服务推荐方法具体包括如下步骤:
S402,获取用户标识和相应的用户场景信息。
其中,用户标识用于唯一标识出相应的用户。用户标识相应的用户场景信息,是表示用户所处场景的信息。用户场景信息可以是用户即时场景信息。用户即时场景信息是表示用户所处场景的即时状态的信息。即时是指当前时间或者近似当前时间。用户即时场景信息可以包括用户终端即时时间、用户即时地理位置或者用户即时操作信息中的一种或几种的组合。
用户终端即时时间是指以用户标识登录的终端的即时时间。用户终端即时时间根据需要可采用不同的形式,如通过日期和当日时间点表示的形式,或者仅通过当日时间点表示的形式。用户终端即时时间可以精确到小时或分钟或秒钟或毫秒等。
用户即时地理位置用于表示用户即时位置。用户即时地理位置可采用终端即时的地理位置,如终端所在位置的经度和纬度。用户即时地理位置也可以采用用户或终端所在信息点(POI,全称为Point ofInformation)。
用户即时操作信息是记录用户对终端即时操作的信息,如终端上当前正被使用的应用的应用标识、终端当前正被使用的应用中正被使用的功能的功能标识以及当前操作所产生的数据等中的一种或几种的组合。
S404,查询获取的用户标识所属的用户集群;用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到。
其中,服务推荐方式是指进行服务推荐所依据的数据,也可以称之为服务推荐策略。个人服务推荐方式是针对个人的服务推荐方式。个人服务推荐方式对应了用户标识,表示针对该用户标识采用该用户标识对应的个人服务推荐方式进行服务推荐。个人服务推荐方式集合是多用户源的,表示该集合包括了多于一个用户的个人服务推荐方式。个人服务推荐方式包括触发条件和相对应的
服务信息,表示触发条件被满足时向相应用户标识对应的终端推荐该触发条件对应的服务信息。
用户集群是聚类得到的用户标识的集合。聚类是把对象分成不同的子集,使得相似的对象属于相同子集的处理过程。参照图5,多用户源的个人服务推荐方式集合中的各个人服务推荐方式对应有用户标识,根据个人服务推荐方式的相似性将用户标识进行聚类,得到多于一个的用户集群。同一个用户集群中的用户标识对应的个人服务推荐方式具有相似性。服务器具体可将个人服务推荐方式集合中的各个人服务推荐方式聚类为个人服务推荐方式子集后,将各个人服务推荐方式子集对应的用户标识构成相应的用户集群。
根据个人服务推荐方式集合中的个人服务推荐方式进行聚类,可采用多种聚类算法,比如K-means聚类算法、人工神经网络算法或者支持向量机(SVM,Support Vector Machine)等。聚类时可将个人服务推荐方式集合中的个人服务推荐方式抽象为向量后进行聚类。采用K-means聚类算法时,可按照如下步骤进行聚类:(1)在个人服务推荐方式集合中随机选取K个聚类中心;(2)遍历个人服务推荐方式集合,将个人服务推荐方式集合中的个人服务推荐方式划分到最近的聚类中心得到相应的聚类;(3)计算每个聚类的平均值并作为新的聚类中心;重复(2)和(3),直至满足迭代停止条件。迭代停止条件比如达到最大迭代次数,或者最后一次计算出的聚类中心相较于上一次计算出的聚类中心的变化量小于预设值。
举例说明,若个人服务推荐方式集合中的一种个人服务推荐方式中,触发条件是中午时段和/或晚餐时段,相应的服务信息是外卖应用服务信息;在另一种个人服务推荐方式中,触发条件是上班时间段和/或下班时间段,相应的服务信息是网上约车应用服务信息。若这两种个人服务推荐方式对应的用户标识数超过预定值,则可将同时对应这两种个人服务推荐方式的用户标识聚类为一个用户集群,表示白领用户群。
S406,获取用户集群对应的集群通用服务推荐方式;集群通用服务推荐方式根据个人服务推荐方式集合中与用户集群对应的个人服务推荐方式生成。
集群通用服务推荐方式,是对相对应的用户集群中的用户标识进行服务推
荐时所共享的服务推荐方式。个人服务推荐方式集合中与用户集群对应的个人服务推荐方式,是指用户集群中的用户标识所对应的个人服务推荐方式集合中的个人服务推荐方式。
集群通用服务推荐方式具体可以采用个人服务推荐方式集合中与用户集群对应的所有个人服务推荐方式。也可以在个人服务推荐方式集合中与用户集群对应的所有个人服务推荐方式中,将对应用户集群中的用户标识数量多于预设值的个人服务推荐方式作为该用户集群对应的集群通用服务推荐方式。集群通用服务推荐方式也可以基于个人服务推荐方式集合中与用户集群对应的个人服务推荐方式进行个人触发条件的调整得到。
S408,当用户场景信息满足集群通用服务推荐方式中的触发条件时,推荐集群通用服务推荐方式中与满足的触发条件对应的服务信息。
其中,集群通用服务推荐方式包括触发条件和相应的服务信息。对于属于用户集群的用户标识,当该用户标识对应的用户场景信息满足该集群通用服务推荐方式所包括的触发条件时,将向该用户标识对应的终端推荐该集群通用服务推荐方式中与该触发条件对应的服务信息。服务信息是向用户提供服务的信息,可以是信息或者链接,链接可以是网页链接、打开指定应用的连接或者打开指定应用的指定功能的连接等。
举例说明,前述的白领用户群,若其中大量用户标识(比如超过预定数量或者超过预定比例的用户标识)对应有相同或相近的个人服务推荐方式,比如触发条件为周六或接近周六、且相应的服务信息为电影票购买应用服务信息的个人服务推荐方式。则可将该个人服务推荐方式作为白领用户群的集群通用服务推荐方式,当白领用户群中的任意用户标识对应的用户终端即时时间属于周六或者接近周六,则向该用户标识对应的终端推荐电影票购买应用服务信息。
上述服务推荐方法,获取到用户标识和相应的用户场景信息后,可以查询用户标识所属的用户集群,该用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到,于是该用户集群表示的是与获取到的用户标识所表示用户在个人服务推荐方式上相类似的用户群体。用户集群对应的集群通用服务推荐方式根据个人服务推荐方式集合中与用
户集群对应的个人服务推荐方式生成,使得各个人服务推荐方式可以在用户集群中通用。当用户场景信息满足集群通用服务推荐方式中的触发条件时,推荐集群通用服务推荐方式中与满足的触发条件对应的服务信息,可以避免推荐服务信息片面化,保证了推荐服务信息的丰富度。
在一个实施例中,在步骤S402之后,该服务推荐方法还包括:当用户场景信息满足与获取的用户标识对应的个人服务推荐方式中的触发条件时,推荐个人服务推荐方式中与满足的触发条件对应的服务信息;当用户场景信息不满足与获取的用户标识对应的个人服务推荐方式中的触发条件时,执行步骤S404。
具体地,参照图6,服务器在获取到用户标识和相应的用户场景信息后,可先判断用户场景信息是否满足用户标识对应的个人服务推荐方式中的触发条件。当满足个人服务推荐方式中的触发条件时,则服务器可直接向用户标识对应的终端推荐个人服务推荐方式中与满足的触发条件对应的服务信息。当不满足人服务推荐方式中的触发条件时,服务器可进一步判断用户场景信息是否满足用户标识所属的用户集群对应的集群通用服务推荐方式中的触发条件。
当满足用户标识所属的用户集群对应的集群通用服务推荐方式中的触发条件时,服务器可向用户标识对应的终端推荐集群通用服务推荐方式中与满足的触发条件对应的服务信息。当不满足用户标识所属的用户集群对应的集群通用服务推荐方式中的触发条件时,或者推荐服务信息后,服务器可将用户场景信息记录为用户行为场景信息。服务器在向终端推荐服务信息后,可接收用户终端针对该服务信息的用户反馈。
本实施例中,优先按照个人服务推荐方式进行服务推荐,可优先保证服务推荐的准确性;在无法按照人服务推荐方式进行服务推荐时再按照集群通用服务推荐方式进行服务推荐,则可以推荐用户潜在需求的服务信息,以尽可能满足用户需求。
在一个实施例中,该服务推荐方法还包括形成个人服务推荐方式的步骤。如图7和图8所示,该形成个人服务推荐方式的步骤具体包括如下步骤:
S702,获取用户场景行为记录以及相应的用户标识;用户场景行为记录包括用户行为信息和相应的用户行为场景信息。
其中,用户场景行为记录是记录用户在特定场景下行为的数据。用户行为可以是用户操作终端的行为。用户场景行为记录可通过用户终端上报的用户场景信息及用户反馈生成,也可以由用户终端直接形成用户场景行为记录后上报至服务器。用户场景行为记录可以包括用户行为场景信息和相应的用户行为信息,用户行为场景信息包括用户行为时间和/或用户行为地理位置。用户行为时间可以表示用户行为发生的时间,用户行为地理位置可以表示用户行为发生的地理位置。用户场景行为记录可以包括用户行为轨迹。用户行为轨迹可以表示多个用户行为发生的次序。
举例说明,一个用户场景行为记录可以记录为:2016-8-19,19:32,深圳科技园XX咖啡厅,XX咖啡厅会员应用-会员卡。该用户场景行为记录可以表示:用户在2016年8月19日19:32这一用户行为时间,在深圳科技园XX咖啡厅这一用户行为地理位置,实施了使用XX咖啡厅会员应用中的会员卡功能这一用户行为。
再比如,一个用户场景行为记录可以记录为:服务A,服务B。该用户场景行为记录表示用户在使用了服务A之后,使用了服务B。该用户场景行为记录表示用户行为轨迹。
S704,根据用户行为场景信息形成触发条件,根据用户行为信息形成服务信息。
具体地,服务器可直接将用户行为场景信息形成触发条件,比如将用户行为时间和/或用户行为地理位置形成触发条件。用户行为轨迹中先发生的用户行为可视为用户行为场景信息,后发生的用户行为可视为用户行为信息。当用户场景行为记录记录的是用户行为轨迹时,可将用户行为轨迹中先发生的用户行为形成触发条件,并将用户行为轨迹中后发生的用户行为形成与该触发条件对应的服务信息。
S706,将形成的触发条件与形成的服务信息对应,形成与获取的用户标识对应的个人服务推荐方式。
本实施例中,利用用户场景行为记录生成个人服务推荐方式,基于该个人服务推荐方式可以准确地进行服务推荐,提高了服务推荐准确性。
参照图9,在一个实施例中,步骤S704具体包括如下步骤:
S902,在统计时间段内统计对应于获取的用户标识且包括同类的用户行为场景信息的用户场景行为记录的数量。
其中,统计时间段可以是预设天数,具体可以是一周或者一个月,可以根据服务推荐的准确性要求选择合适的统计时间段。同类的用户行为场景信息是相同或者近似的用户行为场景信息,比如用户行为时间相差在预设范围内,或者用户行为地理位置相差在预设范围内。同类的用户行为场景信息也可以是包括相同的用户行为轨迹片段,用户行为轨迹片段是用户行为轨迹中截取的一部分。
举例说明,假设有三个用户场景行为记录,分别为场景A-服务A,场景B-服务B,场景C-服务C,若场景A、B和C相同或者非常接近,则场景A、B和C为同类的用户行为场景信息,可以将该三个用户场景行为记录作为相同的用户场景行为记录统计数量。
S904,当统计时间段内统计的数量高于预设阈值时,根据同类的用户行为场景信息形成触发条件,根据与同类的用户行为场景信息对应的用户行为信息形成服务信息。
具体地,预设阈值可根据需要设定。服务器可形成能够同时满足同类的用户行为场景信息的触发条件。服务器可形成能够覆盖与同类的用户行为场景信息对应的用户行为信息的服务信息。
本实施例中,根据包括同类的用户行为场景信息的用户场景行为记录中数量较多的用户场景行为记录形成个人服务推荐方式,使得个人服务推荐方式能够准确反映用户的服务推荐需求。无论基于个人服务推荐方式的服务推荐,还是基于集群通用服务推荐方式的服务推荐,都能够保证一定的准确性。
在一个实施例中,用户行为场景信息包括用户行为时间和/或用户行为地理位置;参照图10,步骤S902包括以下步骤:
S1002,生成与获取的用户标识对应的各用户场景行为记录所包括用户行为场景信息的模糊范围。
具体地,服务器可获取与用户标识对应的所有用户场景行为记录,并生成
各用户场景行为记录所包括的用户行为时间和/或用户行为地理位置的模糊范围。其中,模糊范围是指在用户行为场景信息基础上进行范围扩展而得到的范围。用户行为时间的模糊范围,比如以用户行为时间为中心进行范围扩展而形成的范围,比如用户行为时间为9:00,以9:00为中心左右各扩展1小时得到模糊范围为8:00至10:00。类似地,地理位置的模糊范围可以是以用户行为地理位置为中心进行范围扩展而形成的范围。
S1004,将各用户场景行为记录中的用户行为场景信息按照相应的模糊范围聚类,确定包括同类的用户行为场景信息的用户场景行为记录。
具体地,按照模糊范围将相应的用户行为场景信息进行聚类,可以是将存在交集或者交集面积大于预设面积的用户行为场景信息聚类为同类,从而确定包括同类的用户行为场景信息的用户场景行为记录。
S1006,在统计时间段内统计包括同类的用户行为场景信息的用户场景行为记录的数量。
本实施例中,利用用户行为时间和/或用户行为地理位置的模糊范围确定包括同类的用户行为场景信息的用户场景行为记录,可以更为有效地生成用户场景行为记录,避免用户行为场景信息过于具体而导致用户场景行为记录数量太少。
在一个实施例中,步骤S408中推荐集群通用服务推荐方式中与满足的触发条件对应的服务信息的步骤具体包括:按照获取的各集群通用服务推荐方式对应的权重选择相应的集群通用服务推荐方式;权重与对应的集群通用服务推荐方式作为个人服务推荐方式在用户集群中被共用的用户数或用户数占比相关;推荐选择的集群通用服务推荐方式中与满足的触发条件对应的服务信息。
具体地,用户集群可对应多个集群通用服务推荐方式,每个集群通用服务推荐方式具有权重,在进行服务推荐时可依据权重从多个集群通用服务推荐方式中选择一个集群通用服务推荐方式进行服务推荐。服务器可以根据权重确定各个通用服务推荐方式的概率,从而按照概率选择集群通用服务推荐方式。服务器也可以直接选择权重最大的集群通用服务推荐方式进行服务推荐,当权重相同时可以从权重相同的集群通用服务推荐方式中随机选择。
权重可以与该权重对应的集群通用服务推荐方式作为个人服务推荐方式在用户集群中被共用的用户数或用户数占比正相关,当然也可以是负相关。假设前述的白领用户群,包括用户A、用户B、用户C和用户D,且白领用户群对应的集群通用服务推荐方式为E和F。且集群通用服务推荐方式E为用户A、用户B和用户C共同的个人服务推荐方式,而集群通用服务推荐方式F仅为用户A和用户B的个人服务推荐方式,则E的权重高于F的权重,服务器将优先根据权重高的集群通用服务推荐方式E进行服务推荐。
本实施例中,按照权重选择集群通用服务推荐方式进行服务推荐,可以在存在多个集群通用服务推荐方式时选择合适的集群通用服务推荐方式进行服务推荐,尽量保证服务推荐能够满足用户的服务需求。
在一个实施例中,推荐个人服务推荐方式中与满足的触发条件对应的服务信息的步骤包括:按照各个人通用服务推荐方式对应的权重选择相应的个人通用服务推荐方式;推荐选择的个人通用服务推荐方式中与满足的触发条件对应的服务信息。权重可以与在统计时间段内统计对应于获取的用户标识且包括同类的用户行为场景信息的用户场景行为记录的数量相关。
在一个实施例中,该服务推荐方法还包括:获取针对被推荐的服务信息的用户反馈;根据用户反馈调整被推荐的服务信息所在的集群通用服务推荐方式对应的权重。
其中,用户反馈可以表示接受服务信息或者拒绝服务信息。当用户反馈表示接受服务信息时,服务器可将被推荐的服务信息所在的集群通用服务推荐方式对应的权重调高。当用户反馈表示拒绝服务信息时,服务器可将被推荐的服务信息所在的集群通用服务推荐方式对应的权重调低。
本实施例中,通过用户反馈可以调整集群通用服务推荐方式对应的权重,从而可以动态地调整服务推荐策略,以满足用户不断变化的服务需求。
图11为另一个实施例中服务推荐方法的流程示意图。本实施例主要以该方法应用于上述图2中的终端来举例说明。参照图11,该方法具体包括如下步骤:
S1102,向服务器上报用户标识和相应的用户场景信息,使得服务器查询上报的用户标识所属的用户集群,获取用户集群对应的集群通用服务推荐方式,
当用户场景信息满足集群通用服务推荐方式中的触发条件时,推荐集群通用服务推荐方式中与满足的触发条件对应的服务信息;用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到;集群通用服务推荐方式根据个人服务推荐方式集合中与用户集群对应的个人服务推荐方式生成。
S1104,接收服务器推荐的服务信息。
S1106,展示通知界面,根据服务信息在通知界面中展示相应的服务入口。
其中,通知界面可以通过终端操作***提供的通知渠道生成,比如下拉通知栏或者顶部通知栏。服务信息可以包括链接,终端可将该链接展示为服务入口。服务入口用于触发对相应服务的调用。终端可检测对服务入口的用户操作指令,当检测到用户操作指令时触发该服务入口所指向的服务,比如打开提供服务的应用,或者打开服务中的指定功能,或者链接到指定网页等。
上述服务推荐方法,获取到用户标识和相应的用户场景信息后,可以查询用户标识所属的用户集群,该用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到,于是该用户集群表示的是与获取到的用户标识所表示用户在个人服务推荐方式上相类似的用户群体。用户集群对应的集群通用服务推荐方式根据个人服务推荐方式集合中与用户集群对应的个人服务推荐方式生成,使得各个人服务推荐方式可以在用户集群中通用。当用户场景信息满足集群通用服务推荐方式中的触发条件时,推荐集群通用服务推荐方式中与满足的触发条件对应的服务信息,可以避免推荐服务信息片面化,保证了推荐服务信息的丰富度。
在一个实施例中,步骤S1102包括:向服务器上报用户标识和相应的用户场景信息,使得服务器当用户场景信息满足与上报的用户标识对应的个人服务推荐方式中的触发条件时,推荐个人服务推荐方式中与满足的触发条件对应的服务信息;当用户场景信息不满足与上报的用户标识对应的个人服务推荐方式中的触发条件时,查询上报的用户标识所属的用户集群,获取用户集群对应的集群通用服务推荐方式,当用户场景信息满足集群通用服务推荐方式中的触发条件时,推荐集群通用服务推荐方式中与满足的触发条件对应的服务信息。
举例说明,参照图12,当终端使用图片编辑应用时,终端将图片编辑应用标识和用户标识上报至服务器,且服务器上对应于该用户标识存在表示用户行为轨迹的个人服务推荐方式,且该服务推荐方式的触发条件为该图片编辑应用标识时,则从该服务推荐方式中获取与该触发条件对应的图片分享应用标识推荐至终端。终端则显示通知界面,在通知界面中显示用于进入图片分享应用的服务入口。
参照图13,终端还可以接收服务器发送的根据用户终端即时时间确定的服务信息,并在终端操作***的桌面展示通知界面,在该通知界面中展示服务入口。参照图14,服务信息可由服务器根据非与用户标识对应的个人服务推荐方式的集群通用服务推荐方式进行推荐,终端在接收到该服务信息后,可提示该服务信息根据非与用户标识对应的个人服务推荐方式的集群通用服务推荐方式进行推荐。
在一个实施例中,被推荐的服务信息包括于按照获取的各集群通用服务推荐方式对应的权重所选择的集群通用服务推荐方式。参照图15,该方法还包括触发用户反馈并向服务器发送,使得服务器根据用户反馈调整权重的步骤,具体包括如下步骤:
S1502,根据服务信息在通知界面中展示服务取消控件。
S1504,当检测到针对服务入口的用户操作指令时,进入服务入口所指向的应用。
S1506,当检测到针对服务入口或者服务取消控件的用户操作指令时,向服务器发送用户反馈,使得服务器根据用户反馈调整被推荐的服务信息所在的集群通用服务推荐方式对应的权重。
具体地,终端在检测到针对服务入口的用户操作指令时,生成表示接受服务信息的用户反馈并发送至服务器;终端在检测到针对服务入口或者服务取消控件的用户操作指令时,生成表示拒绝服务信息的用户反馈并发送至服务器,使得服务器根据用户反馈调整推荐服务信息所依据的集群通用服务推荐方式对应的权重。
本实施例中,通过用户反馈可以调整集群通用服务推荐方式对应的权重,
从而可以动态地调整服务推荐策略,以满足用户不断变化的服务需求。
图16为一个实施例中服务器1600的结构框图。服务器1600的内部结构可对应于如图3所示的结构,下述每个模块可全部或部分通过软件、硬件或其组合来实现。
参照图16,该服务器1600包括:数据获取模块1601、集群通用服务推荐方式获取模块1602和服务信息推荐模块1603。
数据获取模块1601,用于获取用户标识和相应的用户场景信息;
集群通用服务推荐方式获取模块1602,用于查询获取的用户标识所属的用户集群;获取用户集群对应的集群通用服务推荐方式;用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到;集群通用服务推荐方式根据个人服务推荐方式集合中与用户集群对应的个人服务推荐方式生成;
服务信息推荐模块1603,用于当用户场景信息满足集群通用服务推荐方式中的触发条件时,推荐集群通用服务推荐方式中与满足的触发条件对应的服务信息。
在一个实施例中,服务信息推荐模块1603还用于当用户场景信息满足与获取的用户标识对应的个人服务推荐方式中的触发条件时,推荐个人服务推荐方式中与满足的触发条件对应的服务信息。
集群通用服务推荐方式获取模块1602还用于当用户场景信息不满足与获取的用户标识对应的个人服务推荐方式中的触发条件时,查询获取的用户标识所属的用户集群。
图17为另一个实施例中服务器1600的结构框图。参照图17,服务器1600还包括:个人服务推荐方式生成模块1604,用于获取用户场景行为记录以及相应的用户标识;用户场景行为记录包括用户行为信息和相应的用户行为场景信息;根据用户行为场景信息形成触发条件,根据用户行为信息形成服务信息;将形成的触发条件与形成的服务信息对应,形成与获取的用户标识对应的个人服务推荐方式。
在一个实施例中,个人服务推荐方式生成模块1604还用于在统计时间段内
统计对应于获取的用户标识且包括同类的用户行为场景信息的用户场景行为记录的数量;当统计时间段内统计的数量高于预设阈值时,根据同类的用户行为场景信息形成触发条件,根据与同类的用户行为场景信息对应的用户行为信息形成服务信息。
在一个实施例中,用户行为场景信息包括用户行为时间和/或用户行为地理位置;个人服务推荐方式生成模块1604还用于生成与获取的用户标识对应的各用户场景行为记录所包括用户行为场景信息的模糊范围;将各用户场景行为记录中的用户行为场景信息按照相应的模糊范围聚类,确定包括同类的用户行为场景信息的用户场景行为记录;在统计时间段内统计包括同类的用户行为场景信息的用户场景行为记录的数量。
在一个实施例中,服务信息推荐模块1603还用于按照获取的各集群通用服务推荐方式对应的权重选择相应的集群通用服务推荐方式;权重与对应的集群通用服务推荐方式作为个人服务推荐方式在用户集群中被共用的用户数或用户数占比相关;推荐选择的集群通用服务推荐方式中与满足的触发条件对应的服务信息。
在一个实施例中,服务器1600还包括:权重调整模块1605,用于获取针对被推荐的服务信息的用户反馈;根据用户反馈调整被推荐的服务信息所在的集群通用服务推荐方式对应的权重。
上述服务器1600,获取到用户标识和相应的用户场景信息后,可以查询用户标识所属的用户集群,该用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到,于是该用户集群表示的是与获取到的用户标识所表示用户在个人服务推荐方式上相类似的用户群体。用户集群对应的集群通用服务推荐方式根据个人服务推荐方式集合中与用户集群对应的个人服务推荐方式生成,使得各个人服务推荐方式可以在用户集群中通用。当用户场景信息满足集群通用服务推荐方式中的触发条件时,推荐集群通用服务推荐方式中与满足的触发条件对应的服务信息,可以避免推荐服务信息片面化,保证了推荐服务信息的丰富度。
图18为一个实施例中终端1800的结构框图。终端1800的内部结构可对应
于如图2所示的结构,下述每个模块可全部或部分通过软件、硬件或其组合来实现。
参照图18,终端1800包括:上报模块1801、服务信息接收模块1802和通知界面展示模块1803。
上报模块1801,用于向服务器上报用户标识和相应的用户场景信息,使得服务器查询上报的用户标识所属的用户集群,获取用户集群对应的集群通用服务推荐方式,当用户场景信息满足集群通用服务推荐方式中的触发条件时,推荐集群通用服务推荐方式中与满足的触发条件对应的服务信息;用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到;集群通用服务推荐方式根据个人服务推荐方式集合中与用户集群对应的个人服务推荐方式生成。
服务信息接收模块1802,用于接收服务器推荐的服务信息。
通知界面展示模块1803,用于展示通知界面,根据服务信息在通知界面中展示相应的服务入口。
在一个实施例中,上报模块1801还用于向服务器上报用户标识和相应的用户场景信息,使得服务器当用户场景信息满足与上报的用户标识对应的个人服务推荐方式中的触发条件时,推荐个人服务推荐方式中与满足的触发条件对应的服务信息;当用户场景信息不满足与上报的用户标识对应的个人服务推荐方式中的触发条件时,查询上报的用户标识所属的用户集群,获取用户集群对应的集群通用服务推荐方式,当用户场景信息满足集群通用服务推荐方式中的触发条件时,推荐集群通用服务推荐方式中与满足的触发条件对应的服务信息。
在一个实施例中,被推荐的服务信息包括于按照获取的各集群通用服务推荐方式对应的权重所选择的集群通用服务推荐方式。
通知界面展示模块1803还用于根据服务信息在通知界面中展示服务取消控件。
参照图19,终端1800还包括应用调用模块1804和用户反馈发送模块1805。
应用调用模块1804,用于当检测到针对服务入口的用户操作指令时,进入服务入口所指向的应用。
用户反馈发送模块1805,用于当检测到针对服务入口或者服务取消控件的用户操作指令时,向服务器发送用户反馈,使得服务器根据用户反馈调整被推荐的服务信息所在的集群通用服务推荐方式对应的权重。
上述服务器1600,获取到用户标识和相应的用户场景信息后,可以查询用户标识所属的用户集群,该用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到,于是该用户集群表示的是与获取到的用户标识所表示用户在个人服务推荐方式上相类似的用户群体。用户集群对应的集群通用服务推荐方式根据个人服务推荐方式集合中与用户集群对应的个人服务推荐方式生成,使得各个人服务推荐方式可以在用户集群中通用。当用户场景信息满足集群通用服务推荐方式中的触发条件时,推荐集群通用服务推荐方式中与满足的触发条件对应的服务信息,可以避免推荐服务信息片面化,保证了推荐服务信息的丰富度。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程,是可以通过计算机程序来指令相关的硬件来完成,该程序可存储于一非易失性计算机可读取存储介质中,该程序在执行时,可包括如上述各方法的实施例的流程。其中,该存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory,ROM)等。
以上实施例的各技术特征可以进行任意的组合,为使描述简洁,未对上述实施例中的各个技术特征所有可能的组合都进行描述,然而,只要这些技术特征的组合不存在矛盾,都应当认为是本说明书记载的范围。
以上实施例仅表达了本申请的几种实施方式,其描述较为具体和详细,但并不能因此而理解为对发明专利范围的限制。应当指出的是,对于本领域的普通技术人员来说,在不脱离本申请构思的前提下,还可以做出若干变形和改进,这些都属于本申请的保护范围。因此,本申请专利的保护范围应以所附权利要求为准。
Claims (21)
- 一种服务推荐方法,执行于服务器,所述服务器包括存储器和处理器,所述方法包括:获取用户标识和相应的用户场景信息;查询获取的用户标识所属的用户集群;所述用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到;获取所述用户集群对应的集群通用服务推荐方式;所述集群通用服务推荐方式根据所述个人服务推荐方式集合中与所述用户集群对应的个人服务推荐方式生成;及当所述用户场景信息满足所述集群通用服务推荐方式中的触发条件时,推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息。
- 根据权利要求1所述的方法,其特征在于,所述获取用户标识和相应的用户场景信息的步骤之后,所述方法包括:当所述用户场景信息满足与获取的所述用户标识对应的个人服务推荐方式中的触发条件时,推荐所述个人服务推荐方式中与满足的触发条件对应的服务信息;当所述用户场景信息不满足与获取的所述用户标识对应的个人服务推荐方式中的触发条件时,执行所述查询获取的用户标识所属的用户集群的步骤。
- 根据权利要求1所述的方法,其特征在于,所述方法还包括:获取用户场景行为记录以及相应的用户标识;所述用户场景行为记录包括用户行为信息和相应的用户行为场景信息;根据所述用户行为场景信息形成触发条件,根据所述用户行为信息形成服务信息;及将形成的触发条件与形成的服务信息对应,形成与获取的用户标识对应的个人服务推荐方式。
- 根据权利要求3所述的方法,其特征在于,所述根据所述用户行为场 景信息形成触发条件,根据所述用户行为信息形成服务信息包括:在统计时间段内统计对应于获取的所述用户标识且包括同类的用户行为场景信息的用户场景行为记录的数量;及当所述统计时间段内统计的数量高于预设阈值时,根据所述同类的用户行为场景信息形成触发条件,根据与所述同类的用户行为场景信息对应的所述用户行为信息形成服务信息。
- 根据权利要求4所述的方法,其特征在于,所述用户行为场景信息包括用户行为时间和/或用户行为地理位置;所述在统计时间段内统计对应于获取的所述用户标识且包括同类的用户行为场景信息的用户场景行为记录的数量包括:生成与获取的所述用户标识对应的各用户场景行为记录所包括用户行为场景信息的模糊范围;将各用户场景行为记录中的用户行为场景信息按照相应的模糊范围聚类,确定包括同类的用户行为场景信息的用户场景行为记录;及在统计时间段内统计所述包括同类的用户行为场景信息的用户场景行为记录的数量。
- 根据权利要求1所述的方法,其特征在于,所述推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息包括:按照获取的各集群通用服务推荐方式对应的权重选择相应的集群通用服务推荐方式;所述权重与对应的集群通用服务推荐方式作为个人服务推荐方式在所述用户集群中被共用的用户数或用户数占比相关;及推荐选择的集群通用服务推荐方式中与满足的触发条件对应的服务信息。
- 根据权利要求6所述的方法,其特征在于,所述方法还包括:获取针对被推荐的所述服务信息的用户反馈;及根据所述用户反馈调整被推荐的所述服务信息所在的集群通用服务推荐方式对应的权重。
- 一种服务推荐方法,执行于终端,所述终端包括存储器和处理器,所述方法包括:向服务器上报用户标识和相应的用户场景信息,使得所述服务器查询上报的用户标识所属的用户集群,获取所述用户集群对应的集群通用服务推荐方式,当所述用户场景信息满足所述集群通用服务推荐方式中的触发条件时,推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息;所述用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到;所述集群通用服务推荐方式根据所述个人服务推荐方式集合中与所述用户集群对应的个人服务推荐方式生成;接收所述服务器推荐的所述服务信息;展示通知界面,根据所述服务信息在所述通知界面中展示相应的服务入口。
- 根据权利要求8所述的方法,其特征在于,所述向服务器上报用户标识和相应的用户场景信息,使得所述服务器查询上报的用户标识所属的用户集群,获取所述用户集群对应的集群通用服务推荐方式,当所述用户场景信息满足所述集群通用服务推荐方式中的触发条件时,推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息包括:向服务器上报用户标识和相应的用户场景信息,使得所述服务器当所述用户场景信息满足与上报的所述用户标识对应的个人服务推荐方式中的触发条件时,推荐所述个人服务推荐方式中与满足的触发条件对应的服务信息;当所述用户场景信息不满足与上报的所述用户标识对应的个人服务推荐方式中的触发条件时,查询上报的用户标识所属的用户集群,获取所述用户集群对应的集群通用服务推荐方式,当所述用户场景信息满足所述集群通用服务推荐方式中的触发条件时,推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息。
- 根据权利要求8所述的方法,其特征在于,被推荐的服务信息包括于按照获取的各集群通用服务推荐方式对应的权重所选择的集群通用服务推 荐方式;所述方法还包括:根据所述服务信息在所述通知界面中展示服务取消控件;当检测到针对所述服务入口的用户操作指令时,进入所述服务入口所指向的应用;当检测到针对所述服务入口或者所述服务取消控件的用户操作指令时,向所述服务器发送用户反馈,使得所述服务器根据所述用户反馈调整被推荐的所述服务信息所在的集群通用服务推荐方式对应的权重。
- 一种服务器,包括存储器和处理器,所述存储器中储存有计算机可读指令,所述计算机可读指令被所述处理器执行时,使得所述处理器执行以下步骤:获取用户标识和相应的用户场景信息;查询获取的用户标识所属的用户集群;所述用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到;获取所述用户集群对应的集群通用服务推荐方式;所述集群通用服务推荐方式根据所述个人服务推荐方式集合中与所述用户集群对应的个人服务推荐方式生成;及当所述用户场景信息满足所述集群通用服务推荐方式中的触发条件时,推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息。
- 根据权利要求11所述的服务器,其特征在于,所述获取用户标识和相应的用户场景信息的步骤之后,所述计算机可读指令还使得所述处理器执行以下步骤:当所述用户场景信息满足与获取的所述用户标识对应的个人服务推荐方式中的触发条件时,推荐所述个人服务推荐方式中与满足的触发条件对应的服务信息;当所述用户场景信息不满足与获取的所述用户标识对应的个人服务推荐方式中的触发条件时,执行所述查询获取的用户标识所属的用户集群的步骤。
- 根据权利要求11所述的服务器,其特征在于,所述计算机可读指令还使得所述处理器执行以下步骤:获取用户场景行为记录以及相应的用户标识;所述用户场景行为记录包括用户行为信息和相应的用户行为场景信息;根据所述用户行为场景信息形成触发条件,根据所述用户行为信息形成服务信息;及将形成的触发条件与形成的服务信息对应,形成与获取的用户标识对应的个人服务推荐方式。
- 根据权利要求13所述的服务器,其特征在于,所述根据所述用户行为场景信息形成触发条件,根据所述用户行为信息形成服务信息包括:在统计时间段内统计对应于获取的所述用户标识且包括同类的用户行为场景信息的用户场景行为记录的数量;及当所述统计时间段内统计的数量高于预设阈值时,根据所述同类的用户行为场景信息形成触发条件,根据与所述同类的用户行为场景信息对应的所述用户行为信息形成服务信息。
- 根据权利要求14所述的服务器,其特征在于,所述用户行为场景信息包括用户行为时间和/或用户行为地理位置;所述在统计时间段内统计对应于获取的所述用户标识且包括同类的用户行为场景信息的用户场景行为记录的数量包括:生成与获取的所述用户标识对应的各用户场景行为记录所包括用户行为场景信息的模糊范围;将各用户场景行为记录中的用户行为场景信息按照相应的模糊范围聚类,确定包括同类的用户行为场景信息的用户场景行为记录;及在统计时间段内统计所述包括同类的用户行为场景信息的用户场景行为记录的数量。
- 根据权利要求11所述的服务器,其特征在于,所述推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息包括:按照获取的各集群通用服务推荐方式对应的权重选择相应的集群通用服务推荐方式;所述权重与对应的集群通用服务推荐方式作为个人服务推荐方式在所述用户集群中被共用的用户数或用户数占比相关;及推荐选择的集群通用服务推荐方式中与满足的触发条件对应的服务信息。
- 根据权利要求16所述的服务器,其特征在于,所述计算机可读指令还使得所述处理器执行以下步骤:获取针对被推荐的所述服务信息的用户反馈;及根据所述用户反馈调整被推荐的所述服务信息所在的集群通用服务推荐方式对应的权重。
- 一种终端,包括存储器和处理器,所述存储器中储存有计算机可读指令,所述计算机可读指令被所述处理器执行时,使得所述处理器执行以下步骤:向服务器上报用户标识和相应的用户场景信息,使得所述服务器查询上报的用户标识所属的用户集群,获取所述用户集群对应的集群通用服务推荐方式,当所述用户场景信息满足所述集群通用服务推荐方式中的触发条件时,推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息;所述用户集群根据多用户源的个人服务推荐方式集合中的各个人服务推荐方式将对应的用户标识进行聚类得到;所述集群通用服务推荐方式根据所述个人服务推荐方式集合中与所述用户集群对应的个人服务推荐方式生成;接收所述服务器推荐的所述服务信息;展示通知界面,根据所述服务信息在所述通知界面中展示相应的服务入口。
- 根据权利要求18所述的终端,其特征在于,所述向服务器上报用户标识和相应的用户场景信息,使得所述服务器查询上报的用户标识所属的用户集群,获取所述用户集群对应的集群通用服务推荐方式,当所述用户场景信息满足所述集群通用服务推荐方式中的触发条件时,推荐所述集群通用服 务推荐方式中与满足的触发条件对应的服务信息包括:向服务器上报用户标识和相应的用户场景信息,使得所述服务器当所述用户场景信息满足与上报的所述用户标识对应的个人服务推荐方式中的触发条件时,推荐所述个人服务推荐方式中与满足的触发条件对应的服务信息;当所述用户场景信息不满足与上报的所述用户标识对应的个人服务推荐方式中的触发条件时,查询上报的用户标识所属的用户集群,获取所述用户集群对应的集群通用服务推荐方式,当所述用户场景信息满足所述集群通用服务推荐方式中的触发条件时,推荐所述集群通用服务推荐方式中与满足的触发条件对应的服务信息。
- 根据权利要求18所述的终端,其特征在于,被推荐的服务信息包括于按照获取的各集群通用服务推荐方式对应的权重所选择的集群通用服务推荐方式;所述计算机可读指令还使得所述处理器执行以下步骤:根据所述服务信息在所述通知界面中展示服务取消控件;当检测到针对所述服务入口的用户操作指令时,进入所述服务入口所指向的应用;当检测到针对所述服务入口或者所述服务取消控件的用户操作指令时,向所述服务器发送用户反馈,使得所述服务器根据所述用户反馈调整被推荐的所述服务信息所在的集群通用服务推荐方式对应的权重。
- 一种非易失性的计算机可读存储介质,存储有计算机可读指令,所述计算机可读指令被处理器执行时,使得所述处理器执行如权利要求1至10中任一项所述方法的步骤。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610865652.0A CN107885742B (zh) | 2016-09-29 | 2016-09-29 | 服务推荐方法和装置 |
CN201610865652.0 | 2016-09-29 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2018059122A1 true WO2018059122A1 (zh) | 2018-04-05 |
Family
ID=61763299
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2017/095946 WO2018059122A1 (zh) | 2016-09-29 | 2017-08-04 | 服务推荐方法、终端、服务器和存储介质 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN107885742B (zh) |
WO (1) | WO2018059122A1 (zh) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110532474A (zh) * | 2019-08-30 | 2019-12-03 | 腾讯科技(深圳)有限公司 | 信息推荐方法、服务器、***以及计算机可读存储介质 |
CN111831903A (zh) * | 2020-06-15 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | 服务站点推荐方法、装置、设备及存储介质 |
CN112988559A (zh) * | 2019-12-17 | 2021-06-18 | 北京沃东天骏信息技术有限公司 | 请求的分流方法和装置 |
CN113449184A (zh) * | 2021-06-23 | 2021-09-28 | 平安科技(深圳)有限公司 | 触达渠道的推荐方法、装置、计算机设备和存储介质 |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108989547B (zh) * | 2018-06-20 | 2021-03-16 | Oppo广东移动通信有限公司 | 发光控制方法、装置、终端及存储介质 |
CN108959053B (zh) * | 2018-07-10 | 2021-12-14 | 海信视像科技股份有限公司 | 一种用户行为日志的生成方法及装置 |
CN109255606A (zh) * | 2018-08-31 | 2019-01-22 | 阿里巴巴集团控股有限公司 | 默认支付工具的推荐方法、装置及支付终端 |
CN111071182A (zh) * | 2019-12-31 | 2020-04-28 | 长城汽车股份有限公司 | 一种车辆配置功能的推荐方法、装置及*** |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101957834A (zh) * | 2010-08-12 | 2011-01-26 | 百度在线网络技术(北京)有限公司 | 一种基于用户特征进行内容推荐的方法与设备 |
CN102594905A (zh) * | 2012-03-07 | 2012-07-18 | 南京邮电大学 | 一种基于场景的社交网络位置兴趣点推荐方法 |
CN103248658A (zh) * | 2012-02-10 | 2013-08-14 | 富士通株式会社 | 服务推荐装置、服务推荐方法和移动设备 |
CN103886090A (zh) * | 2014-03-31 | 2014-06-25 | 北京搜狗科技发展有限公司 | 基于用户喜好的内容推荐方法及装置 |
CN103914536A (zh) * | 2014-03-31 | 2014-07-09 | 北京百度网讯科技有限公司 | 一种用于电子地图的兴趣点推荐方法及*** |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101685458B (zh) * | 2008-09-27 | 2012-09-19 | 华为技术有限公司 | 一种基于协同过滤的推荐方法和*** |
CN103034508B (zh) * | 2011-10-10 | 2015-08-19 | 腾讯科技(深圳)有限公司 | 软件推荐方法和*** |
CN103514496B (zh) * | 2012-06-21 | 2017-05-17 | 腾讯科技(深圳)有限公司 | 一种推荐目标软件的处理方法及*** |
CN104008184A (zh) * | 2014-06-10 | 2014-08-27 | 百度在线网络技术(北京)有限公司 | 信息的推送方法和装置 |
-
2016
- 2016-09-29 CN CN201610865652.0A patent/CN107885742B/zh active Active
-
2017
- 2017-08-04 WO PCT/CN2017/095946 patent/WO2018059122A1/zh active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101957834A (zh) * | 2010-08-12 | 2011-01-26 | 百度在线网络技术(北京)有限公司 | 一种基于用户特征进行内容推荐的方法与设备 |
CN103248658A (zh) * | 2012-02-10 | 2013-08-14 | 富士通株式会社 | 服务推荐装置、服务推荐方法和移动设备 |
CN102594905A (zh) * | 2012-03-07 | 2012-07-18 | 南京邮电大学 | 一种基于场景的社交网络位置兴趣点推荐方法 |
CN103886090A (zh) * | 2014-03-31 | 2014-06-25 | 北京搜狗科技发展有限公司 | 基于用户喜好的内容推荐方法及装置 |
CN103914536A (zh) * | 2014-03-31 | 2014-07-09 | 北京百度网讯科技有限公司 | 一种用于电子地图的兴趣点推荐方法及*** |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110532474A (zh) * | 2019-08-30 | 2019-12-03 | 腾讯科技(深圳)有限公司 | 信息推荐方法、服务器、***以及计算机可读存储介质 |
CN110532474B (zh) * | 2019-08-30 | 2023-08-08 | 腾讯科技(深圳)有限公司 | 信息推荐方法、服务器、***以及计算机可读存储介质 |
CN112988559A (zh) * | 2019-12-17 | 2021-06-18 | 北京沃东天骏信息技术有限公司 | 请求的分流方法和装置 |
CN111831903A (zh) * | 2020-06-15 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | 服务站点推荐方法、装置、设备及存储介质 |
CN113449184A (zh) * | 2021-06-23 | 2021-09-28 | 平安科技(深圳)有限公司 | 触达渠道的推荐方法、装置、计算机设备和存储介质 |
Also Published As
Publication number | Publication date |
---|---|
CN107885742A (zh) | 2018-04-06 |
CN107885742B (zh) | 2021-09-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2018059122A1 (zh) | 服务推荐方法、终端、服务器和存储介质 | |
USRE48194E1 (en) | Systems and methods for monitoring and applying data related to shareable links associated with content items stored in an online content management service | |
CN110476176B (zh) | 用户目标辅助技术 | |
CN107924506B (zh) | 用于推断用户可用性的方法、***及计算机存储介质 | |
US10748121B2 (en) | Enriching calendar events with additional relevant information | |
US20180260490A1 (en) | Method and system for recommending text content, and storage medium | |
CN105051719B (zh) | 用于使得众包内容动态失效的装置和方法 | |
US10185973B2 (en) | Inferring venue visits using semantic information | |
TWI533246B (zh) | 使用者未知興趣之探索方法與系統 | |
CN107851243B (zh) | 推断物理会议位置 | |
CN111615712A (zh) | 多日历协调 | |
EP3288622A1 (en) | Unusualness of events based on user routine models | |
JP2018077821A (ja) | ユーザによって訪問される施設のカテゴリの予測モデルを生成する方法、プログラム、サーバ装置、及び処理装置 | |
US20130254152A1 (en) | Distributed system and methods for modeling population-centric activities | |
US11521238B2 (en) | Method and system for determining fact of visit of user to point of interest | |
US20220078135A1 (en) | Signal upload optimization | |
CN113454669A (zh) | 通过用户访问的特征来表征地点 | |
US20240224223A1 (en) | Evaluating IP Location On A Client Device | |
US10733244B2 (en) | Data retrieval system | |
EP4024906B1 (en) | Method for identifying a device using attributes and location signatures from the device | |
JP5813052B2 (ja) | 情報処理装置、方法及びプログラム | |
US11403324B2 (en) | Method for real-time cohort creation based on entity attributes derived from partially observable location data | |
WO2023105568A1 (ja) | 情報提供装置、情報提供方法及びプログラム | |
CA3009700C (en) | On-demand location visit conversion metrics | |
Soikkeli et al. | The 19th ITS Biennial Conference 2012 “Moving Forward with Future Technologies: Opening |
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: 17854575 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: 17854575 Country of ref document: EP Kind code of ref document: A1 |