WO2019174379A1

WO2019174379A1 - Operation path navigation

Info

Publication number: WO2019174379A1
Application number: PCT/CN2019/070617
Authority: WO
Inventors: 穆毅鹏; 廉洁
Original assignee: 阿里巴巴集团控股有限公司
Priority date: 2018-03-16
Filing date: 2019-01-07
Publication date: 2019-09-19
Also published as: CN108647064A; TW201938993A

Abstract

Provided in an embodiment of the present description are a method and a device for user operation path navigation. The method comprises: first acquiring an operation sequence and user feature data of a current user, and then employing a path navigation model to predict a target operation page of the current user on the basis of the operation sequence and the user feature data; next, recommending the target operation page to the current user; determining a modification to the path navigation model depending on whether the current user accepts the recommended target operation page. As such, the present invention helps a user to quickly reach a target operation page.

Description

Operation path navigation

Cross-reference to related applications

The present application claims priority to Chinese Patent Application No. 201 810 020 305 8.5 filed on March 16, 2018, the entire disclosure of which is incorporated herein by reference. in.

Technical field

One or more embodiments of the present specification relate to the field of computer technology, and more particularly to a method and apparatus for operating path navigation.

Background technique

With the development of Internet technology, people often use a variety of client software, such as mobile app, for a variety of transactions, including shopping, payment, social, financial management. This makes the interaction between the client software and the user very important. Complex client software has many functions. If the interaction design is not reasonable enough, users can not reach and execute the target operation in a simple and quick way, which will reduce the user experience and reduce the service capability of the product. In particular, for some operations that require more timeliness, for example, the operation of funds and stock trading in some wealth management software products, if the target operation cannot be performed quickly, it may even cause the user's property loss. And, on the other hand, different users have different operating habits and appeals. This puts higher demands on the user interaction design of the client software.

Therefore, it is desirable to have an improved solution to help users perform target operations quickly and efficiently through effective operation path navigation.

Summary of the invention

One or more embodiments of the present specification describe a method and apparatus for facilitating a user to quickly perform a target operation by operating a path navigation.

According to a first aspect, a method for user operation path navigation is provided, including: acquiring an operation sequence and user feature data of a current user; and using a path navigation model to predict the current user based on the operation sequence and user feature data a target operation page; recommending the target operation page to the current user; determining a modification to the path navigation model according to whether the current user accepts the recommended target operation page.

In one embodiment, the path navigation model is configured to establish a correspondence between a plurality of user groups and an operation mode based on a historical operation data set and a user feature data set of the plurality of users.

According to a possible design, the target operation page of the current user is predicted by: determining, according to the user feature data, a specific user group corresponding to the current user; and a plurality of user groups and operation modes established according to the path navigation model Corresponding relationship, determining a specific operation mode corresponding to a specific user group; determining, according to the specific operation mode, a specific operation path that matches an operation sequence of the current user; determining, according to the specific operation path, a target operation of the current user page.

According to a possible design, the target operation page is provided by one of the following ways: providing a floating layer icon on the original interface, the floating layer icon pointing to the target operation page; adding an icon in the interface, the adding The icon points to the target operation page; the target operation page is provided to the current user as a next operation page.

According to an embodiment, determining the modification to the path navigation model may include determining a recommendation accuracy of the path navigation model according to whether the current user accepts the recommended target operation page; and the recommendation accuracy is lower than a preset threshold In the case of the path navigation model, the modification is performed.

Further, in an embodiment, modifying the path navigation model includes: recording a subsequent operation of the current user; forming a target operation path according to the operation sequence and the subsequent operation; updating according to the target operation path Correspondence between multiple user groups and operation modes established by the path navigation model.

According to a second aspect, an apparatus for user operation path navigation is provided, comprising: an obtaining unit configured to acquire an operation sequence and user feature data of a current user; and a prediction unit configured to adopt a path navigation model, based on the operation sequence And user feature data, predicting a target operation page of the current user; a recommendation unit configured to recommend the target operation page to the current user; and a determining unit configured to determine, according to whether the current user accepts the recommended target operation page, A modification to the path navigation model is determined.

According to a third aspect, there is provided a computer readable storage medium having stored thereon a computer program for causing a computer to perform the method of the first aspect when the computer program is executed in a computer.

According to a fourth aspect, there is provided a computing device comprising a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, implementing the method of the first aspect .

Through the method and device provided by the embodiments of the present specification, the path navigation model is used to recommend the target operation page to the user, and the model is continuously trained and improved according to the user feedback, so that the recommendation accuracy of the model is continuously improved, thereby helping the user to be more quickly. Reach the target page, simplify user operations and enhance the user experience.

DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the present invention, Those skilled in the art can also obtain other drawings based on these drawings without any creative work.

1 is a schematic diagram of an implementation scenario of an embodiment disclosed in the present specification;

Figure 2 shows a schematic diagram of the user's operating path in one example;

3 illustrates a flow diagram of a method of user operation path navigation, in accordance with one embodiment;

4 shows a flow diagram of a predicted target operation page, in accordance with one embodiment;

Figure 5 shows a schematic diagram of a new icon in accordance with one embodiment;

Figure 6 shows a schematic diagram of a floating layer icon in accordance with one embodiment.

Figure 7 shows a schematic block diagram of an operational path navigation device in accordance with one embodiment.

detailed description

The solution provided in this specification will be described below with reference to the accompanying drawings.

FIG. 1 is a schematic diagram of an implementation scenario of an embodiment disclosed in the present specification. As shown in the figure, each user can form user history operation information by using client software, such as Alipay App, Ant Fortune App, and the like. A computing platform, such as the server of the client software described above, can construct a path navigation model based on the historical operation information of the user and the feature information of the user. The path navigation model obtains usage habits of various users by analyzing the above historical operation information and user characteristic information. Then, the model can be tested and trained while using the path navigation model to navigate the user operation path. When the operation sequence of the user A entering the client software is obtained, the captured operation sequence and the user characteristic information of the user A are input into the path navigation model. The path navigation model predicts the next target operation page of the user A based on the previous analysis of the usage habits of various users. The computing platform then recommends the target operation page to the user. Next, according to whether the user A selects the recommended target operation, the prediction accuracy of the target operation is determined, and the path navigation model is modified with the prediction accuracy as feedback. Through such a process, while using the model to recommend the target operation to the user, the model is modified according to the user's selection as a feedback result, thereby continuously improving the accuracy of the model recommendation, better recommending the target operation for the user, and helping the user to efficiently Achieve the target operation. The specific implementation steps of the above process are described below.

In accordance with an embodiment described in this specification, in a first phase, a path navigation model is established for a user history operational data set and a user feature information data set. In the modeling phase, at least two aspects of information are obtained: a user history operation information data set, and a user feature information data set. A path navigation model is established for the information of the two aspects, and the model is used for analyzing and learning the historical operation information and the user characteristic information of the user, thereby obtaining the operation habits of various users.

The user history operation information includes a user history operation track, or a history operation path. In order to obtain such information, in general, when the user operates on the terminal interface, the user's operational behavior and related pages are captured. The user's operational behavior may include various operations performed by the user on the page, such as clicking, swiping, dragging, character typing, zooming in or out, and the like. The above page may be a web page, a web application page, an app page or other page that can interact with the user. The user's operation can be captured by means of burying the terminal application. For example, various interactive modules in the page, such as buttons, scroll bars, banners, etc., are numbered in advance to facilitate recording of user operations.

Figure 2 shows a schematic diagram of the user's operational path in one example. In FIG. 2, taking Alipay as an example, the user's operation on the page is shown. As shown in FIG. 2, the user starts from the Alipay main page A1, clicks "ant wealth" (performs operation P1), and enters page A2. On the A2 page, click on “Financial Management” (perform operation P2) to enter the financial page A3. Further financial items are displayed on the A3 page. Assume that the user clicks on "Fund" on the A3 page (performs operation P3), thereby entering page A4, which shows various fund selections. In the page A4, the user clicks on "Fund Ranking" (execute operation P4), and then proceeds to page A5, where page A6 is displayed by clicking "Evaluation Ranking" (perform operation P5). After the user browses the content of page A6, the client is closed. Thus, A1.P1-A2.P2-A3.P3-A4.P4-A5.P5-A6 shows a user's operation path OP1.

In one embodiment, relevant time information is also recorded while capturing user operations. As such, the user history operation information further includes time parameter information such as an operation time and a dwell time. For example, for the operation path OP1: A1.P1-A2.P2-A3.P3-A4.P4-A5.P5-A6 shown in FIG. 3, it is also possible to record that the user performs various operations P1-P2-P3-P4-P5. The moments, such as T1-T2-T3-T4-T5, and the user's dwell time on each page, such as ST1-ST2-ST3-ST4-ST5-ST6. It is also possible to determine information such as the time interval of the operation based on the operation time and/or the dwell time.

In one embodiment, the above user history operation information may be recorded in various forms such as a sequence, a vector, a matrix, and the like. For example, in one example, an operational path is recorded as a vector, each vector element corresponding to the number of an interactive module for which the user operation is directed. In one example, the operation path, and the time information corresponding to the operation path, are organized into sequences of the same length, and the sequences are organized into a matrix form. It can be understood that user history operation information can also be recorded in other formats or forms.

In addition to the user history operation information, on the other hand, the user's feature information is also acquired. In one embodiment, the user feature information includes basic attribute information of the user, such as attribute information included in the registration information, such as age, gender, region, and the like. In one embodiment, the user profile information further includes other attribute information of the user related to the client software function. For example, for financial client software, such as Alipay, Ant Wealth, it is also possible to obtain information such as the balance of the user, the total net value of the fund, and the frequency of transactions as its characteristic information.

In one embodiment, the user feature information includes user portrait information. User portrait information is a tagged user model abstracted from the user's basic attribute data. There have been some methods for assigning users a number of tags based on the user's basic attribute data (eg, registration information, consumption records, etc.), and the collection of these tags constitutes user portrait information. For example, in a specific example, the user portrait information includes: "registered new person", "financial white", "saving expert", "medium anti-risk level", "full-time mother" and other tag information, wherein "registered new person" The tag is determined according to the registration time in the user registration information, and the label on the financial management, the savings, and the anti-risk ability is comprehensively determined according to the user's information such as balance, wealth management income, savings level, and consumption record.

In one embodiment, the user's basic attribute data is obtained, and then the user portrait information is determined based on the basic attribute data using the above method. In another embodiment, the client software has rendered a user portrait for the user for other functional purposes. In such a case, such user portrait information is directly acquired as user feature information.

It can be understood that in order to analyze the operation habits of various user groups, it is necessary to obtain a large number of user's respective historical operation information and user feature information. A path navigation model is established for the information of the two aspects, and the model is used for analyzing and learning the historical operation information and the user characteristic information of the user, thereby obtaining the operation habits of various users.

According to an embodiment, a clustering algorithm is adopted in the path navigation model to classify and analyze a large amount of user information, and improve the rapid regression of the model. For example, user feature information may be clustered to divide a large number of users into a plurality of user groups according to a certain granularity, so as to construct an operational behavior model for each user group. It is also possible to perform cluster analysis on user historical operation information, thereby processing complex user operation behaviors into some operational behavior patterns.

In the clustering process, a direct partitioning clustering method such as K-means, a hierarchical clustering algorithm, such as a BIRCH algorithm, a CURE algorithm, etc., a density-based clustering algorithm, such as a DBSCAN algorithm, an OPTICS algorithm, etc., may be adopted. And other possible clustering algorithms.

According to one embodiment, the path navigation model described above may employ algorithms and structures of neural networks, such as convolutional neural network CNN, recurrent neural network RNN, and the like. The path navigation model can adopt the algorithm and structure of the above neural network to perform deep learning on input data, that is, user history operation information and user feature information of a large number of users.

By adopting various algorithms, the path navigation model analyzes and learns the operation habits of each user group based on a large number of user history operation information and user feature information, that is, establishes a correspondence between multiple user groups and operation modes. The process of analyzing and learning may specifically include: counting the frequency of various historical operation paths, operating time, forming a plurality of operation modes according to the historical operation path and the operation time parameter, and counting the correspondence between each user group and the operation mode, thereby User groups build operational behavior models. More specifically, the above-mentioned user group may be a group set according to various granularities, for example, in one example, the division of the user group may correspond to a crowd tag included in the user portrait; in another example, The division of the user group may correspond to a combination of various population tags in the user portrait. For example, the user group 1 corresponds to a user who has both a "financial white" and a "saving expert", and the user group 2 corresponds to a "financial expert" at the same time. ", "Getting started with the stock market" and other users of the label. In yet another example, the division of the user group can be based on independent partitioning rules. The granularity of the user group can be set according to business needs. In one example, a user group can even contain only one user.

In another aspect, the mode of operation can include a historical operational path that exceeds a certain threshold. Further, the operation mode may further include a time-related parameter corresponding to the historical operation path, such as an operation time, a dwell time, and the like.

In a specific example, the user user1 performs the operation path OP1 shown in FIG. 2 frequently (for example, more than one threshold, 10 times) per day: A1.P1-A2.P2-A3.P3-A4.P4-A5.P5 -A6, and another operating path OP2: A1.P1-A2.P2-A3.P7-A8.P8-A9.P9-A10. Such information is recorded as historical operation information. The path navigation model first analyzes the feature information of the user user1, and divides the user user1 into the user group G1. Moreover, the path navigation model can find the high frequency operation paths OP1 and OP2 of the user user1 by analyzing historical operation information, in particular, counting the frequency of occurrence of various operation paths. Such a high frequency operation path can form an operational mode M1. More specifically, the path navigation model can be found by analyzing the operation time of the user user1. The user user1 performs the above-mentioned operation path OP1 5 times on average every 10 o'clock every day, and performs the above operation path OP2 6 times every day around 12 noon. The above operation path OP1 is performed 7 times on average at around 11:00 pm. Based on such information, the path navigation model can establish a correspondence between the user group G1 to which the user user1 belongs and the operation mode M1, and the operation mode M1 includes the above-mentioned high frequency operation paths OP1 and OP2. Further, in one embodiment, the operational mode M1 may also include time-related parameters of the respective operational paths, such as high frequency execution periods (OP1: 10:00, 23:00; OP2: 12:00), dwell time, and the like.

In this way, the path navigation model establishes a correspondence between multiple user groups and operation modes based on a historical operation data set and a user feature data set of a large number of users. The path navigation model constructed in this way can be used for path prediction and navigation of user operations. FIG. 3 illustrates a flow diagram of a method of user operation path navigation, in accordance with one embodiment. As shown in FIG. 3, the method process includes the following steps: Step 31: Acquire an operation sequence and user feature data of a current user; Step 32, adopt a path navigation model, and predict the current user based on the operation sequence and user feature data. The target operation page; step 33, recommending the target operation page to the current user; and step 34, determining the modification of the path navigation model according to whether the current user accepts the recommended target operation page. The specific execution process of each of the above steps will be described below.

First, in step 31, the current user's operation sequence and user feature data are acquired. In order to distinguish from the description of the foregoing modeling stage, the following example describes the user as the current user and the user A as the current user. However, it can be understood that the current user A may be a historical user included in the historical data set on which the path navigation model modeling stage is based, or may be a new user not included in the historical data set.

Similar to the acquisition of historical operation information, the operation sequence SA of the user A can be captured by means of page burying or the like. The sequence of operations SA may contain only one operation or multiple consecutive operations. These operations may include clicking, sliding, dragging, character typing, zooming in or out, etc. of the interactive modules in the page. In a specific example, it is captured in step 31 that the current user A starts from the Alipay main page A1, clicks "ant wealth" (performs the operation P1), enters the page A2, and then on the A2 page, clicks "finance" (performs the operation) P2), enter the financial page A3. The sequence of operations SA is thus formed: A1.P1-A2.P2-A3. In another example, user A continues to operate and the sequence of operations SA is updated to: A1.P1-A2.P2-A3.P3-A4.

On the other hand, the user feature data of the current user A is also acquired.

In one embodiment, the attribute data of the current user A is acquired as user feature data, and the attribute data may include data in the registration information, such as age, gender, registration time, and the like, and other attribute data related to the main function of the client. For example, the balance, income and other data in the wealth management client.

In another embodiment, the acquired attribute data of the user A is converted into user portrait data as user feature data. Specifically, first, the attribute values (such as age, registration time, balance, income, and the like) of the attributes of the user A are obtained, and the crowd label to which the user A belongs is determined according to the attribute values (for example, “registered new person”, “financial white” , "saving expert", etc.), determining the portrait data of the user A as the user characteristic data based on the crowd tag.

In yet another embodiment, the client software has rendered a user portrait for each user for other functional purposes. In such a case, the portrait data of the current user A is directly acquired as its user characteristic information.

Next, in step 32, the path navigation model is used to predict the current user's target operation page based on the captured user A's operation sequence and user feature data. As mentioned above, a path navigation model has been constructed, which establishes a correspondence between multiple user groups and operation modes based on historical operation data sets and user feature data sets of multiple users. In this way, the path navigation model can be adopted to predict the target operation page of the current user A based on the established correspondence.

4 shows a flow diagram of a predicted target operational page, that is, a sub-step of step 32, in accordance with one embodiment. As shown in FIG. 4, in step 321, a specific user group corresponding to the current user is determined according to user characteristic data of the current user.

Specifically, for the current user A, if the historical data set on which the path navigation model modeling stage is based already includes the user (for example, the user A is actually equal to the user user1 described above), then the user group can be directly divided according to the modeling stage. The user group to which the user A belongs, that is, the user group G1 is determined.

If the user A does not include the user A, the user group to which the user A belongs may be determined based on the user profile data of the user A and the user group division rule of the path navigation model modeling stage. For example, the division of the user group in the modeling phase may correspond to the crowd tag included in the user portrait, or may correspond to a combination of various crowd tags in the user portrait. In this case, the user profile data of the user A is first obtained based on the user profile data of the user A, thereby obtaining the crowd tag of the user A, and the user group to which the user A belongs is determined according to the crowd tag and the user group division rule. In other examples, the modeling phase can also be based on other user group partitioning rules. It is assumed here that the user A belongs to the user group G1 according to the user characteristic data of the user A and the user group division rule.

On the basis of determining the specific user group to which the user A belongs, in step 322, the specific operation mode corresponding to the specific user group is determined according to the correspondence between the plurality of user groups established by the path navigation model and the operation mode.

As mentioned above, the path navigation model establishes the correspondence between multiple user groups and operation modes by analyzing and learning the historical operation data sets and user feature data sets of a large number of users. On the basis of the determination of the user group G1 to which the user A belongs, the operation mode M1 corresponding to the user group G1 can be determined by the established correspondence.

Next, at step 323, a particular operational path that matches the current user's sequence of operations is determined based on the particular mode of operation described above. In general, the mode of operation may include a historical operational path that exceeds a certain threshold frequency. Further, the operation mode may further include a time-related parameter corresponding to the historical operation path, such as an operation time, a dwell time, and the like. In one embodiment, in step 323, a historical operation path included in the specific operation mode is obtained, from which an operation path matching the current user's operation sequence is determined as the specific operation path. The above operation path matching the current user's operation sequence may be an operation path including the operation sequence. In one embodiment, time dependent parameters for each of the historical operational paths included in the operational mode may also be further considered. For example, in a case where a plurality of historical operation paths in a specific operation mode include a current operation sequence, the current operation time of the current operation sequence and the historical execution time of each historical operation path recorded in the operation mode may be determined, and the history thereof The historical operation path whose execution time is closest to the current operation time is determined as a specific operation path that matches the current operation sequence.

For the user group G1 in the foregoing example, the corresponding operation mode M1 includes at least the historical high frequency operation path OP1: A1.P1-A2.P2-A3.P3-A4.P4-A5.P5-A6, and OP2 : A1.P1-A2.P2-A3.P7-A8.P8-A9.P9-A10.

In one example, assume that user A's current sequence of operations SA includes, A1.P1-A2.P2-A3.P3-A4. By comparing the elements of the current operation sequence with the historical operation path, it can be seen that the historical operation path OP1 contains the current operation sequence SA, so OP1 can be used as a specific operation path matching the SA.

In another example, assume that user A's current sequence of operations SA includes A1.P1-A2.P2-A3. It can be seen that the historical operation paths OP1 and OP2 included in the operation mode M1 each contain the current operation sequence SA, so both OP1 and OP2 can be used as alternative operation paths. Further, reference may be made to time dependent parameters in the operational mode M1, such as high frequency execution periods. Assuming that the operation time of the current operation sequence SA is 10:05, the history execution period recorded by the operation mode M1 includes: OP1: 10:00 and 23:00, and OP2: 12:00. The current operating time 10:05 is closer to the high frequency execution time of the historical operating path OP1, and therefore, OP1 is determined as a specific operating path that matches the current operating sequence SA.

Next, in step 324, the target operation page of the current user is determined according to the specific operation path determined above. Specifically, the end page in the specific operation path is used as the target operation page of the current user. In the foregoing example, assuming that OP1 is determined to be a specific operation path, the end page A6 of OP1 (the page corresponding to "valuation ranking" in FIG. 2) can be used as the target operation page of user A.

Thus, in the manner shown in FIG. 4, in step 32, the path navigation model is used to predict the target operation page of the current user A. Returning to Figure 3, next, at step 33, the determined target action page is recommended to the current user.

In one embodiment, the determined target action page is provided to the current user as the next action page. For example, assuming that the specific operation path has been determined to be OP1, the end page A6 of OP1 is the target operation page of User A, and the current operation page of User A is A3. If user A performs an operation on page A3, for example, clicking "Fund" on page A3 (performing operation P3) to continue to conform to the specific operation path OP1, then skip the intermediate pages A4 and A5 and directly display the target operation page A6 for user A. .

In one embodiment, in order to avoid interfering with the user's operation, an icon is added to the client interface, and the new icon points to the target operation page. Figure 5 shows a schematic diagram of a new icon in accordance with one embodiment. As shown in FIG. 5, an icon "Quick" is added to the operation icon of the bottom row of the "Alipays" client, and the icon points to the target operation page. Assuming that the page A6 is determined as the target operation page of the user A, in the case of FIG. 5, during the current operation sequence, the user A can directly jump to the page A6 by clicking the icon "fast", and the middle page is omitted. , such as A4 and A5.

In another embodiment, in order to further reduce the modification of the client interface, a floating layer icon is provided on the original interface, and the floating layer icon points to the target operation page. Figure 6 shows a schematic diagram of a floating layer icon in accordance with one embodiment. As shown in Figure 6, on the basis of the original interface of "Alipay", the floating layer icon "Know what you want" is added, and the icon points to the target operation page. In one example, the floating layer icon can be designed to be translucent to minimize the display of the original interface. In the foregoing example, assuming that the page A6 is determined to be the target operation page of the user A, then in the case of FIG. 7, the user A can click on the floating layer icon "understand" what you want under any page of the current operation sequence. ", jump directly to page A6, eliminating the middle page.

It can be understood that the determined target operation page can also be recommended to the user by other means, such as pushing.

By recommending the target operation page to the user, the user can help the target operation to be performed quickly and efficiently, reach the target page, and improve the user experience.

At the same time, according to the user's acceptance of the recommended target page as feedback, the path navigation model is further trained and modified to make it more in line with the user's usage habits. That is, at step 34, a modification to the path navigation model is determined based on whether the current user accepts the recommended target operation page.

In one embodiment, it is first determined whether the current user has accepted the recommended target action page. In the foregoing example, the target operation page A6 is recommended to the user A in various ways. If user A enters the recommended target operation page A6, an exit operation is performed within a short predetermined time (for example, 1 s), for example, by clicking other icons to jump out of the page, or by operating the return key to exit the page, or to close the entire Client software, then you can think that User A has not selected or accepted the recommended target operation page. If the user A stays in the recommended target operation page for a reasonable time, or performs browsing, further interaction, etc., the user A may be considered to have selected or accepted the recommended target operation page.

The recommendation accuracy of the path navigation model can be determined according to whether the current user accepts the recommended target operation page. The recommendation accuracy can be statistically based on the acceptance feedback of a plurality of current users on the recommended target operation page. For example, after recommending the target operation page to 100 users, assuming that 40 of them accepted the recommendation, the recommendation accuracy can be considered to be 40%. For each current user, the recommendation accuracy may be accumulated according to the acceptance of the recommendation, thereby dynamically updating the recommendation accuracy. For example, suppose the current user is user A, and the recommendation of the target operation page has been previously performed to 20 users, among which 9 users have accepted the recommendation, and the recommendation accuracy is 9/20. If the current user A accepts the recommended target operation page, the recommendation accuracy may be updated to 10/21; if the user A does not accept the recommended target operation page, the recommendation accuracy may be updated to 9/21.

In the case where the recommendation accuracy is lower than a preset threshold, the path navigation model may be modified. More specifically, the subsequent operation of the current user, that is, the operation after exiting from the recommended target operation page, can be recorded. It can be understood that such subsequent operations are operations that reflect the actual target of the current user. And forming a target operation path according to the previous operation sequence of the current user and the subsequent operation, and then updating the correspondence between the plurality of user groups established by the path navigation model and the operation mode according to the target operation path. More specifically, in one embodiment, the target operation path is added to the historical operation data set to form an updated historical operation data set, and the path navigation model is re-analyzed for the updated historical operation data set, thereby re-establishing The correspondence between the user group and the operation mode. In another embodiment, an operation mode (eg, operation mode M1) of a user group (eg, user group G1) to which the current user (eg, user A) belongs is updated according to the target operation path described above. Specifically, the target operation path is used as a newly added historical operation path, and the statistics of the historical operation path are updated for the user group to which the current user belongs, thereby updating the content of the operation mode corresponding to the user group.

It can be understood that some conventional algorithms in the model training process, such as gradient transfer, can be used to modify other parameters of the model, so that the model can be better regressed, and the recommendation accuracy is positively improved according to user feedback.

In this way, the closed loop forward feedback is implemented by the path navigation model. While recommending the target operation page to the user, the model is continuously trained and improved, so that the recommendation accuracy of the model is continuously improved, thereby helping the user to reach the target page to a greater extent, simplifying user operations and improving the user experience.

According to another embodiment, an apparatus for user-operated path navigation is also provided. Figure 7 shows a schematic block diagram of an operational path navigation device in accordance with one embodiment. As shown in FIG. 7, the navigation apparatus 700 includes: an obtaining unit 71 configured to acquire an operation sequence and user feature data of a current user; and a prediction unit 72 configured to adopt a path navigation model, based on the operation sequence and user feature data, Predicting a target operation page of the current user; a recommending unit 73 configured to recommend the target operation page to the current user; and a determining unit 74 configured to determine, according to whether the current user accepts the recommended target operation page The modification of the path navigation model.

In an embodiment, the obtaining unit 71 is configured to acquire an attribute value of the at least one attribute of the current user, and determine, according to the attribute value of the at least one attribute, at least one crowd label to which the current user belongs, according to the At least one crowd tag determines the current user's portrait data as the user feature data.

According to an embodiment, the path navigation model is configured to establish a correspondence between a plurality of user groups and an operation mode based on a historical operation data set and a user feature data set of the plurality of users.

In one embodiment, the foregoing prediction unit 72 includes (not shown): a first determining module configured to determine a specific user group corresponding to the current user according to the user feature data; and a second determining module configured to be configured according to Corresponding relationship between the plurality of user groups and the operation mode, determining a specific operation mode corresponding to the specific user group; and the third determining module is configured to determine, according to the specific operation mode, a sequence that matches the operation sequence of the current user a specific operation path; and a fourth determining module configured to determine a target operation page of the current user according to the specific operation path.

Further, in an embodiment, the third determining module is configured to: determine, from the specific operation mode, a plurality of historical operation paths including the operation sequence; determine a current operation time of the operation sequence, and Determining a historical execution time of each of the plurality of historical operation paths; determining a historical operation path whose history execution time is closest to the current operation time as the specific operation path.

According to an embodiment, the recommendation unit 73 is configured to provide the target operation page by providing a floating layer icon on the original interface, the floating layer icon pointing to the target operation page, and new in the interface. An icon is added, the added icon points to the target operation page; and the target operation page is provided to the current user as a next operation page.

In one embodiment, the determining unit 74 includes (not shown): an accuracy determining module configured to determine a recommended accuracy of the path navigation model according to whether the current user accepts the recommended target operation page; The method is configured to modify the path navigation model if the recommendation accuracy is lower than a preset threshold.

Further, in an embodiment, the modifying module is configured to: record a subsequent operation of the current user; form a target operation path according to the operation sequence and the subsequent operation; and update the location according to the target operation path Corresponding relationship between multiple user groups and operation modes established by the path navigation model. Further, in an embodiment, the modifying module is further configured to: use the target operating path as a new historical operating path, and update statistics of the historical operating path for the specific user group to which the current user belongs; Historical operation path statistics, updating the operation mode corresponding to the specific user group.

Through the above device, the path navigation model is used to recommend the target operation page to the user, and the model is continuously trained and improved according to the user feedback, so that the recommendation accuracy of the model is continuously improved, thereby helping the user to reach the target page to a greater extent and simplifying the user. Operation to enhance the user experience.

According to another embodiment, there is also provided a computer readable storage medium having stored thereon a computer program that, when executed in a computer, causes the computer to perform the method described in connection with FIG.

According to still another embodiment, there is also provided a computing device comprising a memory and a processor, the memory storing executable code, and when the processor executes the executable code, implementing the method described in connection with FIG. method.

Those skilled in the art will appreciate that in one or more examples described above, the functions described herein can be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, the functions may be stored in a computer readable medium or transmitted as one or more instructions or code on a computer readable medium.

The specific embodiments of the present invention have been described in detail with reference to the preferred embodiments of the present invention. The scope of the protection, any modifications, equivalent substitutions, improvements, etc., which are made on the basis of the technical solutions of the present invention, are included in the scope of the present invention.

Claims

A method for a user to navigate a path includes:

Obtaining the current user's operation sequence and user feature data;

Using a path navigation model, predicting a target operation page of the current user based on the operation sequence and user feature data;

Recommending the target operation page to the current user;

The modification of the path navigation model is determined according to whether the current user accepts the recommended target operation page.
The method of claim 1, wherein the obtaining the user feature data comprises: acquiring an attribute value of the at least one attribute of the current user, and determining, according to the attribute value of the at least one attribute, at least one group to which the current user belongs And determining, by the tag, the portrait data of the current user as the user feature data according to the at least one crowd tag.
The method of claim 1, wherein the path navigation model is configured to establish a correspondence between a plurality of user groups and an operation mode based on a historical operation data set and a user feature data set of the plurality of users.
The method of claim 3, wherein said predicting a target operation page of said current user comprises:

Determining, according to the user feature data, a specific user group corresponding to the current user;

Determining, according to a correspondence between the multiple user groups and an operation mode, a specific operation mode corresponding to the specific user group;

Determining, according to the specific operation mode, a specific operation path that matches an operation sequence of the current user;

Determining a target operation page of the current user according to the specific operation path.
The method of claim 4, wherein determining a particular operational path that matches the current user's sequence of operations based on the particular mode of operation comprises:

Determining, from the particular mode of operation, a plurality of historical operational paths including the sequence of operations;

Determining a current operating time of the sequence of operations, and a historical execution time of each of the plurality of historical operating paths;

A historical operation path whose history execution time is closest to the current operation time is determined as the specific operation path.
The method of claim 1, wherein recommending the target operation page to the current user comprises providing the target operation page by one of:

Providing a floating layer icon on the original interface, the floating layer icon pointing to the target operation page;

Add an icon to the interface, and the added icon points to the target operation page;

The target operation page is provided to the current user as a next operation page.
The method of claim 3, wherein determining the modification of the path navigation model according to whether the current user accepts the recommended target operation page comprises:

Determining the recommendation accuracy of the path navigation model according to whether the current user accepts the recommended target operation page;

The path navigation model is modified if the recommendation accuracy is lower than a preset threshold.
The method of claim 7 wherein modifying the path navigation model comprises:

Recording subsequent operations of the current user;

Forming a target operation path according to the operation sequence and the subsequent operation;

And updating, according to the target operation path, a correspondence between a plurality of user groups established by the path navigation model and an operation mode.
The method according to claim 8, wherein the updating the correspondence between the plurality of user groups established by the path navigation model and the operation mode comprises:

Using the target operation path as a new historical operation path, updating statistics of the historical operation path for the specific user group to which the current user belongs;

Updating the operation mode corresponding to the specific user group according to the updated historical operation path statistics.
A device for user operation path navigation, comprising:

An obtaining unit configured to acquire an operation sequence and user feature data of a current user;

a prediction unit configured to adopt a path navigation model, and predict a target operation page of the current user based on the operation sequence and user feature data;

a recommendation unit, configured to recommend the target operation page to the current user;

The determining unit is configured to determine a modification to the path navigation model according to whether the current user accepts the recommended target operation page.
The apparatus according to claim 10, wherein the obtaining unit is configured to acquire an attribute value of at least one attribute of the current user, and determine at least one crowd label to which the current user belongs according to an attribute value of the at least one attribute Determining the current user's portrait data as the user feature data according to the at least one crowd tag.
The apparatus of claim 10, wherein the path navigation model is configured to establish a correspondence between a plurality of user groups and an operation mode based on a historical operation data set and a user feature data set of the plurality of users.
The apparatus of claim 12 wherein said predicting unit comprises:

a first determining module, configured to determine, according to the user feature data, a specific user group corresponding to the current user;

a second determining module, configured to determine a specific operation mode corresponding to the specific user group according to a correspondence between the multiple user groups and an operation mode;

a third determining module, configured to determine a specific operation path that matches an operation sequence of the current user according to the specific operation mode;

The fourth determining module is configured to determine a target operation page of the current user according to the specific operation path.
The apparatus of claim 13 wherein said third determining module is configured to:

Determining, from the particular mode of operation, a plurality of historical operational paths including the sequence of operations;

Determining a current operating time of the sequence of operations, and a historical execution time of each of the plurality of historical operating paths;

A historical operation path whose history execution time is closest to the current operation time is determined as the specific operation path.
The apparatus of claim 10, wherein the recommendation unit is configured to provide the target operation page in one of the following ways:

Providing a floating layer icon on the original interface, the floating layer icon pointing to the target operation page;

Add an icon to the interface, and the added icon points to the target operation page;

The target operation page is provided to the current user as a next operation page.
The apparatus of claim 12 wherein said determining unit comprises:

The accuracy determining module is configured to determine a recommendation accuracy of the path navigation model according to whether the current user accepts the recommended target operation page;

And modifying the module, configured to modify the path navigation model if the recommendation accuracy is lower than a preset threshold.
The apparatus of claim 16 wherein said modifying module is configured to:

Recording subsequent operations of the current user;

Forming a target operation path according to the operation sequence and the subsequent operation;

And updating, according to the target operation path, a correspondence between a plurality of user groups established by the path navigation model and an operation mode.
The apparatus of claim 17 wherein said modifying module is further configured to:

Using the target operation path as a new historical operation path, updating statistics of the historical operation path for the specific user group to which the current user belongs;

Updating the operation mode corresponding to the specific user group according to the updated historical operation path statistics.
A computer readable storage medium having stored thereon a computer program for causing a computer to perform the method of any one of claims 1-9 when the computer program is executed in a computer.
A computing device, comprising a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, implementing the method of any one of claims 1-9 method.