Method, device and electronic device for classifying operation paths

Abstract

The present application discloses a method, device, and electronic device for classifying operation paths. The method comprises: obtaining an operation path set; wherein the operation path set includes multiple operation paths, the operation path is generated based on operation data, the operation data records multiple operations involved in the process of accessing a specified page and the execution order between the multiple operations, and the path nodes in the operation path indicate the operation or the page object on which the operation acts; classifying the operation paths in the operation path set that meet preset matching requirements to obtain a classification result for guiding operation analysis. Classifying the operation path set involved in the same page can provide accurate and effective reference data for data processing of the operation path.

Description

Technical Field

[0001] The present application relates to the field of Internet communication technology, and in particular to a method, device, and electronic device for classifying operation paths. Background Art

[0002] With the development of Internet communication technology, various Internet products have emerged in an endless stream, and Internet products can provide service experience for users. In related technologies, for the operations of users on related pages, the number of operations in the page position dimension (such as the coordinate dimension) is often counted, so as to obtain statistical results displayed in the form of a heat map. By reading the heat map, the distribution of operations of users on related pages can be obtained. However, the statistics of the number of operations in the page position dimension are relatively rough and cannot provide accurate and effective reference data for operation analysis. Summary of the Invention

[0003] To address the problems in existing technologies where, when providing reference data for operation analysis, the statistical results (corresponding to the number of operations in the page position dimension) used as reference data have coarse evaluation granularity and cannot guarantee the accuracy of subsequent operation analysis using the reference data, the present application provides an operation path classification method, device, and electronic device:

[0004] According to a first aspect of the present application, a method for classifying operation paths is provided, the method comprising:

[0005] Obtaining an operation path set; wherein the operation path set includes multiple operation paths, the operation paths are generated based on operation data, the operation data records multiple operations involved in a process of accessing a specified page and the execution order of the multiple operations, and the path nodes in the operation paths indicate the operations or the page objects acted upon by the operations;

[0006] The operation paths in the operation path set that meet preset matching requirements are classified to obtain classification results for guiding operation analysis; wherein the preset matching requirements include node attribute matching requirements of relevant path nodes and position matching requirements of relevant path nodes.

[0007] According to a second aspect of the present application, a device for classifying operation paths is provided, the device comprising:

[0008] A set acquisition module is configured to acquire an operation path set, wherein the operation path set includes multiple operation paths, each of which is generated based on operation data. The operation data records multiple operations involved in accessing a specified page and the execution order of the multiple operations. Path nodes in the operation paths indicate the operations or the page objects on which the operations are applied.

[0009] Classification module: used to classify the operation paths in the operation path set that meet the preset matching requirements to obtain classification results for guiding operation analysis; wherein, the preset matching requirements include node attribute matching requirements of the relevant path nodes and position matching requirements of the relevant path nodes.

[0010] According to a third aspect of the present application, an electronic device is provided, comprising a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the operation path classification method as described in the first aspect.

[0011] According to the fourth aspect of the present application, a computer-readable storage medium is provided, in which at least one instruction or at least one program is stored. The at least one instruction or the at least one program is loaded and executed by a processor to implement the operation path classification method as described in the first aspect.

[0012] According to a fifth aspect of the present application, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method for classifying operation paths as described in the first aspect.

[0013] The present application provides a method, device, and electronic device for classifying operation paths, which have the following technical effects:

[0014] The present application obtains an operation path set, and then classifies the operation paths in the operation path set that meet the preset matching requirements to obtain a classification result for guiding operation analysis. The present application classifies the operation path set involved in the same page. The classification result can reflect the different types of operation paths mined from the operation path set. The classification result can be used as accurate and effective reference data for restoring a reference access operation process. On this basis, the convenience and efficiency of using the operation path to mine operation rules and capture abnormal operations can be improved, thereby achieving the effect of obtaining deep and hidden information through data processing based on the operation path, and improving the operation analysis effect. Compared with the reference data obtained by counting the number of operations in the page position dimension in the related art, the classification result for guiding operation analysis obtained in the present application is derived from the classification processing of the operation path set. The operation path characterizes the multiple operations involved in the process of accessing a specified page and the execution order between the multiple operations. Therefore, the evaluation granularity information carried by the reference data is richer and can better help improve the accuracy of the operation analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0016] Figure 1 is a schematic diagram of an application environment provided by an embodiment of the present application;

[0017] Figure 2 This is a flow chart of a method for classifying operation paths provided in an embodiment of the present application;

[0018] Figure 3 This is a schematic diagram of a process for matching local node attributes of two path nodes to be matched, provided by an embodiment of the present application;

[0019] Figure 4 This is a flow chart of classifying operation paths that meet preset matching requirements in an operation path set provided by an embodiment of the present application;

[0020] Figure 5 This is a schematic diagram of the operating points provided in the embodiment of the present application;

[0021] Figure 6 This is a schematic diagram of the operation path provided by the embodiment of the present application;

[0022] Figure 7 This is a schematic diagram of performing similar matching of operating points provided in an embodiment of the present application;

[0023] Figure 8 This is also a schematic diagram of the operation path provided by the embodiment of the present application;

[0024] Figure 9 This is also a schematic diagram of the operation path provided by the embodiment of the present application;

[0025] Figure 10 is a schematic diagram of outliers provided in an embodiment of the present application;

[0026] Figure 11 This is a schematic diagram of city selection provided in an embodiment of the present application;

[0027] Figure 12 This is a schematic diagram of the experimental version selection provided in the embodiment of the present application;

[0028] Figure 13 This is also a schematic diagram of city selection provided by an embodiment of the present application;

[0029] Figure 14 This is also a schematic diagram of city selection provided by an embodiment of the present application;

[0030] Figure 15 This is a schematic diagram of a form provided in an embodiment of the present application;

[0031] Figure 16 is a schematic diagram of an operation path before improvement provided in an embodiment of the present application;

[0032] Figure 17 is a schematic diagram of an improved operation path provided in an embodiment of the present application;

[0033] Figure 18 This is a block diagram of a device for classifying operation paths provided in an embodiment of the present application;

[0034] Figure 19 This is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0035] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0036] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products or devices.

[0037] See also Figure 1 , Figure 1 This is a schematic diagram of an application environment provided by an embodiment of the present application, which may include a client 10 and a server 20. The client 10 and the server 20 may be connected directly or indirectly via wired or wireless communication. The user may send an operation path classification instruction to the server 20 via the client 10. The server 20 responds to the received operation path classification instruction by obtaining an operation path set; then, the operation paths in the operation path set that meet the preset matching requirements are classified to obtain a classification result for guiding operation analysis. It should be noted that Figure 1 Just an example.

[0038] The client 10 may be a physical device such as a smartphone, a computer (e.g., a desktop computer, tablet computer, laptop computer), an augmented reality (AR) / virtual reality (VR) device, a digital assistant, an intelligent voice interaction device (e.g., a smart speaker), a smart wearable device, a smart home appliance, or an in-vehicle terminal. It may also be software running on a physical device, such as a computer program. The operating system supported by the client may be Android, iOS (a mobile operating system developed by Apple), Linux, or Microsoft Windows.

[0039] The server side 20 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. The server can include a network communication unit, a processor, a memory, etc. The server side can provide background services for the corresponding client.

[0040] In actual applications, each operation path in the operation path set is generated based on the corresponding operation data, and the operation data records the multiple operations involved in the process of accessing the specified page and the execution order of the multiple operations. The specified page can be provided by a specified Internet product, and the specified Internet product can be a cloud technology product, an artificial intelligence product, a smart transportation product, an assisted driving product, a live broadcast product, an online office product, an e-commerce product, a game product, a local life product, an instant messaging product, a social product, etc. The services provided by the specified Internet product can be supported by relevant clients and relevant server ends. The relevant client server end can be the same as or different from the client end 10 that sends the operation path classification instruction; the relevant server end can be the same as or different from the server end 20 that receives the operation path classification instruction.

[0041] Operational data can be generated by a related object (such as a user, simulator, etc.) accessing a specified page, such as generated by accessing a specified page through the above-mentioned related client. It should be noted that for operational data involving user information, etc. in the embodiments of this application, when the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0042] The following describes a specific embodiment of a method for classifying operation paths of the present application. Figure 2 It is a flow chart of a method for classifying an operation path provided in an embodiment of the present application. The present application provides method operation steps as described in the embodiment or flow chart, but may include more or fewer operation steps based on conventional or non-creative labor. The order of steps listed in the embodiment is only one way of executing the steps among many, and does not represent the only execution order. When the actual system or product is executed, it can be executed in sequence or in parallel (for example, in a parallel processor or multi-threaded processing environment) according to the method shown in the embodiment or the accompanying drawings. Specifically, Figure 2 As shown, the method may include:

[0043] S201: Obtain an operation path set; wherein the operation path set includes multiple operation paths, each of which is generated based on operation data, wherein the operation data records multiple operations involved in accessing a specified page and the execution order of the multiple operations, and a path node in the operation path indicates the operation or the page object on which the operation acts;

[0044] In an embodiment of the present application, the server side obtains an operation path set. The operation path set can be composed of multiple operation paths for a specified page. Each operation path in the operation path set is generated based on corresponding operation data. The operation data records the multiple operations involved in the process of accessing the specified page and the execution order between the multiple operations. The operation data can be generated by the relevant objects (such as users, simulators, etc.) accessing the specified page. The process of the relevant objects accessing the specified page can start with the start of the display of the specified page on the user interaction interface (corresponding to the relevant objects) and end with the end of the display of the specified page on the user interaction interface. In addition, it can also end with the display screen of the specified page on the user interaction interface remaining unchanged within a preset time period. The specified page can be a single page, or it can be composed of a main page and at least one secondary page. The specified page can be limited by the access address. In addition, the specified page can also be at least two pages with an associated relationship. For example, page 1 indicates version 2.0, page 2 indicates version 4.0, and page 1 and page 2 are both specified pages. The operation path set may include one type of operation path generated based on the operation data for page 1 , and another type of operation path generated based on the operation data for page 2 .

[0045] The operations involved in the access process can indicate touchscreen, mouse, keyboard, and other actions performed by the relevant objects on a specified page. For example, a user interacts with a web page (including mobile devices) by clicking, double-clicking, moving the mouse, or entering content via the keyboard. Accordingly, based on the execution order of the relevant actions, a corresponding operation path can be derived. For example, multiple user operations on a web page can be linked together in chronological order to form an operation path. If a user first performs click action A on page object a, then click action B on page object b, then click action F on page object f, and finally click action D on page object d, when path nodes are used to indicate the operation, the operation path can be formed: click action A -> click action B -> click action F -> click action D. When path nodes are used to select the page object to be operated on, the operation path can be formed: page object a -> page object b -> page object f -> page object d. Page objects can be buttons, links, images, text, input boxes, div areas, etc. The page object affected by the operation can be determined by the coordinates at the time of the operation. For example, although the coordinates (891px, 541px) and (890px, 540px) are slightly different, they actually fall on the same button, so they are operations on the same page object. Determining the page object on which the operation is performed through the user operation layer can provide support for clearly defining the operation path and operation purpose. Figure 7-9 .

[0046] Exemplarily, an operation path is generated based on a corresponding operation data, then the data source of the operation path set including multiple operation paths is multiple operation data. Multiple operation data may be generated by different objects accessing the specified page respectively. For example, multiple operation data include operation data 1-5, operation data 1-2 are generated by user a visiting the specified page twice, operation data 3-4 are generated by user b visiting the specified page twice, and operation data 5 is generated by user c visiting the specified page. Of course, the operation path set obtained here can indicate the same preset time period, for example, the relevant operation data is generated within the same preset time period. The operation path set obtained here can indicate a user group using the same model of device, for example, the relevant operation data is generated by a user group using device a accessing the specified page.

[0047] The following describes the process of generating an operation path based on operation data:

[0048] 1) See Figure 5As shown in the figure, the points in the figure can be regarded as the user's operation points on the page. When the user clicks, inputs, and other operations on the page, the user's relevant operation information can be recorded. The recorded operation information can include time information, as shown in the following code:

[0049]

[0050] 2) See Figure 6 As shown, based on the time information corresponding to each operation, the corresponding operation path can be generated. When the user clicks, enters, and performs other operations on the page, the operation information involved in the user accessing the page can be recorded. The recorded operation information can include basic page information (such as page URL, operator, etc.), as shown in the following code:

[0051]

[0052]

[0053] An operation path can be formed by recording the operation information involved in the user accessing the page.

[0054] S202: Classify the operation paths in the operation path set that meet preset matching requirements to obtain classification results for guiding operation analysis; wherein the preset matching requirements include node attribute matching requirements of relevant path nodes and position matching requirements of relevant path nodes.

[0055] In an embodiment of the present application, the server side classifies the operation paths in the operation path set that meet the preset matching requirements to obtain a classification result for guiding operation analysis. The operation paths that meet the preset matching requirements, that is, two operation paths that can be classified into the same category, can be considered to be the same or similar operation paths. If operation path 1 and operation path 2 can be classified into the same category, then operation path 1 and operation path 2 simultaneously meet the node attribute matching requirements of the relevant path nodes and the position matching requirements of the relevant path nodes. In other words, the path nodes included in operation path 1 and the path nodes included in operation path 2 not only match the node attributes but also match the positions. For example, operation path 1: A->B->C->D, operation path 2: A->C->B->D, then it can be considered that their node attributes match but the positions do not match. If operation path 2: A->B->C->D, then it is considered that their node attributes match and the positions match. It should be noted that at this time, the difference between the time information indicated by the path nodes at the same position can be ignored.

[0056] The following will introduce node attribute matching, specifically the local node attribute matching of two path nodes to be matched.

[0057] 1) If Figure 3 As shown, when the path node in the operation path indicates the page object on which the operation is performed, the method further includes:

[0058] S301: For two path nodes to be matched, respectively determine corresponding page objects to obtain two page objects to be matched;

[0059] S302: When the two page objects to be matched belong to the same upper-level page object, it is determined whether the two path nodes to be matched meet the node attribute matching requirement of the relevant path nodes.

[0060] This article provides an implementation for local node attribute matching based on the page object dimension. Compared to matching based on identical page objects, matching based on belonging to the same parent page object can better consider page layout habits during matching, improve the capture of user operation habits during matching, overcome the physical limitations of page objects during matching, and achieve more accurate and adaptable local node attribute matching.

[0061] For example, if the path node in the operation path selects the page object on which the operation is performed, when the page object is the same as the matching condition, it is often necessary that the two page objects to be matched are both buttons A, so that the two path nodes to be matched can be determined to meet the node attribute matching requirements of the relevant path nodes. If belonging to the same parent page object is used as the matching condition, if page object 1 to be matched is button A and page object 2 to be matched is button B, and button A and button B belong to the same parent page object (such as picture A), the two path nodes to be matched can also be determined to meet the node attribute matching requirements of the relevant path nodes.

[0062] Page layout can guide related user operations, and user operations can also guide page layout. Generally speaking, the subordinate page objects of the same level that belong to the same parent page object have certain commonalities. The subordinate page objects of the same level can serve the presentation of the same user intent. For example, the subordinate page objects of the same level are candidates for the same user intent. Figure 11 , the figure shows multiple checkboxes, which can be candidates for the user intent of "select a city". No matter which one or more of them are clicked, it can be regarded as a specific action that reflects the user intent of "select a city". Figure 12 ,The figure shows the candidates for the user intent of “select experimental version”.

[0063] Combined with the above examples of button A and button B, button A and button B belonging to the same parent page object can serve the presentation of the same user intention, thereby determining that the two path nodes to be matched meet the node attribute matching requirements of the relevant path nodes, and no longer considering that the two path nodes to be matched do not meet the node attribute matching requirements of the relevant path nodes. This can avoid the lack of globality in the classification results due to excessive focus on physical details, which affects the effect of operation analysis.

[0064] Furthermore, after the above-mentioned determination that the two path nodes to be matched meet the node attribute matching requirements of the relevant path nodes, the method also includes: first, when the upper-level page object is an attribute class object (value), determining a candidate page object at the same level as the upper-level page object; then, when the candidate page object is a label class object (label), and the element label matches the element attribute, determining the candidate page object as the target page object; finally, using the target page object to generate a classification label corresponding to the two path nodes to be matched.

[0065] At least two of the subordinate page objects of the attribute class object have the same attribute. In other words, the attribute class object can be used to define the attributes of its subordinate page objects. Attribute class objects and label class objects can appear in pairs. See Figure 15 This is common in forms, which can include multiple form items, such as an "input box"-based form item, a "selector"-based form item, a "switch"-based form item, a "radio box"-based form item, a "multi-select box"-based form item, etc. It should be noted that due to page display needs, the display transparency of attribute objects can be adjusted to achieve transparent display of attribute objects.

[0066] See also Figure 13 For example, consider "City A," "City B," and "City C" as peer page objects. Their parent page object is "value," and the candidate page object at the same level as "value" is "label." Furthermore, "value" and "label" serve the same user intent and are therefore a good match. Therefore, "label" can be used as the target page object to generate classification labels. For example, "Select a city" corresponding to "label" can be used as the classification label.

[0067] You can use div technology to define the subordinate page objects of the attribute class object. If the attribute class object "value" is displayed transparently in the relevant page, you can refer to Figure 12In the figure, "Wildcard Condition A" and "Wildcard Condition B" are two subordinate page objects of the attribute class object. The candidate page object at the same level as "value" is "label." Furthermore, "value" and "label" serve the same user intent and are therefore a match. Therefore, "label" can be used as a target page object to participate in the generation of classification labels, for example, using "experimental version" corresponding to "label" as the classification label.

[0068] In line with page layout conventions, element tags can have element attributes, which can provide a wealth of information about page elements. Using relevant element tags to generate classification tags improves the efficiency and convenience of generating classification tags. Of course, when there are no paired attribute objects and tag objects, classification tags can also be generated based on the characteristic information of the parent page object (such as attributes, identifiers, etc.).

[0069] 2) See Figure 14 , an embodiment of the present application provides an implementation method for matching local node attributes from the page object dimension / operation dimension based on a distance threshold.

[0070] A) When the path node in the operation path indicates the page object on which the operation acts, the page positions of the corresponding page objects are determined for the two path nodes to be matched respectively to obtain two page positions to be matched; when the distance between the two page positions to be matched is less than the first preset distance threshold, it is determined that the two path nodes to be matched meet the node attribute matching requirements of the relevant path nodes. The page objects shown in the figure are "City A", "City B", "City C" and "Delete". The page position of the page object can be the center of the page object or a custom position. It should be noted that when the same distance threshold is selected as the measurement standard, the page positions of the path nodes to be matched need to be determined based on the same selection principle.

[0071] B) When a path node in the operation path indicates the operation, for two path nodes to be matched, the endpoint coordinates of the corresponding operation are determined respectively to obtain two endpoint coordinates to be matched; when the distance between the two endpoint coordinates to be matched is less than a second preset distance threshold, it is determined that the two path nodes to be matched meet the node attribute matching requirements of the relevant path nodes. Figure 14 "City A" can correspond to "Perform Operation A on City A," "City B" can correspond to "Perform Operation B on City B," "City C" can correspond to "Perform Operation C on City C," and "Delete" can correspond to "Perform Operation D on Delete." The end coordinates of the operation can be taken from the x-axis and y-axis positions in the preceding code.

[0072] Taking into account page layout and user operation habits, when a user performs actions such as clicking or typing on a page, if the distance between the endpoint coordinates of related actions is within a certain range, then the related actions belong to the same action type; otherwise, they do not belong to the same action type. If the distance between the page objects affected by the related actions is within a certain range, then the related actions belong to the same action type; otherwise, they do not belong to the same action type. If the preset distance threshold is 100px, then if the distance between an endpoint coordinate and its surrounding endpoint coordinates is within 100px, then the related actions belong to the same action type; if the distance between a page location and its surrounding page locations is within 100px, then the related actions belong to the same action type. In the figure, the three buttons "City A," "City B," and "City C" are less than 100px apart, so the related actions belong to the same action type. However, the nearby "Delete" button, although belonging to a common element at a higher level, does not belong to the same action type because it is more than 100px away from the three buttons.

[0073] By comparing the distance between page positions or the distance between end point coordinates with the preset distance threshold, it is possible to determine whether the two path nodes to be matched meet the node attribute matching requirements of the relevant path nodes based on whether they belong to the same operation type. This can increase the attention to the operation type in the matching and ensure that the matched path nodes can cover more operations and the page objects affected by the operations.

[0074] 3) See Figure 10 The present application provides an implementation method for using artificial intelligence (AI) capabilities, such as machine learning, to perform local node attribute matching from the page object dimension and the operation dimension. Specifically, a preset model is used to determine multiple path nodes to be matched, and node attribute matching is performed to obtain a matching result that filters out outlier path nodes. The preset model is obtained by performing machine learning training on multiple sample nodes, and each sample node carries a label indicating whether it is an outlier.

[0075] By training with relevant machine learning models to obtain a pre-set model with high generalization capabilities, node attribute matching using this pre-set model can improve the ability to recognize and adapt to a large number of path nodes to be matched, thereby significantly improving the efficiency and reliability of node attribute matching. The relevant machine learning model utilized can adopt algorithms such as local anomaly factor algorithms and clustering algorithms. The input to the model during training and prediction can be the coordinates of the endpoint of the aforementioned operation, or the page location of the page object affected by the aforementioned operation.

[0076] In practical applications, in order to balance matching efficiency and accuracy, the implementation methods 1)-3) above can be used in combination with each other or even all of them. For example, use the implementation method in 1) first, and then use the implementation method in 2). Because all operations (combined with the above example including "performing operation D for "delete") may be under the note, then if only the "common superior" method is used to classify the operations, it may not be accurate, so the "distance threshold" method is used to continue to determine whether several operations are of the same operation type.

[0077] By classifying the operation paths in the operation path set, a classification result can be obtained to guide operation analysis. The classification result can be obtained after completing the classification process for all operation paths in the operation path set, or after completing the classification process for any two operation paths from the operation path set. The following describes the process of obtaining the former classification result:

[0078] See also Figure 4 , the classifying of the operation paths in the operation path set that meet the preset matching requirements includes:

[0079] S401: arbitrarily selecting two operation paths to be matched from the operation path set to obtain two target operation paths;

[0080] S402: Extracting path nodes from each target operation path to obtain a target path node group corresponding to each target operation path; wherein the target path node group includes a plurality of target path nodes;

[0081] S403: Matching node attribute dimensions and position dimensions between the target path node groups corresponding to the two target operation paths;

[0082] S404: When the matching result meets the preset matching requirement, the two target operation paths are classified;

[0083] S405: Repeat the above steps of obtaining two target operation paths to classifying the two target operation paths until the classification process for all operation paths in the operation path set is completed.

[0084] Taking target operation path 1: A->B->C->D and target operation path 2: A->C->B->D as examples, target path node group 1 corresponding to target operation path 1 includes A, B, C, and D, and target path node group 2 corresponding to target operation path 2 includes A, B, C, and D. Target path node group 1 and target path node group 2 are matched in terms of node attribute dimensions and position dimensions. The matching of node attribute dimensions can include global matching and local matching. The former focuses on comparing whether the constituent elements of the two target path node groups are consistent; the latter will be combined with the matching of the position dimension and focus on comparing whether the path nodes at the same position in the two target path node groups match. The matching here can adopt the implementation method of local node attribute matching introduced above, which will not be repeated here.

[0085] When the matching result meets the preset matching requirements, the two target operation paths are classified. Exemplarily, target operation path 1 and target operation path 2 may be operation paths in the operation path set that have not been used for matching. If the matching result meets the preset matching requirements, a new category attribute may be generated to label target operation path 1 and target operation path 2. Target operation path 1 may be an operation path in the operation path set that has not been used for matching, and target operation path 2 may be an operation path that carries a category attribute, that is, an operation path that has been used for matching in the operation path set. If the matching result meets the preset matching requirements, the category attribute carried by target operation path 2 may be used to label target operation path 1.

[0086] Through the classification processing method provided in the embodiment of the present application, it is possible to complete the classification processing for all operation paths in the operation path set, and it is possible to ensure that a global classification result is obtained to provide an effective reference for operation analysis.

[0087] Furthermore, the step of matching the node attribute dimension and the position dimension between the target path node groups corresponding to the two target operation paths can be achieved by the following two methods:

[0088] The first method involves local matching and positional matching. Specifically, for each target path node group corresponding to the two target operation paths, local node attribute matching is performed on the two target path nodes indicating the same location, one by one, in order from the start node to the end node. If two target path nodes indicate mismatched local node attributes during the matching process, the matching result is determined to not meet the preset matching requirements.

[0089] Taking target operation path 1: A->B->C->D and target operation path 2: A->C->B->D as examples, the first position in target operation path 1 is A, the second position is B, the third position is C, and the last position is D. The first position in target operation path 2 is A, the second position is C, the third position is B, and the last position is D. In the matching, if the two target path nodes indicating the first position are both A, then the two target path nodes indicating the first position match; if the two target path nodes indicating the second position are B and C respectively, then the two target path nodes indicating the second position do not match. Accordingly, it can be determined that target operation path 1 and target operation path 2 do not meet the preset matching requirements.

[0090] Second type of method): first perform global matching, then perform local matching and position dimension matching. Global matching involves: first, performing global node attribute matching on the target path node groups corresponding to the two target operation paths; then, when the result of the global node attribute matching indicates a match, triggering the step of performing local node attribute matching on the two target path nodes indicating the same position, one by one, in the order from the start node to the end node. The matching of local matching and position dimension matching can refer to the first type of method described above and will not be repeated here.

[0091] The matching metrics for the global node attribute match include the number of nodes within a group and the node attributes within the group. This means that the comparison focuses on whether the constituent elements of the two target path node groups are consistent. For example, if target operation path 1 (A->B->C->D) and target operation path 2 (A->C->B->D) correspond to target path node group 1 (A, B, C, and D), and target operation path 2 (A->C->B->D), the global node attribute match result indicates a match.

[0092] Taking target operation path 3: A->B->C and target operation path 4: C->B->D as examples, target path node group 3 corresponding to target operation path 3 includes A, B, and C, and target path node group 4 corresponding to target operation path 4 includes B, C, and D. Although the number of nodes in the groups is the same, the node attributes in the groups are different. Therefore, it can be determined that the result of the global node attribute matching indicates a mismatch.

[0093] In practical applications, operational analysis refers to analyzing user actions and interactions on a page, also known as behavioral analysis. This analysis includes areas and buttons clicked after entering a page, information viewed, searches for content, and whether users performed actions A, B, and C. Categorized results can be used to guide operational analysis, allowing for analysis based on clusters. During analysis, operation playback can be performed to restore and display past user actions on a specific page at a specific time. Specifically, this can be done by restoring and displaying the user's operating environment and actions in chronological order.

[0094] The significance of classification is that when you find that many people have the same operation path, you can analyze the same type of problems. If you solve and optimize the same type of problems, you will get greater benefits. Figure 16 If many people follow this path, you will find that each time they return from the bottom of the page (E) to the top of the page (F), they will then return to the bottom of the page (G). This shows that the design may be unreasonable and F should be moved to the bottom of the page (see Figure 17 ), so that users do not have to return to the top and then to the bottom every time, which can optimize the user experience. By moving the F that was originally placed at the top of the page to the bottom of the page, the user's operation path can be greatly shortened, thereby improving the page operation experience. By utilizing the classification scheme of the operation path provided in the embodiment of the present application, with the help of the different types of operation paths mined from the operation path set reflected in the classification results, operation analysis can be performed in the dimension of a type of operation path to locate abnormal operations, thereby improving the efficiency and accuracy of page optimization (such as adjusting the page layout), thereby improving the page operation experience. Abnormal operations can be determined by abnormal path node segments. Abnormal operations can be related operations that indicate abnormal path node segments. Abnormal operations can be Figure 16 The operation corresponding to the abnormal path node segment EFG can also be the operation corresponding to the abnormal path node segment consisting of at least two path nodes with too small a distance. For page optimization of the former, please refer to Figure 17 , no longer elaborate; for the latter page optimization, the page position spacing of the page objects affected by the abnormal operation can be increased on the relevant pages, thereby avoiding repeated operations, such as repeated clicks caused by inaccurate clicks.

[0095] The embodiments of the present application intelligently categorize user operation paths on a page in an automated manner. This is achieved by not only recording the x- and y-axis coordinate positions of a user's operation but also automatically identifying whether the user operation is acting on the same page object and what that page object is, based on the coordinates. This allows for accurate identification even when the coordinates differ. By categorizing identical or similar operation paths, it is possible to more clearly analyze user behavior and identify commonalities.

[0096] It can be seen from the technical solutions provided by the above embodiments of the present application that the embodiments of the present application obtain an operation path set, and then classify the operation paths in the operation path set that meet the preset matching requirements to obtain a classification result for guiding operation analysis. The present application classifies the operation path set involved in the same page. The classification result can reflect the different types of operation paths mined from the operation path set. The classification result can be used as accurate and effective reference data for restoring a reference access operation process. On this basis, the convenience and efficiency of using the operation path to mine operation rules and capture abnormal operations can be improved, thereby achieving the effect of obtaining deep and hidden information through data processing based on the operation path, and improving the operation analysis effect. Compared with the reference data obtained by counting the number of operations in the page position dimension in the related art, the classification result for guiding operation analysis obtained in the present application is derived from the classification processing of the operation path set. The operation path characterizes the multiple operations involved in the process of accessing a specified page and the execution order between the multiple operations. Therefore, the evaluation granularity information carried by the reference data is richer and can better help improve the accuracy of the operation analysis.

[0097] The embodiment of the present application also provides a classification device for operation paths, such as Figure 18 As shown, the operation path classification device 180 includes:

[0098] Set acquisition module 1801: configured to acquire an operation path set; wherein the operation path set includes multiple operation paths, each of which is generated based on operation data, wherein the operation data records multiple operations involved in accessing a specified page and the execution order of the multiple operations, and the path nodes in the operation path indicate the operation or the page object on which the operation acts;

[0099] Classification module 1802: used to classify the operation paths in the operation path set that meet the preset matching requirements to obtain classification results for guiding operation analysis; wherein, the preset matching requirements include node attribute matching requirements of the relevant path nodes and position matching requirements of the relevant path nodes.

[0100] It should be noted that the device and method embodiments in the device embodiment are based on the same inventive concept.

[0101] An embodiment of the present application provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the classification method of the operation path provided in the above method embodiment.

[0102] Furthermore, Figure 19 A hardware structure diagram of an electronic device for implementing the method for classifying the operation paths provided in the embodiment of the present application is shown. The electronic device may participate in or include the device for classifying the operation paths provided in the embodiment of the present application. Figure 19 As shown, the electronic device 100 may include one or more (illustrated as 1002a, 1002b, ..., 1002n in the figure) processors 1002 (the processor 1002 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 1004 for storing data, and a transmission device 1006 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, a power supply and / or a camera. It will be understood by those skilled in the art that Figure 19 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 19 More or fewer components than shown, or with Figure 19 Different configurations shown.

[0103] It should be noted that the one or more processors 1002 and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuitry may be a single independent processing module, or may be incorporated in whole or in part into any of the other components of the electronic device 100 (or mobile device). As described in the embodiments of the present application, the data processing circuitry serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0104] The memory 1004 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the classification method of the operation path described in the embodiment of the present application. The processor 1002 executes various functional applications and data processing by running the software programs and modules stored in the memory 1004, that is, implementing the above-mentioned classification method of the operation path. The memory 1004 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 1004 may further include a memory remotely located relative to the processor 1002, and these remote memories may be connected to the electronic device 100 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0105] The transmission device 1006 is used to receive or send data via a network. Specific examples of the aforementioned network may include a wireless network provided by the communication provider of the electronic device 100. In one embodiment, the transmission device 1006 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In one embodiment, the transmission device 1006 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0106] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the electronic device 100 (or mobile device).

[0107] An embodiment of the present application also provides a computer-readable storage medium, which can be set in an electronic device to store at least one instruction or at least one program related to a classification method for an operation path in a method embodiment. The at least one instruction or the at least one program is loaded and executed by the processor to implement the classification method for the operation path provided in the above method embodiment.

[0108] Optionally, in this embodiment, the storage medium may be located in at least one of a plurality of network servers in a computer network. Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0109] It should be noted that the order of the embodiments of the present application described above is for descriptive purposes only and does not represent the superiority or inferiority of the embodiments. The above description is of specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0110] The various embodiments in this application are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the device and electronic device embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the partial description of the method embodiments.

[0111] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or by a program to instruct the relevant hardware, and the program may be stored in a computer-readable storage medium, which may be a read-only memory, a disk, or an optical disk, etc.

[0112] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for classifying operation paths, characterized in that: The method comprises: Obtaining an operation path set; the operation path set includes multiple operation paths, the operation paths are generated based on operation data, the operation data records multiple operations involved in the process of accessing a specified page and the execution order of the multiple operations, and the path nodes in the operation paths indicate the operations or the page objects affected by the operations; Classifying the operation paths in the operation path set that meet preset matching requirements to obtain classification results for guiding operation analysis; the preset matching requirements include node attribute matching requirements for relevant path nodes and position matching requirements for relevant path nodes, wherein the position matching requirements for relevant path nodes are used to constrain two path nodes to be matched to have the same position in the corresponding operation paths; Wherein, when the path node in the operation path indicates the page object on which the operation is performed, the method further includes: For the two path nodes to be matched, respectively determine corresponding page objects to obtain two page objects to be matched; When the two page objects to be matched belong to the same upper-level page object, determining whether the two path nodes to be matched meet the node attribute matching requirement of the relevant path nodes; When the upper-level page object is an attribute-type object, determining a candidate page object of the same level as the upper-level page object, wherein at least two lower-level page objects of the attribute-type object have the same attribute; When the candidate page object is a tag-type object, and the candidate page object and the upper-level page object match because they serve the same user intention, determining the candidate page object as the target page object; The target page object is used to generate classification labels corresponding to the two path nodes to be matched.

2. The method according to claim 1, characterized in that The method further comprises: When a path node in the operation path indicates a page object on which the operation is performed, for the two path nodes to be matched, respectively determining the page positions of the corresponding page objects to obtain two page positions to be matched; when a distance between the two page positions to be matched is less than a first preset distance threshold, determining that the two path nodes to be matched meet the node attribute matching requirement of the relevant path nodes; Alternatively, when the path node in the operation path indicates the operation, the endpoint coordinates of the corresponding operations are determined respectively for the two path nodes to be matched to obtain two endpoint coordinates to be matched; when the distance between the two endpoint coordinates to be matched is less than a second preset distance threshold, it is determined that the two path nodes to be matched meet the node attribute matching requirements of the relevant path nodes.

3. The method according to claim 1, characterized in that The method further comprises: A preset model is used to determine multiple path nodes to be matched for node attribute matching, and a matching result is obtained by filtering out outlier path nodes; wherein the preset model is obtained by machine learning training on multiple sample nodes, and each sample node carries a label indicating whether it is an outlier.

4. The method according to claim 1, wherein The classifying the operation paths in the operation path set that meet the preset matching requirements includes: Randomly select two operation paths to be matched from the operation path set to obtain two target operation paths; Extracting path nodes from each target operation path respectively to obtain a target path node group corresponding to each target operation path; wherein the target path node group includes a plurality of target path nodes; Matching node attribute dimensions and position dimensions between the target path node groups corresponding to the two target operation paths; When the matching result meets the preset matching requirement, the two target operation paths are classified; Repeat the steps of obtaining two target operation paths to classifying the two target operation paths until the classification process for all operation paths in the operation path set is completed.

5. The method according to claim 4, characterized in that The matching of node attribute dimensions and position dimensions between the target path node groups corresponding to the two target operation paths includes: For the target path node groups corresponding to the two target operation paths, respectively, in order from the start node to the end node, local node attribute matching is performed on the two target path nodes indicating the same position one by one; When there are two target path nodes indicating that the local node attributes do not match during the matching process, it is determined that the matching result does not meet the preset matching requirement.

6. The method according to claim 5, characterized in that The method further comprises: Performing global node attribute matching on the target path node groups corresponding to the two target operation paths: wherein the matching index of the global node attribute matching includes the number of nodes in the group and the attributes of the nodes in the group; When the result of the global node attribute matching indicates a match, the step of performing local node attribute matching on two target path nodes indicating the same position one by one in the order from the start node to the end node is triggered.

7. An operation path classification device, characterized in that: The device comprises: A set acquisition module is used to acquire an operation path set; the operation path set includes multiple operation paths, each of which is generated based on operation data. The operation data records multiple operations involved in accessing a specified page and the execution order of the multiple operations. The path nodes in the operation path indicate the operation or the page object on which the operation acts. A classification module is configured to classify the operation paths in the operation path set that meet preset matching requirements, and obtain classification results for guiding operation analysis; the preset matching requirements include node attribute matching requirements for the relevant path nodes and position matching requirements for the relevant path nodes, and the position matching requirements for the relevant path nodes are used to constrain two path nodes to be matched to have the same position in the corresponding operation paths; Wherein, when the path node in the operation path indicates the page object on which the operation is performed, the apparatus is further configured to: For the two path nodes to be matched, respectively determine corresponding page objects to obtain two page objects to be matched; When the two page objects to be matched belong to the same upper-level page object, determining whether the two path nodes to be matched meet the node attribute matching requirement of the relevant path nodes; When the upper-level page object is an attribute-type object, determining a candidate page object of the same level as the upper-level page object, wherein at least two lower-level page objects of the attribute-type object have the same attribute; When the candidate page object is a tag-type object, and the candidate page object and the upper-level page object match because they serve the same user intention, determining the candidate page object as the target page object; The target page object is used to generate classification labels corresponding to the two path nodes to be matched.

8. The device according to claim 7, characterized in that The device is also used for: When a path node in the operation path indicates a page object on which the operation is performed, for the two path nodes to be matched, respectively determining the page positions of the corresponding page objects to obtain two page positions to be matched; when a distance between the two page positions to be matched is less than a first preset distance threshold, determining that the two path nodes to be matched meet the node attribute matching requirement of the relevant path nodes; Alternatively, when the path node in the operation path indicates the operation, the endpoint coordinates of the corresponding operations are determined respectively for the two path nodes to be matched to obtain two endpoint coordinates to be matched; when the distance between the two endpoint coordinates to be matched is less than a second preset distance threshold, it is determined that the two path nodes to be matched meet the node attribute matching requirements of the relevant path nodes.

9. The device according to claim 7, characterized in that The device is also used for: A preset model is used to determine multiple path nodes to be matched for node attribute matching, and a matching result is obtained by filtering out outlier path nodes; wherein the preset model is obtained by machine learning training on multiple sample nodes, and each sample node carries a label indicating whether it is an outlier.

10. The device according to claim 7, characterized in that The classifying the operation paths in the operation path set that meet the preset matching requirements includes: Randomly select two operation paths to be matched from the operation path set to obtain two target operation paths; Extracting path nodes from each target operation path respectively to obtain a target path node group corresponding to each target operation path; wherein the target path node group includes a plurality of target path nodes; Matching node attribute dimensions and position dimensions between the target path node groups corresponding to the two target operation paths; When the matching result meets the preset matching requirement, the two target operation paths are classified; Repeat the steps of obtaining two target operation paths to classifying the two target operation paths until the classification process for all operation paths in the operation path set is completed.

11. The device according to claim 10, characterized in that The matching of node attribute dimensions and position dimensions between the target path node groups corresponding to the two target operation paths includes: For the target path node groups corresponding to the two target operation paths, respectively, in order from the start node to the end node, local node attribute matching is performed on the two target path nodes indicating the same position one by one; When there are two target path nodes indicating that the local node attributes do not match during the matching process, it is determined that the matching result does not meet the preset matching requirement.

12. The device according to claim 11, characterized in that The device is also used for: Performing global node attribute matching on the target path node groups corresponding to the two target operation paths: wherein the matching index of the global node attribute matching includes the number of nodes in the group and the attributes of the nodes in the group; When the result of the global node attribute matching indicates a match, the step of performing local node attribute matching on two target path nodes indicating the same position one by one in the order from the start node to the end node is triggered.

13. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the operation path classification method according to any one of claims 1 to 6.

14. A computer-readable storage medium, characterized in that The storage medium stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the operation path classification method according to any one of claims 1 to 6.

15. A computer program product, characterized in that The computer program product includes computer instructions, which are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the operation path classification method according to any one of claims 1 to 6.

Images