CN114140031A - Method and device for attribution analysis of user behaviors - Google Patents
Method and device for attribution analysis of user behaviors Download PDFInfo
- Publication number
- CN114140031A CN114140031A CN202210103677.2A CN202210103677A CN114140031A CN 114140031 A CN114140031 A CN 114140031A CN 202210103677 A CN202210103677 A CN 202210103677A CN 114140031 A CN114140031 A CN 114140031A
- Authority
- CN
- China
- Prior art keywords
- behavior
- user
- graph
- node
- network graph
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/26—Visual data mining; Browsing structured data
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Databases & Information Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- General Engineering & Computer Science (AREA)
- Educational Administration (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Debugging And Monitoring (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the specification provides a method and a device for attribution analysis of user behaviors, wherein the method is executed by a server and comprises the following steps: acquiring user behavior log data corresponding to a first user reported by a client; the user behavior log data at least comprises behavior information of a plurality of preorder behaviors and target behaviors which are executed in sequence; extracting a first sub-graph corresponding to the first user from a pre-established relationship network graph; the relational network graph at least comprises user nodes representing users, object nodes representing behavior objects and connecting edges among the user nodes and the object nodes; analyzing the user behavior log data to obtain an increment graph; merging the first subgraph and the increment graph to obtain a merged subgraph; and determining an effective path formed by at least one behavior in the plurality of preamble behaviors and the target behavior by carrying out graph calculation on the merged subgraph. Can give consideration to both timeliness and accuracy.
Description
Technical Field
One or more embodiments of the present specification relate to the field of computers, and more particularly, to a method and apparatus for attribution analysis of user behavior.
Background
In the great trend of digitization and informatization, the internet brings great convenience to both consumers and merchants, and channel attribution and path analysis of user behaviors are the core of the field of flow analysis. By calculating the effective behavior path of the user, a complete conversion path is constructed, the value of a product can be quickly seen by a service, and the operation idea can be timely adjusted.
The existing common method for carrying out attribution analysis on user behaviors has poor timeliness or low accuracy, and cannot give consideration to both timeliness and accuracy.
Disclosure of Invention
One or more embodiments of the present specification describe a method and apparatus for attribution analysis of user behavior, which can take account of both timeliness and accuracy.
In a first aspect, a method for attribution analysis of user behavior is provided, where the method is performed by a server and includes:
acquiring user behavior log data corresponding to a first user reported by a client; the user behavior log data at least comprises behavior information of a plurality of preorder behaviors and target behaviors which are executed in sequence;
extracting a first sub-graph corresponding to the first user from a pre-established relationship network graph; the relational network graph at least comprises user nodes representing users, object nodes representing behavior objects and connecting edges among the user nodes and the object nodes;
analyzing the user behavior log data to obtain an increment graph;
merging the first subgraph and the increment graph to obtain a merged subgraph;
and determining an effective path formed by at least one behavior in the plurality of preamble behaviors and the target behavior by carrying out graph calculation on the merged subgraph.
In a possible implementation manner, the incremental graph includes a first user node corresponding to the first user, a plurality of object nodes corresponding to behavior objects of the plurality of preamble behaviors, a behavior node corresponding to the target behavior, a first connection edge between the first user node and the plurality of object nodes, and a second connection edge between the first user node and the behavior node.
In a possible implementation manner, the user behavior log data is obtained by performing a buried point analysis on a behavior object in the client.
In one possible embodiment, the target behavior comprises a trading behavior or a downloading behavior;
the plurality of preorder behaviors comprise at least one behavior of clicking, browsing and collecting.
In one possible implementation, the behavior information includes information of a behavior object and time information of occurrence of a behavior; the connection edge has corresponding edge attribute information, and the edge attribute information at least includes time information of behavior occurrence.
In a possible implementation manner, the relationship network graph further includes a behavior node corresponding to the target behavior, and a connection edge between the user node and the behavior node.
In a possible implementation manner, the extracting a first sub-graph corresponding to the first user from a pre-established relationship network graph includes:
searching a first user node representing the first user from the relational network graph;
and extracting the first user node, the object node with a connecting edge with the first user node and the connecting edge between the first user node and the object node from the relational network graph to serve as the first subgraph.
In one possible embodiment, the method further comprises:
and updating the relational network graph according to the increment graph.
Further, the updating the relationship network graph includes:
if the object node included in the incremental graph does not exist in the relational network graph and the user node included in the incremental graph exists in the relational network graph, adding a corresponding object node in the relational network graph and adding a connecting edge between the object node and the original user node in the relational network graph;
and if the user node and the object node included in the incremental graph exist in the relational network graph, adding a connecting edge between the object node and the user node in the relational network graph.
In a second aspect, an apparatus for performing attribution analysis on user behavior is provided, where the apparatus is disposed at a server, and includes:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring user behavior log data corresponding to a first user reported by a client; the user behavior log data at least comprises behavior information of a plurality of preorder behaviors and target behaviors which are executed in sequence;
the extracting unit is used for extracting a first sub-graph corresponding to the first user from a pre-established relationship network graph; the relational network graph at least comprises user nodes representing users, object nodes representing behavior objects and connecting edges among the user nodes and the object nodes;
the analysis unit is used for analyzing the user behavior log data acquired by the acquisition unit to obtain an increment graph;
the merging unit is used for merging the first subgraph obtained by the extracting unit and the increment graph obtained by the analyzing unit to obtain a merged subgraph;
and the graph calculation unit is used for performing graph calculation on the merged subgraph obtained by the merging unit and determining an effective path formed by at least one behavior in the plurality of preamble behaviors and the target behavior.
In a third aspect, there is provided a computer readable storage medium having stored thereon a computer program which, when executed in a computer, causes the computer to perform the method of the first aspect.
In a fourth aspect, there is provided a computing device comprising a memory having stored therein executable code and a processor that, when executing the executable code, implements the method of the first aspect.
According to the method and the device provided by the embodiment of the specification, firstly, a server side obtains user behavior log data corresponding to a first user reported by a client side; the user behavior log data at least comprises behavior information of a plurality of preorder behaviors and target behaviors which are executed in sequence; then extracting a first sub-graph corresponding to the first user from a pre-established relationship network graph; the relational network graph at least comprises user nodes representing users, object nodes representing behavior objects and connecting edges among the user nodes and the object nodes; analyzing the user behavior log data to obtain an increment graph; merging the first subgraph and the increment graph to obtain a merged subgraph; and finally, performing graph calculation on the merged subgraph to determine an effective path formed by at least one behavior in the plurality of preamble behaviors and the target behavior. It can be seen from the above that, unlike a conventional offline calculation scheme, in the embodiment of the present specification, instead of collecting user data for a period of time and analyzing the user data in a unified manner, a user is abstracted into a type of node, a behavior object is abstracted into another type of node, a relational network graph reflecting the fact of user behavior is constructed, on the basis of this, an incremental graph is obtained by analyzing log data of user behavior in real time, a subgraph of a corresponding user is extracted from a pre-constructed relational network graph, and the subgraph and the incremental graph are merged to obtain a merged subgraph, so that attribution analysis of user behavior is converted into graph calculation for the merged subgraph, thereby considering timeliness and accuracy.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic diagram illustrating an implementation scenario of an embodiment disclosed herein;
FIG. 2 illustrates a flow diagram of a method of attribution analysis of user behavior, according to one embodiment;
FIG. 3 illustrates a structural schematic of a relational network diagram according to one embodiment;
FIG. 4 illustrates a schematic diagram of a first sub-graph according to one embodiment;
FIG. 5 shows a schematic block diagram of an apparatus for attribution analysis of user behavior according to one embodiment.
Detailed Description
The scheme provided by the specification is described below with reference to the accompanying drawings.
Fig. 1 is a schematic view of an implementation scenario of an embodiment disclosed in this specification. The implementation scenario involves attribution analysis of user behavior. Attribution analysis: and calculating the effective behavior path of the user, constructing a reasonable conversion path, and performing channel attribution and path analysis on the reasonable conversion path. It is understood that the target behavior of the user is generally referred to as conversion, the target behavior may include, but is not limited to, transaction behavior, download behavior, etc., before the target behavior occurs, the user often occurs a series of pre-oriented behaviors, such as clicking or browsing a plurality of pages, a causal relationship may exist between the pre-oriented behaviors and the target behavior, and the attribution analysis is to analyze the causal relationship to make the target behavior interpretable.
Referring to fig. 1, before completing a transaction, a user 1 performs a series of operations, such as opening a home page of an Application (APP), clicking a plurality of pages, for example, page 1, page 2, page 3, and page 4, and then performing a transaction. The payment is completed in one transaction, and the front-end series of clicking, accessing and staying on the pages can form a behavior path of the user.
The effective behavior path of the user may refer to a main path from the opening of the first page of the APP to the completion of payment, where the main path is a path obtained after some operation behaviors that do not contribute to achieving the transaction are deleted on the basis of all operation behaviors of the user, for example, an operation behavior of clicking on the page 2 and an operation behavior of clicking on the page 3 are considered to not contribute to achieving the transaction, and therefore, the two operation behaviors are deleted in the behavior path of the user, so as to obtain the effective behavior path of the user.
The conversion path is equal to the effective behavior path in most scenes, but according to the difference of scenes and the difference of algorithms, the effective behavior path can be further processed to obtain the conversion path.
Channel attribution means that path calculation is respectively carried out according to different service categories to which target behaviors belong, such as fund, insurance, activity pages and the like, and the same user is likely to achieve conversion of a plurality of channels at a certain time and needs to respectively carry out conversion path calculation on the channels.
The path analysis means that service analysis is performed according to the calculated conversion path so as to provide data support for service and operation decision.
In the embodiment of the specification, the relational network diagram is constructed based on the user behavior data, and the user behavior is subjected to attribution analysis in a diagram calculation mode, so that both timeliness and accuracy can be considered.
Fig. 2 is a flowchart illustrating a method for attribution analysis of user behavior according to an embodiment, which may be executed by a server based on the implementation scenario illustrated in fig. 1, where the server corresponds to a client and cooperates with the client to ensure normal operation of an application. As shown in fig. 2, the method for attribution analysis of user behavior in this embodiment includes the following steps: step 21, obtaining user behavior log data corresponding to a first user reported by a client; the user behavior log data at least comprises behavior information of a plurality of preorder behaviors and target behaviors which are executed in sequence; step 22, extracting a first sub-graph corresponding to the first user from a pre-established relationship network graph; the relational network graph at least comprises user nodes representing users, object nodes representing behavior objects and connecting edges among the user nodes and the object nodes; step 23, analyzing the user behavior log data to obtain an increment graph; step 24, combining the first subgraph and the increment graph to obtain a combined subgraph; and 25, determining an effective path formed by at least one behavior in the plurality of preamble behaviors and the target behavior by carrying out graph calculation on the merged subgraph. Specific execution modes of the above steps are described below.
Firstly, in step 21, user behavior log data corresponding to a first user reported by a client is obtained; the user behavior log data at least comprises behavior information of a plurality of preorder behaviors and target behaviors which are executed in sequence. It can be understood that the user behavior log data may be user behavior log data in a current time period periodically reported by the client.
In one example, the user behavior log data is obtained by performing a buried point analysis on a behavior object in the client.
In one example, the target behavior comprises a trading behavior or a downloading behavior;
the plurality of preorder behaviors comprise at least one behavior of clicking, browsing and collecting.
In one example, the behavior information includes information of a behavior object and time information of a behavior occurrence.
The behavior object may be a page or a page position, and the information of the behavior object may include: one or more of shopping guide page mark (SPM) codes of corresponding pages of the behavior actions, SPM codes of corresponding controls of the behavior actions and corresponding positions of the behavior actions; the time information of the behavior occurrence may include: one or more of a timestamp corresponding to the behavior action, a retention time of a corresponding page of the behavior action, and an interval time between the behavior action and its neighboring actions.
SPM coding may be used to accurately identify behavioral objects (e.g., web pages, controls) corresponding to behavioral actions. For example, the SPM code may comprise a.b.c.d level four code. Where a may be a site ID, and the site ID is used to identify a root site of a particular service. b may be a page ID that identifies a particular page under a particular service, such as the top page of the service, i.e., the first two levels of encoding (a.b) of the SPM uniquely identify a given page. c may be a region ID for identifying a specific region on the page. d may be a control ID for indicating an operable control such as a corresponding link, picture, card, etc. in a specific area of the page.
Then, in step 22, extracting a first sub-graph corresponding to the first user from a pre-established relationship network graph; the relational network graph at least comprises user nodes representing users, object nodes representing behavior objects and connecting edges among the user nodes and the object nodes. It will be appreciated that the relationship network graph includes a plurality of user nodes, for example, a second user node representing a second user, a third user node representing a third user, and so on, in addition to a first user node representing a first user.
In one example, the connecting edges have corresponding edge attribute information that includes at least time information at which the behavior occurred.
In one example, the relationship network graph further includes a behavior node corresponding to the target behavior, and a connection edge between a user node and the behavior node.
FIG. 3 illustrates a structural schematic of a relational network diagram according to one embodiment. Referring to fig. 3, u1, u2, u3 and u4 are user nodes representing users, and it can be understood that each user node represents different users; s1, s2, s3, s4, s5, s6, s7, s8 and s9 are object nodes representing behavior objects, and it can be understood that each object node represents a different behavior object; t1, t2, t3, t4, t5 and t6 are behavior nodes representing target behaviors, and it is understood that each behavior node represents a different target behavior. In the relational network graph, the behavior node is optional, and the behavior node may not be included, and only the user node and the object node are included. There may be multiple connection edges between one user node and one object node, where one connection edge corresponds to one user behavior, and the connection edge has corresponding edge attribute information, and the edge attribute information may include statistical information required for later analysis. The real-time graph calculation engine is accessed into the user behavior log data, and the behavior fact graphs of all users can be constructed in real time by analyzing corresponding nodes and connecting edges, namely the relationship network graph is constructed.
In one example, the extracting a first sub-graph corresponding to the first user from a pre-established relationship network graph includes:
searching a first user node representing the first user from the relational network graph;
and extracting the first user node, the object node with a connecting edge with the first user node and the connecting edge between the first user node and the object node from the relational network graph to serve as the first subgraph.
Fig. 4 shows a schematic structural diagram of a first sub-graph according to an embodiment, the first sub-graph being extracted based on the relationship network graph shown in fig. 3. Referring to fig. 4, u1 is a first user node representing the first user, s1, s2, s4, s9 are object nodes having connecting edges with u1, and these nodes and the connecting edges therebetween constitute a first sub-graph. It is understood that the way of extracting the first sub-graph is not unique, and how to extract the first sub-graph depends on a preset rule.
Then, in step 23, the user behavior log data is analyzed to obtain an increment map. It can be understood that, after acquiring the user behavior log data corresponding to the first user reported by the client, the server analyzes the user behavior log data in real time.
In one example, the incremental graph includes a first user node corresponding to the first user, a number of object nodes corresponding to behavior objects of the number of preceding behaviors, a behavior node corresponding to the target behavior, a first connecting edge between the first user node and the number of object nodes, and a second connecting edge between the first user node and the behavior node.
In one example, the method further comprises:
and updating the relational network graph according to the increment graph.
Further, the updating the relationship network graph includes:
if the object node included in the incremental graph does not exist in the relational network graph and the user node included in the incremental graph exists in the relational network graph, adding a corresponding object node in the relational network graph and adding a connecting edge between the object node and the original user node in the relational network graph;
and if the user node and the object node included in the incremental graph exist in the relational network graph, adding a connecting edge between the object node and the user node in the relational network graph.
And step 24, merging the first subgraph and the increment graph to obtain a merged subgraph. It will be appreciated that the first sub-graph and the incremental graph correspond to the same user, both having the same user nodes, and in addition, both may have the same object nodes, that is, the user may access the same behavior object twice, and the merging process includes merging the nodes that are common in the first sub-graph and the incremental graph into one, and keeping the respective unique nodes.
Finally, in step 25, by performing graph calculation on the merged subgraph, an effective path formed by at least one behavior in the plurality of the preceding behaviors and the target behavior is determined. It will be appreciated that the graph computation typically involves multiple iterations, with the valid path being obtained after the last iteration.
For example, in multiple rounds of iteration, a sub-graph of a user corresponding to a transaction is loaded in a first round of computation, a preliminary effective path is obtained according to a matching rule, if it is determined that information of other sub-graphs is needed, a current effective path is temporarily stored, a sub-graph of a trigger point is continuously loaded in a next round of iteration, an existing effective path is optimized again according to the matching rule, the effective path obtained in the previous round and the sub-graph of the trigger point in the current round, and finally, computation is finished according to a termination condition of the matching rule or the maximum iteration number, wherein the effective path is the final effective path.
According to the method provided by the embodiment of the specification, firstly, a server side obtains user behavior log data corresponding to a first user reported by a client side; the user behavior log data at least comprises behavior information of a plurality of preorder behaviors and target behaviors which are executed in sequence; then extracting a first sub-graph corresponding to the first user from a pre-established relationship network graph; the relational network graph at least comprises user nodes representing users, object nodes representing behavior objects and connecting edges among the user nodes and the object nodes; analyzing the user behavior log data to obtain an increment graph; merging the first subgraph and the increment graph to obtain a merged subgraph; and finally, performing graph calculation on the merged subgraph to determine an effective path formed by at least one behavior in the plurality of preamble behaviors and the target behavior. It can be seen from the above that, unlike a conventional offline calculation scheme, in the embodiment of the present specification, instead of collecting user data for a period of time and analyzing the user data in a unified manner, a user is abstracted into a type of node, a behavior object is abstracted into another type of node, a relational network graph reflecting the fact of user behavior is constructed, on the basis of this, an incremental graph is obtained by analyzing log data of user behavior in real time, a subgraph of a corresponding user is extracted from a pre-constructed relational network graph, and the subgraph and the incremental graph are merged to obtain a merged subgraph, so that attribution analysis of user behavior is converted into graph calculation for the merged subgraph, thereby considering timeliness and accuracy.
According to another aspect of embodiments, an apparatus for attribution analysis of user behavior is further provided, where the apparatus is disposed at a server and is configured to perform the method provided by the embodiments of the present specification. FIG. 5 shows a schematic block diagram of an apparatus for attribution analysis of user behavior according to one embodiment. As shown in fig. 5, the apparatus 500 includes:
an obtaining unit 51, configured to obtain user behavior log data corresponding to a first user reported by a client; the user behavior log data at least comprises behavior information of a plurality of preorder behaviors and target behaviors which are executed in sequence;
an extracting unit 52, configured to extract a first sub-graph corresponding to the first user from a pre-established relationship network graph; the relational network graph at least comprises user nodes representing users, object nodes representing behavior objects and connecting edges among the user nodes and the object nodes;
an analyzing unit 53, configured to analyze the user behavior log data acquired by the acquiring unit 51 to obtain an incremental map;
a merging unit 54, configured to merge the first sub-graph obtained by the extracting unit 52 and the increment graph obtained by the parsing unit 53 to obtain a merged sub-graph;
a graph calculating unit 55, configured to determine an effective path formed by at least one of the behaviors and the target behavior by performing graph calculation on the merged subgraph obtained by the merging unit 54.
Optionally, as an embodiment, the incremental graph includes a first user node corresponding to the first user, a plurality of object nodes corresponding to behavior objects of the plurality of preamble behaviors, a behavior node corresponding to the target behavior, a first connection edge between the first user node and the plurality of object nodes, and a second connection edge between the first user node and the behavior node.
Optionally, as an embodiment, the user behavior log data is obtained by performing a buried point analysis on a behavior object in the client.
Optionally, as an embodiment, the target behavior comprises a trading behavior or a downloading behavior;
the plurality of preorder behaviors comprise at least one behavior of clicking, browsing and collecting.
Optionally, as an embodiment, the behavior information includes information of a behavior object and time information of occurrence of a behavior; the connection edge has corresponding edge attribute information, and the edge attribute information at least includes time information of behavior occurrence.
Optionally, as an embodiment, the relationship network graph further includes a behavior node corresponding to the target behavior, and a connection edge between the user node and the behavior node.
Optionally, as an embodiment, the extracting unit 52 includes:
a searching subunit, configured to search, from the relationship network graph, a first user node representing the first user;
and the extracting subunit is configured to extract, from the relationship network graph, the first user node found by the searching subunit, the object node having a connection edge with the first user node, and a connection edge therebetween, as the first sub-graph.
Optionally, as an embodiment, the apparatus further includes:
and the updating unit is used for updating the relational network graph according to the increment graph.
Further, the update unit includes:
a first updating subunit, configured to add, if an object node included in the incremental graph does not exist in the relational network graph and a user node included in the incremental graph exists in the relational network graph, a corresponding object node in the relational network graph and add a connection edge between the object node and an original user node in the relational network graph;
and a second updating subunit, configured to add, if both the user node and the object node included in the incremental graph exist in the relational network graph, a connecting edge between the object node and the user node in the relational network graph.
With the apparatus provided in this specification, first, an obtaining unit 51 of a server obtains user behavior log data corresponding to a first user reported by a client; the user behavior log data at least comprises behavior information of a plurality of preorder behaviors and target behaviors which are executed in sequence; then, the extracting unit 52 extracts a first sub-graph corresponding to the first user from a pre-established relationship network graph; the relational network graph at least comprises user nodes representing users, object nodes representing behavior objects and connecting edges among the user nodes and the object nodes; then, the analyzing unit 53 analyzes the user behavior log data to obtain an increment map; merging the first subgraph and the increment graph by a merging unit 54 to obtain a merged subgraph; and the final graph calculation unit 55 determines an effective path formed by at least one behavior in the plurality of the preceding behaviors and the target behavior by performing graph calculation on the merged subgraph. It can be seen from the above that, unlike a conventional offline calculation scheme, in the embodiment of the present specification, instead of collecting user data for a period of time and analyzing the user data in a unified manner, a user is abstracted into a type of node, a behavior object is abstracted into another type of node, a relational network graph reflecting the fact of user behavior is constructed, on the basis of this, an incremental graph is obtained by analyzing log data of user behavior in real time, a subgraph of a corresponding user is extracted from a pre-constructed relational network graph, and the subgraph and the incremental graph are merged to obtain a merged subgraph, so that attribution analysis of user behavior is converted into graph calculation for the merged subgraph, thereby considering timeliness and accuracy.
According to an embodiment of another aspect, there is also provided a computer-readable storage medium having stored thereon a computer program which, when executed in a computer, causes the computer to perform the method described in connection with fig. 2.
According to an embodiment of yet another aspect, there is also provided a computing device comprising a memory having stored therein executable code, and a processor that, when executing the executable code, implements the method described in connection with fig. 2.
Those skilled in the art will recognize that, in one or more of the examples described above, the functions described in this invention may be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium.
The above-mentioned embodiments, objects, technical solutions and advantages of the present invention are further described in detail, it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made on the basis of the technical solutions of the present invention should be included in the scope of the present invention.
Claims (20)
1. A method for attribution analysis of user behavior, the method being performed by a server and comprising:
acquiring user behavior log data corresponding to a first user reported by a client; the user behavior log data at least comprises behavior information of a plurality of preorder behaviors and target behaviors which are executed in sequence;
extracting a first sub-graph corresponding to the first user from a pre-established relationship network graph; the relational network graph at least comprises user nodes representing users, object nodes representing behavior objects and connecting edges among the user nodes and the object nodes;
analyzing the user behavior log data to obtain an increment graph;
merging the first subgraph and the increment graph to obtain a merged subgraph;
and determining an effective path formed by at least one behavior in the plurality of preamble behaviors and the target behavior by carrying out graph calculation on the merged subgraph.
2. The method of claim 1, wherein the delta graph includes a first user node corresponding to the first user, a number of object nodes corresponding to behavior objects of the number of preceding behaviors, a behavior node corresponding to the target behavior, a first connecting edge between the first user node and the number of object nodes, and a second connecting edge between the first user node and the behavior node.
3. The method of claim 1, wherein the user behavior log data is derived by performing a buried point analysis on behavior objects in the client.
4. The method of claim 1, wherein the target behavior comprises a trading behavior or a downloading behavior;
the plurality of preorder behaviors comprise at least one behavior of clicking, browsing and collecting.
5. The method of claim 1, wherein the behavior information includes information of a behavior object and time information of behavior occurrence; the connection edge has corresponding edge attribute information, and the edge attribute information at least includes time information of behavior occurrence.
6. The method of claim 1, wherein the relational network graph further includes behavior nodes corresponding to the target behaviors, and connection edges between user nodes and the behavior nodes.
7. The method of claim 1, wherein the extracting the first sub-graph corresponding to the first user from the pre-established relationship network graph comprises:
searching a first user node representing the first user from the relational network graph;
and extracting the first user node, the object node with a connecting edge with the first user node and the connecting edge between the first user node and the object node from the relational network graph to serve as the first subgraph.
8. The method of claim 1, wherein the method further comprises:
and updating the relational network graph according to the increment graph.
9. The method of claim 8, wherein the updating the relational network graph comprises:
if the object node included in the incremental graph does not exist in the relational network graph and the user node included in the incremental graph exists in the relational network graph, adding a corresponding object node in the relational network graph and adding a connecting edge between the object node and the original user node in the relational network graph;
and if the user node and the object node included in the incremental graph exist in the relational network graph, adding a connecting edge between the object node and the user node in the relational network graph.
10. An apparatus for attribution analysis of user behavior, the apparatus being provided at a server and comprising:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring user behavior log data corresponding to a first user reported by a client; the user behavior log data at least comprises behavior information of a plurality of preorder behaviors and target behaviors which are executed in sequence;
the extracting unit is used for extracting a first sub-graph corresponding to the first user from a pre-established relationship network graph; the relational network graph at least comprises user nodes representing users, object nodes representing behavior objects and connecting edges among the user nodes and the object nodes;
the analysis unit is used for analyzing the user behavior log data acquired by the acquisition unit to obtain an increment graph;
the merging unit is used for merging the first subgraph obtained by the extracting unit and the increment graph obtained by the analyzing unit to obtain a merged subgraph;
and the graph calculation unit is used for performing graph calculation on the merged subgraph obtained by the merging unit and determining an effective path formed by at least one behavior in the plurality of preamble behaviors and the target behavior.
11. The apparatus of claim 10, wherein the delta graph comprises a first user node corresponding to the first user, a number of object nodes corresponding to behavior objects of the number of preceding behaviors, a behavior node corresponding to the target behavior, a first connecting edge between the first user node and the number of object nodes, and a second connecting edge between the first user node and the behavior node.
12. The apparatus of claim 10, wherein the user behavior log data is obtained by performing a buried point analysis on behavior objects in the client.
13. The apparatus of claim 10, wherein the target behavior comprises a trading behavior or a downloading behavior;
the plurality of preorder behaviors comprise at least one behavior of clicking, browsing and collecting.
14. The apparatus of claim 10, wherein the behavior information includes information of a behavior object and time information of occurrence of a behavior; the connection edge has corresponding edge attribute information, and the edge attribute information at least includes time information of behavior occurrence.
15. The apparatus of claim 10, wherein the relational network graph further comprises behavior nodes corresponding to the target behaviors, and connection edges between user nodes and the behavior nodes.
16. The apparatus of claim 10, wherein the extraction unit comprises:
a searching subunit, configured to search, from the relationship network graph, a first user node representing the first user;
and the extracting subunit is configured to extract, from the relationship network graph, the first user node found by the searching subunit, the object node having a connection edge with the first user node, and a connection edge therebetween, as the first sub-graph.
17. The apparatus of claim 10, wherein the apparatus further comprises:
and the updating unit is used for updating the relational network graph according to the increment graph.
18. The apparatus of claim 17, wherein the updating unit comprises:
a first updating subunit, configured to add, if an object node included in the incremental graph does not exist in the relational network graph and a user node included in the incremental graph exists in the relational network graph, a corresponding object node in the relational network graph and add a connection edge between the object node and an original user node in the relational network graph;
and a second updating subunit, configured to add, if both the user node and the object node included in the incremental graph exist in the relational network graph, a connecting edge between the object node and the user node in the relational network graph.
19. A computer-readable storage medium, on which a computer program is stored which, when executed in a computer, causes the computer to carry out the method of any one of claims 1-9.
20. A computing device comprising a memory having stored therein executable code and a processor that, when executing the executable code, implements the method of any of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210103677.2A CN114140031A (en) | 2022-01-28 | 2022-01-28 | Method and device for attribution analysis of user behaviors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210103677.2A CN114140031A (en) | 2022-01-28 | 2022-01-28 | Method and device for attribution analysis of user behaviors |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114140031A true CN114140031A (en) | 2022-03-04 |
Family
ID=80382027
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210103677.2A Pending CN114140031A (en) | 2022-01-28 | 2022-01-28 | Method and device for attribution analysis of user behaviors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114140031A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140068407A1 (en) * | 2012-08-28 | 2014-03-06 | Adobe System Incorporated | Identifying web pages that are likely to guide browsing viewers to improve conversion rate |
| CN108876479A (en) * | 2018-07-18 | 2018-11-23 | 口口相传(北京)网络技术有限公司 | The channel attribution method and device of object entity |
| CN110033315A (en) * | 2019-03-18 | 2019-07-19 | 北京品友互动信息技术股份公司 | The attribution method and device of advertising information conversion, storage medium, electronic device |
| CN111310061A (en) * | 2018-11-27 | 2020-06-19 | 百度在线网络技术（北京）有限公司 | Full-link multi-channel attribution method, device, server and storage medium |
| CN112000848A (en) * | 2020-08-20 | 2020-11-27 | 南京智慧图谱信息技术有限公司 | Graph data processing method and device, electronic equipment and storage medium |
| CN112000863A (en) * | 2020-08-14 | 2020-11-27 | 北京百度网讯科技有限公司 | User behavior data analysis method, device, equipment and medium |
-
2022
- 2022-01-28 CN CN202210103677.2A patent/CN114140031A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140068407A1 (en) * | 2012-08-28 | 2014-03-06 | Adobe System Incorporated | Identifying web pages that are likely to guide browsing viewers to improve conversion rate |
| CN108876479A (en) * | 2018-07-18 | 2018-11-23 | 口口相传(北京)网络技术有限公司 | The channel attribution method and device of object entity |
| CN111310061A (en) * | 2018-11-27 | 2020-06-19 | 百度在线网络技术（北京）有限公司 | Full-link multi-channel attribution method, device, server and storage medium |
| CN110033315A (en) * | 2019-03-18 | 2019-07-19 | 北京品友互动信息技术股份公司 | The attribution method and device of advertising information conversion, storage medium, electronic device |
| CN112000863A (en) * | 2020-08-14 | 2020-11-27 | 北京百度网讯科技有限公司 | User behavior data analysis method, device, equipment and medium |
| CN112000848A (en) * | 2020-08-20 | 2020-11-27 | 南京智慧图谱信息技术有限公司 | Graph data processing method and device, electronic equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 刘宝锺: "《大数据分类模型和算法研究》", 31 January 2020, 云南大学出版社 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Haupt et al. | A framework for the structural analysis of REST APIs | |
| CN111339427B (en) | Book information recommendation method, device and system and storage medium | |
| CN110944048B (en) | Service logic configuration method and device | |
| CN106294614A (en) | Method and apparatus for access service | |
| CN105446706B (en) | Method and device for evaluating form page use effect and providing original data | |
| Fayzrakhmanov et al. | Browserless web data extraction: challenges and opportunities | |
| CN111432003B (en) | Data pushing method and device applied to cloud computing, electronic equipment and system | |
| CN103440199A (en) | Method and device for guiding test | |
| CN113689270B (en) | Method for determining black product device, electronic device, storage medium, and program product | |
| CN114139022B (en) | Subgraph extraction method and device | |
| CN106202440B (en) | Data processing method, device and equipment | |
| CN113761219A (en) | Knowledge graph-based retrieval method and device, electronic equipment and storage medium | |
| US9324026B2 (en) | Hierarchical latent variable model estimation device, hierarchical latent variable model estimation method, supply amount prediction device, supply amount prediction method, and recording medium | |
| CN115174231A (en) | AI-Knowledge-Base-based network fraud analysis method and server | |
| US8368698B2 (en) | Calculating a webpage importance from a web browsing graph | |
| CN114140031A (en) | Method and device for attribution analysis of user behaviors | |
| KR20230095866A (en) | Method, apparatus, and program for providing business analysis information to provide benchmark information through collecting and analyzing data based on automatization | |
| CN112182320B (en) | Cluster data processing method, device, computer equipment and storage medium | |
| CN113590667A (en) | Real-time data updating and managing method based on Spark Streaming | |
| CN114579826A (en) | Task processing method and device based on knowledge graph | |
| CN115455426A (en) | Business error analysis method based on vulnerability analysis model development and cloud AI system | |
| CN113468604A (en) | Big data privacy information analysis method and system based on artificial intelligence | |
| CN114781517A (en) | Risk identification method and device and terminal equipment | |
| CN113946717A (en) | Sub-map index feature obtaining method, device, equipment and storage medium | |
| CN110597216A (en) | Transaction arrangement method and device for manufacturing execution system and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220304 |
|
| RJ01 | Rejection of invention patent application after publication |