CN113177079A

CN113177079A - Interactive event updating method based on artificial intelligence and cloud computing interactive center

Info

Publication number: CN113177079A
Application number: CN202110519386.7A
Authority: CN
Inventors: 崔秀芬
Original assignee: Individual
Current assignee: Individual
Priority date: 2020-12-04
Filing date: 2020-12-04
Publication date: 2021-07-27
Also published as: CN113190597A; CN112434086B; CN112434086A

Abstract

The embodiment of the application provides an interactive event updating method based on artificial intelligence and a cloud computing interactive center, after interactive event updating information of an interactive window track is obtained, the subsequent operation of pushing business information is carried out on the interactive event updating information based on the interactive window track, therefore, the target interactive behavior migration characteristic of the interactive window track and the interactive content migration characteristic information of the interactive window track are synthesized through interactive linkage events, the independent movable window information of the interactive window track and the interactive content migration characteristic information of the interactive window track are integrated, rich business relation characteristic information of the interactive window track is extracted, and data support is provided for accurate interactive mining; in addition, interactive event updating is carried out on the interactive window track through the target interactive behavior migration characteristic of the interactive window track, and interactive event updating information of the interactive window track is obtained, so that an accurate interactive mining process is realized.

Description

Interactive event updating method based on artificial intelligence and cloud computing interactive center

The application is a divisional application of Chinese application with the name of 'information flow mining method based on cloud computing and big data and cloud computing interaction center' invented and created by application number 202011417365.6 and application date of 04.12.2020.

Technical Field

The application relates to the technical field of data mining based on cloud computing, in particular to an interactive event updating method based on artificial intelligence and a cloud computing interactive center.

Background

Cloud computing (cloud computing) is one type of distributed computing, and means that a huge data computing processing program is decomposed into countless small programs through a network "cloud", and then the small programs are processed and analyzed through a system consisting of a plurality of servers to obtain results and are returned to a user. In the early stage of cloud computing, simple distributed computing is adopted, task distribution is solved, and computing results are merged. Thus, cloud computing is also known as grid computing. By the technology, tens of thousands of data can be processed in a short time (several seconds), so that strong network service is achieved.

With the continuous development of high-speed internet technology and audio-video technology, multi-platform multi-object internet interaction is more and more popular, and information stream interaction processing by adopting online cloud computing is more and more common. The internet interactive video is a novel video which integrates interactive experience into a linearly played video through various technical means, and the played multi-object interactive information stream is expanded and displayed on the cloud platform, and personalized interactive function options are configured for a user, so that personalized watching requirements of different audiences can be met.

In the related art, the generation of the interactive event verification stream can be performed on the multi-object interactive information stream, so that the behavior portrait basis formed by the user in the interactive process is summarized, and the subsequent function improvement is facilitated. However, the inventor of the present application researches and discovers that in the process of carrying out intention mining on an interactive event verification stream on a multi-object interactive information stream, the generated interactive intention represents that the mining information has the problem of single interactive intention path.

Disclosure of Invention

In order to overcome at least the above-mentioned deficiencies in the prior art, the present application aims to provide an interactive event updating method and a cloud computing interactive center based on artificial intelligence, wherein interactive intention path information for guiding an interactive intention path structure of interactive intention representation mining information to be generated is obtained based on interactive intention path characteristics of an interactive event verification stream, then intention subject information of the interactive intention representation mining information to be generated corresponding to the interactive intention path structure indicated by the interactive intention path characteristics is determined according to the interactive intention path information and the interactive event characteristics of the interactive event verification stream, and finally the interactive intention representation mining information is generated for the interactive event verification stream according to the intention subject information, thereby not only ensuring that the generated interactive intention representation mining information is similar to the interactive intention path structure of the interactive event verification stream, but also ensuring that the generated interactive intention representation mining information is related to the intention subject of the interactive event verification stream, that is, the interaction intention indicates that mining information can accurately describe the content in the interaction event verification stream. Because the interaction intention path of the generated interaction intention representation mining information is controlled by the interaction intention path characteristics of the interaction event verification flow, for the same interaction event verification flow, if the interaction event verification flows with different interaction intention path structures are selected to constrain the interaction intention path structures of the interaction intention representation mining information, the interaction intention representation mining information with different interaction intention path structures can be generated, so that the interaction intention representation mining information with different interaction intention path structures can be generated for the same interaction event verification flow by changing the interaction event verification flow, so that diversified interaction intention representation mining information can be generated for the interaction event verification flow, and the problem that the interaction intention path of the interaction intention representation mining information in the prior art is single is effectively solved.

In a first aspect, the present application provides an information stream mining method based on cloud computing and big data, which is applied to a cloud computing interaction center, where the cloud computing interaction center is in communication connection with a plurality of information stream node terminals, and the method includes:

acquiring an interactive event verification stream confirmed by the information flow node terminal, and acquiring interactive intention path characteristics of the interactive event verification stream;

determining an interaction intention path of the generated interaction intention representing mining information according to the interaction intention path characteristics to obtain interaction intention path information;

determining the interaction intention representation mining information to be generated to correspond to the intention theme of the interaction intention path according to the interaction intention path information and the interaction event characteristics of the interaction event verification stream to obtain intention theme information;

and generating interaction intention representation mining information of the interaction event verification stream according to the intention topic information.

In a possible implementation manner of the first aspect, the determining, according to the interaction intention path feature, an interaction intention path in which an interaction intention is to be generated to represent mining information to obtain interaction intention path information includes:

generating, by a first intention knowledge unit included in an interaction intention knowledge network, first intention classification information indicating the interaction intention path information from the interaction intention path features, the interaction intention knowledge network further including a second intention knowledge unit, the first and second intention knowledge units being artificial intelligence based cyclic deep neural network units;

determining that the mining information of the interaction intention representation to be generated corresponds to the intention theme of the interaction intention path according to the interaction intention path information and the interaction event characteristics of the interaction event verification stream to obtain intention theme information, wherein the method comprises the following steps:

generating, by the second intention knowledge unit, second intention classification information indicating the intention topic information according to the first intention classification information and the interaction event feature.

In a possible implementation manner of the first aspect, the generating interaction intention representation mining information of the interaction event verification stream according to the intention topic information includes:

determining intention classification components of the n knowledge rule nodes according to second intention classification information generated by the second intention knowledge unit at the n knowledge rule nodes;

generating the interaction intention representation mining information according to intention classification components output by all knowledge rule nodes;

the first intention knowledge unit contained by the interaction intention knowledge network generates first intention classification information according to the interaction intention path characteristics, and the first intention classification information comprises the following steps:

carrying out weighted calculation on the interaction intention path characteristics according to the first intention classification information of the n-1 knowledge rule nodes to obtain target interaction intention path characteristics corresponding to the n knowledge rule nodes;

fusing the target interaction intention path characteristics corresponding to the n knowledge rule nodes with intention classification components of the n-1 knowledge rule nodes to obtain first fusion components corresponding to the n knowledge rule nodes;

correspondingly outputting first intention classification information of the n knowledge rule nodes by the first intention knowledge unit by taking the first fusion components corresponding to the n knowledge rule nodes as input;

generating, by the second intent knowledge unit, second intent classification information from the first intent classification information and the interactivity event features, comprising:

performing weighted calculation on the interactive event characteristics according to the second intention classification information of the n-1 knowledge rule nodes to obtain an intention theme vector of the interactive event verification stream corresponding to the n knowledge rule nodes;

fusing the intention topic vector of the interactive event verification stream corresponding to the n knowledge rule nodes with the first intention classification information of the n knowledge rule nodes to obtain second fusion components corresponding to the n knowledge rule nodes;

and correspondingly outputting second intention classification information of the n knowledge rule nodes by the second intention knowledge unit by taking the second fusion components corresponding to the n knowledge rule nodes as input.

In a possible implementation manner of the first aspect, the first intention knowledge unit includes a first connection layer, a first feature extraction layer, and a first classification layer, and the outputting, by the first intention knowledge unit, first intention classification information of n knowledge rule nodes in a corresponding manner with the first fused component corresponding to the n knowledge rule nodes as an input includes:

calculating by the first feature extraction layer according to the first fusion component corresponding to the n knowledge rule nodes to obtain first feature extraction information of the n knowledge rule nodes, and calculating by the first connection layer according to the first fusion component corresponding to the n knowledge rule nodes to obtain first feature connection information of the n knowledge rule nodes;

calculating according to first feature extraction information of the n knowledge rule nodes, first feature connection information of the n knowledge rule nodes, first dependency transition features of the n knowledge rule nodes and first target dependency transition features of n-1 knowledge rule nodes corresponding to the first intention knowledge unit to obtain first target dependency transition features of the n knowledge rule nodes, wherein the first dependency transition features of the n knowledge rule nodes are obtained by performing service mining according to the first fusion components corresponding to the n knowledge rule nodes, and the dependency transition features are used for representing intention transition feature information with dependency relationship;

calculating first intention classification information of the n knowledge rule nodes according to first target dependency transition characteristics of the n knowledge rule nodes and first feature classification information of the n knowledge rule nodes, wherein the first feature classification information of the n knowledge rule nodes is calculated by the first classification layer according to the first fusion components corresponding to the n knowledge rule nodes;

before the first target dependent transition characteristics of the n knowledge rule nodes are obtained by calculating according to the first characteristic extraction information of the n knowledge rule nodes, the first characteristic connection information of the n knowledge rule nodes, the first dependent transition characteristics of the n knowledge rule nodes and the first target dependent transition characteristics of the n-1 knowledge rule nodes corresponding to the first intention knowledge unit, the method further includes:

respectively normalizing first feature connection information, first feature extraction information, first feature classification information and first dependency transition features in the first intention knowledge unit;

respectively transforming the normalized first feature connection information, the normalized first feature extraction information, the normalized first feature classification information and the normalized first dependent transition feature according to a first mining template and first structural adjustment information to obtain target first feature connection information, target first feature extraction information, target first feature classification information and target first dependent transition feature, wherein the first mining template is output by a first mining control according to the target interaction intention path feature corresponding to n knowledge rule nodes, the first structural adjustment information is output by a second mining control according to the target interaction intention path feature corresponding to n knowledge rule nodes, and the first mining control is independent of the second mining control;

the calculating the first target dependent transition characteristics of the n knowledge rule nodes according to the first characteristic extraction information of the n knowledge rule nodes, the first characteristic connection information of the n knowledge rule nodes, the first dependent transition characteristics of the n knowledge rule nodes and the first target dependent transition characteristics of the n-1 knowledge rule nodes corresponding to the first intention knowledge unit includes:

and calculating to obtain first target dependent transition characteristics of the n knowledge rule nodes according to the target first characteristic extraction information, the target first characteristic connection information, the target first dependent transition characteristics and the first target dependent transition characteristics of the n-1 knowledge rule nodes.

In a possible implementation manner of the first aspect, the second intention knowledge unit includes a second connection layer, a second feature extraction layer, and a second classification layer, and the outputting, by the second intention knowledge unit, second intention classification information of the n knowledge rule nodes in a corresponding manner with the second fused component corresponding to the n knowledge rule nodes as an input includes:

calculating by the second feature extraction layer according to the second fusion components corresponding to the n knowledge rule nodes to obtain second feature extraction information of the n knowledge rule nodes; calculating by the second connection layer according to the second fusion components corresponding to the n knowledge rule nodes to obtain second characteristic connection information of the n knowledge rule nodes;

calculating according to second feature extraction information of the n knowledge rule nodes, second feature connection information of the n knowledge rule nodes, second dependency transition features of the n knowledge rule nodes and second target dependency transition features of n-1 knowledge rule nodes corresponding to the second intention knowledge unit to obtain second target dependency transition features of the n knowledge rule nodes, wherein the second dependency transition features of the n knowledge rule nodes are obtained by performing service mining according to the second fusion components corresponding to the n knowledge rule nodes;

calculating second intention classification information of the n knowledge rule nodes according to second target dependency transition characteristics of the n knowledge rule nodes and second feature classification information of the n knowledge rule nodes, wherein the second feature classification information of the n knowledge rule nodes is calculated by the second classification layer according to the second fusion components corresponding to the n knowledge rule nodes;

before the calculating according to the second feature extraction information of the n knowledge rule nodes, the second feature connection information of the n knowledge rule nodes, the second dependency transition features of the n knowledge rule nodes, and the second target dependency transition features of the n-1 knowledge rule nodes corresponding to the second intention knowledge unit to obtain the second target dependency transition features of the n knowledge rule nodes, the method further includes:

respectively normalizing second feature connection information, second feature extraction information, second feature classification information and second dependency transition features in the second intention knowledge unit;

respectively transforming the normalized second feature connection information, the normalized second feature extraction information, the normalized second feature classification information and the normalized second dependent transition feature according to a second mining template and second structured adjustment information to obtain target second feature connection information, target second feature extraction information, target second feature classification information and target second dependent transition feature, wherein the second mining template is output by a third mining control according to the intention theme vector of the interactive event verification stream corresponding to the n knowledge rule nodes, the second structured adjustment information is output by a fourth mining control according to the intention theme vector of the interactive event verification stream corresponding to the n knowledge rule nodes, and the third mining control is independent of the fourth mining control;

the calculating the second target-dependent transition characteristics of the n knowledge rule nodes according to the second characteristic extraction information of the n knowledge rule nodes, the second characteristic connection information of the n knowledge rule nodes, the second dependent transition characteristics of the n knowledge rule nodes and the second target-dependent transition characteristics of the n-1 knowledge rule nodes corresponding to the second intention knowledge unit includes:

calculating according to the target second feature extraction information, the target second feature connection information, the target second dependency transition feature and the second target dependency transition feature of the n-1 knowledge rule nodes to obtain second target dependency transition features of the n knowledge rule nodes;

the calculating to obtain second intention classification information of the n knowledge rule nodes according to the second target dependency transition characteristics of the n knowledge rule nodes and the second characteristic classification information of the n knowledge rule nodes includes:

and calculating second intention classification information of the n knowledge rule nodes according to the second target dependency transition characteristics of the n knowledge rule nodes and the target second characteristic classification information.

In a possible implementation manner of the first aspect, the obtaining the interaction intention path feature of the interaction event verification stream includes:

acquiring interactive migration process result information of an interactive migration process included by each interactive event object in the interactive event verification stream, wherein the interactive migration process result information is obtained by tracking the interactive migration process;

outputting third intention classification information corresponding to each interactive migration process by a third intention knowledge unit according to the interactive migration process result information of each interactive migration process;

calculating each interactive event object in the interactive event verification stream according to third intention classification information corresponding to each interactive migration process in the interactive event object to obtain result information of the interactive event object;

and outputting a fourth intention classification information sequence by a fourth intention knowledge unit according to result information of each interactive event object in the interactive event verification stream, wherein the fourth intention classification information sequence is used as the interactive intention path characteristic, and the third intention knowledge unit and the fourth intention knowledge unit are artificial intelligence based cyclic depth neural network units.

For example, in a possible implementation manner of the first aspect, the method further includes:

acquiring a training sample set, wherein the training sample set comprises a plurality of interactive event verification stream samples and mining information samples corresponding to the interactive event verification stream samples;

extracting intention theme characteristics of the interactive event verification stream sample to obtain sample interactive event characteristics of the interactive event verification stream sample, and extracting interaction intention path characteristics of the mining information sample corresponding to the interactive event verification stream sample to obtain sample interaction intention path characteristics of the mining information sample;

outputting a first intention classification information sequence by the first intention knowledge unit according to the sample interaction intention path characteristics, and classifying the mining information samples to obtain a classification result by a sixth intention knowledge unit according to the first intention classification information sequence, wherein the sixth intention knowledge unit is a cyclic deep neural network unit based on artificial intelligence;

calculating to obtain interaction intention path difference according to the classified classification result and the actual classification result of the mining information sample;

outputting, by the second intention knowledge unit, a second intention classification information sequence according to the first intention classification information sequence and sample interactivity event features of the interactivity event verification stream sample, and outputting, by a fifth mining control, target mining information for the interactivity event verification stream sample according to the second intention classification information sequence;

calculating to obtain the intention theme difference according to the target mining information and mining information samples corresponding to the interactive event verification stream samples;

calculating to obtain a target difference according to the interaction intention path difference and the intention theme difference;

adjusting parameters of the interaction intention knowledge network based on the target difference.

In a possible implementation manner of the first aspect, the interactivity event characteristics of the interactivity event verification stream are obtained by:

acquiring a segmentation interactive event sequence obtained by performing interactive event segmentation on the interactive event verification stream;

and extracting event characteristic information matched with a preset event characteristic template in the segmentation interactive events as the interactive event characteristics of the interactive event verification stream aiming at each segmentation interactive event in the segmentation interactive event sequence.

In a possible implementation manner of the first aspect, the step of obtaining the interactivity event verification stream confirmed by the information flow node terminal includes:

acquiring interactive event updating information of an interactive window track in the multi-object interactive information stream of the information stream node terminal;

acquiring a drawing and loading element to be simulated which is matched with a plurality of interaction event fragments to be loaded and a target simulation drawing control corresponding to the drawing and loading element to be simulated based on the interaction event update information, wherein the target simulation drawing control is a simulation drawing control which is served by an interaction component to which the event loading information of the drawing and loading element to be simulated belongs, and the target simulation drawing control comprises at least one control drawing object;

screening and matching a plurality of interaction event fragments to be loaded to obtain a target loading interaction event having a drawing association relation with at least one control drawing object, and generating loading drawing control information between the target loading interaction event and the target control drawing object according to drawing parameters of the target loading interaction event and the at least one control drawing object under a target drawing attribute category;

and inputting loading drawing control information between the target loading interactive event and a target control drawing object under each drawing attribute category in each target simulation drawing control, selecting a target simulation drawing resource matched with the drawing loading element to be simulated from a preset target simulation drawing resource set according to an input result, and pushing an interactive event verification stream of the target simulation drawing resource to the information flow node terminal, so that the interactive event verification stream is used for information mining of a user of the information flow node terminal after the information flow node terminal verifies and confirms the interactive event verification stream.

In a possible implementation manner of the first aspect, the obtaining of the interaction event update information of the interaction window trajectory in the multi-object interaction information stream of the information stream node terminal includes:

acquiring a multi-object interactive information stream of the information stream node terminal, and performing independent movable window extraction processing on the multi-object interactive information stream to obtain independent movable window information of interactive window tracks in the multi-object interactive information stream, wherein the multi-object interactive information stream is an interactive information stream formed by object interactive information recorded by each interactive window track acquired based on a single interactive request;

carrying out interactive behavior tracking extraction based on the independent movable window information of the interactive window track to obtain target interactive behavior migration characteristics of the interactive window track;

extracting interactive content migration characteristics of the multi-object interactive information stream based on an artificial intelligence model to obtain interactive content migration characteristic information of the interactive window track;

and performing interactive linkage event synthesis on the target interactive behavior migration characteristic of the interactive window track in the multi-object interactive information stream and the interactive content migration characteristic information of the interactive window track to obtain interactive linkage event synthesis information of the interactive window track, and performing interactive event update on the interactive event record control of the multi-object interactive information stream based on the interactive linkage event synthesis information of the interactive window track to obtain interactive event update information of the interactive window track.

In a second aspect, an embodiment of the present application further provides an information stream mining device based on cloud computing and big data, which is applied to a cloud computing interaction center, where the cloud computing interaction center is in communication connection with a plurality of information stream node terminals, and the device includes:

an obtaining module, configured to obtain an interactive event verification stream confirmed by the information flow node terminal, and obtain an interaction intention path feature of the interactive event verification stream;

the first determining module is used for determining an interaction intention path of the generated interaction intention representing mining information according to the interaction intention path characteristics to obtain interaction intention path information;

a second determining module, configured to determine, according to the interaction intention path information and the interaction event feature of the interaction event verification stream, an intention theme of which the to-be-generated interaction intention representation mining information corresponds to the interaction intention path, so as to obtain intention theme information;

and the generating module is used for generating the interaction intention representation mining information of the interaction event verification stream according to the intention topic information.

In a third aspect, an embodiment of the present application further provides an information flow mining system based on cloud computing and big data, where the information flow mining system based on cloud computing and big data includes a cloud computing interaction center and a plurality of information flow node terminals in communication connection with the cloud computing interaction center;

the cloud computing interaction center is used for:

In a fourth aspect, an embodiment of the present application further provides a cloud computing interaction center, where the cloud computing interaction center includes a processor, a machine-readable storage medium, and a network interface, where the machine-readable storage medium, the network interface, and the processor are connected by a bus system, the network interface is used for being in communication connection with at least one information flow node terminal, the machine-readable storage medium is used for storing a program, an instruction, or a code, and the processor is used for executing the program, the instruction, or the code in the machine-readable storage medium to execute the cloud computing and big data-based information flow mining method in the first aspect or any one of possible implementation manners in the first aspect.

In a fifth aspect, an embodiment of the present application provides a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, and when the instructions are executed, the computer is caused to execute the cloud computing and big data based information stream mining method in the first aspect or any one of the possible implementation manners of the first aspect.

Based on any one of the above aspects, the application first obtains the interaction intention path information for guiding the interaction intention path structure of the generated interaction intention representation mining information based on the interaction intention path characteristics of the interaction event verification flow, then, according to the interaction intention path information and the interaction event characteristics of the interaction event verification flow, the intention theme information of the interaction intention expression mining information to be generated, which corresponds to the interaction intention path structure indicated by the interaction intention path characteristics, is determined, and finally, the interaction intention expression mining information is generated for the interaction event verification flow according to the intention theme information, so that the generated interaction intention expression mining information is ensured to be similar to the interaction intention path structure of the interaction event verification flow, but also ensures that the generated interaction intention representation mining information is related to the intention topic of the interaction event verification stream, that is, the interaction intention indicates that mining information can accurately describe the content in the interaction event verification stream. Because the interaction intention path of the generated interaction intention representation mining information is controlled by the interaction intention path characteristics of the interaction event verification flow, for the same interaction event verification flow, if the interaction event verification flows with different interaction intention path structures are selected to constrain the interaction intention path structures of the interaction intention representation mining information, the interaction intention representation mining information with different interaction intention path structures can be generated, so that the interaction intention representation mining information with different interaction intention path structures can be generated for the same interaction event verification flow by changing the interaction event verification flow, so that diversified interaction intention representation mining information can be generated for the interaction event verification flow, and the problem that the interaction intention path of the interaction intention representation mining information in the prior art is single is effectively solved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that need to be called in the embodiments are briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.

Fig. 1 is a schematic view of an application scenario of an information flow mining system based on cloud computing and big data according to an embodiment of the present application;

fig. 2 is a schematic flowchart of an information flow mining method based on cloud computing and big data according to an embodiment of the present application;

fig. 3 is a functional module schematic diagram of an information flow mining device based on cloud computing and big data according to an embodiment of the present application;

fig. 4 is a schematic block diagram of structural components of a cloud computing interaction center for implementing the above information flow mining method based on cloud computing and big data according to the embodiment of the present application.

Detailed Description

The present application will now be described in detail with reference to the drawings, and the specific operations in the method embodiments may also be applied to the apparatus embodiments or the system embodiments.

Fig. 1 is an interaction diagram of an information flow mining system 10 based on cloud computing and big data according to an embodiment of the present application. The cloud computing and big data based information stream mining system 10 may include a cloud computing interaction center 100 and an information stream node terminal 200 communicatively connected to the cloud computing interaction center 100. The cloud computing and big data based information flow mining system 10 shown in fig. 1 is only one possible example, and in other possible embodiments, the cloud computing and big data based information flow mining system 10 may also include only a portion of the components shown in fig. 1 or may also include other components.

Based on the inventive concept of the technical solution provided by the present application, the cloud computing interaction center 100 provided by the present application can be applied to scenes such as smart medical, smart city management, smart industrial internet, general service monitoring management, and the like, in which a big data technology or a cloud computing technology can be applied, and for example, the cloud computing interaction center can also be applied to scenes such as but not limited to new energy automobile system management, smart cloud office, cloud platform data processing, cloud game data processing, cloud live broadcast processing, cloud automobile management platform, block chain financial data service platform, and the like, but is not limited thereto.

In this embodiment, the cloud computing interaction center 100 and the information flow node terminal 200 in the information flow mining system 10 based on cloud computing and big data may cooperatively execute the information flow mining method based on cloud computing and big data described in the following method embodiment, and the detailed description of the following method embodiment may be referred to in the specific steps executed by the cloud computing interaction center 100 and the information flow node terminal 200.

In order to solve the technical problem in the foregoing background art, fig. 2 is a schematic flowchart of an information flow mining method based on cloud computing and big data according to an embodiment of the present application, where the information flow mining method based on cloud computing and big data according to the present embodiment may be executed by the cloud computing interaction center 100 shown in fig. 1, and the information flow mining method based on cloud computing and big data is described in detail below.

Step S110, acquiring the interactive event verification stream confirmed by the information flow node terminal 200, and acquiring the interactive intention path feature of the interactive event verification stream.

And step S120, determining an interaction intention path of the generated interaction intention representing mining information according to the interaction intention path characteristics to obtain interaction intention path information.

Step S130, according to the interaction intention path information and the interaction event characteristics of the interaction event verification stream, determining an intention theme of which the interaction intention representation mining information to be generated corresponds to the interaction intention path, and obtaining intention theme information.

Step S140, generating the interactive intention expression mining information of the interactive event verification flow according to the intention topic information.

In this embodiment, the interactive event verification stream may refer to an interactive event verification stream confirmed by the information flow node terminal 200 after the multi-object interactive information stream is pushed, and may summarize a behavior portrait basis formed by a user in an interactive process, for example, behavior big data record information under each interactive event.

In this embodiment, the interaction intention path may refer to a path formed by a change in each interaction intention formed by the user in the interaction process. For example, after the user initiates the live interaction behavior of the live e-commerce commodity a during the live e-commerce interaction, the user initiates response interactions with various intentions (including but not limited to purchase, collection, subscription, for example) on the live interaction behavior of the live e-commerce commodity a, and the path process of one interaction intention can be understood.

In this embodiment, the interactivity event features may be used to represent certain specific features of the interactivity event verification stream, which will be described in detail later.

In this embodiment, the intention topic may be used to represent topic attributes summarized by various interaction intentions formed by the user in the interaction process, for example, when a response interaction (including, but not limited to, purchase, collection, subscription, for example) with multiple intentions for a certain clothing category topic is initiated for a live interaction behavior of a certain clothing tag type, it may be understood that the intention topic is a certain clothing category topic under the clothing tag type.

In this embodiment, after the intention topic information is obtained, the interaction intention representation mining information of the interaction event verification stream may be generated. For example, the intention topic information may be used to represent a reference basis for information pushing for the user subsequently, so that the interaction intention expression mining information of the interaction event verification stream may be composed of the intention topic information, and in some alternative implementations, a weighted value may be set in advance for the intention topic information according to a topic subscribed by the user, for example, a higher weighted value may be set for the topic subscribed by the user, and conversely, a slightly lower weighted value may be set, and the specifically set weighted value may be selected according to actual design requirements, which is not limited in this embodiment.

Based on the above steps, the present embodiment first obtains the interaction intention path information for guiding the interaction intention path structure of the generated interaction intention representation mining information based on the interaction intention path characteristics of the interaction event verification stream, then, according to the interaction intention path information and the interaction event characteristics of the interaction event verification flow, the intention theme information of the interaction intention expression mining information to be generated, which corresponds to the interaction intention path structure indicated by the interaction intention path characteristics, is determined, and finally, the interaction intention expression mining information is generated for the interaction event verification flow according to the intention theme information, so that the generated interaction intention expression mining information is ensured to be similar to the interaction intention path structure of the interaction event verification flow, but also ensures that the generated interaction intention representation mining information is related to the intention topic of the interaction event verification stream, that is, the interaction intention indicates that mining information can accurately describe the content in the interaction event verification stream. Because the interaction intention path of the generated interaction intention representation mining information is controlled by the interaction intention path characteristics of the interaction event verification flow, for the same interaction event verification flow, if the interaction event verification flows with different interaction intention path structures are selected to constrain the interaction intention path structures of the interaction intention representation mining information, the interaction intention representation mining information with different interaction intention path structures can be generated, so that the interaction intention representation mining information with different interaction intention path structures can be generated for the same interaction event verification flow by changing the interaction event verification flow, so that diversified interaction intention representation mining information can be generated for the interaction event verification flow, and the problem that the interaction intention path of the interaction intention representation mining information in the prior art is single is effectively solved.

In one possible implementation manner, with respect to step S120, in the process of determining an interaction intention path to generate interaction intention representation mining information according to the interaction intention path feature to obtain the interaction intention path information, the first intention knowledge unit included in the interaction intention knowledge network may generate the first intention classification information according to the interaction intention path feature.

It should be noted that the first intention classification information is used to indicate interaction intention path information, the interaction intention knowledge network further includes a second intention knowledge unit, and the first intention knowledge unit and the second intention knowledge unit are artificial intelligence based cyclic deep neural network units, and may be used to perform mining of intention knowledge points, for example, information classification and prediction operations may be performed after learning a large amount of feature information.

In this way, with respect to step S130, in the process of determining the intention topic of the interaction intention representation mining information to be generated corresponding to the interaction intention path according to the interaction intention path information and the interaction event characteristics of the interaction event verification stream, and obtaining the intention topic information, the second intention knowledge unit may generate second intention classification information according to the first intention classification information and the interaction event characteristics, wherein the second intention classification information is used for indicating the intention topic information.

In one possible implementation, step S140 may be implemented by the following exemplary sub-steps, which are described in detail below.

And a substep S141 of determining intention classification components of the n knowledge rule nodes based on the second intention classification information generated at the n knowledge rule nodes by the second intention knowledge unit.

And a substep S142, generating interaction intention expression mining information according to the intention classification components output by the knowledge rule nodes.

Thus, step S120 may be implemented by the following exemplary substeps, described in detail below.

And a substep S121, carrying out weighted calculation on the interaction intention path characteristics according to the first intention classification information of the n-1 knowledge rule nodes to obtain target interaction intention path characteristics corresponding to the n knowledge rule nodes.

And a substep S122, fusing the target interaction intention path characteristics corresponding to the n knowledge rule nodes with the intention classification components of the n-1 knowledge rule nodes to obtain first fusion components corresponding to the n knowledge rule nodes.

In sub-step S123, the first intention knowledge unit outputs the first intention classification information of the n knowledge rule nodes with the first fusion components corresponding to the n knowledge rule nodes as input.

Thus, step S130 may be implemented by the following exemplary substeps, described in detail below.

And a substep S131, carrying out weighted calculation on the interactive event characteristics according to the second intention classification information of the n-1 knowledge rule nodes to obtain an intention topic vector of the interactive event verification stream corresponding to the n knowledge rule nodes.

And a substep S132, fusing the intention topic vector of the interactive event verification stream corresponding to the n knowledge rule nodes and the first intention classification information of the n knowledge rule nodes to obtain second fusion components corresponding to the n knowledge rule nodes.

The substep S133 outputs second intention classification information of the n knowledge rule nodes corresponding to the second intention knowledge unit with the second fusion components corresponding to the n knowledge rule nodes as input.

In one possible implementation, the first intention knowledge unit may include a first connection layer, a first feature extraction layer and a first classification layer, and in the sub-step S123, the following implementation may be implemented:

(1) and calculating by the first feature extraction layer first feature extraction information of the n knowledge rule nodes according to the first fusion components corresponding to the n knowledge rule nodes, and calculating by the first connection layer first feature connection information of the n knowledge rule nodes according to the first fusion components corresponding to the n knowledge rule nodes.

(2) And calculating according to the first feature extraction information of the n knowledge rule nodes, the first feature connection information of the n knowledge rule nodes, the first dependency transition feature of the n knowledge rule nodes and the first target dependency transition feature of the n-1 knowledge rule nodes corresponding to the first intention knowledge unit to obtain the first target dependency transition feature of the n knowledge rule nodes, wherein the first dependency transition feature of the n knowledge rule nodes is obtained by performing service mining according to the first fusion component corresponding to the n knowledge rule nodes.

In this embodiment, the dependency transition feature is used to indicate the intention transition feature information in which a dependency relationship exists.

(3) And calculating first intention classification information of the n knowledge rule nodes according to the first target dependency transition characteristics of the n knowledge rule nodes and the first feature classification information of the n knowledge rule nodes, wherein the first feature classification information of the n knowledge rule nodes is calculated by the first classification layer according to the first fusion components corresponding to the n knowledge rule nodes.

In an alternative embodiment, before the step (2) of the sub-step S123, the first feature connection information, the first feature extraction information, the first feature classification information, and the first dependent transition feature in the first intention knowledge unit may be normalized respectively.

On this basis, the normalized first feature connection information, the normalized first feature extraction information, the normalized first feature classification information and the normalized first dependent transition feature may be transformed according to the first mining template and the first structural adjustment information, so as to obtain the target first feature connection information, the target first feature extraction information, the target first feature classification information and the target first dependent transition feature.

In this embodiment, the first mining template is output by the first mining control according to the target interaction intention path features corresponding to the n knowledge rule nodes, the first structured adjustment information is output by the second mining control according to the target interaction intention path features corresponding to the n knowledge rule nodes, and the first mining control and the second mining control are independent from each other.

In this way, in the step (2) of the above substep S123, the first target dependent transition characteristics of the n knowledge rule nodes may be calculated based on the target first feature extraction information, the target first feature connection information, the target first dependent transition characteristics, and the first target dependent transition characteristics of the n-1 knowledge rule nodes.

In another possible implementation, the second intention knowledge unit may include a second connection layer, a second feature extraction layer and a second classification layer, and the sub-step S133 may be implemented by the following exemplary embodiments:

(1) and calculating by the second feature extraction layer according to the second fusion components corresponding to the n knowledge rule nodes to obtain second feature extraction information of the n knowledge rule nodes, and calculating by the second connection layer according to the second fusion components corresponding to the n knowledge rule nodes to obtain second feature connection information of the n knowledge rule nodes.

(2) And calculating according to the second feature extraction information of the n knowledge rule nodes, the second feature connection information of the n knowledge rule nodes, the second dependency transition features of the n knowledge rule nodes and the second target dependency transition features of the n-1 knowledge rule nodes corresponding to the second intention knowledge unit to obtain the second target dependency transition features of the n knowledge rule nodes.

In this embodiment, the second dependent transition characteristics of the n knowledge rule nodes are obtained by performing service mining according to the second fusion components corresponding to the n knowledge rule nodes.

(3) And calculating second intention classification information of the n knowledge rule nodes according to second target dependency transition characteristics of the n knowledge rule nodes and second feature classification information of the n knowledge rule nodes, wherein the second feature classification information of the n knowledge rule nodes is calculated by a second classification layer according to second fusion components corresponding to the n knowledge rule nodes.

In an alternative embodiment, before the step (2) of the sub-step S133, the second feature connection information, the second feature extraction information, the second feature classification information, and the second dependent transition feature in the second intention knowledge unit may be normalized respectively.

On the basis, the normalized second feature connection information, the normalized second feature extraction information, the normalized second feature classification information and the normalized second dependent transition feature can be transformed according to the second mining template and the normalized second structural adjustment information, so that the target second feature connection information, the target second feature extraction information, the target second feature classification information and the target second dependent transition feature are obtained.

In this embodiment, the second mining template is output by the third mining control according to the intention theme vector of the interactive event verification stream corresponding to the n knowledge rule nodes, the second structured adjustment information is output by the fourth mining control according to the intention theme vector of the interactive event verification stream corresponding to the n knowledge rule nodes, and the third mining control is independent of the fourth mining control.

In this way, in the above-mentioned substep S133 (2), the second target dependent transition characteristics of the n knowledge rule nodes may be calculated and obtained based on the target second feature extraction information, the target second feature connection information, the target second dependent transition characteristics, and the second target dependent transition characteristics of the n-1 knowledge rule nodes.

In the above-mentioned step (3) of the substep S133, second intention classification information of the n knowledge rule nodes may be calculated from the second target dependent transition feature and the target second feature classification information of the n knowledge rule nodes.

In one possible implementation manner, for step S110, in the process of obtaining the interaction intention path characteristic of the interaction event verification stream, the following exemplary embodiments may be implemented, and the following detailed description is given below.

And a substep S111, obtaining the interactive migration process result information of the interactive migration process included by each interactive event object in the interactive event verification stream.

In this embodiment, the result information of the interactive migration process may be obtained by tracking the interactive migration process.

And a substep S112, outputting third intention classification information corresponding to each interactive migration process by the third intention knowledge unit according to the interactive migration process result information of each interactive migration process.

And a substep S113, calculating each interactive event object in the interactive event verification stream according to the third intention classification information corresponding to each interactive migration process in the interactive event object to obtain result information of the interactive event object.

In sub-step S114, a fourth intention knowledge unit outputs a fourth intention classification information sequence according to the result information of each interactivity event object in the interactivity event verification stream.

In this embodiment, the fourth intention classification information sequence is used as the interactive intention path feature, and the third intention knowledge unit and the fourth intention knowledge unit are artificial intelligence based cyclic deep neural network units.

In one possible implementation, for step S130, the interactivity event features of the interactivity event verification stream may be obtained by:

(1) and acquiring a segmentation interactive event sequence obtained by performing interactive event segmentation on the interactive event verification stream.

(2) And extracting event characteristic information matched with a preset event characteristic template in the segmented interactive events as interactive event characteristics of the interactive event verification stream aiming at each segmented interactive event in the segmented interactive event sequence.

In this embodiment, the interactive event segmentation mode may specifically be that the start node and the end node of each interactive event are taken as a unit to perform segmentation, or any other feasible implementation mode to perform segmentation, which is not specifically limited herein. In addition, the preset event feature template may be selected or designed according to actual requirements, and is not specifically limited herein.

On the basis of the above description, the aforementioned interaction intention knowledge network may be obtained by:

(1) a set of training samples is obtained.

In this embodiment, the training sample set includes a plurality of interactive event verification stream samples and mining information samples corresponding to the interactive event verification stream samples.

(2) Extracting intention theme characteristics of the interactive event verification stream sample to obtain sample interactive event characteristics of the interactive event verification stream sample, and extracting interactive intention path characteristics of the mining information sample corresponding to the interactive event verification stream sample to obtain sample interactive intention path characteristics of the mining information sample.

(3) And the first intention knowledge unit outputs a first intention classification information sequence according to the characteristics of the sample interaction intention path, and the sixth intention knowledge unit classifies the mining information samples according to the first intention classification information sequence to obtain a classification result.

In this embodiment, the sixth intention knowledge unit is an artificial intelligence based cyclic deep neural network unit.

(4) And calculating to obtain the interaction intention path difference according to the classification result obtained by classification and the actual classification result of the mining information sample.

(5) And outputting a second intention classification information sequence by the second intention knowledge unit according to the first intention classification information sequence and the sample interactive event characteristics of the interactive event verification stream samples, and outputting target mining information for the interactive event verification stream samples according to the second intention classification information sequence through a fifth mining control.

(6) And calculating to obtain the intention theme difference according to the mining information samples corresponding to the target mining information and the interactive event verification stream samples.

(7) And calculating to obtain target difference according to the interaction intention path difference and the intention subject difference.

(8) Parameters of the interaction intention knowledge network are adjusted based on the target differences.

In a possible implementation manner, still referring to step S110, in the process of obtaining the interactivity event verification stream confirmed by the information flow node terminal 200, the following exemplary sub-steps can be implemented, which are described in detail below.

In the substep S101, the interaction event update information of the interaction window trajectory in the multi-object interaction information stream of the information stream node terminal 200 is obtained.

And step S102, acquiring a drawing loading element to be simulated and matched with a plurality of interaction event fragments to be loaded and a target simulation drawing control corresponding to the drawing loading element to be simulated based on the interaction event update information.

In this embodiment, the target simulation drawing control may be understood as a simulation drawing control of an interaction component service to which event loading information of a to-be-simulated drawing loading element belongs, where the target simulation drawing control may include at least one control drawing object. In an alternative implementation manner, the interaction event update information may have one or more interaction event segments to be loaded, where an interaction event segment to be loaded may be understood as a specific interaction process included in a loading interaction event (for example, an e-commerce live broadcast interaction process for a specific e-commerce commodity), and a loading interaction event may be understood as a complete interaction process (for example, a process set of e-commerce live broadcast interaction processes for a plurality of specific e-commerce commodities in an overall live broadcast process), where each interaction process may be an interaction process for a certain individual object or an interaction process of an interaction task composed of a plurality of individual objects.

In addition, the to-be-simulated drawing and loading elements matched with each to-be-loaded interactive event fragment may be obtained correspondingly from a current real-time simulation drawing and loading element library based on the to-be-loaded interactive event fragment, or obtained correspondingly from a pre-configured simulation drawing and loading element library, which is not limited specifically. The simulation drawing loading element may be understood as information such as scene description, voting problem, background information, social media interaction, and the like loaded during simulation drawing, and the target simulation drawing control corresponding to the drawing loading element to be simulated may be obtained based on a control parameter (for example, an SDK (software development kit interface) and the like) associated in advance with each drawing loading element to be simulated.

Step S103, the multiple interactive event fragments to be loaded are screened and matched to obtain a target loading interactive event with a drawing association relation with at least one control drawing object, and loading drawing control information between the target loading interactive event and the target control drawing object is generated according to drawing parameters of the target loading interactive event and the at least one control drawing object under the target drawing attribute category.

In this embodiment, the rendering parameters may be understood as rendering instructions when the target loading interaction event and the at least one control rendering object match the same rendering attribute in the target rendering attribute category, and the specific determination manner may refer to an existing common rendering attribute algorithm model. In addition, the load rendering control information may be used to represent control instructions for the target load interaction event and the target control rendering object to be load rendered.

Step S104, inputting the loading drawing control information between the target loading interaction event and the target control drawing object under each drawing attribute category in each target simulation drawing control, selecting a target simulation drawing resource matched with the drawing loading element to be simulated from a preset target simulation drawing resource set according to an input result, and pushing an interaction event verification stream of the target simulation drawing resource to the information flow node terminal 200, so that the interaction event verification stream is used for information mining of a user of the information flow node terminal 200 after the information flow node terminal 200 verifies and confirms the interaction event verification stream.

In this embodiment, the simulation rendering resource may be understood as a specific rendering resource that is finally pushed to the information flow node terminal 200, and may include, but is not limited to, a content prompt rendering resource (such as some special effect schemes), a commodity object rendering resource, a session rendering resource, and the like, for example, but is not limited thereto.

For example, in the process of selecting a target simulation drawing resource matched with a drawing load element to be simulated from a preset target simulation drawing resource set according to an input result and pushing an interaction event verification stream of the target simulation drawing resource to the information flow node terminal 200, a target simulation drawing load element matched with a loading interaction event fragment included in a target loading interaction event whose loading drawing control information is greater than preset loading drawing control information may be determined, and then after obtaining target simulation drawing resources corresponding to the target simulation drawing load elements from the preset target simulation drawing resource set, drawing the target simulation drawing resource to generate an interaction event verification stream, and pushing the interaction event verification stream of the target simulation drawing resource to the information flow node terminal 200.

Based on the above steps, after generating the loading rendering control information between the target loading interactivity event and the target control rendering object according to the rendering parameters of the target loading interactivity event and the target control rendering object under at least one control rendering object, by entering load rendering control information between the target load interaction event and the target control rendering object under each rendering attribute category into each of the target simulated rendering controls, a large number of reference bases based on the loading interaction events can be utilized to ensure that more target control drawing objects are obtained, which is beneficial to improving the accuracy of information flow matching of subsequent simulation drawing resources, and, when the loading interaction event is taken as the independent interaction processing unit, the situation of error occurring when the loading element to be simulated and drawn is subjected to simulation drawing can be avoided, and therefore the accuracy of information flow matching of the simulation drawing resource is improved.

In one possible implementation manner, for step S103, in the process of generating the load rendering control information between the target load interaction event and the target control rendering object according to the rendering parameters of the target load interaction event and the at least one control rendering object in the target rendering attribute category, the load rendering control information may be generated through the following exemplary sub-steps, which are described in detail below.

And step S1031, determining a target drawing attribute category corresponding to each control drawing object according to the existing drawing incidence relation between the target loading interaction event and the control drawing object.

And a substep S1032 of calling the drawing parameters of the target loading interaction event and at least one control drawing object in the determined target drawing attribute category based on the determined target drawing attribute category, and determining the control drawing object of which the drawing parameters meet the preset drawing service range as the target control drawing object.

And a substep S1033, generating loading rendering control information between the target loading interaction event and the target control rendering object according to the rendering parameters of the target loading interaction event and the target control rendering object under the at least one rendering attribute category.

For example, in sub-step S1032, it can be realized by the following three implementations.

And (I) calculating first drawing parameters of the target loading interaction event and at least one control drawing object in the same drawing attribute category, and determining the control drawing object with the first drawing parameters meeting a preset drawing service range as a first target control drawing object.

And (II) calculating a second drawing parameter of the target loading interaction event and at least one control drawing object under the hierarchy drawing attribute category, and determining the control drawing object of which the second drawing parameter meets a preset drawing service range as a second target control drawing object.

And (III) calculating a third drawing parameter of the target loading interaction event and at least one control drawing object in the partition drawing attribute category, and determining the control drawing object of which the third drawing parameter meets a preset drawing service range as a third target control drawing object.

In (one), a same-drawing property sequence may be selected in the target simulated drawing control, and the same-drawing property sequence may include a plurality of same-drawing property lists, each of which includes control drawing objects with at least two same drawing property description vectors. Then, a same drawing attribute list which has the same drawing attribute as the loading level drawing attribute of the target loading interaction event is determined, and a target same drawing attribute list is obtained. On the basis, a first drawing parameter between the target loading interaction event and each control drawing object in the target and drawing attribute list is called, and the control drawing object with the first drawing parameter meeting a preset drawing service range is determined to be a first target control drawing object.

Thus, in sub-step S1033, first load rendering control information between the target load interactivity event and the first target control rendering object may be generated based on the first rendering parameters between the target load interactivity event and the first target control rendering object.

Alternatively, in the second step (b), the hierarchical rendering property relationship between the target load interaction event and the at least one control rendering object may be determined according to the load hierarchical rendering property of the target load interaction event and the load hierarchical rendering property of each control rendering object. And then, based on the determined hierarchy drawing attribute relationship, calling a second drawing parameter between the target loading interaction event and the corresponding upper control drawing object, and determining the control drawing object of which the second drawing parameter meets a preset drawing service range as a second target control drawing object.

In this way, in sub-step S1033, second load rendering control information between the target load interactivity event and the second target control rendering object may be generated based on the second rendering parameters between the target load interactivity event and the second target control rendering object.

Or, in the third step, a drawing attribute partition pre-established for each control drawing object may be collected, then a drawing mapping occurrence value between the target loading interaction event and each control drawing object is calculated, and the control drawing object with the drawing mapping occurrence value greater than a preset value is determined as the key loading interaction event to be selected. On the basis, a third drawing parameter of the target loading interactive event and the key loading interactive event to be selected with the drawing attribute partition covering the preset partition can be calculated, and the key loading interactive event to be selected with the third drawing parameter meeting the preset drawing service range is determined as a third target control drawing object.

In this way, in sub-step S1033, third load rendering control information between the target load interactivity event and the third target control rendering object may be generated according to a third rendering parameter between the target load interactivity event and the third target control rendering object.

Further, in a possible implementation manner, for step S104, in the process of entering, in each of the target simulation rendering controls, load rendering control information between the target load interaction event and the target control rendering object under each of the rendering attribute categories, the following exemplary implementation manner may be implemented, and the following is described in detail.

And a substep S1041 of obtaining a preset partition template corresponding to each drawing attribute type.

And in the substep S1042, the partition template matching information for loading the rendering control information between the obtained partition template and the target loading interaction event under the corresponding rendering attribute category and the target control rendering object is called to obtain the partition loading rendering control information corresponding to each rendering attribute category.

And a substep S1043 of entering partition loading rendering control information corresponding to each rendering attribute category into each target simulation rendering control.

Further, in a possible implementation manner, for step S104, in the process of performing filtering and matching on a plurality of interaction event fragments to be loaded to obtain a target loading interaction event having a drawing association relationship with at least one control drawing object, the following exemplary implementation manner may be implemented, and the following detailed description is provided.

And a substep S1031, identifying a page pushing category of each interactive event fragment to be loaded.

And a substep S1032, removing the interactive event fragments to be loaded with the page pushing category being the blacklist loading category, and performing arrangement, screening and coordination on the reserved interactive event fragments to be loaded to obtain the target loading interactive event with the drawing association relation with the at least one control drawing object.

For example, the arrangement, screening and matching manner may refer to a rule or configuration information of a page layout for a current interaction request, or perform adaptive arrangement, screening and matching according to the size of each interactive event fragment to be loaded, which are all within the protection scope of the embodiment of the present application, so that a target loading interactive event having a drawing association relationship with at least one control drawing object may be obtained.

In a possible implementation manner, for step S101, in the process of obtaining the interaction event update information of the interaction window trajectory in the multi-object interaction information stream of the information stream node terminal 200, the following exemplary sub-steps may be implemented, and the following detailed description is provided below

Step S1011, obtaining the multi-object interactive information stream of the information stream node terminal 200, and performing independent movable window extraction processing on the multi-object interactive information stream to obtain independent movable window information of the interactive window trajectory in the multi-object interactive information stream.

Step S1012, performing interactive behavior tracking extraction based on the independent movable window information of the interactive window trajectory to obtain a target interactive behavior migration feature of the interactive window trajectory.

And S1013, extracting interactive content migration characteristics of the multi-object interactive information stream based on the artificial intelligence model to obtain interactive content migration characteristic information of the interactive window track.

Step S1014, performing interactive linkage event synthesis on the target interactive behavior migration characteristic of the interactive window track in the multi-object interactive information stream and the interactive content migration characteristic information of the interactive window track to obtain interactive linkage event synthesis information of the interactive window track, and updating the interactive event of the interactive event record control of the multi-object interactive information stream based on the interactive linkage event synthesis information of the interactive window track to obtain interactive event update information of the interactive window track.

In this embodiment, the multi-object interaction information stream may be understood as an interaction information stream formed by object interaction information recorded in each interaction window track acquired based on a single interaction request. The interaction request may be an interaction instruction specifically initiated by the information flow node terminal 200 to another interaction object. The interactive window trajectory may refer to a trajectory formed by interactive windows formed under an interactive request, and an interactive window may be understood as an interactive unit for providing various functions required for interaction, and different service functions may generally have an association relationship therebetween, so that interactive windows having an association relationship may be formed into an interactive window trajectory based on the association relationship. In addition, the independent movable window information can be used for characterizing the service interaction condition of the independent movable window related to the interactive window data.

In this embodiment, the interactive behavior migration feature may be used to represent behavior feature information of a migration situation of an interactive behavior initiated by a user at a certain time node or a certain space node, for example, when another user initiates a response interaction (for example, but not limited to purchasing, collecting, and subscribing) on a live broadcast interactive behavior of an e-commerce live broadcast commodity a after the user initiates the live broadcast interactive behavior of the e-commerce live broadcast commodity a in the e-commerce live broadcast process, the interactive behavior migration feature may be recorded. In addition, the interactive content migration characteristic information may be content characteristic information that characterizes that the interactive behavior initiated by the user is concerned when the migration condition occurs at a certain time node or a certain space node, for example, taking the above example as an example, it may be understood as content of a specific interaction when another user initiates a response interaction (for example, including but not limited to purchasing, collecting, subscribing) on the live interactive behavior of the live commercial product a.

In this embodiment, after obtaining the update information of the interactive event of the interactive window trajectory, the subsequent operation of pushing the service information is performed based on the update information of the interactive event of the interactive window trajectory, so that the present embodiment synthesizes the target interactive behavior migration characteristic of the interactive window trajectory and the interactive content migration characteristic information of the interactive window trajectory through the interactive linkage event, extracts the rich service relationship characteristic information of the interactive window trajectory by integrating the independent movable window information of the interactive window trajectory and the interactive content migration characteristic information of the interactive window trajectory, and provides data support for accurate interactive mining; in addition, interactive event updating is carried out on the interactive window track through the target interactive behavior migration characteristic of the interactive window track, and interactive event updating information of the interactive window track is obtained, so that an accurate interactive mining process is realized.

In a possible implementation manner, for step S1011, in the process of performing independent movable window extraction processing on the multi-object interactive information stream to obtain independent movable window information of an interactive window trajectory in the multi-object interactive information stream, the following exemplary sub-steps may be implemented, which are described in detail below.

Sub-step S10111, obtaining a set of interactive graphic elements in the interactive graphic track recorded by the window service of each object interaction event in the multi-object interaction information stream.

In this embodiment, it is worth to be noted that the interactive graphics element set in the interactive graphics track includes interactive graphics elements each of which takes each interactive graphics track as an interactive region, and the interactive graphics elements include graphics interaction trigger information and graphics attribute information of the interactive graphics track, and interactive graphics records in the interactive graphics track. For example, the interactive graphics track may be used to represent a time recording interval related to an interactive window updating process, the graphics interaction trigger information may be used to represent a trigger node when the graphics is captured (for example, a click or browse operation by a user may be used as a trigger node), and the graphics attribute information may be used to represent graphics attribute information indicated after the graphics is captured.

Substep S10112, for each interactive graphic track, according to each content editing graphic in the plurality of content editing graphics in the updated graphic record of the interactive graphic track of each object interaction event, according to the page interaction elements of the content editing interaction pages in the content editing graphic, determining whether each content editing interaction page in the content editing graphic is a reference target content editing interaction page, according to the number of the reference target content editing interaction pages in the content editing graphic, determining each reference interaction indication control corresponding to the content editing graphic, for each reference interaction indication control, dividing the reference interaction indication control into a plurality of sub-interaction indication controls, according to the editing object and the preset object range of each content editing interaction page in each sub-interaction indication control, determining whether the reference interaction indication control is a target interaction indication control, wherein each content editing interaction page corresponds to each content editing interaction behavior.

And step S10113, obtaining interactive window partition information of each content editing interactive page in a preset interactive window rule matching target interactive indication control, wherein the interactive window partition information comprises interactive window calling information and interactive window component information, and the preset interactive window rule comprises matching modes corresponding to different interactive window services.

Substep S10114, determining an interactive window rendering attribute characteristic of each interactive window rendering attribute map and an interactive window precondition of each interactive window abstract map according to the interactive window partition information of each updated graphic record of each different interactive graphic element set in the interactive graphic track, and determining an interactive window label object of each object interactive event in the interactive graphic track according to the interactive window rendering attribute characteristic of each interactive window rendering attribute map and the interactive window precondition of each interactive window abstract map in the target interactive indication control, after taking the characteristic of the interactive window range of the interactive window label object and the characteristic of the interactive window range outside the interactive window range of the interactive window label object and associated with the interactive window label object as the independent movable window characteristic of each object interactive event in the interactive graphic track, and after the characteristics of the independent movable windows of each object interaction event in all interactive graphic tracks are gathered, the independent movable window information of the interaction window track in the multi-object interaction information flow is obtained.

In a possible implementation manner, for step S1012, in the process of performing interactive behavior tracking extraction based on the independent movable window information of the interactive window trajectory to obtain the target interactive behavior migration feature of the interactive window trajectory, the following exemplary sub-steps may be implemented, which are described in detail below.

And a substep S10121, obtaining interaction behavior request information of a user interaction behavior set added to window demonstration information of each independent movable window feature in the independent movable window information of the interaction window trajectory, and determining a first interaction flow segment list corresponding to the interaction behavior request information, where the interaction behavior request information includes interaction behavior result information of interaction behavior running information determined according to interaction behavior input information and interaction behavior output information of the user interaction behavior set, and the first interaction flow segment list includes a sequence of a plurality of interaction flow segments of the interaction behavior result information.

Sub-step S10122, determining window presentation information for each of the independently movable window features based on a first interactive behavior vector of the interactive behavior input information and a second interactive behavior vector of the interactive behavior output information.

And a substep S10123, determining a migration analysis parameter for performing migration analysis on the first interactive flow segment list according to the interactive flow segment sequence relationship between the first interactive behavior vector and the second interactive behavior vector.

And a substep S10124, performing migration analysis on the first interactive flow segment list based on the migration analysis parameters to obtain a second interactive flow segment list.

And a substep S10125, performing migration node positioning on the second interactive flow segment list to obtain a plurality of migration node positioning portions, and performing feature extraction on each migration node positioning portion to obtain migration node positioning features.

And a substep S10126 of determining the interactive behavior migration characteristic as the interactive behavior migration characteristic of each independent movable window characteristic according to the interactive behavior migration characteristics corresponding to the plurality of migration node positioning characteristics corresponding to the second interactive flow segment list.

And a substep S10127 of obtaining a target interactive behavior migration characteristic of the interactive window trajectory based on the interactive behavior migration characteristic of each of the independent movable window characteristics.

Further, in a possible implementation manner, for step S1013, in the process of performing interactive content migration feature extraction on the multi-object interactive information stream based on the artificial intelligence model to obtain the interactive content migration feature information of the interactive window trajectory, the following exemplary sub-steps may be implemented, which are described in detail below.

And a substep S10131, inputting the multi-object interaction information stream into a pre-trained artificial intelligence model, and obtaining the confidence coefficient of the multi-object interaction information stream matched with each interactive content label.

It is worth to be noted that the artificial intelligence model is obtained by training based on training samples and training label information corresponding to the training samples, the training samples are multi-object interactive information stream samples, and the training label information is interactive content index information label information. The specific training process may refer to a conventional training mode provided in the prior art, and the training process does not belong to the technical problem intended to be solved by the embodiment of the present application, and is not described in detail herein.

And a substep S10132, determining target interactive content index information corresponding to the multi-object interactive information stream according to the confidence degree that the multi-object interactive information stream is matched with each interactive content label.

For example, the interactive content tag with the confidence degree greater than the preset confidence degree threshold value may be determined as the target interactive content index information corresponding to the multi-object interactive information stream.

And a substep S10133 of extracting interactive content migration characteristic information matched with each interactive window track from interactive content index information description information of target interactive content index information corresponding to the multi-object interactive information stream.

For example, in the extraction process, the feature information with the transition node description, which is matched with each interactive window track, in the interactive content index information description information may be specifically extracted.

In a possible implementation manner, still referring to step S1014, in the process of updating the interactivity event recording control of the multi-object interactive information stream based on the interactive linkage event synthesis information of the interactive window track to obtain the interactivity event update information of the interactive window track, the following exemplary sub-steps may be implemented, which are described in detail below.

And a substep S10145 of obtaining the interactive event information of the interactive window track under the interactive request.

In sub-step S10146, the interactivity event items under the interactivity event information and the event relationship configuration information corresponding to each interactivity event item are obtained.

And a substep S10147, covering and configuring the interactive linkage event synthesis information of the interactive window track under the event relation configuration information corresponding to each interactive event item, and obtaining interactive event update information of the interactive window track.

Fig. 3 is a schematic diagram of functional modules of an information stream mining device 300 based on cloud computing and big data according to an embodiment of the present disclosure, in this embodiment, the information stream mining device 300 based on cloud computing and big data may be divided into the functional modules according to an embodiment of a method executed by the cloud computing interaction center 100, that is, the following functional modules corresponding to the information stream mining device 300 based on cloud computing and big data may be used to execute various embodiments of the method executed by the cloud computing interaction center 100. The cloud computing and big data based information flow mining apparatus 300 may include an obtaining module 310, a first determining module 320, a second determining module 330, and a generating module 340, and the functions of the functional modules of the cloud computing and big data based information flow mining apparatus 300 are described in detail below.

An obtaining module 310, configured to obtain the interactive event verification stream confirmed by the information flow node terminal 200, and obtain an interactive intention path feature of the interactive event verification stream. The obtaining module 310 may be configured to perform the step S110, and the detailed implementation of the obtaining module 310 may refer to the detailed description of the step S110.

The first determining module 320 is configured to determine, according to the interaction intention path feature, an interaction intention path of the generated interaction intention representation mining information, so as to obtain interaction intention path information. The first determining module 320 may be configured to perform the step S120, and for a detailed implementation of the first determining module 320, reference may be made to the detailed description of the step S120.

The second determining module 330 is configured to determine, according to the interaction intention path information and the interaction event feature of the interaction event verification stream, an intention topic of the interaction intention representation mining information to be generated, where the intention topic corresponds to the interaction intention path, so as to obtain intention topic information. The second determining module 330 may be configured to perform the step S130, and the detailed implementation of the second determining module 330 may refer to the detailed description of the step S130.

A generating module 340, configured to generate interaction intention representation mining information of the interaction event verification stream according to the intention topic information. The generating module 340 may be configured to execute the step S140, and the detailed implementation of the generating module 340 may refer to the detailed description of the step S140.

It should be noted that the division of the modules of the above apparatus is only a logical division, and the actual implementation may be wholly or partially integrated into one physical entity, or may be physically separated. And these modules may all be implemented in software invoked by a processing element. Or may be implemented entirely in hardware. And part of the modules can be realized in the form of calling software by the processing element, and part of the modules can be realized in the form of hardware. For example, the obtaining module 310 may be a processing element separately set up, or may be implemented by being integrated into a chip of the apparatus, or may be stored in a memory of the apparatus in the form of program code, and the processing element of the apparatus calls and executes the functions of the obtaining module 310. Other modules are implemented similarly. In addition, all or part of the modules can be integrated together or can be independently realized. The processing element described herein may be an integrated circuit having signal processing capabilities. In implementation, each step of the above method or each module above may be implemented by an integrated logic circuit of hardware in a processor element or an instruction in the form of software.

Fig. 4 is a schematic diagram illustrating a hardware structure of a cloud computing interaction center 100 for implementing the cloud computing and big data-based information stream mining method, according to an embodiment of the present disclosure, as shown in fig. 4, the cloud computing interaction center 100 may include a processor 110, a machine-readable storage medium 120, a bus 130, and a transceiver 140.

In a specific implementation process, at least one processor 110 executes computer-executable instructions stored in the machine-readable storage medium 120 (for example, the obtaining module 310, the first determining module 320, the second determining module 330, and the generating module 340 included in the cloud computing and big data based information stream mining apparatus 300 shown in fig. 3), so that the processor 110 may execute the cloud computing and big data based information stream mining method according to the above method embodiment, where the processor 110, the machine-readable storage medium 120, and the transceiver 140 are connected through the bus 130, and the processor 110 may be configured to control the transceiving action of the transceiver 140, so as to perform data transceiving with the aforementioned information stream node terminal 200.

For a specific implementation process of the processor 110, reference may be made to the above-mentioned method embodiments executed by the cloud computing interaction center 100, and implementation principles and technical effects thereof are similar, and details of this embodiment are not described herein again.

In the embodiment shown in fig. 4, it should be understood that the Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the present invention may be embodied directly in a hardware processor, or in a combination of hardware and software modules within the processor.

The machine-readable storage medium 120 may comprise high-speed RAM memory and may also include non-volatile storage NVM, such as at least one disk memory.

The bus 130 may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended ISA (EISA) bus, or the like. The bus 130 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, the buses in the figures of the present application are not limited to only one bus or one type of bus.

In addition, the embodiment of the application also provides a readable storage medium, wherein the readable storage medium stores computer execution instructions, and when a processor executes the computer execution instructions, the information flow mining method based on cloud computing and big data is realized.

The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

Having thus described the basic concept, it will be apparent to those skilled in the art that the foregoing detailed disclosure is to be regarded as illustrative only and not as limiting the present specification. Various modifications, improvements and adaptations to the present description may occur to those skilled in the art, although not explicitly described herein. Such modifications, improvements and adaptations are proposed in the present specification and thus fall within the spirit and scope of the exemplary embodiments of the present specification.

Also, particular push elements are used in this description to describe embodiments of this description. Reference to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic described in connection with at least one embodiment of the specification. Therefore, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, some features, structures, or characteristics of one or more embodiments of the specification may be combined as appropriate.

Moreover, those skilled in the art will appreciate that aspects of the present description may be illustrated and described in terms of several patentable species or contexts, including any new and useful combination of processes, machines, manufacture, or materials, or any new and useful improvement thereof. Accordingly, aspects of this description may be performed entirely by hardware, entirely by software (including firmware, resident software, micro-code, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data block," module, "" engine, "" unit, "" component, "or" system. Furthermore, aspects of the present description may be represented as a computer product, including computer readable program code, embodied in one or more computer readable media.

The computer storage medium may comprise a propagated data signal with the computer program code embodied therewith, for example, on baseband or as part of a carrier wave. The propagated signal may take any of a variety of forms, including electromagnetic, optical, etc., or any suitable combination. A computer storage medium may be any computer-readable medium that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, or device. Program code located on a computer storage medium may be propagated over any suitable medium, including radio, cable, fiber optic cable, RF, or the like, or any combination of the preceding.

Computer program code required for the operation of various portions of this specification may be written in any one or more programming languages, including an object oriented programming language such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C + +, C #, VB.NET, Python, and the like, a conventional programming language such as C, VisualBasic, Fortran2003, Perl, COBOL2002, PHP, ABAP, a passive programming language such as Python, Ruby, and Groovy, or other programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any network format, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet), or in a cloud computing environment, or as a service, such as a software as a service (SaaS).

Additionally, the order in which the elements and sequences are processed, the use of alphanumeric characters, or the use of other designations in this specification is not intended to limit the order of the processes and methods in this specification, unless otherwise specified in the claims. While various presently contemplated embodiments of the invention have been discussed in the foregoing disclosure by way of example, it is to be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments herein. For example, although the system components described above may be implemented by hardware devices, they may also be implemented by software-only solutions, such as installing the described system on an existing server or mobile device.

Similarly, it should be noted that in the preceding description of embodiments of the present specification, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of one or more of the embodiments. This method of disclosure, however, is not intended to imply that more features than are expressly recited in a claim. Indeed, the embodiments may be characterized as having less than all of the features of a single embodiment disclosed above.

Finally, it should be understood that the examples in this specification are only intended to illustrate the principles of the examples in this specification. Other variations are also possible within the scope of this description. Thus, by way of example, and not limitation, alternative configurations of the embodiments of the specification can be considered consistent with the teachings of the specification. Accordingly, the embodiments of the present description are not limited to only those embodiments explicitly described and depicted herein.

Claims

1. An interactive event updating method based on artificial intelligence is applied to a cloud computing interactive center, wherein the cloud computing interactive center is in communication connection with a plurality of information flow node terminals, and the method comprises the following steps:

acquiring a multi-object interactive information stream of the information stream node terminal, and performing independent movable window extraction processing on the multi-object interactive information stream to obtain independent movable window information of an interactive window track in the multi-object interactive information stream;

carrying out interactive behavior tracking extraction based on independent movable window information of the interactive window track to obtain target interactive behavior migration characteristics of the interactive window track;

extracting interactive content migration characteristics of the multi-object interactive information stream based on an artificial intelligence model to obtain interactive content migration characteristic information of an interactive window track;

carrying out interactive linkage event synthesis on target interactive behavior migration characteristics of an interactive window track and interactive content migration characteristics of the interactive window track in the multi-object interactive information stream to obtain interactive linkage event synthesis information of the interactive window track and obtain interactive event information of the interactive window track under an interactive request;

acquiring interaction event items under the interaction event information and event relation configuration information corresponding to each interaction event item;

the interactive linkage event synthesis information of the interactive window track is covered and configured under the event relation configuration information corresponding to each interactive event item, and interactive event updating information of the interactive window track is obtained;

the interactive behavior migration feature is used for representing behavior feature information of a migration situation of an interactive behavior initiated by a user at a certain time node or a certain space node, and the interactive content migration feature information represents content feature information concerned by the interactive behavior initiated by the user when the migration situation of the interactive behavior at the certain time node or the certain space node occurs.

2. The artificial intelligence based interaction event updating method of claim 1, wherein the method further comprises:

inputting loading drawing control information between the target loading interactive event and a target control drawing object under each drawing attribute category in each target simulation drawing control, selecting a target simulation drawing resource matched with the drawing loading element to be simulated from a preset target simulation drawing resource set according to an input result, and pushing an interactive event verification stream of the target simulation drawing resource to the information flow node terminal, so that the interactive event verification stream is used for information mining of a user of the information flow node terminal after the information flow node terminal verifies and confirms the interactive event verification stream;

3. The method for updating interaction events based on artificial intelligence according to claim 2, wherein the step of determining the interaction intention path of the generated interaction intention representation mining information according to the interaction intention path features to obtain the interaction intention path information comprises:

4. The artificial intelligence based interaction event updating method according to claim 3, wherein the generating of the interaction intention representation mining information of the interaction event verification stream according to the intention topic information comprises:

5. The artificial intelligence based interaction event updating method of claim 4, wherein the first intention knowledge unit comprises a first connection layer, a first feature extraction layer and a first classification layer, and the outputting, by the first intention knowledge unit, the first intention classification information of the n knowledge rule nodes with the first fused component corresponding to the n knowledge rule nodes as an input, comprises:

6. The artificial intelligence based interaction event updating method of claim 5, wherein the second intention knowledge unit comprises a second connection layer, a second feature extraction layer and a second classification layer, and the outputting of the second intention classification information of the n knowledge rule nodes by the second intention knowledge unit with the second fusion component corresponding to the n knowledge rule nodes as input comprises:

7. The artificial intelligence based interactivity event updating method according to claim 2, wherein the obtaining of the interactivity intention path characteristics of the interactivity event verification stream comprises:

8. The information flow mining system based on cloud computing and big data is characterized by comprising a cloud computing interaction center and a plurality of information flow node terminals in communication connection with the cloud computing interaction center;

the cloud computing interaction center is used for:

9. A cloud computing interaction center, characterized in that the cloud computing interaction center comprises a processor, a machine-readable storage medium, and a network interface, the machine-readable storage medium, the network interface, and the processor are connected through a bus system, the network interface is used for being connected with at least one information flow node terminal in a communication manner, the machine-readable storage medium is used for storing programs, instructions, or codes, and the processor is used for executing the programs, instructions, or codes in the machine-readable storage medium to execute the artificial intelligence based interaction event updating method according to any one of claims 1 to 8.