CN114500101A

CN114500101A - Cloud game vulnerability analysis method based on cloud computing and server

Info

Publication number: CN114500101A
Application number: CN202210216999.8A
Authority: CN
Inventors: 叶成立; 席乐娟
Original assignee: Guilin Gaosheng Network Technology Co ltd
Current assignee: Shanghai Weimai Wenbo Digital Technology Co ltd
Priority date: 2022-03-07
Filing date: 2022-03-07
Publication date: 2022-05-13
Anticipated expiration: 2042-03-07
Also published as: CN115695034A; CN114500101B

Abstract

The invention provides a cloud game vulnerability analysis method and a server based on cloud computing, which can ensure that a detection report conversion strategy deeply analyzes game interaction data of game types in a cloud game vulnerability detection report according to operation vulnerability expression information, so that an expected vulnerability detection report obtained by attention-based conversion processing of the detection report conversion strategy can be associated with the game types, and the attention-based conversion processing quality of the expected vulnerability detection report is ensured; the operation vulnerability expression information of the cloud game vulnerability detection report is obtained according to an operation vulnerability expression analysis strategy which is not related to the detection report conversion strategy, accurate and reliable operation vulnerability expression information can be obtained through the differentiated use of the strategy, the association degree of the expected vulnerability detection report obtained through the attention-based conversion processing and the game type of the cloud game vulnerability detection report is guaranteed, and the expected vulnerability detection report and the corresponding game type can be conveniently subjected to combined vulnerability repair processing.

Description

Cloud game vulnerability analysis method based on cloud computing and server

Technical Field

The invention relates to the technical field of cloud games, in particular to a cloud game vulnerability analysis method and a cloud game vulnerability analysis server based on cloud computing.

Background

Cloud gaming (also known as game on demand) is an online game technology based on Cloud computing technology. Cloud game technology enables light-end equipment with relatively limited graphics processing and data computing capabilities to run high-quality games.

In a cloud game scene, a game is not operated in a player game terminal but in a cloud server, and the cloud server renders the game scene into a video and audio stream which is transmitted to the player game terminal through a network. Therefore, the player game terminal does not need to have strong graphic operation and data processing capacity, and only needs to have basic streaming media playing capacity and capacity of acquiring player input instructions and sending the instructions to the cloud server.

At present, vulnerability analysis of the cloud game is essential for ensuring upgrading and updating of various services of the cloud game. However, the related vulnerability analysis technology is difficult to guarantee the analysis accuracy and reliability, so that an effective basis is difficult to provide for vulnerability repair/service upgrade and update.

Disclosure of Invention

The invention provides a cloud game vulnerability analysis method and a server based on cloud computing, and adopts the following technical scheme in order to achieve the technical purpose.

The first aspect is a cloud game vulnerability analysis method based on cloud computing, which is applied to a cloud game server and comprises the following steps: responding to a game vulnerability analysis instruction, and collecting cloud game vulnerability detection reports which trigger game vulnerability analysis conditions; collecting operation vulnerability expression information of a cloud game vulnerability detection report triggering game vulnerability analysis conditions according to an operation vulnerability expression analysis strategy; and according to the operation vulnerability expression information, performing attention-based conversion processing on the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to a detection report conversion strategy to determine an expected vulnerability detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

By means of the design, the game interaction data of the game types in the cloud game vulnerability detection report triggering the game vulnerability analysis conditions can be deeply analyzed by the detection report conversion strategy as far as possible according to the operation vulnerability expression information by collecting the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis conditions, so that the expected vulnerability detection report obtained by the attention-based conversion processing of the detection report conversion strategy can be associated with the game types of the cloud game vulnerability detection report triggering the game vulnerability analysis conditions, and the attention-based conversion processing quality of the expected vulnerability detection report is guaranteed; in addition, the running vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition is obtained according to a running vulnerability expression analysis strategy which is not related to the detection report conversion strategy, and accurate and reliable running vulnerability expression information can be obtained through the differentiated use of the strategies, so that the association degree of an expected vulnerability detection report obtained through attention-based conversion processing and the game type of the cloud game vulnerability detection report triggering the game vulnerability analysis condition is guaranteed, the expected vulnerability detection report and the corresponding game type can be subjected to joint vulnerability repair processing in the follow-up process, and an effective basis is provided for vulnerability repair/service upgrade updating.

For an independent solution, it is desirable that the vulnerability detection report includes a vulnerability risk level detection report; according to the operation vulnerability expression information, performing attention-based conversion processing on the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to a detection report conversion strategy to determine an expected vulnerability detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, and the method comprises the following steps: processing the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to a detection report conversion strategy to determine the game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition; and determining a vulnerability risk level detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis conditions according to the game type risk description and the operation vulnerability expression information.

In this way, by collecting game type risk descriptions of cloud game vulnerability detection reports that trigger game vulnerability analysis conditions, the effect of attention-based conversion processing of expected vulnerability detection reports of detection report conversion strategies under game types with frequent game type risk level changes can be improved.

For an independent technical scheme, the game type risk is described as a running vulnerability presentation improving characteristic of a running vulnerability presentation improving indication covering different vulnerability detection entries of a cloud game vulnerability detection report triggering a game vulnerability analysis condition, and the running vulnerability presentation information is a running vulnerability presentation characteristic of a running vulnerability presentation covering different vulnerability detection entries of a cloud game vulnerability detection report triggering a game vulnerability analysis condition. The vulnerability risk level detection report for determining the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the game type risk description and the operation vulnerability expression information comprises the following steps: weighting the operation vulnerability expression improvement characteristics and the operation vulnerability expression characteristics to determine a vulnerability risk level detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis conditions.

Therefore, by operating the vulnerability presentation improvement characteristics, the adjustable risk description can be determined, and the effect of attention-based conversion processing of the expected vulnerability detection report of the detection report conversion strategy under the game type with frequent change of the game type risk level is guaranteed.

For an independent technical solution, the detection report conversion strategy includes a global significant information capture sub-strategy and a vulnerability risk level positioning sub-strategy. The method for processing the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the detection report conversion strategy to determine the game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition comprises the following steps: according to the global significant information capturing sub-strategy, significant information capturing is carried out on a cloud game vulnerability detection report triggering a game vulnerability analysis condition to obtain a vulnerability detection significant information cluster, and the vulnerability detection significant information cluster and a first game type social relationship information cluster generated by a vulnerability expression capturing sub-strategy of a vulnerability expression analysis strategy are operated to be sorted so as to determine a first sorted significant information cluster; and positioning the first sorted significant information cluster according to the vulnerability risk level positioning sub-strategy to determine game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

Therefore, the vulnerability detection significant information cluster and the first game type social relationship information cluster generated by the vulnerability expression analysis strategy operation execution vulnerability expression capture sub-strategy are sorted according to the global significant information capture sub-strategy, so that the detection report conversion strategy can ensure the attention-based conversion processing quality of the expected vulnerability detection report according to the social relationship information of the first game type social relationship information cluster.

For an independent technical solution, the global significant information capturing sub-policy includes at least one significant information capturing node satisfying a set rule, and each significant information capturing node includes a running vulnerability performance improvement sub-node. The sorting of the first game type social relationship information cluster generated by the vulnerability detection significant information cluster and the vulnerability expression capturing sub-operation strategy of the vulnerability expression analysis strategy to determine a first sorted significant information cluster comprises the following steps: transmitting the vulnerability detection salient information cluster to a first salient information capturing node; for each salient information capture node: sorting the significant information cluster generated by the previous significant information capturing node and the first game type social relationship information cluster by the child node according to the operation vulnerability performance improvement to determine a second sorted significant information cluster pointed by the significant information capturing node; the information quality of the game type social relationship information cluster pointed by each significant information capturing node is different; a second sorted salient information cluster from the last salient information capture node is determined to be the first sorted salient information cluster.

Therefore, the game type social relationship information cluster generated by the vulnerability expression analysis strategy and the detection report conversion strategy are used for capturing the significant information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the vulnerability expression improvement sub-node, and the vulnerability detection significant information cluster is obtained and is arranged, so that the purpose of transmitting the significant information obtained by the vulnerability expression analysis strategy to the detection report conversion strategy for utilization can be realized according to the game type relationship information of the game type in the cloud game vulnerability detection report triggering the game vulnerability analysis condition in the game type social relationship information cluster at the later stage of the detection report conversion strategy.

For an independent technical solution, before the sorting of the significant information cluster generated by the previous significant information capturing node and the game type social relationship information cluster by the child node according to the operation vulnerability performance improvement to determine a second sorted significant information cluster pointed to by the significant information capturing node, the method further includes: carrying out redundancy removing operation on the significant information cluster generated by the previous significant information capturing node; and/or sorting the significant information cluster and the game type social relationship information cluster generated by the previous significant information capturing node by the child node according to the operation vulnerability performance improvement function to determine a second sorted significant information cluster pointed by the significant information capturing node, wherein the method comprises the following steps: improving the implementation of child nodes according to the operation vulnerability performance: and updating the game type social relationship information cluster into a game type social relationship information cluster with a target dimension, and performing joint optimization on the updated game type social relationship information cluster and the significant information cluster generated by the previous significant information capturing node to determine a second sorted significant information cluster pointed by the significant information capturing node.

In this way, by means of redundancy removal operation, the significant information cluster generated by the last significant information capturing node can be simplified. In addition, the operation vulnerability expression improvement sub-node performs joint optimization through the game type social relationship information cluster and the significant information cluster generated by the last significant information capture node, and realizes efficient sorting of the game type social relationship information cluster and the significant information cluster generated by the last significant information capture node.

For an independent technical solution, the step of locating the first sorted significant information cluster according to the vulnerability risk level locating sub-policy to determine a game type risk description of a cloud game vulnerability detection report triggering a game vulnerability analysis condition includes: and positioning the first sorted significant information cluster and at least one second sorted significant information cluster of the operation vulnerability performance improvement sub-node according to the vulnerability risk level positioning sub-strategy to determine game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

Therefore, the vulnerability risk level positioning sub-strategy can accurately obtain game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis conditions by the second sorted significant information cluster generated by the vulnerability performance improvement sub-node according to the first sorted significant information cluster and the operation vulnerability performance improvement sub-node.

For an independent technical scheme, the detection report conversion strategy further comprises a vulnerability fix time-consuming positioning sub-strategy; according to the operation vulnerability expression information, performing attention-based conversion processing on the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to a detection report conversion strategy to determine an expected vulnerability detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, and further comprising the following steps of: and positioning the first sorted significant information cluster according to a bug fixing time-consuming positioning sub-strategy to determine a bug fixing time-consuming detection report of the cloud game bug detection report triggering the game bug analysis condition.

Therefore, the bug fixing time-consuming positioning sub-strategy can accurately obtain the bug fixing time-consuming detection report of the cloud game bug detection report triggering the game bug analysis condition according to the first sorted significant information cluster.

For an independent technical solution, the determining a bug-fix time-consuming detection report of a cloud game bug detection report that triggers a game bug analysis condition by locating a first sorted significant information cluster according to a bug-fix time-consuming locator policy includes: and positioning the first sorted significant information cluster and at least one second sorted significant information cluster of the operation vulnerability performance improvement sub-node according to a vulnerability repair time-consuming positioning sub-strategy to determine a vulnerability repair time-consuming detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

Therefore, the vulnerability repair time-consuming positioning sub-strategy can accurately obtain the vulnerability repair time-consuming detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition by improving the second sorted significant information cluster of the sub-node according to the first sorted significant information cluster and at least one operation vulnerability expression.

For an independent technical scheme, the vulnerability expression analysis strategy comprises a vulnerability expression capturing sub-strategy, a vulnerability expression positioning sub-strategy and a vulnerability expression sorting sub-strategy. The method for collecting the running vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis conditions according to the running vulnerability expression analysis strategy comprises the following steps: capturing a cloud game vulnerability detection report triggering a game vulnerability analysis condition according to an operation vulnerability expression capturing sub-strategy to determine a first game type social relationship information cluster; positioning the first game type social relationship information cluster according to the operation vulnerability presentation positioning sub-strategy to determine a positioning significant information cluster; and sorting the first game type social relationship information cluster and the positioning significant information cluster according to the operation vulnerability expression sorting sub-strategy to determine operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

In this way, the cloud game vulnerability detection report triggering the game vulnerability analysis condition is processed according to the operation vulnerability expression capture sub-strategy, the operation vulnerability expression positioning sub-strategy and the operation vulnerability expression sorting sub-strategy of the operation vulnerability expression analysis strategy, so that the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition can be comprehensively obtained.

For an independent technical solution, the capturing a cloud game vulnerability detection report triggering a game vulnerability analysis condition according to a runtime vulnerability expression capturing sub-policy to determine a first game type social relationship information cluster includes: and capturing the cloud game vulnerability detection report triggering the game vulnerability analysis condition for a plurality of rounds according to the operation vulnerability expression capturing sub-strategy to determine the first game type social relationship information cluster pointed by each round, wherein the information quality in the first game type social relationship information cluster pointed by each round is different, and the first game type social relationship information cluster pointed by the last round of remarkable information capturing sub-strategy is transmitted to the operation vulnerability expression positioning sub-strategy. The method for sorting the first game type social relationship information cluster and the positioning significant information cluster according to the operation vulnerability expression sorting sub-strategy to determine the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition comprises the following steps: and (3) arranging sub-strategies according to the operation vulnerability expression implementation: and combining the first game type social relationship information clusters pointed by each round to obtain a second game type social relationship information cluster, combining the second game type social relationship information cluster and the positioning significant information cluster to obtain a third game type social relationship information cluster, and determining the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis conditions according to the third game type social relationship information cluster.

Therefore, by capturing the cloud game vulnerability detection report triggering the game vulnerability analysis condition for a plurality of rounds, more comprehensive and complete significant information can be extracted more deeply, so that the social relationship of the game type of the cloud game vulnerability detection report triggering the game vulnerability analysis condition can be more accurate.

For an independent technical scheme, no association exists between the deployment process of the vulnerability expression analysis strategy and the deployment process of the detection report conversion strategy.

For an independent technical solution, before collecting, according to a runtime vulnerability expression analysis policy, runtime vulnerability expression information of a cloud game vulnerability detection report that triggers a game vulnerability analysis condition, the method further includes: deploying according to a first example set to obtain a running vulnerability expression analysis strategy, wherein running vulnerability expression information is added to a cloud game vulnerability detection report in the first example set; and collecting example operation vulnerability expression information of the cloud game vulnerability detection report in the second example set according to the successfully deployed operation vulnerability expression analysis strategy, and deploying the detection report conversion strategy according to the second example set and the example operation vulnerability expression information.

In this way, by independently deploying the operation vulnerability expression analysis strategy, the operation vulnerability expression analysis strategy can be deployed only according to the operation vulnerability expression paradigm data, the attention-based conversion processing quality of the expected vulnerability detection report can be improved, and meanwhile, the interference on the attention-based conversion processing quality of the expected vulnerability detection report caused by insufficient expected vulnerability detection report paradigm data can be weakened.

For an independent solution, the second exemplar set includes a first phased exemplar set and a second phased exemplar set, and the deploying the detection report conversion policy according to the second exemplar set and the exemplar operation vulnerability performance information includes: deploying the detection report conversion strategy according to the first staged example set and the example operation vulnerability expression information pointed by the first staged example set so as to update the strategy weight of the global significant information capturing sub-strategy and the vulnerability risk level positioning sub-strategy in the detection report conversion strategy; and deploying the detection report conversion strategy according to the second phased example set and the example operation vulnerability expression information pointed by the second phased example set so as to update the strategy weight of the global significant information capturing sub-strategy and the vulnerability fixing time-consuming positioning sub-strategy in the detection report conversion strategy.

In this way, by respectively deploying the global significant information capturing sub-strategy, the vulnerability risk level positioning sub-strategy, the global significant information capturing sub-strategy and the vulnerability repair time-consuming positioning sub-strategy, when the cloud game vulnerability detection report triggering the game vulnerability analysis condition is subjected to attention-based conversion processing by the detection report conversion strategy, a vulnerability risk level detection report and a vulnerability repair time-consuming detection report with high quality can be obtained.

A second aspect is a cloud gaming server comprising a memory and a processor; the memory and the processor are coupled; the memory for storing computer program code, the computer program code comprising computer instructions; wherein the computer instructions, when executed by the processor, cause the cloud gaming server to perform the method of the first aspect.

According to one embodiment of the invention, by collecting the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, the detection report conversion strategy can be ensured to deeply analyze the game interaction data of the game type in the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the operation vulnerability expression information, so that the expected vulnerability detection report obtained by the attention-based conversion processing of the detection report conversion strategy can be associated with the game type of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, and the attention-based conversion processing quality of the expected vulnerability detection report is ensured; in addition, the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition is obtained according to an operation vulnerability expression analysis strategy which is not related to the detection report conversion strategy, and accurate and reliable operation vulnerability expression information can be obtained through the differentiated use of the strategies, so that the association degree of an expected vulnerability detection report obtained through attention-based conversion processing and the game type of the cloud game vulnerability detection report triggering the game vulnerability analysis condition is guaranteed, and the expected vulnerability detection report and the corresponding game type can be subjected to combined vulnerability repair processing conveniently.

Drawings

Fig. 1 is a schematic flow chart of a cloud game vulnerability analysis method based on cloud computing according to an embodiment of the present invention.

Fig. 2 is a block diagram of a cloud game vulnerability analysis apparatus based on cloud computing according to an embodiment of the present invention.

Detailed Description

In the following, the terms "first", "second" and "third", etc. are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second" or "third", etc., may explicitly or implicitly include one or more of the features.

Fig. 1 is a schematic flowchart illustrating a cloud computing-based cloud game vulnerability analysis method according to an embodiment of the present invention, where the cloud computing-based cloud game vulnerability analysis method may be implemented by a cloud game server, and the cloud game server may include a memory and a processor; the memory and the processor are coupled; the memory for storing computer program code, the computer program code comprising computer instructions; wherein the computer instructions, when executed by the processor, cause the cloud game server to perform the following steps.

step11, responding to the game vulnerability analysis instruction, and collecting cloud game vulnerability detection reports triggering game vulnerability analysis conditions.

In the embodiment of the present invention, the cloud game vulnerability detection report triggering the game vulnerability analysis condition may be an initially imported cloud game vulnerability detection report for obtaining an expected vulnerability detection report directed thereto based on the attention conversion processing. The cloud game vulnerability detection report triggering the game vulnerability analysis condition can be an MOBA cloud game vulnerability detection report or an RPG cloud game vulnerability detection report and the like. For example, the game vulnerability analysis condition may be set according to the actual situation of the cloud game vulnerability analysis. Such as: cloud game running time, cloud game interaction scenes, the number of cloud game players and the like.

step12, collecting the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the operation vulnerability expression analysis strategy.

In the embodiment of the invention, the operation vulnerability expression analysis strategy can be an AI model generated according to artificial intelligence and used for analyzing the significant information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition so as to obtain the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition. And operating a vulnerability expression analysis strategy to perform significant information analysis on the cloud game vulnerability detection report triggering the game vulnerability analysis conditions to obtain a plurality of groups of significant information clusters. The operation vulnerability expression information can be understood as operation vulnerability expression information of each vulnerability detection entry in the cloud game vulnerability detection report triggering the game vulnerability analysis condition, and game interaction data to be imported into the cloud game vulnerability detection report can be obtained through the operation vulnerability expression information, such as: and triggering relationship information of game types in the cloud game vulnerability detection report of the game vulnerability analysis conditions. For another example, the vulnerability presentation analysis policy may be a GCN (graph convolution neural network), and the GCN can organize the multi-modal significant information clusters to obtain an expected vulnerability detection report with a large personality difference.

step13, according to the operation vulnerability expression information, performing attention-based conversion processing on the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the detection report conversion strategy to determine an expected vulnerability detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

It can be understood that after the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition is obtained, the detection report conversion strategy can perform attention-based conversion processing on the cloud game vulnerability detection report to be imported according to the operation vulnerability expression information. In actual implementation, the detection report conversion strategy can perform attention-based conversion processing on the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the operation vulnerability expression information of each vulnerability detection entry in the operation vulnerability expression information and the relationship information of the game type contained in the operation vulnerability expression information to obtain an expected vulnerability detection report, so as to obtain a vulnerability risk level detection report and a vulnerability repair time-consuming detection report.

For example, the detection report conversion strategy may be, for example, a detection report analysis model or a detection analysis thread, and the underlying technology thereof may also refer to the existing artificial intelligence technology. The detection report conversion strategy is used for carrying out attention-based conversion processing on the cloud game vulnerability detection report triggering the game vulnerability analysis condition, and can be understood as analyzing the cloud game vulnerability detection report triggering the game vulnerability analysis condition at different angles by means of the detection report conversion strategy so as to obtain an expected vulnerability detection report. In addition, the expected vulnerability detection report may be a detection report at different angles, such as the vulnerability risk level detection report and the vulnerability repair time-consuming detection report described above, or may be a detection report at the remaining angles, such as a potential vulnerability detection report or a malicious Trojan horse detection report. For another example, the detection report conversion policy may be a GCN or an RNN.

Therefore, by collecting the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, the game interaction data of the game type in the cloud game vulnerability detection report triggering the game vulnerability analysis condition can be deeply analyzed by the detection report conversion strategy as far as possible according to the operation vulnerability expression information, so that the expected vulnerability detection report obtained by the attention-based conversion processing of the detection report conversion strategy can be associated with the game type of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, and the attention-based conversion processing quality of the expected vulnerability detection report is guaranteed; in addition, the running vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition is obtained according to a running vulnerability expression analysis strategy which is not related to the detection report conversion strategy, and accurate and reliable running vulnerability expression information can be obtained through differentiated use of the strategy, so that the association degree of an expected vulnerability detection report obtained through attention-based conversion processing and the game type of the cloud game vulnerability detection report triggering the game vulnerability analysis condition is guaranteed, and the expected vulnerability detection report and the corresponding game type can be subjected to joint vulnerability repair processing in the follow-up process.

For an independently implementable embodiment, the cloud game vulnerability analysis method based on cloud computing shown in the present invention can also be implemented as follows.

step21, responding to the game vulnerability analysis instruction, and collecting cloud game vulnerability detection reports triggering game vulnerability analysis conditions. It is understood that step21 may refer to the operation steps recorded in step11, and the embodiment of the present invention is not limited thereto.

step22, collecting the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the operation vulnerability expression analysis strategy. It is understood that step22 may refer to the operation steps recorded in step12, and the embodiment of the present invention is not limited thereto.

In the embodiment of the invention, the operation vulnerability expression information is the operation vulnerability expression characteristics of the operation vulnerability expressions of different vulnerability detection items of the cloud game vulnerability detection report which covers the game vulnerability analysis condition, in other words, each vulnerability detection item in the cloud game vulnerability detection report which triggers the game vulnerability analysis condition has a pointed operation vulnerability expression.

It can be understood that the operation vulnerability expression parsing strategy comprises an operation vulnerability expression capturing sub-strategy, an operation vulnerability expression positioning sub-strategy and an operation vulnerability expression sorting sub-strategy. The vulnerability expression capturing sub-strategy is operated to capture the significant information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, the vulnerability expression positioning sub-strategy is operated to position the significant information (extract the significant information) and derive a significant information cluster, and the vulnerability expression sorting sub-strategy is operated to deeply analyze the derivation of the positioning sub-strategy.

For an independently implementable embodiment, the cloud game vulnerability analysis method based on cloud computing shown in the present invention can also be implemented as follows. In actual implementation, the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition is collected according to the operation vulnerability expression analysis strategy, and may include the following technical scheme recorded by step221-step 223.

step221, capturing a cloud game vulnerability detection report triggering a game vulnerability analysis condition according to the operation vulnerability expression capturing sub-strategy to determine a first game type social relationship information cluster.

In the embodiment of the invention, the cloud game vulnerability detection report triggering the game vulnerability analysis condition can be captured according to the operation vulnerability expression capturing sub-strategy of the operation vulnerability expression analysis strategy so as to capture the significant information in the cloud game vulnerability detection report triggering the game vulnerability analysis condition. The significant information obtained by capturing the cloud game vulnerability detection report triggering the game vulnerability analysis condition by running the vulnerability expression capturing sub-strategy can be social relationship information of the game type in the cloud game vulnerability detection report triggering the game vulnerability analysis condition, wherein the social relationship information comprises player relationship network information and game network security environment information. Finally, the vulnerability presentation capture sub-policy may be run to derive a first game type social relationship information cluster, in other words, a social relationship information cluster for a game type in the cloud game vulnerability detection report that triggers the game vulnerability analysis conditions.

On the premise that the vulnerability expression capturing sub-strategy is operated to have a plurality of groups of social relations, a cloud game vulnerability detection report triggering game vulnerability analysis conditions can be subjected to a plurality of rounds of capturing (can be understood as significant information capturing) according to the vulnerability expression capturing sub-strategy, and a significant information cluster obtained by each round of significant information capturing sub-strategy is a first game type social relation information cluster. For example, when the salient information capture sub-policy includes four capture units to perform capture, the first capture unit captures a cloud game vulnerability detection report that triggers a game vulnerability analysis condition, and then derives a first game type social relationship information cluster. And the second capturing unit takes the first game type social relationship information cluster generated by the previous capturing unit as a reference, captures again, and then derives the pointed first game type social relationship information cluster. In addition, the information quality of the first game type social relationship information cluster generated by each unit of the operation vulnerability expression capture sub-strategy can be configured to be different. The information quality may include the personality differences of the first game type social relationship information cluster, the level of the salient information, and the like. At this time, the first game type social relationship information cluster pointed by the last round is transmitted to the vulnerability expression positioning sub-strategy. Through a plurality of rounds of capture of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, more comprehensive and complete significant information can be extracted more deeply, so that the social relationship of the game types of the cloud game vulnerability detection report triggering the game vulnerability analysis condition can be more comprehensive.

step222, positioning the first game type social relationship information cluster according to the operation vulnerability expression positioning sub-strategy to determine a positioning significant information cluster.

In the embodiment of the invention, after the cloud game vulnerability detection report triggering the game vulnerability analysis condition is captured according to the operation vulnerability expression capturing sub-strategy and the first game type social relationship information cluster is derived, the first game type social relationship information cluster can be positioned according to the operation vulnerability expression positioning sub-strategy to determine the positioning significant information cluster. When the vulnerability expression positioning sub-strategy is operated to position the first game type social relationship information cluster, the positioning of the significant information captured by the vulnerability expression capturing sub-strategy can be specifically performed, and a positioning significant information cluster with a set size and a set personalized difference degree is generated.

It can be understood that when the vulnerability expression positioning sub-strategy is operated under the premise that a plurality of groups of social relations exist, the vulnerability expression positioning sub-strategy can also be operated to perform multi-round positioning on the first game type social relation information cluster, the first positioning sub-strategy positions the first game type social relation information cluster, and then the pointed set positioning significant information cluster is derived. The second positioning sub-strategy positions the positioning significant information cluster generated by the first positioning sub-strategy and then derives the pointed set positioning significant information cluster. And calculating one by one, and setting the positioning significant information cluster generated in the last round to be the positioning significant information cluster.

step223, sorting the first game type social relationship information cluster and the positioning significant information cluster according to the operation vulnerability expression sorting sub-strategy to determine operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

In the embodiment of the invention, after the positioning is carried out according to the operation vulnerability expression capturing sub-strategy, in order to deeply analyze the significant information generated by the operation vulnerability expression capturing sub-strategy and obtain more accurate and reliable game type relation information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, the first game type social relation information cluster and the positioning significant information cluster can be sorted according to the operation vulnerability expression sorting sub-strategy to obtain the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition. For example, the saliency information in the first game type social relationship information cluster and the saliency information locating the saliency information cluster may be collated together to obtain the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

It can be understood that the operation vulnerability expression information is an operation vulnerability expression characteristic of operation vulnerability expressions of different vulnerability detection entries of the cloud game vulnerability detection report which triggers the game vulnerability analysis condition, in other words, each vulnerability detection entry in the cloud game vulnerability detection report which triggers the game vulnerability analysis condition has a pointed operation vulnerability expression.

On the premise that the vulnerability expression capturing sub-strategy is operated to have a plurality of groups of social relations, the first game type social relation information clusters pointed by each round can be combined according to the vulnerability expression sorting sub-strategy to obtain a second game type social relation information cluster, and the second game type social relation information cluster and the positioning significant information cluster are combined to obtain a third game type social relation information cluster.

Further, the first game type social relationship information cluster generated according to the local capture unit of the operation vulnerability expression capture sub-strategy can be configured to be combined.

In some examples, the first game type social relationship information clusters generated by each unit significant information capture sub-policy may be first processed according to the run-bug performance sorting sub-policy, such that the significant information size and the personality difference of each second game type social relationship information cluster are the same.

After the third game type social relationship information cluster is obtained, the vulnerability expression sorting sub-strategy can be further positioned according to the significant information of the third game type social relationship information cluster to obtain vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, for example, the vulnerability expression information can be vulnerability expression characteristics.

Further, the vulnerability expression parsing strategy can comprise a vulnerability expression capturing sub-strategy, a vulnerability expression positioning sub-strategy and a vulnerability expression sorting sub-strategy. After the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition is obtained, the cloud game vulnerability detection report triggering the game vulnerability analysis condition can be subjected to attention-based conversion processing according to the obtained operation vulnerability expression information, so that an expected vulnerability detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition can be obtained. When a vulnerability risk level detection report needs to be obtained, the step "performing attention-based conversion processing on the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the operation vulnerability expression information and the detection report conversion strategy to determine an expected vulnerability detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition" may include the following contents by way of example.

step23, processing the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the detection report conversion strategy to determine the game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

In the embodiment of the present invention, the detection report switching policy may be a GCN. The detection report conversion strategy can perform significant information capture operation on the cloud game vulnerability detection report triggering the game vulnerability analysis condition so as to obtain game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition. For example, the game type risk description may be understood as risk description information of a game type in a cloud game vulnerability detection report that triggers a game vulnerability analysis condition. In actual implementation, the game type risk is described as a running vulnerability expression improvement characteristic (optimization vector set) covering running vulnerability expression improvement indications (optimization vectors) of different vulnerability detection items of a cloud game vulnerability detection report triggering a game vulnerability analysis condition, and reference can be made to a dynamic adjustment technology in artificial intelligence and automatic control theory. The performance-improving features of the execution vulnerability can be used to capture game-type risk profiles.

For an independently implementable embodiment, the detection report conversion policy includes a global significant information capture sub-policy, a vulnerability risk level location sub-policy, and a vulnerability fix time-consuming location sub-policy. The cloud game vulnerability detection report triggering the game vulnerability analysis condition is processed according to the detection report conversion strategy to determine the game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, and the following technical solutions can be exemplarily included.

step231, conducting salient information capturing on the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the global salient information capturing sub-strategy to obtain a vulnerability detection salient information cluster, and sorting the vulnerability detection salient information cluster and a first game type social relationship information cluster generated by the vulnerability expression capturing sub-strategy of the vulnerability expression analysis strategy to determine a first sorted salient information cluster.

In the embodiment of the present invention, the global significant information capturing sub-policy means that the significant information captured by the significant information capturing sub-policy is simultaneously used to obtain a vulnerability risk level detection report and a vulnerability repair time consumption detection report. The first sorted significant information cluster obtained after sorting may include social relationship information of game types in the cloud game vulnerability detection report that triggers the game vulnerability analysis condition, and remaining significant information.

It is to be understood that the global significant information capturing sub-policy includes at least one significant information capturing node satisfying a set rule, each significant information capturing node including a run-time vulnerability performance improving sub-node. Wherein setting rules may be understood as sequential connections. The operation vulnerability representation improvement child node may be understood as a dynamic regulator of the operation vulnerability representation for performing adaptive parameter improvement.

For an independently implementable embodiment, the game-type social relationship information clusters generated by the vulnerability detection salient information cluster and the vulnerability expression capture sub-strategy of the vulnerability expression parsing strategy are collated to determine the first collated salient information cluster, which may exemplarily include the following technical solutions recorded in step2311-step 2313.

step2311, transmitting the vulnerability detection salient information cluster to a first salient information capturing node.

In the embodiment of the invention, firstly, the remaining significant information capturing sub-strategy of the detection report conversion strategy can capture significant information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition to obtain the vulnerability detection significant information cluster. Secondly, the vulnerability detection salient information cluster is transmitted to the first salient information capturing node. And carrying out subsequent processing on the vulnerability detection significant information cluster by the significant information capture node. For example, the salient information capture node may be understood as a feature information capture unit.

step2312, sorting the significant information cluster and the game type social relationship information cluster generated by the previous significant information capturing node by each significant information capturing node according to the operation vulnerability performance improvement sub-node to determine a second sorted significant information cluster pointed by the significant information capturing node; wherein the information quality in the game type social relationship information cluster pointed to by each salient information capture node is different.

In the embodiment of the invention, after the significant information capturing node collects the vulnerability detection significant information cluster, the significant information cluster and the game type social relationship information cluster generated by the previous significant information capturing node can be sorted according to the operation vulnerability performance improvement sub-node to determine a second sorted significant information cluster pointed by the significant information capturing node. Wherein the quality of information in the game type social relationship information cluster pointed to by each salient information capture node is different. The difference of the information quality can understand the individual difference degree of the game type social relationship information cluster and the difference of the levels of the remarkable information.

For the first significant information capturing node, the collected significant information cluster of vulnerability detection after significant information capturing is performed by the remaining key information extracting nodes. For the second salient information capture node, the vulnerability detection salient information cluster collected by the second salient information capture node can be the second sorted salient information cluster generated by the last salient information capture node. When the vulnerability expression capturing sub-strategy of the vulnerability expression analysis strategy is operated for only one round, the vulnerability expression capturing sub-strategy shows that only one group of first game type social relationship information clusters are derived, and all the significant information capturing nodes can arrange the single first game type social relationship information clusters and the significant information clusters generated by the previous significant information capturing node. When the vulnerability expression capture sub-strategy of the vulnerability expression analysis strategy has multiple rounds, the first game type social relationship information cluster generated by the multiple rounds of vulnerability expression capture sub-strategies can be respectively collated with the significant information cluster generated by the significant information capture node, for example, the first game type social relationship information cluster obtained by the first vulnerability expression capture sub-strategy is transmitted to the first significant information capture node, the first game type social relationship information cluster obtained by the second vulnerability expression capture sub-strategy is transmitted to the second significant information capture node, and the second significant information capturing node can sort the first game type social relationship information cluster obtained by the second operation vulnerability expression capturing sub-strategy according to the significant information cluster generated by the last significant information capturing node.

For an independently implementable embodiment, the vulnerability expression improvement child node sorts the salient information cluster and the game-type social relationship information cluster generated by the previous salient information capture node to determine a second sorted salient information cluster pointed to by the salient information capture node, which may include the following exemplary: and the vulnerability expression improvement child node is operated to update the game type social relationship information cluster into a game type social relationship information cluster with target dimensionality, for example, the individual difference and the level of the significant information of the game type social relationship information cluster are updated. And operating the vulnerability expression improvement child node, and then performing joint optimization on the updated game type social relationship information cluster and the significant information cluster generated by the previous significant information capturing node to determine a second sorted significant information cluster pointed by the significant information capturing node. For example, the operation vulnerability performance improvement sub-node of the second significant information capturing node may perform joint optimization on the second sorted significant information cluster generated by the first significant information capturing node and the game type social relationship information cluster transmitted thereto to obtain the second sorted significant information cluster pointed by the second significant information capturing node. Therefore, the operation vulnerability expression improvement sub-node also performs joint optimization through the game type social relationship information cluster and the significant information cluster generated by the last significant information capture node, and the sorting of the game type social relationship information cluster and the significant information cluster generated by the last significant information capture node is realized.

It is to be appreciated that, for each salient information capturing node, the salient information cluster generated by the previous salient information capturing node may be subjected to a redundancy removing operation (e.g., a downsampling operation) before the salient information cluster generated by the previous salient information capturing node and the game-type social relationship information cluster are sorted by the operation vulnerability performance improving sub-node to determine a second sorted salient information cluster pointed by the salient information capturing node. For example, the second salient information capture node performs a redundancy elimination operation on the second sorted salient information cluster generated by the first node. By means of the redundancy removal operation, the second sorted salient information clusters can be compacted such that the second sorted salient information clusters meet the criteria.

step2313, a second sorted salient information cluster from the last salient information capture node is determined to be the first sorted salient information cluster.

In the embodiment of the present invention, on the premise that the last round of the global significant information capturing sub-policy of the detection report conversion policy is not the last significant information capturing node, it indicates that the global significant information capturing sub-policy has a plurality of capturing units after the last significant information capturing node, and the capturing units are used for continuing to capture the second sorted significant information cluster after the last significant information capturing node derives the second sorted significant information cluster, so as to further process the sorted significant information. The salient information cluster generated after the last round of processing of the global salient information capturing sub-strategy can be the first sorted salient information cluster. For example, a second sorted salient information cluster derived from the last salient information capture node may be subjected to redundancy elimination, the second sorted salient information cluster is further refined, and then capture is performed again according to the capture unit to capture the salient information. At this time, the generated saliency cluster may be understood as the first sorted saliency cluster.

In some examples, the second sorted salient information cluster may also be directly used as the first sorted salient information cluster. Therefore, the game type social relation information cluster generated by the vulnerability expression analysis strategy and the detection report conversion strategy are sorted according to the vulnerability expression improvement sub-node, the vulnerability detection significant information cluster obtained by capturing significant information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition is enabled to be sorted, the detection report conversion strategy can be subsequently enabled to transmit the significant information obtained by the vulnerability expression analysis strategy to the detection report conversion strategy for utilization according to the game type relation information of the game type in the cloud game vulnerability detection report triggering the game vulnerability analysis condition in the game type social relation information cluster, and the attention-based conversion processing quality of the expected vulnerability detection report is improved.

It can be understood that after the first sorted significant information cluster is obtained, attention-based conversion processing can be continuously performed on the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the first sorted significant information cluster to obtain an expected vulnerability detection report.

step232, positioning the first sorted significant information cluster according to the vulnerability risk level positioning sub-strategy to determine game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

In the embodiment of the present invention, since the first sorted significant information cluster includes social relationship information of a game type and remaining significant information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, the first sorted significant information cluster may be located according to the vulnerability risk level locator policy to obtain a game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, for example, obtain a running vulnerability performance improvement characteristic indicating a running vulnerability performance improvement for each vulnerability detection entry in the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

It can be understood that the first sorted significant information cluster and the second sorted significant information cluster of at least one operation vulnerability performance improvement sub-node may be located according to the vulnerability risk level location sub-policy to determine a game type risk description of the cloud game vulnerability detection report that triggers the game vulnerability analysis condition. The vulnerability risk level positioning sub-strategy can simultaneously collect a first sorted significant information cluster generated at the end of a global significant information capturing sub-strategy of a detection report conversion strategy and a second sorted significant information cluster of at least one operation vulnerability performance improvement sub-node, and positions the two significant information clusters to obtain game type risk description of the cloud game vulnerability detection report triggering game vulnerability analysis conditions. When one round at the end of the global significant information capturing sub-strategy is a significant information capturing node, a first sorted significant information cluster generated by the last significant information capturing node and a second sorted significant information cluster generated by the operation vulnerability performance improvement sub-nodes of the rest significant information capturing nodes can be collected for positioning.

In one possible example, the number of significant information capturing nodes of the global significant information capturing sub-strategy is multiple, and the vulnerability risk level positioning sub-strategy can collect the second sorted significant information clusters generated by the multiple significant information capturing nodes at the same time to perform positioning. For example, the vulnerability risk level positioning sub-policy collects second sorted significant information clusters generated by a plurality of significant information capturing nodes having transfer relationships, and then a plurality of key information extraction sub-networks having transfer relationships are configured in the vulnerability risk level positioning sub-policy to respectively collect the second sorted significant information clusters generated by the plurality of significant information capturing nodes for positioning. For example, the first key information extraction sub-network of the vulnerability risk level positioning sub-policy may collect the first sorted significant information cluster generated by the global significant information capturing sub-policy and the second sorted significant information cluster generated by the first operation vulnerability performance improvement sub-node to perform positioning, and derive the significant information cluster. The second key information extraction sub-network of the vulnerability risk level positioning sub-strategy can perform positioning according to the significant information cluster generated by the previous key information extraction sub-network and the second sorted significant information cluster generated by the second operation vulnerability performance improvement sub-node.

It can be understood that after the key information extraction sub-network of the vulnerability risk level positioning sub-strategy is positioned according to the first sorted significant information cluster and the second sorted significant information cluster, the key information extraction sub-network is positioned according to a plurality of key information extraction sub-networks so as to update the vulnerability risk level detection report finally generated by the vulnerability risk level positioning sub-strategy.

Therefore, by collecting game type risk descriptions of cloud game vulnerability detection reports which trigger game vulnerability analysis conditions, for example, obtaining operation vulnerability expression improvement characteristics of operation vulnerability expression improvement instructions of each vulnerability detection item, adjustable risk descriptions can be determined, and the effect of attention-based conversion processing of expected vulnerability detection reports of a detection report conversion strategy under game types with frequent game type risk level changes is guaranteed.

step24, determining a vulnerability risk level detection report of the cloud game vulnerability detection report triggering game vulnerability analysis conditions according to the game type risk description and the operation vulnerability performance information.

In the embodiment of the invention, after the game type risk description of the game type in the cloud game vulnerability detection report triggering the game vulnerability analysis condition is obtained, attention-based conversion processing can be performed on the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the game type risk description and the operation vulnerability expression information generated by the operation vulnerability expression analysis strategy so as to obtain the vulnerability risk level detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition. For example, a vulnerability performance improvement feature and a vulnerability performance feature may be run to obtain a vulnerability risk level detection report that triggers a cloud game vulnerability detection report of the game vulnerability analysis condition.

It may be appreciated that the vulnerability performance improvement characteristics and the vulnerability performance characteristics may be weighted to determine vulnerability risk level detection reports that trigger cloud game vulnerability detection reports of game vulnerability analysis conditions.

Therefore, the operation vulnerability expression improvement characteristic fully depends on the player relationship network information and the game network security environment information in the game type social relationship information provided by the operation vulnerability expression analysis strategy, so that the vulnerability risk level detection report obtained by the attention-based conversion processing of the detection report conversion strategy can realize the association between the game type of the vulnerability repair time-consuming detection report and the game type of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

In the embodiment of the present invention, the detection report conversion policy may further include a bug fix time-consuming location sub-policy. Therefore, the significance information captured by the global significance information capturing sub-strategy can be used for obtaining a vulnerability repair time-consuming detection report. In view of this, after the global saliency information capture sub-policy performs saliency information capture on the cloud game bug detection report that triggers the game bug analysis condition, in other words, after step231, the method may further continue to implement the following: and positioning the first sorted significant information cluster according to a bug fixing time-consuming positioning sub-strategy to determine a bug fixing time-consuming detection report of the cloud game bug detection report triggering the game bug analysis condition.

Based on the above, the last round of the global significant information capturing sub-strategy generates a first sorted significant information cluster, where the first sorted significant information cluster includes game type social relationship information of a game type in a cloud game vulnerability detection report that triggers a game vulnerability analysis condition. Therefore, the first sorted significant information cluster can be located according to the bug fixing time-consuming location sub-strategy to determine a bug fixing time-consuming detection report of the cloud game bug detection report which triggers the game bug analysis condition.

It can be understood that the first sorted significant information cluster and the second sorted significant information cluster of at least one operation vulnerability performance improvement sub-node may be located according to the vulnerability repair time-consuming location sub-policy to determine a vulnerability repair time-consuming detection report of the cloud game vulnerability detection report that triggers the game vulnerability analysis condition. The vulnerability repair time-consuming positioning sub-strategy can simultaneously collect a first sorted significant information cluster generated one round at the end of a global significant information capturing sub-strategy of a detection report conversion strategy and a second sorted significant information cluster of at least one operation vulnerability performance improvement sub-node, and positions the two significant information clusters to obtain a vulnerability repair time-consuming detection report of the cloud game vulnerability detection report triggering game vulnerability analysis conditions. When one round at the end of the global significant information capturing sub-strategy is a significant information capturing node, a first sorted significant information cluster generated by the last significant information capturing node and a second sorted significant information cluster generated by the operation vulnerability performance improvement sub-nodes of the rest significant information capturing nodes can be collected for positioning.

In some examples, there are multiple salient information capture nodes of the global salient information capture sub-policy, and the vulnerability fix time-consuming localization sub-policy may collect a second sorted salient information cluster generated by the multiple salient information capture nodes simultaneously to perform localization. For example, if the vulnerability repair time-consuming positioning sub-policy collects second sorted significant information clusters generated by a plurality of significant information capturing nodes meeting the set rule, a plurality of key information extraction sub-networks meeting the set rule may be configured in the vulnerability repair time-consuming positioning sub-policy to respectively collect the second sorted significant information clusters generated by the plurality of significant information capturing nodes to perform positioning. For example, the first key information extraction sub-network of the vulnerability fix time-consuming positioning sub-policy may collect the first sorted significant information cluster generated by the global significant information capturing sub-policy and the second sorted significant information cluster generated by the first operation vulnerability performance improvement sub-node to perform positioning, and derive the significant information cluster. The second key information extraction sub-network of the vulnerability fix time-consuming positioning sub-strategy can perform positioning according to the significant information cluster generated by the last key information extraction sub-network and the second sorted significant information cluster generated by the second operation vulnerability performance improvement sub-node.

In some examples, after the key information extraction sub-network of the vulnerability repair time-consuming positioning sub-policy is positioned according to the first sorted significant information cluster and the second sorted significant information cluster, the key information extraction sub-network is positioned according to a plurality of key information extraction sub-networks to update the vulnerability repair time-consuming detection report finally generated by the vulnerability repair time-consuming positioning sub-policy.

Therefore, the cloud game vulnerability detection report triggering the game vulnerability analysis condition is subjected to attention-based conversion processing according to the first sorted significant information cluster containing the game type social relationship information of the game type in the cloud game vulnerability detection report triggering the game vulnerability analysis condition, adaptive vulnerability repair time is added to each vulnerability item of the game type in the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the game type social relationship information, and the attention-based conversion processing quality of the expected vulnerability detection report is guaranteed.

For an independently implementable embodiment, the detection report translation policy Stra _0010 includes: the system comprises a global significant information capturing sub-strategy Stra _0011, a vulnerability risk level positioning sub-strategy Stra _0012 and a vulnerability repair time-consuming positioning sub-strategy Stra _ 0013. The detection report conversion policy Stra _0010 may be a GCN. The global saliency information capture sub-policy Stra _0011 includes a key information extraction node _00111 and a plurality of saliency information capture nodes node _ 00112. The significant information capturing node _00112 comprises a running vulnerability performance improvement child node _ 001121. The operation vulnerability performance improvement child node note _001121 may be associated with a partial salient information capture node of the operation vulnerability performance resolution policy. The vulnerability risk level positioning sub-strategy Stra _0012 comprises a plurality of key information extraction nodes note _ 00121. The vulnerability fix time-consuming positioning sub-strategy Stra _0013 comprises a plurality of key information extraction nodes note _ 00131. The vulnerability expression improvement child node _001121 is intermittently associated with a part of key information extraction node _00121 of the vulnerability risk level positioning child strategy Stra _0012 and a part of key information extraction node _00131 of the vulnerability repair time-consuming positioning child strategy Stra _0013 respectively.

The detection report conversion policy Stra _0010 may process the cloud game vulnerability detection reports that trigger the game vulnerability analysis conditions to determine a game type risk description for the cloud game vulnerability detection reports that trigger the game vulnerability analysis conditions. The detection report conversion strategy Stra _0010 can also determine a vulnerability risk level detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the game type risk description and the operation vulnerability expression information generated by the operation vulnerability expression analysis strategy. In addition, the detection report conversion strategy Stra _0010 can also derive a bug fix time-consuming detection report.

The global significant information capturing sub-strategy Stra _0011 can capture significant information of a cloud game vulnerability detection report triggering a game vulnerability analysis condition to obtain a vulnerability detection significant information cluster, and arrange the vulnerability detection significant information cluster and a first game type social relationship information cluster generated by the vulnerability expression capturing sub-strategy of the vulnerability expression analysis strategy, so as to arrange a first arranged significant information cluster. In practical implementation, the key information extraction node _001111 disposed before the significant information capture node _00112 may capture significant information of a cloud game vulnerability detection report triggering a game vulnerability analysis condition to obtain the vulnerability detection significant information cluster mentioned in the embodiment. The salient information capturing node note _00112 may sort the salient information cluster generated by the previous salient information capturing node and the game type social relationship information cluster generated by the salient information capturing sub-policy of the vulnerability expression analysis policy according to the vulnerability expression improvement sub-node note _001121 to determine a second sorted salient information cluster pointed to by the salient information capturing node. The key information extraction node _001112, which is located after the salient information capture node _00112, may further capture the second sorted salient information cluster generated by the last salient information capture node, and finally derive the first sorted salient information cluster. The salient information capturing node _00112 may further include a redundancy removing key information extracting node _001012, which is used for performing a redundancy removing operation on the salient information cluster generated by the previous salient information capturing node.

The vulnerability risk level positioning sub-policy Stra _0012 comprises 5 key information extraction nodes. And the last round of key information extraction node note _00121 derives game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, for example, the game type risk description can be a running vulnerability performance improvement characteristic.

The vulnerability repair time-consuming positioning sub-strategy Stra _0013 comprises 5 key information extraction nodes note _00131, the key information extraction nodes note _00131 can be used for sequentially positioning a first arrangement social relationship information cluster generated by the globally significant information capturing sub-strategy, and a vulnerability repair time-consuming detection report is directly derived from the last round of key information extraction nodes note _ 00131.

For an embodiment that can be implemented independently, the deployment method of the vulnerability expression analysis policy and the detection report conversion policy shown in the present invention may exemplarily include the following contents.

Before the operation vulnerability expression analysis strategy is implemented to collect the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, the operation vulnerability expression analysis strategy and the detection report conversion strategy can be deployed.

The operation vulnerability expression analysis strategy comprises a single operation vulnerability expression capturing sub-strategy, an operation vulnerability expression positioning sub-strategy and an operation vulnerability expression sorting sub-strategy. Therefore, the operation vulnerability expression analysis strategy can be deployed according to implementation. Meanwhile, the deployment of the detection report conversion strategy can be realized.

In some examples, there is no correlation between the deployment process running the vulnerability presentation resolution policy and the detection report translation policy. When the operation vulnerability expression analysis strategy and the detection report conversion strategy are deployed, the operation vulnerability expression analysis strategy and the detection report conversion strategy can be deployed singly.

The single operation vulnerability expression capturing sub-strategy, the operation vulnerability expression positioning sub-strategy and the operation vulnerability expression sorting sub-strategy are configured for the operation vulnerability expression analysis strategy, so that the operation vulnerability expression analysis strategy can be independently deployed, the operation vulnerability expression analysis strategy can be deployed only according to the operation vulnerability expression paradigm data, the attention-based conversion processing quality of the expected vulnerability detection report can be improved, and meanwhile, the interference of the attention-based conversion processing quality of the expected vulnerability detection report caused by insufficient expected vulnerability detection report paradigm data can be weakened.

In some examples, when the operation vulnerability expression analysis policy is deployed, the operation vulnerability expression analysis policy may be deployed according to a first paradigm set, wherein operation vulnerability expression information is added to a cloud game vulnerability detection report in the first paradigm set. The operation vulnerability expression information may be, for example, operation vulnerability expressions pointed by each vulnerability detection entry in the cloud game vulnerability detection report.

It can be understood that after the operation vulnerability expression analysis policy is deployed, the paradigm operation vulnerability expression information of the cloud game vulnerability detection report in the second paradigm set can be collected according to the successfully deployed operation vulnerability expression analysis policy, and the detection report conversion policy is deployed according to the second paradigm set and the paradigm operation vulnerability expression information. The cloud game vulnerability detection report of the second exemplar set can be added with vulnerability risk level detection report actual values and vulnerability repair time-consuming detection report actual values. The second set of examples includes the first set of staged examples and the second set of staged examples. The cloud game vulnerability detection report of the first phased example set can be added with vulnerability risk level detection report actual values, and the cloud game vulnerability detection report of the second phased example set can be added with vulnerability repair time-consuming detection report actual values. Further, the data formats of the first exemplar set and the second exemplar set are different.

For an independently implementable embodiment, when the detection report conversion policy is deployed according to the second example set and the example operation vulnerability expression information, the following contents recorded in Step301 and Step302 can be implemented exemplarily.

Step301, deploying the detection report conversion strategy according to the first staged example set and the example operation vulnerability expression information pointed by the first staged example set so as to update strategy weights of the global significant information capturing sub-strategy and the vulnerability risk level positioning sub-strategy in the detection report conversion strategy.

It is to be appreciated that the runtime vulnerability expression information pointed to by the first phased set of paradigms may be obtained according to a runtime vulnerability expression resolution policy through deployment. Deployment of a global significant information capturing sub-strategy and a vulnerability risk level positioning sub-strategy in a detection report conversion strategy can be realized according to the first-stage example set added with the actual value of the vulnerability risk level detection report.

step302, deploying the detection report conversion strategy according to the second phased example set and the example operation vulnerability expression information pointed by the second phased example set so as to update the strategy weight of the global significant information capturing sub-strategy and the vulnerability fixing time-consuming positioning sub-strategy in the detection report conversion strategy.

It is to be appreciated that the second set of phased exemplars may be directed to runtime vulnerability expression information that is obtained according to a deployed runtime vulnerability expression resolution policy. After the globally significant information capturing sub-strategy and the vulnerability risk level positioning sub-strategy in the detection report conversion strategy are deployed according to the first staged example set added with the actual value of the vulnerability risk level detection report, the globally significant information capturing sub-strategy and the vulnerability repair time-consuming positioning sub-strategy in the detection report conversion strategy can be deployed. Deployment of the globally significant information capture sub-strategy and the vulnerability repair time-consuming positioning sub-strategy in the detection report conversion strategy can be achieved through the second phased example set added with the actual value of the vulnerability repair time-consuming detection report.

In the deployment process of the operation vulnerability expression analysis strategy and the detection report conversion strategy, the deployment effect can be determined according to the relevant model evaluation parameters, and then the strategy variables of each strategy are updated by using the specific values of the parameter evaluation results, so that the deployment is completed.

Based on the same inventive concept, fig. 2 shows a block diagram of a cloud computing-based cloud game vulnerability analysis apparatus according to an embodiment of the present invention, and the cloud computing-based cloud game vulnerability analysis apparatus may include the following modules for implementing the relevant method steps shown in fig. 1.

And the report collection module 210 is configured to collect a cloud game vulnerability detection report triggering a game vulnerability analysis condition in response to the game vulnerability analysis instruction.

The information obtaining module 220 is configured to collect, according to the operation vulnerability expression analysis policy, operation vulnerability expression information of the cloud game vulnerability detection report that triggers the game vulnerability analysis condition.

And the report conversion module 230 is configured to perform attention-based conversion processing on the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the detection report conversion policy according to the operation vulnerability expression information to determine an expected vulnerability detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

The related embodiment applied to the invention can achieve the following technical effects: by collecting the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, the detection report conversion strategy can be ensured to deeply analyze game interaction data of game types in the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the operation vulnerability expression information, so that an expected vulnerability detection report obtained by the attention-based conversion processing of the detection report conversion strategy can be associated with the game types of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, and the attention-based conversion processing quality of the expected vulnerability detection report is ensured; in addition, the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition is obtained according to an operation vulnerability expression analysis strategy which is not related to the detection report conversion strategy, and accurate and reliable operation vulnerability expression information can be obtained through the differentiated use of the strategies, so that the association degree of an expected vulnerability detection report obtained through attention-based conversion processing and the game type of the cloud game vulnerability detection report triggering the game vulnerability analysis condition is guaranteed, and the expected vulnerability detection report and the corresponding game type can be subjected to combined vulnerability repair processing conveniently.

The foregoing is only illustrative of the present invention. Those skilled in the art can conceive of changes or substitutions based on the specific embodiments provided by the present invention, and all such changes or substitutions are intended to be included within the scope of the present invention.

Claims

1. A cloud game vulnerability analysis method based on cloud computing is applied to a cloud game server and comprises the following steps:

responding to a game vulnerability analysis instruction, and collecting cloud game vulnerability detection reports which trigger game vulnerability analysis conditions;

collecting the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to an operation vulnerability expression analysis strategy;

and according to the operation vulnerability expression information, performing attention-based conversion processing on the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to a detection report conversion strategy to determine an expected vulnerability detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

2. The method of claim 1, wherein the expected vulnerability detection report comprises a vulnerability risk level detection report; the method for determining the expected vulnerability detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition by performing attention-based conversion processing on the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the operation vulnerability expression information and a detection report conversion strategy comprises the following steps:

processing the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to a detection report conversion strategy to determine the game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition;

and determining a vulnerability risk level detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the game type risk description and the operation vulnerability performance information.

3. The method of claim 2, wherein the game type risk is described as a performance vulnerability performance improvement characteristic of a performance vulnerability performance improvement indication covering different vulnerability detection entries of the cloud game vulnerability detection report of the triggered game vulnerability analysis condition, and the performance vulnerability performance information is a performance vulnerability performance characteristic of a performance vulnerability performance covering different vulnerability detection entries of the cloud game vulnerability detection report of the triggered game vulnerability analysis condition;

the vulnerability risk level detection report for determining the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the game type risk description and the operation vulnerability performance information comprises the following steps: weighting the operation vulnerability performance improvement characteristics and the operation vulnerability performance characteristics to determine a vulnerability risk level detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis conditions.

4. The method of claim 3, wherein the detection report conversion policy comprises a global significant information capture sub-policy and a vulnerability risk level location sub-policy; the processing the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the detection report conversion strategy to determine the game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition comprises the following steps:

according to the global significant information capturing sub-strategy, significant information capturing is carried out on the cloud game vulnerability detection report triggering the game vulnerability analysis condition to obtain a vulnerability detection significant information cluster, and the vulnerability detection significant information cluster and a first game type social relationship information cluster generated by the vulnerability expression capturing sub-strategy of the vulnerability expression analysis strategy are sorted to determine a first sorted significant information cluster;

positioning the first sorted significant information cluster according to the vulnerability risk level positioning sub-strategy to determine game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition; and the deployment process of the operation vulnerability expression analysis strategy and the detection report conversion strategy is not associated.

5. The method of claim 4, wherein the global significant information capture sub-policy includes no less than one significant information capture node that satisfies a set rule, each of the significant information capture nodes including a run-time vulnerability performance improvement sub-node; the sorting the vulnerability detection significant information cluster and a first game type social relationship information cluster generated by a vulnerability expression capturing sub-strategy of the vulnerability expression analysis strategy to determine a first sorted significant information cluster comprises:

transmitting the vulnerability detection salient information cluster to the first salient information capturing node;

for each significant information capturing node, sorting a significant information cluster generated by the previous significant information capturing node and the first game type social relationship information cluster according to the operation vulnerability performance improvement sub-node to determine a second sorted significant information cluster pointed by the significant information capturing node; wherein the quality of information in the game type social relationship information cluster pointed to by each salient information capture node is different;

determining the first sorted salient information cluster from the second sorted salient information cluster of the target salient information capture node.

6. The method of claim 5, wherein prior to the sorting of the salient information cluster generated by the last one of the salient information capture nodes and the game-type social relationship information cluster by the performance-improving child node in accordance with the vulnerability to determine a second sorted salient information cluster to which the salient information capture node points, the method further comprises: carrying out redundancy removing operation on the significant information cluster generated by the previous significant information capturing node;

the sorting of the significant information cluster generated by the previous significant information capturing node and the game type social relationship information cluster by the child node according to the operation vulnerability performance improvement to determine a second sorted significant information cluster pointed by the significant information capturing node includes: updating the game type social relationship information cluster into a game type social relationship information cluster with a target dimension according to the operation vulnerability performance improvement sub-node, and performing joint optimization on the updated game type social relationship information cluster and a significant information cluster generated by a previous significant information capture node to determine a second sorted significant information cluster pointed by the significant information capture node;

wherein, according to the vulnerability risk level positioning sub-strategy, positioning the first sorted significant information cluster to determine the game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition includes: and positioning the first sorted significant information cluster and at least one second sorted significant information cluster of the operation vulnerability performance improvement sub-node according to the vulnerability risk level positioning sub-strategy to determine the game type risk description of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

7. The method of claim 6, wherein the detection report conversion policy further comprises a vulnerability fix time-consuming location sub-policy; the method comprises the following steps of performing attention-based conversion processing on a cloud game vulnerability detection report triggering a game vulnerability analysis condition according to a detection report conversion strategy according to the operation vulnerability expression information to determine an expected vulnerability detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition, and further comprises the following steps:

positioning the first sorted significant information cluster according to the bug fixing time-consuming positioning sub-strategy to determine a bug fixing time-consuming detection report of the cloud game bug detection report triggering the game bug analysis condition;

the method for detecting the bug fixing time consumption of the cloud game bug detection report triggering the game bug analysis condition comprises the following steps of: and positioning the first sorted significant information cluster and at least one second sorted significant information cluster of the operation vulnerability performance improvement sub-node according to the vulnerability repair time-consuming positioning sub-strategy so as to determine a vulnerability repair time-consuming detection report of the cloud game vulnerability detection report triggering the game vulnerability analysis condition.

8. The method according to any one of claims 1 to 7, wherein the vulnerability expression analysis strategy comprises a vulnerability expression capture sub-strategy, a vulnerability expression positioning sub-strategy and a vulnerability expression sorting sub-strategy; the method for collecting the running vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis conditions according to the running vulnerability expression analysis strategy comprises the following steps:

capturing the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the operation vulnerability expression capturing sub-strategy to determine a first game type social relationship information cluster;

positioning the first game type social relationship information cluster according to the operation vulnerability presentation positioning sub-strategy to determine a positioning significant information cluster;

sorting the first game type social relationship information cluster and the positioning significant information cluster according to the operation vulnerability expression sorting sub-strategy to determine operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition;

before collecting the operation vulnerability expression information of the cloud game vulnerability detection report triggering the game vulnerability analysis condition according to the operation vulnerability expression analysis strategy, the method further comprises the following steps: deploying and obtaining the operation vulnerability expression analysis strategy according to a first example set, wherein operation vulnerability expression information is added to a cloud game vulnerability detection report in the first example set; and collecting example operation vulnerability expression information of the cloud game vulnerability detection report in a second example set according to the successfully deployed operation vulnerability expression analysis strategy, and deploying the detection report conversion strategy according to the second example set and the example operation vulnerability expression information.

9. The method of claim 8, wherein the second set of paradigms comprises a first set of phased paradigms and a second set of phased paradigms, the deploying the detection report conversion policy in accordance with the second set of paradigms and the paradigms operating vulnerability performance information comprises:

deploying the detection report conversion strategy according to the first staged example set and the example operation vulnerability performance information pointed by the first staged example set so as to update strategy weights of a global significant information capturing sub-strategy and a vulnerability risk level positioning sub-strategy in the detection report conversion strategy;

deploying the detection report conversion strategy according to the second phased example set and the example operation vulnerability performance information pointed by the second phased example set so as to update strategy weights of a global significant information capturing sub-strategy and a vulnerability fixing time-consuming positioning sub-strategy in the detection report conversion strategy.

10. A cloud gaming server, comprising: a memory and a processor; the memory and the processor are coupled; the memory for storing computer program code, the computer program code comprising computer instructions; wherein the computer instructions, when executed by the processor, cause the cloud gaming server to perform the method of any of claims 1-9.