Risk assessment method and device, computing equipment and computer readable storage medium

Abstract

Embodiments of the invention provide a risk assessment method and apparatus, a computing device and a computer readable storage medium. The risk assessment method comprises the steps of obtaining a planning text for a change activity of a target project; utilizing a rule extraction model to perform structured rule analysis on the planning text, and extracting at least one piece of activity rule data of the change activity; comparing the at least one activity rule data with the activity rule code of the changed activity to obtain a comparison result; and performing change risk assessment on the change activity based on the comparison result by using a risk assessment model to obtain a risk assessment result of the planning text. The planning text is analyzed through the structured rule, the activity rule data is automatically and accurately extracted, the activity rule data is compared with the code, the potential difference between the text and the code is accurately positioned, and the potential risk of the planning text in actual execution is analyzed based on the comparison result. And the flexibility, accuracy and efficiency of risk assessment are enhanced.

Description

Technical Field

[0001] The embodiments of this specification relate to the technical field of information technology, and in particular, to a risk assessment method, apparatus, computing device, and computer-readable storage medium. Background Art

[0002] With the development of information technology, organizations often launch complex project activities, which involve various specific project activity rules, such as customer rules and reward mechanisms in digital marketing and e-commerce activities.

[0003] At present, the activity rule codes and planning texts of project activities rely on manual configuration and writing. However, with the rapid changes in project requirements, project activity rules have become more complex and updated more frequently. There may be potential inconsistencies between the actual execution methods corresponding to the planning text and the activity rule code. It is impossible to accurately judge the logical conflicts or potential compliance issues between complex rules. The release of erroneous planning texts may increase the risks in the actual project implementation process. The management of planning texts is insufficiently flexible, accurate, and efficient. Therefore, a highly flexible, accurate, and efficient risk assessment method is urgently needed. Summary of the invention

[0004] In view of this, an embodiment of this specification provides a risk assessment method. One or more embodiments of this specification also relate to a risk assessment device, a computing device, a computer-readable storage medium and a computer program product to solve the technical defects existing in the prior art.

[0005] According to a first aspect of an embodiment of this specification, a risk assessment method is provided, comprising: Obtain the planning text of the change activities for the target project; Using the rule extraction model, the planning text is parsed for structured rules to extract at least one activity rule data of the change activity; Compare at least one activity rule data with the activity rule code of the changed activity to obtain a comparison result; Using the risk assessment model, conduct a change risk assessment on the change activity based on the comparison results to obtain the risk assessment results of the planning text.

[0006] According to a second aspect of an embodiment of this specification, a risk assessment device is provided, comprising: A text acquisition module is configured to acquire a planning text of a change activity for a target project; A rule extraction module is configured to use a rule extraction model to perform structured rule parsing on the planning text to extract at least one activity rule data of the change activity; a change detection module configured to compare at least one activity rule data with an activity rule code of a changed activity to obtain a comparison result; The risk assessment module is configured to use the risk assessment model to perform change risk assessment on the change activity based on the comparison result to obtain the risk assessment result of the planning text.

[0007] According to a third aspect of an embodiment of this specification, a computing device is provided, including: Memory and processor; The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the above method are implemented.

[0008] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided, which stores computer-executable instructions, and the instructions implement the steps of the above method when executed by a processor.

[0009] According to a fifth aspect of the embodiments of this specification, a computer program product is provided, wherein when the computer program product is executed in a computer, the computer is caused to execute the steps of the above method.

[0010] In one embodiment of the present specification, a planning text of a change activity for a target project is obtained; a rule extraction model is used to perform structured rule parsing on the planning text to extract at least one activity rule data of the change activity; the at least one activity rule data is compared with the activity rule code of the change activity to obtain a comparison result; a risk assessment model is used to perform a change risk assessment on the change activity based on the comparison result to obtain a risk assessment result of the planning text.

[0011] By parsing the planning text with structured rules, the automatic and accurate extraction of activity rule data is achieved. The extracted activity rule data is compared with the activity rule code, realizing the conversion from unstructured text to structured data, accurately locating the potential differences between text and code, and improving the accuracy and consistency of understanding of change activities. Using the structured comparison results as input, the risk assessment model analyzes the potential risks of the planning text in actual implementation, obtains risk assessment results, and enhances the flexibility, accuracy and efficiency of risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 is a flow chart of a risk assessment method provided by an embodiment of this specification; Figure 2 is a process flow chart of a risk assessment method applied to digital marketing and e-commerce activities provided by an embodiment of this specification; Figure 3 It is a schematic diagram of the structure of a risk assessment device provided by an embodiment of this specification; Figure 4 It is a structural function diagram of a risk assessment device provided by an embodiment of this specification; Figure 5 It is a structural block diagram of a computing device provided by an embodiment of this specification. DETAILED DESCRIPTION

[0013] Many specific details are described in the following description to facilitate a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of this specification, so this specification is not limited to the specific implementation disclosed below.

[0014] The terms used in one or more embodiments of this specification are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of this specification. The singular forms of "a", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or possible combination of one or more associated listed items.

[0015] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0016] In addition, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0017] First, the terms involved in one or more embodiments of this specification are explained.

[0018] Large Language Model (LLM): Also known as a large model, it is a large-scale neural network model trained by a deep learning algorithm. It has powerful natural language understanding and generation capabilities and can handle complex text parsing and information extraction tasks.

[0019] Marketing campaign configuration: The rules and parameter settings involved in the company's marketing activities are constructed into the activity rule code of the marketing activities, such as activity time, participation conditions, award rules, etc.

[0020] Marketing activity rules for customers: marketing activity rules for customers, such as conditions and reward mechanisms specified in the planning copy.

[0021] Activity rule extraction: Use natural language processing technology to extract structured data from unstructured text.

[0022] Marketing activity risk assessment: In the process of project management and decision-making, risk assessment is the process of identifying, analyzing and prioritizing potential risks associated with specific changes or activities. It involves quantitatively or qualitatively evaluating the likelihood of risk and its impact to determine the necessary response measures. For marketing activity configuration, risk assessment aims to analyze the planning text of the marketing activity and the corresponding activity rule code through a systematic approach, detect the consistency and logical integrity between the two, and thus discover and quantify possible risk factors in advance, such as compliance issues, logical conflicts or execution deviations. Using large language models (LLM) and rule extraction technology, it is possible to automatically parse and predict the risks of customer rules in marketing activity configuration, provide scientific risk assessment results, and ensure the effectiveness and safety of marketing activities.

[0023] In this specification, a risk assessment method is provided. This specification also relates to a risk assessment device, a computing device, a computer-readable storage medium and a computer program product, which are described in detail one by one in the following embodiments.

[0024] See also Figure 1 , Figure 1 A flow chart of a risk assessment method provided according to an embodiment of the present specification is shown, comprising the following specific steps: Step 102: Obtain the planning text of the change activity for the target project.

[0025] The embodiments of this specification are applied to applications, websites or system platforms with risk assessment capabilities, for example, the backend system of an e-commerce platform for managing promotional and marketing activities, or the risk assessment module integrated in the enterprise resource planning (ERP) system to support risk analysis of changes in internal enterprise processes, or the digital marketing service platform for optimizing online marketing and advertising strategies.

[0026] A target project is a specific project of a specific organization, including a set of activity rules with clear goals, scope and time frame. The target project contains a series of activity rules with predefined operation logic and parameters to guide the execution process of the project. Activity rules are configured in the form of code to ensure that the project is automatically executed according to the predetermined plan. For example, product development and release projects, enterprise management projects, platform marketing projects, advertising projects, etc.

[0027] Change activities for target projects are modifications or updates to the activity rules for target projects. Change activities are adjustments made in response to new project requirements, which may involve changes in operating logic and parameters. For example, in a platform marketing project, a marketing campaign needs to be launched to increase user engagement.

[0028] The planning text of the change activity is a natural language description text of the change activity, which describes the specific activity content of the change activity, including the purpose of the change, implementation method, time schedule, etc. The planning text is generally manually written and published to ensure that the relevant parties understand the content and implementation method of the change.

[0029] To obtain the planning text of the change activity for the target project, one optional way is to receive the planning text of the change activity for the target project uploaded by the user from the data interface. Another optional way is to obtain the planning text of the change activity for the target project from the activity planning text library. Another optional way is to use a data acquisition tool to obtain the planning text of the change activity for the target project from the front-end page of the target project. There is no limitation here.

[0030] For example, on a resource management platform, in order to increase user participation in a resource management product, a change activity needs to be launched. Receive the planning text of the change activity uploaded by the user from the data interface, which is completed and about to be released: 1. This event is open to a specific user group. Invited users can participate in the event on the designated page and complete relevant operations on the page to receive rewards.

[0031] 2. The user needs to follow the instructions on the page, select "Transfer in" or "Transfer resource quota", and complete the transfer of resource quota from other accounts to the specified account according to the prompts. After successful completion, there will be a chance to receive corresponding rewards. Please note that only resource quota transfers completed through the above specified process are eligible for rewards.

[0032] 3. Each time a certain amount of resources is successfully transferred, one reward will be obtained. Specifically, one reward will be obtained for each two units of resources transferred, and the maximum cumulative reward can be five. For example, if two units of resources are transferred, one reward will be obtained; if ten units of resources are transferred, a total of five rewards will be obtained.

[0033] 4. This activity starts on December 11, 2024 and ends on December 31, 2024.

[0034] By obtaining the planned text of the change activities for the target project, text input is provided for the subsequent structured rule parsing.

[0035] Step 104: Utilize the rule extraction model to perform structured rule parsing on the planning text, and extract at least one activity rule data of the change activity.

[0036] The rule extraction model is a neural network model with structured parsing and information extraction capabilities. The rule extraction model uses natural language understanding capabilities to complete the structured parsing of input data, and further extracts structured information on this basis. The rule extraction model can be a pre-supervised neural network model, such as the Long Short-Term Memory (LSTM) network model, the self-attention deep learning model (Transformer), the Bidirectional Encoder Representations from Transformers (BERT), the T5 model (Text-to-Text Transfer Transformer), or it can be directly implemented by a large language model under the guidance of prompt information.

[0037] Structured rule parsing is the process of converting unstructured natural language text into structured data. This process usually involves text segmentation, named entity recognition, relationship extraction, etc., and then extracts structured data that is easy to be processed by computers. Structured rule parsing can capture the semantic information in the text and convert it into a data structure with clear meaning, such as key-value pairs, tables, or JSON objects.

[0038] The activity rule data of the change activity is a structured data representation that describes the operation logic and parameters of the change activity extracted from the planning text. The activity rule data of the change activity exists in a form that is easy for computers to process, ensuring the clarity and consistency of the rules. The activity rule data contains the necessary rules and conditions for executing the change activity, ensuring that the change activity is executed as expected, and is usually presented in a structured format, such as JSON, XML, or a table.

[0039] Using a rule extraction model, the planning text is parsed according to structured rules to extract at least one activity rule data for the change activity. An optional method is to use a rule extraction model that has been pre-supervised and trained to perform structured rule parsing on the planning text to extract at least one activity rule data for the change activity, wherein the rule extraction model is pre-trained based on sample planning text and labeled activity rule data. Another optional method is to use a large language model and prompt information to perform structured rule parsing on the planning text to extract at least one activity rule data for the change activity.

[0040] For example, the prompt information is: You are now a marketing rule risk assessment expert. You can perform JSON structured rule parsing on the planning text and extract the activity rule data for the changed activity (such as activity cycle, award rules, changes in the activity audience, etc.): enter: 1. This event is open to a specific user group. Invited users can participate in the event on the designated page and complete relevant operations on the page to receive rewards.

[0041] 2. The user needs to follow the instructions on the page, select "Transfer in" or "Transfer resource quota", and complete the transfer of resource quota from other accounts to the specified account according to the prompts. After successful completion, there will be a chance to receive corresponding rewards. Please note that only resource quota transfers completed through the above specified process are eligible for rewards.

[0042] 3. Each time a certain amount of resources is successfully transferred, one reward will be obtained. Specifically, one reward will be obtained for each two units of resources transferred, and the maximum cumulative reward can be five. For example, if two units of resources are transferred, one reward will be obtained; if ten units of resources are transferred, a total of five rewards will be obtained.

[0043] 4. This activity starts on December 11, 2024 and ends on December 31, 2024.

[0044] Using the large language model, the above prompt information is used to perform structured rule analysis on the planning text, and the three activity rule data of the change activity are extracted: Activity period: "gmtBegin": "2024-12-11T00:00:00Z", / / Activity start time.

[0045] "gmtEnd": "2024-12-31T23:59:59Z", / / Activity end time.

[0046] Award Rules: "prizeCountFrequencyModel": { / / Reward frequency restriction rules "countFrequencyDimension": ["USER_ID"], / / The user ID (USER_ID) that defines the frequency limit means that the frequency limit is for each user.

[0047] "userFrequencyLimit": 1, / / Sets the maximum number of rewards for each user within a specified time unit. The value here is 1, which means that each user can be rewarded at most once a day.

[0048] "userFrequencyLimitControl": "notLimit" / / Specifies whether to enable additional frequency limits. Setting it to notLimit means that no other limits are imposed except for the frequency defined above.

[0049] }, "prizeCountLimitModel": { / / Defines the total number of times a user can receive rewards during the entire event.

[0050] "limitControl": "limit", / / Control whether to limit the number of rewards. If set to limit, it means to enable the limit on the number of rewards.

[0051] "limitCount": 5, / / specifies the maximum number of rewards each user can receive during the entire activity. The value here is 5, which means each user can receive a maximum of 5 rewards.

[0052] "limitDimension": ["USER_ID"] / / Defines the dimension of the number limit, which is the same as the frequency limit. Here it is the user ID (USER_ID), which means that the number limit is also for each user.

[0053] }, Changes to the event audience: "populationRule": { / / Activity audience.

[0054] "target_audience": { / / Target users.

[0055] "eligible_users": "invited_users", / / The event is only open to invited users.

[0056] "participation_conditions": ["complete_resource_transfer_via_designated_page"] / / Users must complete the resource transfer via the designated page to be eligible for the reward.

[0057] } }, By parsing the planning text with structured rules, the automatic and accurate extraction of activity rule data is achieved, providing a data basis for subsequent comparison.

[0058] Step 106: Compare at least one activity rule data with the activity rule code of the changed activity to obtain a comparison result.

[0059] The activity rule code of a change activity is a specific implementation written in a specific code language that defines the operation logic and parameters in the change activity. The activity rule code of a change activity is usually written by developers and deployed to guide the automated execution process of the change activity. The activity rule code contains the operation logic such as the rules and conditions for executing the change activity to ensure that the change activity is executed as expected, and is usually presented in a structured format such as JSON, XML or a table. In theory, the activity rule code needs to be consistent with the planning text to ensure that the marketing activities are executed accurately according to the planning requirements. However, in reality, with the rapid changes in project requirements, project activity rules have become more complex and updated more frequently. There may be potential inconsistencies between the actual execution methods corresponding to the planning text and the activity rule code. Therefore, a comparison is required.

[0060] The comparison result is the result determined by comparing the extracted activity rule data with the configured activity rule code, which is used to evaluate the consistency and difference between the two. The comparison result includes whether there is a difference, the specific difference content, the difference degree of the difference content and the affected object.

[0061] Compare at least one activity rule data with the activity rule code of the changed activity to obtain a comparison result. An optional way is to use a change detection model to compare at least one activity rule data with the activity rule code of the changed activity to obtain a comparison result. On this basis, further, an optional way is to use a change detection model that has been pre-supervised and trained to compare at least one activity rule data with the activity rule code of the changed activity to obtain a comparison result, wherein the change detection model is pre-trained based on sample activity rule data, sample activity rule code and label comparison results. Another optional way is to use a large language model to compare at least one activity rule data with the activity rule code of the changed activity using prompt information to obtain a comparison result.

[0062] For example, the prompt information is: You are now a marketing rule risk assessment expert. You can compare the activity rule data and activity rule code to obtain the comparison results (whether there is a conclusion that the activity rule data and activity rule code do not match; if there is a configuration that does not match the rule description, point out the specific content of the discrepancy in the planning text): enter: Activity rules data: Activity period: "gmtBegin": "2024-12-11T00:00:00Z", / / Activity start time.

[0063] "gmtEnd": "2024-12-31T23:59:59Z", / / Activity end time.

[0064] Award Rules: "prizeCountFrequencyModel": { / / Reward frequency restriction rules "countFrequencyDimension": ["USER_ID"], / / The user ID (USER_ID) that defines the frequency limit means that the frequency limit is for each user.

[0065] "userFrequencyLimit": 1, / / Sets the maximum number of rewards for each user within a specified time unit. The value here is 1, which means that each user can be rewarded at most once a day.

[0066] "userFrequencyLimitControl": "notLimit" / / Specifies whether to enable additional frequency limits. Setting it to notLimit means that no other limits are imposed except for the frequency defined above.

[0067] }, "prizeCountLimitModel": { / / Defines the total number of times a user can receive rewards during the entire event.

[0068] "limitControl": "limit", / / Control whether to limit the number of rewards. If set to limit, it means to enable the limit on the number of rewards.

[0069] "limitCount": 5, / / specifies the maximum number of rewards each user can receive during the entire activity. The value here is 5, which means each user can receive a maximum of 5 rewards.

[0070] "limitDimension": ["USER_ID"] / / Defines the dimension of the number limit, which is the same as the frequency limit. Here it is the user ID (USER_ID), which means that the number limit is also for each user.

[0071] }, Changes to the event audience: "populationRule": { / / Activity audience.

[0072] "target_audience": { / / Target users.

[0073] "eligible_users": "invited_users", / / The event is only open to invited users.

[0074] "participation_conditions": ["complete_resource_transfer_via_designated_page"] / / Users must complete the resource transfer via the designated page to be eligible for the reward.

[0075] } }, Activity rule JSON code: { "bizOwner": "xxx", / / Project owner, refers to the project person or department responsible for this activity.

[0076] "id": "Rule ID", / / The unique identifier of the rule, used to identify and manage specific activity rules within the system.

[0077] "campTag": "", / / Campaign tags can be used to classify or mark different marketing activities.

[0078] "campTopic": "", / / Activity topic, describing the main content or purpose of the activity.

[0079] "countFrequencyModel": { / / Dimension of reward frequency limit "countFrequencyDimension": [ "USER_ID" ], / / Defines the user ID (USER_ID) for frequency limit, which means that the frequency limit is for each user.

[0080] "userFrequencyLimit": "xx", / / Maximum number of rewards a user can receive within a specified time unit. A specific value should be filled in here.

[0081] "userFrequencyLimitControl": "notLimit" / / Whether to enable additional frequency limits. Setting it to notLimit means that no other limits are imposed except for the frequency defined above.

[0082] }, "countLimitModel": { / / Dimension of the reward limit "limitControl": "notLimit", / / Controls whether to limit the number of rewards. If set to notLimit, it means that the number of rewards is not limited.

[0083] "limitCount": "xx", / / Specifies the maximum number of rewards each user can receive during the entire activity. A specific value should be filled in here.

[0084] "limitDimension": [ "USER_ID" ] / / Defines the dimension of the number of times limit, here it is the user ID (USER_ID), which means the number of times limit is also for each user.

[0085] }, "creator": "xxx", / / Creator information, records the person or system that created the activity rule.

[0086] "gmtBegin": "Activity start time", / / The start time of the activity, which should use a specific date and time format (such as ISO 8601).

[0087] "gmtEnd": "Event end time", / / The end time of the event, which should also use a specific date and time format.

[0088] "groupCountLimit": "", / / Group-level reward limit, empty string if not defined.

[0089] "hasPageConfig": false, / / Whether there is a page configuration. False means there is no dedicated page configuration.

[0090] "idCardNoIsAllowEmpty": "false", / / Whether to allow the identity card to be empty. Set to false to disallow empty values.

[0091] "isMobileOsFilterWithTagAllow": "true", / / Mobile operating system filter flag. Setting this to true means filtering based on mobile OS tags is allowed.

[0092] "limitCampConsultProcess": "false", / / Whether to limit the activity consultation process, false means no limit.

[0093] "limitCampProcessTemplate": "false", / / Whether to limit the activity process template, false means no limit.

[0094] "name": "xxx", / / Activity name, used for internal management and external display.

[0095] "notify": "", / / Notification settings, which can include information such as the method and conditions for sending notifications.

[0096] "outBizId": "xxx", / / External project ID, used to associate the project entity of the external system.

[0097] "planId": "xxxx", / / Activity plan ID, used to distinguish activity plans of different stages or versions.

[0098] "populationRule": "", / / Activity audience rule, empty string if undefined.

[0099] "prizeConfigModel": { "prizeAddType": "manualAdd", / / Reward adding type, manualAdd means manually adding rewards.

[0100] "prizeDecision": { "decisionType": "WEIGHT", / / Decision type. WEIGHT means the reward is determined by weight.

[0101] "multiplePrizeSendLimit": 1 / / The limit for issuing multiple rewards at a time. 1 means only one reward can be issued at a time.

[0102] } }, "prizeCountFrequencyModel": { / / Reward frequency limit "countFrequencyDimension": [ "USER_ID" ], / / Defines the user ID (USER_ID) for frequency limit, which means that the frequency limit is for each user.

[0103] "userFrequencyLimit": 1, / / Sets the maximum number of rewards for each user within a specified time unit. The value here is 1, which means that each user can be rewarded at most once a day.

[0104] "userFrequencyLimitControl": "notLimit" / / Specifies whether to enable additional frequency limits. Setting it to notLimit means that no other limits are imposed except for the frequency defined above.

[0105] }, "prizeCountLimitModel": { / / Reward count limit "limitControl": "limit", / / Controls whether to limit the number of rewards. If set to limit, it means enabling the limit on the number of rewards.

[0106] "limitCount": 1, / / Note: The original value here is 1, while the value in the planning text is 5, which specifies the maximum number of rewards each user can receive during the entire activity. This value should be consistent with the planning text, that is, changed to 5.

[0107] "limitDimension": [ "USER_ID" ] / / Defines the dimension of the number limit, which is the same as the frequency limit. Here it is the user ID (USER_ID), which means that the number limit is also for each user.

[0108] }, "userRuleModel": { / / Activity crowd rules "crowdLimitType": "xxxx", / / Crowd restriction type, used to define which groups can participate in the activity.

[0109] "promoRuleUUid": "xxxx", / / Promotion rule UUID, used to uniquely identify the promotion rule.

[0110] "promoRuleUUidControl": "limit", / / Promotion rule control. Setting it to limit means applying this rule for restriction.

[0111] "verified": false / / Whether it has been verified. False means it has not been verified.

[0112] } } Using the large language model, the three activity rule data and the activity rule JSON code of the changed activity are compared using the above prompt information to obtain the comparison results: The activity audience is not configured in the activity rule code, which is inconsistent with the changes in the activity audience ("eligible_users": "invited_users",; "participation_conditions": ["complete_resource_transfer_via_designated_page"]).

[0113] The extracted activity rule data is compared with the activity rule code to achieve the conversion from unstructured text to structured data, accurately locate the potential differences between text and code, and improve the accuracy and consistency of understanding of change activities.

[0114] Step 108: Using the risk assessment model, perform a change risk assessment on the change activity based on the comparison results to obtain the risk assessment results of the planning text.

[0115] The risk assessment model is a neural network model that analyzes the potential risks of the planned text in actual execution. By comparing and processing the input of the results, it identifies potential risk factors, such as compliance issues, logical conflicts or execution deviations, and predicts the possibility of their occurrence and their impact. The risk assessment model can be a pre-supervised neural network model, such as a long short-term memory network model, a self-attention deep learning model, a bidirectional encoding representation Transformer model, a T5 model, or it can be directly implemented by a large language model under the guidance of prompt information.

[0116] Change risk assessment is a process of discovering and quantifying possible risk factors in advance based on the consistency and logical integrity between change activities. This process involves not only qualitative analysis, but also quantitative data support to better understand the possible impact and severity of risks.

[0117] The risk assessment results of the planning document are the risk assessment results of the planned document of the change activity that may be faced during the actual implementation process. The risk assessment results may include the specific content of the risk, the probability of occurrence, the possible impact, and the feasible strategy.

[0118] Using the risk assessment model, a change risk assessment is performed on the change activity based on the comparison results to obtain the risk assessment results of the planned text. An optional method is: using a risk assessment model trained in advance through supervision, a change risk assessment is performed on the change activity based on the comparison results to obtain the risk assessment results of the planned text, wherein the risk assessment model is pre-trained based on sample comparison results and label risk assessment results. Another optional method is: using a large language model, using prompt information to perform a change risk assessment on the change activity based on the comparison results to obtain the risk assessment results of the planned text.

[0119] Based on the comparison results, a change risk assessment is performed on the change activity to obtain the risk assessment results of the planning text. An optional method is: based on the comparison results, determine whether there is a change risk in the change activity. If so, identify the risk type of the change activity, and based on the risk type, generate the risk assessment results of the planning text.

[0120] For example, the prompt information is: You are now a marketing rule risk assessment expert. Based on the comparison results, determine whether the change activity has change risks. If so, identify the risk type of the change activity and generate risk assessment results for the planning text based on the risk type: enter: Comparison results: The activity audience is not configured in the activity rule code, which is inconsistent with the changes in the activity audience ("eligible_users": "invited_users",; "participation_conditions": ["complete_resource_transfer_via_designated_page"]).

[0121] Using the large language model, the above prompt information is used to conduct a change risk assessment on the change activity based on the comparison results to obtain the risk assessment results of the planning text: The activity audience rules are not configured in the activity rule code, but the planning text clearly stipulates that the activity is only open to invited users, and users must complete the resource quota transfer through the designated page to be eligible for the reward.

[0122] Risk Type: Compliance risk: Lack of clear definition of campaign audiences may result in non-compliance with internal or external compliance requirements.

[0123] Participation risk: If all users can participate in the event instead of just invited users, it may lead to unexpectedly high participation, increasing operating costs and system load.

[0124] User experience risk: Incorrect rule implementation may cause user confusion and reduce user satisfaction, especially when users find that they do not meet the conditions.

[0125] Probability of occurrence: Due to the lack of necessary restrictions, any user may try to participate in the activity, which is contrary to the provisions of the planning text.

[0126] Impact: Serious. Not only will it affect the cost control of the activity, it may also damage the reputation of the resource management platform and require additional customer service support to handle user questions or complaints.

[0127] Suggested actions: Update the activity rule code immediately, add the activity audience rules, and ensure that they are consistent with the planning text. At the same time, check whether there are other places involving similar rule settings to avoid similar problems.

[0128] In the embodiments of the present specification, the planning text is parsed by structured rules to achieve automatic and accurate extraction of activity rule data, and the extracted activity rule data is compared with the activity rule code to achieve conversion from unstructured text to structured data, accurately locate the potential differences between the text and the code, improve the accuracy and consistency of understanding of the change activity, use the structured comparison results as input, and the risk assessment model analyzes the potential risks of the planning text in actual implementation, obtains risk assessment results, and enhances the flexibility, accuracy and efficiency of risk assessment.

[0129] In an optional embodiment of the present specification, before step 106, the following specific steps are also included: Retrieve the activity rule code for the change activity from the activity rule code repository of the target project.

[0130] The target project's activity rule code base is a database that stores the configuration code of the target project's project activities. The code base records the target project's historical project activities and the current project activity rule code through version control, so as to track the change history, manage the differences between different versions, and support team collaborative development. Such as open source code repositories, enterprise-level code bases, etc.

[0131] Retrieve the activity rule code for the change activity from the activity rule code library of the target project. One optional way is to retrieve the activity rule code for the change activity from the activity rule code library of the target project based on the activity identifier of the change activity. Another optional way is to retrieve the activity rule code for the change activity from the activity rule code library of the target project based on a timestamp. Another optional way is to retrieve the activity rule code for the change activity from the activity rule code library of the target project based on an object name and / or variable name. There is no limitation here.

[0132] Exemplarily, based on the activity identifier "Marketing Campaign 2024-Q4-001" of the change activity for increasing user participation in a resource management product, the activity rule JSON code of the change activity is retrieved from the activity rule enterprise-level Github code library of the resource management product.

[0133] In the embodiment of the present specification, the code for changing the activity is retrieved from the activity rule code library, thereby ensuring the validity of the subsequent comparison between the activity rule data and the activity rule code.

[0134] In an optional embodiment of the present specification, before retrieving the activity rule code of the change activity from the activity rule code library of the target project, the following specific steps are also included: For the change activity of the target project, configure the activity rule code of the change activity, store the activity rule code in the activity rule code library of the target project, and assign an activity identifier to the change activity; Retrieve the activity rule code for the change activity from the activity rule code library of the target project, including the following specific steps: Based on the activity identifier of the change activity, the activity rule code of the change activity is retrieved from the activity rule code library of the target project.

[0135] The activity identifier of the change activity is an identifier that uniquely identifies the activity rule code of the change activity in the activity rule code library of the target project, ensuring that the configuration records of the change activity during its project life cycle can be accurately tracked and managed.

[0136] Exemplarily, for a change activity that increases user engagement in a resource management product, the activity rule JSON code of the change activity is configured, and the activity rule JSON code is stored in the enterprise-level Github code repository of the activity rules of the resource management product, and the change activity is assigned an activity identifier "Marketing Activity 2024-Q4-001". Based on the activity identifier "Marketing Activity 2024-Q4-001" of the change activity, the activity rule JSON code of the change activity is retrieved from the enterprise-level Github code repository of the activity rules of the resource management product.

[0137] In the embodiments of the present specification, by assigning an activity identifier as an identification index of the activity rule code of the change activity in the code library, it is ensured that the configuration records of each change activity during its project life cycle can be accurately tracked and managed, and the code of the change activity can be accurately retrieved from the activity rule code library, ensuring the validity of subsequent comparisons of activity rule data and activity rule codes.

[0138] In an optional embodiment of the present specification, any activity rule data includes activity rule parameters and activity rule logic, and the activity rule code includes code parameters and code logic; step 106 includes the following specific steps: An activity rule parameter of at least one activity rule data is compared with a code parameter of the activity rule code of the changed activity, and an activity rule logic of at least one activity rule data is compared with a code logic of the activity rule code of the changed activity to obtain a comparison result.

[0139] Activity rule parameters are variables or constant values ​​of the operation logic extracted from the planning text of the change activity. Activity rule parameters usually do not have a clearly defined data type.

[0140] Activity rule logic is the operation process and conditional judgment of the operation logic extracted from the planning text of the change activity. Activity rule logic potentially contains the decision points and path selection required to achieve the activity goals, and may involve multiple condition combinations, loop processing, branch structures and other concepts to adapt to complex activity rules.

[0141] Code parameters are variables or constant values ​​in the activity rule code of the change activity and the operation logic. Code parameters usually have clearly defined data types, such as strings, numbers, Boolean values, etc.

[0142] Code logic is the operation flow and conditional judgment of the operation logic in the activity rule code of the change activity. Code logic clearly defines the decision points and path selection required to achieve the activity goals, which may involve multiple condition combinations, loop processing, branch structures and other concepts to adapt to complex activity rules.

[0143] Compare the activity rule parameters of at least one activity rule data with the code parameters of the activity rule code of the changed activity, and compare the activity rule logic of at least one activity rule data with the code logic of the activity rule code of the changed activity to obtain a comparison result. An optional method is: using a change detection model, compare the activity rule parameters of at least one activity rule data with the code parameters of the activity rule code of the changed activity, and compare the activity rule logic of at least one activity rule data with the code logic of the activity rule code of the changed activity to obtain a comparison result. Going further on this basis, an optional method is: using a pre-supervised trained change detection model, compare the activity rule parameters of at least one activity rule data with the code parameters of the activity rule code of the changed activity, and compare the activity rule logic of at least one activity rule data with the code logic of the activity rule code of the changed activity to obtain a comparison result, wherein the change detection model is pre-trained based on sample activity rule parameters, sample activity rule logic, sample code parameters, sample code logic and label comparison results. Another optional method is: using a large language model, using prompt information to compare the activity rule parameters of at least one activity rule data with the code parameters of the activity rule code of the changed activity, and compare the activity rule logic of at least one activity rule data with the code logic of the activity rule code of the changed activity to obtain a comparison result.

[0144] For example, the prompt information is: You are now a marketing rule risk assessment expert. You can compare the activity rule parameters and activity rule logic in the activity rule data with the code parameters and code logic in the activity rule code to obtain the comparison results (whether there is a conclusion that the activity rule data and the activity rule code do not match; if there is a configuration that does not match the rule description, point out the specific content of the discrepancy in the planning text): enter: Activity rules data: Activity rule parameters: "gmtBegin": "2024-12-11T00:00:00Z", "gmtEnd": "2024-12-31T23:59:59Z", "limitCount": 5……

[0145] Activity rule logic: Users need to complete the resource quota transfer through the designated page. For every two units of resource quota successfully transferred, one reward will be obtained, with a maximum cumulative reward of five...

[0146] Activity rule JSON code: Code parameters: { "gmtBegin": "2024-12-11T00:00:00Z", "gmtEnd": "2024-12-31T23:59:59Z", "limitCount": 1 }…….

[0147] Code logic: After the user completes the resource quota transfer within the specified time range, whether to give a reward and the amount of reward will be determined based on the transfer amount. Each time two units of resource quota are transferred, one reward can be obtained, but each user can only receive a reward once at most (not in line with the planning text)...

[0148] Using the large language model, the above prompt information is used to compare the activity rule parameters of the three activity rule data with the code parameters of the activity rule code of the changed activity, and the activity rule logic of the three activity rule data is compared with the code logic of the activity rule code of the changed activity to obtain the comparison result: The activity audience is not configured in the activity rule code, which is inconsistent with the changes in the activity audience ("eligible_users": "invited_users",; "participation_conditions": ["complete_resource_transfer_via_designated_page"]).

[0149] In the embodiments of this specification, the comparison between activity rule data and activity rule code of the change activity is completed from two dimensions: parameters and logic, thereby realizing the conversion from unstructured text to structured data, comprehensively and accurately locating the potential differences between text and code, and improving the accuracy and consistency of understanding of the change activity.

[0150] In an optional embodiment of the present specification, step 108 includes the following specific steps: Use the risk assessment model to determine whether there is a change risk in the change activity based on the difference in the comparison results; If so, use the risk assessment model to identify the risk type of the change activity based on the difference degree and / or impact object of the difference content, and generate the risk assessment results of the planning text based on the risk type.

[0151] The difference content of the comparison result is the difference determined when comparing the activity rule data with the activity rule code of the changed activity. The difference content can be the inconsistency of parameter values ​​and logical structure. The difference content directly reflects the potential inconsistency between the planning text and the actual execution code.

[0152] The difference degree of the difference content is a quantitative indicator to measure the degree of difference of the difference content. The difference degree can determine which differences need to be corrected immediately and which can be handled in subsequent versions. For example, high difference: the limitCount value in the activity rule code is 1, while it is specified as 5 in the planning text, which directly affects the number of rewards that users can receive and is a key difference. Another example is low difference: minor differences in some non-key fields (such as descriptive tags) do not affect the main function of the activity.

[0153] The impact objects of the difference content are indicators of the project objects in the target project affected by the difference content. By clarifying the impact objects, it is helpful to evaluate the actual impact scope of the difference. Understanding the impact objects can also help the team prioritize those key issues for the target project. For example, the lack of activity audience rules in the activity rule code may cause the user management system to be unable to correctly screen out qualified participants, which in turn affects the system's user management module and participation qualification verification process.

[0154] The risk type of a change activity is the potential risk category identified in the actual execution of the change activity. Risk type is a way to categorize the problems that a change activity may face, which helps to evaluate and quantify different types of potential risks and provide a basis for taking targeted prevention or mitigation measures. For example, compliance risk: the lack of a clear definition of the activity audience may lead to non-compliance with internal or external compliance requirements. Another example is participation risk: if all users can participate in the activity instead of only invited users, it may lead to unexpectedly high participation, increase operating costs and system load. Another example is user experience risk: incorrect rule implementation may cause user confusion and reduce user satisfaction, especially when users find that they do not meet the conditions.

[0155] For example, the prompt information is: You are now a marketing rule risk assessment expert. Based on the difference in the comparison results, determine whether there is a change risk in the change activity. If so, use the risk assessment model to identify the risk type of the change activity based on the difference in the difference content and the affected objects, and generate the risk assessment results of the planning text based on the risk type: enter: Comparison results: The activity audience is not configured in the activity rule code, which is inconsistent with the changes in the activity audience ("eligible_users": "invited_users",; "participation_conditions": ["complete_resource_transfer_via_designated_page"]).

[0156] Using the large language model, the above prompt information is used to determine whether there is a change risk in the change activity based on the difference content (activity audience) of the comparison results. If there is, the risk assessment model is used to identify the risk type of the change activity based on the difference degree of the difference content (high difference: the activity audience is not configured in the activity rule code, which is inconsistent with the change of the activity audience) and the affected object (because the activity audience rule is not configured, the user management system cannot distinguish which users are invited when screening invited users, resulting in all users seeing and trying to participate in the activity. This not only increases the load on the system, but may also confuse or dissatisfy users who do not meet the conditions; the original planning text stipulates that only invited users can participate in the activity, and they need to complete the resource quota transfer through the specified page to obtain the reward qualification. However, since the relevant rules are not configured in the activity rule code, all registered users can participate in the activity), and based on the risk type, the risk assessment result of the planning text is generated: The activity audience rules are not configured in the activity rule code, but the planning text clearly stipulates that the activity is only open to invited users, and users must complete the resource quota transfer through the designated page to be eligible for the reward.

[0157] Risk Type: Compliance risk: Lack of clear definition of campaign audiences may result in non-compliance with internal or external compliance requirements.

[0158] Participation risk: If all users can participate in the event instead of just invited users, it may lead to unexpectedly high participation, increasing operating costs and system load.

[0159] User experience risk: Incorrect rule implementation may cause user confusion and reduce user satisfaction, especially when users find that they do not meet the conditions.

[0160] Probability of occurrence: Due to the lack of necessary restrictions, any user may try to participate in the activity, which is contrary to the provisions of the planning text.

[0161] Impact: Serious. Not only will it affect the cost control of the activity, it may also damage the reputation of the resource management platform and require additional customer service support to handle user questions or complaints.

[0162] Suggested actions: Update the activity rule code immediately, add the activity audience rules, and ensure that they are consistent with the planning text. At the same time, check whether there are other places involving similar rule settings to avoid similar problems.

[0163] In the embodiments of this specification, through the automated comparison and risk assessment process, the differences between the activity rule data and the code are accurately identified, and the enhanced risk assessment is completed in detail, further improving the flexibility, accuracy and efficiency of the enhanced risk assessment.

[0164] In an optional embodiment of the present specification, after step 108, the following specific steps are also included: Integrate the planning text, at least one activity rule data, the comparison result and the risk assessment result to generate a risk assessment report of the planning text.

[0165] The risk assessment report of the planning text is a document that evaluates the potential risks of the planning text in actual implementation. It is a comprehensive risk assessment report of the planning text, providing users with detailed references to ensure that the change activities can be smoothly implemented as expected and take necessary preventive or mitigation measures. The risk assessment report of the planning text includes the planning text, at least one activity rule data, comparison results and risk assessment results.

[0166] Integrate the planning text, at least one activity rule data, comparison results and risk assessment results to generate a risk assessment report for the planning text. One optional method is to use a preset report template to integrate the planning text, at least one activity rule data, comparison results and risk assessment results to generate a risk assessment report for the planning text. Another optional method is to use a large language model and prompt information to integrate the planning text, at least one activity rule data, comparison results and risk assessment results to generate a risk assessment report for the planning text.

[0167] For example, the prompt information is: You are now a marketing rules risk assessment expert. Your task is to generate a comprehensive risk assessment report based on the provided planning text, at least one campaign rule data, comparison results and risk assessment results. Please ensure that the report is detailed and easy to understand, and includes the following sections: 1. Title page.

[0168] 2. Report title.

[0169] 3. Directory.

[0170] 4. Introduction: Summarize the background and purpose of the change activity.

[0171] 5. Planning text summary: briefly summarize the content of the change activity, including activity time, participation conditions, reward mechanism, etc.

[0172] 6. Activity rule data analysis: List in detail the structured activity rule data extracted from the planning text, such as activity cycle, award rules, target audience rules, etc.

[0173] 7. Comparison results: Describe the differences between the planning text and the activity rule code, and point out the specific differences and their impact.

[0174] 8. Risk assessment results: Analyze the identified risk types (such as compliance risk, engagement risk, user experience risk, etc.). Evaluate the probability and impact of each risk. Propose corresponding recommended measures to reduce or eliminate these risks.

[0175] 9. Conclusion and Suggestions: Summarize the key issues found and the coping strategies.

[0176] 10. Appendix: List of relevant document links or attachments (if any).

[0177] Utilize the large language model and use prompt information to integrate the planning text, at least one activity rule data, comparison results and risk assessment results to generate a doc document of the risk assessment report of the planning text.

[0178] In the embodiments of this specification, a detailed risk assessment report is automatically generated by integrating planning text, activity rule data, comparison results and risk assessment results, which significantly improves the efficiency and accuracy of report generation, ensures that the risks of change activities can be fully identified and assessed, provides a solid basis for decision-making, effectively reduces potential risks in actual implementation, and ensures the smooth implementation of activities.

[0179] In an optional embodiment of the present specification, after step 108, the following specific steps are also included: When the risk assessment results meet the preset alarm conditions, an alarm message is sent to the user.

[0180] Preset alarm conditions are rules or thresholds that are set in advance during the risk assessment process to trigger the alarm mechanism. Preset alarm conditions are usually determined based on indicators such as risk type, probability of occurrence, and degree of impact. Preset alarm conditions can be quantitative (such as the risk score exceeds a certain value) or qualitative (such as the occurrence of a specific type of risk). For example, high risk threshold: If the risk assessment results show that the combined score of any risk type exceeds 80 points (out of 100), an alarm is triggered. Another example is a critical risk type: Whenever a compliance risk is identified, an alarm is immediately triggered regardless of its score. Another example is a composite condition: When the probability of occurrence of the participation risk is "high" and the degree of impact is "serious", an alarm is triggered.

[0181] Alert messages are notification messages sent to users when preset alert conditions are met. Alert messages usually contain detailed alert reasons, risk types involved, current status, and recommended response measures. Alert messages can be sent through a variety of channels, such as email, instant messaging tools, and text messages.

[0182] Users are stakeholders involved in the change activity, including but not limited to event planners, development engineers, testers, operations team members, and other individuals or teams who need to understand the risk assessment results and respond to alerts.

[0183] Exemplarily, the preset alarm condition is: once the activity audience group rule is included, sending of an alarm message is triggered.

[0184] When the risk assessment result reaches the preset alarm condition, an alarm message is sent to the user: Change Activity - Compliance and Engagement Risk Alerts: The event audience rules are missing from the event rule code, which results in users other than invited users being able to participate in the event.

[0185] It is recommended to update the event rule code immediately to limit participant qualifications to prevent unnecessary high participation and related risks.

[0186] In the embodiments of the present specification, by triggering the alarm mechanism through preset alarm conditions, key risks in change activities can be identified and notified to relevant users in a timely manner, ensuring rapid response and handling of potential problems.

[0187] The following combination Figure 2 , taking the application of the risk assessment method provided in this specification in digital marketing and e-commerce activities as an example, the risk assessment method is further explained. Figure 2 A process flow chart of a risk assessment method for digital marketing and e-commerce activities provided by an embodiment of this specification is shown, including the following specific steps: Step 202: Obtain a planning text of an online e-commerce promotion activity to be launched for a digital marketing project.

[0188] Step 204: Use the large language model to perform structured rule parsing on the planning text, and extract multiple activity rule data for the e-commerce promotion activities to be launched.

[0189] Step 206: Based on the activity identifier of the e-commerce promotion activity to be launched, the activity rule code of the e-commerce promotion activity to be launched is retrieved from the activity rule code library of the digital marketing project.

[0190] Step 208: Using the large language model, compare the activity rule parameters of the multiple activity rule data with the code parameters of the activity rule code of the e-commerce promotion activity to be launched, and compare the activity rule logic of the multiple activity rule data with the code logic of the activity rule code of the e-commerce promotion activity to be launched, to obtain a comparison result.

[0191] Step 210: Using the large language model, based on the difference content of the comparison results, determine whether there is a risk of change in the e-commerce promotion activities to be launched. If so, using the risk assessment model, based on the difference degree of the difference content and the affected objects, identify the risk type of the e-commerce promotion activities to be launched, and based on the risk type, generate a risk assessment result for the planning text.

[0192] Step 212: When the risk assessment result reaches a preset alarm condition, a warning message is sent to the user.

[0193] Step 214: Utilize the large language model to integrate the planning text, multiple activity rule data, comparison results, and risk assessment results to generate a risk assessment report for the planning text.

[0194] In the embodiments of this specification, a large language model is used to implement deep semantic analysis and rule extraction of e-commerce promotion activity planning texts, structured comparison of activity rule parameters and code logic, accurate identification of potential change risks and generation of detailed evaluation reports, and the ability to detect complex logical conflicts and compliance issues to ensure consistency and accuracy of activity configurations. The highly automated workflow forms a closed-loop management from obtaining planning texts to sending alarm messages, significantly improving the flexibility, accuracy and risk management capabilities of digital marketing projects, and overcoming the technical lag and insufficient risk identification of traditional systems.

[0195] Corresponding to the above method embodiment, this specification also provides a risk assessment device embodiment, Figure 3 FIG. 1 is a schematic diagram showing the structure of a risk assessment device provided by an embodiment of the present specification. Figure 3 As shown, the device comprises: A text acquisition module 302 is configured to acquire a planning text of a change activity for a target project; The rule extraction module 304 is configured to use the rule extraction model to perform structured rule analysis on the planning text to extract at least one activity rule data of the change activity; The change detection module 306 is configured to compare at least one activity rule data with the activity rule code of the changed activity to obtain a comparison result; The risk assessment module 308 is configured to use the risk assessment model to perform a change risk assessment on the change activity based on the comparison result to obtain a risk assessment result of the planning text.

[0196] Optionally, the change detection module 306 is further configured to retrieve the activity rule code of the changed activity from the activity rule code library of the target project.

[0197] Optionally, it further includes a rule configuration module configured to configure an activity rule code for the change activity of the target project, store the activity rule code in an activity rule code library of the target project, and assign an activity identifier to the change activity; The change detection module 306 is further configured to retrieve the activity rule code of the change activity from the activity rule code library of the target project based on the activity identifier of the change activity.

[0198] Optionally, any activity rule data includes activity rule parameters and activity rule logic, and the activity rule code includes code parameters and code logic; The change detection module 306 is further configured to compare the activity rule parameters of at least one activity rule data with the code parameters of the activity rule code of the changed activity, and compare the activity rule logic of at least one activity rule data with the code logic of the activity rule code of the changed activity to obtain a comparison result.

[0199] Optionally, the risk assessment module 308 is further configured to: Use the risk assessment model to determine whether there is a change risk in the change activity based on the difference in the comparison results; If any, use the risk assessment model to identify the risk type of the change activity based on the degree of difference and / or scope of impact of the difference content, and generate the risk assessment results of the planning text based on the risk type.

[0200] Optionally, it also includes a first alarm module, which is configured to integrate the planning text, at least one activity rule data, the comparison result and the risk assessment result to generate a risk assessment report for the planning text.

[0201] Optionally, a second alarm module is also included, which is configured to send an alarm message to the user when the risk assessment result reaches a preset alarm condition.

[0202] In the embodiments of the present specification, the rule extraction module parses the planning text through structured rules to realize the automatic and accurate extraction of activity rule data. The change detection module compares the extracted activity rule data with the activity rule code to realize the conversion from unstructured text to structured data, accurately locates the potential differences between the text and the code, and improves the accuracy and consistency of the understanding of the change activity. The risk assessment module uses the structured comparison results as input, and the risk assessment model analyzes the potential risks of the planning text in actual execution, obtains the risk assessment results, and enhances the flexibility, accuracy and efficiency of the risk assessment.

[0203] With the above Figure 3 The device embodiment corresponds to the following: Figure 4 FIG. 1 shows a schematic diagram of the structure and function of a risk assessment device provided by an embodiment of the present specification. Figure 4 As shown: The function of the rule extraction module is to perform structured rule analysis on the planning text of the change activity and extract the activity rule data of the change activity, thus converting the unstructured planning text into comparable structured data.

[0204] The function of the change detection module is to retrieve the activity rule code of the change activity and compare the activity rule data with the activity rule code to obtain the comparison result. By comparing the activity rule data with the activity rule code, potential differences can be identified.

[0205] The functions of the risk assessment module are: determine whether there is a change risk in the change activity, if so, identify the risk type of the change activity, and generate risk assessment results for the planning text based on the risk type. Analyze and compare the results to assess the risk type and degree of the change activity.

[0206] The functions of the alarm module are: generating risk assessment reports of planning texts and sending alarm messages to users. Generate risk assessment reports and notify relevant users in a timely manner to ensure that risks are handled in a timely manner.

[0207] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the risk assessment device, since it is basically similar to the risk assessment method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the risk assessment method embodiment.

[0208] Figure 5The block diagram of a computing device provided by one embodiment of the present specification is shown. The components of the computing device 500 include but are not limited to a memory 510 and a processor 520. The processor 520 is connected to the memory 510 via a bus 530, and the database 550 is used to store data.

[0209] The computing device 500 also includes an access device 540 that enables the computing device 500 to communicate via one or more networks 560. Examples of these networks include a Public Switched Telephone Network (PSTN), a Local Area Network (LAN), a Wide Area Network (WAN), a Personal Area Network (PAN), or a combination of communication networks such as the Internet. The access device 540 may include one or more of any type of network interface (e.g., a Network Interface Controller (NIC)) of wired or wireless type, such as an IEEE 802.11 Wireless Local Area Network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, and a Near Field Communication (NFC).

[0210] In one embodiment of the present specification, the above components of the computing device 500 and Figure 5 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Figure 5 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0211] The computing device 500 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 500 may also be a mobile or stationary server.

[0212] The processor 520 is used to execute the following computer program / instructions, which implement the steps of the above-mentioned risk assessment method when executed by the processor.

[0213] The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the risk assessment method described above are of the same concept, and the details not described in detail in the technical scheme of the computing device can be found in the description of the technical scheme of the risk assessment method described above.

[0214] An embodiment of the present specification also provides a computer-readable storage medium storing a computer program / instruction, which implements the steps of the above-mentioned risk assessment method when executed by a processor.

[0215] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the risk assessment method described above are of the same concept, and the details not described in detail in the technical scheme of the storage medium can be found in the description of the technical scheme of the risk assessment method described above.

[0216] An embodiment of the present specification also provides a computer program product, including a computer program / instruction, which implements the steps of the above risk assessment method when executed by a processor.

[0217] The above is a schematic scheme of a computer program product of this embodiment. It should be noted that the technical scheme of the computer program product and the technical scheme of the risk assessment method described above are of the same concept, and the details not described in detail in the technical scheme of the computer program product can be found in the description of the technical scheme of the risk assessment method described above.

[0218] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0219] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the contents contained in the computer readable medium may be appropriately increased or decreased according to the requirements of patent practice. For example, in some regions, according to patent practice, computer readable media do not include electric carrier signals and telecommunication signals.

[0220] It should be noted that the above is a description of a specific embodiment of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of the present specification.

[0221] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0222] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The optional embodiments do not describe the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that technicians in the relevant technical field can understand and use this specification well. This specification is limited only by the claims and their full scope and equivalents.

Claims

1. A risk assessment method comprising: Obtain the planning text of the change activities for the target project; Using a rule extraction model, the planning text is subjected to structured rule parsing to extract at least one activity rule data of the change activity; Comparing the at least one activity rule data with the activity rule code of the changed activity to obtain a comparison result; Using the risk assessment model, a change risk assessment is performed on the change activity based on the comparison result to obtain a risk assessment result of the planning text.

2. The method according to claim 1, before comparing the at least one activity rule data with the activity rule code of the changed activity to obtain a comparison result, further comprising: The activity rule code of the change activity is retrieved from the activity rule code library of the target project.

3. The method according to claim 2, before retrieving the activity rule code of the change activity from the activity rule code library of the target project, further comprises: For a change activity of a target project, configuring an activity rule code of the change activity, storing the activity rule code in an activity rule code library of the target project, and assigning an activity identifier to the change activity; The retrieving the activity rule code of the change activity from the activity rule code library of the target project includes: Based on the activity identifier of the change activity, the activity rule code of the change activity is retrieved from the activity rule code library of the target project.

4. The method according to claim 1, wherein any activity rule data comprises activity rule parameters and activity rule logic, and the activity rule code comprises code parameters and code logic; The step of comparing the at least one activity rule data with the activity rule code of the changed activity to obtain a comparison result includes: The activity rule parameters of the at least one activity rule data are compared with the code parameters of the activity rule code of the changed activity, and the activity rule logic of the at least one activity rule data is compared with the code logic of the activity rule code of the changed activity to obtain a comparison result.

5. According to the method according to any one of claims 1 to 4, the step of using a risk assessment model to perform a change risk assessment on the change activity based on the comparison result to obtain a risk assessment result of the planning text comprises: Using a risk assessment model, based on the difference content of the comparison results, determining whether the change activity has a change risk; If so, the risk assessment model is used to identify the risk type of the change activity based on the difference degree and / or impact object of the difference content, and based on the risk type, a risk assessment result of the planning text is generated.

6. The method according to claim 1, after using the risk assessment model to perform change risk assessment on the change activity based on the comparison result to obtain the risk assessment result of the planning text, further comprises: The planning text, the at least one activity rule data, the comparison result and the risk assessment result are integrated to generate a risk assessment report of the planning text.

7. The method according to claim 1, after using the risk assessment model to perform a change risk assessment on the change activity based on the comparison result to obtain the risk assessment result of the planning text, further comprises: When the risk assessment result reaches a preset alarm condition, an alarm message is sent to the user.

8. A risk assessment device, comprising: A text acquisition module is configured to acquire a planning text of a change activity for a target project; A rule extraction module is configured to perform structured rule parsing on the planning text using a rule extraction model to extract at least one activity rule data of the change activity; a change detection module, configured to compare the at least one activity rule data with the activity rule code of the changed activity to obtain a comparison result; The risk assessment module is configured to use a risk assessment model to perform a change risk assessment on the change activity based on the comparison result to obtain a risk assessment result of the planning text.

9. The apparatus according to claim 8, wherein the change detection module is further configured to retrieve the activity rule code of the changed activity from the activity rule code library of the target project.

10. The apparatus according to claim 9, further comprising a rule configuration module, configured to configure an activity rule code for a change activity of a target project, store the activity rule code in an activity rule code library of the target project, and assign an activity identifier to the change activity; The change detection module is further configured to retrieve the activity rule code of the change activity from the activity rule code library of the target project based on the activity identifier of the change activity.

11. The apparatus according to claim 8, wherein any activity rule data comprises activity rule parameters and activity rule logic, and the activity rule code comprises code parameters and code logic; The change detection module is further configured to compare the activity rule parameters of the at least one activity rule data with the code parameters of the activity rule code of the changed activity, and to compare the activity rule logic of the at least one activity rule data with the code logic of the activity rule code of the changed activity to obtain a comparison result.

12. The device according to any one of claims 8 to 11, wherein the risk assessment module is further configured to: Using a risk assessment model, based on the difference content of the comparison results, determining whether the change activity has a change risk; If so, the risk assessment model is used to identify the risk type of the change activity based on the difference degree and / or impact scope of the difference content, and based on the risk type, a risk assessment result of the planning text is generated.

13. The device according to claim 8 further includes a first alarm module configured to integrate the planning text, the at least one activity rule data, the comparison result and the risk assessment result to generate a risk assessment report for the planning text.

14. The device according to claim 8, further comprising a second alarm module, configured to send an alarm message to a user when the risk assessment result reaches a preset alarm condition.

15. A computing device comprising: Memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the computer program / instructions are executed by the processor, the steps of the method described in any one of claims 1 to 7 are implemented.

16. A computer-readable storage medium storing a computer program / instruction, wherein the computer program / instruction, when executed by a processor, implements the steps of the method according to any one of claims 1 to 7.

17. A computer program product, comprising a computer program / instruction, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 7.

Images