Gas safety knowledge base updating method and device, equipment and storage medium

Abstract

The embodiment of the invention provides a gas safety knowledge base updating method, device and equipment and a storage medium, and the method aims at a target model, based on a gas safety knowledge base and a target knowledge result corresponding to any safety scene problem output by real-time safety data, and based on a model processing process corresponding to the target knowledge result. Obtaining processing process data, and obtaining result feedback information corresponding to the target knowledge result; determining a performance evaluation value and an effect evaluation value based on the processing process data and the result feedback information; the performance evaluation value is used for representing the processing performance of the knowledge processing process corresponding to the target model, and the effect evaluation value is used for representing the effect of the target knowledge result output by the target model; and updating the gas safety knowledge base based on the performance evaluation value and the effect evaluation value. Through continuous positive feedback optimization of the gas safety knowledge base, the accuracy and timeliness of contents in the gas knowledge base are ensured, and the updating efficiency and updating accuracy of the knowledge base are improved.

Description

Technical Field

[0001] The present invention relates to the field of gas safety technology, and in particular to a method, device, equipment and storage medium for updating a gas safety knowledge base. Background Art

[0002] With the rapid development of information technology, the field of artificial intelligence is undergoing unprecedented changes. In this context, manually updating the model knowledge base content has become a key link to ensure the continuous evolution of AI systems and adapt to complex and changing environments.

[0003] In related technologies, AI models rely on pre-trained data sets to build their knowledge systems, but these data sets often fail to fully cover all scenarios and changes in the real world. At this time, the technology of manually updating the model knowledge base content came into being. It combines expert experience, domain knowledge, and the latest data analysis technology to regularly evaluate the completeness and accuracy of the current knowledge base and identify knowledge points that need to be supplemented or corrected. However, with the continuous advancement of technology and the increasing richness of application scenarios, artificial intelligence systems urgently need an efficient and flexible way to continuously update the content of their knowledge base to ensure the accuracy and timeliness of their decisions. Summary of the invention

[0004] In order to overcome the problems existing in the related art, the present invention provides a method, device, equipment and storage medium for updating a gas safety knowledge base.

[0005] In a first aspect, the present invention provides a method for updating a gas safety knowledge base, the method comprising:

[0006] For the target knowledge result corresponding to any safety scenario problem output by the target model based on the gas safety knowledge base and real-time safety data, based on the model processing process corresponding to the target knowledge result, obtain the processing process data, and obtain the result feedback information corresponding to the target knowledge result;

[0007] Based on the processing data and the result feedback information, a performance evaluation value and an effect evaluation value are determined; the performance evaluation value is used to characterize the processing performance of the knowledge processing process corresponding to the target model, and the effect evaluation value is used to characterize the effect of the target knowledge result output by the target model;

[0008] Based on the performance evaluation value and the effect evaluation value, the gas safety knowledge base is updated.

[0009] Optionally, updating the gas safety knowledge base based on the performance evaluation value and the effect evaluation value includes:

[0010] When the performance evaluation value does not meet the preset performance condition, updating the output result corresponding to the target model to the gas safety knowledge base;

[0011] When the effect evaluation value does not meet the preset effect condition, the gas safety knowledge base is updated based on the result feedback information.

[0012] Optionally, the method further comprises:

[0013] Based on target knowledge in other data sources, obtain target answers to the security scenario questions;

[0014] In a case where the first timestamp information of the target knowledge corresponding to the target answer is later than the second timestamp information of the knowledge item corresponding to the target knowledge result, the gas safety knowledge base is updated based on the target knowledge.

[0015] Optionally, the method further comprises:

[0016] For any first knowledge in the gas safety knowledge base, when at least two knowledge results obtained based on the first knowledge have negative user feedback information, the gas safety knowledge base is updated based on answer suggestions in the negative user feedback information.

[0017] Optionally, the result feedback information includes user feedback information, result evaluation information and result verification information, the user feedback information is determined based on the user's evaluation and suggestions on the target knowledge result, the result evaluation information is obtained based on the comparison between the target answer corresponding to the security scenario question and the target knowledge result, and the result verification information is obtained by evaluating the effect of the target knowledge result based on at least two effect evaluation models.

[0018] Optionally, the processing data includes model response time, knowledge memory length corresponding to the target model, and text processing volume.

[0019] In a second aspect, the present invention provides a device for updating a gas safety knowledge base, the device comprising:

[0020] The first acquisition module is used to obtain the target knowledge result corresponding to any safety scenario problem output by the target model based on the gas safety knowledge base and the real-time safety data, based on the model processing process corresponding to the target knowledge result, obtain the processing process data, and obtain the result feedback information corresponding to the target knowledge result;

[0021] A first determination module is used to determine a performance evaluation value and an effect evaluation value based on the processing process data and the result feedback information; the performance evaluation value is used to characterize the processing performance of the knowledge processing process corresponding to the target model, and the effect evaluation value is used to characterize the effect of the target knowledge result output by the target model;

[0022] The first updating module is used to update the gas safety knowledge base based on the performance evaluation value and the effect evaluation value.

[0023] Optionally, the first update module includes:

[0024] A first updating submodule, configured to update the output result corresponding to the target model to the gas safety knowledge base when the performance evaluation value does not meet the preset performance condition;

[0025] The second updating submodule is used to update the gas safety knowledge base based on the result feedback information when the effect evaluation value does not meet the preset effect condition.

[0026] Optionally, the device further comprises:

[0027] A second acquisition module is used to acquire a target answer to the security scenario question based on target knowledge in other data sources;

[0028] The second updating module is used to update the gas safety knowledge base based on the target knowledge when the first timestamp information of the target knowledge corresponding to the target answer is later than the second timestamp information of the knowledge item corresponding to the target knowledge result.

[0029] Optionally, the device further comprises:

[0030] The third updating module is used to update the gas safety knowledge base based on the answer suggestions in the user negative feedback information for any first knowledge in the gas safety knowledge base, when at least two knowledge results obtained based on the first knowledge have user negative feedback information.

[0031] Optionally, the result feedback information includes user feedback information, result evaluation information and result verification information, the user feedback information is determined based on the user's evaluation and suggestions on the target knowledge result, the result evaluation information is obtained based on the comparison between the target answer corresponding to the security scenario question and the target knowledge result, and the result verification information is obtained by evaluating the effect of the target knowledge result based on at least two effect evaluation models.

[0032] Optionally, the processing data includes model response time, knowledge memory length corresponding to the target model, and text processing volume.

[0033] In a third aspect, the present invention provides an electronic device comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for updating the gas safety knowledge base described in any one of the first aspects above is implemented.

[0034] In a fourth aspect, the present invention provides a readable storage medium, which, when the instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to perform the steps in the method for updating the gas safety knowledge base in any one of the embodiments of the first aspect above.

[0035] In the embodiment of the present invention, for the target knowledge result corresponding to any safety scenario problem output by the target model based on the gas safety knowledge base and real-time safety data, the processing process data is obtained based on the model processing process corresponding to the target knowledge result, and the result feedback information corresponding to the target knowledge result is obtained; based on the processing process data and the result feedback information, the performance evaluation value and the effect evaluation value are determined; the performance evaluation value is used to characterize the processing performance of the knowledge processing process corresponding to the target model, and the effect evaluation value is used to characterize the effect of the target knowledge result output by the target model; based on the performance evaluation value and the effect evaluation value, the gas safety knowledge base is updated. In this way, through the processing process data of the target model in the actual question-and-answer process and the output target knowledge result, the performance evaluation value that can evaluate the model processing performance and the effect evaluation value that can evaluate the question-and-answer effect of the target knowledge result are determined, and then based on the performance evaluation value and the effect evaluation value, the gas safety knowledge base is updated, and continuous positive feedback optimization of the gas safety knowledge base is achieved, the accuracy and timeliness of the content in the gas knowledge base are ensured, and the updating efficiency and accuracy of the knowledge base are improved. At the same time, each time the target model outputs the target knowledge result, the gas safety knowledge base will be automatically and timely updated, reducing human intervention and reducing the cost of knowledge base update. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 It is a flowchart of the steps of a method for updating a gas safety knowledge base provided by an embodiment of the present invention;

[0038] Figure 2 It is a structural diagram of a gas safety knowledge base updating device provided by an embodiment of the present invention;

[0039] Figure 3 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] Figure 1 is a flowchart of a method for updating a gas safety knowledge base provided by an embodiment of the present invention, such as Figure 1 As shown, the method may include:

[0042] Step 101, for a target knowledge result corresponding to any safety scenario problem output by a target model based on a gas safety knowledge base and real-time safety data, based on a model processing process corresponding to the target knowledge result, obtain processing process data, and obtain result feedback information corresponding to the target knowledge result.

[0043] In an embodiment of the present invention, users can raise different safety scenario questions for different safety scenarios, and input the safety scenario questions into a pre-trained target model. The target model will output the target knowledge results corresponding to the safety scenario questions based on the gas safety knowledge base and the real-time safety data related to the safety scenario questions. Among them, the target model is used to conduct knowledge questions and answers, and propose solutions that can solve the safety scenario questions, that is, the target knowledge results. The gas safety knowledge base can be a gas safety knowledge base constructed by collecting, organizing and integrating resources such as books, articles, research reports, patents, and cases related to different gas safety scenarios.

[0044] For example, when it is necessary to establish a gas safety knowledge base related to gas safety, a gas safety knowledge base related to gas safety can be constructed by vectorizing and structuring different types of gas safety knowledge (maintenance work orders, emergency response plans, response cases, pipeline problem summaries, testing experience, industry specifications and standards, etc.).

[0045] Different safety scenarios may include pipeline networks, projects, plants, pan-energy, indoor and other scenarios. Correspondingly, safety scenario issues may include monitoring, emergency response, construction safety and other related issues that may exist in pipeline networks, projects, plants, pan-energy, indoor and other scenarios. Input the safety scenario issue into the target model, and the target model will obtain real-time safety data related to the safety scenario issue. For example, if the safety scenario issue is how to conduct gas safety monitoring in the plant, the real-time safety data may include the data of the IoT devices set up in the plant collected in real time (such as pressure value, flow rate, etc.). The target model will output the target knowledge results for gas safety issues based on the gas safety knowledge base and real-time safety data.

[0046] For any target knowledge result output by the target model, obtain the processing data involved in the model processing process of the target model outputting the target knowledge result. Among them, the processing data may include data such as model response time, knowledge memory length corresponding to the target model, and text (tokens) processing volume. Obtain result feedback information corresponding to the target knowledge result. The result feedback information may include user feedback information, result evaluation information, and result verification information. The user feedback information is determined based on the user's evaluation and suggestions for the target knowledge result. The result evaluation information is obtained based on the comparison between the target answer corresponding to the security scenario question and the target knowledge result. The result verification information is obtained based on at least two effect evaluation models to evaluate the effect of the target knowledge result.

[0047] Specifically, after the target knowledge result is fed back to the user, the user will evaluate the target knowledge result. For example, the user can like, dislike or reject the answer to the target knowledge result. Correspondingly, after the user evaluates the target knowledge result, the user's answer suggestion for the target knowledge result can be obtained. For example, after the user dislikes the target knowledge result, the correct answer corresponding to the security scenario question that the user believes can be provided as the suggestion content. Based on the user's evaluation and suggestion of the target knowledge result, the user feedback information is determined. Among them, the user feedback information may include positive user feedback information and negative user feedback information. For example, after the user likes the target knowledge result, positive user feedback information will be generated, and if the user dislikes or rejects the answer to the target knowledge result, negative user feedback information will be generated.

[0048] Obtain the target answer corresponding to the safety scenario question, compare the target answer with the indicated result, and determine the result evaluation information based on the comparison result. The target answer can be the answer content determined by the knowledge in other data sources or the standard answer obtained by the expert evaluation in the gas field. For example, the other data source can be any knowledge source other than the gas safety knowledge base, such as the Internet. Exemplarily, the standard answer obtained by the expert evaluation can be obtained by randomly selecting experts for evaluation, and the standard answer can be compared with the target knowledge result for similarity, and the result evaluation information can be determined based on the similarity comparison result.

[0049] At least two effect evaluation models for evaluating the target knowledge results are pre-trained, and the effect evaluation models can be used to output evaluation results of the accuracy, reliability and effectiveness of the target knowledge results as result verification information. Exemplarily, cross-validation can be performed based on at least two effect evaluation models to evaluate the effect of the target knowledge results and obtain result verification information.

[0050] Step 102: determine a performance evaluation value and an effect evaluation value based on the processing data and the result feedback information; the performance evaluation value is used to characterize the processing performance of the knowledge processing process corresponding to the target model, and the effect evaluation value is used to characterize the effect of the target knowledge result output by the target model.

[0051] In an embodiment of the present invention, a performance evaluation value is determined based on the processing data, and an effect evaluation value is determined based on the result feedback information, wherein the performance evaluation value is used to characterize the processing performance of the knowledge processing process corresponding to the target model, and the effect evaluation value is used to characterize the effect of the target knowledge result output by the target model. A performance evaluation standard is predetermined, the processing data is evaluated based on the performance evaluation standard, and a performance evaluation value corresponding to the processing data is obtained. Similarly, an effect evaluation standard is predetermined, the effect of the result feedback information is evaluated based on the effect evaluation standard, and an effect evaluation value corresponding to the result feedback information is obtained. The performance evaluation value and the effect evaluation value can be expressed in numerical form or in the form of a grade, and can be specifically set according to actual needs. Exemplarily, the value range of the performance evaluation value and the effect evaluation value can be 0-100, and can also be divided into different grades such as excellent, good, and poor.

[0052] For example, the user feedback score (0-40 points), the result evaluation score (0-30 points) and the result verification score (0-30 points) can be obtained for the user feedback information, the result evaluation information and the result verification information, and the user feedback score, the result evaluation score and the result verification score are added together to obtain the effect evaluation value. The grade (e.g., good) can also be determined as the effect evaluation value based on the correspondence between the predetermined score threshold and the grade.

[0053] Step 103: Update the gas safety knowledge base based on the performance evaluation value and the effect evaluation value.

[0054] In the embodiment of the present invention, the gas safety knowledge base is updated based on the performance evaluation value and the effect evaluation value. The updating of the gas safety knowledge base can automatically adjust the content of the gas safety knowledge base, including but not limited to adding new knowledge, updating existing knowledge, and deleting outdated knowledge.

[0055] Optionally, step 103 may include the following steps:

[0056] Step 201: When the performance evaluation value does not meet the preset performance condition, the output result corresponding to the target model is updated to the gas safety knowledge base.

[0057] In the embodiment of the present invention, when the performance evaluation value does not meet the preset performance conditions, it indicates that there is a shortcoming in the performance of the target model, and the output result corresponding to the target model can be updated to the gas safety knowledge base. In this way, based on the questions and results of the knowledge quiz conducted by the target model this time, the content in the gas safety knowledge base can be reversely optimized, so that the target model can improve its processing performance when answering safety scenario questions similar to or the same as this question in the future.

[0058] Step 202: When the effect evaluation value does not meet the preset effect condition, the gas safety knowledge base is updated based on the result feedback information.

[0059] In an embodiment of the present invention, when the effect evaluation value does not meet the preset effect condition, it indicates that the accuracy of the content in the gas safety knowledge base is low, resulting in the reliability and accuracy of the target knowledge results output by the target model being low. In this case, the gas safety knowledge base can be updated based on the result feedback information. Exemplarily, the gas safety knowledge base can be updated based on the user's suggestions in the result feedback information or the target answer corresponding to the safety scenario question, which can specifically include adding the user's suggestions or the target answer to the gas safety knowledge base. Furthermore, the original knowledge in the gas safety knowledge base whose similarity with the user's suggestions or the target answer content is greater than a preset similarity threshold can be deleted to avoid knowledge redundancy in the gas safety knowledge base.

[0060] In summary, in an embodiment of the present invention, for a target knowledge result corresponding to any safety scenario problem output by a target model based on a gas safety knowledge base and real-time safety data, the processing process data is obtained based on the model processing process corresponding to the target knowledge result, and the result feedback information corresponding to the target knowledge result is obtained; based on the processing process data and the result feedback information, a performance evaluation value and an effect evaluation value are determined; the performance evaluation value is used to characterize the processing performance of the knowledge processing process corresponding to the target model, and the effect evaluation value is used to characterize the effect of the target knowledge result output by the target model; based on the performance evaluation value and the effect evaluation value, the gas safety knowledge base is updated. In this way, through the processing process data of the target model in the actual question-and-answer process and the output target knowledge result, a performance evaluation value that can evaluate the model processing performance and an effect evaluation value that can evaluate the question-and-answer effect of the target knowledge result are determined, and then based on the performance evaluation value and the effect evaluation value, the gas safety knowledge base is updated, and continuous positive feedback optimization of the gas safety knowledge base is achieved, the accuracy and timeliness of the content in the gas knowledge base are ensured, and the updating efficiency and accuracy of the knowledge base are improved. At the same time, every time the target model outputs the target knowledge result, it will automatically and timely update the gas safety knowledge base, reducing human intervention and lowering the cost of updating the knowledge base.

[0061] Furthermore, the update of the gas safety knowledge base also provides data support for the optimization of the target model. By analyzing the performance evaluation value and the effect evaluation value, it is possible to identify the deficiencies of the target model in understanding and answering questions, and then to optimize and improve the target model in a targeted manner, thereby improving the adaptability and robustness of the model and the overall performance of the system.

[0062] Optionally, the embodiment of the present invention may further include the following steps:

[0063] Step 301: Based on target knowledge in other data sources, obtain a target answer to the security scenario question.

[0064] In the embodiment of the present invention, based on the target knowledge in other data sources, a target answer corresponding to the same security scenario problem is obtained, wherein the target answer is used to represent the answer of the knowledge from different data sources to the same security scenario problem.

[0065] Step 302: When the first timestamp information of the target knowledge corresponding to the target answer is later than the second timestamp information of the knowledge item corresponding to the target knowledge result, the gas safety knowledge base is updated based on the target knowledge.

[0066] In an embodiment of the present invention, the target knowledge corresponding to the target answer is determined. The target knowledge may be one or more other data source knowledge items in other data sources that can solve the safety scenario problem and form the target answer. The first timestamp information corresponding to the target knowledge is obtained. The first timestamp information corresponding to the target knowledge may be the knowledge formation time corresponding to the knowledge item. For example, when the knowledge item is an industry standard, the knowledge formation time may be the publication time of the industry standard. When the knowledge item is the content of an article, the knowledge formation time may be the publication time of the article. Accordingly, the knowledge formation time may also include the upload time, etc. The knowledge item corresponding to the target knowledge result refers to one or more knowledge in the gas safety knowledge base that can solve the safety scenario problem and form the target knowledge result. Specifically, in the process of the target model processing the safety scenario problem and querying the gas safety knowledge base, the system may record detailed logs. These logs include the input of the question, the query keywords, the accessed knowledge base items, the intermediate processing steps, and the final answer output. By viewing and analyzing these logs, it is possible to track which knowledge in the gas safety knowledge base the target model specifically refers to when obtaining the answer. Correspondingly, the knowledge formation time of each knowledge item is also recorded in the gas safety knowledge base. Based on the knowledge formation time of each knowledge item in the gas safety knowledge base, the second timestamp information of the knowledge item corresponding to the target knowledge result is obtained. When the first timestamp information is later than the second timestamp information, it indicates that the target answer is more updated and more meaningful than the answer of the target knowledge result. Therefore, the gas safety knowledge base can be updated based on the target knowledge.

[0067] In an embodiment of the present invention, when the first timestamp information of the target knowledge corresponding to the target answer is later than the second timestamp information of the knowledge entry corresponding to the target knowledge result, the gas safety knowledge base is updated. In this way, it is ensured that the knowledge entries in the gas safety knowledge base are up-to-date, thereby ensuring the timeliness and reliability of the knowledge.

[0068] Optionally, the embodiment of the present invention may further include the following steps:

[0069] Step 401: For any first knowledge in the gas safety knowledge base, if at least two knowledge results obtained based on the first knowledge have negative user feedback information, update the gas safety knowledge base based on answer suggestions in the negative user feedback information.

[0070] In an embodiment of the present invention, if there is any first knowledge in the gas safety knowledge base, and at least two knowledge results obtained based on the first knowledge both have negative user feedback information, the gas safety knowledge base is updated based on the answer suggestions in the negative user feedback information. Specifically, the negative user feedback information can be obtained first, and the knowledge item referenced by the knowledge result targeted by each user negative feedback information can be confirmed. If there are at least two knowledge results targeted by the negative user feedback information that are both from the same knowledge item, the knowledge item is determined, that is, the first knowledge needs to be updated.

[0071] Since if the first knowledge obtains negative user feedback information on at least two knowledge results generated based on the first knowledge, it indicates that the first knowledge has omissions, resulting in the knowledge results output based on the first knowledge failing to meet user needs, the content in the gas safety knowledge base can be updated in a targeted manner based on the answer suggestions in the user's negative feedback information. Exemplarily, the user feedback information can be added to the first knowledge through text matching and text comparison, or the answer suggestions in the user's negative feedback information can be integrated, and the integrated answer suggestions can directly replace the first knowledge. It can be understood that the embodiment of the present invention does not limit the method of updating the gas safety knowledge base based on the answer suggestions in the user's negative feedback information, and can be specifically set according to actual needs.

[0072] In an embodiment of the present invention, the user negative feedback information corresponding to the knowledge results can be used to determine whether the knowledge items in the gas safety knowledge base are accurate and reliable. Furthermore, when at least two knowledge results obtained from the first knowledge have user negative feedback information, they are updated in a timely manner based on the user negative feedback information, so that the knowledge content in the gas safety knowledge base is richer and more in line with actual scenarios, thereby improving the reliability and practicality of the gas safety knowledge base.

[0073] In a possible implementation, the data collection module can collect real-time safety data and user feedback information; the effect evaluation module can evaluate the actual effect of the target model in the knowledge application process based on the result feedback information to obtain the effect evaluation value; the performance evaluation module can evaluate the performance of the target model in processing tasks related to safety scenario problems based on the processing process data; the self-feedback loop module can automatically generate adjustment suggestions for adjusting the content of the knowledge base based on the effect evaluation value and the performance evaluation value, including but not limited to adding new knowledge, updating existing knowledge, and deleting outdated knowledge; the knowledge base update module can implement the adjustment suggestions of the self-feedback loop module to update the gas safety knowledge base. At the same time, users can update the feedback information of the gas safety knowledge base through the user interface, and view the update history and performance improvement.

[0074] Figure 2is a schematic diagram of a structure of a gas safety knowledge base updating device provided by an embodiment of the present invention, such as Figure 2 As shown, the device may specifically include:

[0075] The first acquisition module 501 is used to obtain the target knowledge result corresponding to any safety scenario problem output by the target model based on the gas safety knowledge base and the real-time safety data, based on the model processing process corresponding to the target knowledge result, obtain the processing process data, and obtain the result feedback information corresponding to the target knowledge result;

[0076] A first determination module 502 is used to determine a performance evaluation value and an effect evaluation value based on the processing process data and the result feedback information; the performance evaluation value is used to characterize the processing performance of the knowledge processing process corresponding to the target model, and the effect evaluation value is used to characterize the effect of the target knowledge result output by the target model;

[0077] The first updating module 503 is used to update the gas safety knowledge base based on the performance evaluation value and the effect evaluation value.

[0078] The embodiment of the present invention provides a device for updating a gas safety knowledge base, which obtains processing process data and result feedback information corresponding to the target knowledge result corresponding to any safety scenario problem output by the target model based on the gas safety knowledge base and real-time safety data; determines a performance evaluation value and an effect evaluation value based on the processing process data and the result feedback information; the performance evaluation value is used to characterize the processing performance of the knowledge processing process corresponding to the target model, and the effect evaluation value is used to characterize the effect of the target knowledge result output by the target model; and updates the gas safety knowledge base based on the performance evaluation value and the effect evaluation value. In this way, the performance evaluation value that can evaluate the model processing performance and the effect evaluation value that can evaluate the question-answering effect of the target knowledge result are determined through the processing process data of the target model in the actual question-answering process and the output target knowledge result, and then the gas safety knowledge base is updated based on the performance evaluation value and the effect evaluation value, so as to realize continuous positive feedback optimization of the gas safety knowledge base, ensure the accuracy and timeliness of the content in the gas knowledge base, and improve the updating efficiency and accuracy of the knowledge base. At the same time, every time the target model outputs the target knowledge result, it will automatically and timely update the gas safety knowledge base, reducing human intervention and lowering the cost of updating the knowledge base.

[0079] Furthermore, the update of the gas safety knowledge base also provides data support for the optimization of the target model. By analyzing the performance evaluation value and the effect evaluation value, it is possible to identify the deficiencies of the target model in understanding and answering questions, and then to optimize and improve the target model in a targeted manner, thereby improving the adaptability and robustness of the model and the overall performance of the system.

[0080] Optionally, the first updating module 503 includes:

[0081] A first updating submodule, configured to update the output result corresponding to the target model to the gas safety knowledge base when the performance evaluation value does not meet the preset performance condition;

[0082] The second updating submodule is used to update the gas safety knowledge base based on the result feedback information when the effect evaluation value does not meet the preset effect condition.

[0083] Optionally, the device further comprises:

[0084] A second acquisition module is used to acquire a target answer to the security scenario question based on target knowledge in other data sources;

[0085] The second updating module is used to update the gas safety knowledge base based on the target knowledge when the first timestamp information of the target knowledge corresponding to the target answer is later than the second timestamp information of the knowledge item corresponding to the target knowledge result.

[0086] Optionally, the device further comprises:

[0087] The third updating module is used to update the gas safety knowledge base based on the answer suggestions in the user negative feedback information for any first knowledge in the gas safety knowledge base, when at least two knowledge results obtained based on the first knowledge have user negative feedback information.

[0088] Optionally, the result feedback information includes user feedback information, result evaluation information and result verification information, the user feedback information is determined based on the user's evaluation and suggestions on the target knowledge result, the result evaluation information is obtained based on the comparison between the target answer corresponding to the security scenario question and the target knowledge result, and the result verification information is obtained by evaluating the effect of the target knowledge result based on at least two effect evaluation models.

[0089] Optionally, the processing data includes model response time, knowledge memory length corresponding to the target model, and text processing volume.

[0090] The present invention also provides an electronic device, see Figure 3 , including: a processor 601, a memory 602, and a computer program 6021 stored in the memory and executable on the processor, wherein when the processor executes the program, the method for updating the gas safety knowledge base of the aforementioned embodiment is implemented.

[0091] The present invention also provides a readable storage medium, when the instructions in the storage medium are executed by a processor of an electronic device, the electronic device can execute the method for updating the gas safety knowledge base of the aforementioned embodiment.

[0092] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0093] The algorithm and display provided herein are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to realize the content of the present invention described herein, and the description of the above specific languages ​​is for disclosing the best mode of the present invention.

[0094] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0095] Similarly, it should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the following intention: that the claimed invention requires more features than the features explicitly recited in each claim. More specifically, as reflected in the claims below, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself serving as a separate embodiment of the present invention.

[0096] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0097] The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in the sorting device according to the present invention. The present invention may also be implemented as a device or apparatus program for executing part or all of the methods described herein. Such a program for implementing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0098] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc., does not indicate any order. These words may be interpreted as names.

[0099] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0100] It should be noted that all actions of acquiring signals, information or data in this application are carried out in compliance with the relevant data protection laws and policies of the country where they are located and with the authorization given by the owner of the corresponding device.

[0101] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0102] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for updating a gas safety knowledge base, characterized in that: The method comprises: For the target knowledge result corresponding to any safety scenario problem output by the target model based on the gas safety knowledge base and real-time safety data, based on the model processing process corresponding to the target knowledge result, obtain the processing process data, and obtain the result feedback information corresponding to the target knowledge result; Based on the processing data and the result feedback information, a performance evaluation value and an effect evaluation value are determined; the performance evaluation value is used to characterize the processing performance of the knowledge processing process corresponding to the target model, and the effect evaluation value is used to characterize the effect of the target knowledge result output by the target model; Based on the performance evaluation value and the effect evaluation value, the gas safety knowledge base is updated.

2. The method according to claim 1, characterized in that The updating of the gas safety knowledge base based on the performance evaluation value and the effect evaluation value includes: When the performance evaluation value does not meet the preset performance condition, updating the output result corresponding to the target model to the gas safety knowledge base; When the effect evaluation value does not meet the preset effect condition, the gas safety knowledge base is updated based on the result feedback information.

3. The method according to claim 1, characterized in that The method further comprises: Based on target knowledge in other data sources, obtain target answers to the security scenario questions; In a case where the first timestamp information of the target knowledge corresponding to the target answer is later than the second timestamp information of the knowledge item corresponding to the target knowledge result, the gas safety knowledge base is updated based on the target knowledge.

4. The method according to claim 1, characterized in that: The method further comprises: For any first knowledge in the gas safety knowledge base, when at least two knowledge results obtained based on the first knowledge have negative user feedback information, the gas safety knowledge base is updated based on answer suggestions in the negative user feedback information.

5. The method according to claim 1, characterized in that The result feedback information includes user feedback information, result evaluation information and result verification information. The user feedback information is determined based on the user's evaluation and suggestions on the target knowledge result. The result evaluation information is obtained based on the comparison between the target answer corresponding to the security scenario question and the target knowledge result. The result verification information is obtained by evaluating the effect of the target knowledge result based on at least two effect evaluation models.

6. The method according to claim 1, characterized in that The processing data includes model response time, knowledge memory length corresponding to the target model, and text processing volume.

7. A gas safety knowledge base updating device, characterized in that: The device comprises: The first acquisition module is used to obtain the target knowledge result corresponding to any safety scenario problem output by the target model based on the gas safety knowledge base and the real-time safety data, based on the model processing process corresponding to the target knowledge result, obtain the processing process data, and obtain the result feedback information corresponding to the target knowledge result; A first determination module is used to determine a performance evaluation value and an effect evaluation value based on the processing process data and the result feedback information; the performance evaluation value is used to characterize the processing performance of the knowledge processing process corresponding to the target model, and the effect evaluation value is used to characterize the effect of the target knowledge result output by the target model; The first updating module is used to update the gas safety knowledge base based on the performance evaluation value and the effect evaluation value.

8. The method according to claim 7, characterized in that The first update module includes: A first updating submodule, configured to update the output result corresponding to the target model to the gas safety knowledge base when the performance evaluation value does not meet the preset performance condition; The second updating submodule is used to update the gas safety knowledge base based on the result feedback information when the effect evaluation value does not meet the preset effect condition.

9. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for updating the gas safety knowledge base as described in any one of claims 1 to 6 is implemented.

10. A readable storage medium, characterized in that: When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method for updating the gas safety knowledge base as described in any one of claims 1-6.

Images