SAAS-based internal control management method and system for multi-organizations of public institutions

Abstract

The present invention discloses a SAAS-based internal control management method and system for multiple organizations of public institutions, and relates to the field of intelligent management. It first obtains a text description of an internal control system and a text description of the internal control execution situation, then performs semantic coding based on word granularity on the text description of the internal control system to obtain an internal control system semantic coding feature vector, then performs semantic coding based on word granularity on the text description of the internal control execution situation to obtain an internal control execution situation semantic coding feature vector, then performs sparse semantic matching fusion processing on the internal control system semantic coding feature vector and the internal control execution situation semantic coding feature vector, and finally determines whether the execution of the internal control system complies with the system specification based on the obtained internal control system-internal control execution semantic matching feature, which can realize the automation, intelligence and standardization of the internal control management of multiple organizations of public institutions, and improve the efficiency and effect of internal control management.

Description

Technical Field

[0001] The present application relates to the field of intelligent management, and more specifically, to a SAAS-based multi-organization internal control management method and system for public institutions. Background Art

[0002] Public institutions are affiliated institutions of state organs and social groups, and they undertake the function of public services. The effectiveness of their internal control management directly affects the interests of the state and society. However, public institutions are usually composed of multiple departments or branches, and each organization has its own internal control system and implementation. Therefore, due to the diversity, multi-level and multi-organizational nature of public institutions, their internal control management faces problems such as inconsistent system norms, opaque implementation, and unclear evaluation results.

[0003] Therefore, a SAAS-based multi-organization internal control management solution for public institutions is desired. Summary of the invention

[0004] In order to solve the above technical problems, this application is proposed. This application provides a SAAS-based multi-organization internal control management method and system for public institutions, which can provide unified internal control standards and processes for different departments and branches in public institutions, and realize collaboration and supervision among multiple organizations.

[0005] According to one aspect of the present application, a SAAS-based internal control management method for multiple organizations of a public institution is provided, which includes:

[0006] Get a text description of the internal control system;

[0007] Obtain a textual description of the internal control implementation status;

[0008] Performing semantic coding based on word granularity on the text description of the internal control measure to obtain a semantic coding feature vector of the internal control measure;

[0009] Performing semantic coding based on word granularity on the text description of the internal control execution status to obtain a semantic coding feature vector of the internal control execution status;

[0010] Performing sparse semantic matching fusion processing on the semantic encoding feature vector of the internal control system and the semantic encoding feature vector of the internal control execution status to obtain an internal control system-internal control execution semantic matching feature; and

[0011] Based on the internal control system-internal control execution semantic matching characteristics, determine whether the execution of the internal control system complies with institutional norms.

[0012] According to another aspect of the present application, a SAAS-based multi-organization internal control management system for public institutions is provided, which includes:

[0013] An internal control system acquisition module is used to obtain a text description of the internal control system;

[0014] An internal control execution status acquisition module is used to obtain a text description of the internal control execution status;

[0015] An internal control measure semantic encoding module, used for performing semantic encoding on the text description of the internal control measure based on word granularity to obtain an internal control measure semantic encoding feature vector;

[0016] An internal control execution status semantic encoding module, used for performing word-granularity-based semantic encoding on the text description of the internal control execution status to obtain a semantic encoding feature vector of the internal control execution status;

[0017] A sparse semantic matching fusion module, used for performing sparse semantic matching fusion processing on the semantic encoding feature vector of the internal control system and the semantic encoding feature vector of the internal control execution status to obtain an internal control system-internal control execution semantic matching feature; and

[0018] The execution analysis module is used to determine whether the execution of the internal control system complies with the system specifications based on the internal control system-internal control execution semantic matching characteristics.

[0019] Compared with the prior art, the SAAS-based multi-organization internal control management method and system for public institutions provided in the present application first obtains a text description of the internal control system and a text description of the internal control execution status, then performs word-granularity-based semantic encoding on the text description of the internal control system to obtain a semantic encoding feature vector of the internal control system, then performs word-granularity-based semantic encoding on the text description of the internal control execution status to obtain a semantic encoding feature vector of the internal control execution status, then performs sparse semantic matching and fusion processing on the semantic encoding feature vector of the internal control system and the semantic encoding feature vector of the internal control execution status, and finally determines whether the execution of the internal control system complies with institutional norms based on the obtained internal control system-internal control execution semantic matching features, thereby realizing the automation, intelligence and standardization of the internal control management of multiple organizations in public institutions and improving the efficiency and effectiveness of internal control management. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. The following drawings are not intentionally scaled to the actual size, and the focus is on illustrating the main purpose of the present application.

[0021] Figure 1 It is a flow chart of a SAAS-based multi-organization internal control management method for a public institution according to an embodiment of the present application.

[0022] Figure 2It is a schematic diagram of the architecture of the SAAS-based multi-organization internal control management method for public institutions according to an embodiment of the present application.

[0023] Figure 3 It is a flowchart of sub-step S160 of the SAAS-based multi-organization internal control management method for public institutions according to an embodiment of the present application.

[0024] Figure 4 It is a block diagram of a SAAS-based multi-organization internal control management system for public institutions according to an embodiment of the present application.

[0025] Figure 5 This is an application scenario diagram of the SAAS-based multi-organization internal control management method for public institutions according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0027] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0028] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0029] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0030] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0031] It should be understood that SAAS is a model that deploys software in the cloud and provides services to users through the network. It allows users to access and use software applications through the Internet. The SAAS model provides flexibility and scalability. Using the SAAS platform, it can provide unified internal control standards and processes for different departments and branches in public institutions, realize collaboration and supervision among multiple organizations, make the internal control management of multiple organizations more convenient and efficient, and improve the efficiency and quality of internal control.

[0032] Based on this, in the technical solution of this application, a SAAS-based internal control management method for multiple organizations of public institutions is proposed, which can collect text descriptions of internal control systems and internal control execution, and introduce data processing and semantic understanding algorithms in the back end to perform semantic analysis and matching association of the text descriptions of the internal control systems and internal control execution, so as to judge whether the internal control execution complies with the system norms. In this way, the automation, intelligence and standardization of the internal control management of multiple organizations of public institutions can be realized, and the efficiency and effectiveness of internal control management can be improved.

[0033] Figure 1 It is a flow chart of a SAAS-based multi-organization internal control management method for a public institution according to an embodiment of the present application. Figure 2 FIG. 1 is a schematic diagram of the architecture of a SAAS-based multi-organization internal control management method for a public institution according to an embodiment of the present application. Figure 1 and Figure 2 As shown, according to the SAAS-based multi-organization internal control management method for public institutions of the embodiment of the present application, the steps include: S110, obtaining a text description of the internal control system; S120, obtaining a text description of the internal control execution status; S130, performing word-granularity-based semantic encoding on the text description of the internal control system to obtain a semantic encoding feature vector of the internal control system; S140, performing word-granularity-based semantic encoding on the text description of the internal control execution status to obtain a semantic encoding feature vector of the internal control execution status; S150, performing sparse semantic matching and fusion processing on the semantic encoding feature vector of the internal control system and the semantic encoding feature vector of the internal control execution status to obtain an internal control system-internal control execution semantic matching feature; and, S160, based on the internal control system-internal control execution semantic matching feature, determining whether the implementation of the internal control system complies with institutional specifications.

[0034] It should be understood that in step S110, the internal control system refers to a series of norms, policies, procedures and measures formulated within the organization to achieve its goals. By obtaining the text description of the internal control system, it can be used as the input of the subsequent steps. In step S120, the internal control implementation situation refers to the description of the actual implementation of the internal control system within the organization. By obtaining the text description of the internal control implementation situation, it can be used as the input of the subsequent steps. In step S130, the text description of the internal control system is semantically encoded and converted into a feature vector representation. The semantic encoding based on word granularity mentioned here may involve converting words in the text into vector representations to capture the semantic relationship between words. In step S140, the text description of the internal control implementation situation is semantically encoded and converted into a feature vector representation. The semantic encoding based on word granularity mentioned here may involve converting words in the text into vector representations to capture the semantic relationship between words. In step S150, the semantic coding feature vector of the internal control system and the semantic coding feature vector of the internal control execution status are subjected to sparse semantic matching and fusion processing. The purpose of this step is to obtain the semantic matching features between the internal control system and the internal control execution status by matching and fusing the semantic features of the two. In step S160, the implementation of the internal control system can be evaluated, and it can be determined whether it complies with the specifications and requirements established within the organization. In general, these steps constitute the process of the SAAS-based multi-organization internal control management method for public institutions, from obtaining the text description of the internal control system and the internal control execution status, to converting it into a semantic coding feature vector and matching and fusion processing, and finally determining whether the implementation of the internal control system complies with the specifications. This method can help public institutions effectively manage and evaluate the implementation of internal internal control systems.

[0035] Specifically, in the technical solution of the present application, first, a text description of the internal control system is obtained, and a text description of the internal control implementation is obtained. It should be understood that since there is a large amount of semantic information in the text description of the internal control system and the text description of the internal control implementation, the word granularity is the smallest semantic unit in the text description. Therefore, in order to be able to perform more comprehensive and accurate semantic analysis and understanding of the text description of the internal control system and the text description of the internal control implementation, it is necessary to perform word segmentation processing on these two text descriptions, and perform semantic analysis based on word granularity on the text description after word segmentation. This is because word segmentation can divide long sentences or paragraphs into smaller semantic units, which helps to capture key information and semantic structure in the text.

[0036] Based on this, in the technical solution of the present application, after the text description of the internal control system is processed by word segmentation, semantic encoding is performed in a semantic encoder including a word embedding layer to extract contextual semantic feature information based on word granularity in the text description of the internal control system, thereby obtaining a semantic encoding feature vector of the internal control system. In addition, after the text description of the internal control execution status is processed by word segmentation, semantic encoding is performed in a semantic encoder including a word embedding layer to extract contextual semantic feature information based on word granularity in the text description of the internal control execution status, thereby obtaining a semantic encoding feature vector of the internal control execution status.

[0037] Accordingly, in step S130, the text description of the internal control measure is semantically encoded based on word granularity to obtain a semantic encoding feature vector of the internal control measure, including: performing word segmentation processing on the text description of the internal control measure and then passing it through a semantic encoder including a word embedding layer to obtain the semantic encoding feature vector of the internal control measure.

[0038] It is worth mentioning that the word embedding layer is a technology in natural language processing, which is used to convert words in the text into vector representations. It is part of the neural network model and is usually located after the input layer of the model. The role of the word embedding layer is to convert discrete word representations into continuous vector representations for computer processing and understanding. The main purpose of the word embedding layer is to map the semantic relationship between words to the distance relationship in the vector space. Through this mapping, words with similar semantic meanings will be closer in the vector space, while semantically unrelated words will be relatively far away. This continuous vector representation can provide richer semantic information and help the model better understand the meaning of the text. In the semantic encoding based on word granularity, after the word embedding layer converts each word into a vector representation, these vectors are passed as input to the semantic encoder. The semantic encoder can be a variety of neural network models, such as recurrent neural network (RNN), long short-term memory network (LSTM), convolutional neural network (CNN), etc. The semantic encoder will further process the input word embedding vector to capture the semantic relationship between words and generate the semantic encoding feature vector of the internal control mechanism. By using word embedding layers and semantic encoders, words in the text can be converted into continuous vector representations, while retaining certain semantic information. Such vector representations can be used for subsequent semantic matching, classification, clustering and other tasks, helping to improve the performance and effectiveness of the model in natural language processing tasks.

[0039] In step S140, the text description of the internal control execution status is semantically encoded based on word granularity to obtain a semantic encoding feature vector of the internal control execution status, including: performing word segmentation processing on the text description of the internal control execution status and then passing it through the semantic encoder including the word embedding layer to obtain the semantic encoding feature vector of the internal control execution status.

[0040] It should be understood that in internal control management, the internal control system is a document or system that stipulates internal control requirements and processes, and the internal control implementation situation refers to the implementation of the internal control system in actual operation. In order to evaluate whether the internal control implementation complies with the system specifications, it is necessary to compare the degree of semantic matching between the internal control system and the internal control implementation situation. Therefore, in the technical solution of the present application, a sparse semantic matching fusion module is further used to process the semantic encoding feature vector of the internal control system and the semantic encoding feature vector of the internal control implementation situation to obtain an internal control system-internal control execution semantic matching feature vector. By performing sparse semantic matching processing on the semantic encoding feature vectors of the internal control system and the internal control implementation situation, the semantic relevance and sparse semantic matching degree between the two can be comprehensively considered, so that it is possible to more accurately judge whether the internal control implementation complies with the system specifications, and it is also possible to speed up the calculation speed and highlight the difference between the semantic features of the text descriptions of the two.

[0041] Correspondingly, in step S150, the internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector are subjected to sparse semantic matching fusion processing to obtain an internal control system-internal control execution semantic matching feature, including: using a sparse semantic matching fusion module to process the internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector to obtain an internal control system-internal control execution semantic matching feature vector as the internal control system-internal control execution semantic matching feature.

[0042] It is worth mentioning that the sparse semantic matching fusion module is a technology used in step S150 to process the semantic encoding feature vector of the internal control system and the semantic encoding feature vector of the internal control execution situation to obtain the internal control system-internal control execution semantic matching feature vector. The main purpose of the sparse semantic matching fusion module is to match and fuse the two feature vectors to capture the semantic matching relationship between the internal control system and the internal control execution situation. Specifically, the sparse semantic matching fusion module may use some technologies and algorithms, such as attention mechanism, similarity calculation, feature fusion, etc., to achieve the matching and fusion of feature vectors. During the matching process, the sparse semantic matching fusion module can calculate the similarity or correlation between the semantic encoding feature vector of the internal control system and the semantic encoding feature vector of the internal control execution situation. This can be achieved by calculating the cosine similarity, Euclidean distance and other metrics between the two vectors. The calculation of similarity can help determine the degree of semantic matching between the internal control system and the internal control execution situation. During the fusion process, the sparse semantic matching fusion module can perform weighted fusion or splicing fusion of the two feature vectors. Weighted fusion can perform weighted summation of the two feature vectors according to the matching results to obtain a comprehensive feature representation. Splicing fusion can splice two feature vectors in the feature dimension to form a longer feature vector. Such a fusion operation can help combine the semantic information of the internal control system and the internal control execution status to obtain a more comprehensive feature representation. Through the sparse semantic matching fusion module, the internal control system-internal control execution semantic matching feature vector can be obtained. This feature vector can be used for subsequent judgment, evaluation or classification tasks to determine whether the implementation of the internal control system meets the specifications. The use of the sparse semantic matching fusion module can improve the modeling ability of the semantic matching relationship between the internal control system and the internal control execution status, thereby improving the accuracy and effectiveness of the internal control management method.

[0043] Specifically, in one example, a sparse semantic matching fusion module is used to process the internal control system semantic coding feature vector and the internal control execution situation semantic coding feature vector to obtain an internal control system-internal control execution semantic matching feature vector as the internal control system-internal control execution semantic matching feature, including: using a sparse semantic matching fusion module to process the internal control system semantic coding feature vector and the internal control execution situation semantic coding feature vector using the following fusion formula to obtain the internal control system-internal control execution semantic matching feature vector; wherein the fusion formula is:

[0044] Z=X * +Y *

[0045] X * =W x X

[0046] Y * =Wy Y

[0047] Where X is the semantic coding feature vector of the internal control system, Y is the semantic coding feature vector of the internal control execution status, and W x and W y are the conversion matrix of the semantic coding feature vector of the internal control system and the conversion matrix of the semantic coding feature vector of the internal control execution situation, respectively, * and Y * are respectively the inter-class scatter vector of the semantic encoding feature vector of the internal control system and the inter-class scatter vector of the semantic encoding feature vector of the internal control execution situation, and Z is the internal control system-internal control execution semantic matching feature vector.

[0048] Then, the internal control system-internal control execution semantic matching feature vector is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether the execution of the internal control system complies with the system specifications. In other words, the semantic matching feature information between the internal control system and the internal control execution is used for classification processing to determine whether the internal control execution complies with the system specifications. In this way, it is possible to automatically determine whether the internal control execution complies with the system specifications based on the semantic matching analysis of the internal control system and the internal control execution, thereby improving the efficiency and effectiveness of internal control management.

[0049] Accordingly, in step S160, if Figure 3 As shown, based on the internal control system-internal control execution semantic matching feature, determining whether the execution of the internal control system complies with the system specifications includes: S161, performing feature correction on the internal control system-internal control execution semantic matching feature vector to obtain a corrected internal control system-internal control execution semantic matching feature vector; and, S162, passing the corrected internal control system-internal control execution semantic matching feature vector through a classifier to obtain a classification result, and the classification result is used to indicate whether the execution of the internal control system complies with the system specifications.

[0050] It should be understood that in step S161, the purpose of feature correction of the internal control system-internal control execution semantic matching feature vector is to further optimize the representation of the feature vector to improve the semantic matching accuracy between the internal control system and the internal control execution. Feature correction can be achieved through various technologies and methods, such as feature selection, feature weighting, feature conversion, etc. The corrected internal control system-internal control execution semantic matching feature vector will more accurately reflect the execution of the internal control system. In step S162, the corrected internal control system-internal control execution semantic matching feature vector is classified by a classifier, the purpose of which is to determine whether the execution of the internal control system complies with the system specifications. A classifier is a machine learning model that can predict corresponding categories or labels based on the input feature vector. In this case, the classifier will predict whether the execution of the internal control system complies with the specifications based on the corrected feature vector. The classification result is the result of the classifier output, which is used to indicate whether the execution of the internal control system complies with the system specifications. Generally, the classification result can be binary, indicating whether it complies with the specifications, or multivariate, indicating different degrees of compliance. The classification results can be used as an evaluation indicator of the internal control management method to judge the quality of implementation and compliance of the internal control system. Through these two steps, the implementation of the internal control system can be evaluated and judged. Feature correction can improve the accuracy and expression ability of the feature vector, while the classifier can convert the feature vector into an interpretable classification result, which helps determine whether the implementation of the internal control system complies with the system specifications. The combination of these steps can provide an effective method to manage and evaluate the internal control implementation within the institution.

[0051] Among them, in step S162, the corrected internal control system-internal control execution semantic matching feature vector is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether the implementation of the internal control system complies with the system specifications, including: using the fully connected layer of the classifier to fully connect encode the corrected internal control system-internal control execution semantic matching feature vector to obtain a coded classification feature vector; and, inputting the coded classification feature vector into the Softmax classification function of the classifier to obtain the classification result.

[0052] That is, in the technical solution of the present application, the label of the classifier includes that the execution of the internal control system complies with the system norms (first label), and that the execution of the internal control system does not comply with the system norms (second label), wherein the classifier determines which classification label the corrected internal control system-internal control execution semantic matching feature vector belongs to by a soft maximum function. It is worth noting that the first label p1 and the second label p2 here do not contain artificially set concepts. In fact, during the training process, the computer model does not have the concept of "whether the execution of the internal control system complies with the system norms". It only has two classification labels and the probability of the output feature under these two classification labels, that is, the sum of p1 and p2 is one. Therefore, the classification result of whether the execution of the internal control system complies with the system norms is actually converted into a binary class probability distribution that complies with the laws of nature through the classification label. In essence, the physical meaning of the natural probability distribution of the label is used, rather than the linguistic text meaning of "whether the execution of the internal control system complies with the system norms".

[0053] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression and SVM are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification. However, this is prone to errors and is not efficient. Commonly used multi-classification methods include the Softmax classification function.

[0054] In the above technical scheme, the internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector respectively express the serialized text semantic features of the text description of the internal control system and the text description of the internal control execution status based on word segmentation semantic space embedding, that is, the text semantic feature content of the internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector has semantic feature expression variability under different source text semantic calculation dimensions based on the text description of the internal control system and the text description of the internal control execution status. Therefore, when the internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector are processed using a sparse semantic matching fusion module, considering the sparse correspondence of the text semantic features of the internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector distributed in a common fusion feature space, it is expected to improve the sparse semantic matching fusion expression effect therebetween.

[0055] Based on this, the applicant of the present application fuses and corrects the semantic coding feature vector of the internal control system and the semantic coding feature vector of the internal control execution status.

[0056] Accordingly, in one example, in step S161, feature correction is performed on the internal control system-internal control execution semantic matching feature vector to obtain a corrected internal control system-internal control execution semantic matching feature vector, including: fusing and correcting the internal control system semantic encoding feature vector and the internal control execution situation semantic encoding feature vector to obtain an internal control system-internal control execution semantic fusion feature vector; and fusing the internal control system-internal control execution semantic fusion feature vector and the internal control system-internal control execution semantic matching feature vector to obtain the corrected internal control system-internal control execution semantic matching feature vector.

[0057] The internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector are fused and corrected to obtain the internal control system-internal control execution semantic fusion feature vector, including: fusing and correcting the internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector with the following correction formula to obtain the internal control system-internal control execution semantic fusion feature vector; wherein the correction formula is:

[0058]

[0059] Among them, V1 is the semantic coding feature vector of the internal control system, V2 is the semantic coding feature vector of the internal control execution status, and v 1i and v 2i are the i-th eigenvalues ​​of the semantic encoding feature vector V1 of the internal control system and the semantic encoding feature vector V2 of the internal control execution status, respectively, and ‖·‖2 1 / 2 denote the square of the first norm of the feature vector and the square root of the second norm of the feature vector respectively, the first feature vector and the second feature vector have the same length L, and α is a weight hyperparameter, represents vector addition, represents vector subtraction, exp(·) represents the exponential operation of a value, and the exponential operation of the value represents the calculation of the value of the natural exponential function with the value as the power, v ci It is the i-th eigenvalue of the internal control measure-internal control execution semantic fusion feature vector.

[0060] Here, the above-mentioned optimization fusion is based on the correspondence at the feature value granularity to divide the serialized fusion representation of the internal control system semantic coding feature vector V1 and the internal control execution situation semantic coding feature vector V2 into foreground manifold and background manifold based on the vector scale, so as to stack the dynamic feature value channelization association of the internal control system semantic coding feature vector V1 and the internal control execution situation semantic coding feature vector V2 under the feature corresponding channel supermanifold aggregation mechanism, thereby marking the feature semantic information of the change between the internal control system semantic coding feature vector V1 and the internal control execution situation semantic coding feature vector V2, and realizing the fully connected stacking fusion according to the variability of the semantic content between the internal control system semantic coding feature vector V1 and the internal control execution situation semantic coding feature vector V2 in different computing dimensions. Then, based on v ci The correction feature vector composed of (i.e., the internal control mechanism-internal control execution semantic fusion feature vector), for example, is denoted as V c Further fusion with the internal control system-internal control execution semantic matching feature vector can improve the sparse semantic matching fusion effect of the internal control system-internal control execution semantic matching feature vector for the internal control system semantic coding feature vector V1 and the internal control execution situation semantic coding feature vector V2, thereby improving the accuracy of the classification result obtained by the internal control system-internal control execution semantic matching feature vector through the classifier. In this way, it is possible to judge whether the internal control execution complies with the system norms based on the semantic matching analysis of the internal control system and the internal control execution situation, thereby realizing the automation, intelligence and standardization of the internal control management of multiple organizations in public institutions, and improving the efficiency and effectiveness of internal control management.

[0061] In summary, the SAAS-based internal control management method for multiple organizations of public institutions based on the embodiment of the present application is explained, which can realize the automation, intelligence and standardization of the internal control management of multiple organizations of public institutions, and improve the efficiency and effectiveness of internal control management.

[0062] Figure 4 FIG. 1 is a block diagram of a SAAS-based multi-organization internal control management system 100 for a public institution according to an embodiment of the present application. Figure 4As shown, according to the SAAS-based multi-organization internal control management system 100 of the public institution according to the embodiment of the present application, it includes: an internal control system acquisition module 110, which is used to obtain a text description of the internal control system; an internal control execution status acquisition module 120, which is used to obtain a text description of the internal control execution status; an internal control system semantic encoding module 130, which is used to perform word-granularity-based semantic encoding on the text description of the internal control system to obtain an internal control system semantic encoding feature vector; an internal control execution status semantic encoding module 140, which is used to perform word-granularity-based semantic encoding on the text description of the internal control execution status to obtain an internal control execution status semantic encoding feature vector; a sparse semantic matching fusion module 150, which is used to perform sparse semantic matching fusion processing on the internal control system semantic encoding feature vector and the internal control execution status semantic encoding feature vector to obtain an internal control system-internal control execution semantic matching feature; and an execution analysis module 160, which is used to determine whether the execution of the internal control system complies with the system specifications based on the internal control system-internal control execution semantic matching feature.

[0063] In one example, in the SAAS-based multi-organization internal control management system 100 for public institutions, the internal control system semantic encoding module 130 is used to: perform word segmentation processing on the text description of the internal control system and then pass it through a semantic encoder including a word embedding layer to obtain the semantic encoding feature vector of the internal control system.

[0064] Here, those skilled in the art can understand that the specific functions and operations of each module in the SAAS-based multi-organization internal control management system 100 have been described in the above reference. Figures 1 to 3 It has been introduced in detail in the description of the SAAS-based multi-organization internal control management method for public institutions, and therefore, its repeated description will be omitted.

[0065] As described above, the SAAS-based multi-organization internal control management system 100 for public institutions according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a SAAS-based multi-organization internal control management algorithm for public institutions, etc. In one example, the SAAS-based multi-organization internal control management system 100 for public institutions according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the SAAS-based multi-organization internal control management system 100 for public institutions can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the SAAS-based multi-organization internal control management system 100 for public institutions can also be one of the many hardware modules of the wireless terminal.

[0066] Alternatively, in another example, the SAAS-based public institution multi-organization internal control management system 100 and the wireless terminal may also be separate devices, and the SAAS-based public institution multi-organization internal control management system 100 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0067] Figure 5 FIG. 1 is an application scenario diagram of the SAAS-based multi-organization internal control management method for a public institution according to an embodiment of the present application. Figure 5 As shown, in this application scenario, first, a text description of the internal control system is obtained (for example, Figure 5 D1), and a textual description of the performance of internal controls (e.g. Figure 5 Then, the text description of the internal control system and the text description of the internal control execution status are input into a server (for example, Figure 5 In S) shown in , the server is able to use the SAAS-based public institution multi-organization internal control management algorithm to process the text description of the internal control system and the text description of the internal control implementation status to obtain a classification result indicating whether the implementation of the internal control system complies with system norms.

[0068] According to another aspect of the present application, a non-volatile computer-readable storage medium is provided, on which computer-readable instructions are stored. When the instructions are executed by a computer, the above-mentioned method can be executed.

[0069] The program part of the technology can be considered as a "product" or "manufactured product" in the form of executable code and / or related data, which is participated in or realized by computer-readable media. Tangible and permanent storage media can include any memory or storage used by any computer, processor, or similar device or related module. For example, various semiconductor memories, tape drives, disk drives or any similar devices that can provide storage functions for software.

[0070] In addition, it will be appreciated by those skilled in the art that various aspects of the present application may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may all be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of the present application may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.

[0071] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0072] The above is an explanation of the present application and should not be considered as a limitation thereof. Although several exemplary embodiments of the present application are described, it will be readily understood by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application. Therefore, all of these modifications are intended to be included within the scope of the present application as defined by the claims. It should be understood that the above is an explanation of the present application and should not be considered to be limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present application is defined by the claims and their equivalents.

Claims

1. A SAAS-based internal control management method for multiple organizations in public institutions, characterized in that: include: Get a text description of the internal control system; Obtain a textual description of the internal control implementation status; Performing semantic coding based on word granularity on the text description of the internal control measure to obtain a semantic coding feature vector of the internal control measure; Performing semantic coding based on word granularity on the text description of the internal control execution status to obtain a semantic coding feature vector of the internal control execution status; Performing sparse semantic matching fusion processing on the semantic coding feature vector of the internal control system and the semantic coding feature vector of the internal control execution status to obtain an internal control system-internal control execution semantic matching feature; Based on the semantic matching characteristics of the internal control system and internal control execution, determine whether the execution of the internal control system complies with the system specifications; Based on the semantic matching characteristics of the internal control system and internal control execution, determine whether the execution of the internal control system complies with the system specifications, including: Performing feature correction on the internal control system-internal control execution semantic matching feature vector to obtain a corrected internal control system-internal control execution semantic matching feature vector; Pass the corrected internal control system-internal control execution semantic matching feature vector through a classifier to obtain a classification result, and the classification result is used to indicate whether the execution of the internal control system complies with the system specifications; Performing feature correction on the internal control system-internal control execution semantic matching feature vector to obtain a corrected internal control system-internal control execution semantic matching feature vector, including: The internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector are fused and corrected to obtain an internal control system-internal control execution semantic fusion feature vector; Fusing the internal control system-internal control execution semantic fusion feature vector and the internal control system-internal control execution semantic matching feature vector to obtain the corrected internal control system-internal control execution semantic matching feature vector; The internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector are fused and corrected to obtain the internal control system-internal control execution semantic fusion feature vector, including: fusing and correcting the internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector according to the following correction formula to obtain the internal control system-internal control execution semantic fusion feature vector; wherein the correction formula is: ； in, is the semantic encoding feature vector of the internal control mechanism, is the semantic encoding feature vector of the internal control execution status, and They are the semantic encoding feature vectors of the internal control mechanism and the semantic encoding feature vector of the internal control execution status No. eigenvalues, and The square of the first norm of the feature vector and the square root of the second norm of the feature vector are respectively represented, and the semantic encoding feature vector of the internal control mechanism and the semantic encoding feature vector of the internal control execution status have the same length ,and is the weight hyperparameter, represents vector addition, represents vector subtraction, represents an exponential operation of a value, wherein the exponential operation of the value represents calculating the value of a natural exponential function with the value as a power, is the first character of the internal control mechanism-internal control execution semantic fusion feature vector feature values.

2. The SAAS-based internal control management method for multiple organizations in public institutions according to claim 1 is characterized in that: The text description of the internal control system is semantically encoded based on word granularity to obtain a semantic encoding feature vector of the internal control system, including: The text description of the internal control measure is processed by word segmentation and then passed through a semantic encoder including a word embedding layer to obtain a semantic encoding feature vector of the internal control measure.

3. The SAAS-based internal control management method for multiple organizations in public institutions according to claim 2 is characterized in that: The text description of the internal control execution status is semantically encoded based on word granularity to obtain a semantic encoding feature vector of the internal control execution status, including: The text description of the internal control execution status is processed by word segmentation and then passed through the semantic encoder including the word embedding layer to obtain the semantic encoding feature vector of the internal control execution status.

4. The SAAS-based internal control management method for multiple organizations in public institutions according to claim 3 is characterized in that: The internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector are subjected to sparse semantic matching fusion processing to obtain an internal control system-internal control execution semantic matching feature, including: The internal control system semantic encoding feature vector and the internal control execution status semantic encoding feature vector are processed using a sparse semantic matching fusion module to obtain an internal control system-internal control execution semantic matching feature vector as the internal control system-internal control execution semantic matching feature.

5. The SAAS-based internal control management method for multiple organizations in public institutions according to claim 4 is characterized in that: The internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector are processed using a sparse semantic matching fusion module to obtain an internal control system-internal control execution semantic matching feature vector as the internal control system-internal control execution semantic matching feature, including: The sparse semantic matching fusion module is used to process the internal control system semantic encoding feature vector and the internal control execution status semantic encoding feature vector using the following fusion formula to obtain the internal control system-internal control execution semantic matching feature vector; wherein the fusion formula is: ； ； ； in, is the semantic encoding feature vector of the internal control mechanism, A semantic encoding feature vector for the internal control execution status, and are respectively the conversion matrix of the semantic coding feature vector of the internal control system and the conversion matrix of the semantic coding feature vector of the internal control execution situation, and are respectively the inter-class scatter vector of the semantic coding feature vector of the internal control system and the inter-class scatter vector of the semantic coding feature vector of the internal control execution situation, A semantic matching feature vector is performed for the internal control mechanism-internal control.

6. The SAAS-based internal control management method for multiple organizations in public institutions according to claim 5 is characterized in that: The corrected internal control system-internal control execution semantic matching feature vector is passed through a classifier to obtain a classification result, which is used to indicate whether the execution of the internal control system complies with the system specifications, including: Using the fully connected layer of the classifier to perform fully connected encoding on the corrected internal control mechanism-internal control execution semantic matching feature vector to obtain an encoded classification feature vector; The encoded classification feature vector is input into the Softmax classification function of the classifier to obtain the classification result.

7. A SAAS-based internal control management system for multiple organizations in public institutions, characterized in that: include: An internal control system acquisition module is used to obtain a text description of the internal control system; An internal control execution status acquisition module is used to obtain a text description of the internal control execution status; An internal control measure semantic encoding module, used for performing semantic encoding on the text description of the internal control measure based on word granularity to obtain an internal control measure semantic encoding feature vector; An internal control execution status semantic encoding module, used for performing word-granularity-based semantic encoding on the text description of the internal control execution status to obtain a semantic encoding feature vector of the internal control execution status; A sparse semantic matching fusion module, used for performing sparse semantic matching fusion processing on the semantic encoding feature vector of the internal control system and the semantic encoding feature vector of the internal control execution status to obtain an internal control system-internal control execution semantic matching feature; An execution analysis module, used to determine whether the execution of the internal control system complies with the system specifications based on the internal control system-internal control execution semantic matching characteristics; The execution analysis module is used to: Performing feature correction on the internal control system-internal control execution semantic matching feature vector to obtain a corrected internal control system-internal control execution semantic matching feature vector; Pass the corrected internal control system-internal control execution semantic matching feature vector through a classifier to obtain a classification result, and the classification result is used to indicate whether the execution of the internal control system complies with the system specifications; The execution analysis module is used to: The internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector are fused and corrected to obtain an internal control system-internal control execution semantic fusion feature vector; Fusing the internal control system-internal control execution semantic fusion feature vector and the internal control system-internal control execution semantic matching feature vector to obtain the corrected internal control system-internal control execution semantic matching feature vector; The execution analysis module is used to: fuse and correct the internal control system semantic coding feature vector and the internal control execution status semantic coding feature vector using the following correction formula to obtain the internal control system-internal control execution semantic fusion feature vector; wherein the correction formula is: ； in, is the semantic encoding feature vector of the internal control mechanism, is the semantic encoding feature vector of the internal control execution status, and They are the semantic encoding feature vectors of the internal control mechanism and the semantic encoding feature vector of the internal control execution status No. eigenvalues, and The square of the first norm of the feature vector and the square root of the second norm of the feature vector are respectively represented, and the semantic encoding feature vector of the internal control mechanism and the semantic encoding feature vector of the internal control execution status have the same length ,and is the weight hyperparameter, represents vector addition, represents vector subtraction, represents an exponential operation of a value, wherein the exponential operation of the value represents calculating the value of a natural exponential function with the value as a power, is the first character of the internal control mechanism-internal control execution semantic fusion feature vector feature values.

8. The SAAS-based internal control management system for multiple organizations of public institutions according to claim 7 is characterized in that: The internal control system semantic encoding module is used to: The text description of the internal control measure is processed by word segmentation and then passed through a semantic encoder including a word embedding layer to obtain a semantic encoding feature vector of the internal control measure.

Images