Conference record evidence storage method and device, equipment and storage medium

Abstract

The invention discloses a conference record evidence storage method, device and equipment and a storage medium, and relates to the technical field of block chain evidence storage, and the method comprises the steps: obtaining a block chain evidence storage interface, conference organizer information and a real-time conference record corresponding to a target conference; inputting the real-time conference record into a conference evidence storage analysis model to obtain conference key content output by the conference evidence storage analysis model, and comparing the conference key content with a preset dynamic rule to obtain target evidence storage content; generating a digital signature and a public key certificate based on the conference organizer information and the target evidence content; and connecting the block chain evidence storage interface, generating a signature data packet based on the digital signature, the public key certificate and the target evidence storage content, and storing the signature data packet to a distributed account book corresponding to the block chain evidence storage interface. According to the method, automatic evidence storage is triggered by identifying the target evidence storage content, so that accurate identification of the key content and automatic triggering of evidence storage are realized, and the processing efficiency and the legal potency of the conference record are improved.

Description

Technical Field

[0001] This application relates to the field of blockchain evidence storage technology, and particularly to a method, device, equipment and storage medium for meeting record evidence storage. Background Art

[0002] With the rapid development of information technology, NLP (Natural Language Processing) technology has been widely used in the field of meeting records. In recent years, the emergence of deep learning models (such as WavLM, Transformer) has significantly improved the accuracy of speech recognition and text understanding, providing strong support for the automated processing of meeting records. At the same time, the maturity of technologies such as blockchain technology and timestamp has provided a reliable solution for data evidence storage, ensuring the immutability and traceability of information.

[0003] However, although related technologies have made significant progress in their respective fields, the application of combining natural language processing technology with data evidence storage technology still has gaps. The traditional method of organizing meeting records mainly relies on manual work, which is time-consuming and prone to missing key information. Especially when the meeting content is complex, involving multiple topics and a large number of details, the efficiency of manual sorting is low, and it is difficult to meet the needs of rapid decision-making and information traceability. At the same time, the traditional evidence storage method has a cumbersome process and lacks technical means to accurately identify and solidify key content, resulting in the difficulty of guaranteeing the efficiency and reliability of evidence storage.

[0004] Therefore, there is an urgent need for a method that can accurately identify key content and automatically trigger evidence storage to improve the processing efficiency and legal effect of meeting records. Summary of the Invention

[0005] The main purpose of this application is to provide a method, device, equipment and storage medium for meeting record evidence storage, aiming to accurately identify key content and automatically trigger evidence storage to improve the processing efficiency and legal effect of meeting records.

[0006] To achieve the above object, this application proposes a method for meeting record evidence storage, and the method includes:

[0007] Obtain the blockchain evidence storage interface, the meeting organizer information, and the real-time meeting record corresponding to the target meeting;

[0008] Input the real-time meeting record into the meeting evidence storage analysis model to obtain the key meeting content output by the meeting evidence storage analysis model, and compare the key meeting content with a preset dynamic rule to obtain the target evidence storage content;

[0009] Generate a digital signature and a public key certificate based on the meeting organizer information and the target evidence storage content;

[0010] Connect to the blockchain evidence storage interface, generate a signature data packet based on the digital signature, the public key certificate, and the target evidence content, and store the signature data packet in the distributed ledger corresponding to the blockchain evidence storage interface.

[0011] In one embodiment, the training process of the conference evidence analysis model includes:

[0012] Obtain several groups of historical conference data, and perform data augmentation on each group of historical conference data;

[0013] Through a preset model encoder, perform entity recognition, decision point detection, and task relationship extraction on each group of historical conference data after data augmentation to obtain several groups of model training data;

[0014] Input each group of historical conference data and each group of model training data into the initial conference evidence analysis model for iterative training to obtain the conference evidence analysis model.

[0015] In one embodiment, the step of performing entity recognition, decision point detection, and task relationship extraction on each group of historical conference data after data augmentation through a preset model encoder to obtain several groups of model training data includes:

[0016] Through the first task head in the preset model encoder, perform entity recognition on each group of historical conference data after data augmentation to obtain several groups of model training data;

[0017] Through the second task head in the preset model encoder, perform decision point detection on each group of historical conference data after data augmentation to obtain several groups of model training data;

[0018] Through the third task head in the preset model encoder, perform task relationship extraction on each group of historical conference data after data augmentation to obtain several groups of model training data.

[0019] In one embodiment, the step of generating a digital signature and a public key certificate based on the conference organizer information and the target evidence content includes:

[0020] Obtain an encryption key pair, where the encryption key pair includes a public key and a private key;

[0021] Generate a public key certificate according to the public key and the conference organizer information;

[0022] Perform a hash calculation on the target evidence content to obtain a first hash value, and encrypt the first hash value according to the private key to generate a digital signature.

[0023] In one embodiment, generating a signature data packet based on the digital signature, the public key certificate, and the target content to be deposited, and storing the signature data packet in a distributed ledger corresponding to the blockchain deposit interface includes:

[0024] Obtain the deposit metadata and deposit timestamp corresponding to the blockchain deposit interface;

[0025] Associate and combine the deposit metadata and the deposit timestamp with the digital signature, the public key certificate, and the target content to be deposited to generate a signature data packet;

[0026] Store the signature data packet in a distributed ledger corresponding to the blockchain deposit interface.

[0027] In one embodiment, after generating a signature data packet based on the digital signature, the public key certificate, and the target content to be deposited, and storing the signature data packet in a distributed ledger corresponding to the blockchain deposit interface, it further includes:

[0028] Obtain the complete original meeting record and the deposit index information of the signature data packet;

[0029] Input the complete original meeting record into a meeting deposit analysis model to obtain a meeting record report output by the meeting deposit analysis model;

[0030] Associate and combine the deposit index information, the meeting record report, and the complete original meeting record, and store them in an off-chain storage location.

[0031] In one embodiment, after generating a signature data packet based on the digital signature, the public key certificate, and the target content to be deposited, and storing the signature data packet in a distributed ledger corresponding to the blockchain deposit interface, it further includes:

[0032] Connect to the blockchain deposit interface, download the signature data packet, and parse the signature data packet to obtain the digital signature and the public key certificate;

[0033] Calculate a second hash value of the target content to be deposited, and decrypt the digital signature based on the public key in the public key certificate to verify whether the first hash value of the decrypted digital signature is consistent with the second hash value;

[0034] If so, verify the validity of the public key certificate and generate a meeting deposit verification result.

[0035] In addition, to achieve the above object, the present application also proposes a meeting record deposit device, and the meeting record deposit device includes:

[0036] An acquisition module, configured to acquire a blockchain evidence storage interface, information of a meeting organizer, and a real-time meeting record corresponding to a target meeting;

[0037] An analysis module, configured to input the real-time meeting record into a meeting evidence storage analysis model to obtain key meeting content output by the meeting evidence storage analysis model, and compare the key meeting content with a preset dynamic rule to obtain target evidence storage content;

[0038] A generation module, configured to generate a digital signature and a public key certificate based on the information of the meeting organizer and the target evidence storage content;

[0039] An evidence storage module, configured to connect to the blockchain evidence storage interface, generate a signature data packet based on the digital signature, the public key certificate, and the target evidence storage content, and store the signature data packet in a distributed ledger corresponding to the blockchain evidence storage interface.

[0040] In addition, to achieve the above object, the present application further provides a meeting record evidence storage device, where the device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the meeting record evidence storage method as described above.

[0041] In addition, to achieve the above object, the present application further provides a storage medium, where the storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium, and when the computer program is executed by a processor, the steps of the meeting record evidence storage method as described above are implemented.

[0042] In addition, to achieve the above object, the present application further provides a computer program product, where the computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the meeting record evidence storage method as described above are implemented.

[0043] The present application provides a meeting record evidence storage method, device, equipment, and storage medium. The meeting record evidence storage method acquires a blockchain evidence storage interface, information of a meeting organizer, and a real-time meeting record corresponding to a target meeting, and then inputs the real-time meeting record into a meeting evidence storage analysis model to obtain key meeting content output by the meeting evidence storage analysis model, and compares the key meeting content with a preset dynamic rule to obtain target evidence storage content. Then, based on the information of the meeting organizer and the target evidence storage content, a digital signature and a public key certificate are generated, and then the blockchain evidence storage interface is connected. Based on the digital signature, the public key certificate, and the target evidence storage content, a signature data packet is generated, and the signature data packet is stored in a distributed ledger corresponding to the blockchain evidence storage interface, so as to achieve accurate identification of key content and automatic triggering of evidence storage, and improve the processing efficiency and legal effect of meeting records. Brief Description of the Drawings

[0044] The drawings herein are incorporated into the specification and form a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0045] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0046] Figure 1 It is a schematic flowchart provided for the first embodiment of the method for depositing meeting record evidence of the present application;

[0047] Figure 2 It is a schematic flowchart provided for the second embodiment of the method for depositing meeting record evidence of the present application;

[0048] Figure 3 It is a schematic flowchart provided for the third embodiment of the method for depositing meeting record evidence of the present application;

[0049] Figure 4 It is a schematic flowchart provided for the fourth embodiment of the method for depositing meeting record evidence of the present application;

[0050] Figure 5 It is a schematic diagram of the module structure of the device for depositing meeting record evidence in the embodiments of the present application;

[0051] Figure 6 It is a schematic diagram of the device structure of the hardware operating environment involved in the method for depositing meeting record evidence in the embodiments of the present application.

[0052] The realization of the purpose, functional features, and advantages of the present application will be further described in conjunction with the embodiments with reference to the drawings. Detailed Embodiments

[0053] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0054] To better understand the technical solutions of the present application, the following will be described in detail in conjunction with the drawings of the specification and the specific embodiments.

[0055] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication, and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device, a big data service platform, a meeting record deposit system, etc. that can implement the above functions. Hereinafter, the meeting record deposit system will be taken as an example to illustrate this embodiment and the following embodiments.

[0056] Based on this, an embodiment of the present application provides a method for depositing meeting records for evidence. Refer to Figure 1 , Figure 1 which is a schematic flowchart provided for the first embodiment of the method for depositing meeting records for evidence of the present application.

[0057] In this embodiment, the method for depositing meeting records for evidence includes steps S11 to S14:

[0058] Step S11, obtain a blockchain evidence deposit interface, meeting organizer information, and the real-time meeting record corresponding to the target meeting;

[0059] It should be noted that the blockchain evidence deposit interface refers to an API (Application Programming Interface) or technical interface used to interact with the blockchain platform, allowing the system to upload evidence deposit data (such as hash values, signatures, etc.) to the blockchain, retrieve data from the blockchain, and provide communication capabilities with the blockchain platform (such as Ant Chain, Tencent Zhixin Chain, Ethereum, etc.).

[0060] Furthermore, it should be noted that the meeting organizer information refers to the relevant information of the person or entity responsible for organizing the meeting, including name, position, ID number, enterprise certification information, etc., which is used to generate a digital signature to ensure that the evidence deposit content is bound to a specific identity, and in a legal scenario, prove that the evidence deposit content is confirmed by a specific organizer.

[0061] Furthermore, the target meeting refers to a specific meeting instance that needs to be deposited for evidence, and the real-time meeting record refers to the text record generated during the meeting, which can be obtained by means such as voice transcription or PPT screenshots. For example, the audio-visual code stream or auxiliary stream code stream generated by the target meeting is used to provide the original data of the meeting content for subsequent analysis and evidence deposit, ensuring that every word or decision in the meeting is recorded.

[0062] Specifically, obtain a blockchain evidence deposit interface, meeting organizer information, and the real-time meeting record corresponding to the target meeting. Among them, the meeting organizer information can verify the authenticity of the identity by docking with the public security / industry and commerce database API (such as the Alipay real-name authentication interface), and at the same time attach biometric features (face recognition / voiceprint verification) to ensure "consistency between the person and the certificate". It can also synchronize the organizer's identity based on the enterprise AD (Active Directory) / LDAP (Lightweight Directory Access Protocol) system, associate with the internal permission system, and use a hardware token (such as a UKey) or a dynamic password for secondary verification, so as to collect and verify the meeting organizer information.

[0063] Furthermore, the real-time meeting record supports real-time voice transcription (such as through the WavLM model) and text import, and is also compatible with meeting images (such as PPT screenshots, etc.).

[0064] Step S12: Input the real-time meeting record into the meeting evidence storage analysis model to obtain the key meeting content output by the meeting evidence storage analysis model, and compare the key meeting content with the preset dynamic rules to obtain the target evidence storage content.

[0065] It should be noted that the meeting evidence storage analysis model refers to a natural language processing (NLP) model used to process and analyze meeting records, which usually includes functions such as entity recognition, decision point detection, and task relationship extraction to extract key information (such as decision points, task assignments, responsible persons, etc.) from real-time meeting records, thereby improving the efficiency and accuracy of evidence storage and reducing manual intervention.

[0066] Furthermore, it should be noted that the key meeting content refers to the important information extracted from the meeting record, such as decision points, task assignments, responsible persons, time nodes, etc., to determine the core information that needs to be stored as evidence, reduce the redundancy of evidence storage data, and improve the efficiency of evidence storage.

[0067] Furthermore, the preset dynamic rules refer to the rules set according to enterprise requirements or compliance requirements to determine which meeting content needs to be stored as evidence. The preset dynamic rules can be set based on keywords, speaker identities, task types, etc., without limitation here, to automatically trigger evidence storage, reduce manual intervention, and ensure that the evidence storage content meets the requirements of the enterprise or law. Additionally, the target evidence storage content refers to the key meeting content that needs to be stored as evidence after being screened by the dynamic rules, that is, the data that ultimately needs to be stored as evidence, to ensure that the evidence storage content meets the requirements of the preset rules.

[0068] Specifically, inputting the real-time meeting record into the meeting evidence storage analysis model to obtain the key meeting content output by the meeting evidence storage analysis model, thereby automatically extracting the key content of the meeting (such as decision points, task assignments, responsible persons, etc.) through the NLP model, reducing the time and workload of manually sorting out the meeting record, and then improving the efficiency of evidence storage to ensure that important information can be quickly identified and stored as evidence.

[0069] Furthermore, comparing the key meeting content with the preset dynamic rules to obtain the target evidence storage content, where the preset dynamic rules can be flexibly configured according to the needs and compliance requirements of the enterprise. In one embodiment, the preset dynamic rules include the following types:

[0070] (1) Keyword Trigger Rule: Set specific keywords or phrases. When these keywords appear in the meeting content, it triggers evidence preservation. For example, the set keywords are: "Contract signing", "Project start", "Budget adjustment". If "Contract signing has been completed" is mentioned in the meeting, it triggers automatic evidence preservation.

[0071] (2) Speaker Identity Rule: Determine whether to trigger evidence preservation based on the identity or role of the speaker. For example, the set dynamic rule is: If the speaker is "Company executive" or "Legal department", then their speech content automatically triggers evidence preservation. For example, the compliance suggestions mentioned by the legal department in the meeting will trigger automatic evidence preservation.

[0072] (3) Task Assignment Rule: When task assignment appears in the meeting, trigger evidence preservation according to the type or importance of the task. For example, the set dynamic rule is: Any task involving "Fund appropriation" or "Project extension" automatically triggers evidence preservation. For example, if the meeting decides that "The project is extended to June 2024", then this decision point and related tasks are automatically preserved as evidence.

[0073] (4) Sentiment Analysis Rule: Trigger evidence preservation according to the sentiment tendency of the meeting content (such as negative sentiment). For example, the set dynamic rule is: If negative sentiment appears in the meeting (such as "Serious problem", "No consensus reached"), then it triggers evidence preservation. For example, if "There are serious quality problems with the product" is mentioned in the meeting, then this content is automatically preserved as evidence.

[0074] (5) Time-related Rule: Trigger evidence preservation according to the time node in the meeting content. For example, the set dynamic rule is: Any task involving "Tasks to be completed within the next three months" automatically triggers evidence preservation. For example, if the meeting decides that "The product needs to be launched before the end of May", then this task is automatically preserved as evidence.

[0075] Thus, through the compliance design of dynamic rules, flexible configuration can be achieved according to the compliance requirements of the enterprise to ensure that all content that needs to be preserved as evidence is recorded and stored on the blockchain.

[0076] Step S13, generate a digital signature and a public key certificate based on the meeting organizer information and the target content to be preserved as evidence;

[0077] It should be noted that the digital signature refers to the signature generated by encrypting data (such as the hash value of the key content of the meeting) using a private key, which is used to verify the integrity and source of the data. The public key certificate refers to the digital certificate issued by a certificate authority or within the enterprise, which contains the public key and identity information and is used to verify the digital signature.

[0078] Specifically, an encrypted key pair is obtained, where the encrypted key pair includes a public key and a private key. Furthermore, according to the public key and the conference organizer information, a public key certificate is generated, and then the target evidence storage content is subjected to a hash calculation to obtain a first hash value, and the first hash value is encrypted according to the private key to generate a digital signature, thereby ensuring the reliability of the source and non-repudiation of the evidence storage content, because only the conference organizer with the corresponding private key can generate a valid signature, further enhancing the security of the data, preventing unauthorized tampering or forgery, meeting the legal requirements for electronic evidence, so that in legal proceedings or compliance inspections, the evidence storage content has higher and stable legal effect.

[0079] Step S14, connect to the blockchain evidence storage interface, and based on the digital signature, the public key certificate, and the target evidence storage content, generate a signature data packet, and store the signature data packet in the distributed ledger corresponding to the blockchain evidence storage interface.

[0080] It should be noted that the signature data packet refers to a data structure that encapsulates the key conference content, hash value, digital signature, and public key certificate for evidence storage. The distributed ledger refers to the data storage structure in blockchain technology, which records all verified transactions or data and has the characteristics of being immutable and traceable.

[0081] Specifically, obtain the evidence storage metadata and evidence storage timestamp corresponding to the blockchain evidence storage interface, and then associate and combine the evidence storage metadata and the evidence storage timestamp with the digital signature, the public key certificate, and the target evidence storage content to generate a signature data packet, and then store the signature data packet in the distributed ledger corresponding to the blockchain evidence storage interface, thereby ensuring the authenticity and integrity of the evidence storage content, and providing credible evidence even in the face of legal disputes or audits.

[0082] In this embodiment, by obtaining the blockchain evidence storage interface, the information of the meeting organizer, and the real-time meeting record corresponding to the target meeting, the real-time meeting record is then input into the meeting evidence analysis model to obtain the key meeting content output by the meeting evidence analysis model, and the key meeting content is compared with the preset dynamic rules to obtain the target evidence content. Based on the information of the meeting organizer and the target evidence content, a digital signature and a public key certificate are generated. Then, the blockchain evidence storage interface is connected, and based on the digital signature, the public key certificate, and the target evidence content, a signature data packet is generated and stored in the distributed ledger corresponding to the blockchain evidence storage interface, thereby realizing accurate identification of the key content and automatically triggering evidence storage, improving the processing efficiency and legal effect of the meeting record. At the same time, by automatically extracting the key content, manual intervention is reduced, thereby improving the evidence storage efficiency. Through the blockchain and digital signature, the authenticity and immutability of the data are ensured, the evidence storage security is improved, and the legal compliance is enhanced. In addition, multi-modal input and dynamic rule configuration are supported, the flexibility of data management is improved, and the auditing and traceability capabilities are enhanced, facilitating subsequent auditing and traceability, enhancing the transparency of the system, and thus reducing the human and operation costs through the automated process and efficient data management.

[0083] Based on this, an embodiment of the present application provides a method for storing meeting records. Refer to Figure 2 , Figure 2 which is a schematic flowchart provided for the second embodiment of the method for storing meeting records of the present application.

[0084] In a feasible implementation manner, the training process of the meeting evidence analysis model includes:

[0085] Step S21, obtaining a plurality of groups of historical meeting data and performing data augmentation on each of the historical meeting data;

[0086] It should be noted that the historical meeting data refers to the records and related materials of past completed meetings, which are used to train and optimize the meeting evidence analysis model, including meeting record texts, voice transcripts, PPTs or other presentation materials, meeting minutes, information of participants, timestamp information, and other relevant documents, such as the agenda before the meeting, the action plan after the meeting, etc., which are not limited here.

[0087] Specifically, a plurality of groups of historical meeting data are obtained and data augmentation is performed on each of the historical meeting data. In one embodiment, the data augmentation methods are shown in Table 1:

[0088] Table 1

[0089]

[0090]

[0091] Step S22: Through a preset model encoder, perform entity recognition, decision point detection, and task relationship extraction on each of the augmented historical meeting data to obtain several groups of model training data;

[0092] It should be noted that the preset model encoder refers to a pre-trained language model used to encode the input text data into high-dimensional feature representations. It is usually implemented based on deep learning architectures such as BERT, Transformer, etc., so as to convert the text data into feature vectors that the model can process, capture the semantics and context information of the text, and provide a unified feature representation for multiple downstream tasks (such as entity recognition, decision point detection, etc.), thereby reducing the time and computing resources for training the model from scratch.

[0093] Furthermore, it should be noted that the model training data refers to the data that has been preprocessed and augmented and is used to train the meeting evidence analysis model, which is the annotation information corresponding to the original historical meeting data.

[0094] Specifically, through the first task head in the preset model encoder, perform entity recognition on each of the augmented historical meeting data to obtain several groups of model training data. Then, through the second task head in the preset model encoder, perform decision point detection on each of the augmented historical meeting data to obtain several groups of model training data. Thus, through the third task head in the preset model encoder, perform task relationship extraction on each of the augmented historical meeting data to obtain several groups of model training data.

[0095] Step S23: Input each of the historical meeting data and each of the model training data into the initial meeting evidence analysis model for iterative training to obtain the meeting evidence analysis model.

[0096] It should be noted that the initial meeting evidence analysis model can be further trained by MLM (Masked Language Modeling) using each of the historical meeting data and each of the model training data based on the general domain BERT model.

[0097] Specifically, perform data preprocessing on each of the historical meeting data and each of the model training data, such as data cleaning, normalization, etc., so as to input each of the historical meeting data and each of the model training data into the initial meeting evidence analysis model. Among them, use adversarial training (Gradient Reversal Layer) to reduce the distribution difference of meeting styles in different industries, and at the same time introduce curriculum learning, that is, first train meeting records with strong structure, and then expand to free discussion scenarios, to obtain the predicted values output by the initial meeting evidence analysis model. Furthermore, based on the predicted values and the sample labels, use the loss function to calculate the model loss value. In this embodiment, the loss function can be set according to actual needs and will not be specifically limited here. After calculating the model loss value, this training process ends, and then use the error backpropagation algorithm to update the model parameters in the initial meeting evidence analysis model, and then obtain a new round of loss function values through weighted summation of the loss function for the next training. During the training process, determine whether the updated initial meeting evidence analysis model meets the preset training end conditions. If it meets, use the updated initial meeting evidence analysis model as the pending meeting evidence analysis model. If it does not meet, continue to train the model. Among them, the preset training end conditions include loss convergence and reaching the maximum iteration number threshold, etc.

[0098] In this embodiment, by obtaining several groups of historical meeting data, and performing data augmentation on each of the historical meeting data, and then through a preset model encoder, perform entity recognition, decision point detection, and task relationship extraction on each of the historical meeting data after data augmentation to obtain several groups of model training data, so as to input each of the historical meeting data and each of the model training data into the initial meeting evidence analysis model for iterative training to obtain the meeting evidence analysis model. Furthermore, generate more diverse training data through data augmentation (such as back translation, synonym replacement, sentence recombination, etc.) to reduce the model's dependence on specific data formats, so that the model can more accurately identify key information when facing meeting records with different styles and different expressions, and through multi-task learning, simultaneously perform entity recognition, decision point detection, and task relationship extraction, so that the model can learn the shared features and correlations between different tasks, and further improve the performance of the model on each task and the generalization ability of the model.

[0099] Based on this, the embodiment of the present application provides a method for depositing meeting records, referring to Figure 3 , Figure 3 which is the schematic flowchart provided for Embodiment 3 of the method for depositing meeting records of the present application.

[0100] In a feasible implementation manner, the preset model encoder is used to perform entity recognition, decision point detection, and task relationship extraction on each of the augmented historical meeting data to obtain several groups of model training data, including:

[0101] Step S31: Use the first task head in the preset model encoder to perform entity recognition on each of the augmented historical meeting data to obtain several groups of model training data;

[0102] It should be noted that the first task head refers to the output layer in the model specifically for entity recognition (Named Entity Recognition, NER), which is used to identify entities with specific meanings from the text, such as person names, dates, locations, organization names, etc. The output format is usually in the sequence annotation format. For example, the BIO (Begin / Inside / Outside) annotation method is used to assign an entity label to each word.

[0103] Specifically, use the first task head in the preset model encoder to perform entity recognition on each of the augmented historical meeting data to obtain several groups of model training data.

[0104] Step S32: Use the second task head in the preset model encoder to perform decision point detection on each of the augmented historical meeting data to obtain several groups of model training data;

[0105] It should be noted that the second task head refers to the output layer in the model specifically for decision point detection (Decision Point Detection), which is used to identify key decision points in the text, such as key information like decisions, approvals, and passes in the meeting. The output format is usually a sentence-level label, marking which sentences contain decision points.

[0106] Specifically, use the second task head in the preset model encoder to perform decision point detection on each of the augmented historical meeting data to obtain several groups of model training data. Among them, keyword triggering can be achieved through semantic feature extraction, such as screening sentences containing verbs like "decide", "approve", "pass", etc. At the same time, the dependency tree with the `ROOT` node as a decision-making verb is detected through dependency syntax analysis. Furthermore, context-aware classification, such as using the BiLSTM-CRF model, is used to combine the context before and after to determine whether it is a valid decision (to avoid misjudging negative situations like "undecided").

[0107] Step S33: Use the third task head in the preset model encoder to perform task relationship extraction on each of the augmented historical meeting data to obtain several groups of model training data.

[0108] It should be noted that the third task head refers to the output layer in the model specifically used for task relation extraction, which is used to extract the task assignment relationships from the text, such as the responsible person, task content, time node, etc. The output format is usually in the form of triples, representing the subject, relation, and object of the task. Among them, the task classification system can refer to Table 2, and the task assignment system needs to be customized according to the actual workflow of the enterprise. In the initial stage, it can focus on high-frequency task types (such as responsible person / time node), and then gradually expand to complex scenarios such as resource collaboration.

[0109] Table 2

[0110]

[0111]

[0112] Specifically, through the third task head in the preset model encoder, task relation extraction is performed on each of the augmented historical meeting data to obtain several sets of model training data. In one embodiment, the structured extraction of task assignment first performs joint entity relation extraction, such as using a Span-based annotation scheme to identify task elements and establish associations, and then post-processes the rules to merge scattered expressions. For example, splitting "Engineer Wang is responsible for front-end development and Engineer Li is responsible for back-end" into two independent tasks, and normalizing the time, such as converting "next Wednesday" to the specific date "2024-03-20".

[0113] In addition, during the process of entity recognition, decision point detection, and task relation extraction for each of the augmented historical meeting data, the data output by the task head is evaluated respectively. If it does not meet the standard, continue to learn. The evaluation indicators and passing values can refer to Table 3.

[0114] Table 3

[0115]

[0116] In this embodiment, through the first task head in the preset model encoder, entity recognition is performed on each of the historical meeting data after data augmentation to obtain several groups of model training data. Then, through the second task head in the preset model encoder, decision point detection is performed on each of the historical meeting data after data augmentation to obtain several groups of model training data. Thus, through the third task head in the preset model encoder, task relationship extraction is performed on each of the historical meeting data after data augmentation to obtain several groups of model training data. Furthermore, a unified feature representation is provided for all tasks, reducing duplicate calculations, saving computing resources, and being able to learn common language patterns from multiple tasks, reducing the dependence on a large amount of labeled data, and potentially improving the performance of the model on each task. Because the shared feature representation can capture more comprehensive language information, the model learns multiple tasks simultaneously and can learn more general features instead of overfitting the data of a specific task, reducing data overfitting.

[0117] Based on this, an embodiment of the present application provides a method for depositing meeting records. Refer to Figure 4 , Figure 4 which is a schematic flowchart provided for Embodiment 4 of the method for depositing meeting records of the present application.

[0118] In a feasible implementation manner, the generating of the digital signature and the public key certificate based on the meeting organizer information and the target content to be deposited includes:

[0119] Step S41, obtain an encryption key pair, where the encryption key pair includes a public key and a private key;

[0120] It should be noted that the encryption key pair refers to a pair of asymmetric encryption keys, including a public key and a private key. They are generated based on asymmetric encryption algorithms (such as the RSA (Rivest-Shamir-Adleman) algorithm, the ECC (Elliptic Curve Cryptography) algorithm) and are used for encrypting and decrypting data. The public key is used for encrypting data or verifying signatures and can be publicly distributed, while the private key is used for decrypting data or generating signatures and must be kept confidential and can only be held by the key owner.

[0121] Specifically, in one embodiment, the organizer locally generates an asymmetric encryption key pair (RSA 2048 / ECC), and the private key is only stored in a secure environment (such as an HSM hardware encryption module, a TEE trusted execution environment).

[0122] Step S42, generate a public key certificate according to the public key and the meeting organizer information;

[0123] Specifically, a digital public key certificate is issued by a third-party CA (Certificate Authority) institution (such as CFCA - China Financial Certification Authority), and the public key certificate contains the organizer's identity information + public key. Alternatively, a self-built PKI (Public Key Infrastructure) system can be adopted and signed by the enterprise root certificate (which needs to comply with the requirements of the Electronic Signature Law).

[0124] Step S43: Perform a hash calculation on the target evidence content to obtain a first hash value, and encrypt the first hash value according to the private key to generate a digital signature.

[0125] It should be noted that the first hash value refers to the unique digital fingerprint obtained by performing a hash calculation on the target evidence content (such as the key information of the meeting record). Among them, the first hash value can be generated by a hash algorithm (such as SHA-256) and has irreversibility and uniqueness.

[0126] In this embodiment, an encryption key pair is obtained, where the encryption key pair includes a public key and a private key. Then, according to the public key and the meeting organizer information, a public key certificate is generated. Thus, a hash calculation is performed on the target evidence content to obtain a first hash value, and the first hash value is encrypted according to the private key to generate a digital signature. Furthermore, it is realized that only the meeting organizer with the private key can generate a valid signature, while the public key can be publicly used to verify the signature, ensuring the security of data during transmission and storage. Furthermore, the integrity and immutability of the data are ensured through the digital signature. Any modification to the target evidence content will cause a change in the hash value, resulting in a failed signature verification, ensuring the authenticity and integrity of the evidence content. At the same time, the public key certificate is issued by a trusted certificate authority or within the enterprise, ensuring the credibility of the public key and the identity binding. In a legal scenario, the public key certificate can be used as evidence to prove that the evidence content is generated by an organizer with a specific identity, enhancing the legal effect of the evidence content. Moreover, the digital signature complies with the requirements of laws and regulations such as the Electronic Signature Law and has the same legal effect as a handwritten signature. Therefore, in a legal lawsuit or compliance inspection, the authenticity and integrity of the evidence content can be verified through the digital signature, reducing legal disputes.

[0127] In a feasible implementation manner, generating a signature data packet based on the digital signature, the public key certificate, and the target evidence content, and storing the signature data packet in the distributed ledger corresponding to the blockchain evidence interface includes:

[0128] Step S51: Obtain the deposition metadata and deposition timestamp corresponding to the blockchain deposition interface;

[0129] It should be noted that the deposition metadata refers to the additional information related to the deposition content, which describes the source, storage location, generation method, etc. of the deposition, but does not include the deposition content itself. The deposition timestamp refers to the specific time point when the deposition content is generated or stored on the blockchain.

[0130] Step S52: Associate and combine the deposition metadata and the deposition timestamp with the digital signature, the public key certificate, and the target deposition content to generate a signature data packet;

[0131] Step S53: Store the signature data packet in the distributed ledger corresponding to the blockchain deposition interface.

[0132] It should be noted that the characteristics of the blockchain (tamper-proof, traceable) and the storage cost limitation determine the hybrid architecture of "only key vouchers are stored on the chain, and the complete report is stored off-chain". Among them, the content stored on the chain can be referred to in Table 4.

[0133] Table 4

[0134]

[0135] Specifically, when storing the signature data packet in the distributed ledger corresponding to the blockchain deposition interface, in one embodiment, the signature data packet is uploaded to a consortium chain (such as Ant Chain, Tencent Zhixin Chain) to generate a deposition hash and a timestamp, and is deposited at a notary office. For example, when docking with a judicial deposition platform (such as Notary Cloud), a deposition number and judicial endorsement are obtained.

[0136] In this embodiment, by obtaining the deposition metadata and deposition timestamp corresponding to the blockchain deposition interface, and then associating and combining the deposition metadata and the deposition timestamp with the digital signature, the public key certificate, and the target deposition content to generate a signature data packet, thereby storing the signature data packet in the distributed ledger corresponding to the blockchain deposition interface. Furthermore, through blockchain and digital signature technologies, the authenticity and integrity of the deposition content are ensured, the security and immutability of the data are improved, thereby enhancing the legal effect, and the audit efficiency and system transparency are improved through the traceability of the deposition metadata and the blockchain, enhancing the credibility of the system, and reducing the dependence on a single centralized system.

[0137] In a feasible implementation manner, after generating the signature data packet based on the digital signature, the public key certificate, and the target deposition content, and storing the signature data packet in the distributed ledger corresponding to the blockchain deposition interface, it further includes:

[0138] Step S61: Obtain the complete original meeting record and the deposit index information of the signature data packet;

[0139] It should be noted that the complete original meeting record refers to the original text record generated during the meeting without modification, including all the content and details of the meeting. The deposit index information refers to the metadata related to the deposited content, which is used to describe the storage location and related information of the deposit on the blockchain, provide the specific location of the deposited content on the blockchain, and facilitate quick retrieval and verification.

[0140] Step S62: Input the complete original meeting record into the meeting deposit analysis model to obtain the meeting record report output by the meeting deposit analysis model;

[0141] It should be noted that the meeting record report refers to a "standardized, machine-readable, and extensible" structured data format report generated by parsing the original meeting content through the meeting deposit analysis model, which contains the key information and decision points of the meeting. The core features can be referred to in Table 5.

[0142] Table 5

[0143]

[0144]

[0145] Step S63: Associate and combine the deposit index information, the meeting record report, and the complete original meeting record, and store them in an off-chain storage location.

[0146] It should be noted that the off-chain storage location refers to storing data in a storage system outside the blockchain, including private clouds, distributed storage systems, or traditional databases, etc., so as to reduce storage costs, improve storage efficiency, enhance privacy protection. Off-chain storage supports data encryption and access control, enhancing privacy protection. Among them, the content of off-chain storage can be referred to in Table 6.

[0147] Table 6

[0148]

[0149] In this embodiment, by obtaining the complete original meeting record and the deposit index information of the signature data packet, the complete original meeting record is then input into the meeting deposit analysis model to obtain a meeting record report output by the meeting deposit analysis model. Then, the deposit index information, the meeting record report, and the complete original meeting record are associated and combined and stored in an off-chain storage location. Furthermore, off-chain storage allows the system to process a large amount of data, while using the immutable feature of the blockchain to ensure the security of the deposited content. That is, off-chain storage is used to reduce costs while maintaining the immutable feature of the blockchain, and at the same time quickly locate and verify the deposited content, improve the traceability of the system, and ensure the integrity and consistency of the deposited content.

[0150] In a feasible implementation manner, after generating a signature data packet based on the digital signature, the public key certificate, and the target deposited content, and storing the signature data packet in the distributed ledger corresponding to the blockchain deposit interface, it further includes:

[0151] Step S71: Connect to the blockchain deposit interface, download the signature data packet, and parse the signature data packet to obtain the digital signature and the public key certificate.

[0152] Step S72: Calculate the second hash value of the target deposited content, and decrypt the digital signature based on the public key in the public key certificate to verify whether the first hash value of the decrypted digital signature is consistent with the second hash value.

[0153] It should be noted that the second hash value refers to the hash value recalculated for the target deposited content (such as the key information of the meeting record) during the verification stage, and is used to verify whether the deposited content has been tampered with during storage and transmission.

[0154] Specifically, the second hash value of the target deposited content can be calculated through the SHA-256 algorithm, and the digital signature is decrypted based on the public key in the public key certificate to verify whether the first hash value of the decrypted digital signature is consistent with the second hash value.

[0155] Step S73: If so, verify the validity of the public key certificate and generate a meeting deposit verification result.

[0156] Specifically, if so, it means that the deposited content has not been tampered with. Then, verify the validity of the public key certificate, such as whether it has expired and whether the issuing agency is trustworthy, etc. Thus, according to the verification result and the comparison result, a meeting deposit verification result is generated.

[0157] In this embodiment, by connecting to the blockchain evidence storage interface, downloading the signature data packet, and parsing the signature data packet, the digital signature and the public key certificate are obtained. Then, the second hash value of the target evidence storage content is calculated, and the digital signature is decrypted based on the public key in the public key certificate to verify whether the first hash value of the decrypted digital signature is consistent with the second hash value. Thus, if they are consistent, the validity of the public key certificate is verified, and a meeting evidence storage verification result is generated. Furthermore, it is ensured that the evidence storage content has not been tampered with during storage and transmission. That is, if the two hash values are consistent, it indicates that the evidence storage content has not been modified since its generation, ensuring the integrity and authenticity of the data. At the same time, the digital signature is generated by a private key, and only the corresponding public key can decrypt it. That is, verifying the digital signature can confirm that the evidence storage content is generated by an organizer with a specific identity, thereby ensuring the reliability and non-repudiation of the source of the evidence storage content. Furthermore, by verifying the validity of the public key certificate (such as whether the certificate has expired, whether the issuing agency is trustworthy, etc.), the legal effect of the evidence storage content is further ensured. Finally, through a multi-layer mechanism of hash value verification, digital signature verification, and public key certificate verification, the credibility of the evidence storage content is ensured, enabling users to trust the authenticity and integrity of the evidence storage content, reducing doubts about the data authenticity, improving the transparency and credibility of the system, and enhancing users' trust in the system.

[0158] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not imply the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0159] This application also provides a meeting record evidence storage device. Please refer to Figure 5 , the meeting record evidence storage device includes:

[0160] An acquisition module 51, configured to acquire a blockchain evidence storage interface, meeting organizer information, and a real-time meeting record corresponding to a target meeting;

[0161] An analysis module 52, configured to input the real-time meeting record into a meeting evidence storage analysis model to obtain key meeting content output by the meeting evidence storage analysis model, and compare the key meeting content with a preset dynamic rule to obtain target evidence storage content;

[0162] A generation module 53, configured to generate a digital signature and a public key certificate based on the meeting organizer information and the target evidence storage content;

[0163] An evidence storage module 54, configured to connect to the blockchain evidence storage interface, generate a signature data packet based on the digital signature, the public key certificate, and the target evidence storage content, and store the signature data packet in a distributed ledger corresponding to the blockchain evidence storage interface.

[0164] The meeting record certification device is further configured to:

[0165] Obtain a number of groups of historical meeting data, and perform data augmentation on each group of the historical meeting data;

[0166] Through a preset model encoder, perform entity recognition, decision point detection, and task relationship extraction on each group of the historical meeting data after data augmentation, and obtain a number of groups of model training data;

[0167] Input each group of the historical meeting data and each group of the model training data into an initial meeting certification analysis model for iterative training to obtain a meeting certification analysis model.

[0168] The meeting record certification device is further configured to:

[0169] Through a first task head in the preset model encoder, perform entity recognition on each group of the historical meeting data after data augmentation to obtain a number of groups of model training data;

[0170] Through a second task head in the preset model encoder, perform decision point detection on each group of the historical meeting data after data augmentation to obtain a number of groups of model training data;

[0171] Through a third task head in the preset model encoder, perform task relationship extraction on each group of the historical meeting data after data augmentation to obtain a number of groups of model training data.

[0172] The meeting record certification device is further configured to:

[0173] Obtain an encryption key pair, where the encryption key pair includes a public key and a private key;

[0174] Generate a public key certificate according to the public key and the meeting organizer information;

[0175] Perform a hash calculation on the target certified content to obtain a first hash value, and encrypt the first hash value according to the private key to generate a digital signature.

[0176] The meeting record certification device is further configured to:

[0177] Obtain the certified metadata and the certification timestamp corresponding to the blockchain certification interface;

[0178] Associate and combine the certified metadata and the certification timestamp with the digital signature, the public key certificate, and the target certified content to generate a signature data packet;

[0179] Store the signature data packet in the distributed ledger corresponding to the blockchain certification interface.

[0180] The meeting record evidence storage device is further configured to:

[0181] Obtain the complete original meeting record and the evidence storage index information of the signature data packet;

[0182] Input the complete original meeting record into the meeting evidence analysis model to obtain a meeting record report output by the meeting evidence analysis model;

[0183] Associate and combine the evidence storage index information, the meeting record report, and the complete original meeting record, and store them in an off-chain storage location.

[0184] The meeting record evidence storage device is further configured to:

[0185] Connect to the blockchain evidence storage interface, download the signature data packet, and parse the signature data packet to obtain the digital signature and the public key certificate;

[0186] Calculate the second hash value of the target evidence storage content, and decrypt the digital signature based on the public key in the public key certificate to verify whether the first hash value of the decrypted digital signature is consistent with the second hash value;

[0187] If so, verify the validity of the public key certificate and generate a meeting evidence storage verification result.

[0188] The meeting record evidence storage device provided by this application adopts the meeting record evidence storage method in the above embodiment, and can solve the technical problems in the background technology. Compared with the prior art, the beneficial effects of the meeting record evidence storage device provided by this application are the same as those of the meeting record evidence storage method provided by the above embodiment, and other technical features in the meeting record evidence storage device are the same as the features disclosed in the above embodiment method, which will not be elaborated here.

[0189] This application provides a meeting record evidence storage device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the meeting record evidence storage method in the first embodiment above.

[0190] Next, refer to Figure 6, which shows a schematic structural diagram of a meeting record evidence storage device suitable for implementing the embodiments of the present application. The meeting record evidence storage device in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The shown meeting record evidence storage device is only an example and should not impose any limitations on the functions and usage scope of the embodiments of the present application.

[0191] As Figure 6 shown, the meeting record evidence storage device may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory 1002 or the program loaded from the storage device 1003 into the random access memory 1004. In the random access memory 1004, various programs and data required for the operation of the meeting record evidence storage device are also stored. The processing device 1001, the read-only memory 1002, and the random access memory 1004 are connected to each other through a bus 1005. An input / output interface 1006 is also connected to the bus. Generally, the following systems may be connected to the input / output interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the meeting record evidence storage device to communicate with other devices wirelessly or wiredly to exchange data. Although the figure shows a meeting record evidence storage device with various systems, it should be understood that it is not required to implement or have all the shown systems. More or fewer systems may be implemented or had alternatively.

[0192] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product that includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a read-only memory 1002. When the computer program is executed by a processing device 1001, the above functions defined in the methods of the embodiments disclosed in the present application are executed.

[0193] The meeting record certification device provided by the present application adopts the meeting record certification method in the above embodiment and can solve the technical problems in the background art. Compared with the prior art, the beneficial effects of the meeting record certification device provided by the present application are the same as those of the meeting record certification method provided by the above embodiment, and other technical features in the meeting record certification device are the same as the features disclosed in the method of the previous embodiment, which will not be elaborated here.

[0194] It should be understood that the various parts disclosed in the present application can be implemented by hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in a suitable manner in any one or more embodiments or examples.

[0195] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0196] The present application provides a computer-readable storage medium having computer-readable program instructions (i.e., computer programs) stored thereon, and the computer-readable program instructions are used to execute the meeting record certification method in the above embodiment.

[0197] The computer-readable storage medium provided by this application can be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or flash memory, optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this embodiment, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system or device. The program code contained on the computer-readable storage medium can be transmitted by any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0198] The above computer-readable storage medium can be included in the meeting record certification device; it can also exist separately without being assembled into the meeting record certification device.

[0199] The above computer-readable storage medium carries one or more programs. When the above one or more programs are executed by the meeting record certification device, the meeting record certification device is caused to:

[0200] Obtain the blockchain certification interface, meeting organizer information, and the real-time meeting record corresponding to the target meeting;

[0201] Input the real-time meeting record into the meeting certification analysis model to obtain the key meeting content output by the meeting certification analysis model, and compare the key meeting content with the preset dynamic rules to obtain the target certification content;

[0202] Generate a digital signature and a public key certificate based on the meeting organizer information and the target certification content;

[0203] Connect to the blockchain certification interface, generate a signature data packet based on the digital signature, the public key certificate, and the target certification content, and store the signature data packet in the distributed ledger corresponding to the blockchain certification interface.

[0204] Computer program code for performing the operations of this application can be written in one or more programming languages or combinations thereof. The above-mentioned programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, by connecting through the Internet using an Internet service provider).

[0205] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of the code, and this module, program segment, or part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0206] The modules described in the embodiments of this application can be implemented in software or in hardware. Among them, the name of the module does not constitute a limitation to the unit itself in some cases.

[0207] The readable storage medium provided by this application is a computer-readable storage medium. The computer-readable storage medium stores computer-readable program instructions (i.e., computer programs) for performing the above-mentioned meeting record certification method, and can solve the technical problems in the background technology. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by this application are the same as those of the meeting record certification method provided by the above embodiments, and will not be elaborated here.

[0208] An embodiment of the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the steps of the meeting record deposit method as described above.

[0209] The computer program product provided by the present application can solve the technical problems in the background art. Compared with the prior art, the beneficial effects of the computer program product provided by the embodiment of the present application are the same as those of the meeting record deposit method provided by the above embodiment, and will not be elaborated here.

[0210] The above are only partial embodiments of the present application, and thus do not limit the patent scope of the present application. Any equivalent structural transformation made by using the content of the specification and drawings of the present application under the technical concept of the present application, or direct / indirect application in other related technical fields, is included in the patent protection scope of the present application.

Claims

1. A method for preserving meeting records, characterized in that: include: Obtain blockchain evidence storage interface, meeting organizer information, and real-time meeting records corresponding to the target meeting; Input the real-time meeting record into the meeting evidence analysis model to obtain the key content of the meeting output by the meeting evidence analysis model, and compare the key content of the meeting with the preset dynamic rules to obtain the target evidence content; Generate a digital signature and a public key certificate based on the conference organizer information and the target evidence content; Connect to the blockchain evidence storage interface, generate a signature data packet based on the digital signature, the public key certificate and the target evidence content, and store the signature data packet in the distributed ledger corresponding to the blockchain evidence storage interface.

2. The conference record storage method according to claim 1, characterized in that: The training process of the conference evidence analysis model includes: Acquire several groups of historical meeting data, and perform data enhancement on each of the historical meeting data; By using a preset model encoder, entity recognition, decision point detection, and task relationship extraction are performed on each of the historical meeting data after data enhancement to obtain several groups of model training data; The historical meeting data and the model training data are input into the initial meeting evidence analysis model for iterative training to obtain the meeting evidence analysis model.

3. The conference record storage method as claimed in claim 2, characterized in that: The preset model encoder is used to perform entity recognition, decision point detection, and task relationship extraction on each of the historical meeting data after data enhancement to obtain several groups of model training data, including: By using the first task head in the preset model encoder, entity recognition is performed on each of the historical meeting data after data enhancement to obtain several groups of model training data; By using the second task head in the preset model encoder, decision point detection is performed on each of the historical meeting data after data enhancement to obtain several groups of model training data; By using the third task head in the preset model encoder, task relationships are extracted from each of the historical meeting data after data enhancement to obtain several groups of model training data.

4. The conference record storage method according to claim 1, characterized in that: The generating of a digital signature and a public key certificate based on the conference organizer information and the target evidence content includes: Obtaining an encryption key pair, wherein the encryption key pair includes a public key and a private key; Generate a public key certificate according to the public key and the conference organizer information; A hash calculation is performed on the target evidence content to obtain a first hash value, and the first hash value is encrypted according to the private key to generate a digital signature.

5. The conference record storage method according to claim 1, characterized in that: The generating of a signature data packet based on the digital signature, the public key certificate and the target evidence content, and storing the signature data packet in a distributed ledger corresponding to the blockchain evidence interface, includes: Obtain the evidence metadata and evidence timestamp corresponding to the blockchain evidence interface; Associating and combining the evidence metadata and the evidence timestamp with the digital signature, the public key certificate and the target evidence content to generate a signature data packet; The signed data packet is stored in the distributed ledger corresponding to the blockchain evidence storage interface.

6. The conference record storage method according to claim 1, characterized in that: After generating a signature data packet based on the digital signature, the public key certificate and the target evidence content, and storing the signature data packet in the distributed ledger corresponding to the blockchain evidence interface, the method further includes: Obtain the complete record of the original meeting and the evidence index information of the signed data package; Inputting the original complete meeting record into the meeting evidence analysis model to obtain a meeting record report output by the meeting evidence analysis model; The evidence index information, the meeting record report and the original meeting complete record are associated and combined and stored in an off-chain storage location.

7. The conference record storage method according to claim 1, characterized in that: After generating a signature data packet based on the digital signature, the public key certificate and the target evidence content, and storing the signature data packet in the distributed ledger corresponding to the blockchain evidence interface, the method further includes: Connecting to the blockchain evidence storage interface, downloading the signature data packet, and parsing the signature data packet to obtain the digital signature and the public key certificate; Calculate a second hash value of the target stored evidence content, and decrypt the digital signature based on the public key in the public key certificate to verify whether the first hash value of the digital signature after decryption is consistent with the second hash value; If so, verify the validity of the public key certificate and generate a conference evidence verification result.

8. A conference record and evidence storage device, characterized in that: include: The acquisition module is used to obtain the blockchain evidence storage interface, meeting organizer information, and real-time meeting records corresponding to the target meeting; An analysis module, used to input the real-time meeting record into a meeting evidence analysis model, obtain the key content of the meeting output by the meeting evidence analysis model, and compare the key content of the meeting with a preset dynamic rule to obtain the target evidence content; A generation module, used to generate a digital signature and a public key certificate based on the conference organizer information and the target evidence content; The evidence storage module is used to connect to the blockchain evidence storage interface, generate a signature data packet based on the digital signature, the public key certificate and the target evidence content, and store the signature data packet in the distributed ledger corresponding to the blockchain evidence storage interface.

9. A conference record and evidence storage device, characterized in that: The conference record and evidence storage device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the conference record and evidence storage method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the meeting record storage method as described in any one of claims 1 to 7 are implemented.

Images