Cooperative task management method and system

Abstract

The invention discloses a collaborative task management method and system. The method comprises the steps of obtaining a target task file corresponding to a target task; performing text extraction and semantic analysis on the target task file to obtain text summary information of the target task file, and sending the target task file and the text summary information to an approval node; under the condition that the target task is a project task and the approval is passed, analyzing the text summary information by using a pre-trained target and key achievement OKR generation model, generating a plurality of sub-tasks corresponding to the target task, and distributing each sub-task to a corresponding execution node; and periodically detecting the task execution state of each execution node, and generating early warning prompt information when the task execution state is abnormal. According to the method and the device, the technical problems of low collaborative task approval efficiency and inaccurate task tracking management in related office scenes are solved.

Description

Technical Field

[0001] The present application relates to the field of computer software technology, and in particular to a collaborative task management method and system. Background Art

[0002] In today's highly information-rich enterprise environment, office automation (OA) systems, as core tools supporting daily operations and management, face unprecedented challenges. Traditional collaborative OA systems rely on manual, step-by-step review and decision-making during the document approval process. While this model offers stability and rigor, it can be inefficient in the fast-paced, high-efficiency world of modern enterprises. From drafting to final approval, documents undergo multiple levels of review, resulting in lengthy approval cycles. This is particularly true for large and complex documents, requiring reviewers to spend significant time reading and understanding them, then making decisions based on their personal experience and business knowledge. This process not only consumes valuable management resources but can also lead to decision delays due to information asymmetry or misunderstandings, hindering the smooth flow of business processes. Task management, on the other hand, is an essential function of collaborative OA systems. However, traditional tracking methods, which rely on manual progress reports and regular meetings, are no longer able to meet the demands of modern enterprise management. Under this model, task execution and monitoring lack real-time and accuracy. Progress reports can be distorted by subjective factors, and the completion status of key milestones is difficult to promptly reflect. This prevents managers from quickly identifying problems and risks and taking effective measures. More importantly, the lack of an effective goal management mechanism leads to inconsistent understanding of work objectives among team members, inefficient communication, and potentially irrational task allocation, which in turn impacts the efficiency and effectiveness of team collaboration.

[0003] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0004] The embodiments of the present application provide a collaborative task management method and system to at least solve the technical problems of low collaborative task approval efficiency and inaccurate task tracking management in relevant office scenarios.

[0005] According to one aspect of an embodiment of the present application, a collaborative task management method is provided, including: obtaining a target task file corresponding to a target task; performing text extraction and semantic analysis on the target task file to obtain text summary information of the target task file, and sending the target task file and the text summary information to an approval node; when the target task is a project-type task and is approved, using a pre-trained objectives and key results (OKR) generation model to analyze the text summary information, generate multiple subtasks corresponding to the target task, and distribute each subtask to a corresponding execution node; periodically detecting the task execution status of each execution node, and generating early warning prompt information when the task execution status is abnormal.

[0006] Optionally, obtaining the target task file corresponding to the target task includes: receiving the task file uploaded by the target object, and performing file signature verification and Multipurpose Internet Mail Extensions type verification on the task file to determine the file format of the task file; when the file format meets the preset format requirements, determining that the task file is the target task file; when the file format does not meet the preset format requirements, generating an exception prompt information, wherein the exception prompt information is used to prompt that the format of the uploaded task file is abnormal and the task file needs to be uploaded again.

[0007] Optionally, text extraction and semantic analysis are performed on the target task file to obtain text summary information of the target task file, including: calling a file parsing tool corresponding to the file format of the target task file to extract the first text content of the target task file; performing text cleaning on the first text content to obtain the second text content, wherein the second text content retains the paragraph structure identifier; semantically segmenting the second text content based on the paragraph structure identifier and / or the semantic similarity between different sentences in the second text content to obtain multiple structured data objects; mapping each structured data object into a multi-dimensional semantic vector; and analyzing the multiple semantic vectors using a target macro model to obtain text summary information of the target task file, wherein the target macro model is obtained by fine-tuning the general macro model using knowledge of the technical field to which the target task belongs.

[0008] Optionally, after sending the target task file and text summary information to the approval node, the method further includes: periodically detecting the task approval status of the approval node, and generating an approval reminder message if the approval node fails to perform the approval operation within a preset number of consecutive cycles.

[0009] Optionally, a pre-trained OKR generation model is used to analyze the text summary information to generate multiple subtasks corresponding to the target task, including: extracting multiple task goals from the text summary information using the OKR generation model; for each task goal, deducing multiple key results corresponding to the task goal based on the thinking chain model, and generating a subtask based on each key result, wherein each key result includes at least: a subtask type and a subtask indicator.

[0010] Optionally, before generating multiple subtasks based on multiple key results, the above method also includes: calling a pre-trained feasibility assessment model to analyze each key result obtained by deduction, determining the feasibility score of each key result, and correcting the key results whose feasibility score is lower than a preset score threshold; and / or, based on the retrieval enhancement generation algorithm, retrieving each historical key result corresponding to the historical task matching the target task from the historical task database, and correcting each key result obtained by deduction based on each historical key result; and / or, correcting each key result obtained by deduction based on the knowledge graph of the technical field to which the target task belongs.

[0011] Optionally, the task execution status of each execution node is periodically detected, and a warning prompt message is generated when the task execution status is abnormal, including: for each execution node, the task execution status of the execution node is periodically detected, and the task execution status is compared with the subtask indicators of the subtask executed by the execution node; if after a preset number of consecutive cycles, the task execution status and the subtask indicators of the subtask executed by the execution node are always mismatched, it is determined that the task execution status is abnormal, and a warning prompt message is generated, wherein the warning prompt message is used to prompt that the task execution status of the execution node is abnormal, and it is necessary to adjust the subtask executed by the execution node, or reassign the execution node to the subtask.

[0012] According to another aspect of an embodiment of the present application, a collaborative task management system is also provided, including: an acquisition module for acquiring a target task file corresponding to a target task; an analysis module for performing text extraction and semantic analysis on the target task file to obtain text summary information of the target task file, and sending the target task file and text summary information to an approval node; a generation module for analyzing the text summary information using a pre-trained OKR generation model when the target task is a project-type task and is approved, generating multiple subtasks corresponding to the target task, and distributing each subtask to a corresponding execution node; a monitoring module for periodically detecting the task execution status of each execution node, and generating early warning prompt information when the task execution status is abnormal.

[0013] According to another aspect of an embodiment of the present application, a computer program product is further provided, the computer program product comprising: a computer program, wherein the computer program implements the above-mentioned collaborative task management method when executed by a processor.

[0014] According to another aspect of an embodiment of the present application, an electronic device is provided, which includes: a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to execute the above-mentioned collaborative task management method through the computer program.

[0015] In an embodiment of the present application, by performing text extraction and semantic analysis on the acquired target task file, text summary information of the target task file is obtained, and the target task file and text summary information are sent to the approval node, which can help the approval personnel quickly understand the key information in the content to be approved and improve the approval efficiency; when the target task is a project task and the approval is passed, the pre-trained goal and key result OKR generation model is used to analyze the text summary information, generate multiple subtasks corresponding to the target task, and distribute each subtask to the corresponding execution node, and periodically detect the task execution status of each execution node, which can help managers track the current task execution status and provide early warning of risks in the task execution process, thereby solving the technical problems of low efficiency in collaborative task approval and inaccurate task tracking management in related office scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0017] Figure 1 is a flowchart of an optional collaborative task management method according to an embodiment of the present application;

[0018] Figure 2 is a schematic structural diagram of an optional collaborative task management system according to an embodiment of the present application;

[0019] Figure 3 This is a schematic diagram of an optional electronic device structure according to an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0021] It should be noted that the terms "first", "second", etc. in the specification, claims, and drawings of the present application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0022] In order to better understand the embodiments of the present application, some nouns or terms that appear in the description of the embodiments of the present application are first translated and explained as follows:

[0023] OKR (Objectives and Key Results) is a widely adopted goal-setting and performance tracking framework. Originally introduced by Intel and popularized by tech giants like Google, it has become an effective tool for many companies to improve organizational effectiveness and team execution. The core concept of OKR is to guide and motivate the work of teams and individuals by setting clear objectives and measurable key results, ensuring their efforts are directly aligned with the company's strategic direction.

[0024] Multipurpose Internet Mail Extensions (MIME) type verification is a technical process used to verify that the actual type of a file or data transmitted over a network matches its declared type. It is primarily used in email protocols. The verification process typically includes the following steps: reading the MIME header, checking the file signature, analyzing the content, determining the actual type, and handling exceptions. In email protocols, the MIME type is typically declared in the header information of the data being transmitted. The recipient first reads the MIME type declaration from the header. Many file formats have a unique byte sequence that serves as a signature, such as the %PDF- prefix in PDF files. The verification process reads the first few bytes of the file and compares them to known MIME type signatures. For files without an obvious signature, the verification process may need to analyze the file content, looking for specific formatting features to confirm its true type. Through these steps, the verification process can determine the actual type of the file. If the actual type does not match the declared MIME type, it may indicate that the file has been tampered with or contains some form of security threat. If the MIME type verification fails, the system should take measures such as rejecting the file, requesting resend, or performing a more in-depth security scan to protect the system from potential malicious data attacks.

[0025] Hybrid Document Embedding (HDE): This is an advanced technique for document representation that combines traditional document analysis methods (such as the bag-of-words model) with modern deep learning techniques (such as word embeddings, sentence vectors, and document vectors) to create more comprehensive and expressive document representations. This technique is particularly effective when processing long documents and complex document structures because it can capture both local and global information about the document.

[0026] Named Entity Recognition (NER) is a natural language processing technique used to identify meaningful entities in text, such as names of people, places, organizations, time periods, and monetary values. NER technology is crucial in applications such as information extraction, question-answering systems, machine translation, and public opinion analysis. Its workflow includes: 1) text preprocessing, which involves performing basic natural language processing on the input text, such as word segmentation and part-of-speech tagging; 2) entity annotation, which involves using a trained machine learning model or rule-based methods to annotate entities in the text; 3) entity classification, which involves classifying identified entities into predefined categories, such as names of people, places, and organizations; and 4) output, where NER technology outputs the location and category of all entities in the text. This information can be used for further text understanding and information extraction. In addition to the aforementioned steps, NER models typically require a large amount of training data with entity annotations, trained using algorithms such as deep learning, and continuously optimized to improve recognition accuracy.

[0027] Sentence-BERT (Sentence-Bidirectional Encoder Representations from Transformers): The Sentence-BERT model is a variant of the BERT model, primarily designed for sentence-level semantic representation tasks such as semantic similarity calculation, text clustering, and text matching. Pre-trained on a large-scale corpus and then fine-tuned on specific sentence-level tasks, Sentence-BERT effectively captures the semantic features of sentences and generates high-quality sentence vector representations, demonstrating outstanding performance in multiple areas of natural language processing.

[0028] The SMART (Specific, Measurable, Achievable, Relevant, Time-bound) principle is a widely accepted goal-setting method that helps individuals and teams set clear, achievable goals, thereby improving work performance and project success. In the OKR methodology, the SMART principle is also used as a guide for developing key results, ensuring they are specific, measurable, and time-bound, thereby enhancing the effectiveness and actionability of goal management.

[0029] Example 1

[0030] According to an embodiment of the present application, a collaborative task management method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0031] Figure 1 is a flowchart of a collaborative task management method provided according to an embodiment of the present application, such as Figure 1 As shown, the method includes the following steps:

[0032] Step S102, obtaining the target task file corresponding to the target task;

[0033] Step S104: performing text extraction and semantic analysis on the target task file to obtain text summary information of the target task file, and sending the target task file and text summary information to the approval node;

[0034] Step S106: If the target task is a project-type task and has been approved, the pre-trained OKR generation model is used to analyze the text summary information, generate multiple subtasks corresponding to the target task, and distribute each subtask to the corresponding execution node;

[0035] Step S108: Periodically detect the task execution status of each execution node, and generate warning information when the task execution status is abnormal.

[0036] The following describes the various steps of the collaborative task management method in conjunction with a specific implementation process.

[0037] In the technical solution provided in the above step S102 of the present application, when a user uploads a task file to the OA system, the task file must first be verified to obtain a target task file corresponding to the target task, which can be specifically achieved in the following ways: receiving the task file uploaded by the target object, and performing file signature verification and multipurpose internet mail extension type verification on the task file to determine the file format of the task file; when the file format meets the preset format requirements, determining the task file as the target task file; when the file format does not meet the preset format requirements, generating an exception prompt information, wherein the exception prompt information is used to prompt that the format of the uploaded task file is abnormal and the task file needs to be uploaded again.

[0038] Specifically, the above process embodies an efficient and comprehensive mechanism for file reception and preliminary verification, ensuring that all task files uploaded to the collaborative OA system meet pre-set formatting standards and security requirements before entering the subsequent processing flow. Specifically, when the target user—an employee, department, or partner within the company—uploads a task file through the system interface, the system immediately activates an intelligent verification mechanism to conduct an in-depth analysis of the uploaded file. The system utilizes a dual verification mechanism: file signature verification and Multipurpose Internet Mail Extensions (MIMEE) type verification.

[0039] First, the system performs a file signature check. This involves examining the hexadecimal identifier in the file header, such as %PDF-1.4 for PDF files or PK for DOCX files. This confirms the file's true type and prevents formatting or malicious infiltration. Furthermore, the system utilizes Multipurpose Internet Mail Extensions (MIME) type verification to further verify the consistency of the file format, ensuring that the file type matches the actual content, avoiding decoding errors or security risks caused by inconsistent types.

[0040] After completing this preliminary verification, the system will determine whether the uploaded file type complies with the system's supported formats, such as PDF, DOCX, and XLSX, based on pre-set formatting requirements. If the file format fully complies, the system will automatically identify it as the target task file and prepare to proceed to the next step of intelligent parsing and process management. If the file format does not meet the pre-set standards, the system will immediately generate an exception message, clearly informing the uploader of the specific file format anomaly and providing guidance and suggestions for re-uploading to ensure a smooth process.

[0041] The generation and feedback of exception notifications is a key component of this technical solution. The system not only simply identifies the problem but also provides a user-friendly interface with a detailed explanation of why the uploaded file format doesn't match, including the specific file type and possible formatting risks. It also offers solutions, such as recommending the use of specific file conversion tools, prompting to check file header information, or directly providing online format conversion services. This improves the user experience and reduces technical barriers to use.

[0042] For the target task files obtained, text extraction and semantic analysis can be performed on the content of the target task files to generate text summary information of the target task files. Through intelligent processing of the target task files, key information can be quickly extracted from massive documents to provide accurate data support for subsequent file approval and task management.

[0043] As an optional implementation, performing text extraction and semantic analysis on the target task file to obtain text summary information of the target task file can be achieved in the following manner: calling a file parsing tool corresponding to the file format of the target task file to extract the first text content of the target task file; performing text cleaning on the first text content to obtain the second text content, wherein the second text content retains the paragraph structure identifier; performing semantic segmentation on the second text content based on the paragraph structure identifier and / or the semantic similarity between different sentences in the second text content to obtain multiple structured data objects; mapping each structured data object into a multi-dimensional semantic vector; using the target big model to analyze the multiple semantic vectors to obtain text summary information of the target task file, wherein the target big model is obtained by fine-tuning the general big model using the knowledge of the technical field to which the target task belongs.

[0044] During specific implementation, text extraction and semantic analysis are performed on the target task file to obtain text summary information of the target task file, which can be achieved through steps S1 to S5.

[0045] Step S1: Text Parsing and Content Extraction. A dedicated file parsing tool matching the target task's file format is used, such as PyMuPDF for PDF files or python-docx for DOCX files, to fully parse the file and extract the original first text content. This process not only includes the recognition and conversion of textual information but also includes the identification and conversion of non-textual elements such as tables and images, ensuring the integrity of the original information.

[0046] Step S2: Text Cleaning. The first text content undergoes in-depth text cleaning, removing redundant and meaningless formatting marks, such as headers, footers, and copyright information. Punctuation in both Chinese and English is also standardized to ensure consistency and readability. Crucially, the system retains and enhances paragraph structure markers, such as headings and bullet points. These markers are not only an important part of the document's visual structure, but also provide a clear logical framework for subsequent semantic analysis.

[0047] Step S3, generating structured data objects. Advanced semantic segmentation technology can be used to structure the cleaned text content based on the retained paragraph structure identifiers or the calculation of semantic similarity between different sentences in the second text content. Alternatively, the retained paragraph structure identifiers and the calculation of semantic similarity between different sentences in the second text content can be combined to structure the cleaned text content using semantic segmentation technology. This technology integrates a rule engine and a deep learning model. It can dynamically segment the text into multiple structured data objects with clear semantic boundaries while respecting the original logical structure of the document. Each object focuses on a specific topic or task.

[0048] Step S4: Generate semantic vectors. To further enhance the depth and accuracy of analysis, the system maps each structured data object into a multi-dimensional semantic vector. This process utilizes pre-trained models such as Sentence-BERT. The resulting 768-dimensional vector not only captures the literal meaning of the text but also implicitly captures the semantic relationships within the context, providing rich semantic information for subsequent intelligent analysis.

[0049] Step S5: Intelligent analysis and summary generation. Multiple semantic vectors are input into the target macromodel for in-depth analysis and intelligent summary generation. The target macromodel is derived from the general macromodel, fine-tuned with knowledge from the target task's technical domain. This model not only possesses broad language understanding and generation capabilities but is also optimized for domain-specific terminology and business logic, enabling a more accurate understanding of document content and extraction of key information.

[0050] The target large model utilizes advanced natural language processing technologies, such as hybrid document embedding technology to generate a document's implicit structural representation. Combined with named entity recognition technology, it can identify and extract key elements from documents, such as strategic goals, timelines, and budgets. This process generates concise and clear text summaries that cover the document's core points. It also intelligently analyzes task priorities, execution difficulty, and expected outcomes, providing reviewers and team members with a comprehensive and in-depth understanding of tasks, accelerating the document approval process, optimizing task allocation, and ensuring efficient task execution.

[0051] Through steps S1 to S5 above, efficient and accurate processing of target task files is achieved, providing strong support for improving the efficiency of document approval and task management in collaborative OA systems. This solution not only simplifies the document reading process but also enhances the intelligence of information processing through deep semantic understanding, opening up new paths for agile office work and intelligent decision-making in modern enterprises.

[0052] After intelligent analysis of the target task file, the target task file and text summary information can be sent to the approval node. Since the efficient operation of the document approval process is crucial to ensuring the smooth operation of the business process, and considering that the approval personnel may cause approval delays due to various reasons (such as workload, distraction, etc.), the embodiment of the present application also provides an optional intelligent approval reminder and status monitoring mechanism, which can be implemented in the following ways: periodically detecting the task approval status of the approval node, and generating an approval reminder prompt message if the approval node does not perform the approval operation within a preset number of consecutive cycles.

[0053] Specifically, after the target task file and its summary information are sent to the designated approval node, the intelligent approval status monitoring mechanism can be immediately activated. This mechanism is based on an event-driven architecture and can actively query the task approval status of the approval node within a preset time period (such as every hour or every day) without manual intervention. The query scope covers all pending tasks of the approval node, ensuring comprehensive status monitoring, not just limited to a single file. When the system detects that an approval node has not performed an approval operation on any pending document within a preset number of cycles (for example, two consecutive days), it is regarded as an approval delay, and the system immediately triggers the approval reminder mechanism. The generated approval reminder prompt information is detailed, including not only basic information such as the title, summary, and upload time of the pending document, but also the urgency of the approval, the importance assessment of the document, and an analysis of the possible impact of approval delays. It aims to stimulate the attention of the approvers through comprehensive information presentation and urge them to process the pending documents as soon as possible.

[0054] In an OA system, the approval process formally recognizes document content, ensuring compliance with corporate policies and standards. For project-related documents, approval means senior management has acknowledged the project's theoretical framework, objectives, and implementation plan. Only then does generating OKRs have practical guiding significance and implementation value. Documents that have not been approved or have failed approval may still be under discussion or revision, and the resulting OKRs may be based on immature or uncertain information, making them ineligible for OKR generation. Common project-related documents include project plans and marketing documents; common non-project-related documents include meeting minutes, financial statements, and training materials.

[0055] As an optional implementation method, when the target task is a project-type task and is approved, the pre-trained OKR generation model is used to analyze the text summary information to generate multiple subtasks corresponding to the target task. This can be achieved in the following way: multiple task goals are extracted from the text summary information using the OKR generation model; for each task goal, multiple key results corresponding to the task goal are deduced based on the thinking chain model, and a subtask is generated based on each key result, wherein each key result includes at least: subtask type and subtask indicator.

[0056] Specifically, after receiving the text summary information related to the target task file, the system first calls upon a pre-trained OKR generation model. This model, based on deep learning technology and trained on a large amount of corporate documents and task objective data, accurately identifies and extracts the core task objectives from the summary. For each identified task objective, the system employs a thought chain model, leveraging the model's deductive capabilities to generate the corresponding key results. This thought chain model allows the model to demonstrate its reasoning process, making the model's decision-making process more transparent and improving the accuracy and rationality of generated key results.

[0057] For each key result generated, it needs to be converted into a specific subtask. The generation of subtasks needs to consider factors such as task type, indicators, deadlines, required resources, and responsible persons to ensure the feasibility of the subtasks and consistency with the goals. Specific steps include: subtask type definition, subtask indicator setting, and subtask parameter configuration. Subtask type definition: Based on the attributes of the key results, the system automatically matches the subtask type, such as market research, product development, internal training, etc.; subtask indicator setting: Set specific measurable indicators for each subtask. These indicators should be consistent with the quantitative indicators in the key results, such as "Complete the competitive product analysis report within two weeks", "Increase user satisfaction of product feature X to 85%", etc.; subtask parameter configuration: The system automatically generates other necessary parameters for the subtask, including information such as expected completion time, required resources, task priority, and responsible person.

[0058] In order to ensure the feasibility, innovation and consistency with historical practice of the key results, the embodiment of the present application can correct the key results in the following ways before generating multiple subtasks based on multiple key results: call a pre-trained feasibility evaluation model to analyze each key result obtained by deduction, determine the feasibility score of each key result, and correct the key results whose feasibility score is lower than the preset score threshold; and / or, based on the retrieval enhancement generation algorithm, retrieve each historical key result corresponding to the historical task matching the target task from the historical task database, and correct each key result obtained by deduction based on each historical key result; and / or, correct each key result obtained by deduction based on the knowledge graph of the technical field to which the target task belongs.

[0059] This process leverages a pre-trained feasibility assessment model, a historical task database, and a knowledge graph from the technical field to enable comprehensive analysis and intelligent revision of key achievements. These revisions can be made using any of the three aforementioned methods, individually or in any combination.

[0060] Specifically, when using a pre-trained feasibility assessment model for correction, it can be achieved in the following way: each key result is input into the feasibility assessment model, and the model will identify potential risks and challenges based on its description and contextual information; the model generates a feasibility score between 0 and 1 based on the identified risk level, where 0 means completely infeasible and 1 means highly feasible; for key results with a feasibility score below the preset threshold, the system will automatically trigger the correction process, and by calling the expert system or industry knowledge base, it will supplement the missing conditions for the key results or adjust the indicator values to make them more in line with reality.

[0061] Specifically, when using the historical task database for corrections, this can be achieved by leveraging the semantic matching capabilities of the large model to retrieve task examples from the historical task database that are similar or related to the current target task. By comparing the differences between historical key achievements and current key achievements, the system can identify any unrealistic or overlooked details in the current key achievement. Based on the actual performance and feedback from historical key achievements, the system appropriately adjusts the indicators for the current key achievement, ensuring that it is both innovative and compliant with the company's actual execution capabilities.

[0062] Specifically, when using the knowledge graph in the technical field for correction, this can be achieved in the following way: the system uses natural language processing technology to semantically link key results with related terms in the knowledge graph to obtain more context and detailed information. Through the knowledge graph, the system can verify whether the indicators of key results meet industry standards and whether there are technical bottlenecks or market uncertainties. Based on the information provided by the knowledge graph, the system will revise the settings of key results, such as adjusting the target value to match the latest technological development trends, or refining the task description to cover a wider industry perspective, to ensure the foresight and accuracy of the key results.

[0063] By revising the generated key results, we can obtain key results that meet the SMART principles, ensuring that the generated key results are specific, the indicators are quantifiable, achievable, linked to the extracted core objectives, and time-bound. The revised key results can also be used as optimized sample data and fed back into the OKR generation model to enable iterative training of the OKR generation model. After each training batch, low-rank adaptive technology is used to fine-tune the model parameters, thereby updating and optimizing the OKR generation model.

[0064] According to the task type, indicators and person in charge information in the key results, the generated subtasks will be automatically assigned to the corresponding team members. An intelligent allocation algorithm based on historical data and member capabilities can be built into the collaborative OA system to ensure the fairness and efficiency of task allocation.

[0065] Throughout the subtask execution process, it is crucial to ensure that the task execution status matches the corresponding subtask indicators and to promptly detect and address execution deviations. In order to achieve dynamic monitoring and effective adjustment of task execution, the present application embodiment also provides an optional implementation method for monitoring the execution status of nodes and providing early warning prompts when the status is abnormal.

[0066] As an optional implementation method, periodically detecting the task execution status of each execution node and generating a warning prompt message when the task execution status is abnormal can be achieved in the following way: for each execution node, periodically detecting the task execution status of the execution node, and comparing the task execution status with the subtask indicators of the subtask executed by the execution node; if after a preset number of consecutive cycles, the task execution status and the subtask indicators of the subtask executed by the execution node are always mismatched, it is determined that the task execution status is abnormal, and a warning prompt message is generated, wherein the warning prompt message is used to prompt that the task execution status of the execution node is abnormal, and it is necessary to adjust the subtask executed by the execution node, or reassign the execution node to the subtask.

[0067] Specifically, the system automatically monitors the task execution status of each execution node by setting up regular detection tasks. The detection cycle can be adjusted according to the urgency and complexity of the task. For example, for high-priority or critical path subtasks, detection can be set once an hour or a day, while for non-critical subtasks in long-term projects, detection can be set once a week or every two weeks. Execution status detection includes: data collection, status analysis, and status comparison. Among them, the system automatically collects task-related data from multiple channels such as activity records of execution nodes, application logs, external system interfaces, etc., and parses the collected text data through natural language processing technology to understand the current progress of the task, status description, and obstacles encountered, etc. to achieve status analysis. The parsed task execution status is compared with the subtask indicators to determine whether the expected milestones have been achieved or deviated from the plan.

[0068] If, after a preset number of consecutive cycles, the task execution status consistently mismatches the subtask metrics being executed by the execution node, the system will determine that the task execution status is abnormal and generate a warning message. The preset number of cycles can be set based on the nature of the task and the company's response time requirements; typically, three to five cycles are suitable to ensure timely and accurate warnings. Using a large model, the warning message is intelligently analyzed and recommendations are generated for adjusting subtasks or reallocating execution nodes.

[0069] During specific execution, the cause is analyzed based on warning information and historical information. If the risk of task execution is caused by changes in the external environment, suggestions for adjusting subtask indicators can be considered. If it is due to insufficient resources or manpower, suggestions for reallocating task execution nodes can be given based on the task priority and the built-in execution node matching algorithm.

[0070] Through the above steps, the target task file is first subjected to text extraction and semantic analysis to obtain text summary information of the target task file, which can help approvers quickly understand the file content and help make approval decisions; and when the target task is a project-type task and is approved, the pre-trained OKR generation model is used to analyze the text summary information to obtain key results that meet the SMART principle and generate the final subtasks, laying the foundation for the subsequent distribution and execution of tasks and the achievement of task goals; thereafter, the execution status of subtasks is periodically tracked and early warning prompts are given to timely feedback the task execution status to managers, facilitating task tracking and making relevant adjustments in a timely manner, thereby reducing the risk of task execution.

[0071] Example 2

[0072] According to an embodiment of the present application, a collaborative task management system is also provided for implementing the collaborative task management method in Example 1, such as Figure 2 As shown, the collaborative task management system includes at least: an acquisition module 21, an analysis module 22, a generation module 23 and a monitoring module 24, wherein:

[0073] Acquisition module 21, acquires the target task file corresponding to the target task;

[0074] The analysis module 22 performs text extraction and semantic analysis on the target task file to obtain text summary information of the target task file, and sends the target task file and text summary information to the approval node;

[0075] Generation module 23, when the target task is a project-type task and has been approved, uses the pre-trained objectives and key results OKR generation model to analyze the text summary information, generates multiple subtasks corresponding to the target task, and distributes each subtask to the corresponding execution node;

[0076] The monitoring module 24 periodically detects the task execution status of each execution node and generates an early warning message when the task execution status is abnormal.

[0077] The following describes the functions of each module of the collaborative task management system in combination with the specific implementation process.

[0078] When the acquisition module receives a task file uploaded by the user to the OA system, it must first verify the task file to obtain the target task file corresponding to the target task. This can be achieved in the following ways: receive the task file uploaded by the target object, and perform file signature verification and Multipurpose Internet Mail Extension type verification on the task file to determine the file format of the task file; if the file format meets the preset format requirements, determine the task file as the target task file; if the file format does not meet the preset format requirements, generate an exception prompt message, wherein the exception prompt message is used to prompt that the uploaded task file format is abnormal and the task file needs to be uploaded again.

[0079] For the target task files obtained, text extraction and semantic analysis can be performed on the content of the target task files to generate text summary information of the target task files. Through intelligent processing of the target task files, key information can be quickly extracted from massive documents to provide accurate data support for subsequent file approval and task management.

[0080] As an optional implementation, the analysis module performs text extraction and semantic analysis on the target task file to obtain text summary information of the target task file, which can be achieved in the following way: calling a file parsing tool corresponding to the file format of the target task file to extract the first text content of the target task file; performing text cleaning on the first text content to obtain the second text content, wherein the second text content retains the paragraph structure identifier; performing semantic segmentation on the second text content based on the paragraph structure identifier and / or the semantic similarity between different sentences in the second text content to obtain multiple structured data objects; mapping each structured data object into a multi-dimensional semantic vector; using the target big model to analyze the multiple semantic vectors to obtain text summary information of the target task file, wherein the target big model is obtained by fine-tuning the general big model using the knowledge of the technical field to which the target task belongs.

[0081] After the analysis module performs intelligent analysis on the target task file, the target task file and text summary information can be sent to the approval node. Since the efficient operation of the document approval process is crucial to ensuring the smoothness of the business process, and considering that the approval personnel may cause approval delays due to various reasons (such as workload, distraction, etc.), as an optional implementation method, the embodiment of the present application also provides an intelligent approval reminder and status monitoring mechanism. The intelligent approval reminder and status monitoring mechanism can be implemented by scheduling the monitoring module in the following ways: periodically detecting the task approval status of the approval node, and generating an approval reminder prompt message if the approval node does not perform the approval operation within a preset number of consecutive cycles.

[0082] As an optional implementation method, when the target task is a project-type task and is approved, the generation module uses the pre-trained OKR generation model to analyze the text summary information to generate multiple subtasks corresponding to the target task. This can be achieved in the following way: use the OKR generation model to extract multiple task goals from the text summary information; for each task goal, deduce multiple key results corresponding to the task goal based on the thinking chain model, and generate a subtask based on each key result, wherein each key result includes at least: subtask type and subtask indicator.

[0083] As an optional implementation method, in order to ensure the feasibility, innovation and consistency with historical practice of the key results, the embodiment of the present application can correct the key results in the following ways before generating multiple subtasks based on multiple key results: call the pre-trained feasibility evaluation model to analyze each key result obtained by deduction, determine the feasibility score of each key result, and correct the key results whose feasibility score is lower than the preset score threshold; and / or, based on the retrieval enhancement generation algorithm, retrieve each historical key result corresponding to the historical task matching the target task from the historical task database, and correct each key result obtained by deduction based on each historical key result; and / or, correct each key result obtained by deduction based on the knowledge graph of the technical field to which the target task belongs.

[0084] Throughout the subtask execution process, it is crucial to ensure that the task execution status matches the corresponding subtask indicators and to promptly detect and address execution deviations. In order to achieve dynamic monitoring and effective adjustment of task execution, the present application embodiment also provides an optional implementation method for monitoring the execution status of nodes and providing early warning prompts when the status is abnormal.

[0085] As an optional implementation method, the monitoring module periodically detects the task execution status of each execution node and generates an early warning prompt message when the task execution status is abnormal. This can be achieved in the following way: for each execution node, the task execution status of the execution node is periodically detected, and the task execution status is compared with the subtask indicators of the subtask executed by the execution node; if after a preset number of consecutive cycles, the task execution status and the subtask indicators of the subtask executed by the execution node are always mismatched, it is determined that the task execution status is abnormal, and an early warning prompt message is generated, wherein the early warning prompt message is used to prompt that the task execution status of the execution node is abnormal, and it is necessary to adjust the subtask executed by the execution node, or reassign the execution node to the subtask.

[0086] It should be noted that each module in the collaborative task management system in the embodiment of the present application corresponds one-to-one to each implementation step of the collaborative task management method in Example 1. Since a detailed description has been given in Example 1, some details not reflected in this embodiment can be referred to Example 1 and will not be elaborated here.

[0087] Example 3

[0088] According to an embodiment of the present application, a computer program product is further provided, which includes a computer program, wherein when the computer program is executed by a processor, the collaborative task management method in Example 1 is implemented.

[0089] According to an embodiment of the present application, a non-volatile storage medium is also provided, which includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the collaborative task management method in Example 1 by running the computer program.

[0090] According to an embodiment of the present application, a processor is further provided, which is used to run a computer program, wherein the collaborative task management method in Example 1 is executed when the computer program is running.

[0091] According to an embodiment of the present application, an electronic device is also provided, which includes: a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to execute the collaborative task management method in Example 1 through the computer program.

[0092] Specifically, when the computer program is running, the following steps are executed: obtaining the target task file corresponding to the target task; performing text extraction and semantic analysis on the target task file to obtain text summary information of the target task file, and sending the target task file and text summary information to the approval node; when the target task is a project-type task and is approved, using the pre-trained objectives and key results OKR generation model to analyze the text summary information, generate multiple subtasks corresponding to the target task, and distribute each subtask to the corresponding execution node; periodically detecting the task execution status of each execution node, and generating early warning prompt information when the task execution status is abnormal.

[0093] As an optional implementation, the electronic device may be in the form of a mobile terminal, a computer terminal or a similar computing device. Figure 3 FIG1 shows a hardware structure block diagram of an electronic device for implementing a collaborative task management method. Figure 3As shown, the electronic device 30 may include one or more (illustrated as 302a, 302b, ..., 302n in the figure) processors 302 (the processor 302 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 304 for storing data, and a transmission device 306 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It will be understood by those skilled in the art that Figure 3 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 3 More or fewer components than shown, or with Figure 3 Different configurations shown.

[0094] It should be noted that the one or more processors 302 and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuitry may be a single independent processing module, or may be incorporated in whole or in part into any of the other components of the electronic device 30. As described in the embodiments of the present application, the data processing circuitry serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0095] The memory 304 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the collaborative task management method in the embodiment of the present application. The processor 302 executes various functional applications and data processing by running the software programs and modules stored in the memory 304, that is, implementing the vulnerability detection method of the above-mentioned application. The memory 304 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 304 may further include a memory remotely located relative to the processor 302, and these remote memories may be connected to the electronic device 30 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0096] The transmission device 306 is used to receive or send data via a network. Specific examples of the aforementioned network may include a wireless network provided by the communications provider of the electronic device 30. In one embodiment, the transmission device 306 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In one embodiment, the transmission device 306 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0097] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the electronic device 30 .

[0098] The serial numbers of the above embodiments are for description only and do not represent the advantages or disadvantages of the embodiments.

[0099] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0100] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0101] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected to achieve the purpose of the present embodiment according to actual needs.

[0102] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0103] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program code.

[0104] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A collaborative task management method, characterized in that: include: Get the target task file corresponding to the target task; Performing text extraction and semantic analysis on the target task file to obtain text summary information of the target task file, and sending the target task file and the text summary information to the approval node; If the target task is a project-type task and has been approved, the text summary information is analyzed using the pre-trained objectives and key results OKR generation model to generate multiple subtasks corresponding to the target task, and each subtask is distributed to the corresponding execution node; The task execution status of each execution node is periodically detected, and an early warning prompt message is generated when the task execution status is abnormal.

2. The method according to claim 1, characterized in that Get the target task file corresponding to the target task, including: receiving a task file uploaded by a target object, and performing a file signature check and a Multipurpose Internet Mail Extensions (MIMEX) type check on the task file to determine a file format of the task file; If the file format meets the preset format requirements, determining the task file as a target task file; When the file format does not meet the preset format requirements, an abnormal prompt message is generated, wherein the abnormal prompt message is used to prompt that the uploaded task file format is abnormal and the task file needs to be uploaded again.

3. The method according to claim 1, characterized in that Perform text extraction and semantic analysis on the target task file to obtain text summary information of the target task file, including: Invoking a file parsing tool corresponding to the file format of the target task file to extract the first text content of the target task file; Performing text cleaning on the first text content to obtain second text content, wherein the second text content retains paragraph structure identifiers; Semantically segmenting the second text content based on the paragraph structure identifier and / or semantic similarity between different sentences in the second text content to obtain a plurality of structured data objects; Mapping each of the structured data objects into a multi-dimensional semantic vector; The target large model is used to analyze the multiple semantic vectors to obtain text summary information of the target task file, wherein the target large model is obtained by fine-tuning the general large model using knowledge of the technical field to which the target task belongs.

4. The method according to claim 1, wherein After sending the target task file and the text summary information to the approval node, the method further includes: The task approval status of the approval node is periodically detected. If the approval node fails to perform the approval operation within a preset number of consecutive cycles, an approval reminder message is generated.

5. The method according to claim 1, wherein The text summary information is analyzed using a pre-trained OKR generation model to generate multiple subtasks corresponding to the target task, including: Extracting multiple task objectives from the text summary information using the OKR generation model; For each task objective, multiple key results corresponding to the task objective are deduced based on the thought chain model, and a subtask is generated based on each key result, wherein each key result includes at least: a subtask type and a subtask indicator.

6. The method according to claim 5, characterized in that Before generating a plurality of subtasks based on the plurality of key achievements, the method further includes: Calling the pre-trained feasibility assessment model to analyze each key result obtained from the deduction, determine the feasibility score of each key result, and revise the key results whose feasibility score is lower than the preset score threshold; and / or, Retrieving each historical key achievement corresponding to the historical task matching the target task from the historical task database based on the retrieval enhancement generation algorithm, and revising each key achievement obtained by deduction according to each historical key achievement; and / or, The key results obtained by deduction are revised based on the knowledge graph of the technical field to which the target task belongs.

7. The method according to claim 5, characterized in that Periodically detecting the task execution status of each execution node and generating early warning information when the task execution status is abnormal, including: For each execution node, periodically detecting the task execution status of the execution node, and comparing the task execution status with the subtask indicator of the subtask executed by the execution node; If after a preset number of consecutive cycles, the task execution status still does not match the subtask indicator of the subtask executed by the execution node, it is determined that the task execution status is abnormal, and an early warning prompt information is generated, wherein the early warning prompt information is used to prompt that the task execution status of the execution node is abnormal, and it is necessary to adjust the subtask executed by the execution node, or re-assign the execution node to the subtask.

8. A collaborative task management system, characterized in that: include: The acquisition module is used to obtain the target task file corresponding to the target task; An analysis module is used to perform text extraction and semantic analysis on the target task file to obtain text summary information of the target task file, and send the target task file and the text summary information to the approval node; A generation module is configured to analyze the text summary information using a pre-trained OKR generation model when the target task is a project-type task and has been approved, generate multiple subtasks corresponding to the target task, and distribute each subtask to a corresponding execution node; The monitoring module is used to periodically detect the task execution status of each execution node and generate early warning information when the task execution status is abnormal.

9. A computer program product, characterized in that include: A computer program, wherein when the computer program is executed by a processor, the collaborative task management method according to any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the collaborative task management method according to any one of claims 1 to 7 through the computer program.

Images