Project full-process data tracking and processing methods and systems

Abstract

This invention relates to the field of artificial intelligence technology, specifically providing a method and system for tracking and processing data throughout the entire project process. The method includes: acquiring project planning data; extracting task information and relationships between tasks from the project planning data using a pre-trained language model; constructing a knowledge graph based on the task information and relationships between tasks; acquiring project execution data from multiple data sources; identifying task execution information from the project execution data; predicting task progress based on a data sequence composed of all task execution information; adding state attributes to corresponding nodes in the knowledge graph based on the task progress; analyzing abnormal situations in task execution status based on the updated state attributes of the knowledge graph; analyzing abnormal factors based on project execution data; and generating processing solutions for the abnormal factors using a pre-built knowledge base. This invention improves the precision and intelligence of project management.

Description

Technical Field

[0001] This invention belongs to the field of artificial intelligence technology, specifically relating to a method and system for tracking and processing data throughout the entire project process. Background Technology

[0002] With the increasing complexity of project management, especially in large-scale engineering projects, software development, scientific research, and government public projects, the need for real-time tracking of project progress and accurate problem identification has become increasingly urgent. Traditional project management tools mainly rely on manual data entry and static report analysis. This model is inadequate when faced with the frequently changing requirements and diverse data inputs in complex projects. They struggle to effectively capture and respond to subtle changes in project dynamics, and are even less likely to keenly identify potential risk factors or signs of deviation in the early stages of a project.

[0003] Especially in terms of timeline management and problem summarization, traditional methods reveal deficiencies in understanding the semantics of massive amounts of project documents and a lack of in-depth correlation analysis capabilities. This not only makes it difficult to identify and correct deviations in projects in a timely manner, but also subtly slows down the overall project progress and reduces the final quality. Therefore, exploring more efficient and intelligent project management methods has become a critical issue that urgently needs to be addressed in the field of project management. Summary of the Invention

[0004] In view of the above-mentioned shortcomings of the prior art, the present invention provides a method and system for tracking and processing data throughout the entire project process to solve the above-mentioned technical problems.

[0005] In a first aspect, the present invention provides a method for tracking and processing data throughout the entire project process, including:

[0006] Acquire project planning data, use a pre-trained language model to extract task information and relationships between tasks from the project planning data, the task information including planning time, and construct a knowledge graph based on the task information and relationships between tasks;

[0007] Project execution data is obtained from multiple data sources, and task execution information is identified from the project execution data;

[0008] The task progress is predicted based on a data sequence composed of all task execution information. The task progress is matched with the corresponding planning time, and state attributes are added to the corresponding nodes of the knowledge graph based on the matching results.

[0009] Based on the knowledge graph analysis of the updated state attributes, abnormal situations in the task execution state are identified. Based on the project execution data, abnormal factors are analyzed, and a processing solution is generated for the abnormal factors using a pre-built knowledge base.

[0010] In one optional implementation, project planning data is acquired, and a pre-trained language model is used to extract task information and relationships between tasks from the project planning data. The task information includes planning time. A knowledge graph is then constructed based on the task information and the relationships between tasks, including:

[0011] Collect project planning data from project management platforms, document storage email systems, and project logs;

[0012] The project planning data is preprocessed, including data cleaning and data formatting.

[0013] The preprocessed project planning data is input into the pre-trained language model to obtain task information and task relationships. The task information includes resource information and planning time, and the task relationships include the dependency relationship between tasks, the resource allocation relationship, and the time sequence relationship.

[0014] By using tasks, resources, and time as nodes and task relationships as edges, a knowledge graph is constructed using a graph database.

[0015] Importance labels are generated for tasks within the knowledge graph.

[0016] Set up a query interface for the knowledge graph.

[0017] In one optional implementation, project execution data is obtained from multiple data sources, and task execution information is identified from the project execution data, including:

[0018] By using keyword technology to identify filenames, project execution data is collected from project management platforms, document storage email systems, and project logs.

[0019] The pre-trained language model is used to extract task execution information from the project execution data. The task execution information includes executed information and approval information. The executed information includes executed task information and execution time, and the approval information includes pending task information and execution time.

[0020] In an optional implementation, the method further includes:

[0021] Based on the information of the task to be executed, query the corresponding target task information and the status of the preceding tasks that the target task depends on from the knowledge graph;

[0022] Extract the person in charge information and the planned time from the target task information;

[0023] The approval information is matched with the person in charge information and the planned time in the target task information. If the match is not complete, an error message is generated; if the match is complete, a message indicating that the approval information is normal is generated.

[0024] If the approval information is confirmed to be normal, and all the prerequisite tasks that the target task depends on are in the completed state, then the approval information is determined to meet the approval conditions and is sent to the review terminal; if any of the prerequisite tasks that the target task depends on are incomplete, then the approval information is determined to not meet the approval conditions and a prompt message indicating that the approval has failed is generated.

[0025] In one optional implementation, the task progress is predicted based on a data sequence composed of all task execution information, the task progress is matched with the corresponding planning time, and state attributes are added to the corresponding nodes of the knowledge graph based on the matching results, including:

[0026] Save the execution information of the same task to the same list;

[0027] The task is broken down into multiple execution phases, and execution time is allocated to each phase based on the task's planned time.

[0028] Based on the execution time and execution order of the execution phase, set corresponding progress values ​​for the execution phase;

[0029] Get the completed execution stage corresponding to the executed task information, and retrieve the progress value corresponding to the executed information based on the completed execution stage and the correspondence between the execution stage and the progress value.

[0030] The executed information in the list is converted into a data group consisting of progress value and execution time;

[0031] The data groups are arranged into a time data sequence according to their execution time.

[0032] The time data sequence and the current system time are input into a pre-trained long short-term memory neural network model to predict the task completion time.

[0033] Calculate the difference between the completion time and the planned completion time of the task, and generate status attributes based on the difference. The status attributes include early completion, normal progress, and overdue completion.

[0034] Add the state attributes to the corresponding task nodes in the knowledge graph.

[0035] In one optional implementation, the system analyzes anomalies in task execution status based on the updated state attributes of the knowledge graph, analyzes anomaly factors based on project execution data, and generates processing solutions for the anomalies using a pre-built knowledge base, including:

[0036] Set anomaly detection indicators, including the proportion of overdue tasks, task progress deviation, resource utilization rate, and abnormal task dependencies;

[0037] Traverse the task nodes in the knowledge graph and obtain the state attributes of each task node;

[0038] Calculate the abnormal indicator values ​​for each task;

[0039] Tasks whose abnormal indicator values ​​exceed the threshold are marked as abnormal tasks and specially marked in the knowledge graph;

[0040] By analyzing the task relationships, resource allocation, execution time, and task execution information of abnormal tasks, and using the thinking chain reasoning technique to locate the abnormal causes of the task abnormality based on the analysis results;

[0041] Based on the causes of anomalies in abnormal tasks, a handling plan is generated using a pre-built knowledge base.

[0042] In one optional implementation, a handling plan is generated based on the cause of the anomaly in the abnormal task using a pre-built knowledge base, including:

[0043] The cause of the abnormal task is input into the knowledge base to obtain multiple historical processing solutions;

[0044] Using the chain of reasoning technique, we can conduct causal analysis on multiple historical solutions and deduce multiple implementation steps and effects from the problem-solution-result chain.

[0045] Based on the effectiveness of the implementation steps, multiple implementation schemes with the best results are generated, and each implementation scheme includes multiple implementation steps;

[0046] A feasibility assessment is conducted on the implementation plan, and the implementation plan that passes the feasibility assessment is output as a treatment plan.

[0047] In an optional implementation, the method further includes:

[0048] Collect the corresponding handling solutions for historical abnormal tasks from multiple projects, and save the historical data of multiple projects into multiple corresponding datasets;

[0049] The knowledge base is trained using a federated learning mechanism.

[0050] Secondly, the present invention provides a project end-to-end data tracking and processing system, comprising:

[0051] The knowledge graph construction module is used to acquire project planning data, extract task information and relationships between tasks from the project planning data using a pre-trained language model, the task information including planning time, and construct a knowledge graph based on the task information and relationships between tasks.

[0052] The execution monitoring module is used to obtain project execution data from multiple data sources and identify task execution information from the project execution data;

[0053] The status setting module is used to predict the task progress based on a data sequence composed of all task execution information, match the task progress with the corresponding planning time, and add status attributes to the corresponding nodes of the knowledge graph based on the matching results.

[0054] The exception handling module is used to analyze the abnormal situation of task execution status based on the knowledge graph after updating the status attributes, analyze the abnormal factors based on the project execution data, and generate a handling solution for the abnormal factors using a pre-built knowledge base.

[0055] The beneficial effects of this invention lie in the fact that the project end-to-end data tracking and processing method and system provided by this invention, through the integration of technologies such as pre-trained language models, knowledge graphs, progress prediction, and anomaly analysis, achieve intelligent processing of project planning and execution data. The system extracts tasks and relationships from multi-source data, constructs a knowledge graph to enhance planning visualization and decision support; monitors execution status in real time, predicts progress and matches plans, and dynamically updates the status of graph nodes; automatically detects anomalies and generates handling solutions, optimizes approval processes, and improves response efficiency. This solution significantly enhances the refinement, intelligence, and overall efficiency of project management, and has achieved good economic and social benefits in practical applications. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is a schematic flowchart of a method according to an embodiment of the present invention.

[0058] Figure 2 This is a schematic flowchart illustrating a method for constructing a knowledge graph based on project planning data, according to an embodiment of the present invention.

[0059] Figure 3 This is a schematic flowchart illustrating the method for extracting task execution information according to an embodiment of the present invention.

[0060] Figure 4 This is a schematic flowchart illustrating a method for setting state attributes for a task node according to an embodiment of the present invention.

[0061] Figure 5 This is a schematic flowchart illustrating the abnormal task screening and processing method according to an embodiment of the present invention.

[0062] Figure 6 This is a schematic block diagram of a system according to an embodiment of the present invention. Detailed Implementation

[0063] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0064] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0065] The key terms used in this invention will be explained below.

[0066] Pre-trained language models (PLMs) are an important tool in the field of natural language processing (NLP). They learn rich language knowledge and patterns by pre-training on large-scale text data, and can then be applied to various downstream tasks, such as text classification, named entity recognition, sentiment analysis, question answering systems, machine translation, etc.

[0067] Main features:

[0068] Large-scale data training: Pre-trained language models are typically trained using hundreds of millions of text data, covering a wide range of fields and topics, enabling the model to learn rich language knowledge and contextual information.

[0069] Unsupervised learning: The pre-training process usually adopts an unsupervised learning approach, that is, the model does not need manually labeled, but automatically learns the statistical regularities and patterns of language from text data.

[0070] Transfer learning: Through pre-training, the knowledge learned by the model can be transferred to different downstream tasks. Downstream tasks only need to use a small amount of labeled data to fine-tune the pre-trained model to achieve good performance.

[0071] Common models include:

[0072] BERT (Bidirectional Encoder Representations from Transformers): BERT is a pre-trained language model based on the Transformer architecture that learns text representations through bidirectional encoding. BERT's pre-training tasks include Masked Language Model (MLM) and NextSentence Prediction (NSP).

[0073] GPT (Generative Pre-trained Transformer): The GPT series of models are autoregressive pre-trained language models based on the Transformer architecture. They learn text representations in a generative manner and can generate coherent text sequences. The pre-training task of GPT is language modeling, that is, predicting the next word based on the preceding text.

[0074] T5 (Text-to-Text Transfer Transformer): T5 is a pre-trained language model that unifies various NLP tasks into a text-to-text format. It uses the same model architecture and training process to handle different tasks by concatenating the task description and input text as input to the model.

[0075] RoBERTa (Robustly optimized BERT pretraining approach): RoBERTa is an improvement and optimization of BERT. It improves the performance of the model by increasing the amount of training data, the number of training steps, the batch size, and removing the NSP task.

[0076] ERNIE (Enhanced Representation through kNowledge IntEgration): ERNIE is a pre-trained language model developed by Baidu. It emphasizes knowledge enhancement by introducing semantic information such as entities and relationships to enhance the model's representation capabilities.

[0077] By using pre-trained language models to process project-related data, we can accurately identify task information in the project and clarify the relationships between tasks.

[0078] LSTM (Long Short-Term Memory) is a special type of recurrent neural network (RNN) architecture designed to address the vanishing and exploding gradient problems inherent in traditional RNNs when processing long sequences of data. Proposed by Hochreiter and Schmidhuber in 1997, the LSTM model's core lies in the introduction of three special "gate" structures: an input gate, a forget gate, and an output gate, along with a memory cell. These structures work together to enable LSTM to maintain and update memories over longer timeframes, effectively capturing and preserving long-term dependencies.

[0079] Specifically, the input gate controls how much new information is stored in the cell state; the forget gate controls how much information is retained in the current cell state; and the output gate determines how much information is output from the cell state. This mechanism allows LSTM to flexibly decide which information should be retained, updated, or forgotten, thereby avoiding the gradient vanishing problem and preserving information over long time sequences while eliminating irrelevant or noisy information.

[0080] Advantages of LSTM in predicting task progress:

[0081] Long-term dependency capture: LSTM, through its unique memory units and gating mechanism, can effectively capture and retain long-term dependencies. This is particularly important for predicting task progress, as task progress is often influenced by multiple past time steps.

[0082] Information filtering capability: The gating mechanism allows LSTM to flexibly decide which information should be retained, updated, or forgotten. This capability enables LSTM to filter out irrelevant or noisy information when processing complex task progress data, thereby improving prediction accuracy.

[0083] Nonlinear Modeling: LSTM networks have the ability to learn nonlinear functions, which is crucial for most real-world task progress data. Because task progress often exhibits complex nonlinear patterns and trends, LSTMs can capture these patterns and make accurate predictions.

[0084] End-to-end learning: LSTMs can learn directly from raw data without manual feature engineering. This simplifies the model building process and improves the automation and efficiency of prediction. For task progress prediction, this means that progress patterns can be learned directly from historical task data without tedious feature extraction and selection.

[0085] Flexibility and scalability: LSTM can be easily combined with other network architectures (such as Convolutional Neural Networks, attention mechanisms, etc.) or its expressive power can be enhanced by stacking multiple layers. This flexibility and scalability enable LSTM to adapt to more complex and varied task progress prediction tasks.

[0086] In summary, LSTM models have significant advantages in predicting task progress. Their long-term reliance on data capture, information filtering, nonlinear modeling, end-to-end learning, flexibility, and scalability all provide strong support for task progress prediction.

[0087] The project end-to-end data tracking and processing method provided in this embodiment of the invention is executed by a computer device, and correspondingly, the project end-to-end data tracking and processing system runs on the computer device.

[0088] Figure 1 This is a schematic flowchart illustrating a method according to an embodiment of the present invention. Wherein, Figure 1 The implementing entity can be a project-wide data tracking and processing system. Depending on different needs, the order of the steps in this flowchart can be changed, and some can be omitted.

[0089] like Figure 1 As shown, the method includes:

[0090] Step 110. Obtain project planning data, use a pre-trained language model to extract task information and relationships between tasks from the project planning data, the task information including planning time, and construct a knowledge graph based on the task information and relationships between tasks.

[0091] Obtaining Project Planning Data: First, collect project planning data from project management tools, documents, or databases. This data should include an overall overview of the project, the Work Breakdown Structure (WBS), task descriptions, planned timelines (such as start date, end date, milestones, etc.), and dependencies between tasks.

[0092] Extracting task information and relationships using pre-trained language models: Using pre-trained language models (such as BERT, GPT, etc.), natural language processing is performed on project planning data to extract key information such as task names, descriptions, and planning times, as well as dependencies such as the order and parallel relationships between tasks.

[0093] Building a knowledge graph: Based on the extracted task information and the relationships between tasks, a knowledge graph is constructed. In the knowledge graph, each task is represented as a node, and the relationships between tasks (such as dependencies) are represented as edges between nodes. Additionally, attributes such as planning time are added to task nodes.

[0094] Step 120. Obtain project execution data from multiple data sources, and identify task execution information from the project execution data.

[0095] Obtain project execution data from multiple data sources: Collect project execution data from various data sources such as project management tools, monitoring systems, and log records. This data should include task execution status, actual start and end times, resource usage, and progress percentage.

[0096] Identify task execution information: Process and analyze project execution data to identify the execution information of each task, including actual execution time, resource input, progress status, etc.

[0097] Step 130. Predict the task progress based on the data sequence composed of all task execution information of the task, match the task progress with the corresponding planning time, and add state attributes to the corresponding nodes of the knowledge graph based on the matching results.

[0098] Predicting task progress based on data sequences: For each task, using a data sequence (such as time series data) composed of all its task execution information, LSTM or other time series prediction models are used to predict the future progress of the task.

[0099] Matching planned time with projected progress: Matching projected task progress with corresponding planned time to assess whether tasks are proceeding as planned.

[0100] Update knowledge graph state attributes: Based on the matching results, update the state attributes of the corresponding task nodes in the knowledge graph. For example, if the task progress is behind schedule, set the state attribute to "behind"; if the task is completed ahead of schedule, set it to "completed ahead of schedule".

[0101] Step 140. Analyze the abnormal situations of task execution status based on the knowledge graph after updating the status attributes, analyze the abnormal factors based on the project execution data, and generate a processing solution for the abnormal factors using the pre-built knowledge base.

[0102] Knowledge graph-based analysis of task execution status anomalies: Based on the updated knowledge graph of status attributes, analyze abnormal situations in task execution status. For example, identify which tasks are lagging behind or lack sufficient resources.

[0103] Analysis of anomalies based on project execution data: Further analysis of project execution data to identify specific factors causing abnormal task execution status. This may include uneven resource allocation, skill shortages, external interference, etc.

[0104] Generate solutions using a knowledge base: Based on the anomalies, generate solutions using a pre-built knowledge base (containing common problems and solutions). These solutions may include adjusting resource allocation, providing training support, and optimizing workflows.

[0105] In one embodiment of the present invention, based on step 110, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner, such as... Figure 2 As shown.

[0106] Project planning data is collected from multiple channels, including project management platforms, document storage systems, email systems, and project logs. This data contains key information such as the overall project plan, task division for each stage, resource allocation, and time nodes, forming the basis for subsequent analysis and processing. Before entering the pre-trained language model, the collected project planning data undergoes a series of preprocessing operations. These operations include data cleaning to remove duplicate, erroneous, or irrelevant information; and data formatting to unify the data structure for easier subsequent processing. The quality of the pre-processed data is significantly improved, laying a solid foundation for subsequent information extraction. Next, this invention utilizes the pre-trained language model for deep analysis of the project planning data. This model possesses powerful natural language understanding capabilities, accurately identifying and extracting task information (including resource information and planning time) and relationships between tasks (such as dependencies, resource allocation relationships, and time sequence relationships). This information is a key element in constructing a knowledge graph. After extracting task information and task relationships, this invention uses graph database technology to construct a knowledge graph. In this graph, tasks, resources, and time are abstracted as nodes, while relationships between tasks are represented as edges. This graph-structured data representation makes the various tasks in a project and their relationships readily apparent, providing strong support for subsequent task execution status analysis and anomaly handling. Furthermore, this invention generates importance tags for task nodes in the knowledge graph to quickly identify critical tasks. Simultaneously, by setting up a query interface, users can easily retrieve and query information in the graph, meeting diverse needs in the project management process.

[0107] In a specific example, the following specific process is included:

[0108] Step 1101. Obtain project planning data

[0109] Data source:

[0110] Export project planning data from project management software (such as Jira or Trello). This data typically includes task names, descriptions, resource allocations, start and end dates, etc.

[0111] Alternatively, extract data from project documents (such as Word, Excel, and PDF), which may contain detailed planning information about the project.

[0112] Data preprocessing:

[0113] Clean the data to remove irrelevant information, such as redundant spaces and incorrect date formats.

[0114] Format the data to ensure a consistent format for easier subsequent processing.

[0115] Step 1102. Extract task information and relationships between tasks using a pre-trained language model.

[0116] Choose a suitable pre-trained language model:

[0117] Pre-trained language models such as BERT and GPT-3 are selected, as these models perform well in natural language processing tasks.

[0118] Task information extraction:

[0119] Resource Information: The model is used to parse the task description and identify the resources involved (such as personnel and equipment). For example, for the task description "Zhang San is responsible for designing the UI interface", the model can identify "Zhang San" as a resource.

[0120] Planning Time: Extract start and end dates from the task description or a separate date field. For example, for the task "2023-04-01 to 2023-04-10: Develop backend functionality", the model can extract the time range.

[0121] Task Relationship Extraction:

[0122] Analyze the dependencies and sequence of tasks in the task descriptions. For example, task A "complete UI design" may be a prerequisite for task B "develop backend functionality".

[0123] The model is used to perform semantic understanding of task descriptions and identify the logical relationships between tasks.

[0124] Step 1103. Construct a knowledge graph based on task information and the relationships between tasks.

[0125] Define the nodes and edges of the knowledge graph:

[0126] Nodes: tasks, resources, time, etc.

[0127] Edges: Dependencies, resource allocation relationships, and time sequence relationships between tasks.

[0128] Building a knowledge graph:

[0129] Use graph databases (such as Neo4j) to build knowledge graphs.

[0130] The extracted task information (task name, resources, time) is added as nodes to the graph.

[0131] Based on the extracted task relationships, create corresponding edges in the graph.

[0132] Visualization and querying of knowledge graphs:

[0133] Use visualization tools (such as Neo4j Browser, Graphileon) to display the knowledge graph so that you can intuitively see the relationships between tasks.

[0134] It provides a query interface that allows users to search and filter based on keywords, resources, time, and other conditions in project planning data.

[0135] Step 1104. Set importance labels for task nodes.

[0136] Definition of Importance Tags: Importance tags are used to indicate the relative importance of a task within a project. This can be assessed based on multiple dimensions such as the task's scale, contribution to project goals, amount of resources invested, and time urgency. Tags can be qualitative (e.g., "high," "medium," "low") or quantitative (e.g., a rating of 1-5).

[0137] Importance assessment method: Assessment based on project objectives: Analyze the relationship between tasks and the core objectives of the project, as well as the contribution of tasks to the success of the project.

[0138] Assessment based on resource input: Consider the amount of human, material, and time resources required for the task.

[0139] Time-urgency-based assessment: Assess the urgency of a task based on its deadline and the current time.

[0140] Task dependency-based assessment: Analyze the position of a task in the task chain and the degree to which it depends on or is depended on by other tasks.

[0141] Generate importance labels: Based on the evaluation method described above, calculate an importance score or category for each task. Add the importance score or category as a label to the task node in the knowledge graph.

[0142] Tag visualization and updates: The knowledge graph visualization tool uses different colors, sizes, or icons to represent task importance tags. Users can update and adjust these importance tags based on project progress and actual conditions.

[0143] Example application:

[0144] Suppose we have simple project planning data that includes the following tasks:

[0145] Task 1: Zhang San is responsible for designing the UI interface (April 1, 2023 to April 5, 2023).

[0146] Task 2: Li Si is responsible for developing backend functions (April 6, 2023 to April 15, 2023), which depends on Task 1;

[0147] Task 3: Wang Wu is responsible for testing (April 16, 2023 to April 20, 2023), which depends on Task 2.

[0148] Using the methods described above, the following knowledge graph can be constructed:

[0149] Nodes: Task 1, Task 2, Task 3, Zhang San, Li Si, Wang Wu, 2023-04-01, 2023-04-05, 2023-04-06, ...

[0150] Edge: Task 1 → Task 2 (dependency relationship), Zhang San → Task 1 (resource allocation relationship), 2023-04-01 → Task 1 (time relationship), ...

[0151] This knowledge graph clearly shows the task allocation, time planning, and dependencies between tasks in a project, which helps project managers to better manage and make decisions.

[0152] In one embodiment of the present invention, based on step 120, the following will provide a possible embodiment and its specific implementation will be described in a non-limiting manner, such as... Figure 3 As shown.

[0153] Step 1201. Identify filenames using keyword technology and collect project execution data from the project management platform, document storage email system, and project logs.

[0154] This method utilizes keyword technology to identify filenames and collect project execution data from project management platforms, document storage email systems, and project logs. Keyword technology involves using keyword extraction algorithms from natural language processing techniques, such as TF-IDF, TextRank, or deep learning-based keyword generation models, to identify keywords relevant to project execution. These keywords can include task names, resource names, timestamps, etc. Project management platforms, such as Jira and Trello, typically record task status updates, comments, attachments, and other information during project execution. Project execution data from these platforms can be obtained through API interfaces or data export functions. Document storage email systems, such as Google Drive, Dropbox, and Outlook, store project-related documents and email communications. Documents and emails relevant to project execution can be collected through keyword searches or folder traversal. Project logs record important events, decisions, and changes during project execution. Data relevant to project execution can be extracted by parsing log files.

[0155] Step 1202. Extract task execution information from the project execution data using a pre-trained language model. The task execution information includes executed information and approval information. The executed information includes executed task information and execution time, and the approval information includes pending task information and execution time.

[0156] Pre-trained language models: Select pre-trained language models such as BERT and GPT-3. These models perform well in natural language processing tasks and can accurately understand the semantic information in the text.

[0157] Task execution information: including execution information and approval information.

[0158] Execution Information: This includes information on executed tasks and their execution time. For example, for the task description "Zhang San completed the UI design on 2023-04-05", the model can identify "Zhang San" as the executor, "UI design" as the executed task, and "2023-04-05" as the execution time.

[0159] Approval information: This includes information about the task to be executed and the execution time. For example, for the task description "Li Si's backend development function has been submitted for approval and is expected to be approved on 2023-04-07", the model can identify "Li Si" as the submitter, "backend development function" as the task to be executed, and "2023-04-07" as the approval time.

[0160] Based on the above implementation method, after the approval information is identified, the approval data management based on knowledge graph can automatically exclude approval information that does not meet the approval conditions, thereby improving approval efficiency.

[0161] (1) Based on the information of the task to be executed, query the corresponding target task information from the knowledge graph.

[0162] Extract key information from the task information to be executed, such as task name and task ID.

[0163] Use this key information to search the knowledge graph and find the corresponding target task node.

[0164] Read detailed task information from the target task node, including task description, person in charge, planned time, priority, etc.

[0165] (2) Query the status of the prerequisite tasks that the target task depends on.

[0166] Find all the preceding task nodes that are dependent on the target task from the knowledge graph.

[0167] Check the status attributes of these preceding task nodes to confirm whether they have been completed.

[0168] If there are any unfinished prerequisite tasks, record their task names and the reasons for not completing them (if they are recorded in the knowledge graph).

[0169] (3) Extract the person in charge information and planning time from the target task information.

[0170] Extract the person in charge of the task and the planned time from the retrieved target task information.

[0171] (4) Match the approval information with the responsible person information and planning time in the target task information.

[0172] Compare the approval information of the task to be executed (including approver, approval time and approval content) with the person in charge and the planned time in the target task information.

[0173] If the approval information does not completely match the target task information (e.g., the approver is not the task leader, or the approval time is earlier than the planned time), proceed to the next step to generate an error message.

[0174] (5) Generate error message

[0175] When the approval information does not match the target task information, a corresponding error message will be generated based on the specific content of the mismatch.

[0176] For example, if the approver is not the task leader, the message "The approver is not the task leader, please re-designate the approver" will be displayed.

[0177] (6) Generate a prompt message indicating that the approval information is normal.

[0178] When the approval information matches the target task information perfectly, a message indicating that the approval information is normal is generated.

[0179] For example, "The approval information is consistent with the target task information, and the approval process is normal."

[0180] (7) Determine the approval conditions and send the approval information.

[0181] After confirming that the approval information is normal, check the status of the prerequisite tasks that the target task depends on again.

[0182] If all prerequisite tasks are in the completed state, then the approval information is deemed to meet the approval conditions.

[0183] The approval information is sent to the review terminal (such as the project manager, department head, etc.) for further review.

[0184] If any of the prerequisite tasks that the target task depends on are incomplete, then the approval information is determined to be unsuitable for approval.

[0185] Generate a message indicating that the approval was not approved, and specify the incomplete prerequisite tasks and the reasons for them.

[0186] The information indicating that the approval was not approved will be fed back to the submitter of the task information so that they can understand the problem and take appropriate measures to resolve it.

[0187] In one embodiment of the present invention, based on step 130, the following will provide a possible embodiment and its specific implementation will be described in a non-limiting manner, such as... Figure 4 As shown.

[0188] Step 1301. Extract all execution information for the same task, including but not limited to task name, executor, execution time, and execution result. Organize and save this information to a dedicated list for subsequent processing.

[0189] Step 1302. Decompose the task into multiple execution phases and allocate execution time to the multiple execution phases based on the task's planned time.

[0190] Based on the complexity and workflow of the task, it is broken down into multiple specific execution phases.

[0191] Based on the planned timeframe for the task, allocate reasonable execution time to each execution phase to ensure that the overall task can be completed within the planned time.

[0192] For example, the complex task of software development can be broken down into multiple specific execution phases for better management and monitoring. These phases might include:

[0193] Requirements analysis phase: Communicate with the client to clarify the software's functional requirements and performance requirements.

[0194] Design phase: Based on the requirements analysis results, design the overall architecture and interface of the software.

[0195] Coding phase: Write the software source code according to the design document.

[0196] Testing phase: Unit testing, integration testing, and system testing are performed on the software to ensure its quality.

[0197] Deployment phase: Deploy the software to the production environment and perform necessary configuration and debugging.

[0198] Maintenance phase: Provide technical support and maintenance services after the software goes live.

[0199] Based on the planned timeline of the software development project (e.g., a total project cycle of 6 months), allocate reasonable execution time for each phase. These times may vary depending on the specific project and the team's efficiency, but can generally be roughly estimated as follows:

[0200] Requirements analysis phase: 1 month;

[0201] Design phase: 1.5 months;

[0202] Coding phase: 2 months;

[0203] Testing phase: 1 month;

[0204] Deployment phase: 0.3 months;

[0205] Maintenance phase (continued after project completion): to be arranged according to actual needs.

[0206] Step 1303. Based on the execution time and execution order of the execution phase, set the corresponding progress value for the execution phase.

[0207] Assign a progress value to each execution phase, which can be determined based on execution time and execution order. The progress value can be a percentage, representing the proportion or degree of completion of that phase in the overall task.

[0208] Assign a progress value to each stage. This progress value can be determined based on execution time and execution order, and can typically be represented as the proportion or degree of completion of that stage within the overall task. For example:

[0209] Requirements Analysis Phase: Progress value is 1 / 6 (approximately 16.67%), indicating that after this phase is completed, the overall task progress will be approximately 16.67%.

[0210] Design phase: The progress value is 1.5 / 6 (approximately 25%), indicating that after this phase is completed, the overall task progress will be approximately 25%.

[0211] Coding phase: Progress value is 2 / 6 (approximately 33.33%), indicating that after this phase is completed, the overall task progress will be approximately 33.33%.

[0212] Testing phase: Progress value 3 / 6 (approximately 50%), indicating that after this phase is completed, the overall task progress is approximately 50%.

[0213] Deployment phase: The progress value is 3.3 / 6 (approximately 55%), indicating that after this phase is completed, the overall task progress will be approximately 55%.

[0214] Maintenance phase: Since this phase continues after the project ends, no separate progress value is set. Instead, the overall task progress is updated as needed.

[0215] Step 1304. Obtain the completed execution stage corresponding to the executed task information. Based on the completed execution stage and the correspondence between the execution stage and the progress value, retrieve the progress value corresponding to the executed information.

[0216] Based on the information on executed tasks, determine which execution phases have been completed.

[0217] The total progress value of the executed information is calculated by establishing the correspondence between the completed execution phases and the execution phases and their progress values.

[0218] Step 1305. Convert the executed information in the list into a data group consisting of progress value and execution time.

[0219] The progress value and execution time of each executed message are combined into a data group.

[0220] These data sets will be used for subsequent time series construction and forecasting.

[0221] Step 1306. Arrange the data groups into a time data sequence according to their execution time.

[0222] Arrange all data groups in chronological order of execution time to form a time data sequence.

[0223] This time data series will reflect the historical progress and current status of the task execution.

[0224] Step 1307. Input the time data sequence and the current system time into the pre-trained long short-term memory neural network model to predict the task completion time.

[0225] A pre-trained Long Short-Term Memory (LSTM) neural network model is used, with a time data sequence and the current system time as input.

[0226] The model will predict the future completion time of the task based on patterns learned from historical data.

[0227] Step 1308. Calculate the difference between the completion time and the completion time limited by the planned time of the task, and generate status attributes based on the difference. The status attributes include early completion, normal progress, and overdue completion.

[0228] Compare the predicted completion time with the planned completion time and calculate the difference between them.

[0229] Based on the magnitude of the difference, corresponding status attributes are generated, such as "Completed Ahead of Schedule", "Normal Progress", and "Completed After the Deadline".

[0230] Step 1309. Add the state attribute to the corresponding task node in the knowledge graph.

[0231] Find the node in the knowledge graph that corresponds to the prediction task.

[0232] Add the generated status attributes to the node's attribute list to quickly obtain task status information for project monitoring and decision-making.

[0233] In one embodiment of the present invention, based on step 140, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner, such as... Figure 5 As shown.

[0234] Step 1401. Set anomaly detection indicators, including the proportion of overdue tasks, task progress deviation, resource utilization rate, and abnormal task dependencies.

[0235] Overdue task ratio: This calculates the percentage of tasks that are currently past their planned completion time out of the total number of tasks. This metric reflects the overall project schedule delays.

[0236] Task schedule deviation: This compares the actual progress of a task with the planned progress. This metric quantifies the degree of delay or ahead-of-time in task execution.

[0237] Resource utilization rate: This analyzes the allocation and use of resources during project execution, including human resources, time resources, and material resources. Low resource utilization may indicate resource waste, while high utilization may lead to resource shortages.

[0238] Task dependency anomalies: Check whether the dependencies between tasks meet the planning requirements. If a task's prerequisite task has not been completed but the task itself has already started executing, or if the execution order of tasks is disrupted, the overall progress and quality of the project may be affected.

[0239] Step 1402. Traverse the task nodes in the knowledge graph and obtain the state attributes of each task node.

[0240] After setting the anomaly detection metrics, it is necessary to traverse the task nodes in the knowledge graph and obtain the status attributes of each task node (such as progress, completion time, resource allocation, etc.).

[0241] Step 1403. Calculate the abnormal indicator value for each task.

[0242] Based on these state attributes and the calculation formulas for anomaly detection metrics, the anomaly metric value for each task is calculated.

[0243] Step 1404. Mark tasks with abnormal indicator values ​​exceeding the threshold as abnormal tasks and make special markings in the knowledge graph.

[0244] The calculated anomaly indicator values ​​are compared with preset thresholds. If an anomaly indicator value for a task exceeds the threshold, the task is marked as an anomaly and specially labeled in the knowledge graph (e.g., using different colors, icons, or text). This allows project managers to visually identify which tasks exhibit anomalies and take timely action.

[0245] Step 1405. By analyzing the task relationships, resource allocation, execution time, and task execution information of the abnormal task, and using the thinking chain reasoning technique, the abnormal cause of the task abnormality is located based on the analysis results.

[0246] First, it is necessary to collect and analyze relevant information for the "user interface design" task, including:

[0247] Task Relationships: Check the dependencies between this task and other tasks to confirm whether there are any pre-requisite tasks that have not been completed or subsequent tasks that have started ahead of schedule.

[0248] Resource allocation: Analyze whether the human resources (such as designers, developers, etc.) and time resources (such as the planned design cycle) allocated to this task are sufficient.

[0249] Execution time: Compare the actual execution time of the task with the planned execution time to determine if there is a delay.

[0250] Task execution information: View logs, meeting minutes, communication emails, etc. during the task execution process to understand the details of the task execution and the problems encountered.

[0251] After gathering sufficient information, the mental chain reasoning technique was used to deduce the cause of the anomaly in the "User Interface Design" task. Possible causes of the anomaly include:

[0252] Incomplete prerequisite tasks: Ambiguity or omissions in the requirements analysis phase can lead to an unclear design direction.

[0253] Insufficient resources: Insufficient number of designers or unreasonable time allocation leads to slow design progress.

[0254] Technical challenges: Difficult technical problems were encountered during the design process, such as the high learning cost of new design tools.

[0255] Poor communication: There were problems with communication with other team members, which resulted in changes in requirements not being communicated to the designers in a timely manner.

[0256] Step 1406. Input the abnormal cause of the abnormal task into the knowledge base to obtain multiple historical processing solutions.

[0257] The derived cause of the anomaly is input into a pre-built knowledge base to search for related historical handling solutions. Possible historical handling solutions include:

[0258] Increase resources: Increase the number of designers or adjust time allocation to speed up the design process.

[0259] Technical training: Organizing training courses for designers to learn new design tools or solve technical problems.

[0260] Enhance communication: Establish a more effective communication mechanism to ensure that changes in requirements are communicated to designers in a timely manner.

[0261] Step 1407. Use the thinking chain reasoning technique to conduct causal analysis on multiple historical processing solutions, and deduce multiple implementation steps and effects from the problem-solution-result chain.

[0262] Perform causal analysis on the historical solutions found, deduce the implementation steps and expected effects of each solution from the problem-solution-result chain. For example:

[0263] Additional resource options:

[0264] Implementation steps: Assess the workload of existing designers and determine the number of designers needed; coordinate with the human resources department to recruit or reassign designers; reallocate design tasks to ensure that each designer's workload is reasonable.

[0265] Expected results: faster design progress and improved resource utilization.

[0266] Technical training program:

[0267] Implementation steps: Identify the technical content requiring training; organize training courses or seminars; evaluate the training effectiveness to ensure that designers have mastered the necessary skills.

[0268] Expected outcome: Designers' technical skills will be improved, enabling them to solve technical problems in the design process more efficiently.

[0269] Step 1408. Based on the effects of the implementation steps, generate multiple implementation schemes with the best effects, wherein each implementation scheme includes multiple implementation steps.

[0270] The most effective implementation steps are randomly combined to form multiple implementation schemes.

[0271] Step 1409. Conduct a feasibility assessment of the implementation plan, and output the implementation plan that passes the feasibility assessment as a treatment plan.

[0272] Feasibility assessments were conducted on these options, taking into account factors such as the project's actual situation, resource constraints, and priorities, to select the optimal implementation plan.

[0273] Based on the above embodiments, in order to further improve the adaptability and usability of the knowledge base provided by the above embodiments, in an implementable manner, corresponding processing schemes for historical abnormal tasks of multiple projects are collected, and the historical data of multiple projects are saved as corresponding multiple datasets; the knowledge base is trained using a federated learning mechanism.

[0274] In project management practice, every project may encounter various exceptional tasks, and the handling process and results of these tasks are valuable experiences. In order to build a more comprehensive and effective knowledge base, it is necessary to collect corresponding handling solutions for historical exceptional tasks from multiple projects.

[0275] Data collection channels include:

[0276] Project documentation: Review the documentation for each project, including project plans, progress reports, meeting minutes, issue tracking records, etc., and extract information on abnormal tasks and their handling solutions.

[0277] Team communication: Conduct interviews or questionnaires with project team members to understand their experiences and lessons learned in handling unusual tasks.

[0278] Knowledge Management System: If a company has a knowledge management system or project management platform, it can directly extract historical data from it.

[0279] Data cleaning and preparation:

[0280] Organize the collected data to ensure that the handling plan for each abnormal task includes key information such as task description, cause of abnormality, handling steps, and handling results.

[0281] Data cleaning removes duplicate, incomplete, or invalid information to ensure data quality and accuracy.

[0282] Dataset Construction:

[0283] The processed data is categorized according to project or anomaly type, creating multiple datasets. Each dataset contains a series of anomaly tasks and their handling schemes for subsequent analysis and training.

[0284] Training the knowledge base using a federated learning mechanism

[0285] After collecting historical data from multiple projects, a federated learning mechanism can be used to train the knowledge base, enabling it to more accurately identify and handle anomalous tasks.

[0286] Overview of Federated Learning:

[0287] Federated learning is a distributed machine learning framework that allows multiple participants to collaboratively train a model without sharing the original data. This mechanism protects data privacy while leveraging the data resources of multiple participants to improve model performance.

[0288] Knowledge base training process:

[0289] Model initialization: Select a suitable machine learning model as the basis of the knowledge base, such as a decision tree, neural network, or deep learning model.

[0290] Data distribution: Distribute multiple datasets to different stakeholders (such as different project teams or departments) according to the requirements of federated learning.

[0291] Model training: On each participant, the model is trained using the local dataset to obtain local model parameters.

[0292] Parameter aggregation: Aggregate the local model parameters of each participant to obtain the global model parameters. This process can be repeated multiple times until the model performance reaches stability or a preset stopping condition is met.

[0293] Model evaluation and optimization: The global model is evaluated using the validation dataset, and the model is optimized and adjusted based on the evaluation results.

[0294] Knowledge base applications:

[0295] After training, the knowledge base can be used to identify new anomalous tasks and provide corresponding handling suggestions. When a new anomalous task appears, the knowledge base can retrieve similar cases from the training data based on the task description and the cause of the anomaly, and provide suggestions on handling steps and expected results.

[0296] Continuous updates and maintenance:

[0297] As the project progresses and new exceptional tasks emerge, the knowledge base needs to be continuously updated and maintained. This includes adding new exceptional task handling solutions, optimizing model parameters, and adjusting data distribution.

[0298] By collecting historical exception handling solutions from multiple projects and training the knowledge base using a federated learning mechanism, a more comprehensive, accurate, and intelligent knowledge base can be built, providing stronger support for project management.

[0299] The following is another embodiment of the present invention. In project management, real-time progress display and problem feedback mechanisms can help management grasp the project status in a timely manner and respond to problems quickly. The S7 module provides an interactive visualization interface to present project progress, anomalies, and risk information in intuitive charts, and supports user feedback on model identification and recommendation results. User feedback further refines and optimizes the model output, thereby continuously improving the accuracy and adaptability of the system. The specific steps are as follows:

[0300] Real-time data visualization provides an interactive user interface with real-time data visualization, displaying project progress, anomalies, and risks; it also integrates user feedback to refine model output.

[0301] By monitoring and graphically displaying key project data in real time, it provides a global view of project management, enabling users to quickly understand the current status of the project and potential risks.

[0302] Project Progress Dashboard: Provides an overall project progress dashboard, displaying the current status, completion percentage, and deadlines for each task through visualizations such as Gantt charts and timelines. Color-coded indicators help management quickly identify schedule deviations for delayed or ahead-of-time tasks.

[0303] Risk and Anomaly Charts: Display risk points and anomalies in the project. Through charts such as heatmaps and radar charts, the risk level of tasks, resource bottlenecks, and abnormal tasks in the critical path are displayed intuitively, supporting management's risk assessment and decision-making.

[0304] Resource Allocation and Usage: Visualize the current allocation and usage of resources (such as manpower, budget, and equipment), and display the distribution of resource consumption through bar charts, pie charts, and other formats to help users rationally plan resource allocation and quickly identify resource shortages or over-allocation issues.

[0305] User feedback integration and management:

[0306] By integrating user feedback during actual use and recording user preferences and opinions, we can further improve the model's output and personalized recommendations.

[0307] Feedback collection mechanism: Embed feedback buttons in each data visualization page and recommendation results to allow users to provide feedback on results such as progress display, risk identification, and solution recommendation (e.g., "accurate", "partially accurate", "inaccurate"), forming a user feedback dataset.

[0308] Feedback data classification and labeling: The feedback data is classified and labeled according to the type of feedback content (such as progress tracking, risk identification, and solution recommendation) to provide directional data for subsequent model optimization.

[0309] Personalized preference recording: Records user preference information in feedback (such as prioritizing certain types of risks, focusing on specific resource allocations, etc.), providing personalized customization support for the model, making subsequent recommendation results more in line with user needs.

[0310] Model output refinement and optimization:

[0311] By incorporating user feedback into the model training process and optimizing and fine-tuning the model output through feedback data, the model's response to project status and user needs becomes more accurate.

[0312] Feedback-driven model fine-tuning: This involves using feedback data to fine-tune and retrain the model to correct identification errors and improve recommendation accuracy. For example, training can be used to correct risk identification items that users have labeled as "inaccurate," thereby reducing errors.

[0313] Personalized optimization of the recommendation algorithm: Based on user feedback, the weights and strategies of the recommendation algorithm are optimized to make the recommended solutions and risk warnings more closely match user needs. For frequently selected specific solution types or high-risk levels, the system will prioritize recommendations and reminders in subsequent updates.

[0314] Dynamic updates and feedback adaptation: After receiving multiple similar feedbacks, the model automatically generates adaptive rules and algorithm adjustment strategies to adapt to user habits and project-specific needs, and updates them in real time in the system background.

[0315] Multi-dimensional project dynamic display:

[0316] Through multi-dimensional data display and perspective switching, users can view the project status from different angles and fully grasp the project dynamics.

[0317] Multi-level view switching: Provides multi-level views such as task-level, milestone-level, and overall project-level, allowing users to switch between detailed and global perspectives. For example, the milestone view displays a summary of progress for each stage, while the task-level view focuses on showing the specific status and risks of each task.

[0318] Historical data comparison view: By comparing current data with historical project data, management can understand project trends and progress efficiency, and identify differences from historical projects. For mission-critical tasks, it provides historical latency rates and resource consumption comparisons to facilitate the identification of potential improvement areas.

[0319] Future Trend Forecast View: Based on time series models, this view predicts future progress and risk trends, displaying future task completion rates and risk changes through line charts or trend graphs, providing decision-making references and early warnings.

[0320] Real-time notifications and interactive alerts:

[0321] By establishing a real-time notification and interactive alert system, users can be notified in a timely manner when critical developments and risk events occur, facilitating rapid response.

[0322] Real-time notification push: When the system detects new major anomalies, risks, or schedule deviations, it automatically pushes notifications to users to ensure that management can be informed of key project developments and respond immediately.

[0323] Customizable reminder settings: Allows users to set reminder rules according to their own needs (such as paying attention to delay warnings for specific tasks or the usage status of important resources), thereby achieving personalized dynamic reminders and ensuring that users receive the most relevant information.

[0324] Interactive task reminders: When a task is about to expire or resources are about to exceed limits, the system will trigger an interactive reminder, suggesting that users take appropriate adjustment measures (such as task rescheduling and resource scheduling) to improve the response speed and execution efficiency of project management.

[0325] In some embodiments, the project end-to-end data tracking and processing system may include multiple functional modules composed of computer program segments. The computer programs for each program segment in the project end-to-end data tracking and processing system may be stored in the memory of a computer device and executed by at least one processor to perform (see details). Figure 1 (Description) Functionality for tracking and processing data throughout the entire project process.

[0326] In this embodiment, the project end-to-end data tracking and processing system can be divided into multiple functional modules based on its functions, such as... Figure 6As shown. The functional modules of system 600 may include: a map construction module 610, an execution monitoring module 620, a status setting module 630, and an exception handling module 640. The module referred to in this invention is a series of computer program segments that can be executed by at least one processor and perform a fixed function, and are stored in memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

[0327] The knowledge graph construction module is used to acquire project planning data, extract task information and relationships between tasks from the project planning data using a pre-trained language model, the task information including planning time, and construct a knowledge graph based on the task information and relationships between tasks.

[0328] The execution monitoring module is used to obtain project execution data from multiple data sources and identify task execution information from the project execution data;

[0329] The status setting module is used to predict the task progress based on a data sequence composed of all task execution information, match the task progress with the corresponding planning time, and add status attributes to the corresponding nodes of the knowledge graph based on the matching results.

[0330] The exception handling module is used to analyze the abnormal situation of task execution status based on the knowledge graph after updating the status attributes, analyze the abnormal factors based on the project execution data, and generate a handling solution for the abnormal factors using a pre-built knowledge base.

[0331] In one optional implementation, the atlas construction module includes:

[0332] The data collection unit is used to collect project planning data from the project management platform, document storage email system, and project logs.

[0333] A preprocessing unit is used to preprocess the project planning data, the preprocessing including data cleaning and data formatting;

[0334] The model processing unit is used to input the pre-processed project planning data into the pre-trained language model to obtain task information and task relationships. The task information includes resource information and planning time, and the task relationships include the dependency relationship between tasks, the resource allocation relationship, and the time sequence relationship.

[0335] The graph construction unit is used to construct a knowledge graph by using tasks, resources, and time as nodes and task relationships as edges, and utilizing a graph database.

[0336] A tag generation unit is used to generate importance tags for tasks in the knowledge graph;

[0337] The interface setting unit is used to set the query interface for the knowledge graph.

[0338] Although the present invention has been described in detail with reference to the accompanying drawings and preferred embodiments, the present invention is not limited thereto. Various equivalent modifications or substitutions can be made to the embodiments of the present invention by those skilled in the art without departing from the spirit and essence of the invention, and such modifications or substitutions should all be within the scope of the present invention. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should also be covered within the protection scope of the present invention.

Claims

1. A project lifecycle data tracking process method, characterized by, The method comprises the following steps: acquiring project planning data, extracting task information and relationships between tasks from the project planning data by using a pre-trained language model, the task information including planning time, and constructing a knowledge graph based on the task information and the relationships between tasks; acquiring project execution data from multiple data sources and identifying task execution information from the project execution data; predicting task progress based on data sequences composed of all task execution information of a task, matching the task progress with corresponding planning time, and adding state attributes to corresponding nodes of the knowledge graph based on matching results; analyzing abnormal situations of task execution based on the knowledge graph with updated state attributes, analyzing abnormal factors based on project execution data, and generating a processing scheme for the abnormal factors by using a pre-constructed knowledge base; acquiring project execution data from multiple data sources and identifying task execution information from the project execution data, comprising: collecting project execution data from a project management platform, a document storage and mailing system, and project logs by keyword technology to identify file names; extracting task execution information from the project execution data by using a pre-trained language model, the task execution information including executed information and approval information, the executed information including executed task information and execution time, and the approval information including to-be-executed task information and execution time; The method further comprises: based on the to-be-executed task information, querying corresponding target task information and the status of pre-requisite tasks dependent on the target task from the knowledge graph; extracting responsible person information and planning time from the target task information; matching the approval information with the responsible person information and the planning time in the target task information, if not completely matched, generating an abnormal prompt information, if completely matched, generating a prompt information that the approval information is normal; if the status of pre-requisite tasks dependent on the target task are all completed, determining that the approval information meets the approval conditions, and sending the approval information to an audit terminal, if there is an uncompleted status among the pre-requisite tasks dependent on the target task, determining that the approval information does not meet the approval conditions, and generating a prompt information that the approval is not passed; predicting task progress based on data sequences composed of all task execution information of a task, matching the task progress with corresponding planning time, and adding state attributes to corresponding nodes of the knowledge graph based on matching results, comprising: saving executed information of the same task into the same list; decomposing a task into multiple execution stages and assigning execution times to the multiple execution stages based on the planning time of the task; setting corresponding progress values for the execution stages based on the execution times and execution sequences of the execution stages; acquiring completed execution stages corresponding to the executed task information, based on the completed execution stages and the corresponding relationship between the execution stages and the progress values, calling the progress values corresponding to the executed information; converting the executed information in the list into a data group composed of progress values and execution times; arranging the data group into a time data sequence according to execution times in chronological order; inputting the time data sequence and a current system time into a pre-trained long short-term memory neural network model to predict a completion time of the task; calculating a difference between the completion time and a completion time defined by a planning time of the task, generating a state attribute based on the difference, the state attribute including early completion, normal progress, and overdue completion; adding the state attribute to a corresponding task node in a knowledge graph.

2. The method of claim 1, wherein, acquiring project planning data, extracting task information and relationships between tasks from the project planning data using a pre-trained language model, the task information including a planning time, and constructing a knowledge graph based on the task information and the relationships between tasks, including: collecting project planning data from a project management platform, a document storage and mailing system, and project logs; preprocessing the project planning data, the preprocessing including data cleaning and data formatting; inputting the preprocessed project planning data into a pre-trained language model to obtain task information and task relationships, the task information including resource information and a planning time, and the task relationships including dependency relationships between tasks, resource allocation relationships, and time sequence relationships; constructing a knowledge graph using a graph database by taking tasks, resources, and times as nodes and taking task relationships as edges; generating importance labels for tasks in the knowledge graph; setting up a query interface for the knowledge graph.

3. The method of claim 1, wherein, analyzing abnormal situations of task execution based on the knowledge graph with updated state attributes, analyzing abnormal factors based on project execution data, and generating a processing scheme for the abnormal factors using a pre-constructed knowledge base, including: setting abnormal detection indicators, the abnormal detection indicators including overdue task proportion, task progress deviation, resource utilization rate, and task dependency relationship abnormality; iterating through task nodes in the knowledge graph to obtain state attributes of each task node; calculating abnormal indicator values for each task; labeling tasks with abnormal indicator values exceeding a threshold as abnormal tasks and marking them specially in the knowledge graph; analyzing task relationships, resource allocation, execution time, and task execution information of abnormal tasks, and using a thought chain reasoning technology to locate abnormal causes of task abnormalities based on the analysis results; generating a processing scheme based on abnormal causes of abnormal tasks using a pre-constructed knowledge base.

4. The method of claim 3, wherein, Generating a processing scheme based on abnormal causes of abnormal tasks using a pre-constructed knowledge base, including: inputting the abnormal causes of the abnormal tasks into the knowledge base to obtain multiple historical processing schemes; using a thought chain reasoning technology to perform causal analysis on the multiple historical processing schemes, and deducing multiple implementation steps and effects from a problem-scheme-result chain; based on the effects of the implementation steps, generating multiple implementation schemes with the best effects, the implementation schemes including multiple implementation steps; feasibility assessment of the implementation schemes, and outputting implementation schemes that pass the feasibility assessment as processing schemes.

5. The method of claim 4, wherein, The method further includes: collecting corresponding processing schemes of historical abnormal tasks of multiple projects, and saving historical data of the multiple projects as corresponding multiple data sets; training the knowledge base using a federated learning mechanism.

6. A project lifecycle data tracking processing system characterized by, including: The atlas construction module is configured to obtain project planning data, extract task information and relationships between tasks from the project planning data by using a pre-trained language model, the task information including planning time, and construct a knowledge graph based on the task information and relationships between tasks; The execution monitoring module is configured to obtain project execution data from multiple data sources and identify task execution information from the project execution data; The state setting module is configured to predict task progress based on a data sequence composed of all task execution information of a task, match the task progress with corresponding planning time, and add a state attribute to a corresponding node of the knowledge graph based on a matching result; The exception handling module is configured to analyze an abnormal situation of task execution based on the knowledge graph with updated state attributes, analyze an abnormal factor based on the project execution data, and generate a processing scheme for the abnormal factor by using a pre-constructed knowledge base; The execution monitoring module is configured to obtain project execution data from multiple data sources and identify task execution information from the project execution data, including: Collecting project execution data from a project management platform, a document storage and mail system, and a project log by keyword technology to identify file names; Extracting task execution information from the project execution data by using a pre-trained language model, the task execution information including executed information and approval information, the executed information including executed task information and execution time, and the approval information including to-be-executed task information and execution time; Further comprising: Based on the to-be-executed task information, querying corresponding target task information and a state of a preceding task dependent on the target task information from the knowledge graph; Extracting responsible person information and planning time from the target task information; Matching the approval information with the responsible person information and the planning time in the target task information, generating an abnormal prompt information if the matching is not complete, and generating a prompt information that the approval information is normal if the matching is complete; If the states of the preceding tasks dependent on the target task are all completed, determining that the approval information meets approval conditions and sending the approval information to an audit terminal; if there is an uncompleted state among the preceding tasks dependent on the target task, determining that the approval information does not meet approval conditions and generating a prompt information that the approval is not passed; The state setting module is configured to predict task progress based on a data sequence composed of all task execution information of a task, match the task progress with corresponding planning time, and add a state attribute to a corresponding node of the knowledge graph based on a matching result, including: Saving executed information of the same task into the same list; Decomposing a task into multiple execution stages and assigning execution times to the multiple execution stages based on planning time of the task; Setting corresponding progress values for the execution stages based on execution times and execution sequences of the execution stages; Obtaining completed execution stages corresponding to the executed task information, calling progress values corresponding to the executed information based on the completed execution stages and a corresponding relationship between execution stages and progress values; Converting the executed information in the list into a data group composed of progress values and execution times; Arranging the data group into a time data sequence according to execution times in chronological order; inputting the time data sequence and a current system time into a pre-trained long short-term memory neural network model to predict a completion time of the task; calculating a difference between the completion time and a planned completion time of the task, generating a state attribute based on the difference, the state attribute including early completion, normal progress, and overdue completion; adding the state attribute to a corresponding task node in a knowledge graph.

7. The system of claim 6, wherein, The graph construction module comprises: a data collection unit configured to collect project planning data from a project management platform, a document storage and mailing system, and a project log; a preprocessing unit configured to preprocess the project planning data, the preprocessing including data cleaning and data formatting; a model processing unit configured to input the preprocessed project planning data into a pre-trained language model to obtain task information and task relationships, the task information including resource information and planning time, and the task relationships including dependency relationships between tasks, resource allocation relationships, and time sequence relationships; a graph construction unit configured to construct a knowledge graph using a graph database by taking tasks, resources, and time as nodes and taking task relationships as edges; a label generation unit configured to generate importance labels for tasks in the knowledge graph; an interface setting unit configured to set a query interface for the knowledge graph.

Images