A system and method for intellectual property traceability management

Abstract

The present application relates to the technical field of intellectual property management, and particularly relates to a kind of intellectual property traceability management system and method, in research and development process, the task node with innovation potential is marked, automatically generates intellectual property seed object and constructs virtual cultivation space;Continuously monitor newly generated research and development data, actively establish association with existing seed object, dynamically update whole process asset atlas with seed as core;In response to intellectual property application instruction, automatically filter matching process evidence from cultivation space according to application type, encapsulate to form evidence package containing time stamp and personnel contribution record;Based on the correlation atlas, the quantitative analysis of technical correlation degree, team innovation capability and technical evolution path of intellectual property is carried out.The present application realizes the deep integration of intellectual property management and research and development process, improves the efficiency and quality of intellectual property application, and forms traceable and assessable core digital assets.

Description

Technical Field

[0001] This invention relates to the field of intellectual property management technology, and in particular to an intellectual property traceability management system and method. Background Technology

[0002] As corporate innovation activities deepen, intellectual property (especially patents and software copyrights) has become a crucial manifestation of a company's core competitiveness. However, in current intellectual property management practices, most companies still face the problem of a disconnect between "R&D" and "IP." Typically, patent or software copyright applications are only initiated after R&D projects are completed, making it difficult to systematically trace back to the background of the technical problem, the evolution of the solution, and supporting experimental data during the application process.

[0003] Existing intellectual property management systems are mostly results-oriented, only registering, maintaining, and charging fees for existing intellectual property achievements, lacking the ability to track and manage the "incubation process" of intellectual property. Even in systems that support document uploads, it is difficult to achieve dynamic correlation between documents and specific R&D tasks, personnel, and intellectual property seeds, resulting in a serious lack of traceability and completeness of the evidence chain for intellectual property. This not only affects the success rate and quality of intellectual property applications but also makes it difficult to obtain credible historical evidence for the subsequent operation, evaluation, and transformation of intellectual property. Summary of the Invention

[0004] The purpose of this invention is to provide an intellectual property traceability management system and method, which realizes the deep integration of intellectual property management and R&D process, improves the efficiency and quality of intellectual property application, and forms traceable and assessable core digital assets.

[0005] To achieve the above objectives, in a first aspect, the present invention provides a method for managing the traceability of intellectual property rights, comprising the following steps: During the R&D project management process, the set task nodes are marked, and intellectual property seed objects are automatically generated based on the marking operations. The type, creator, creation time, project and task node of the intellectual property seed object are recorded. A virtual cultivation space is created for each intellectual property seed object, and the project information, participant information, and process data generated during task execution of the intellectual property seed object are automatically associated to form a preliminary association map with the intellectual property seed object as the core. Continuously monitor newly generated data during the R&D process, establish associations between the new data and existing intellectual property seed objects through proactive or manual association, and dynamically update the full-process asset map with the intellectual property seed objects as the core. Based on the intellectual property application instructions, and according to the type of intellectual property applied for, matching full-process evidence data is extracted from the cultivation space and packaged into an intellectual property creation process evidence package containing timestamps and personnel contribution records. Based on the full-process asset map, the technological relevance, team innovation capability, and technological evolution path of the intellectual property seed objects are analyzed, and quantitative analysis results are generated.

[0006] In the R&D project management process, designated task nodes are marked, and intellectual property seed objects are automatically generated based on the marking operations. The type, creator, creation time, project, and task node of the intellectual property seed object are recorded, including: In the R&D project management platform, several task nodes are established for each R&D project. Each task node has a unique identifier and is associated with its respective project. Based on the user's operation on the marker control, the seed type and technology topic selected by the user are obtained; The system automatically captures the context information of the current task node, including the task name, the project it belongs to, the person in charge, and the start and end times; Based on the seed type, technology topic, and automatically captured context information filled in by the user, an intellectual property seed object is created in the database, and a unique seed number is assigned to the intellectual property seed object. At the same time, the intellectual property seed object is established with the task node that triggers the marker to establish a "source" relationship, and the association time and associated person are recorded.

[0007] Specifically, a virtual cultivation space is created for each intellectual property seed object, automatically linking it to project information, participant information, and process data generated during task execution, forming a preliminary association map centered on the intellectual property seed object, including: Simultaneously with the generation of the seed object, a virtual cultivation space instance uniquely corresponding to the intellectual property seed object is automatically constructed; Based on the source task identifier recorded in the intellectual property seed object, retrieve all information of the task node and its attached documents and comment records from the task database, and import them into the cultivation space as initial association data; Based on the project identifier of the source task, retrieve the basic information of the project and import it into the cultivation space; Identify the list of participants in the source task, obtain other output data of the participants within a preset time window, and import them into the cultivation space after filtering by keyword similarity calculation; For each imported piece of related data, record its relationship type with the seed object, as well as the establishment time and method, to form a preliminary association graph with the intellectual property seed object as the core. The relationship types include project affiliation, task origin, task output, personnel contribution, content relevance, and time proximity.

[0008] This includes continuously monitoring newly generated data during the R&D process, establishing associations between the new data and existing intellectual property seed objects through proactive or manual methods, and dynamically updating the entire process asset map centered on the aforementioned intellectual property seed objects, including: Scan for data changes in the document management library, code repository, task management system, and experimental data platform in real time or on a scheduled basis; When new data is detected, text extraction, word segmentation and cleaning, and keyword extraction are performed on the new data to generate feature vectors; The similarity between the feature vector and the technology topic vector of the intellectual property seed object in the cultivation state is calculated. The technology topic vector is dynamically updated according to the technology topic when the seed is created and subsequent associated documents. When the similarity exceeds a preset threshold, a correlation suggestion is generated and pushed to the cultivation space of the corresponding seed object; Based on the user's confirmation of the association suggestion, an association record between the new data and the seed object is established in the relational database, recording the relationship type, confirmation time, confirmer, and similarity score, and the new data is included in the cultivation space.

[0009] The method further includes: Nodes and edges are constructed based on a graph database. The node types include intellectual property seeds, projects, tasks, personnel, documents, code submission records, and experiment records. The edge types represent the relationships between nodes, and each edge is accompanied by the creation time, creation method, and operator attributes. Whenever a new relationship is established, check if the new data node already exists in the graph. If it does not exist, create the node and create an edge of the appropriate type between the seed node and the new data node to build a full-process asset graph centered on the seed object.

[0010] Specifically, based on the intellectual property application instructions and according to the type of intellectual property applied for, matching full-process evidence data is extracted from the cultivation space and packaged into an intellectual property creation process evidence package containing timestamps and personnel contribution records, including: A pre-stored mapping model between intellectual property types and evidence requirements defines the core evidence categories, data sources, relationship types, and priorities required for each type of intellectual property. Obtain the intellectual property application type selected by the user, traverse the associated data in the nurturing space according to the mapping model, and filter by data type matching, relationship type matching, and priority sorting; The screened evidence data is checked for timeline integrity to detect whether there is a time gap exceeding a preset threshold between the first occurrence of the technical problem and the application decision date. If there is a gap, the user is prompted to supplement it. The filtered evidence data is automatically categorized into core evidence categories and a catalog is generated. The original files are extracted from each data source and packaged. Metadata information is embedded in each evidence file, including file identifier, association type, uploader, upload time, and version number. Generate a digital fingerprint and timestamp for the entire evidence package, and output it as a compressed file with an evidence list.

[0011] The method further includes: When a user switches the intellectual property application type after generating an evidence package, the selected evidence data is re-filtered according to the new type's mapping model, retaining matching evidence and prompting for missing core evidence categories; If evidence data is manually selected or unselected before the evidence package is generated, the catalog will be updated in real time and a warning will be given about potential gaps in the chain of evidence caused by custom operations. If there is no data under a certain core evidence category, supplementary suggestions are provided based on the mapping model to guide users to upload or associate data of the missing type.

[0012] Among these, based on the entire process asset map, the technical relevance analysis of the intellectual property seed objects includes: Count the number of different projects, different tasks, and different technical documents directly associated with the seed object, and calculate the breadth of association; Each association is assigned a weight based on its type, with core association types having a higher weight than weak association types. Duplicate associations and version iterations are counted to calculate the association strength. A comprehensive index of technical correlation is generated by combining the breadth and strength of correlation, which is used to compare the technical foundation of different seed objects horizontally.

[0013] Specifically, based on the aforementioned full-process asset map, the analysis of the technological evolution path of the intellectual property seed objects includes: Arrange all nodes associated with the seed object in chronological order to form a timeline, with each node representing an event and labeling the event type. By analyzing the reference relationships and content similarity between nodes, the main path and branch paths of technological evolution are identified. The main path starts from the earliest event, and each time the event with the highest content similarity or the closest relationship with the current event is selected as the next node. Identify key nodes in the evolution path, including turning points where the technology direction changes, milestones that produce significant results, and bottlenecks where there is no progress for a long time, and generate an evolution path description.

[0014] Secondly, the present invention provides an intellectual property traceability management system, applied to an intellectual property traceability management method as provided in the first aspect, comprising: The seed management module is used to mark the set task nodes during the R&D project management process, and automatically generate intellectual property seed objects based on the marking operations, and record the type, creator, creation time, project and task node of the intellectual property seed object; The space construction module is used to create a virtual cultivation space for each intellectual property seed object, automatically associate the project information, participant information and process data generated during task execution of the intellectual property seed object, and form a preliminary association map with the intellectual property seed object as the core. The dynamic association module is used to continuously monitor newly generated data during the R&D process. Through active or manual association, it establishes association relationships between new data and existing intellectual property seed objects, and dynamically updates the full-process asset map with the intellectual property seed objects as the core. The evidence encapsulation module is used to extract matching full-process evidence data from the cultivation space based on the intellectual property application instructions and according to the type of intellectual property applied for, and encapsulate it into an intellectual property creation process evidence package containing timestamps and personnel contribution records; The value analysis module is used to analyze the technological relevance, team innovation capability, and technological evolution path of the intellectual property seed objects based on the full-process asset map, and generate quantitative analysis results.

[0015] This invention discloses an intellectual property traceability management system and method. During the R&D process, it marks task nodes with innovative potential, automatically generates intellectual property seed objects, and constructs a virtual cultivation space. It continuously monitors newly generated R&D data, proactively establishes associations with existing seed objects, and dynamically updates the entire process asset map centered on the seeds. Responding to intellectual property application instructions, it automatically selects matching process evidence from the cultivation space according to the application type, encapsulating it into an evidence package containing timestamps and personnel contribution records. Based on the association map, it performs quantitative analysis of the technological relevance, team innovation capabilities, and technological evolution paths of intellectual property. This invention achieves deep integration of intellectual property management and the R&D process, improves the efficiency and quality of intellectual property applications, and forms traceable and assessable core digital assets. It can be widely applied in the fields of enterprise intellectual property management and R&D process management. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0017] Figure 1 This is a schematic diagram illustrating the steps of an intellectual property traceability management method according to the first embodiment of the present invention.

[0018] Figure 2 This is a simplified flowchart illustrating a method for managing the traceability of intellectual property rights provided by this invention.

[0019] Figure 3 This is a complete flowchart of an intellectual property traceability management method provided by the present invention.

[0020] Figure 4 This is a structural schematic diagram of an intellectual property traceability management system according to the second embodiment of the present invention.

[0021] Figure 5 This is a schematic diagram of the electronic device of the present invention.

[0022] In the diagram: 101-Seed Management Module, 102-Space Construction Module, 103-Dynamic Association Module, 104-Evidence Encapsulation Module, 105-Value Analysis Module. Detailed Implementation

[0023] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0024] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0025] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0026] The first embodiment of this application is as follows: Please see Figures 1-3 This invention provides a method for managing the traceability of intellectual property rights, comprising the following steps: S1. During the R&D project management process, the set task nodes are marked, and intellectual property seed objects are automatically generated based on the marking operation. The type, creator, creation time, project and task node of the intellectual property seed object are recorded.

[0027] Specifically, within the system, each R&D project is defined as an independent data entity, containing basic information such as project number, project name, start and end dates, project objectives, and project leader. After a project is created, the project leader or administrator can break it down into several executable task nodes according to the R&D plan. A task node is the smallest management unit for R&D activities and can be a specific task such as technical research, solution design, experimental verification, code development, or test review.

[0028] The methods for creating task nodes include: Planned creation: During the project initiation phase, task nodes are pre-created based on the project task breakdown structure (WBS), and attributes such as task name, description, estimated working hours, responsible person, and deadline are set.

[0029] Dynamic creation: During project execution, project members can temporarily create task nodes (such as analysis of sudden technical problems, temporary experiments, etc.) as needed, and add necessary attributes.

[0030] Each task node in the system has a unique identifier and is associated with its project, creator, creation time, and status (such as pending, in progress, or completed). Task nodes also support features such as attachment uploads, comments, and version history, used to accumulate various process data generated during the development process.

[0031] During task execution, project members (such as R&D engineers and technical leads) can determine, based on their own experience or team discussions, whether the current task has produced technologically innovative results, for example: The technical solution received breakthrough approval during the review process; The experiment yielded unexpectedly excellent data; Algorithm design achieves new functionality or performance improvements; They discovered process techniques that could be registered as trade secrets.

[0032] When a member believes that a task node has the potential to form intellectual property rights (such as patents, trade secrets, or software copyrights), they can trigger it through the "Mark as Intellectual Property Seed" function provided by the platform. This marking function is embedded in the task details page in the form of a visual control (such as a button or menu item), and a marking window pops up when the user clicks it.

[0033] In the tagging window, the system prompts the user to provide the following information to ensure an accurate description of the seed: Seed type: Select from preset types, including invention patent, utility model, design, trade secret, software copyright, etc. (customizable according to enterprise needs).

[0034] Technical topic: Briefly describe the core content of this innovation, such as "an image recognition method based on deep learning".

[0035] Note: Please provide supplementary explanations regarding the value, application scenarios, or subsequent development suggestions of the innovative points.

[0036] The system automatically captures the context information of the current task node, including task name, task description, project to which it belongs, current person in charge, task start and end time, etc., as the initial attributes of the seed object.

[0037] After the user confirms the marking, the system background performs the following operations to generate an intellectual property seed object: Create a seed entity: Create a new "Intellectual Property Seed" record in the database, assign a unique seed number (such as IP_Seed_202310001), and store information such as the seed type, technology topic, and remarks filled in by the user.

[0038] Associating Tasks and Projects: Establishing a "origin" relationship between the seed object and the task node that triggered the marker, while indirectly associating it with the project through the task node. The system records the association time and the associated person (i.e., the marker user).

[0039] Record creation information: The seed object stores the creator (marked user) and creation timestamp to ensure the traceability of the seed's origin.

[0040] Initialization status: The initial status of the seed object is set to "under cultivation", indicating that it has not yet entered the formal application process.

[0041] In addition, the system automatically establishes a preliminary association between the existing process data of the task nodes (such as uploaded documents, experimental records, code submissions, etc.) and the seed objects, forming the initial content of the seed cultivation space. This data serves as the basis for the subsequent dynamic association graph.

[0042] Once a seed is generated, the system displays a "Marked as Intellectual Property Seed" icon on the task node page and provides an entry point for users to view the seed cultivation space at any time. Simultaneously, the system can send notifications to project leaders or intellectual property specialists, alerting them to the generation of new seeds so they can monitor and cultivate them promptly.

[0043] If the same innovation involves multiple task nodes (e.g., the scheme design task and the experimental verification task both contribute to the same invention), the system supports manually associating other task nodes with the generated seed to achieve multi-task convergence.

[0044] S2. Create a virtual cultivation space for each intellectual property seed object, automatically associate the project information, participant information, and process data generated during task execution of the intellectual property seed object, and form a preliminary association map with the intellectual property seed object as the core.

[0045] Specifically, as a seed is generated, the system automatically creates a unique "nurturing space" instance in the background. This instance is logically a virtual container that does not occupy physical storage space. Instead, it aggregates data scattered across different modules such as the project management system, document management system, code repository, and experimental data platform through dynamic queries and association rules. The nurturing space is represented as a separate page or panel in the user interface, containing the following core areas: Seed basic information area: Displays seed number, seed type, technical topic, creator, creation time, current status, etc.

[0046] Related Data Aggregation Area: Displays related projects, tasks, personnel, documents, code, experiment records, and other data in the form of lists, cards, or graph thumbnails.

[0047] Operation Area: Provides entry points for functions such as manually adding associations, confirming system-suggested associations, viewing details, and exporting evidence packages.

[0048] The basic framework of the cultivation space is initialized when the seed is created, and the content is dynamically updated as data is continuously generated and association rules are executed.

[0049] To achieve comprehensive process traceability, the system needs to obtain various types of data that may be related to the seeds from multiple data sources on the enterprise's R&D platform. These data sources include, but are not limited to: Project Database: Stores basic project information (project name, number, start and end dates, objective description), project members, project phases, etc.

[0050] Task database: Stores task node information (task name, description, status, person in charge, start and end time, project), task comments, task attachments, etc.

[0051] Document Management Library: Stores various documents generated during the R&D process (such as meeting minutes, technical solutions, design drawings, experimental reports, and test records). Each document contains metadata such as uploader, upload time, version number, document tags, and keywords.

[0052] Code repository: Stores code commit records (committer, commit time, commit information, file change list) and associates them with specific tasks or projects.

[0053] Experimental data platform: Stores experimental records, raw data files, analysis results, etc., and associates them with experimental personnel, experimental time, and experimental purpose.

[0054] The system uses a unified data access interface to retrieve and extract data from the aforementioned data sources based on conditions such as seed identifier, project identifier, personnel identifier, and time range.

[0055] After the seed is created, the system immediately performs the first data aggregation operation, automatically acquiring and associating the initial data according to the following rules: Direct association based on task nodes: Since the seed originates directly from a specific task node (referred to as the "source task"), the system first retrieves all information about that task node from the task database based on the "source task identifier" stored in the seed object, including: Basic attributes of a task (task name, description, start and end times, status); List of personnel involved in the task (person in charge, participating members); All attachments uploaded during task execution; All comments related to the task; If the task is associated with a code repository, then retrieve all code records submitted under that task.

[0056] The above data is automatically imported into the cultivation space and a direct association is established with the seeds, with the association type marked as "originating from task".

[0057] Indirect connections based on projects: Seeds are indirectly associated with their respective projects through source tasks. Based on the project identifier of the source task, the system retrieves the project's basic information (project name, objective description, project members, etc.) and imports it as associated data into the nurturing space. Simultaneously, the system further retrieves other task nodes under that project, but to control the breadth of the initial associations, it only retrieves tasks that are close in time to the source task (e.g., within one month before or after) or related to the technical topic. These are considered potentially related data and are not automatically associated; instead, they are added to a "suggested association list" for later user confirmation.

[0058] Based on personnel association: The system identifies the list of participants in the source task and retrieves their other outputs both inside and outside the project from the following dimensions: Documents and code submitted by this person in other tasks; Other projects in which this person participated and their results; This person's professional field tags and historical intellectual property contribution record.

[0059] To avoid over-association, in the initial stage, only outputs that are close in time to the source task (as mentioned above) and match the technical keywords will be automatically associated with the nurturing space, and the rest will be considered as suggestions.

[0060] Supplementary associations based on time windows: The system uses the seed creation time as a baseline and traces back a certain time window (e.g., three months) to retrieve all data generated within the project during that window that is not covered by the aforementioned rules (such as independently uploaded documents and experimental records). Keywords are extracted using natural language processing techniques, and similarity is calculated between these keywords and the seed's technical theme. Data with similarity exceeding a preset threshold is automatically associated with the cultivation space, and the association type is marked as "time proximity and content relevance."

[0061] To support subsequent dynamic graph construction and refined analysis, the system records the relationship type and attributes of each association when establishing it. Relationship types mainly include: Project Affiliation: The relationship between the seed and its affiliated project.

[0062] Originating from the task: the relationship between the seed and the task node that directly triggers its creation.

[0063] Task outputs: The relationship between the seed and the documents, code, and other data generated under the source task.

[0064] Personnel contribution: The relationship between the seed and the participants can be further divided into roles such as person in charge and participants.

[0065] Referenced documents: The relationship between a seed and documents matched via keywords or manually associated.

[0066] Code commit: The relationship between a seed and a code commit record.

[0067] Experimental record: The relationship between seeds and experimental data.

[0068] Time proximity: The relationship between a seed and data that is automatically associated based on a time window.

[0069] Suggested associations: Potentially related data identified by the system but not yet confirmed by the user.

[0070] Each relationship type comes with the following attributes: creation time, creation method (automatic / manual), operator (if manual), and similarity score (if content-based matching). This typified information is stored in the relation database, forming edges with seed nodes, providing a structured foundation for subsequent graph queries, analysis, and evidence chain encapsulation.

[0071] The system presents the automatically acquired and associated data in a visual manner within the nurturing space. The default display mode is a list view, categorized by data type (projects, tasks, personnel, documents, etc.). Users can switch to a graph view, where the seed is the central node, associated data are surrounding nodes, and nodes are connected by labeled edges (such as "belongs to," "output," "participates," etc.), initially forming an association graph centered on the seed.

[0072] For data that is "suggested to be associated", the system displays it with a special identifier (such as gray or dashed border) and provides a one-click confirmation or rejection option, making it convenient for users to participate in improving the association relationship.

[0073] S3. Continuously monitor newly generated data during the R&D process, establish associations between the new data and existing intellectual property seed objects through active or manual association, and dynamically update the full-process asset map with the intellectual property seed objects as the core.

[0074] Specifically, the system deploys a data monitoring service in the background to scan for changes in various data sources within the enterprise's R&D platform in real time or periodically. The monitoring scope includes: Document Management Library: Monitors document upload, update, and deletion events, and captures the metadata (filename, uploader, upload time, project, tags) and full text content of newly added documents.

[0075] Code repository: Listen for code commit events, capture commit information, list of changed files, committer, commit time for each commit, and extract key comments or technical terms from code snippets.

[0076] Task Management System: Monitors events such as changes in task status, addition of task comments, and addition of task attachments.

[0077] Experimental data platform: Listen for events such as the creation of experimental records and the upload of data files.

[0078] Project information change: Listen for events such as project member adjustment and project goal update.

[0079] The monitoring service can adopt a message queue mechanism. When the data source changes, the system automatically generates corresponding event messages and pushes them to the associated update processor. For data sources that do not support real-time pushing, the system can set up a scheduled task (such as every hour) to scan for incremental data to ensure that new data is captured in a timely manner.

[0080] Active association means that the system automatically analyzes the semantic content of new data, judges its relevance to existing intellectual property seeds, and recommends establishing an association. The specific processing process is as follows: Data preprocessing and feature extraction: When new data is detected (taking a newly uploaded document as an example), the system first preprocesses it: Text extraction: If it is a parsable document (such as PDF, Word, TXT), the system calls a text parsing engine to extract the full text content; if it is a picture or a scanned copy, it first performs OCR recognition; if it is a code submission, it extracts the submission information and key comments, function names, variable names, etc. in the code.

[0081] Word segmentation and cleaning: Segment the extracted text, remove stop words (such as meaningless words like "de", "shi", etc.), and perform stemming processing to obtain a list of effective technical terms.

[0082] Keyword extraction: Use algorithms such as TF-IDF and TextRank to extract the keywords that best represent the technical theme of the document from the text to form a document feature vector. At the same time, if the document is accompanied by manual tags, they are used as weighted features.

[0083] Vectorization of seed technical themes: The system pre-maintains a technical theme vector for each intellectual property seed in the "cultivation" state. This vector is generated based on the "technical theme" filled in by the user and the relevant documents of the source task when the seed is created, and is dynamically updated as subsequent associated documents are added. The update method can adopt incremental learning. For example, the keyword vectors of the newly added associated documents are weighted and merged into the seed theme vector so that the seed theme can reflect the evolution of its technical connotation.

[0084] Similarity calculation and candidate matching: For the feature vector of new data, the system calculates its similarity with the technology topic vectors of all active seeds. The similarity metric can be cosine similarity, Euclidean distance, or the Jaccard coefficient. A dynamic threshold (e.g., 0.6) can be set during the calculation; seeds with similarities above this threshold are considered "candidate matches." If no matching results are found, no association suggestions are generated.

[0085] To avoid excessive computation, the system can adopt index optimization strategies, such as using inverted indexes or vector retrieval libraries (like Faiss) to accelerate the matching process, and only perform fine-grained calculations on seeds that are similar in technical field.

[0086] Related suggestion generation and push: For each candidate match, the system generates a related suggestion record containing the following information: A unique identifier for new data (such as a document ID); Matching seed ID and similarity score; Suggested relationship types (e.g., "content-related documents"); The basis for the suggested generation (e.g., a list of terms based on keyword matching).

[0087] The system will push the association suggestions to the corresponding seed's cultivation space and display them in a list in the "Suggestion Association" area. At the same time, it will remind the relevant seed personnel (such as the seed creator, project leader, and intellectual property specialist) to confirm via in-site notification or email.

[0088] User confirmation and association retention: After viewing the suggestions in the nurturing space, users can choose to "Confirm Association" or "Ignore". If the association is confirmed, the system will perform the following operations: Create a new record in the relational database that links the data to the seed. The record should be of type "content related" (or other preset type) and include information such as confirmation time, confirmer, and similarity score.

[0089] The new data will be formally incorporated into the associated data aggregation area of ​​the cultivation space, and the map display will be updated.

[0090] Trigger incremental updates to the seed topic vector, incorporating the feature vectors of the new data into the seed topic to make subsequent matching more accurate.

[0091] If the user ignores the suggestion, it will be marked as "ignored". If the same data is matched again in the future, the system may adjust the priority or stop making suggestions.

[0092] Manual association provides users with the ability to actively intervene, suitable for the following scenarios: users believe that certain data is related to the seed, but it was not automatically captured by the system (e.g., the data was generated before the seed was created, or the semantic matching degree did not reach the threshold). The manual association process is as follows: Starting from the seeds: In the operation area of ​​the seed cultivation space, a "Manually Add Association" button is provided. Clicking it will bring up a data selection window. In the window, data can be filtered by project, task, personnel, time range, etc. The user can confirm the association after checking the selected items.

[0093] Starting with data: On data display interfaces such as document details pages and code submission pages, a "Link to Intellectual Property Seed" button is provided. Users can click on it to search for and select the target seed to complete the association.

[0094] Batch association: Supports batch association of a batch of data to a specified seed at the project or task level.

[0095] When manually establishing associations, the system also records the relationship type (selected by the user, such as "referenced document", "experimental support", "code implementation", etc.), association time, operator, and incorporates them into the cultivation space and graph.

[0096] As relationships are continuously established, the system needs to construct a dynamic graph that can intuitively display the complex relationships between seeds and various assets. The construction and updating of the graph are based on graph databases (such as Neo4j) or graph query extensions of relational databases, specifically implemented as follows: Node and edge definitions of a graph: Node types include intellectual property seeds, projects, tasks, personnel, documents, code commit records, experiment records, etc. Each node stores its unique identifier, type, name, creation time, and other attributes.

[0097] Edge type: Represents the relationship between nodes, such as: Seed - [Belongs to] → Project Seed - [Source] → Task Task - [Output] → Documentation Personnel - [Participation] → Task Document - [Content Related] → Seed Code - [Implementation] → Seed Each edge has attributes such as creation time, creation method, and operator.

[0098] Whenever a new relationship is established (whether actively confirmed or manually added), the system automatically performs the following graph update operations: Node check: If a new data node (such as a new document) does not yet exist in the graph, create the node and populate its attributes.

[0099] Edge creation: Create an edge between the seed node and the data node. The edge type is determined according to the association scenario (such as "content related" or "manual association"), and the attributes described above are attached.

[0100] Related propagation: If new data is already related to other nodes (such as projects or tasks), the system can further explore and suggest including these indirect relationships in the graph. For example, if a new document is related to a task, and that task is already related to another sub-task, the system can suggest whether there is a cross-seed relationship, but manual confirmation is required to avoid over-association.

[0101] The system provides a spectral visualization interface and supports the following functions: Expanding from the seed: By default, the seed and its one-hop related nodes are displayed. Users can click on a node to expand more related nodes and view multi-hop relationships.

[0102] Timeline filtering: Filter nodes and edges by time range to display the interconnected network within a specific time period.

[0103] Relationship filtering: Filter based on relationship type (e.g., only display document associations).

[0104] Path analysis: Query the associated paths between two nodes, such as all intermediate nodes between a seed and a certain experimental data.

[0105] The graph data is stored in a graph database, and the front end renders it using a graph visualization library (such as D3.js or ECharts). After each update, the graph view can be refreshed in real time or manually by the user.

[0106] S4. Based on the intellectual property application instructions, and according to the type of intellectual property applied for, extract matching full-process evidence data from the cultivation space, and package it into an intellectual property creation process evidence package containing timestamps and personnel contribution records.

[0107] Specifically, to support evidence extraction for different types of intellectual property, the system predefines an "evidence requirements model." This model can be sourced from sources such as examination guidelines or corporate experience, or it can be customized. The model uses intellectual property type as a dimension, clearly defining the core evidence categories, priorities, and data sources required for each type of application. The model is built upon intellectual property laws and regulations, examination guidelines, and corporate practice experience, and can be dynamically adjusted. Examples of evidence requirements for major intellectual property types are shown in Table 1. Table 1 Evidence Requirements for Major Intellectual Property Types The model is stored in the system configuration repository, allowing administrators to add, delete, modify, and query data according to the actual needs of the enterprise. The "Data Source and Relationship Type" field in the model specifies which related data should be retrieved from the nurturing space and which relationship types should be prioritized.

[0108] When a user clicks "Generate Evidence Package with One Click" in the seed cultivation space, the system first pops up an intellectual property type selection window for the user to confirm the type of application (such as an invention patent). After the user selects, the system automatically matches the most likely type as the default option based on the current status of the seed (such as the existence of a preliminary technical disclosure) and the content of the cultivation space. The user can manually modify this option.

[0109] The system takes the type selected by the user as a parameter, calls the evidence extraction engine, and begins the extraction process.

[0110] First, retrieve all associated data nodes and their relationship records for the seed from the cultivation space. Each data node is stored with its metadata (such as document name, upload time, version number, author, etc.) as well as its relationship type with the seed (such as "originating from task", "content related", "code implementation", etc.) and relationship attributes (such as creation time, similarity score).

[0111] The system iterates through all related data based on the evidence requirement model corresponding to the selected type and filters it according to the following rules: Type matching: Checks whether the data type of the data node itself (document, code, experiment record, etc.) matches the "data source" in the requirement model. For example, if an invention patent requires "meeting minutes", the system will filter out nodes whose data type is "document" and whose document type label contains "meeting minutes".

[0112] Relationship type matching: Check whether the relationship type between data nodes and seeds is within the range specified in the requirement model. For example, an invention patent requires that data in the "technical solution iteration process" must come from a "task output" or "content related" relationship. If a document and a seed only have a "time proximity" relationship and have not been manually confirmed as content related, it may be excluded.

[0113] Priority sorting: For data that meets the conditions, the system groups them according to the priority (high / medium / low) in the requirement model. Data with higher priority will be included in the evidence package first and highlighted in the evidence package.

[0114] Deduplication and Version Selection: The same data node may have multiple versions (e.g., multiple document updates). The system defaults to selecting the latest version while retaining historical versions as attachments or notes. If the requirement model specifies the need to reflect process iteration, all versions are included in chronological order.

[0115] To ensure the continuity of the evidence chain, the system automatically verifies the integrity of the timeline of the evidence package after screening. The specific procedure is as follows: Extract the timestamps of all selected evidence (document upload time, code submission time, experiment record time, etc.) and sort them in ascending order.

[0116] Identify the time period from the first occurrence of the seed technical issue (typically the creation time of the source task or the earliest associated document) to the application decision date.

[0117] Check for time gaps exceeding a preset threshold (e.g., 30 days). If a gap exists, the system will prompt the user that there may be missing process evidence, list the missing time periods, and suggest that the user supplement the relevant data or upload it manually.

[0118] This mechanism effectively avoids the problem of incomplete evidence chains due to accidental data omissions.

[0119] After screening and verifying the evidence data, the system enters the encapsulation stage. The encapsulation process includes: Evidence classification and catalog generation: Based on the core evidence categories of the demand model, the selected evidence data is automatically categorized, and an evidence package catalog is generated. For example, the invention patent evidence package contains the following sections: Chapter 1: Technical Issues Chapter Two: Iteration of Technical Solutions Chapter 3: Experimental Verification Data Chapter Four: Personnel Contribution Records Each chapter lists the specific evidence documents under that category, along with information such as document name, creation date, author, version number, and type of association.

[0120] Packaged evidence documents: The system extracts the original content of evidence files (such as full-text documents, code snippets, and experimental data files) from various data sources and organizes them into folders or compressed packages according to the directory structure. For data that cannot be directly exported (such as experimental records in a database), the system generates a summary report containing key fields and attaches access links or screenshots.

[0121] Metadata embedding and timestamp solidification: When each evidence file is packaged, the system automatically embeds the following information into its metadata: The file's unique identifier in the system; The type of association with the seed and the time of its establishment; The file uploader, upload time, and last modified time; File version number (if multiple versions exist).

[0122] At the same time, the system can call a third-party timestamp service to generate an immutable digital fingerprint and timestamp for the entire evidence package, ensuring the originality of the evidence package after its generation.

[0123] Evidence package format and output: The evidence package is generated as a ZIP compressed file by default, with the internal files retaining their original formats (such as PDF, DOCX, PNG, and TXT). Simultaneously, the system automatically generates an evidence list file (PDF or Excel) containing metadata summaries, directory structures, integrity verification results, and generation timestamps for all evidence. Users can choose to download the evidence package or directly push it to the intellectual property application system (such as the CPC client).

[0124] To address the complexities that may arise in actual applications, the system is designed with a type-driven dynamic adaptation mechanism, allowing users to intervene and adjust during the extraction process: Type Switching: If a user decides to change the application type after generating the evidence package, the system can re-filter the selected evidence data according to the new type's requirement model, retain matching evidence, supplement missing suggestions, and avoid duplicate operations.

[0125] Custom evidence selection: Users can manually select or deselect certain evidence data before the evidence package is generated. The system updates the catalog and integrity verification results in real time and prompts users to identify any gaps in the evidence chain that may result from the custom operation.

[0126] Evidence supplementation suggestions: If there is no data under a certain core evidence category, the system will provide supplementary suggestions based on the demand model, such as "No experimental verification data found, it is recommended to upload the experimental report or add the association".

[0127] S5. Based on the full-process asset map, analyze the technical relevance, team innovation capability, and technology evolution path of the intellectual property seed objects, and generate quantitative analysis results.

[0128] Specifically, a particular intellectual property seed node in the graph is denoted as S. All nodes and edges in the graph that are directly or indirectly connected to this seed constitute the data foundation for the analysis. To support subsequent calculations, the system first extracts the following basic statistical information from the graph: The set of directly connected nodes: all nodes that are directly connected to the seed by an edge, including projects, tasks, people, documents, code submissions, experiment records, etc.

[0129] Indirectly related node set: Other nodes that are further connected through directly related nodes, such as being related to other people through task nodes, or to other projects through document nodes, etc.

[0130] Relationship type distribution: Statistics on edge types between the seed and each node, such as the number of edges "originating from task", the number of edges "content-related", and the number of edges "code implementation".

[0131] Time distribution: The distribution of timestamps of all related nodes (such as document upload time, code submission time, task creation time) on the timeline.

[0132] Based on the above basic data, the system constructs quantitative indicators for three core analytical dimensions.

[0133] Technological relevance measures the closeness and breadth of the connection between the intellectual property and the company's existing technological assets, reflecting the solidity of its technological foundation, its potential for cross-domain applications, and its influence within the internal technology network. The specific analysis process is as follows: The system counts the number of different projects, different tasks, and documents in different technical fields (based on document tags or project categories) directly associated with the seed. The breadth of association score is calculated as follows: Project diversity: This counts the number of projects a seed belongs to, as well as the number of other projects indirectly associated with it through tasks. If a seed is associated with only a single project, its project diversity score is low; if its associated nodes involve multiple projects, its score is high, indicating that the technology has the potential for cross-project application.

[0134] Domain diversity: Cluster all documents associated with the seed using technical field tags (e.g., natural language processing, image recognition, database optimization, etc.), and count the number of documents in different fields. The greater the domain diversity, the more general or interdisciplinary the technology may be.

[0135] Relationship type richness: This counts the number of different relationship types between the seed and associated nodes. The richer the relationship types (e.g., including task outputs, experimental data, and code implementations), the more complete the intellectual property cultivation process and the stronger the evidence chain.

[0136] Building upon the breadth of associations, the system further evaluates the strength of the associations, considering the following factors: Core association weights: For relationships that directly reflect the origin and output of technology, such as "originating from the task" and "task output", higher weights are assigned; for relationships based on semantic matching, such as "content relevance", the weights are adjusted according to the similarity score; for weak associations, such as "temporal proximity", lower weights are assigned.

[0137] Repeated association and version iteration: If there are multiple versions of the same document, and each version is associated with a seed, it indicates that the document has been continuously used or modified in the process of technological development. When the system merges multiple versions for calculation, it increases the intensity coefficient.

[0138] Personnel involvement depth: This counts the number of people involved in the seed-related tasks and their roles in the tasks (responsible person, participant). A large number of responsible persons with important roles indicates deep involvement from the core team.

[0139] The system integrates the above breadth and intensity indicators to generate a "Comprehensive Index of Technological Relevance," expressed as a score from 0 to 100. This index can be used to compare the technological foundation of different seeds.

[0140] If the graph integrates a company's authorized patents or external literature data, the system can also analyze the citation or similarity relationships between the seed and these external nodes to assess its technological influence. For example, by semantic matching, if the seed's technological topic is found to be similar to a highly cited patent, its potential influence can be inferred.

[0141] Team innovation capability measures the innovation activity and collaboration level of the group of people involved in cultivating the intellectual property, reflecting the team's ability to continuously produce high-quality innovative results. Analysis is based on personnel nodes in the graph and their relationships with other nodes.

[0142] Quantification of individual contributions: For each person associated with the seed, the system calculates the following metrics: Number of tasks participated in: The total number of all seed-related tasks that this person has participated in, including the source task and other related tasks.

[0143] Number of documents produced: The number of documents associated with this person as an author or uploader.

[0144] Code commit count: The number of code commits made by this person related to the seed.

[0145] Role weight: If the person is the task leader or seed creator, they are given a higher weight.

[0146] The system calculates a "personal innovation contribution score" for each individual, which reflects their level of investment in this intellectual property cultivation effort.

[0147] Team collaboration network analysis: The system extracts a set of all participants and analyzes their collaborative relationships in other projects or tasks: Collaboration network density: This measures whether there are direct task collaboration relationships (i.e., joint participation in the same task) among these individuals. Widespread collaboration indicates strong teamwork, which is conducive to knowledge sharing and innovation.

[0148] Core-periphery structure: Identifies individuals at the center of a collaborative network (i.e., those who directly collaborate with multiple people) and those on the periphery. Core individuals typically play a key role in driving innovation.

[0149] Cross-project experience: This counts the number of other projects each person has participated in, reflecting the breadth of their experience. If the team as a whole has extensive cross-project experience, it indicates that the team has a strong ability to transfer knowledge.

[0150] Team Innovation Index Generation: The system generates a "Team Innovation Capability Index" by combining individual contribution scores with collaboration network indicators. This index includes two sub-dimensions: Team Commitment: The weighted sum of the individual contributions of all team members, reflecting the team's total commitment to this intellectual property project.

[0151] Team Collaboration: Based on the density of the collaborative network and the proportion of core personnel, it reflects the effectiveness of team collaboration.

[0152] The higher the index, the more likely the intellectual property is backed by a highly invested and collaborative innovation team, and the more likely its technological achievements are to be continuously iterated and transformed into value.

[0153] The technology evolution path aims to reconstruct the entire process of this intellectual property from the emergence of technical problems to the maturity of the solution, identify key nodes and turning points, and provide a reference for subsequent technology layout or continued research.

[0154] Timeline event extraction: The system arranges all related nodes chronologically to form a timeline. Each node is considered an event, with an attached event type (task creation, document upload, code submission, experiment record, etc.). The system automatically labels the following key events: Technical issue raising event: usually the creation time of the source task associated with the seed, or the upload time of the earliest associated document (such as meeting minutes).

[0155] Solution iteration events: the upload time of multiple versions of documents and the code submission time, especially documents with tags such as "solution" and "design".

[0156] Verification event: the upload time of experimental records and test reports.

[0157] Decision event: If the system records a change in the seed status (such as from "under cultivation" to "under application"), it is considered a decision node.

[0158] Evolution path identification: Based on a timeline, the system identifies the trajectory of technological evolution by analyzing the referencing relationships between nodes (such as document version updates and tasks associated with code submissions) and content similarity. For example: Main Path: Starting with the earliest event, each subsequent event is chosen as the next node, selecting the one with the highest similarity or closest relationship to the current event, thus forming a main evolutionary path. This path typically reflects the iterative process of technology from a rough idea to a concrete implementation.

[0159] Branching paths: If multiple similar subsequent events with different directions occur after a certain event, a branch is formed, representing that the technology may diverge into attempts in different directions (such as parallel exploration of algorithm A and algorithm B).

[0160] The system presents these paths in the form of a timeline, and marks the event type and brief description at key nodes.

[0161] Key node identification: The system automatically identifies key nodes in the evolution path, based on the following criteria: Turning point: A significant change in the technical direction after the event (such as shifting from theoretical solutions to experimental verification), or the emergence of a new topic after a sudden drop in similarity scores.

[0162] Milestone: The event produces important documents (such as the final technical solution) or passes review (such as the task status changing to "completed").

[0163] Bottleneck: A prolonged period of no progress (large time gap) after an event may indicate that the technology has encountered difficulties or stagnated.

[0164] Technology trend forecast: If the evolutionary paths of multiple seeds are accumulated in the graph, the system can perform pattern mining to predict the possible development direction of the current seed. For example, by analyzing what types of intellectual property rights historical seeds typically choose to apply for after reaching a certain stage, or in which modules subsequent technological improvements are concentrated, the system can provide a reference for decision-makers.

[0165] The analysis results from the above three dimensions are presented together in a single "Intellectual Property Value Analysis Report," which includes: Technology relevance: comprehensive index, project diversity, field distribution, and list of key related nodes.

[0166] Team innovation capability: team index, core personnel list, collaboration network diagram.

[0167] Technology evolution path: timeline main line diagram, key event annotation, path description.

[0168] Decision recommendations: Based on the analysis results, the system automatically generates preliminary recommendations, such as "This technology is associated with multiple projects, and it is recommended to consider a patent portfolio layout" and "The team has strong innovation capabilities, and continuous output can be encouraged."

[0169] The report can be exported as PDF or Excel, and can also be viewed interactively in the system (e.g., clicking on a node will jump to the original data).

[0170] The second embodiment of this application is as follows: Please see Figure 4 This invention provides an intellectual property traceability management system, applied to an intellectual property traceability management method as provided in the first embodiment, comprising: The seed management module 101 is used to mark the set task nodes during the R&D project management process, and automatically generate intellectual property seed objects based on the marking operation, and record the type, creator, creation time, project and task node of the intellectual property seed object; The space construction module 102 is used to create a virtual cultivation space for each intellectual property seed object, automatically associate the project information, participant information and process data generated during task execution of the intellectual property seed object, and form a preliminary association map with the intellectual property seed object as the core. The dynamic association module 103 is used to continuously monitor newly generated data during the R&D process, establish association relationships between new data and existing intellectual property seed objects through active or manual association, and dynamically update the whole-process asset map with the intellectual property seed objects as the core. The evidence encapsulation module 104 is used to extract matching full-process evidence data from the cultivation space based on the intellectual property application instructions and according to the type of intellectual property applied for, and encapsulate it into an intellectual property creation process evidence package containing timestamps and personnel contribution records. The value analysis module 105 is used to analyze the technological relevance, team innovation capability, and technological evolution path of the intellectual property seed objects based on the full-process asset map, and generate quantitative analysis results.

[0171] Regarding the system in the above embodiments, the specific ways in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0172] For the system embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0173] Accordingly, this application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; and, when the one or more programs are executed by the one or more processors, causing the one or more processors to implement the intellectual property traceability management method described above. Figure 5 The diagram shown is a hardware structure diagram of any device with data processing capabilities within the intellectual property traceability management system provided in an embodiment of the present invention, except... Figure 5 In addition to the processor, memory, and network interface shown, any data processing device in the embodiment may also include other hardware depending on the actual function of the data processing device, which will not be described in detail here.

[0174] Accordingly, this application also provides a computer-readable storage medium storing computer instructions thereon, which, when executed by a processor, implement the intellectual property traceability management method described above. The computer-readable storage medium can be an internal storage unit of any data-processing device as described in any of the foregoing embodiments, such as a hard disk or memory. The computer-readable storage medium can also be an external storage device, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc., equipped on the device. Furthermore, the computer-readable storage medium can include both internal storage units of any data-processing device and external storage devices. The computer-readable storage medium is used to store the computer program and other programs and data required by the data-processing device, and can also be used to temporarily store data that has been output or will be output.

[0175] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0176] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method for managing the traceability of intellectual property rights, characterized in that, Includes the following steps: During the R&D project management process, the set task nodes are marked, and intellectual property seed objects are automatically generated based on the marking operations. The type, creator, creation time, project and task node of the intellectual property seed object are recorded. A virtual cultivation space is created for each intellectual property seed object, and the project information, participant information, and process data generated during task execution of the intellectual property seed object are automatically associated to form a preliminary association map with the intellectual property seed object as the core. Continuously monitor newly generated data during the R&D process, establish associations between the new data and existing intellectual property seed objects through proactive or manual association, and dynamically update the full-process asset map with the intellectual property seed objects as the core. Based on the intellectual property application instructions, and according to the type of intellectual property applied for, matching full-process evidence data is extracted from the cultivation space and packaged into an intellectual property creation process evidence package containing timestamps and personnel contribution records. Based on the full-process asset map, the technological relevance, team innovation capability, and technological evolution path of the intellectual property seed objects are analyzed, and quantitative analysis results are generated.

2. The intellectual property traceability management method as described in claim 1, characterized in that, During the R&D project management process, designated task nodes are marked, and intellectual property seed objects are automatically generated based on the marking operations. The type, creator, creation time, project, and task node of each intellectual property seed object are recorded, including: In the R&D project management platform, several task nodes are established for each R&D project. Each task node has a unique identifier and is associated with its respective project. Based on the user's operation on the marker control, the seed type and technology topic selected by the user are obtained; The system automatically captures the context information of the current task node, including the task name, the project it belongs to, the person in charge, and the start and end times; Based on the seed type, technology topic, and automatically captured context information filled in by the user, an intellectual property seed object is created in the database, and a unique seed number is assigned to the intellectual property seed object. At the same time, the intellectual property seed object is established with the task node of the trigger mark to establish a "source" relationship, and the association time and associated person are recorded.

3. The intellectual property traceability management method as described in claim 1, characterized in that, A virtual cultivation space is created for each intellectual property seed object, automatically linking it to the project information, participant information, and process data generated during task execution, forming a preliminary association map centered on the intellectual property seed object, including: Simultaneously with the generation of the seed object, a virtual cultivation space instance uniquely corresponding to the intellectual property seed object is automatically constructed; Based on the source task identifier recorded in the intellectual property seed object, retrieve all information of the task node and its attached documents and comment records from the task database, and import them into the cultivation space as initial association data; Based on the project identifier of the source task, retrieve the basic information of the project and import it into the cultivation space; Identify the list of participants in the source task, obtain other output data of the participants within a preset time window, and import them into the cultivation space after filtering by keyword similarity calculation; For each imported piece of related data, record its relationship type with the seed object, as well as the establishment time and method, to form a preliminary association graph with the intellectual property seed object as the core. The relationship types include project affiliation, task origin, task output, personnel contribution, content relevance, and time proximity.

4. The intellectual property traceability management method as described in claim 1, characterized in that, Continuously monitor newly generated data during the R&D process, and establish associations between the new data and existing intellectual property seed objects through proactive or manual methods. Dynamically update the entire process asset map centered on these intellectual property seed objects, including: Scan for data changes in the document management library, code repository, task management system, and experimental data platform in real time or on a scheduled basis; When new data is detected, text extraction, word segmentation and cleaning, and keyword extraction are performed on the new data to generate feature vectors; The similarity between the feature vector and the technology topic vector of the intellectual property seed object in the cultivation state is calculated. The technology topic vector is dynamically updated according to the technology topic when the seed is created and subsequent associated documents. When the similarity exceeds a preset threshold, a correlation suggestion is generated and pushed to the cultivation space of the corresponding seed object; Based on the user's confirmation of the association suggestion, an association record between the new data and the seed object is established in the relational database, recording the relationship type, confirmation time, confirmer, and similarity score, and the new data is included in the cultivation space.

5. The intellectual property traceability management method as described in claim 4, characterized in that, The method further includes: Nodes and edges are constructed based on a graph database. The node types include intellectual property seeds, projects, tasks, personnel, documents, code submission records, and experiment records. The edge types represent the relationships between nodes, and each edge is accompanied by the creation time, creation method, and operator attributes. Whenever a new relationship is established, check if the new data node already exists in the graph. If it does not exist, create the node and create an edge of the appropriate type between the seed node and the new data node to build a full-process asset graph centered on the seed object.

6. The intellectual property traceability management method as described in claim 1, characterized in that, Based on the intellectual property application instructions, and according to the type of intellectual property applied for, matching full-process evidence data is extracted from the cultivation space and packaged into an intellectual property creation process evidence package containing timestamps and personnel contribution records, including: A mapping model between intellectual property types and evidence requirements is pre-stored, wherein the model defines the core evidence categories, data sources, relationship types, and priorities required for each type of intellectual property. Obtain the intellectual property application type selected by the user, traverse the associated data in the nurturing space according to the mapping model, and filter by data type matching, relationship type matching, and priority sorting; The screened evidence data is checked for timeline integrity to detect whether there is a time gap exceeding a preset threshold between the first occurrence of the technical problem and the application decision date. If there is a gap, the user is prompted to supplement it. The filtered evidence data is automatically categorized into core evidence categories and a catalog is generated. The original files are extracted from each data source and packaged. Metadata information is embedded in each evidence file, including file identifier, association type, uploader, upload time, and version number. Generate a digital fingerprint and timestamp for the entire evidence package, and output it as a compressed file with an evidence list.

7. The intellectual property traceability management method as described in claim 6, characterized in that, The method further includes: When a user switches the intellectual property application type after generating an evidence package, the selected evidence data is re-filtered according to the new type's mapping model, retaining matching evidence and prompting for missing core evidence categories; If evidence data is manually selected or unselected before the evidence package is generated, the catalog will be updated in real time and a warning will be given about potential gaps in the chain of evidence caused by custom operations. If there is no data under a certain core evidence category, supplementary suggestions are provided based on the mapping model to guide users to upload or associate data of the missing type.

8. The intellectual property traceability management method as described in claim 1, characterized in that, Based on the aforementioned full-process asset map, an analysis of the technological relevance of the intellectual property seed objects is performed, including: Count the number of different projects, different tasks, and different technical documents directly associated with the seed object, and calculate the breadth of association; Each association is assigned a weight based on its type, with core association types having a higher weight than weak association types. Duplicate associations and version iterations are counted to calculate the association strength. A comprehensive index of technical correlation is generated by combining the breadth and strength of correlation, which is used to compare the technical foundation of different seed objects horizontally.

9. The intellectual property traceability management method as described in claim 1, characterized in that, Based on the aforementioned full-process asset map, an analysis of the technological evolution path of the intellectual property seed objects is performed, including: Arrange all nodes associated with the seed object in chronological order to form a timeline, with each node representing an event and labeling the event type. By analyzing the reference relationships and content similarity between nodes, the main path and branch paths of technological evolution are identified. The main path starts from the earliest event, and each time the event with the highest content similarity or the closest relationship with the current event is selected as the next node. Identify key nodes in the evolution path, including turning points where the technology direction changes, milestones that produce significant results, and bottlenecks where there is no progress for a long time, and generate an evolution path description.

10. An intellectual property traceability management system, applied to the intellectual property traceability management method as described in claim 1, characterized in that, include: The seed management module is used to mark the set task nodes during the R&D project management process, and automatically generate intellectual property seed objects based on the marking operations, and record the type, creator, creation time, project and task node of the intellectual property seed object; The space construction module is used to create a virtual cultivation space for each intellectual property seed object, automatically associate the project information, participant information and process data generated during task execution of the intellectual property seed object, and form a preliminary association map with the intellectual property seed object as the core. The dynamic association module is used to continuously monitor newly generated data during the R&D process. Through active or manual association, it establishes association relationships between new data and existing intellectual property seed objects, and dynamically updates the full-process asset map with the intellectual property seed objects as the core. The evidence encapsulation module is used to extract matching full-process evidence data from the cultivation space based on the intellectual property application instructions and according to the type of intellectual property applied for, and encapsulate it into an intellectual property creation process evidence package containing timestamps and personnel contribution records; The value analysis module is used to analyze the technological relevance, team innovation capability, and technological evolution path of the intellectual property seed objects based on the full-process asset map, and generate quantitative analysis results.

Images