Embedded path evolution tracking system and implementation method thereof

Abstract

The invention discloses an embedded path evolution tracking system and an implementation method thereof, and aims to solve the technical problems of incomplete path record, difficulty in problem positioning and insufficient experience inheritance in innovation process management. The system is embedded in innovation platforms such as DIKWP-TRIZ and comprises a path data acquisition module, a log storage module, an organization visualization module, a path analysis module, an improvement suggestion module and a learning optimization module. Key events such as cognitive engine reasoning jump, TRIZ principle application and user decision are intercepted in real time, standardized path units are generated and stored according to a time sequence, and a multi-level evolution path chain and a visual network diagram are constructed. The path analysis module can diagnose and infer breaking points and unresolved contradictions, the improvement suggestion module prompts introduction of new knowledge or adjustment strategies according to the contradictions, and the learning optimization module dynamically optimizes a scheduling algorithm based on a historical data training model. According to the method, full-path transparent recording and intelligent analysis of the innovation process are achieved, efficient redisk, accurate problem positioning and strategy optimization are supported, the innovation success rate and knowledge management efficiency are remarkably improved, and the method is suitable for the fields of research and development management, patent mining and the like.

Description

Technical Field

[0001] This invention belongs to the fields of innovation process management and artificial intelligence technology, and specifically relates to an embedded path evolution tracing system and its implementation method. More specifically, this invention relates to embedding path evolution tracing technology in the DIKWP-TRIZ innovation platform for recording and analyzing the entire process from problem formulation to solution generation. Background Technology

[0002] The innovation process is often a complex process of iterative iteration and trial and error. In current technologies, innovation teams typically rely on manual records, meeting notes, or scattered documents to track the evolution of ideas. This unsystematic management approach has several shortcomings:

[0003] Difficulty in complete retrospective analysis: Due to the lack of unified tracking tools, a large amount of intermediate data, changes in thinking, and rationale for decisions are often not fully recorded, making it difficult to reconstruct how the solution evolved step by step. This is especially true in scenarios involving AI-assisted innovation (such as the application of TRIZ invention principles and cognitive engine reasoning), where intermediate reasoning steps are even more difficult to capture manually.

[0004] Difficulty in pinpointing the problem: When the innovation process is hindered or fails, there is a lack of evidence to determine where the problem lies. Teams often rely solely on experience and reflection, unable to accurately pinpoint which decision-making error or unresolved key conflict led to the failure. The lack of tracking and analysis of the innovation path also makes it impossible to promptly identify and correct deviations in the process.

[0005] Unclear intellectual property ownership: In collaborative innovation, the lack of clear process records makes it difficult to attribute creative contributions. When applying for patents or reporting results, it is impossible to prove who proposed a specific idea and when, increasing the risks of intellectual property management.

[0006] Experience is difficult to pass on: New employees find it hard to systematically learn from the innovative cases of predecessors because they can only see the final results, not the complete thought process and intermediate decisions. Existing methods lack the means to transform successful or unsuccessful innovation paths into visual teaching materials, which is detrimental to team capability improvement.

[0007] In summary, current innovation management lacks an embedded technological approach to automatically record and analyze the entire innovation process, supporting post-event review, problem diagnosis, and experience accumulation. This has become a major technological bottleneck restricting the efficiency and success rate of complex innovation projects. Summary of the Invention

[0008] The main objective of this invention is to overcome the shortcomings of the prior art and provide an embedded path evolution tracking system and its implementation method. This system can record a large amount of intermediate data and decision-making steps in the innovation process in real time within an innovation platform, and organize and analyze them to form a complete innovation path log. With this system, it is possible to easily review the innovation process, locate process bottlenecks, protect intellectual property rights, and guide future innovation strategies, significantly improving the transparency and efficiency of innovation activities.

[0009] To achieve the above objectives, this invention provides an embedded path evolution tracing system, characterized by being embedded in various modules of innovation support platforms such as DIKWP-TRIZ, for tracking and recording path evolution information throughout the entire process from problem formulation to solution development. The system includes the following modules:

[0010] The path data acquisition module is deployed across platform components such as the cognitive engine, path scheduling algorithm, TRIZ engine, and user interface. It intercepts various key events in real time and formats them into "path unit" records. These events include, but are not limited to: reasoning transitions on the semantic graph (e.g., the reasoning steps of the cognitive engine jumping from a knowledge node to another possible solution), inventive principle events applied by the TRIZ engine (e.g., the operation of applying an inventive principle to resolve a contradiction), and user decision-making interventions on the interface (e.g., the user selecting a solution path or inputting new constraints). Each event is abstracted into a path unit with a unified format, containing metadata such as the event's timestamp, event type, involved knowledge nodes or principles, and decision content.

[0011] Path log storage module: This module writes the aforementioned path units to the storage medium in chronological order, accumulating them to form a complete path log. This module ensures the temporal consistency and data integrity of the records, enabling the linking of all path units generated from the start to the end of an innovative project. Log data can be stored in a relational database, graph database, or distributed log system for subsequent retrieval and processing.

[0012] The Path Organization and Visualization module organizes path units according to chronological order and semantic relationships, constructing multi-level innovation path chains and generating visual representations. This module analyzes the relationships between path units (e.g., which reasoning steps belong to the same branch, which solutions were abandoned), organizing the path structure into a tree-like or network-like topology. The visualization result can be an evolutionary tree or network diagram, intuitively showing the complete path from the initial problem information through the data-knowledge-wisdom layer to the final solution, as well as the branch explorations and decision-making choices encountered along the way.

[0013] The path analysis module is used for intelligent analysis of path logs, especially to diagnose the reasons for bottlenecks in the innovation process when innovation has not yet succeeded or problems have not been resolved. This module includes an algorithm (such as a bug identification algorithm) that can automatically detect whether there are breaks in the reasoning chain (a key reasoning step has not progressed further) or unresolved contradictions in the path chain (identifying nodes where TRIZ contradictions have not been resolved). Once such problematic nodes are found, the analysis module will generate prompts indicating which step the innovation process may be stuck at and the specific bottleneck.

[0014] Improvement Suggestion Module: Based on the diagnostic results of the path analysis module, this module provides targeted improvement suggestions to users or the system. For example, if it detects that the knowledge required for a certain step is missing in the existing semantic graph, the system will suggest introducing new external knowledge to fill the gap; if it detects that the problem statement is ambiguous or contradictory, the system will suggest adjusting the understanding of the innovative intent or trying different inventive principles. Simultaneously, this module can highlight the corresponding path units on the user interface for easy location and modification by the user.

[0015] Path Learning Optimization Module (Optional): This module aggregates and analyzes a large amount of historical innovation path logs and trains machine learning algorithms to extract path patterns that influence the success or failure of innovation. By comparing the evolutionary paths of different problems, this module can identify which path patterns are more likely to lead to success and which decision sequences tend to cause failure. Based on this, the system can automatically optimize future path scheduling strategies. For example, it can adjust the algorithm weights of the cognitive engine in selecting the next inference jump on the semantic graph, prioritize trying path branches with higher historical success rates, and continuously improve the efficiency and success probability of the innovation process.

[0016] Compared with existing technologies, the embedded path evolution tracing system of the present invention has significant practical value and technical advantages:

[0017] Systematic end-to-end recording: This invention provides a "black box"-like function, recording every key event in the innovation process to form a traceable and tamper-proof innovation path log. This not only ensures the clear attribution of intellectual property rights and creative contributions (every creative point is verifiable), but also provides complete data support for post-event analysis.

[0018] Rapid Problem Diagnosis: Through automatic analysis of path logs, this invention can promptly identify bottlenecks or erroneous decisions in the innovation process. When innovation encounters obstacles, the system can accurately pinpoint the problematic step (such as a break in a reasoning chain or an unresolved contradiction), significantly improving the speed and accuracy of problem location compared to manual investigation.

[0019] Path Visualization and Retrospective Analysis: The multi-level evolutionary tree / network diagram generated by this invention concretizes and visualizes the abstract innovation process. Team members can intuitively see how innovative ideas evolve, which solutions were considered along the way, and why they were abandoned, facilitating retrospective analysis after the project ends. This visualized path can also be used to train new employees, transforming the innovation methodology into a transferable process asset.

[0020] Intelligent Decision Support: By learning patterns of success and failure from a large number of cases, the path learning optimization module of this invention can continuously improve the decision-making strategy of the innovation algorithm, making the system increasingly "intelligent." This is equivalent to using collective experience and lessons learned to optimize the inference scheduling of AI, making future innovation processes more efficient and with a higher probability of success.

[0021] Seamless integration and low interference: This system is embedded within various functional modules of the innovation platform, recording data through passive listening and event capture, without interfering with the user's normal innovation workflow. Furthermore, the use of lightweight logging and analysis algorithms ensures high-performance system operation without significantly increasing the platform's workload.

[0022] In summary, this invention provides a novel technical means that makes the innovation process transparent, controllable, diagnosable, and optimizable, bringing great convenience and value to enterprise R&D and intellectual property management. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the innovative path evolution tree generated by the system of this invention.

[0024] Figure 2 This is a schematic diagram of the innovative process path tracing method according to an embodiment of the present invention. Detailed Implementation

[0025] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to explain the present invention and are not intended to limit the scope of the present invention.

[0026] Example 1: Embedded Path Evolution Tracing Method

[0027] This embodiment provides a path evolution tracing method embedded in the DIKWP-TRIZ innovation platform, such as... Figure 2 As shown, it includes the following steps:

[0028] Event interception and path unit recording: When a critical event occurs within the innovation platform, the path data acquisition module is immediately triggered to record the event as a path unit. Specifically:

[0029] Reasoning jump event: Whenever the cognitive engine or path scheduling algorithm jumps from one knowledge node to another potential solution node on the platform's semantic graph, it records the path unit that generates the "reasoning jump" type, including information such as "jump from node A to node B (reason: satisfying a certain reasoning rule)", and records the timestamp and related algorithm parameters.

[0030] Principle Application Event: Whenever the TRIZ engine applies an inventive principle or innovation rule to attempt to resolve the current technical contradiction, it records a path unit of type "Principle Application", which includes information such as: "Applying inventive principle #5 (e.g., 'merging principle') to resolve contradiction X", and attaches the contradiction's identifier and the application result (whether it was successfully resolved).

[0031] User Decision Events: Whenever a user actively intervenes on the innovation platform interface (e.g., selects a recommended solution, modifies problem parameters, or introduces new resources), a path unit of the "User Decision" type is recorded, containing information such as "User introduces data from external database Y" or "User rejects solution C and tries solution D instead", along with metadata such as time and user operation description.

[0032] Each path unit is formatted into a uniform data structure, including: event type, occurrence time, involved objects (knowledge nodes, principles, or operational objects, etc.), executing entity (which module or user), and additional descriptions or parameters. This standardized format facilitates the integrated processing of different types of events.

[0033] Path Log Storage and Chaining: The path data acquisition module transmits each generated path unit to the path log storage module in real time. The storage module writes the path units sequentially into the log database according to the order in which events occur. For concurrent events, they can be sorted according to timestamp precision or predetermined causal relationships. In this way, all path units are chained together to form a complete time series chain, recording every step of the innovation process. At the same time, the storage module can also add indexes or link information to the log based on the semantic association of events (e.g., marking which events belong to the same branch of exploration), supporting subsequent reconstruction of the path structure.

[0034] Path Structure Reconstruction and Visualization: When an innovation process ends (a final solution is obtained or the problem is abandoned), the path organization and visualization module extracts all path unit records of that process from the log database and organizes them into an evolutionary path chain structure according to time sequence and association indexes. This structure can be represented as a tree or a directed acyclic graph.

[0035] The root node of a tree / graph represents the initial problem or initial information, and the leaf nodes represent the final solutions obtained (multi-leaf nodes can also be formed if there are multiple candidate solutions).

[0036] The nodes in the middle represent intermediate states or milestones in the innovation process (such as new ideas introduced at a certain step, or new knowledge points introduced). The arrows connecting the nodes represent the evolution of reasoning or decision-making from the previous step to the next.

[0037] If the process involves branching exploration, multiple branching arrows will appear from a certain node, indicating that the team or system has tried multiple alternative paths in parallel. These branches may eventually converge (to a successful solution) or terminate (an abandoned attempt).

[0038] The visualization module generates an evolutionary tree / network graph based on the above structure (e.g., Figure 1 (As shown in the illustration). Different shapes and colors are used to distinguish different types of path units: for example, reasoning jumps in the cognitive engine are represented by circular nodes, TRIZ principle applications by square nodes, and user decisions by triangular nodes, etc. Furthermore, nodes at different levels (data layer, knowledge layer, wisdom layer) can be drawn on different horizontal planes, allowing the evolutionary path to exhibit characteristics that span abstract levels vertically. The final visualization can be interactively presented on the platform interface, allowing users to click on nodes to view detailed records, etc.

[0039] Path Analysis (Success Scenario): For innovative paths that have successfully yielded solutions, the path analysis module can further calculate and display some statistics and characteristics of the path, such as the total number of iterations, the number of branches generated, and the average time per step. This information helps the team evaluate innovation efficiency. Simultaneously, the analysis module can automatically extract patterns from the successful path, marking key decision points (e.g., the application of a certain principle significantly changed the direction, or a user decision played a crucial role). These key points can be used later to refine successful experiences.

[0040] Path Analysis (Unresolved Cases): If the innovation process ultimately fails to yield a solution (the problem remains unresolved), the path analysis module initiates in-depth diagnostic algorithms to pinpoint the problem. This algorithm, similar to a debugging tool, progressively examines each node and branch of the path chain, looking for the following signs:

[0041] Reasoning chain break: This occurs when a step is detected where there is no further progress or other branches to follow. For example, if an intermediate node in the path chain is found where all its child branches stop and no solution is reached, then this node may be the breakpoint causing the stagnation. The analysis module will further retrieve the context of this node to determine whether the stagnation is due to a lack of necessary information or overly strict constraints that prevent the formation of a feasible path.

[0042] Unresolved contradictions: If the innovation involves the TRIZ method, a new technical contradiction may have been introduced at some point in the process but has not yet been resolved. The analysis module checks for contradictions that have not been successfully resolved by any inventive principle based on the contradiction identifiers recorded in the path unit. If a contradiction is found to persist and hinder subsequent steps, it is determined that the contradiction is unresolved, which is one of the reasons for failure.

[0043] Once the above issues are detected, the path analysis module generates a diagnostic report. For example: "A reasoning break occurred in step 5, possibly due to a lack of knowledge for continuing reasoning from node B. It is recommended to introduce new knowledge about the XX domain" or "An unresolved 'performance vs. security' conflict exists in step 7. It is recommended to try different TRIZ principles (such as principle #12)." These diagnostic results will be presented to the user through the improvement suggestion module.

[0044] Improvement suggestion output: The improvement suggestion module obtains diagnostic information from the analysis module and provides users with actionable suggested solutions. For example:

[0045] In cases of missing knowledge, the system will suggest that users search for and import new information or data sources in the relevant field, and expand the corresponding nodes in the semantic graph to provide support for further attempts.

[0046] If the contradiction remains unresolved, the system may prompt the user to adopt different innovation strategies, such as referencing other TRIZ invention principles, redefining the problem to circumvent the contradiction, or reducing certain constraints.

[0047] If a user's decision is diagnosed as potentially flawed (e.g., prematurely abandoning a promising branch), the system will highlight the decision node in the path visualization interface and suggest that the user re-examine the rationale behind the decision and consider resuming or re-exploring the branch.

[0048] The suggestion module can present these suggestions on the platform front end in the form of dialog boxes, prompt cards, etc., and provide "one-click operation" to execute some suggestions (such as automatically importing the recommended knowledge base). Through these improvement guidelines, users can adjust their innovation direction in a targeted manner and increase the likelihood of problem solving.

[0049] Path data accumulation and learning optimization: Once innovative path logs for multiple different problems have accumulated in the system, the path learning optimization module begins to function. It utilizes machine learning and data mining techniques to find patterns and regularities from historical paths.

[0050] The system extracts feature vectors for each path (including step sequence, set of principles used, number of branches, frequency of user intervention, etc.) and correlates them with innovation results (success / failure) to form a path-result sample set.

[0051] By training classification or sequence models (such as decision trees, sequence pattern mining algorithms, or deep learning models) on this sample set, the module identifies path patterns that significantly influence the results. For example, it discovers patterns such as "successful cases often resolve key conflicts early on" or "failed cases often remain at a certain level for a long time without progress."

[0052] Based on these patterns, the optimization module automatically updates the strategy parameters of the path scheduling algorithm in the innovation platform. For example, it increases the priority of known high-success-rate step sequences, or promptly prompts the user to change the strategy when a certain failure symptom is detected.

[0053] Through the above learning mechanism, the system of the present invention can self-evolve: as time goes by, by referring to a large number of past experiences and lessons, the system's reasoning and decision-making become more intelligent, avoiding repeating past failures and greatly improving innovation efficiency and success probability.

[0054] The path evolution tracking system provided in this embodiment can be integrated into an innovation support platform as software, or it can be implemented in conjunction with hardware sensors (such as user operation log capture devices). Those skilled in the art can develop corresponding software modules or hardware circuits to perform the various functions based on the above scheme. The system and method of this invention provide powerful recording, analysis, and navigation support for the innovation process without affecting the user's normal innovation work, and can be widely applied in fields such as enterprise R&D management, patent mining, and product design optimization.

Claims

1. An embedded path evolution tracing system, characterized in that, Embedded within the innovation support platform, it tracks the innovation path in real time throughout the entire process from problem identification to solution generation. This includes: a path data acquisition module, a path log storage module, a path organization and visualization module, a path analysis module, and an improvement suggestion module; among which, The path data acquisition module is configured to intercept key events generated by each component of the innovation support platform and format them into path unit records. The events include at least: semantic reasoning jump events of the cognitive engine, invention principle application events of the TRIZ engine, and user decision intervention events. The path log storage module stores the path units in chronological order of events and connects them to form a complete innovation path log. The path organization and visualization module organizes path units into multi-level evolutionary path chains according to time sequence and semantic association, and generates corresponding path evolution trees or network graphs for display. The path analysis module analyzes the path chain at the end of the innovation process: extracts path features for cases where a final solution is generated, and automatically detects whether there are breakpoints in the reasoning chain or unresolved technical contradictions in the path chain for cases where problems remain unsolved. The improvement suggestion module generates targeted improvement suggestions based on the path analysis results. When a breakpoint or unresolved contradiction is detected in the innovation path, it suggests introducing new knowledge or adjusting the innovation strategy and provides feedback to the user on the platform interface.

2. The embedded path evolution tracing system according to claim 1, characterized in that, The path data acquisition module uniformly formats different types of events into path units that include timestamps, event types, event source components, relevant knowledge nodes or principle identifiers, user operation content, and additional parameters, so as to ensure that innovation process data from different sources can be uniformly stored and processed.

3. The embedded path evolution tracing system according to claim 1, characterized in that, In the evolutionary path diagram generated by the path organization and visualization module, different types of path units are identified in different ways and displayed in layers according to their respective abstraction levels. This makes the structure of the evolutionary path spanning the data layer, knowledge layer, and wisdom layer clearly visible, and the diagram also marks the direction of each branch of exploration and its merging or termination status.

4. The embedded path evolution tracing system according to claim 1, characterized in that, The path analysis module includes an algorithm for identifying process bottlenecks. When the innovation process fails to find a solution, the algorithm detects broken steps or unresolved contradictions in the evolutionary path chain and generates diagnostic information characterizing the bottleneck.

5. The embedded path evolution tracing system according to claim 1, characterized in that, It also includes a path learning optimization module, which is used to collect path data of multiple innovation projects stored in the path log storage module, use machine learning algorithms to analyze the relationship between different path patterns and innovation success rates, and automatically adjust the path scheduling strategy parameters in the innovation support platform to optimize the inference path selection in the future innovation process.

6. An embedded path evolution tracing method, characterized in that, This is applied to innovation support platforms to track and record the entire process of innovation problems from their inception to resolution, including the following steps: (1) Intercept key events generated by each module in the innovation process in real time, format the event information into path units and record them in the path log, wherein the events include at least three types: reasoning jump, application of invention principle and user decision intervention; (2) The recorded path units are connected in chronological order to form an innovation path chain, which represents the evolution process and branch structure of the innovation process; (3) At the end of the innovation process, a visualized multi-level evolution path diagram is constructed based on the path chain to show the evolution trajectory from the initial problem to the final solution. (4) If the innovation process successfully yields a solution, extract the key features and node information in the path chain and output them; if the innovation process fails to yield a solution, analyze the path chain to detect reasoning breakpoints or unresolved contradictions, and generate a corresponding diagnostic report. (5) Provide the user with improvement suggestions based on the diagnostic report, including suggestions such as introducing external knowledge, adjusting the problem definition or changing the innovation strategy, to help the user modify the innovation path and try again.

7. The embedded path evolution tracing method according to claim 6, characterized in that, Also includes: By aggregating data from multiple innovation path chains and their outcomes, a machine learning model is trained to identify path patterns that lead to innovation success and those that lead to failure. Based on the analysis results of the model, the decision parameters of the path scheduling algorithm in the innovation support platform are dynamically optimized, thereby improving the efficiency and success probability of the subsequent innovation process.

Images