Intellectual Property Case Management System

Abstract

An intellectual property case management system comprising: a receiving module that receives electronic data relating to at least one intellectual property case; a process generation module that is signal-connected to the receiving module and generates a control process set according to a control logic, wherein the control process set comprises a plurality of control processes; and a management module that is signal-linked to the process generation module and performs a marking for each control process, wherein the marking is a case type marking, a priority marking and an execution status marking, and the execution order of the control processes is dynamically adjusted according to the markings of the control processes.

Description

Field of the invention

[0001] The present invention relates to the technical field of process management. State of the art

[0002] Traditional intellectual property case management practices often face challenges such as low efficiency, high risk, collaboration difficulties, and a lack of data support, and are unable to meet the demands of modern businesses for efficient, accurate, and coordinated management. These challenges require innovative solutions to comprehensively improve case management efficiency and enhance companies' competitiveness in the highly competitive market. Object of the invention

[0003] The present invention provides a system for managing intellectual property cases, comprising: a receiving module, a process generation module, and a management module, wherein the receiving module receives electronic data relating to at least one intellectual property case, the process generation module is signal-connected to the receiving module and generates a control process set in accordance with a control logic, the control process set comprises a plurality of control processes, the management module is signal-connected to the process generation module and performs a marking for each control process, the marking being a case type marking, a priority marking, and an execution status marking, and the execution order of the control processes is dynamically adjusted in accordance with the markings of the control processes. Brief description of the drawings Fig. 1, Fig. 3 and Fig. 4 shows schematic block diagrams according to an embodiment of the present invention; Fig. 2 shows a structural diagram of the control process set according to an embodiment of the present invention; Fig. 5 shows a flowchart for case management according to an embodiment of the present invention. Detailed description of the implementation examples

[0004] In the example of Fig. 1, Fig. 3 and Fig. 4, the intellectual property case management system 100 represents a data processing apparatus composed of multiple specific circuit modules and memory interfaces. This apparatus is a technical structure realized by a physical device rather than a simple algorithm or logic flow.

[0005] The system 100 includes a receiving module 110, a process generation module 120, a management module 130, an update module 140, an analysis module 150, a database 160, and a comparison module 170. These modules are implemented as hardware modules that can be physically arranged in a device and each have explicit electrical connections and data transmission functions. These modules can be integrated into a microprocessor (MCU), a server, or an embedded device with a central processing unit (CPU), a memory, an internal bus, and multiple input / output interfaces. The modules are each assigned to specific memory segments and processor subunits to execute specific instructions.

[0006] Here, the receiving module 110 may be a data receiving unit with network input and cache memory. It is located at the front end of the system and is responsible for receiving electronic data related to intellectual property cases and transferring them to the storage. The process generation module 120 and the management module 130 serve as core control elements and, based on control logic stored in the memory, generate a control process set in which the control processes are prioritized according to the assigned labels. The updating module 140 represents an output control element that monitors the process status and issues commands for subsequent operation to external devices (e.g., notification modules or report output units).

[0007] Furthermore, the analysis module 150 and the comparison module 170 belong to the data acquisition element and the comparison circuit element, respectively. The analysis module 150 uses an OCR (optical character recognition) unit, keyword recognition logic, and a semantic inference engine to extract identification information from the electronic data. The comparison module 170 performs a multi-column data comparison with the database 160 and, based on a calculated confidence index, decides whether the information should be archived, a warning issued, or the data ignored. The modules exchange data with the memory via the internal data bus, whereby control signals and execution results can also be output via standardized output interfaces (e.g., USB, HDMI, and Ethernet).

[0008] In the following, an embodiment of the present invention will be described with reference to the modules of the system. Reference will first be made to Fig. 1 referred to. Fig. 1 shows a first schematic block diagram according to an embodiment of the system for managing intellectual property cases according to the invention.

[0009] First, the intellectual property case management system 100 receives and stores electronic data related to intellectual property cases via the receiving module 110. This data can originate, for example, from application documents submitted by customers, official notices, or email correspondence, and is usually in unstructured text form. To extract useful information, the receiving module 110 applies optical character recognition technology such as NLP (natural language processing).

[0010] Specifically, pre-trained language models based on deep learning, such as BERT or GPT, are used, which can fine-tune subject-specific corpora from the field of intellectual property. Through NLP processing, the receiving module 110 is able to automatically recognize and extract key data, such as the case type (such as "patent application" or "trademark registration"), important deadlines (such as deadlines for official comments or renewal fee deadlines), and action requests (such as submitting a response or additional documentation).

[0011] Based on the extracted key data, the intellectual property case management system 100 then uses the process generation module 120 to generate a control process set appropriate for the case according to a control logic and / or an appropriate algorithm. The algorithm combines decision trees with machine learning models, taking into account factors such as case type, importance, official deadlines, and internal control deadlines for the case. The decision trees quickly determine the basic process steps using predefined rules, while the machine learning models rely on historical case data to predict the optimal execution sequence and potential risk points.

[0012] It is based on the Fig. 1 and Fig. 2, each of which shows an architecture diagram of a control process set. In a mandate case in the area of ​​patents or trademarks, the process generation module 120, after receiving a process start command, divides the control processes into an internal control process, an official control process, and a subsequent process. Each process is assigned to a predefined addressee. For example: The internal control process is directed at executors (e.g., sales staff, process handlers, or patent clerks) and the client; The official control process concerns the receipt of official letters and is thus directed at courts or intellectual property examining bodies (e.g., the Patent and Trademark Office); The subsequent process refers to possible further steps that go beyond the currently entrusted project (e.g.,These may go beyond the filing of a new case, such as legal proceedings, objections, corrections, transfers, inheritances, pledges, or annual fee payments. The subsequent process may involve courts, inspection authorities, or contracting authorities.

[0013] Simultaneously with the generation of control processes, all processes are marked by the management module 130. These markings include, but are not limited to: (1) Case type marking: marking the case type to which a process belongs in order to enable better categorisation and management; (2) Priority labeling: Automatically assigning a priority level based on important deadlines and case relevance to ensure that critical processes receive preferential treatment. (3) Execution status marking: Updating the execution status of a process in real time, such as "pending", "in progress", "completed" or "overdue".

[0014] The priority order of importance can be structured as follows: For official notifications (e.g., from courts or the Patent and Trademark Office): litigation > petition > appeal > reopening of proceedings > initial hearing > subsequent filing. For unofficial instructions (e.g., from the client): infringement opinions > patent analyses > FTO (freedom to operate) > patent search.

[0015] To optimize the execution order of processes, the management module 130 implements a priority-based scheduling algorithm. The algorithm considers factors such as important deadlines, resource availability, and process dependencies and dynamically adjusts the execution order of processes. In the event of resource conflicts or process delays, the management module 130 can perform automatic rescheduling to maximize overall efficiency.

[0016] During process execution, the management module 130 continuously tracks the progress of each individual process. Real-time data collection and analysis allows potential delays or risks to be identified early, and appropriate alerts can be generated to notify the responsible personnel to take action. Once the completion condition of a process is met, the update module 140 automatically updates the execution status flag and performs a data update, including archiving the control history and generating the next process step.

[0017] To increase the intelligence level of the process generation module 120, a machine learning model for process optimization is integrated. Based on data from successfully handled intellectual property cases in the past, the model of the process generation module 120 can learn the optimal processing strategies for different case types and predict the optimal control steps and processes for future cases. For example, the process generation module 120 can predict problems that may arise at a specific phase in a particular case type based on historical data and plan appropriate response processes in advance.

[0018] Furthermore, the management module 130 uses a graph database to manage the dependencies between the control processes. The graph-based structure allows the relationships and dependencies between processes to be clearly represented and changes such as adding, deleting, or modifying processes to be easily implemented.

[0019] It will be Fig. 3, which shows a second schematic block diagram according to an embodiment of the intellectual property case management system according to the invention.

[0020] According to a further embodiment, as shown in Fig. 3, the process generation module 120 includes the following sub-modules: a rule engine module 122, a machine learning module 124, and a process optimization module 126, wherein each step has standardized parameters that can be set based on factors such as internal control period, electronic data requirements, planned execution unit, labor cost, time expenditure, importance, and risk.

[0021] First, the rule engine module 122 is responsible for executing a decision tree. The rule engine module 122 uses a decision tree algorithm to quickly determine the basic process steps based on predefined business rules as well as laws and regulations. The decision tree algorithm used in this embodiment quickly determines the basic process steps based on predefined business rules as well as laws and regulations. The decision tree is constructed from nodes and edges, with each node representing a decision condition and path selection based on the key data of a case.

[0022] For example, the decision tree consists of several nodes, each representing a decision condition. Path selection is based on the key data of a case. Example: Node 1: Determine whether the case is a patent, trademark, or copyright. If it is a patent, proceed to node 2. If it is a trademark, proceed to node 3.

[0023] To automate the generation of the above-mentioned process, the Intellectual Property Case Management System 100 provides the following concrete execution process: Step 1: Perform data preprocessing. The extracted key data is standardized and converted into a format that can be processed by the machine learning and optimization models. Step 2: Run the rule engine. The rule engine module 122 quickly generates a preliminary list of process steps based on the decision tree. Step 3: Create the feature vector and combine the case features and preliminary process steps to serve as input to the machine learning model. Step 4: Perform a model prediction. The machine learning module 124 outputs an estimated execution time and a risk assessment for each step. Step 5: Perform optimization. The process optimization module 126 uses the prediction results as input to the optimization model to determine the optimal process sequence. Step 6: Output the results. The final control process set is output, including the execution sequence, estimated execution time, and responsible person(s) for each step. Step 7: Perform continuous learning. During process execution, actual execution data is collected and fed back to the machine learning module to continuously train and optimize the model.

[0024] According to another embodiment, the intellectual property case management system 100 must be able to effectively coordinate the work of all parties involved when handling an intellectual property case involving multiple execution units. This embodiment describes how to manage a control process with multiple execution units.

[0025] If a control process requires the collaboration of multiple departments or external partners, the process generation module 120 first generates a corresponding process branch for each execution unit. Each branch contains the tasks to be performed by a respective execution unit, deadlines, and completion conditions. The process generation module 120 links these process branches to the main process to ensure the coherence of the overall process.

[0026] During execution, the process generation module 120 chronologically lists the execution documents and annotations provided by the execution units and summarizes them in the case history. The management module 130 has a notification function. When a process step is completed or other units need to intervene, a notification is automatically sent to ensure timely information transmission.

[0027] According to a further embodiment, the application of the system is demonstrated in the actual processing of a patent application.

[0028] A customer submits an electronic patent application through the interface provided by the receiving module 110. After receiving the electronic data, the receiving module 110 uses NLP technology to extract key information, including details of the applicant, the title of the invention, the technical field, and the filing deadlines required by the authority.

[0029] Based on the extracted information, the process generation module 120 generates a control process set for the case in question. Since the patent concerns a specific technical field (e.g., semiconductors, artificial intelligence) that is technologically complex or has a high density of patent applications, the management module 130 may have a machine learning model. It is recommended to conduct a patent search and technical evaluation prior to filing to increase the likelihood of success. This can also improve processing speed and reduce the time required for review by a patent engineer.

[0030] During process execution, the management module 130 continuously monitors progress. If the technical assessment indicates that changes to the original application need to be made, the management module 130 automatically notifies the customer, recommends supplementing documentation, and updates the execution status of a process. All change logs and communication content are automatically archived by the management module 130, thus creating a seamless case history.

[0031] Once all process steps are completed, the administration module 130 automatically generates the electronic application documents for the patent office and updates the case status to "closed." The update module 140 also generates the next control process, such as waiting for the review result and preparing a possible statement on an examination action.

[0032] According to yet another embodiment, it is explained how the process generation module 120 can use a machine learning model to predict the completion conditions of a case and optimize the control processes based thereon.

[0033] To this end, the process generation module 120 collects extensive processing data from previous cases, including case type, technical area, audit progress, and final result. The process generation module 120 uses this data as training data sets to create a predictive model. Based on the characteristics and taking into account factors such as costs and risks, the process generation module 120 can predict the possible audit results and required process steps for a new case using the models and mathematical formulas described in the above embodiments.

[0034] For example, for patent applications in a specific technical field, the model of the process generation module 120 predicts that in such cases, an action for examination will typically be issued during the substantive examination phase, which will then require an amendment and response. Based on this prediction, the process generation module 120 integrates preparatory steps into the initial control process, such as preparing a response strategy and relevant technical information in advance. This predictive process design enables more proactive case management and effectively reduces the risk of delays caused by unexpected events. At the same time, the process generation module 120 continuously learns from new case data and updates the model to continuously improve the accuracy of the prediction.

[0035] According to yet another embodiment, to ensure data security and integrity, the update module 140 uses blockchain technology to record important transactions and documents.

[0036] After all control processes have been completed, the update module 140 compiles the associated information, such as execution documents, annotations, and transaction timestamps, into a block and uploads it to the blockchain. Due to the immutability of the blockchain, these records can serve as a trusted audit trail and prevent unauthorized data manipulation.

[0037] According to yet another embodiment, a concrete implementation of the inventive system for managing intellectual property cases is explained, which integrates the techniques for data extraction, transformation and loading from an ERP system.

[0038] The specific process is as follows: Once the receiving module 110 receives data on a new patent application, including the filing date, applicant information, and technical field, it identifies the case type as "Patent Application," determines important deadlines (e.g., priority deadline, deadline for replying to the action), and identifies necessary actions (e.g., "filing the request for substantive examination"). The process generation module 120 creates a series of control processes according to its control logic, for example: filing a request for substantive examination within a certain deadline (Process A), preparing a technical disclosure document (Process B), and paying the filing fee (Process C). Each process is assigned a corresponding label: The case type label is "Patent Application"; the priority label is based on the urgency of the deadline (e.g.,Process A with the highest priority); The execution status flag has the initial value “pending”.

[0039] Based on the settings, the process generation module 120 can determine whether an authorization request should be sent to the administrator to request the administrator to initiate a generated control process. After the administrator's review, approval is granted, and the process continues.

[0040] The management module 130 then schedules the execution sequence of the control processes according to their priority and monitors the progress of each process. If process A is at risk of missing an important deadline, the management module 130 increases its priority and notifies the relevant personnel. Once the completion condition for process A (e.g., the submission of a request for substantive review) is met, the update module 140 automatically updates the execution status flag of process A to "completed," archives the process's control history, records the execution time, the executor, and other information, and generates a next control process, such as "waiting for review decision" (process D). At the same time, the management module 130 proactively sends notifications to the relevant personnel and provides appropriate document templates to assist in the completion of subsequent work.Furthermore, the management module 130 dynamically adjusts the execution order of the control processes based on the actual execution situation and available resources. If it is determined that a process is blocked (in terms of execution: for example, due to operator overload; in terms of hardware: for example, due to insufficient server resources), the management module 130 automatically adjusts the priorities of other processes to improve overall efficiency. Regarding case history and document management, the management module 130 consolidates all execution documents and annotations of a control process into a complete case history, which is listed chronologically for easy query and traceability.

[0041] In the present embodiment, the technical features of data extraction and transfer from ERP systems are fully integrated and applied to the intellectual property case management system 100. Through its modular design and automated process management, the system can efficiently and reliably manage the entire lifecycle of intellectual property cases, thereby increasing the company's competitiveness in the intellectual property field.

[0042] According to yet another embodiment of the present invention, the intellectual property case management system 100 is further enhanced with document analysis and matching functions. This embodiment is particularly suitable for processing official letters submitted by intellectual property authorities worldwide or by foreign representatives. The documents are automatically matched to the cases managed in the system, thereby reducing manual reconciliation errors, improving the accuracy of deadline control, and increasing the efficiency of case management.

[0043] It will be Fig. 4, which shows a third schematic block diagram according to an embodiment of the intellectual property case management system according to the invention.

[0044] The intellectual property case management system 100 may further include an analysis module 150. The analysis module 150 is based on an NLP model and may perform the following two types of document analysis processes separately or alternately:

[0045] The first document analysis process consists of extracting keyword rules. The analysis module 150 can pre-store multiple sets of document keywords (e.g., "application number," "our reference," "filing date," "applicant," "priority claim," etc.) and automatically scan the electronic data during the analysis phase to find the alphanumeric combinations following these keywords. If these combinations conform to certain formats (e.g., application number formats such as PCT, CN, US), they are considered potential identifying information. The analysis module 150 can further apply keyword lists and template rules according to the document structure (such as PCT Notice of Allowance, Office Action, Notice of Approval, Notice of Opposition) to improve extraction accuracy.

[0046] The second document analysis process consists of semantic inference. The analysis module 150 operates without predefined keywords. Instead, the entire content of the electronic data is fed into a semantic inference model (e.g., a large language model such as GPT or an optimized version thereof) as an input layer to capture contextual meaning. Document attributes (e.g., "official decision" and "admission decision") are derived from this. The output layer provides identification information, including application number, applicant, filing date, case type (e.g., design, utility model, trademark), and associated deadline information. Systems that apply this process are suitable for unstructured or multilingual documents, as well as documents without unique keywords. However, some noise information that has no actual relevance (such as numbers of other reference cases) can also be extracted.

[0047] The information extracted by the analysis module 150 can be used as a result of a preliminary file analysis and can also be used alone to assist the user in identifying the source or content attributes of the file. However, to achieve the goal of automatic archiving, the present embodiment may further include a database 160 and a comparison module 170 to enable the automatic comparison and assignment process of documents to the internal cases.

[0048] Database 160 stores the archiving information for individual electronic data, for example, the intellectual property cases represented by a patent and trademark agency. This archiving information includes the filing number, applicant name, application number, case type, filing date, filing country, priority information, and the processing status of a case. The archiving information can be used as a basis for the comparison module 170 to perform a file comparison.

[0049] The comparison module 170 receives the identification information determined by the analysis module 150 through the first document analysis process and / or the second document analysis process and then performs either a single or cross-comparison. The execution logic is as follows:

[0050] If an application number has been successfully extracted in both processes and the two match, the comparison module 170 compares this application number with all cases registered in the database 160;

[0051] If only one of the processes produces a clear result, this result is used as the basis for comparison;

[0052] The comparison module 170 compares the identification information with the archiving information associated with the electronic data in the database 160. For example, the applicant's name, application number, case type, filing date, filing country, priority information, and processing status of a case are compared with the archiving information stored in the database 160 of all intellectual property cases. Based on the match result, a confidence index for the comparison is calculated;

[0053] If the confidence index reaches a predefined threshold (e.g., an initial threshold above 90%), a valid match is assumed. The received document is then automatically assigned to the relevant intellectual property case filing information, marking the match as successful.

[0054] If the comparison fails or the confidence index is within a medium range (e.g., between 70% and 89%, i.e., within the range of the second threshold), a manual review is initiated by the specialist staff. The system interface can simultaneously display the identification information obtained in the first and second processes and, to support manual decision-making, list the archiving information of multiple intellectual property cases with confidence indices in the medium range (e.g., 70% to 89%).

[0055] If the comparison result does not produce an exact match (e.g., if the confidence index is below the third threshold of 70%), this is marked as "requires manual processing" to avoid incorrect archiving. In this case, the comparison module 170 determines by default that the identification information of the electronic data has not yet been archived in the database 160, marks the electronic data as a new intellectual property case, and makes it available for manual confirmation. If manual confirmation determines that this case is a new intellectual property case, an archiving number is generated and used as the unique identification number for this new intellectual property case. All related data is summarized in the archiving information.

[0056] By combining the multi-layered analysis and comparison mechanism of database 160, analysis module 150, and comparison module 170, the present embodiment ensures both high accuracy in data processing and flexible handling in practical document analysis. At the same time, the workload for programmers is effectively reduced and the reliability of automated archiving and case control is improved overall.

[0057] It will be Fig.5, which shows a flowchart of the entire process according to an embodiment of the present invention. The goal is to integrate all steps in the processing of intellectual property cases, from data receipt to the completion of the control processes, in order to demonstrate the technical feasibility and the technically advantageous effects of the system according to the invention in practical operation.

Claims

[1] A system for managing intellectual property cases, comprising: a receiving module that receives electronic data relating to at least one case in the field of intellectual property; a process generation module that is connected to the receiver module via signal technology and generates a control process set according to a control logic, wherein the control process set comprises several control processes; and An administration module that is connected to the process generation module via signal technology and performs a marking for each control process, whereby the marking consists of a case type marking, a priority marking and an execution status marking, and the execution sequence of the control processes is dynamically adjusted according to the markings of the control processes. [2] System according to claim 1, wherein each control process has at least one completion condition, the system for managing intellectual property cases further comprising an update module which is signal-technically connected to the management module, wherein, when at least one completion condition of a control process is met, the update module performs a corresponding data update in that control process, wherein the data update comprises updating the execution status indicator and generating a further control process and a corresponding notification and / or document template in the same control process set. [3] System according to claim 2, wherein the update module compares the document status to determine whether at least one completion condition is met and automatically records this in the case history. [4] System according to claim 1, wherein the process generation module further extracts key data from the electronic data, the key data including case types, important deadlines and action requirements. [5] System according to claim 4, further comprising an analysis module, wherein the analysis module serves to analyze the key data and comprises the following: a first document analysis process, which consists of extracting at least one piece of identification information from the document data according to a keyword set; and a second document analysis process, which consists of analyzing the semantic content of the electronic data with a semantic inference model in order to obtain at least one piece of identification information. [6] System according to claim 5, wherein in the first document analysis process the electronic data are searched according to a predefined set of keywords for strings that correspond to a certain format in order to extract the identification information. [7] System according to claim 5, wherein in the second document analysis process the electronic data are used as the input level of the semantic inference model, the input level containing the identification information of the electronic data. [8] System according to any one of claims 5 to 7, wherein the identification information includes the applicant's name, application number, case type, filing date, country of filing, priority information and the case status. [9] System according to claim 5, further comprising a database, wherein the database contains the archiving information for each electronic data, the archiving information including the archiving number, the applicant's name, the application number, the case type, the filing date, the country of filing, priority information and the processing status of a case. [10] System according to claim 9, further comprising a comparison module, wherein the comparison module is connected to the analysis module and the database via a signal connection, wherein the comparison module receives the identification information obtained from the analysis module in the first and second document analysis processes and compares this identification information with the archiving information corresponding to the associated electronic data in the database, wherein the confidence index is calculated on the basis of the degree of agreement, wherein, when the confidence index reaches a first threshold, the identification information is archived in the archiving information corresponding to the associated electronic data. [11] System according to claim 10, wherein, if the confidence index does not reach the first threshold and lies between the first threshold and a second threshold, the comparison module initiates a manual review request and displays the identification information obtained through the first and second document analysis processes in a display interface and, in this case where the confidence index lies between the first threshold and the second threshold, provides at least one archiving information so that staff can make an assessment. [12] System according to claim 11, wherein, if the confidence index is lower than the second threshold, it is determined that the identification information associated with the electronic data was never archived in the database and the case is marked as a new case in the field of intellectual property.

Images