Intellectual property risk efficient analysis system based on big data

Abstract

The invention provides an intellectual property risk efficient analysis system based on big data, and relates to the technical field of data analysis, the method comprises the following steps: fusing intellectual property related data to form fused data; preprocessing the fused data to obtain processed data; clustering the processed data to obtain a clustering result; according to a clustering result, identifying intellectual property data groups with similar features; analyzing each intellectual property data group, and extracting key features and risk indexes; constructing an intellectual property risk assessment model according to the key features and the risk indexes; calculating a risk coefficient of each intellectual property group according to the intellectual property risk assessment model and the dynamic adjustment factor; and sorting each intellectual property group according to the risk coefficient of each intellectual property group, and identifying potential risk points. According to the invention, key features and risk indexes can be extracted, and the accuracy of risk analysis is improved.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to an efficient intellectual property risk analysis system based on big data. Background Art

[0002] Traditional IP risk analysis methods mainly rely on manual review and professional judgment. This method may be able to cope with small-scale data, but its efficiency and accuracy are greatly reduced when faced with large-scale and diverse IP data.

[0003] First, manual review is inefficient. As IP data is huge and complex, manual review not only takes a lot of time and effort, but also makes it difficult to ensure the timeliness of the review. In the fast-paced modern business environment, this inefficiency may cause companies to miss market opportunities or fail to respond to potential risks in a timely manner.

[0004] Secondly, traditional methods are easily affected by subjective factors. Although the experience and judgment of professionals are valuable, they are inevitably biased and subjective. This subjectivity may lead to deviations in risk analysis, thus affecting the accuracy of decision-making. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide an efficient intellectual property risk analysis system based on big data, which can extract key features and risk indicators and improve the accuracy of risk analysis.

[0006] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0007] First, an efficient analysis method for intellectual property risk based on big data, the method comprising:

[0008] Obtain intellectual property-related data, including patent application records, trademark registration information, copyright registration data, and infringement case records;

[0009] Fusion of intellectual property related data to form fused data;

[0010] Preprocessing the fused data to obtain processed data;

[0011] Clustering the processed data to obtain clustering results;

[0012] Based on the clustering results, identify IP data groups with similar characteristics;

[0013] Analyze each IP data group to extract key features and risk indicators;

[0014] Construct an intellectual property risk assessment model based on key characteristics and risk indicators;

[0015] Calculate the risk factor of each intellectual property group based on the intellectual property risk assessment model and dynamic adjustment factors;

[0016] Sort each intellectual property group according to its risk factor and identify potential risk points;

[0017] According to potential risk points and risk thresholds, potential risk points are divided into different risk levels.

[0018] Furthermore, the intellectual property related data are integrated to form integrated data, including:

[0019] Obtain key fields in intellectual property-related data, including application number, registration number and right holder name;

[0020] Obtain detailed records corresponding to key fields from the intellectual property-related data, and locate corresponding key feature fields in the detailed records, where the key feature fields include technology classification, legal status, and application date;

[0021] The key feature fields are associated with the key fields to form a fused data including the key fields and the key features.

[0022] Further, the fused data is preprocessed to obtain preprocessed data, including:

[0023] Determine the dimensions of the state vector and observation vector, and set the initial state estimate and the initial error covariance matrix P 0 ;

[0024] Define the state transfer matrix A, observation matrix H, control input matrix B, process noise covariance matrix Q and observation noise covariance matrix R; for each time point k, perform the following steps:

[0025] Use the state transition matrix A to predict the next state:

[0026] in, is at time t k The estimated state of A(t k ,t k-1 ) is the state transfer matrix; is at time t k-1 The estimated state of B(t k ,t k-1 ) is the control input matrix; u k-1 is at time t k-1 The control input signal w(t k ,tk-1 ) is the process noise; k is a discrete time index used to represent the kth moment in the time series; t is a continuous time variable, t k and t k-1 Respectively represent the specific time corresponding to time k and time k-1;

[0027] Forecast error covariance:

[0028] in, is at time t k The error covariance matrix represents the error at the current time t k The prediction of the covariance matrix of the state estimation error before the measurement value of P k-1 is at time t k-1 The error covariance matrix of A(t k ,t k-1 ) T is the state transfer matrix A(t k ,t k-1 ) is the transpose of Q(t k ,t k-1 ) is the process noise covariance matrix;

[0029] When there is a new observation z k When , calculate the Kalman gain:

[0030] Among them, K k is at time t k The Kalman gain matrix of is at time t k The error covariance matrix of H(t k ) is the observation matrix; H(t k ) T is the observation matrix H(t k ) is the transpose of k ) is at time t k The measurement noise covariance matrix represents the noise and error in the measurement process; is the inverse of a matrix;

[0031] Update the state using the observed values:

[0032] in, is at time t k The posterior estimated state of ; is at time t k A priori state estimate of K k is the Kalman gain matrix; z k is at time t k The measured value of is the predicted observation value that maps the prior estimated state to the observation space; v k is a term that deals with measurement bias;

[0033] Update the error covariance matrix:

[0034] Among them, P k is at time t k The posterior estimation error covariance matrix of ; I is the identity matrix; IK k H(t k ) is a matrix used to calculate the posterior estimation error covariance matrix, which represents the part of the error reduced by the measurement update minus the prior estimate; (IK k H(t k )) T is the matrix IK k H(t k ) is the error covariance; R(t k ) is the measurement noise covariance matrix;

[0035] After Kalman filtering, a filtered state estimation sequence is obtained.

[0036] The filtered state estimation sequence Normalization is performed to obtain preprocessed data.

[0037] Furthermore, the filtered state estimation sequence Perform normalization to obtain processed data, including:

[0038] The filtered state estimation sequence pass Processing to obtain processed data

[0039] in, is the filtered state estimation sequence The i-th element in x min and x max Sequence ; α is a scaling factor used to adjust the dynamic range of the data; β is a translation parameter used to translate the data before normalization; w i It is a data point The weight of i is a bias term.

[0040] Furthermore, the processed data is clustered to obtain clustering results, including:

[0041] Determine the number of clusters K, and randomly select K points from the processed data as the initial cluster centers;

[0042] For each point in the processed data, calculate its Euclidean distance to each cluster center;

[0043] According to the Euclidean distance, each point is assigned to the corresponding cluster center to form K clusters;

[0044] For each cluster, calculate the mean of all data points within it and use the mean as the new cluster center;

[0045] Reassign data points to the nearest cluster center and update the cluster center until the preset number of iterations is reached to obtain the clustering result.

[0046] Furthermore, the calculation formula of the intellectual property risk assessment model is:

[0047]

[0048] Among them, a ijp is the weight of the pth data point of the jth key feature in the ith risk indicator; x ijp is the value of the pth data point of the jth key feature in the ith risk indicator; w i is the weight of the i-th risk indicator; k ij is the weight of the jth key feature in the ith risk indicator; m is the total number of key features; n is the total number of risk indicators; l is the number of data points used to evaluate each key feature.

[0049] Furthermore, the calculation formula for the risk factor of each intellectual property group is:

[0050] RCg=(w DR ×DC+w DT ×(w DUF ×DUF+w DTN ×DTN))×R g ;

[0051] Where RCg is the risk factor of the g-th intellectual property group; w DR is the total weight of data reliability; DC is the accuracy score of data collection; w DT is the total weight of data timeliness; w DUF is the weight of data update frequency in data timeliness; DUF is the data update frequency score; w DTN is the weight of data timeliness requirement in data timeliness; DTN is the data timeliness requirement score; R gis the original risk score of the g-th intellectual property group calculated according to the intellectual property risk assessment model.

[0052] Second, an efficient intellectual property risk analysis system based on big data, including:

[0053] The acquisition module is used to acquire intellectual property-related data, including patent application records, trademark registration information, copyright registration data, and infringement case records; fuse the intellectual property-related data to form fused data; pre-process the fused data to obtain processed data; cluster the processed data to obtain clustering results; and identify intellectual property data groups with similar characteristics based on the clustering results;

[0054] The processing module is used to analyze each intellectual property data group and extract key features and risk indicators; construct an intellectual property risk assessment model based on the key features and risk indicators; calculate the risk coefficient of each intellectual property group based on the intellectual property risk assessment model and dynamic adjustment factors; sort each intellectual property group according to the risk coefficient of each intellectual property group and identify potential risk points; and classify potential risk points into different risk levels based on potential risk points and risk thresholds.

[0055] According to a third aspect, a computing device includes:

[0056] one or more processors;

[0057] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the above method.

[0058] In a fourth aspect, a computer-readable storage medium stores a program, and the program implements the above method when executed by a processor.

[0059] The above solution of the present invention includes at least the following beneficial effects:

[0060] Through big data technology, the present invention can efficiently process and analyze massive amounts of intellectual property-related data, greatly improving the efficiency of analysis. The present invention can more accurately identify intellectual property data groups with similar characteristics, and then extract key features and risk indicators, thereby improving the accuracy of risk analysis. By constructing an intellectual property risk assessment model and combining it with dynamic adjustment factors, the present invention can conduct a comprehensive risk assessment of each intellectual property group. The present invention can classify potential risk points into different risk levels based on potential risk points and set risk thresholds. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1It is a flowchart of an efficient intellectual property risk analysis method based on big data provided by an embodiment of the present invention.

[0062] Figure 2 It is a schematic diagram of an efficient intellectual property risk analysis system based on big data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0063] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0064] like Figure 1 As shown, an embodiment of the present invention proposes an efficient intellectual property risk analysis method based on big data, the method comprising:

[0065] Step 11, obtaining intellectual property related data, including patent application records, trademark registration information, copyright registration data and infringement case records;

[0066] Step 12, fusing the intellectual property related data to form fused data;

[0067] Step 13, preprocessing the fused data to obtain processed data;

[0068] Step 14, clustering the processed data to obtain a clustering result;

[0069] Step 15, identifying intellectual property data groups with similar characteristics based on the clustering results;

[0070] Step 16, analyzing each IP data group to extract key features and risk indicators;

[0071] Step 17, construct an intellectual property risk assessment model based on key features and risk indicators;

[0072] Step 18, calculating the risk coefficient of each intellectual property group according to the intellectual property risk assessment model and the dynamic adjustment factor;

[0073] Step 19, sorting each intellectual property group according to the risk coefficient of each intellectual property group, identifying potential risk points; and dividing the potential risk points into different risk levels according to the potential risk points and risk thresholds.

[0074] In the embodiment of the present invention, step 11, by comprehensively collecting intellectual property-related data including patent application records, trademark registration information, copyright registration data, and infringement case records, etc., a rich data source is provided for subsequent risk analysis, ensuring the comprehensiveness and accuracy of the analysis. Step 12, data fusion can integrate intellectual property data from different sources and in different formats to form a unified data set, which helps to eliminate data silos and improve data availability and analysis efficiency. Step 13, data preprocessing can clean and organize the fused data and remove noise, thereby improving data quality and analysis accuracy. Step 14, through cluster analysis, intellectual property data with similar characteristics can be classified into one category, which helps to discover hidden patterns and associations in the data. Step 15, by identifying data groups with similar characteristics, the intellectual property set with common risk characteristics can be more accurately located. Step 16, by conducting in-depth analysis on each data group, extracting key features and risk indicators, it is possible to more accurately identify potential risk points within the group. Step 17, based on the extracted key features and risk indicators, a risk assessment model is constructed to quantitatively assess the risk level of intellectual property. Step 18: Calculate the risk coefficient of each intellectual property group through the risk assessment model and dynamic adjustment factor, which can reflect the risk status of the group in real time. Step 19: By sorting and grading the risks of intellectual property groups, it can help decision makers quickly identify high-risk groups, prioritize potential risk points, and improve the efficiency and effectiveness of risk management.

[0075] In a preferred embodiment of the present invention, the above step 12, fusing the intellectual property related data to form fused data, may include:

[0076] Step 121, obtaining key fields in the intellectual property related data, the key fields including application number, registration number and right holder name;

[0077] Step 122, obtaining detailed records corresponding to the key fields from the intellectual property related data, and locating the corresponding key feature fields in the detailed records, the key feature fields including technology classification, legal status and application date;

[0078] Step 123, the key feature field is associated with the key field to form a fused data including the key field and the key feature.

[0079] In an embodiment of the present invention, by extracting key fields and key feature fields and associating them, the integration and standardization of different types of intellectual property data are achieved. This integration makes the originally scattered and heterogeneous data unified and orderly, which is convenient for subsequent data processing and analysis. The fused data not only contains basic identification information (such as application number, registration number and right holder name), but also covers key features such as technical classification, legal status and application date. The addition of this information greatly enriches the dimension of the data, making the analysis more in-depth and comprehensive. Data fusion reduces data redundancy and inconsistency, improves data quality, and thus improves analysis efficiency. Analysts do not need to switch back and forth between different types of data, they only need to operate on a unified data set. By accurately associating key fields and key feature fields, the accuracy and consistency of the data are ensured.

[0080] In another preferred embodiment of the present invention, step 121 may include:

[0081] From patent application records, trademark registration information, copyright registration data and other intellectual property-related databases, through SQL query or other data extraction tools, selectively extract key fields such as application number, registration number and right holder name, and save these key fields to a temporary data table for subsequent processing.

[0082] Step 122 may include:

[0083] For each key field (such as application number, registration number), go back to the original intellectual property database again to find the detailed record corresponding to the key field. In the detailed record, further locate and extract key feature fields such as technical classification, legal status and application date. These fields are usually included in the metadata or attributes of the detailed record, ensuring that the extracted key feature fields can correspond one-to-one with the key fields for subsequent data association operations.

[0084] Step 123 may include:

[0085] Use database connection operations (such as JOIN operations) to associate the temporary data table containing key fields with the detailed record table containing key feature fields. The basis for association is the key fields such as application number and registration number, ensuring that these fields in the two tables can be fully matched. After the association is completed, a new fused data table will be formed, which contains both key fields (such as application number, registration number and right holder name) and corresponding key feature fields (such as technical classification, legal status and application date). Perform a quality check on the fused data to ensure the accuracy and completeness of the data, and then export it to a CSV file or other format for subsequent risk analysis. Through the above specific implementation process, the intellectual property-related data can be efficiently fused to form a fused data set containing rich information.

[0086] In a preferred embodiment of the present invention, the above step 13, preprocessing the fused data to obtain preprocessed data, includes:

[0087] Step 131, determine the dimensions of the state vector and observation vector, and set the initial state estimate and the initial error covariance matrix P 0 ;

[0088] Step 132, define the state transfer matrix A, the observation matrix H, the control input matrix B, the process noise covariance matrix Q and the observation noise covariance matrix R; for each time point k, perform the following steps:

[0089] Use the state transition matrix A to predict the next state:

[0090] in, is at time t k The estimated state of A(t k ,t k-1 ) is the state transfer matrix; is at time t k-1 The estimated state of B(t k ,t k-1 ) is the control input matrix; u k-1 is at time t k-1 The control input signal w(t k ,t k-1 ) is the process noise; k is a discrete time index used to represent the kth moment in the time series; t is a continuous time variable, t k and t k-1 Respectively represent the specific time corresponding to time k and time k-1;

[0091] Forecast error covariance:

[0092] in, is at time t k The error covariance matrix represents the error at the current time t k The prediction of the covariance matrix of the state estimation error before the measurement value of P k-1 is at time t k-1 The error covariance matrix of A(t k ,t k-1 ) T is the state transfer matrix A(t k ,t k-1 ) is the transpose of Q(t k ,t k-1 ) is the process noise covariance matrix;

[0093] When there is a new observation z k When , calculate the Kalman gain:

[0094] Among them, K k is at time t k The Kalman gain matrix of is at time t k The error covariance matrix of H(t k ) is the observation matrix; H(t k ) T is the observation matrix H(t k ) is the transpose of k ) is at time t k The measurement noise covariance matrix represents the noise and error in the measurement process; is the inverse of a matrix;

[0095] Update the state using the observed values:

[0096] in, is at time t k The posterior estimated state of ; is at time t k A priori state estimate of K k is the Kalman gain matrix; z k is at time t k The measured value of is the predicted observation value that maps the prior estimated state to the observation space; v k is a term that deals with measurement bias;

[0097] Update the error covariance matrix:

[0098] Among them, P k is at time t kThe posterior estimation error covariance matrix of ; I is the identity matrix; IK k H(t k ) is a matrix used to calculate the posterior estimation error covariance matrix, which represents the part of the error reduced by the measurement update minus the prior estimate; (IK k H(t k )) T is the matrix IK k H(t k ) is the error covariance; R(t k ) is the measurement noise covariance matrix;

[0099] Step 133, after Kalman filtering, a filtered state estimation sequence is obtained.

[0100] Step 134, the filtered state estimation sequence Normalization is performed to obtain preprocessed data.

[0101] In an embodiment of the present invention, the Kalman filter can estimate the state of a dynamic system from a series of incomplete and noisy measurements. Through filtering, the noise and outliers in the data can be significantly reduced, and the smoothness and accuracy of the data can be improved. By combining the dynamic model of the system and real-time observation data, the Kalman filter can provide a more accurate state estimate. This accuracy improvement is crucial for subsequent intellectual property risk analysis because it ensures that the analysis is based on reliable and high-quality data. The Kalman filter can not only estimate the current state, but also predict the future state based on the current state. This predictive ability is very useful in intellectual property risk analysis and can help decision makers identify potential risk points in advance and take corresponding countermeasures. The Kalman filter can process data streams in real time and automatically update state estimates. This automation and real-time nature enable the preprocessing method to efficiently process large-scale intellectual property data.

[0102] In a preferred embodiment of the present invention, the above step 134 is to estimate the state sequence after filtering. Perform normalization to obtain processed data, including:

[0103] The filtered state estimation sequence pass Processing to obtain processed data

[0104] in, is the filtered state estimation sequence The i-th element in x min and xmax Sequence ; α is a scaling factor used to adjust the dynamic range of the data; β is a translation parameter used to translate the data before normalization; w i It is a data point The weight of i is a bias term.

[0105] In an embodiment of the present invention, the normalization process converts the original data into a specific range (usually between 0 and 1), which can eliminate the differences in the dimensions and numerical ranges between different features of the original data, so that all data features can be compared and analyzed on the same scale. This helps to improve the accuracy and efficiency of subsequent data processing. The normalization method can better adapt to the nonlinear distribution of data, especially when the data is unevenly distributed or there are extreme values. Through nonlinear transformation, the original characteristics of the data can be better preserved, while reducing the impact of extreme values ​​on the overall data distribution. By introducing scaling factors and translation parameters, the dynamic range and translation amount of the data can be flexibly adjusted to meet different analysis requirements. This flexibility makes the normalization process more versatile and customizable, and the representation of data can be optimized according to specific circumstances. By assigning weights and bias terms to each data point, the effect of the normalization process can be further refined. The weight can be adjusted according to the importance or reliability of the data point, and the bias term can be used to correct or adjust the normalization result of a specific data point. This processing method can more accurately reflect the actual characteristics and requirements of the data.

[0106] In a preferred embodiment of the present invention, the above step 14, clustering the processed data to obtain a clustering result, includes:

[0107] Step 141, determine the number of clusters K, and randomly select K points from the processed data as initial cluster centers;

[0108] Step 142, for each point in the processed data, calculating the Euclidean distance between it and each cluster center;

[0109] Step 143, assigning each point to a corresponding cluster center according to the Euclidean distance to form K clusters;

[0110] Step 144, for each cluster, calculate the mean of all data points within it, and use the mean as the new cluster center;

[0111] Step 145, reallocate the data points to the nearest cluster center, and update the cluster center until a preset number of iterations is reached to obtain a clustering result.

[0112] In an embodiment of the present invention, through clustering, a large number of data points are grouped into a few clusters, thereby simplifying the data structure, helping to understand the distribution and characteristics of the data as a whole, and making data analysis more efficient. Clustering can reveal the inherent structure and pattern in the data. In intellectual property risk analysis, this may mean identifying intellectual property groups with similar risk characteristics, which helps to locate and manage risks more accurately. During the clustering process, data points that are far away from all clusters may be regarded as outliers. In the field of intellectual property, these outliers may represent high-risk cases that require special attention. Through clustering, resources can be more concentratedly allocated to clusters with higher risks, thereby achieving optimal allocation of resources. For example, for high-risk intellectual property groups, more manpower and material resources can be invested in detailed review and protection. The clustering results are easy to visualize, which helps to intuitively show the structure and pattern of complex data to non-technical personnel. This is particularly important in intellectual property risk analysis because it can help decision makers better understand the risk distribution and make more informed decisions based on it. The K-means algorithm is relatively efficient when processing large-scale data sets and is easy to parallelize, so that it can cope with the growing amount of data in the field of intellectual property.

[0113] In a preferred embodiment of the present invention, the above step 15 is specifically implemented as follows:

[0114] From the clustering results obtained in step 14, a mapping relationship from data points to cluster centers can be obtained, which indicates to which cluster each data point is assigned. Based on the clustering results, the intellectual property data points belonging to the same cluster can be grouped together. The data points in these groups have similar characteristics because they are divided into the same group by the clustering algorithm. A unique identifier or label is assigned to each identified intellectual property data group to facilitate subsequent analysis and management.

[0115] In a preferred embodiment of the present invention, the above step 16 is specifically implemented as follows:

[0116] From the intellectual property data groups identified in step 15, select a specific group for analysis; analyze the intellectual property data in the group and extract key features. These key features are attributes related to the type of intellectual property, application date, right holder, technical field, etc. Based on the extracted key features, calculate the risk indicators related to the group; record the extracted key features and calculated risk indicators, and generate an analysis report; repeat the steps to analyze each identified intellectual property data group until all groups have been processed. The key feature extraction process may include:

[0117] Count the number and proportion of different types of intellectual property rights (such as patents, trademarks, copyrights, etc.), and analyze the correlation between different types of intellectual property rights and specific risks (such as infringement risks); count the number of intellectual property applications by year or month, observe application trends, analyze the time lag effect between application date and intellectual property value, validity or risk, count the number of intellectual property rights of different right holders, identify major right holders and potential competitors, analyze the application activity of right holders in different time periods, and the correlation with their business strategies, and classify intellectual property rights into different technical fields according to their descriptions or classification codes; analyze application trends in various technical fields, identify hot technologies and emerging technologies, and calculate the technology diversity index of right holders or the entire data set to measure the breadth of technological innovation.

[0118] In a preferred embodiment of the present invention, the calculation formula of the intellectual property risk assessment model is:

[0119]

[0120] Among them, a ijp is the weight of the pth data point of the jth key feature in the ith risk indicator; x ijp is the value of the pth data point of the jth key feature in the ith risk indicator; w i is the weight of the i-th risk indicator; k ij is the weight of the jth key feature in the ith risk indicator; m is the total number of key features; n is the total number of risk indicators; l is the number of data points used to evaluate each key feature.

[0121] In an embodiment of the present invention, the intellectual property risk assessment model can comprehensively consider multiple risk indicators (represented by n), each risk indicator contains multiple key features (represented by m), and each key feature is specifically quantified by multiple data points (represented by l). This comprehensive assessment method can more comprehensively reflect the risk status of intellectual property. The weight coefficient in the intellectual property risk assessment model can be more flexible, and can highlight the key points according to the actual situation, thereby improving the accuracy and pertinence of the assessment. By using multiple data points to evaluate each key feature, the intellectual property risk assessment model can capture more detailed risk information. The intellectual property risk assessment model ultimately outputs a specific risk value R, which is a comprehensive quantitative representation of the intellectual property risk. Compared with qualitative risk assessment, quantitative output is easier to sort, compare and analyze risks, and helps to formulate more accurate risk management strategies.

[0122] In a preferred embodiment of the present invention, the calculation formula of the risk factor of each intellectual property group is:

[0123] RCg=(w DR ×DC+wDT ×(w DUF ×DUF+w DTN ×DTN))×R g ;

[0124] Where RCg is the risk factor of the g-th intellectual property group; w DR is the total weight of data reliability; DC is the accuracy score of data collection; w DT is the total weight of data timeliness; w DUF is the weight of data update frequency in data timeliness; DUF is the data update frequency score; w DTN is the weight of data timeliness requirement in data timeliness; DTN is the data timeliness requirement score; R g is the original risk score of the g-th intellectual property group calculated according to the intellectual property risk assessment model.

[0125] In an embodiment of the present invention, the formula can convert the risk level of each intellectual property group into a specific numerical value, namely, a risk coefficient. This quantitative method makes risk comparison and ranking simple and clear, and helps enterprises give priority to high-risk groups. At the same time, it comprehensively considers multiple dimensions such as the accuracy of data collection, the frequency of data updates, and the timeliness requirements of data, ensuring the comprehensiveness of risk assessment. This multi-dimensional assessment method can more realistically reflect the risk status of intellectual property groups. By introducing different weight coefficients, the formula allows the importance of each factor in risk assessment to be adjusted according to actual conditions. This flexibility makes the assessment more in line with actual needs. The formula places special emphasis on the timeliness of the data, and reflects the latest information and its ability to meet business needs through the data update frequency score and the data timeliness requirement score. This is especially important for the rapidly changing intellectual property environment. The R in the formula g It is the raw risk score calculated based on the IP risk assessment model, which means that the formula can be combined with a broader risk assessment framework to provide a more comprehensive and in-depth risk analysis.

[0126] In a preferred embodiment of the present invention, the above step 19 is specifically implemented as follows:

[0127] Step 191, obtain the risk factor of each intellectual property group calculated in the previous step; use a sorting algorithm (such as quick sort) to sort each intellectual property group in order of risk factor from high to low (or low to high); save the sorted intellectual property group list for subsequent analysis and processing.

[0128] Step 192, according to business needs and historical data, set a risk threshold, which is used to define which risk factors indicate high risk, which indicate medium risk, and which indicate low risk; based on the sorted list of intellectual property groups, mark the groups with risk factors higher than the set threshold as potential risk points.

[0129] Step 193, clarify the risk level classification criteria, for example, you can set three levels of "high risk", "medium risk" and "low risk". Each level corresponds to a range of risk factors; according to the previously set risk threshold and risk level definition, each intellectual property group marked as a potential risk point is classified into the corresponding risk level; the classification results are recorded and a report is generated. The report should include the group identification, risk factor, risk level, possible risk factors and recommended measures of each potential risk point.

[0130] The specific example operation process is as follows:

[0131] Read the risk factor of each intellectual property group from the database or data file, and sort the risk factors in Excel or data analysis software; discuss and set appropriate risk thresholds within the team based on business needs and past experience, such as a high risk threshold of 0.8 and a medium risk threshold of 0.5; mark groups with risk factors above 0.8 as high risk points, those above 0.5 but below 0.8 as medium risk points, and those below 0.5 as low risk points; write a report that lists all potential risk points and their risk levels, and proposes corresponding risk management recommendations for each risk point. Through such concrete steps, organizations can more effectively identify and manage potential risks in their intellectual property groups.

[0132] like Figure 2 As shown, an embodiment of the present invention further provides an efficient intellectual property risk analysis system 20 based on big data, comprising:

[0133] The acquisition module 21 is used to acquire intellectual property related data, including patent application records, trademark registration information, copyright registration data and infringement case records; fuse the intellectual property related data to form fused data; pre-process the fused data to obtain processed data; cluster the processed data to obtain clustering results; and identify intellectual property data groups with similar characteristics based on the clustering results;

[0134] The processing module 22 is used to analyze each intellectual property data group and extract key features and risk indicators; construct an intellectual property risk assessment model based on the key features and risk indicators; calculate the risk coefficient of each intellectual property group based on the intellectual property risk assessment model and the dynamic adjustment factor; sort each intellectual property group according to the risk coefficient of each intellectual property group and identify potential risk points; and classify potential risk points into different risk levels based on potential risk points and risk thresholds.

[0135] It should be noted that the device is a device corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0136] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0137] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0138] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0139] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0140] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0141] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0142] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0143] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical disks.

[0144] In addition, it should be noted that in the apparatus and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. Moreover, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order, and some steps can be performed in parallel or independently of each other. For those of ordinary skill in the art, it is understood that all or any steps or components of the method and apparatus of the present invention can be implemented in any computing device (including processors, storage media, etc.) or a network of computing devices in hardware, firmware, software or a combination thereof, which can be achieved by those of ordinary skill in the art using their basic programming skills after reading the description of the present invention.

[0145] Therefore, the purpose of the present invention can also be achieved by running a program or a group of programs on any computing device. The computing device can be a well-known general device. Therefore, the purpose of the present invention can also be achieved by simply providing a program product containing a program code that implements the method or device. That is to say, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any well-known storage medium or any storage medium developed in the future. It should also be pointed out that in the device and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. In addition, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order. Some steps can be performed in parallel or independently of each other.

[0146] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An efficient analysis method of intellectual property risk based on big data, characterized by: The method comprises: Obtain intellectual property-related data, including patent application records, trademark registration information, copyright registration data, and infringement case records; Fusion of intellectual property related data to form fused data; Preprocessing the fused data to obtain processed data; Clustering the processed data to obtain clustering results; Based on the clustering results, identify IP data groups with similar characteristics; Analyze each IP data group to extract key features and risk indicators; Construct an intellectual property risk assessment model based on key characteristics and risk indicators; Calculate the risk factor of each intellectual property group based on the intellectual property risk assessment model and dynamic adjustment factors; Sort each intellectual property group according to its risk factor and identify potential risk points; According to potential risk points and risk thresholds, potential risk points are divided into different risk levels.

2. The efficient intellectual property risk analysis method based on big data according to claim 1 is characterized in that: Integrate IP-related data to form integrated data, including: Obtain key fields in intellectual property-related data, including application number, registration number and right holder name; Obtain detailed records corresponding to key fields from the intellectual property-related data, and locate corresponding key feature fields in the detailed records, where the key feature fields include technology classification, legal status, and application date; The key feature fields are associated with the key fields to form a fused data including the key fields and the key features.

3. The efficient intellectual property risk analysis method based on big data according to claim 2 is characterized in that: The fused data is preprocessed to obtain preprocessed data, including: Determine the dimensions of the state vector and observation vector, and set the initial state estimate and the initial error covariance matrix P0; Define the state transfer matrix A, observation matrix H, control input matrix B, process noise covariance matrix Q and observation noise covariance matrix R; for each time point k, perform the following steps: Use the state transition matrix A to predict the next state: in, is at time t k The estimated state of A(t k ,t k-1 ) is the state transfer matrix; is at time t k-1 The estimated state of B(t k ,t k-1 ) is the control input matrix; u k-1 is at time t k-1 The control input signal w(t k ,t k-1 ) is the process noise; k is a discrete time index used to represent the kth moment in the time series; t is a continuous time variable, t k and t k-1 Respectively represent the specific time corresponding to time k and time k-1; Forecast error covariance: in, is at time t k The error covariance matrix represents the error at the current time t k The prediction of the covariance matrix of the state estimation error before the measurement value of P k-1 is at time t k-1 The error covariance matrix of A(t k ,t k-1 ) T is the state transfer matrix A(t k ,t k-1 ) is the transpose of Q(t k ,t k-1 ) is the process noise covariance matrix; When there is a new observation z k When , calculate the Kalman gain: Among them, K k is at time t k The Kalman gain matrix of is at time t k The error covariance matrix of H(t k ) is the observation matrix; H(t k ) T is the observation matrix H(t k ) is the transpose of k ) is at time t k The measurement noise covariance matrix represents the noise and error in the measurement process; is the inverse of a matrix; Update the state using the observed values: in, is at time t k The posterior estimated state of ; is at time t k A priori state estimate of K k is the Kalman gain matrix; z k is at time t k The measured value of is the predicted observation value that maps the prior estimated state to the observation space; v k is a term that deals with measurement bias; Update the error covariance matrix: Among them, P k is at time t k The posterior estimation error covariance matrix of ; I is the identity matrix; IK k H(t k ) is a matrix used to calculate the posterior estimation error covariance matrix, which represents the part of the error reduced by the measurement update minus the prior estimate; (IK k H(t k )) T is the matrix IK k H(t k ) is the error covariance; R(t k ) is the measurement noise covariance matrix; After Kalman filtering, a filtered state estimation sequence is obtained. The filtered state estimation sequence Normalization is performed to obtain preprocessed data.

4. The efficient intellectual property risk analysis method based on big data according to claim 3 is characterized in that: The filtered state estimation sequence Perform normalization to obtain processed data, including: The filtered state estimation sequence pass Processing to obtain processed data in, is the filtered state estimation sequence The i-th element in x min and x max Sequence ; α is a scaling factor used to adjust the dynamic range of the data; β is a translation parameter used to translate the data before normalization; w i It is a data point The weight of i is a bias term.

5. The efficient intellectual property risk analysis method based on big data according to claim 4 is characterized in that: Cluster the processed data to obtain clustering results, including: Determine the number of clusters K, and randomly select K points from the processed data as the initial cluster centers; For each point in the processed data, calculate its Euclidean distance to each cluster center; According to the Euclidean distance, each point is assigned to the corresponding cluster center to form K clusters; For each cluster, calculate the mean of all data points within it and use the mean as the new cluster center; Reassign data points to the nearest cluster center and update the cluster center until the preset number of iterations is reached to obtain the clustering result.

6. The efficient intellectual property risk analysis method based on big data according to claim 5 is characterized in that: The calculation formula of the intellectual property risk assessment model is: Among them, a ijp is the weight of the pth data point of the jth key feature in the ith risk indicator; x ijp is the value of the pth data point of the jth key feature in the ith risk indicator; w i is the weight of the i-th risk indicator; k ij is the weight of the jth key feature in the ith risk indicator; m is the total number of key features; n is the total number of risk indicators; l is the number of data points used to evaluate each key feature.

7. The efficient analysis method of intellectual property risk based on big data according to claim 6 is characterized in that: The risk factor for each IP group is calculated as follows: RCg=(in DR ×DC+in DT ×(in DUF ×DUF+in DTN ×DTN))×R g ; Where RCg is the risk factor of the g-th intellectual property group; w DR is the total weight of data reliability; DC is the accuracy score of data collection; w DT is the total weight of data timeliness; w DUF is the weight of data update frequency in data timeliness; DUF is the data update frequency score; w DTN is the weight of data timeliness requirement in data timeliness; DTN is the data timeliness requirement score; R g is the original risk score of the g-th intellectual property group calculated according to the intellectual property risk assessment model.

8. An efficient intellectual property risk analysis system based on big data, characterized by: The method as claimed in any one of claims 1 to 7 comprises: The acquisition module is used to acquire intellectual property-related data, including patent application records, trademark registration information, copyright registration data, and infringement case records; fuse the intellectual property-related data to form fused data; pre-process the fused data to obtain processed data; cluster the processed data to obtain clustering results; and identify intellectual property data groups with similar characteristics based on the clustering results; The processing module is used to analyze each intellectual property data group and extract key features and risk indicators; construct an intellectual property risk assessment model based on the key features and risk indicators; calculate the risk coefficient of each intellectual property group based on the intellectual property risk assessment model and dynamic adjustment factors; sort each intellectual property group according to the risk coefficient of each intellectual property group and identify potential risk points; and classify potential risk points into different risk levels based on potential risk points and risk thresholds.

9. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in any one of claims 1 to 7.

Images