Enterprise intellectual property intelligent management system based on service big data

Abstract

The invention discloses an enterprise intellectual property intelligent management system based on service big data, which belongs to the technical field of enterprise management and comprises a data statistics module, a data management module, a feature extraction module and a risk assessment module. The method comprises the following steps: counting various data of enterprise intellectual property, and preprocessing the counted data to obtain intellectual property processing data; a quick data query function is realized by using an inverted index algorithm, and different query permissions are allocated according to roles and responsibilities of users; performing feature extraction on the intellectual property processing data to obtain key feature data; and constructing an intellectual property risk assessment model according to the logistic regression model, applying the intellectual property risk assessment model to enterprise intellectual property management, performing risk early warning on the existing intellectual property data in real time, and taking measures in time for response when a risk early warning prompt is identified.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise management, and particularly to an intelligent management system for enterprise intellectual property based on service big data. Background Art

[0002] In today's knowledge-based economy era, intellectual property has become an important part of an enterprise's core competitiveness. For enterprises, effective intellectual property management can not only protect their own innovation achievements, but also bring huge commercial value to the enterprises. However, there are many problems in the traditional enterprise intellectual property management methods.

[0003] In terms of data processing, existing management systems are difficult to comprehensively and accurately count the massive intellectual property data.

[0004] In terms of data query, the traditional system has low query efficiency and cannot meet the diverse query needs of different user roles;

[0005] In addition, for the risk assessment of intellectual property, the traditional method often relies on manual experience judgment, lacks a scientific and quantitative assessment model, and is difficult to accurately and timely conduct risk assessment and early warning, bringing potential economic losses to the enterprises. Summary of the Invention

[0006] The purpose of the present invention is to overcome the deficiencies of the prior art and provide an intelligent management system for enterprise intellectual property based on service big data to solve the problems raised in the above background art and achieve efficient and intelligent management of enterprise intellectual property.

[0007] The purpose of the present invention can be achieved through the following technical solutions: An intelligent management system for enterprise intellectual property based on service big data, including:

[0008] A data statistics module, used for counting various data of enterprise intellectual property and preprocessing the counted various data to obtain intellectual property processed data;

[0009] A data management module, used for implementing a data quick query function by using the inverted index algorithm and assigning different query permissions according to the roles and responsibilities of users;

[0010] A feature extraction module, used for extracting features from the intellectual property processed data to obtain key feature data;

[0011] A risk assessment module, used for constructing an intellectual property risk assessment model according to the logistic regression model, taking the key feature data as the input features of the model, and dividing them into a training set and a validation set according to a preset ratio; training, optimizing and evaluating the logistic regression model through the training set and the validation set respectively to obtain an intellectual property risk assessment model;

[0012] Apply the intellectual property risk assessment model to the enterprise's intellectual property management, conduct real-time risk early warning on the existing intellectual property data, and take timely measures to respond when risk early warning prompts are identified.

[0013] Preferably, the data statistics module specifically includes:

[0014] Classify and statistically analyze the various data of the enterprise's intellectual property collected according to different dimensions, including: classify the data into patent data, trademark data, and copyright data according to the type of intellectual property;

[0015] Sort the data in time series respectively according to the application time and authorization time;

[0016] Classify the data according to the technical field to which the patent belongs.

[0017] Preferably, preprocess the statistically analyzed data, including:

[0018] Clean the data, use regular expressions to clean the illegal characters in the patent name, and remove duplicate, incorrect or incomplete data; standardize the data according to a unified standard, and use a data mapping table to unify the data format and coding method to obtain the processed intellectual property data.

[0019] Preferably, the data management module specifically includes:

[0020] Obtain the processed intellectual property data, perform word segmentation on the text content in the processed intellectual property data, decompose the text into individual words or phrases, and record the position information where each word or phrase appears;

[0021] When inputting keywords for retrieval, quickly find the document list containing the keywords through the inverted index algorithm, and sort the results according to the term frequency-inverse document frequency algorithm, and display the information to the user in turn;

[0022] Implement user authentication and authorization mechanisms to require users to authenticate their identities when logging in to the system. The system determines their permission levels according to the user's identity information, and monitors and audits the user's query operations in real time to ensure the secure use of data.

[0023] Preferably, the method for extracting key feature data is as follows:

[0024] Determine the various data indicators that affect intellectual property risks based on the understanding of the enterprise's business; use the decision tree algorithm to calculate the information gain of each data indicator, and mark it as the importance score. By viewing the importance scores of all features, screen out the features with importance scores greater than or equal to the score standard as the key feature data.

[0025] Preferably, the expression of the logistic regression model is as follows:

[0026]

[0027] In the formula, P(Y = 1|X) represents the probability that the output variable Y is equal to 1 when the input is X = (x 1 , x 2 ,..., x n ), that is, the predicted probability that this sample belongs to the existence of intellectual property rights risks; β i is the weight parameter corresponding to the input feature; β 0 is the intercept term; exp is the natural exponential function; among them, X = (x 1 , x 2 ,..., x n ) is the key feature data of the input sample.

[0028] Preferably, the model is trained and optimized using the training set, including:

[0029] Input the training set samples into the model for training, obtain the predicted probability P output by the model, and compare and judge by setting the risk threshold Q with the predicted probability P output to obtain the prediction result;

[0030] If the predicted probability P output ≥ the risk threshold Q, it means that this training set sample has intellectual property rights risks, and a risk warning prompt is given for this sample; otherwise, it means that this training sample has no intellectual property rights risks.

[0031] Preferably, it further includes:

[0032] Calculate the difference between the model prediction result and the true label through the logarithmic loss function, and continuously adjust the weight parameters of the model using the stochastic gradient descent optimization algorithm to minimize the calculated loss function value;

[0033] Among them, the expression of the logarithmic loss function is as follows:

[0034]

[0035] In the formula, L is the loss function value; M is the number of samples; Y m is the true label of the mth sample. When Y m = 1, it means that this sample belongs to the existence of intellectual property rights risks. When Y m = 0, it means that this sample belongs to the non-existence of intellectual property rights risks; represents the prediction result of the mth sample.

[0036] Preferably, the model is evaluated using the validation set, including:

[0037] Count the number of samples in the statistical validation set where the true label and the model prediction result are consistent; through the formula Calculate the accuracy rate of the model prediction; where, H is the accuracy rate, YZ is the number of samples in the validation set where the true label and the model prediction result are consistent; ZS is the total number of samples in the validation set;

[0038] If the calculated accuracy rate is greater than or equal to 85%, it means that the performance of the logistic regression model meets the standard, and an intellectual property risk assessment model is obtained; otherwise, it is necessary to continue to adjust the model parameters using the validation set data until the prediction accuracy rate of the model reaches the standard.

[0039] Compared with the existing solutions, the beneficial effects achieved by the present invention are as follows:

[0040] The present invention provides a solid data foundation for enterprise intellectual property management by counting key information such as the quantity, application time, and authorization status of various types of intellectual property; realizes fast data query by using the inverted index algorithm, and at the same time, assigns different query permissions according to the roles and responsibilities of users, ensuring the security and confidentiality of data;

[0041] The present invention constructs an intellectual property risk assessment model by adopting a logistic regression model, comprehensively considers the influence of various factors on intellectual property risks, and provides objective and accurate risk assessment results for enterprises; when the risk indicators of intellectual property exceed the preset threshold, the system will issue a risk warning prompt in a timely manner to remind relevant personnel of the enterprise to take corresponding measures to deal with it. This risk warning mechanism can help enterprises discover potential risks in advance, adjust intellectual property management strategies in a timely manner, and reduce the losses caused by risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The following further describes the present invention with reference to the accompanying drawings.

[0043] Figure 1 It is a module structure diagram of an enterprise intellectual property intelligent management system based on service big data proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0045] As Figure 1 shown, the present invention is an enterprise intellectual property intelligent management system based on service big data, including: a data statistics module, a data management module, a key feature extraction module, and a risk assessment module;

[0046] The data statistics module is used to count various data of enterprise intellectual property and preprocess the counted various data to obtain intellectual property processed data;

[0047] Classify and statistically analyze the collected enterprise intellectual property data according to different dimensions, including: according to the type of intellectual property, divide the data into patent data, trademark data, and copyright data;

[0048] Sort the data in time series according to the application time and authorization time respectively;

[0049] Classify the data according to the technical fields to which the patents belong, such as information technology, biomedicine, and mechanical manufacturing;

[0050] Preprocess the statistically analyzed data, including: cleaning the data, using regular expressions to clean illegal characters in the patent names, and removing duplicate, incorrect, or incomplete data; standardizing the data according to a unified standard, using a data mapping table to unify the data format and encoding method to obtain the intellectual property processed data;

[0051] In the embodiments of the present invention, by statistically analyzing and preprocessing the enterprise intellectual property data, it is beneficial to provide reliable data support for subsequent refined analysis and management of enterprise intellectual property.

[0052] The data management module is used to implement the function of quickly querying data by using the inverted index algorithm, and allocate different query permissions according to the roles and responsibilities of users;

[0053] Obtain the intellectual property processed data, perform word segmentation on the text content in the intellectual property processed data, decompose the text into individual words or phrases, and record the position information where each word or phrase appears;

[0054] When inputting keywords for retrieval, quickly find the document list containing the keywords by querying the inverted index algorithm, and sort the results according to the term frequency-inverse document frequency algorithm, and display the information to the user in turn;

[0055] Implement user authentication when logging in to the system through the user authentication and authorization mechanism. The system determines the user's permission level according to the user's identity information, and monitors and audits the user's query operations in real time to ensure the safe use of data; among them, the authentication methods include but are not limited to user name / password, digital certificate;

[0056] In the embodiments of the present invention, calculate the correlation score between each document and the search keywords through the TF-IDF algorithm, rank the documents with high correlation in front of the search results, improve the search efficiency and accuracy; establishing a perfect user permission management system is beneficial to the security of intellectual property.

[0057] The feature extraction module is used to extract features from the intellectual property processed data to obtain key feature data, including:

[0058] Determine various data indicators that affect intellectual property risks based on the understanding of enterprise business; among them, the various data indicators include, but are not limited to, the number of patent applications, the number of patent citations, and the proportion of R & D investment.

[0059] Use the decision tree algorithm to calculate the information gain of each data indicator and mark it as the importance score. By viewing the importance scores of all features, select the features with importance scores greater than or equal to the score standard as the key feature data; among them, the score standard is set according to the specific requirements in the enterprise intellectual property management in actual applications.

[0060] It should be noted that the decision tree algorithm adopted is the existing ID3 algorithm, which will not be elaborated here; among them, the greater the information gain, the greater the role of the feature in predicting the target variable, and the higher its importance.

[0061] In the embodiment of the present invention, by calculating the importance scores of each data indicator and screening out the key feature data according to the set score standard, the key feature data is effectively screened out from the numerous data indicators that affect intellectual property risks, which not only improves the efficiency of risk management, but also ensures that the enterprise can concentrate on monitoring the data indicators corresponding to the key feature data.

[0062] A risk assessment module for constructing an intellectual property risk assessment model according to a logistic regression model to achieve intelligent risk assessment.

[0063] Among them, the expression of the logistic regression model is as follows:

[0064]

[0065] In the formula, P(Y = 1|X) represents the probability that the output variable Y is equal to 1 when the input is X = (x 1 , x 2 ,..., x n ), that is, the prediction probability that the sample belongs to the existence of intellectual property risks; β i is the weight parameter corresponding to the input feature; β 0 is the intercept term; exp is the natural exponential function; among them, X = (x 1 , x 2 ,..., x n ) is the key feature data of the input sample.

[0066] Use the key feature data as the input feature of the model and divide it into a training set and a validation set according to a preset ratio; among them, the preset ratio is set to 8:2.

[0067] Use the training set to train and optimize the model, including:

[0068] Input the training set samples into the model for training, obtain the predicted probability P output by the model, and compare and judge by setting the risk threshold Q with the output predicted probability P to obtain the prediction result;

[0069] If the output predicted probability P ≥ the risk threshold Q, it indicates that the training set sample has intellectual property risks, and a risk warning prompt is given for this sample; otherwise, it indicates that the training sample has no intellectual property risks; among them, the risk threshold is jointly determined according to specific business requirements and knowledge in this field;

[0070] Calculate the difference between the model prediction result and the true label through the logarithmic loss function, and continuously adjust the weight parameters of the model using the stochastic gradient descent optimization algorithm to minimize the calculated loss function value;

[0071] Among them, the expression of the logarithmic loss function is as follows:

[0072]

[0073] In the formula, L is the loss function value; M is the number of samples; Y m is the true label of the m-th sample. When Y m = 1, it indicates that this sample belongs to the category with intellectual property risks. When Y m = 0, it indicates that this sample belongs to the category without intellectual property risks; represents the prediction result of the m-th sample;

[0074] Use the validation set to evaluate the model, including:

[0075] Count the number of samples in the validation set where the true label and the model prediction result are consistent; calculate the accuracy rate of the model prediction through the formula ; among them, H is the accuracy rate, YZ is the number of samples in the validation set where the true label and the model prediction result are consistent; ZS is the total number of samples in the validation set;

[0076] If the calculated accuracy rate is greater than or equal to 85%, it indicates that the performance of the logistic regression model meets the standard, and an intellectual property risk assessment model is obtained; otherwise, it is necessary to continue to adjust the model parameters using the validation set data until the prediction accuracy rate of the model reaches the standard;

[0077] Apply the intellectual property risk assessment model to enterprise intellectual property management, and conduct real-time risk warnings on existing intellectual property data. When a risk warning prompt is identified, take timely measures to respond. For example, arrange professional personnel to rectify the data with intellectual property risks;

[0078] In the embodiments of the present invention, an intellectual property risk assessment model is constructed based on a logistic regression model. By reasonably dividing data, scientifically constructing and optimizing the model, and strictly evaluating the model performance, relatively effective intelligent assessment of intellectual property risks is achieved.

[0079] In several embodiments provided by the present invention, it should be understood that the disclosed system can be implemented in other ways. For example, the above-described embodiments of the invention are merely illustrative. For example, the division of modules is only a logical function division, and there may be other division methods in actual implementation.

[0080] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0081] In addition, in each embodiment of the present invention, the functional modules can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a hardware plus software functional module.

[0082] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above-described exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent management system for enterprise intellectual property rights based on service big data, characterized by: include: The data statistics module is used to collect statistics on various data of the enterprise's intellectual property rights and pre-process the statistical data to obtain intellectual property processing data; The data management module is used to implement fast data query using the inverted index algorithm and assign different query permissions based on the user's role and responsibilities; A feature extraction module is used to extract features from intellectual property processing data to obtain key feature data; The risk assessment module is used to construct an intellectual property risk assessment model based on the logistic regression model, taking the key feature data as the input feature of the model and dividing it into a training set and a validation set according to a preset ratio; the logistic regression model is trained, optimized and evaluated through the training set and the validation set respectively to obtain an intellectual property risk assessment model; Apply the intellectual property risk assessment model to corporate intellectual property management, provide real-time risk warnings for existing intellectual property data, and take timely measures to respond when risk warning prompts are identified.

2. According to claim 1, the enterprise intellectual property intelligent management system based on service big data is characterized in that: The data statistics module specifically includes: Classify and count the collected data on corporate intellectual property rights according to different dimensions, including: classifying the data into patent data, trademark data, and copyright data according to the type of intellectual property rights; Sort the data in time series according to the application time and authorization time; The data is classified according to the technical field to which the patent belongs.

3. The enterprise intellectual property intelligent management system based on service big data according to claim 2 is characterized in that: Preprocess the statistical data, including: Clean the data, use regular expressions to clean illegal characters in patent names, and remove duplicate, erroneous or incomplete data; standardize the data according to unified standards, use data mapping tables to unify data formats and encoding methods, and obtain intellectual property processing data.

4. The enterprise intellectual property intelligent management system based on service big data according to claim 3 is characterized in that: The data management module specifically includes: Acquire intellectual property processing data, perform word segmentation on the text content in the intellectual property processing data, decompose the text into individual words or phrases, and record the position information of each word or phrase; When a keyword is entered for retrieval, the inverted index algorithm is used to quickly find a list of documents containing the keyword, and the results are sorted according to the word frequency-inverse document frequency algorithm, and the information is displayed to the user in sequence; Through the user authentication and authorization mechanism, users are required to authenticate their identities when logging into the system. The system determines their authority level based on their identity information, and monitors and audits their query operations in real time to ensure the safe use of data.

5. According to claim 4, the enterprise intellectual property intelligent management system based on service big data is characterized in that: The method of extracting key feature data is as follows: Based on the understanding of the enterprise business, the various data indicators that affect intellectual property risks are determined; the information gain of each data indicator is calculated using the decision tree algorithm and marked as an importance score. By checking the importance scores of all features, the features with importance scores greater than or equal to the scoring criteria are screened out as key feature data.

6. The enterprise intellectual property intelligent management system based on service big data according to claim 5 is characterized in that: The expression of the logistic regression model is as follows: Where P(Y=1|X) means when the input is X=(x1,x2,...,x n ), the probability that the output variable Y is equal to 1, that is, the predicted probability that the sample is subject to intellectual property risk; β i is the weight parameter corresponding to the input feature; β0 is the intercept term; exp is the natural exponential function; where X = (x1, x2, ..., x n ) is the key feature data of the input sample.

7. The enterprise intellectual property intelligent management system based on service big data according to claim 6 is characterized in that: Use the training set to train and optimize the model, including: Input the training set samples into the model for training, obtain the predicted probability P output by the model, and compare and judge the output predicted probability P by setting the risk threshold Q to obtain the prediction result; If the output prediction probability P ≥ risk threshold Q, it means that the training set sample has intellectual property risk, and a risk warning prompt is issued for the sample; otherwise, it means that the training sample does not have intellectual property risk.

8. The enterprise intellectual property intelligent management system based on service big data according to claim 7 is characterized in that: Also includes: The difference between the model prediction result and the true label is calculated through the logarithmic loss function, and the weight parameters of the model are continuously adjusted using the stochastic gradient descent optimization algorithm to minimize the calculated loss function value; Among them, the expression of the logarithmic loss function is as follows: Where L is the loss function value; M is the sample size; Y m is the true label of the mth sample, when Y m =1, it means that the sample has intellectual property risk. m =0, it means that the sample has no intellectual property risk; Represents the prediction result of the mth sample.

9. The enterprise intellectual property intelligent management system based on service big data according to claim 8 is characterized in that: The model is evaluated using the validation set, including: Count the number of samples in the validation set that are consistent with the true labels and model predictions; by formula Calculate the accuracy of the model prediction; where H is the accuracy, YZ is the number of samples in the validation set that are consistent with the true labels and the model prediction results; ZS is the total number of samples in the validation set; If the calculated accuracy is greater than or equal to 85%, it means that the performance of the logistic regression model meets the standard and an intellectual property risk assessment model is obtained; otherwise, it is necessary to use the validation set data to continue to adjust the model parameters until the model's prediction accuracy meets the standard.

Images