KR101932517B1 - Method and system of grading patent using multiple regression analysis - Google Patents

Abstract

The present invention relates to a method and system for evaluating a patent using a multiple regression model. The method includes acquiring patent information, processing patent information, and calculating a plurality of regression coefficients, And generating an evaluation model for the evaluation index by calculating a representative value of a plurality of regression coefficients of the plurality of multiple regression models for each independent variable calculated through the multiple regression analysis .

Description

[0001] METHOD AND SYSTEM FOR GRADING PATENT USING MULTIPLE REGRESSION ANALYSIS [0002]

The present invention relates to a patent evaluation method and system using a multiple regression model, and more particularly, to a patent evaluation method using a multiple regression model that calculates a plurality of patent evaluation indexes statistically using multiple regression analysis, ≪ / RTI >

In recent years, there have been many efforts and achievements in intellectual property rights for technology protection in Korea compared to other developed countries. In this process, it is true that the intellectual property rights holders have a lot of intellectual property rights, and the cost and effort to maintain and maintain the intellectual property rights are one of the burden.

Therefore, intellectual property owners are assessing their intellectual property rights by rating their own intellectual property rights, or requesting commercial or nonprofit organizations to evaluate patented technologies.

These patent evaluations include professional evaluation and automatic evaluation. Expert assessment is the assessment of the value of the patents evaluated by experts in the field of technology. Expert evaluation has the advantage of being able to utilize individual expert knowledge of experts and qualitative valuation, but it has a disadvantage that it is likely that the subjectivity of expert is strongly involved in evaluation and that it is costly and time consuming.

In order to overcome the disadvantages of the expert evaluation, a patent automatic evaluation system has been developed. However, a method of evaluating the patent rating of the sample patents for designing the evaluation model by an expert evaluation or using only a few pieces In many cases, an evaluation model is designed. Even if multiple regression models are used, only one regression model is used. Therefore, the reliability of the evaluation result is often questioned.

Meanwhile, the background art of the present invention is disclosed in Korean Patent Laid-open Publication No. 10-2014-0080593 (Apr.

The present invention provides a multiple regression model that allows a patent evaluation to be carried out by constructing a highly reliable evaluation model that appropriately reflects the evaluation factors considering the relative environment between technically similar patents, while reflecting the structural characteristics of the patent specification And to provide a method and system for evaluating the utilized patent.

A method for evaluating a patent using a multiple regression model according to the present invention includes: acquiring patent information; Performing a multiple regression analysis on each of the plurality of core evaluation factors previously set for the evaluation index as dependent variables by processing the patent information; And generating an evaluation model for the evaluation index by calculating a representative value of a regression coefficient of a plurality of multiple regression models for each independent variable calculated through the multiple regression analysis.

The patent evaluation method using the multiple regression model according to the present invention is characterized by further comprising the step of calculating a numerical evaluation index of the patent to be evaluated using the evaluation model.

The step of generating the evaluation model according to the present invention is characterized by calculating a weighted average or an arithmetic mean of the regression coefficient as the representative value.

In the present invention, the multiple regression model utilizes an evaluation factor having a significance probability of 0.05 or less.

In the present invention, each step of performing the multiple regression analysis is characterized in that a multiple regression analysis is performed by using an evaluation factor preset for each core evaluation element as an independent variable of each multiple regression model.

The present invention is characterized in that an evaluation model is generated for each of a plurality of evaluation indexes, wherein the evaluation index includes at least one of rightness, technicality and usability.

The patent evaluation system utilizing the multiple regression model according to the present invention includes a patent information processor for processing the obtained patent information to extract an evaluation factor; A multiple regression analysis processing unit for performing a multiple regression analysis in which each of a plurality of core evaluation factors previously set for an evaluation index is a dependent variable to calculate regression coefficients of a plurality of multiple regression models for each independent variable; And an evaluation model generation unit for calculating a representative value of a plurality of regression coefficients of a plurality of multiple regression models for each of the independent variables calculated through the multiple regression analysis processing unit to generate an evaluation model for the evaluation index .

A method and system for evaluating a patent using a multiple regression model according to the present invention includes constructing a multiple regression model for each of a plurality of core evaluation factors and evaluating the evaluation model for the patent evaluation index through a combination of regression coefficients of a plurality of multiple regression models It is possible to appropriately reflect an evaluation factor considering the relative environment between technically similar patents while reflecting the structural characteristics of the patent specification.

Accordingly, the method and system for evaluating a patent using the multiple regression model according to the present invention have an effect of enabling cheap and rapid evaluation of a large number of patents under an objective evaluation standard.

Also, the patent evaluation method and system using the multiple regression model according to the present invention has an effect of generating evaluation information for a plurality of evaluation items for one patent.

FIG. 1 is a diagram illustrating a schematic configuration of a patent evaluation system utilizing a multiple regression model according to an embodiment of the present invention.
2 is a diagram illustrating a detailed configuration of a server of a patent evaluation system utilizing a multiple regression model according to an embodiment of the present invention.
FIG. 3 is a diagram illustrating an exemplary operation of a patent evaluation system using a multiple regression model according to an embodiment of the present invention.
4 is a flowchart illustrating a process of generating an evaluation model in a patent evaluation method using a multiple regression model according to an embodiment of the present invention.
FIG. 5 is a flowchart illustrating a process of performing a patent evaluation in a patent evaluation method using a multiple regression model according to an embodiment of the present invention.
FIG. 6 is an exemplary view showing a detailed configuration of a server of a patent evaluation system utilizing a multiple regression model according to another embodiment of the present invention.
FIG. 7 is a flowchart illustrating a process of performing a patent evaluation in a patent evaluation method using a multiple regression model according to another embodiment of the present invention.
8 is a flowchart illustrating a process of providing a patent evaluation service in a patent evaluation method using a multiple regression model according to another embodiment of the present invention.

Hereinafter, embodiments of a method and system for evaluating a patent using a multiple regression model according to the present invention will be described with reference to the accompanying drawings. In this process, the thicknesses of the lines and the sizes of the components shown in the drawings may be exaggerated for clarity and convenience of explanation. In addition, the terms described below are defined in consideration of the functions of the present invention, which may vary depending on the intention or custom of the user, the operator. Therefore, definitions of these terms should be made based on the contents throughout this specification. Furthermore, the singular < RTI ID = 0.0 > terms < / RTI > used herein may include a plurality of representations unless the context clearly dictates otherwise.

FIG. 1 is a diagram illustrating a schematic configuration of a patent evaluation system utilizing a multiple regression model according to an embodiment of the present invention.

1, a patent evaluation system using a multiple regression model according to an embodiment of the present invention may be configured with at least one server 100, and the server 100 may be connected to a wired / So that the user device 200 can provide the result of the patent evaluation. That is, when the server 100 receives a request for an evaluation service for a specific patent from the user device 200, the server 100 can provide an evaluation result of the patent, and a detailed operation method will be described later.

The server 100 according to an embodiment of the present invention may include a processor, a memory for storing and executing program data, a persistent storage, a communication port for communicating with an external device, a user interface device, and the like. Methods implemented with software program modules or algorithms may be stored on a computer readable recording medium as computer readable codes or program instructions executable on the processor. The computer-readable recording medium may be distributed over networked computer systems so that computer readable code in a distributed manner may be executed and executed.

2 is a diagram illustrating a detailed configuration of a server of a patent evaluation system utilizing a multiple regression model according to an embodiment of the present invention.

2, the server 100 includes a patent information processing unit 110, a multiple regression analysis processing unit 120, an evaluation model generation unit 130, an evaluation model DB 140, and a patent evaluation processing unit 150 . According to an embodiment of the present invention, these components may be distributed over one or more servers connected to a wired / wireless network.

The patent information processing unit 110 can extract the evaluation element by processing the obtained patent information.

For example, the patent information processing unit 110 can collect patent information from an external data provider. Specifically, the server 100 may include a data collection unit (not shown) for receiving raw data from a foreign data provider in Korea or overseas (for example, the United States). The data collecting unit may physically include a network interface (NIC), and may logically be a program made up of an application programming interface (API).

The patent information processing unit 110 can extract and parse (or convert) each piece of information from the obtained patent information. For example, the patent information processing unit 110 may extract information such as patent specification, bibliographic information, progress information, and drawings, and the specification may be written in, for example, an XML format. Can be extracted.

The server 100 can acquire patent information on a plurality of patents, and the patent information processor 110 can extract evaluation elements from respective patents and store the extracted data in a DB (not shown).

On the other hand, if the obtained patent information is already processed information (for example, the evaluation element is extracted and provided in advance), the patent information processing unit 110 may omit the process of extracting the evaluation factor.

The multiple regression analysis processing unit 120 constructs a multiple regression model in which each of a plurality of core evaluation factors set in advance for the evaluation index is used as a dependent variable among evaluation factors and performs a multiple regression analysis to determine a plurality of A regression coefficient of a multiple regression model can be calculated, and a more detailed operation method will be described later.

The evaluation model generation unit 130 may generate a evaluation model for evaluation indexes by calculating a representative value of a plurality of regression coefficients of a plurality of multiple regression models for each independent variable calculated through the multiple regression analysis processing unit 120, For example, a weighted average or an arithmetic mean of the regression coefficients can be calculated as the above representative value.

The evaluation model generation unit 130 may generate (build) an evaluation model for an evaluation index by combining regression coefficients of independent variables of a plurality of multiple regression analysis modules into a weighted average (or arithmetic average) It is possible to store the model in the evaluation model DB 140 so that the patent evaluation can be performed.

The patent evaluation processing unit 150 can calculate a numerical evaluation index (e.g., an evaluation score or an evaluation grade) of the patent to be evaluated by using the evaluation model stored in the evaluation model DB 140. Specifically, Average), it is possible to calculate the numerical evaluation index of the patent by inputting the information of the corresponding patent to each independent variable of the multiple regression analysis model combined with the index, It is also possible to calculate a representative value of the evaluation.

FIG. 3 is a diagram illustrating an exemplary operation of a patent evaluation system using a multiple regression model according to an embodiment of the present invention.

That is, the server 100 extracts and processes the collected patent information, extracts evaluation factors, generates evaluation models for each evaluation index (or technology field) through multiple regression analysis, and uses the generated evaluation models to evaluate the patent To provide a patent analysis service to the user.

FIG. 4 is a flowchart illustrating a process of generating an evaluation model in a patent evaluation method using a multiple regression model according to an embodiment of the present invention. FIG. 5 is a flowchart illustrating a process of generating an evaluation model using a multiple regression model according to an exemplary embodiment of the present invention A flowchart for explaining a process of performing a patent evaluation in a patent evaluation method will be described with reference to FIG. 1, and a description will be made of a patent evaluation method using a multiple regression model according to an embodiment of the present invention.

As shown in FIG. 4, the server 100 first obtains patent information (S300), and extracts evaluation elements from the obtained patent information (S310). If the obtained patent information is already processed information, the step of extracting the evaluation element (S310) may be omitted.

Subsequently, the server 100 performs a multiple regression analysis on each of a plurality of core evaluation factors related to the evaluation index (S320). That is, the server 100 can construct a plurality of multiple regression models each having a plurality of core evaluation factors previously set for the evaluation index as dependent variables and an evaluation factor as an independent variable, and perform multiple regression analysis for each core evaluation factor .

In this case, since the core evaluation factor is a factor belonging to the evaluation factor and multiple regression analysis is performed on a plurality of core evaluation factors, when a multivariate analysis is used, one core evaluation factor can not be used as an independent variable for other core evaluation factors . Therefore, in the present invention, instead of using general multivariate analysis, a plurality of regression analysis equations are constructed to perform multiple regression analysis with each of a plurality of core evaluation factors as dependent variables.

Specifically, it is possible to construct a plurality of multiple regression analysis equations as shown in the following Equation 1 to perform regression analysis.

는 오차를 의미하고, 오른쪽 항에서 위첨자는 종속변수의 순번을, 아래첨자는 독립변수(평가요소)의 순번을 의미한다.Here, y denotes a multiple regression model for the core evaluation factor, the subscript denotes the order of the dependent variable (core evaluation factor), β ₀ to β _n denotes the regression coefficient, and β ₀ denotes a constant In addition,

, The superscript in the right-hand side denotes the order of the dependent variable, and the subscript denotes the order of the independent variable (evaluation factor).

Next, the server 100 calculates a representative value of a regression coefficient (including a constant) for each independent variable to generate an evaluation model for the evaluation index (S330). That is, since a plurality of multiple regression equations are constructed for one evaluation index, a final evaluation model for the evaluation index can be constructed by combining regression coefficients. For example, a weighted average or an arithmetic average of the regression coefficient To generate an evaluation model.

Specifically, when combining with a weighted average of regression coefficients, an evaluation model can be generated as shown in Equation 2 below.

Here, Y denotes a numerical evaluation index, and? Denotes a weight value.

On the other hand, at this time, each multiple regression identification can be configured to utilize only the evaluation factors having a probability less than the reference value, and the exclusion of such evaluation factors can be performed in real time in the multiple regression analysis process. According to one embodiment of the present invention, the reference value of the significance probability is 0.1 or less, preferably 0.05.

In addition, evaluation factors can be excluded by multiloccurrence tests between evaluation elements. In other words, if there is a strong correlation between independent variables and a mutual exclusion violation occurs, the evaluation factor can be excluded. In the present embodiment, the evaluation factors excluded by the multi- Quot; number ".

In addition, evaluation factors to be used as independent variables may be preset for each core evaluation element, and a specific evaluation element is excluded in such a manner that a regression coefficient for an evaluation factor excluded from a specific core evaluation factor is fixed to 0 in the corresponding multiple regression equation .

Meanwhile, the patent evaluation method using the multiple regression model according to the present embodiment may be configured to generate evaluation models for a plurality of evaluation indicators, respectively. For example, the evaluation index may include at least one of rightness, . ≪ / RTI >

Here, the rightness means the degree to which the patent to be evaluated can maintain exclusive exclusive status in patent disputes with a third party, and the technicality means that the patent to be evaluated conforms to the technology trend, leading degree, It means the extent to which a patent is utilized in business and its potential for utilization.

The present embodiment reflects the structural characteristics of the patent specification, and is a core evaluation factor for evaluating the rightness so that the evaluation factors considering the relative environment between technically similar patents can be appropriately reflected. (I) Inter Partes Review (IPR); (ii) Post Grant Review (PGR); (iii) Determination of whether the application is a split application or a continuation application; (Hereinafter referred to as "litigation"), and whether or not the patent application continues to be filed, the total number of patents used, and the CPC At least one of the patent increase / decrease rate (for example, US patent increase rate in the case of US patent analysis) at the level of the national patent office may be used, There is still pending, whether litigation or not, international family information (overseas family filed whether or overseas family filed countries can), at least one of whether to extend, and whether the duration standard patent (Standard Essential Patent) as a key evaluation factors can be used for groups. At this time, the standard patents can be used for patents in the fields of electricity, electronics, and IT, and the extension of the duration can be used for patents in the chemical field.

The key evaluation factors may be considered different factors depending on the characteristics of each country's patent law and patent system. For example, in the European patent evaluation, at least one of the right limitation procedure, the number of complainants, the division application, and the total number of citations can be used as a key evaluation factor for evaluating the rightness, The number of citations, the number of complainants, and the number of citations can be used as a core evaluation factor for evaluating the applicability. Whether or not the license is granted, the number of complainants, , And a right limiting procedure may be used.

Meanwhile, the patent evaluation method using the multiple regression model according to the present embodiment may include, for example, a US patent increase rate of a CPC level, a number of interrogation procedures, a number of IPCs, an IPR, a number of PGRs, The number of reexaminations, the number of reissues, the number of re-enrollment, the number of patent holder changes, and the number of re-enrollment (reexamination) The number of inventors, the number of inventors, the number of years of patent registration, the number of years of patent registration, the priority claim for review, the precedent cited by the examiner (hereinafter referred to as " (The average of the period from the filing date of the patent application to the current date of the patent application), the number of information provided, the duration of the patent term, the average depth of the patent term, the number of claims, Number of citations, standard patents, cited references Monju, cited differences with the filing date, international family information (number of countries, whether or not) can be designed to use such as evaluation factors. The CPC described in this embodiment refers to the Cooperative Patent Classification and the IPC refers to the International Patent Classification.

In this case, in the case of the model excluding the evaluation factor according to the probability probability, the IPR · PGR number, the number of interferences, and the number of information provided can be excluded from the evaluation factor in the case of the model in which the US patent increase rate of the CPC level is a dependent variable. In the case of a model with fixed number as a dependent variable, the US patent increase rate, the number of interferences, the number of intervening patents, the number of drawings, the length of the description of the invention, the average age of the patented patents, Number of patents, standard patents, and number of overseas family members may be excluded from the evaluation factors.

In the case of a model with continuous application as a dependent variable, the number of IPR · PGR determinations, the number of RCEs, and the number of claims can be excluded from the evaluation factors. In the case of a model with the dependent variable, And the number of series can be excluded from the evaluation factor. In the case of the model in which the total number of citations is used as a dependent variable, the number of information provided can be excluded from the evaluation factor. In the case of the model having the standard patent as a dependent variable, PGR number, number of interferences, number of RCEs, number of reexaminations, number of information provided, number of articles in cited documents can be excluded from the evaluation factors. In the case of a model with dependent variable of overseas family information, IPR, PGR number (pending), number of interferences, number of information provided, etc. may be excluded from the evaluation factors.

Meanwhile, the patent evaluation method using the multiple regression model according to the present embodiment is characterized in that, in generating the evaluation model for evaluating the domestic (Korean) patent, the number of the independent factors, the length of the independent claim, the average depth of the dependent claims, The length of the description of the invention, the number of filing applications, the number of overseas family members, the number of IPCs, the early disclosure, the priority review request, the number of comments submitted, the number of information provided, the total number of cited documents, The number of foreign patents, the number of foreign patents, the number of annual patent registrations, the number of inventors, the number of inventors, the number of patent holders, the number of patent holders, the number of pledges of financial institutions, · Number of appeals, number of appeals against appeal, number of appeals against appeal, number of referees cited, number of referees cited, confirmation of scope of active rights refusal of referee, withdrawal, The number of citations, withdrawals, ex officio, and whether or not a correction is made can be used as an evaluation factor.

As described above, the patent evaluation method using the multiple regression model according to the present embodiment can be used to generate evaluation models for evaluating patents of countries such as Europe, USA, Korea, Japan, and China.

After generating the evaluation model through the above process, as shown in FIG. 5, in order to provide the patent evaluation service, the server 100 receives the identification information of the patent to be evaluated from the user device (S400). For example, the server 100 may receive, as identification information, an application number or a registration number of a patent to be evaluated.

Subsequently, the server 100 calculates a numerical evaluation index of the patent using the evaluation model generated in the step S330 of FIG. 4 (S410). In other words, the server 100 can acquire an evaluation factor of the patent to be evaluated, substitute the obtained evaluation factor into the evaluation model, and calculate the corresponding numerical evaluation index. Also, at this time, the server 100 may calculate a plurality of numerical evaluation indexes (e.g., rightness, technicality, usability) for the evaluation target patent by using the evaluation model for each of the plurality of evaluation indicators.

On the other hand, when calculating a plurality of digitized evaluation indexes in step S410, the server 100 calculates a representative value of a plurality of digitized evaluation indexes as a representative value of the patent evaluation (S420). That is, for example, the server 100 can calculate a representative value (e.g., a weighted average or an arithmetic average) of the calculated rightness, technicality, and usability, and can calculate the rating of the patent through the calculated value have.

FIG. 6 is a diagram illustrating a detailed configuration of a server of a patent evaluation system using a multiple regression model according to another embodiment of the present invention, and FIG. 7 is a diagram illustrating a patent evaluation using a multiple regression model according to another embodiment of the present invention. FIG. 8 is a flow chart for explaining a process of providing a patent evaluation service in a patent evaluation method using a multiple regression model according to another embodiment of the present invention, A method of evaluating a patent using a multiple regression model according to another embodiment of the present invention will be described with reference to the following.

6, the patent evaluation system includes a patent information processing unit 110, a multiple regression analysis processing unit 120, an evaluation model generation unit 130, an evaluation model DB 140, And may further include an evaluation result DB 160 in addition to the processing unit 150. [

Here, the configurations and operations of the patent information processing unit 110, the multiple regression analysis processing unit 120, the evaluation model generation unit 130, the evaluation model DB 140, and the patent evaluation processing unit 150 are the same as those described with reference to FIG. 2 .

The patent evaluation processing unit 150 performs the evaluation of the collected patents automatically, and notifies the evaluation result DB 160 of the evaluation result, instead of performing the patent evaluation only when there is a patent evaluation request through the user device 200. [ Lt; / RTI > In this case, when there is a patent evaluation request through the user device 200, the evaluation result previously stored in the evaluation result DB 160 is extracted and output, thereby minimizing the time for outputting the evaluation result for the evaluation request.

According to an embodiment of the present invention, the evaluation processing unit 150 can update the evaluation result stored in the evaluation result DB 160 by performing the evaluation of the patent collected at a predetermined date and time again. Even if the same patent is evaluated, the evaluation result may differ depending on the evaluation point.

7 and 8, the operation of the patent evaluation method using the multiple regression model according to another embodiment of the present invention will be described.

As shown in Fig. 7, the operations from step S600 to step S630 may be the same as those according to Fig.

After the step S630, the server 100 evaluates the patent obtained in the step S600 using the evaluation model generated in the step S630 (S640), and stores the evaluation result (S650). In other words, the server 100 may perform evaluation of the collected patents in advance and store them in advance of the user's request, in addition to the evaluation model construction using the collected patents.

In this case, as shown in FIG. 8, when the identification information of the patent to be evaluated is received from the user device 200 (S700), the server 100 reads the stored evaluation result and transmits it to the user device 200 (S710 ). That is, in the process of providing a patent evaluation service, the patent evaluation may be performed in a manner in which a previously-stored patent evaluation is provided rather than being performed in real time.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. I will understand. Accordingly, the technical scope of the present invention should be defined by the following claims.

100: Server
110: Patent Information Processing Department
120: multiple regression analysis processor
130: Evaluation model generation unit
140: Evaluation model DB
150: Patent evaluation processor
160: Evaluation result DB
200: User device

Claims (15)

A method for evaluating a patent utilizing a multiple regression model, the method being performed in a server including a patent information processor, a multiple regression analysis processor, and an evaluation model generator,
Obtaining the patent information by the server;
Wherein the server processes the patent information and performs a plurality of multiple regression analyzes each having a plurality of core evaluation factors previously set for the evaluation index as dependent variables, The evaluation factors are subjected to multiple regression analysis as independent variables of respective multiple regression models, wherein the core evaluation factor is an element belonging to the evaluation factor, and one core evaluation factor is independent of multiple regression models for other core evaluation factors Performing a plurality of multiple regression analyzes, each of which can be used as a variable; And
And generating an evaluation model for the evaluation index by calculating a representative value of a regression coefficient of a plurality of multiple regression models for each independent variable calculated through the multiple regression analysis by the server,
Wherein each of the representative values calculated in the evaluation model is a coefficient for each independent variable of the evaluation model.
The method according to claim 1,
Wherein the server further comprises a patent evaluation processing unit,
Further comprising a step of the server calculating a numerical evaluation index of a patent to be evaluated using the evaluation model generated.
The method according to claim 1,
Wherein the server further comprises a patent evaluation processing unit,
The server calculating a numerical evaluation index of patents collected using the evaluation model; And
Further comprising the step of the server storing the calculated evaluation index in an evaluation result DB.
The method according to claim 1,
Wherein the server generates a weighted average or an arithmetic average of the regression coefficient as the representative value in the step of generating the evaluation model.
The method according to claim 1,
Wherein the server utilizes an evaluation factor whose significance probability is less than or equal to a preset reference value in the multiple regression model.
delete The method according to claim 1,
Wherein the server performs multiple collinearity tests between the evaluation elements to exclude one or more evaluation factors in each step of performing the multiple regression analysis.
The method according to claim 1,
Wherein the server generates an evaluation model for a plurality of evaluation indicators,
Wherein the evaluation index includes at least one of rightness, technicality, and applicability.
delete A patent information processor for processing the obtained patent information and extracting an evaluation factor;
A multiple regression analysis processing unit for performing a plurality of multiple regression analyzes each having a plurality of core evaluation factors set in advance as the dependent variable for the evaluation index to calculate regression coefficients of a plurality of multiple regression models for each independent variable; And
And an evaluation model generation unit for calculating a representative value of a plurality of regression coefficients of a plurality of multiple regression models for each of the independent variables calculated through the multiple regression analysis processing unit to generate an evaluation model for the evaluation index,
Wherein the multiple regression analysis processor performs a multiple regression analysis using the evaluation factors preset for each core evaluation element as independent variables of each multiple regression model, wherein the core evaluation factor is an element belonging to the evaluation factor, Wherein the evaluation element can be used as an independent variable of a multiple regression model for other core evaluation elements and each of the representative values calculated in the evaluation model is a coefficient for each independent variable of the evaluation model. Patent evaluation system utilized.
11. The method of claim 10,
And a patent evaluation processing unit for calculating a numerical evaluation index of the patent to be evaluated by using the evaluation model thus generated.
12. The method of claim 11,
And an evaluation result DB for storing the evaluation result of the patent,
Wherein the patent evaluation processing unit performs an evaluation of the patent on which the information is acquired and stores the evaluation result in the evaluation result DB.
13. The method of claim 12,
Wherein the patent evaluation processing unit performs evaluation of the patent on which the target information is acquired at a preset time point and stores the evaluation result in the evaluation result DB.
11. The method of claim 10,
Wherein the evaluation model generation unit calculates a weighted average or an arithmetic average of the regression coefficients as the representative value.
delete

Images