JP5655305B2 - Patent power calculation device and operation method of patent power calculation device - Google Patents

Description

  The present invention relates to a patent profit calculating apparatus that calculates patent profit using patent history data.

  Conventional evaluation methods are roughly classified into a monetary evaluation method for calculating the value per patent right and a relative evaluation method performed in a think tank or the like.

  In the conventional monetary evaluation method, various methods such as a scoring DCF method, a Black Scholes model, a cost approach, and a market approach have been used. Since these methods are financial and economic evaluations, there is a high demand in the transfer of patent rights, etc., but subject to qualitative analysis (scoring), and let us evaluate all patent rights. Then, there are problems such as enormous costs.

  On the other hand, in the relative evaluation method, a statistical evaluation method that analyzes the number of patents owned, registration rate, number of applications, number of claims, etc., a method that evaluates the technical value from word analysis of patent specifications, technical system diagrams, etc. Is used. Since these methods are data-centered evaluation methods, objectivity is easily secured and their functions are demonstrated in situations such as comparing relative technical capabilities with competitors. There was a problem that it was difficult to specify the causal relationship with the monopoly power of patent rights). For example, it is well known that a company with a large number of patents does not necessarily have high profitability.

In addition, as in Patent Document 1, market information consisting of market data, financial data, or marketing data about a product for which the patent is implemented in calculating the patent value, commercialization power, technical power, and rights of the patented invention. Power, modern social adaptability and overall power rating, and calculate the profit creation output index of the target patent from the profitability of the product and the contribution of profit of the target patent. Inventions have been made to obtain the theoretical price of a target patent at the time of evaluation based on the ability to form a product market in consideration of the risk and the risk rate and expansion capability of the target patent.
JP 2005-174313 A

  However, when using a monetary evaluation method such as Patent Document 1, it is necessary to examine market information and the scope of rights for each patent, and it is not possible to evaluate the value of many patents at once. difficult. In the relative evaluation method, as described above, it is difficult to specify the causal relationship between the evaluation item and the economic activity (the exclusive power of patent rights). Therefore, it is difficult to evaluate the value of patent groups, and no appropriate method has been developed so far. However, in recent years when intellectual property is said to occupy most of the corporate value, a standard indicator for confirming the results of intellectual property management is required.

    Therefore, the inventor uses a patent power calculation device and a patent power calculation device as a device for calculating a patent power according to an evaluation method capable of specifying a causal relationship with an economic activity (a monopoly exclusive power of a patent right). An operation method (JP 2009-3727 A) has been proposed.

    It is an object of the present invention to provide a patent profit calculation apparatus that can use this patent power calculation apparatus and further know the correlation between a patent and a profit generated by the patent.

In the present invention, the following patent profit calculation apparatus is proposed. In the following description, the term “patent” may include a pending application.

Specifically, as the first invention,
A patent acquisition unit that acquires a patent ID that is a plurality of application numbers or patent numbers as a population,
A patent history data acquisition unit for acquiring patent history data of the acquired patent ID;
Search for a combination of standard item names indicating the legal procedure taken for the patent described in the acquired patent history data by pattern matching using a pattern prepared in advance, and the combination of the searched standard item names And an item content extraction unit that extracts the item content described in the patent history data in association with the procedure date,
A search result holding unit that holds the extracted item contents and the procedure date associated therewith in association with the combination of the searched standard item names;
A cost table holding unit that holds a cost table that associates costs prepared in advance for each combination of item contents held in association with a combination of standard item names;
An obsolete function storage section that stores an obsolete function that serves as an indication of obsolete technology for each technical field;
For each combination of item contents extracted according to the combination of the standard item names of each patent , the cost is acquired using the cost table held in the cost table holding unit, and the calculation reference date and the combination of the item contents a procedure date associated with each, and filing date of the patent, and obsolescence after cost calculating unit that calculates the cost after obsolescence of obsolescence function and the calculation reference date with the art to which this patent belongs,
A summation unit that sums all calculated post-stale costs for patents ;
A totaling section for further summing up the totaled values for each name of the applicant or right holder,
A business profit value acquisition unit for acquiring a business profit value for each name;
A correlation calculation unit that calculates the correlation between the aggregated value for each name and the business revenue value for each name in the population using a plurality of pairs of the aggregated value for each registered name and the business revenue value for each acquired name;
The patent profit calculation apparatus which has is provided.

As a second invention,
Instead of the aggregate value for each name, the aggregate value for each name excluding nominal name, which is the aggregate value for each name excluding the influence of the factor representing the company size of each name,
The patent profit calculation device according to the first invention, wherein the business profit value for each name is a business profit value for each name excluding the influence of a factor representing the company size of each company instead of the business profit value for each name. I will provide a.

As a third invention,
The first business revenue calculation unit according to the first invention or the second invention, further comprising a first business revenue calculation unit that calculates a business revenue value corresponding to one or more names using the correlation calculated by the correlation calculation unit. A patent profit calculation device is provided.

As a fourth invention,
The first invention or the second invention further includes a second business revenue calculation unit that calculates a business revenue value corresponding to one or more patent IDs using the correlation calculated by the correlation calculation unit. A patent profit calculation apparatus is provided.

  With the patent profit calculation apparatus of the present invention, it is possible to know the correlation between a patent and the profit generated by the patent.

Functional block diagram of the patent profit calculation apparatus according to the first embodiment An example of part of patent history data Example cost table Diagram explaining how to calculate the obsolescence function Example of obsolescence function An example of a table containing a plurality of pairs of YK values, which are examples of aggregate values for each name, and operating profits, which are examples of business revenue values for each name An example of a scatter diagram using the values in the table of FIG. An example of a scatter diagram (logarithmic display) using the values in the table of FIG. An enlarged view of a part of the scatter diagram in FIG. Hardware configuration diagram of patent profit calculating apparatus according to Embodiment 1 The flowchart which shows the flow of a process of the patent profit calculation apparatus which concerns on Embodiment 1. FIG. Functional block diagram of patent profit calculation apparatus according to Embodiment 2

The form for implementing this invention is demonstrated below. Note that the present invention is not limited to these embodiments, and can be implemented in various modes without departing from the scope of the present invention. In the first embodiment, the invention according to claims 1 and 2 will be mainly described. In the second embodiment, the invention according to claims 3 and 4 will be mainly described.
<< Embodiment 1 >>
<Embodiment 1: Overview>

  The patent profit calculating apparatus according to the present embodiment is characterized in that a correlation between a total value for each holder who totals the calculated total value for each holder and a business profit value for each holder is calculated.

The calculated total value is an index value representing the patent power of one patent or the like that is distinguished from others by a patent ID that is an application number or a patent number, as will be described later. Here, the meaning of the total value calculated by the patent profit calculation apparatus of the present embodiment will be briefly described. The total value is calculated per patent . The patent profit calculation apparatus according to the present embodiment is based on the idea that the value of a patent lies in exclusive exclusivity, and is based on a completely new method that has been mainly aimed at directly measuring the exclusive exclusivity. To calculate. Here, the monopoly power is the power that shows how the patent owner monopolizes the business, in other words, the power that shows how the patent right is an obstacle to the business of competitors. This exclusive exclusivity can be compared to a fence or fence that creates boundaries with other companies. It doesn't make much sense to create fences where there are no enemies or where no one is interested. In other words, even if a fence is made on an uninhabited island, it cannot be used to prevent third parties from entering, so it is meaningless. However, creating a fence where there are actual enemies is of great significance. The significance is even greater if it is a splendid habit that a third party cannot fully enter. In other words, it is of great significance to take a large territory in a crowded place in the middle of Tokyo and make a bag to completely prevent third parties from entering. The act of preventing this third party's entry is the exclusion of competitors, taking a large territory means a wide range of rights, and a splendid trap means a patent that is hard to be invalidated.

  Having a strong patent right with a wide range of rights in a business where many competitors collide has a strong monopoly power. Having a strong monopoly power in the market is a source of profit for the patentee. In other words, evaluating the exclusive power of a patent is considered to be synonymous with evaluating the profitability of a patent.

  Next, the method for evaluating the monopoly power will be explained. With patents with monopoly exclusivity, there are some other parties that must be excluded in order for the patentee to monopolize the business. Therefore, we will consider the actions that the other party to exclude should take against disability patents with exclusive exclusive power.

  What actions will you take if you discover a patent right that hinders your business? First of all, it will be necessary to examine the contents of the patent right and decide whether to negotiate the license, whether it is going to be crushed, or to change the design. At that time, some action should be taken against the patent right. Therefore, it is desirable that the patent power calculation apparatus according to the present embodiment evaluates an action from a third party for this patent right.

  Various actions are taken against patents from the time an invention is actually made, to filing, publication, examination, registration, and extinction. For example, a request for examination, a notice of reasons for refusal, a patent decision or a decision of refusal, a request for browsing, a trial for refusal of decision of refusal, a trial for invalidation, etc. Among these various actions, the third-party action is a request for browsing for knowing patent examination progress information, a trial for invalidation requested to invalidate the patent right, or the like. The patent power calculation apparatus according to the present embodiment can index the exclusive power of patent rights by evaluating actions from such third parties (competitors).

  Now, I will explain why it is desirable to limit the evaluation to only third party actions. For example, the action of the applicant (right holder) rather than a third party is “application”. The action “Apply” is not included in our evaluation. This is because a company that applied for many patents can never be said to have high profitability through patents. For example, if most of the applied patents are deemed to be withdrawn without requesting examination, or if they are rejected in examination and do not become patents, there is no point in filing a large number of applications. Absent. Moreover, if one's own action is included in the evaluation target, it becomes possible to arbitrarily change one's own evaluation.

  However, patents that competitors cannot ignore the existence of, and patents that must be investigated, patents that it is difficult to avoid patents as a result of the investigation and want to invalidate by making a trial for invalidation It can be said that these are highly valuable.

  The value calculated as the total value is not limited to the case where an action from a third party is an evaluation target. For example, based on the idea of the cost approach, it may be calculated by adding the costs that the company has applied to the patent.

Since the calculated total value is indexed, it is possible to estimate a regression equation that expresses the relationship between the aggregate value for each holder and the business revenue value for each holder. The patent profit calculation apparatus according to the present embodiment is characterized in that
<Embodiment 1: Configuration>

  FIG. 1 illustrates a functional block diagram of the patent profit calculation apparatus according to the present embodiment. The patentability calculation apparatus (0100) shown in FIG. 1 includes a “patent acquisition unit” (0101), a “patent history data acquisition unit” (0102), an “item content extraction unit” (0103), Part ”(0104),“ cost table holding part ”(0105),“ obsolete function storage part ”(0106),“ post-stale cost calculation part ”(0107), and“ summation part ”(0108) A “aggregation unit” (0109), a “business profit value acquisition unit” (0110), and a “correlation calculation unit” (0111).

  The “patent acquisition unit” (0101) acquires a patent ID. The patent ID is an application number or a patent number of a plurality of patents serving as a population.

  The “patent history data acquisition unit” (0102) has a function of acquiring patent history data of the acquired patent ID. The patent history data is organized standardized data obtained by organizing and processing various information such as examination progress information held by the JPO in Japan. The patent history data includes, in addition to examination progress information, actions such as invalidation trials that have been filed against registered patent rights. Further, not only the history of patents but also examination progress information and post-registration information on utility models, designs, and trademarks are included. By referring to the patent history data, it is possible to know information such as the filing date, applicant, inventor, right holder, IPC, etc., whether there is an application examination request, and the examination progress status.

The “item content extraction unit” (0103) performs pattern matching processing using a pattern in which a combination of standard item names indicating legal procedures taken for patents described in the acquired patent history data is prepared in advance. It has a function of searching and extracting item contents described in the patent history data in association with the procedure date according to the combination of the searched standard item names.

Here, the legal procedure taken against the patent is a request for browsing or a trial for invalidation.

The combination of standard item names indicating legal procedures taken against patents, for example, in the case where legal proceedings that invalidation trial were taken for the patent, the referee type, trial final disposal type, decision of It is a combination of standard item names such as the conclusion of a decision article. By this combination, a legal procedure called invalidation trial is searched by performing pattern matching processing on the patent history data. In the patent history data, data for searching for a legal procedure called invalidation trial and finding out the result are scattered, so it is necessary to perform pattern matching processing to extract item contents and the like.

  Next, a method for extracting item contents described in the patent history data according to the combination of retrieved standard item names will be described. FIG. 2 shows an example of a part (0200) of the patent history data. The left side is the standard item name (0201), and the right side is the item content (0202). In the case of FIG. 2, the item content for the standard item name “trial type” is “112 (all invalid (new invalid))”, and the item content for the standard item name “trial final disposal type” is “02 ( The item content for the “conclusion” in the standard item name “determination decision article” is “Y (not invalid)”.

  And the date when the procedure was performed is extracted in association with these item contents. For example, if the trial is invalid, the “request date for appeal” is extracted.

The “search result holding unit” (0104) has a function of holding the extracted item content and the procedure date associated therewith in association with the combination of the searched standard item names. That is, taking FIG. 2 as an example, the item contents 112 (all invalid (new invalid)), 02 (billing not established), Y (not invalid) are associated with the combination of standard item names, and the procedure date is 2004/04. / 01 is also stored in association with the combination of standard item names. By referring to the stored search results, it is possible to determine the date of the trial for invalidation trial, which is the legal procedure taken against the patent , the trial type, the final disposal type, and the conclusion of the decision article of the trial decision.

  The “cost table holding unit” (0105) has a function of holding a cost table in which a cost prepared in advance is associated with each combination of item contents held in association with a combination of standard item names. An example of the cost table is shown in FIG. The first line in FIG. 3 shows a combination of standard item names. For example, the standard item name combinations for invalidation referees are a referee type, a final disposition type of a referee, a decision article for a trial decision, and the like. In the second and third lines, examples of item content combinations are shown. First, the example on the second line is a case where a trial for invalidation has occurred, the final disposition is unsuccessful, and the trial decision is not invalidated. In this case, the cost that the third party has tried for invalidation, for example, 1,000,000 (million) yen is held in the cost table as a cost. The example on the third line is a case where an invalidation trial occurs, the final disposition is a claim, and the trial decision is invalid. In this case, the patent becomes invalid and it is considered that the patent has no value, and zero is stored as a cost in the cost table. The cost described in the cost table may be a monetary unit, a value divided by an appropriate value, or an index corresponding to the legal procedure.

  The “obsolete function storage unit” (0106) has a function of storing an obsolete function that is a measure of the obsolete technology for each technical field.

  The obsolescence function is obtained as follows. The upper diagram in FIG. 4 is a graph showing statistics on how many years a patent right has expired in a certain technical field. The vertical axis is the percentage of patent rights that have been extinguished, and the horizontal axis is the number of years since filing. One of the characteristics of this statistical data is that it is based on the time of filing. Perhaps it is natural that technology obsolescence does not start when the right is registered, but it starts at the peak when the invention is made. Therefore, it seems that it is most correct to start from the moment of invention. However, since the statistics cannot be taken, it starts from the time of filing. Looking closely at the top of FIG. 4, it can be seen that the patent rights that are extinguished by the fourth year after the application are almost 0 (zero), and thereafter, the patent rights that gradually disappear are increasing. And the patent right corresponding to 25% to 30% of registered patents will be extinguished 20 years after the application. This is because the duration of the patent right is in principle 20 years from the filing date. If the lifetime is longer than 20 years, all patent rights that would have been maintained for a longer period have expired in the 20th year. The obsolete technology did not happen at the 20th year. Therefore, in accordance with the hypothesis that the patent rights that have been extinguished in the 20th year were gradually extinguished over the years after the 20th year, the ratios shown in the circles in the lower diagram of FIG. I predicted that it would disappear year by year. This is the main graph that represents the obsolescence of technology. FIG. 5 is a graph obtained by approximating the lower diagram of FIG. 4 with a normal distribution and calculating “1-normal cumulative distribution”. This curve is the obsolescence function. This can also be called a technological value obsolescence curve. Here, the reason for approximating the ratio of extinguished patent rights with a normal distribution will be briefly described. Registered patents are established after the judgment of inventive step is made for each. In other words, other patents will not become obsolete due to the fact that one patent has become obsolete as technology advances. Therefore, each patent is considered to be independent and can be approximated by a normal distribution.

  According to FIG. 5, if there is no delimitation that the lifetime is 20 years, almost all patents lose the value maintained in about 25 years from the filing date. This graph is characterized by little staleness in the first few years, but the rate of staleness accelerates as it approaches the number of years that the average patent expires, and then becomes stale once the average number of years expires That is. This function is calculated for each technical field and stored as an obsolete function.

The “post-stale cost calculation unit” (0107) acquires the cost using the cost table held in the cost table holding unit for each combination of item contents extracted according to the combination of the standard item names of each patent. obsolescence of calculation reference date used with a calculation reference date, and the procedure dates are associated with each combination of the item contents, and the filing date of that patent, the obsolescence function in the art to which this patent belongs to It has a function to calculate the post cost.

First, a method for acquiring a cost using a cost table held in a cost table holding unit for each combination of item contents extracted according to a combination of standard item names of each patent will be described. As shown in FIG. 3, the cost table holding unit holds a cost for each combination of item contents corresponding to a combination of standard item names indicating legal procedures. Therefore, the cost table is searched by the combination of the extracted item contents, and the cost of the matching combination is acquired.

Next, a method for calculating the post-stale cost using the stale function will be described with reference to FIG. First, the technical field to which the patent belongs is obtained, and the corresponding obsolescence function is obtained. Then, the calculation reference date, the procedure date associated with each combination of the item contents, and the filing date of the patent are acquired. In the case of a retroactive application, the original filing date may be acquired as the filing date. The reason for this is that, as mentioned above, technology obsolescence does not start when the right is registered, but it starts at the peak when the invention is made.

Assume that a patent invalidation trial has been requested for a certain patent right in the α-year from the application, but a maintenance trial decision has been made. According to the cost table, the series of procedures is 100 points. Furthermore, it is assumed that the calculation reference date is β years from the filing date. In this case, if the technical value residual coefficient in year α is T (α) and the technical value residual coefficient in year β is T (β), the post-stale cost on the calculation base date is
Cost after obsolescence = 100 × T (β) / T (α)
Can be calculated as

  Considering the calculation reference date as the present, if the action date (α year) is 2 years from the filing date and the current date (β year) is 3 years from the filing date, the date is almost obsolete. And if the action date (α year) is 2 years from the filing date and the current date (β year) is 15 years from the filing date, the obsolescence is great. In other words, the legal procedure that took place in the past is a smaller value when it is reduced to the current post-obsolete cost.

The “summing unit” (0108) has a function of summing up the calculated post-stale costs for all patents . Thereby, it is possible to calculate the patent power of one patent .

  The “counting unit” (0109) has a function of further summing up the summed values for each name of the applicant or right holder.

  The “business profit value acquisition unit” (0110) has a function of acquiring the business profit value for each name. The business revenue value includes not only operating income and ordinary income figures but also market capitalization and free cash flow (FCF). In companies that are not obligated to submit securities reports, if financial data such as operating profit and recurring profit, market capitalization, etc. cannot be obtained, available business revenue is used instead.

  The “correlation calculation unit” (0111) uses a plurality of pairs of the aggregated value for each nominal name and the acquired business revenue value for each nominal name to calculate the aggregated value for each nominal name and the business revenue value for each nominal name in the population. It has a function to calculate the correlation.

  Correlation refers to the relationship between two variables. For example, when there are n pairs of data (X1, Y1)... (Xn, Yn), there is a positive correlation when X increases and Y increases. When X increases, there is a negative correlation when Y decreases. When X changes with respect to changes in X and does not correspond to the change, or when it is constant Uncorrelated. In addition, a correlation coefficient is used as an index indicating the correlation. If it is close to 1, there is a positive correlation, if it is close to -1, there is a negative correlation, and if it is close to 0, the correlation is weak.

  FIG. 6 is an example of a table including a plurality of pairs of YK values, which are examples of aggregate values for each name, and operating profits, which are examples of business revenue values for each name. YK value and operating income are averaged over 5 years.

  FIG. 7 is an example of a scatter diagram using the values in the table of FIG. The correlation coefficient of each dot (plot) shown in this figure is about 0.85, and the correlation is high.

  FIG. 8 is an example of a scatter diagram in which FIG. 6 is displayed in common logarithm with base 10 for the total value for each name and the business profit value for each name. Thus, the display of the scatter diagram can be performed in various forms. For example, logarithmic display is used. If the logarithmic display is used, it is possible to display the scatter diagram in an easy-to-see manner when the dots (plots) are close to the origin. The correlation coefficient of each dot (plot) shown in this figure is about 0.81, and the correlation is high.

  The calculation of the correlation means that a regression equation representing the relationship between a plurality of variables is estimated by regression analysis. A variable to be explained is called an explanatory variable, and a variable for explaining the explained variable is called an explanatory variable.

FIG. 9 is an enlarged view of a part of the scatter diagram of FIG. This figure shows an example of a regression equation representative of the dispersion of data composed of the business revenue value for each name, which is the explained variable shown in the table of FIG. 6, and the aggregate value for each name, which is the explanatory variable. This regression equation is estimated as a first-order approximation of the least square method. As shown in this figure, in the single regression analysis with one explanatory variable, the regression equation uses the aggregated value S for each name, the business revenue value R for each name, and constants α, β,
R = α + βS
Can be expressed by a first-order approximation formula. For example, in FIG.
Since S = 0.05997 × R,
R = S / 0.05997 = about 16.67500 × S
It becomes.

By calculating the correlation between a patent and the profit it generates, various analyzes are performed, such as whether the company's patent is linked to profits, what problems it has, and which companies are helpful Is possible. That is, it is possible to compare the business revenue value predicted based on the aggregate value for each name, which is an index value indicating patentability, and the actual business revenue value for each name acquired by the business revenue value acquisition unit described above. . For example, since the G1 company YK value in FIG. 6 is 11169.93, using the above regression equation,
G1 forecast operating profit R '(G1)
= 16.67500 x 11169.93 = 186258.58275 (million yen)
It becomes. This is substantially the same as the actual operating profit 183659 (million yen) of G1 company. On the other hand, the YK value of company H1 is 11139.33, so using the above regression equation,
H1's forecast operating profit R '(H1)
= 16.67500 x 11139.33 = 1857748.32775
It becomes. This is higher than H1's actual operating profit of 31829.4 million yen. From this, it is possible to analyze that company H1 has a problem in that patents are not tied to profits, and the activities of company G1, which is considered to have a patent power of approximately the same level, are problematic. It will be possible to analyze whether it is useful for the solution.

  Further, the target for calculating the correlation can be limited to a specific technical field. For example, it is a technical field such as an in-vehicle fuel cell group. In this case, further detailed analysis is possible. For example, if the holders a and b have the same relationship in the technical field such as the in-vehicle fuel cell group, the difference in profitability is brought about in the technical field such as the in-vehicle fuel cell group. It can be analyzed if there is.

  Furthermore, the explanatory variable is not limited to one. It is conceivable to use a plurality of explanatory variables in addition to the calculated value for each name. For example, the number of employees, company age, total asset turnover ratio, tangible assets per capita, R & D cost ratio, etc., or trend variables for taking into account time fluctuations or industry dummy variables for taking into account industry-specific trends And so on. In this case, it is desirable that the correlation between each explanatory variable is low. This is because there are cases where inconvenience may occur in the interpretation of the results if those having strong correlation with each other are used as explanatory variables. This problem is called multicollinearity.

  In addition, the reference date for calculating each explanatory variable such as the total value for each name and the business profit value for each name may be the same or different. For example, the correlation is calculated for the business profit value for each name after five years from the calculation reference date data of each explanatory variable. In this way, it is possible to perform an analysis according to the characteristics of the data even for a variable in which a certain time lag occurs until the profits are increased after the capital investment.

  Also, as each variable for calculating the correlation, instead of the aggregated value for each name, the aggregated value for each nominal name that excludes the influence of the factor representing the company size of each name, and the aggregated value for each nominal name, excluding the influence of the factor representing the company size of each name, Instead of the value, a configuration may be used in which the business profit value for each name is a business profit value for each name excluding the influence of the factor representing the company size of each company.

  The influence of the scale factor means the influence of the company size on each variable. Exclusion includes the case where the effect is not completely eliminated but reduced. For example, it is considered that large companies generally tend to have a larger number of patents held, but if the number of patents held is large, the aggregated value for each name is likely to increase. Therefore, for example, by dividing the aggregate value for each name by the number of patents held, and using the average patent power per patent as a variable, the analysis should be performed by representing the overall patent power of the company. And so on. Alternatively, the number of employees, the number of inventors, the number of applications, sales, R & D expenses, etc. may be used as the scale factor.

Further, for example, instead of the business profit value, the sales growth rate or the profit rate may be used as the business profit value for each name excluding the scale factor.
<Embodiment 1: Hardware configuration>

  FIG. 10 is a schematic diagram showing an example of a configuration when each component of the patent profit calculation apparatus according to the present embodiment is expressed as hardware.

  All or a part of each component that is a component of the present embodiment is configured by hardware, software, or both hardware and software. For example, as an example for realizing these, when a computer is used, there are hardware constituted by a CPU, a bus, a memory, an interface, a peripheral device, and the like, and software that can be executed on the hardware. As software, the functions of each unit are realized by sequentially executing a program developed on the memory, and processing, storing, and outputting data on the memory and data input via the interface.

  More specifically, as shown in FIG. 10, the computer includes a CPU (1001), a RAM (1002), a ROM (1003), an input / output interface (I / O) (1004), an HDD (1005), and the like. They are connected to each other via a data communication path such as a system bus (1006) to transmit / receive information and process information.

  The RAM (1002) reads a program for performing various processes for execution by the CPU and simultaneously provides a work area for the program. A plurality of memory addresses are assigned to the RAM (1002) and the ROM (1003), and programs executed by the CPU (1001) exchange data with each other by specifying and accessing the memory addresses. It is possible to perform processing.

  The operation of the hardware configuration unit in this embodiment will be described with reference to FIG.

  First, when the power source of the patent profit calculation apparatus is activated, the CPU (1001) causes the patent acquisition program, the patent history data acquisition program, the item content extraction program, the search result held in the storage device such as the ROM (1003). Various programs such as a holding program, an obsolete cost calculation program, a summation program, a tabulation program, a business profit value acquisition unit program, and a correlation calculation program are developed in the work area of the RAM (1002).

  The CPU (1001) executes the patent acquisition program and acquires the patent ID of the patent to be evaluated. The patent ID is an application number or a patent number of a plurality of patents serving as a population. The acquisition of the patent ID may be to acquire a patent ID input by the user, or may be to acquire a patent ID included in the scope of the patent to be evaluated. The range of the patent to be evaluated may be determined in advance or may be input by the user. For example, all the patent IDs are acquired from the stored standardized data, or only the patent ID associated with the IPC symbol input by the user is acquired. The acquired patent ID is, for example, a part of the above-described standardized data. This data is held in a storage area such as a ROM (1003), for example. The acquired patent ID is held in the storage data area of the RAM (1002).

  Next, the CPU (1001) executes a patent history data acquisition program and acquires patent history data of the acquired patent ID. For example, as described above, what is associated with the acquired patent ID is acquired from the patent history data held in the storage area such as the ROM (1003). The acquired patent history data is held in the storage data area of the RAM (1002).

Next, the CPU (1001) executes the item content extraction program and reads the pattern file held in the storage area such as the ROM (1003) into the storage data area of the RAM (1002). In the pattern file, as described above, combinations of standard item names indicating legal procedures taken for patents are prepared in advance. Then, a pattern matching process using the pattern file is performed on the patent history data held in the storage data area of the RAM (1002), thereby searching for a combination of standard item names indicating legal procedures. Then, the item content and the procedure date corresponding to the combination of the standard item names are extracted. For example, for the above-mentioned legal procedure of trial for invalidation of a patent , search for a combination of standard item names such as the type of trial, the type of final judgment of trial, the conclusion of the decision article of the trial decision, and if applicable, the name of each standard item For example, the item contents and the procedure date are extracted from the retained patent history data.

  Next, the CPU (1001) executes the search result holding program, associates the extracted item contents and the procedure date with the combination of the standard item names of the pattern file described above, and the storage data area of the RAM (1002). Hold on.

Next, the CPU (1001) executes an obsolete cost calculation program. At this time, the CPU (1001) reads the cost table and the obsolete function held in the storage area such as the ROM (1003) into the storage data area of the RAM (1002). Then, a cost corresponding to the search result held in the storage data area is acquired by searching the cost table. Further, the obsolescence function is held, for example, in a form corresponding to each classification symbol of the IPC subclass, and the IPC classification associated with the patent ID from the acquired patent history data is used as the technical field to which the patent belongs. What is necessary is just to acquire the obsolescence function corresponding to this acquired IPC classification. Then, the calculation reference date is acquired by a method such as a user input or by executing a program for acquiring patent profit calculation current in advance, and associated with each combination of the item contents from the held search results. And the application date of the patent associated with the patent ID is acquired from the acquired patent history data. As described above, when the cost is 100, the procedure date is α year from the application, and the calculation reference date is β year from the application, α obtained by substituting these values into the obsolescence function If the technical value residual coefficient for the year is T (α) and the technical value residual coefficient for the year β is T (β), the post-stale cost on the calculation base date is
Cost after obsolescence = 100 × T (β) / T (α)
Is calculated as follows. The calculated post-stale cost is held in the storage data area of the RAM (1002).

Then, the CPU (1001) executes the summation program, and sums all the post-stale costs held in the storage data area of the RAM (1002) for the patent . The calculated total value is held in the storage data area of the RAM (1002).

  Next, the CPU (1001) executes a totaling program, extracts the same value of the applicant or right holder from the sum of the values stored in the storage data area of the RAM (1002), and the applicant or right holder. Aggregate by name. The aggregated total value for each name is stored in the storage data area of the RAM (1002).

  Further, the CPU (1001) executes a business profit value acquisition program, and the business profit corresponding to the same name as the aggregate value for each name from the business profit value data stored in the storage area such as the ROM (1003). Get the value. The business profit value data includes at least the name of each holder corresponding to the above-mentioned aggregate value for each name and the business profit value. As described above, the business revenue value acquired is the average of the business revenue values for multiple years corresponding to the name, or the operating profit and sales corresponding to the name, and the operating profit / sales amount is calculated. It may be a profit rate obtained by calculation. The acquired business profit value for each name is held in the storage data area of the RAM (1002).

  Then, the CPU (1001) executes the correlation calculation program, and uses a plurality of pairs of the aggregated value for each nominal name and the acquired business revenue value for each nominal name, and the aggregated value for each nominal name and the business revenue for each nominal name in the population. Calculate the correlation with the value. For example, as described above, the regression equation may be calculated as a linear approximation formula of the least square method. Since the specific procedure is the same as that generally performed by a commercially available program or the like, the description is omitted. The calculated regression equation is held in a storage data area of the RAM (1002).

  Furthermore, the total value for each name, the business revenue value for each name, the calculated regression equation, etc. may be drawn in the form of a graph or the like and displayed via an input / output interface (1004) such as a display. In this way, visual analysis is possible.

  In addition, as described above, each variable for calculating the correlation is configured as an aggregate value for each name excluding nominal name, which is an aggregate value for each name excluding the influence of the factor representing the company size of each name, instead of the aggregate value for each name. You may take In this case, although not shown in the figure, the CPU executes a scale factor acquisition program and acquires, for example, the number of owned patents as the scale factor for each name as described above. For example, the number of patents held may be held corresponding to each holder in the business profit value data held in the storage area such as the ROM (1003). Or you may make CPU (1001) count how many patent numbers are contained for each holder about the arrangement | sequence standardization data mentioned above. The acquired scale factor is held in the storage data area of the RAM (1002). Then, the CPU (1001) calculates the aggregated value for each scale factor exclusion name by dividing the aggregated value for each nominal by the scale factor associated with the same name as the aggregated value for each nominal name. The calculated total value for each scale factor exclusion name is stored in the storage data area of the RAM (1002). And in the above-mentioned correlation calculation program, it replaces with the aggregated value for each nominal name, and this aggregated value for each nominal name exclusion name is used.

  Further, as described above, instead of the business profit value for each name, the business profit value for each name excluding the name, which is the business profit value for each name excluding the influence of the factor representing the company size of each company, may be used. In this case, when executing the business profit value acquisition program, the CPU (1001) acquires the sales growth rate and the profit rate as the business profit value for each name excluding the scale factor in place of the business profit value for each name. What should I do? The acquired business profit value for each scale factor exclusion name is stored in the storage data area of the RAM (1002). And in the above-mentioned correlation calculation program, it replaces with the acquired business profit for each name, and uses the business profit value for each name excluding the scale factor.

  <Embodiment 1: Processing flow>

  FIG. 11 is an example showing a processing flow of the operation method of the patent profit calculation apparatus according to the present embodiment.

First, in step S1101, a patent ID is acquired. Next, in step S1102, patent history data of the acquired patent ID is acquired. Next, in step S1103, a combination of standard item names indicating legal procedures is searched by pattern matching processing using a pattern prepared in advance. Next, in step S1104, the item contents are extracted in association with the procedure date according to the combination of the standard item names retrieved from the patent history data. In step S1105, the extracted item contents and the procedure date associated therewith are stored in association with the combination of standard item names. In step S1106, the cost table is searched to obtain the corresponding cost. In step S1107, an obsolescence function, a calculation reference date, a procedure date, and an application date corresponding to the technical field to which the patent belongs are acquired. Next, in step S1108, the post-stale cost is calculated using the acquired cost, the stale function, the calculation reference date, the procedure date, and the application date. Next, in step S1109, all the calculated post-stale costs are added up for the patent . Next, in step S1110, for the total value, those having the same name of the applicant or right holder are extracted and totaled for each name of the applicant or right holder. Next, in step S1111, the business profit value for each name is acquired. Next, in step S1112, the correlation between the aggregate value for each name and the business revenue value for each name in the population is calculated using a plurality of pairs of the aggregate value for each nominal name and the acquired business revenue value for each name.

  In addition, as described above, each variable for calculating the correlation is configured as an aggregate value for each name excluding nominal name, which is an aggregate value for each name excluding the influence of the factor representing the company size of each name, instead of the aggregate value for each name. You may take In this case, for example, in step S1110, after summing up the total values for each name, a step of obtaining a scale factor for each name and dividing the total value for each name by this size factor may be added. Further, as described above, instead of the business profit value for each name, the business profit value for each name excluding the name, which is the business profit value for each name excluding the influence of the factor representing the company size of each company, may be used. In this case, for example, in step S1111, instead of acquiring the business profit value for each name, the sales growth rate or the profit rate may be acquired as the business profit value for each name excluding the name of the scale factor.

Note that the flowchart of FIG. 11 can also be regarded as a processing flowchart of a program executed by a computer. Further, such a program can be recorded on a medium such as a CD or an IC memory.
<Embodiment 1: Effect>

  The patent profit calculation apparatus according to the present embodiment can know the correlation between a patent and the profit generated by the patent.

  In evaluating a patent, the index indicating the patent power is a macro-evaluation and high-resolution data, and has the characteristic of completely eliminating arbitraryness.

  In other words, until now, it took enormous costs (eg, about 3 million yen per case) and time to micro-evaluate the economic value of each patent. Micro evaluation was considered difficult. Here, the micro evaluation is to conduct a detailed investigation on one patent and calculate its economic value. In the patent profit calculation device of the present invention, the action taken as a result of the third party investigating the disability patent and evaluating the degree of disability in his / her business is subject to evaluation. It will be evaluated. The degree of business discomfort felt by a third party is scattered in the progress information, and since it is an evaluation target, it is a macro evaluation and high resolution data.

Furthermore, since the total value per patent calculated by the patent profit calculation device is calculated using only objective data without using scoring, the characteristic is that the arbitraryness is completely eliminated. Have.

The patent profit calculation apparatus according to the present embodiment is characterized in that it further includes a first business revenue calculation unit that calculates a business revenue value using the correlation calculated by the correlation calculation unit, based on the first embodiment. There is.
<Embodiment 2: Configuration>

  FIG. 12 illustrates a functional block diagram of the patent profit calculation apparatus according to the present embodiment. In the patent power calculation apparatus (1200) shown in FIG. 12, a “patent acquisition unit” (1201), a “patent history data acquisition unit” (1202), an “item content extraction unit” (1203), and a “search result holding” Part "(1204)," cost table holding part "(1205)," obsolete function storage part "(1206)," post-stale cost calculation part "(1207), and" summing part "(1208) Since the “aggregation unit” (1209), “business profit value acquisition unit” (1210), and “correlation calculation unit” (1211) are the same as those described in the first embodiment, the description thereof is omitted. To do. The patent profit calculation apparatus according to the present embodiment further includes a “first business profit calculation unit” (1212).

  The “first business revenue calculation unit” (1212) calculates a business revenue value corresponding to one or more names using the correlation calculated by the “correlation calculation unit” (1211). The business profit value may be calculated, for example, by substituting a YK value, which is an example of the aggregate value for each name, into the regression equation. For example, for Company H1 in FIG. 6, the regression equation R ′ = S / 0.05997 = about 16.67500 × S is used to calculate 16.67500 × 11139.33 = 185748.32775 (million yen). Condition.

  The business profit value calculated by the first business profit calculation unit is a business profit value (predicted value) predicted based on the aggregate value for each name. On the other hand, the business profit value acquired by the business profit value acquisition unit is a numerical value (actually measured value) relating to business profit actually generated. By calculating the predicted business revenue value in the first business revenue calculation unit, the revenue generated from the patent owned by the company can be grasped specifically. Further, the predicted business profit value and the actual business profit value can be compared and analyzed.

  Regarding the comparison between the predicted business revenue value and the actual business revenue value, the difference between the predicted business revenue value and the actual business revenue value may be rated. For example, the rating is classified into 10 levels in descending order. The difference obtained by subtracting the actual business revenue value from the predicted business revenue value indicates that the business revenue may increase from the base time toward the future. Therefore, by performing rating on this difference, it is possible to extract from the holder a bargain stock whose current stock price is cheaper than the stock price from the reference time to the future time. Alternatively, since the predicted business profit value is determined by a regression equation, the business profit value for each name / the total value for each name may be compared. In this case, it is not necessary to calculate the predicted business profit value.

  In addition, it is thought that the total value for each name has the following characteristics. That is, each company often takes action only in its business field, and it is considered that there is little need to take action beyond a certain technical field. Therefore, it is desirable that the correlation calculation is a relative comparison for each technical field. In this case, in order to rank the difference between the predicted business profit value and the actual business profit value, it is desirable to compare the regression equations in each technical field as a reference. For example, the angle θ formed by the straight line connecting the origin, the plot representing the business revenue value for each name and the total value for each name, and the straight line representing the regression equation is ranked in descending order of the angle.

Then, it is possible to display a discounted stock whose actual business revenue value is lower than the predicted business revenue value as a recommended stock on the ordering screen for buying and selling stocks, and it was rated to a certain rank It is good also as having the structure for ordering automatically buying and selling of a holder's stock.
<Embodiment 2: Hardware configuration>

  All or a part of each component that is a component of the present embodiment is configured by either hardware, software, or both hardware and software. As shown in FIG. 10, the computer has a CPU (1001), a RAM ( 1002), ROM (1003), input / output interface (I / O) (1004), HDD (1005), etc., which are interconnected by a data communication path such as a system bus (1006), Information transmission / reception and processing, CPU is patent acquisition program, patent history data acquisition program, item content extraction program, search result retention program, post-stale cost calculation program, summation program, aggregation program, business profit value acquisition unit Interpret various programs such as programs and correlation calculation programs The management of the content is the same as was described above, the description thereof is omitted. Although not shown in this figure, in the patent profit calculation apparatus according to the present embodiment, the first business profit calculation program is further held in the storage device such as the ROM (1003), and the patent profit calculation is performed. When the power supply of the apparatus is activated, the CPU (1001) also develops this first business profit calculation program in the work area of the RAM (1002).

  Then, the CPU (1001) executes the first business revenue calculation program and accepts input of one or more names as calculation targets. The input may be input by selecting individual names or by direct input. If a company group belonging to a certain business field is selected and input, the name associated with the business field is stored in the ROM (1003). It is good also as acquiring from the business profit value data currently hold | maintained at storage areas, etc., and if the IPC classification is selected, the name associated with this IPC classification may be acquired from the patent history data. . Alternatively, the entire range may be targeted. Then, the total value for each name corresponding to each inputted name is acquired from the storage data area of the RAM (1002). Then, the total value obtained for each name is substituted into the calculated regression equation held in the storage data area of the RAM (1002), and the business profit value corresponding to each name is calculated.

In addition, as described above, when rating the difference between the predicted business revenue value and the actual business revenue value, the CPU (1001) further determines the actual business revenue value corresponding to each name. The business profit value is acquired from the storage data area of the RAM (1002). Then, the difference between the calculated business revenue value (predicted business revenue value) corresponding to each name and the business revenue value for each name (actual business revenue value) is calculated. Then, the CPU compares the magnitude relationship of the differences with each other, sorts the data for each name in order of increasing difference, and associates the rating symbols. The rating symbol is expressed by, for example, a 10-level number and is stored in a storage area such as a ROM.
<Embodiment 2: Effect>

With the patent profit calculation device of this embodiment, the future patent profit can be calculated with respect to the reference date for the patent held by the holder.
<< Embodiment 3 >>
<Embodiment 3: Overview>

The patent profit calculation apparatus according to the present embodiment includes a second business profit calculation unit that calculates the business profit value corresponding to the patent ID using the correlation calculated by the correlation calculation unit, based on the first embodiment. Furthermore, there is a feature in having.
<Embodiment 3: Configuration>

  In the patent profit calculation apparatus according to the present embodiment, an example of a functional block diagram is obtained by replacing the “first business revenue calculation unit” (1212) shown in FIG. 12 with a “second business revenue calculation unit”. . That is, the “patent acquisition unit” (1201), “patent history data acquisition unit” (1202), “item content extraction unit” (1203), and “search” in the patent power calculation apparatus (1200) of the present embodiment. “Result holding unit” (1204), “cost table holding unit” (1205), “obsolete function storage unit” (1206), “post-stale cost calculation unit” (1207), “summing unit” ( 1208), “Aggregation Unit” (1209), “Business Revenue Value Acquisition Unit” (1210), and “Correlation Calculation Unit” (1211) are the same as those described in the first embodiment. Is omitted. The patent profit calculation apparatus according to the present embodiment further includes a “second business profit calculation unit”.

The second business revenue calculation unit calculates a business revenue value corresponding to one or more patent IDs using the correlation calculated by the correlation calculation unit (1211). The business profit value may be calculated, for example, by substituting a YK value, which is an example of the aggregate value for each name, into the regression equation. For example, for company H1 in FIG. 6, 16.67500 × 11139.33 = 185748.32775 (million yen) is calculated using the regression equation R ′ = S / 0.05997 = about 16.67500 × S. Condition. The predicted business revenue value for each patent ID is calculated, for example, by dividing the calculated predicted business revenue value for each name by the ratio of the total value for each patent ID in the total value for each name. In order to calculate the business profit value corresponding to one or more patent IDs, the predicted business profit value for each patent ID may be added to the patent ID to be calculated.
<Embodiment 3: Effect>

    The patent profit calculation apparatus according to the present embodiment can calculate future patent profits for a reference date from a certain patent or patent group.

0100, 1200 Patent profit calculation device 0101, 1201 Patent acquisition unit 0102, 1202 Patent history data acquisition unit 0103, 1203 Item content extraction unit 0104, 1204 Search result holding unit 0105, 1205 Cost table holding unit 0106, 1206 Obsolete function storage unit 0107, 1207 Post-stale cost calculation unit 0108, 1208 Total unit 0109, 1209 Total unit 0110, 1210 Business revenue value acquisition unit 0111, 1211 Correlation calculation unit 1212 First business revenue calculation unit

Claims (3)

A patent acquisition unit that acquires a patent ID that is a plurality of application numbers or patent numbers as a population,
A patent history data acquisition unit for acquiring patent history data of the acquired patent ID;
A pattern using a pattern prepared in advance with a combination of standard item names indicating legal procedures taken for patents (including those that are pending, the same applies below) described in the acquired patent history data An item content extracting unit that searches by matching processing and extracts the item content described in the patent history data in association with the procedure date according to the combination of the searched standard item names,
A search result holding unit that holds the extracted item contents and the procedure date associated therewith in association with the combination of the searched standard item names;
A cost table holding unit that holds a cost table that associates costs prepared in advance for each combination of item contents held in association with a combination of standard item names;
An obsolete function storage section that stores an obsolete function that serves as an indication of obsolete technology for each technical field;
For each combination of item contents extracted according to the combination of the standard item names of each patent , the cost is acquired using the cost table held in the cost table holding unit, and the calculation reference date and the combination of the item contents a procedure date associated with each, and filing date of the patent, and obsolescence after cost calculating unit that calculates the cost after obsolescence of obsolescence function and the calculation reference date with the art to which this patent belongs,
A summation unit that sums all calculated post-stale costs for patents ;
A totaling unit for further summing up the totaled value for each name of the applicant or right holder and for each technical field ;
A business profit value acquisition unit for acquiring a business profit value for each name;
Companies of each company based on the aggregated value for each technical field for each nominal name excluding the influence of the factor representing the company size of each name from the aggregated value for each technical field for each nominal name, and the total business revenue value for each nominal name Using multiple pairs of scale factor-excluded nominal business revenue values excluding the influence of the factor representing the scale Correlation between aggregated values for each technical field and scale factor-excluded nominal business revenue values in the population A correlation calculation unit for calculating
A patent profit calculation apparatus having,
The factor representing the company size is the calculation of patent profit, which is one of two factors selected from the number of patents held, the number of employees, the number of inventors, the number of applications, sales, operating profit, market capitalization, and research and development expenses. Equipment . The patent profit calculation apparatus according to claim 1, further comprising a first business profit calculation unit that calculates a business profit value corresponding to one or more names using the correlation calculated by the correlation calculation unit. The patent profit calculation apparatus according to claim 1, further comprising a second business profit calculation unit that calculates a business profit value corresponding to one or more patent IDs using the correlation calculated by the correlation calculation unit.

Images