KR102331992B1 - Method of technology evaluation using artificial intelligence - Google Patents

Abstract

A technology evaluation method using artificial intelligence is disclosed. The technology evaluation method using the artificial intelligence may comprise the steps of: calculating an expert growth score using first growth data; calculating an AI growth score from second growth data using a first growth potential evaluation model trained using machine learning; calculating an expert risk score using the first risk data; calculating an artificial intelligence risk score from second risk data using a first risk assessment model trained using machine learning; and calculating a final score by combining the expert growth score, the AI growth score, the expert risk score, and the artificial intelligence risk score.

Description

Method of technology evaluation using artificial intelligence

The present invention relates to a technology evaluation method using artificial intelligence. In particular, the evaluation target technology is evaluated by combining the rating calculated by an expert evaluating the technology evaluation index and the rating calculated by the artificial intelligence for the technology evaluation index by calculating the final rating. It relates to a technology evaluation method using artificial intelligence that can improve the predictive power of a company's growth potential or risk potential.

In general, the types of technology evaluation can be broadly divided into technology evaluation and technology value evaluation. the type it represents. And technology valuation is a type of converting the value of individual technology into monetary units in order to calculate the technology transaction price or the collateral value of intellectual property. Here, technology evaluation mainly uses a rating model or a rating model, and evaluates how much the business entity has the ability to generate profits by using technology, using items such as technicality, business feasibility, and marketability.

In such a technology evaluation method, the technology evaluation index for evaluating the technology to be evaluated is used, and the items of the technology evaluation index can be evaluated by using quantitative information and the qualitative judgment of the evaluator (expert). items are included. In the case of items evaluated using quantitative information, there is no problem because the same evaluation value is always obtained regardless of the evaluator. There was a problem of poor objectivity, such as different evaluation results. In addition, the technology rating evaluation is performed to determine the growth potential or risk potential of a company commercializing the technology to be evaluated. .

The problem to be solved by the present invention is the growth potential of a company commercializing the technology to be evaluated by combining the rating calculated by the expert evaluating the technology evaluation index and the rating calculated by the artificial intelligence for the technology evaluation index to calculate the final rating. It is to provide a technology evaluation method using artificial intelligence that can improve the predictive power of risk potential.

In a technology evaluation method using artificial intelligence according to an embodiment of the present invention for achieving the above object, an information processing device receives first growth data, which are evaluation values according to a technology evaluation index, and evaluates the first growth data calculating an expert growth score for growth potential according to the technology evaluation index by giving a growth weight to each item and calculating a weighted average value; Receive second growth data including quantitative data for evaluating indicators, and use the first growth potential evaluation model learned using machine learning from the second growth data for growth potential by the technology evaluation index Calculating an artificial intelligence growth score, which is a rating, in the information processing device, receiving first risk data, which are evaluation values according to a technology evaluation index, and giving a risk weight for each evaluation item to the first risk data to calculate a weighted average value calculating, in the information processing device, at least one of the evaluation values and quantitative data for evaluating the technical evaluation index 2 receiving risk data and calculating an artificial intelligence risk score, which is a rating for risk potential by the technology evaluation index, from the second risk data using a first risk assessment model learned using machine learning; and calculating, in the information processing device, the final score by combining the expert growth point, the artificial intelligence growth point, the expert risk score, and the artificial intelligence risk score.

In a technology evaluation method using artificial intelligence according to another embodiment of the present invention for achieving the above object, an information processing device receives first growth data, which are evaluation values according to a technology evaluation index, and applies the first growth data to the first growth data. calculating an expert growth score for growth potential according to the technology evaluation index by giving a growth weight to each evaluation item and calculating a weighted average value, in the information processing device, at least one evaluation value of the evaluation values and the technology Receive second growth data including quantitative data that can evaluate the evaluation index, and use the first growth potential evaluation model learned using machine learning to determine the growth potential by the technology evaluation index from the second growth data It may include calculating an artificial intelligence growth point, which is a rating for a business, and calculating, in the information processing device, a final score, which is a rating for growth potential, by combining the expert growth point and the artificial intelligence growth point.

In a technology evaluation method using artificial intelligence according to another embodiment of the present invention for achieving the above object, the information processing device receives first risk data, which are evaluation values according to the technology evaluation index, and applies the first risk data to the first risk data. calculating an expert risk score for the risk possibility by the technical evaluation index by giving a risk weight to each evaluation item and calculating a weighted average value; Receive the second risk data including quantitative data that can evaluate the evaluation index, and use the first risk assessment model learned using machine learning to evaluate the risk by the technology evaluation index from the second risk data. Calculating an artificial intelligence risk score, which is a rating for risk, and, in the information processing device, combining the expert risk score and the artificial intelligence risk score to calculate a final score, which is a rating for risk potential.

A technology evaluation method according to an embodiment according to the technical idea of the present invention is a rating calculated using an evaluation value evaluated by an evaluator (expert) with respect to a technology evaluation index item with respect to the technology to be evaluated and the technology evaluation index At least one evaluation value and quantitative data calculated by the evaluator are input, and the evaluation model learned through machine learning is combined to calculate the final score for growth potential or risk potential, thereby reflecting the subjective criteria of experts. Even in the case of items, it has the advantage of being more objective. That is, in the case of using the present invention, the evaluation value evaluated by the evaluator after calculating the rating by combining the evaluation value subjectively determined by the evaluator while finding the optimal weight in the evaluation model learned through machine learning without relying only on the subjective judgment of the evaluator It has the advantage of being able to more accurately predict the growth potential or risk potential of a company based on scientific statistics by combining it with the rating calculated using the value.

In addition, the technology evaluation method according to an embodiment according to the technical idea of the present invention uses artificial intelligence to add high growth potential and sustainable growth potential to the method of evaluating growth potential based on the evaluation result of the conventional technology evaluation index. It has the advantage of being able to more accurately judge the growth potential by reflecting it. And the technology evaluation method according to an embodiment according to the technical idea of the present invention uses artificial intelligence in the method of evaluating the risk potential based on the evaluation result of the conventional technology evaluation index to internal environmental factors and external environmental factors of the company It has the advantage of being able to more accurately judge the risk potential by additionally reflecting the risk potential according to the environmental risk calculated according to the environmental risks.

In order to more fully understand the drawings cited in the Detailed Description, a brief description of each drawing is provided.
1 is an embodiment of an information processing apparatus for performing a technology evaluation method using artificial intelligence according to an embodiment according to the technical idea of the present invention.
FIG. 2 is a flowchart of a technology evaluation method using artificial intelligence that calculates a final score reflecting only growth potential using the information processing device of FIG. 1 .
FIG. 3 is a flowchart of a technology evaluation method using artificial intelligence for calculating a final score reflecting only the risk possibility using the information processing device of FIG. 1 .
4 is a flowchart of a technology evaluation method using artificial intelligence to calculate a final score reflecting both growth potential and risk potential using the information processing device of FIG. 1 .
5 is an embodiment of an information processing apparatus for performing a technology evaluation method using artificial intelligence according to another embodiment according to the technical idea of the present invention.
6 is a flowchart of a technology evaluation method using artificial intelligence to calculate a final score reflecting only growth potential using the information processing device of FIG. 5 .
7 is a flowchart of a technology evaluation method using artificial intelligence that calculates a final score reflecting only risk possibility using the information processing device of FIG. 5 .
8 is a flowchart of a technology evaluation method using artificial intelligence to calculate a final score reflecting both growth potential and risk potential using the information processing device of FIG. 5 .

In order to fully understand the present invention, the operational advantages of the present invention, and the objects achieved by the practice of the present invention, reference should be made to the accompanying drawings illustrating preferred embodiments of the present invention and the contents described in the drawings.

Hereinafter, the present invention will be described in detail by describing preferred embodiments of the present invention with reference to the accompanying drawings. Like reference numerals in each figure indicate like elements.

1 is an embodiment of an information processing apparatus 100 for performing a technology evaluation method using artificial intelligence according to an embodiment according to the technical idea of the present invention, and FIG. 2 is the information processing apparatus 100 of FIG. This is a flowchart of a technology evaluation method using artificial intelligence that calculates the final score reflecting only the growth potential. 3 is a flowchart of a technology evaluation method using artificial intelligence that calculates a final score reflecting only risk possibility using the information processing device 100 of FIG. 1 , and FIG. 4 is a flowchart of the information processing device 100 of FIG. This is a flowchart of a technology evaluation method using artificial intelligence that calculates a final score that reflects both growth potential and risk potential.

Referring to FIG. 1 , the information processing apparatus 100 may include at least one of a growth potential evaluation unit 110 , a risk potential evaluation unit 130 , and a final score calculation unit 150 . The information processing apparatus 100 is a device that receives data, information, or a signal, calculates, and outputs it, and may be various devices such as a computer or a server. The growth potential evaluation unit 110 may include an expert growth point calculation unit 120 and a first growth potential evaluation model GM_1 . The risk assessment unit 130 may include an expert risk score calculation unit 140 and a first risk assessment model WM_1. The operation of each component will be described in more detail below with reference to FIGS. 1 to 4 .

The growth potential evaluation unit 110 and the risk potential evaluation unit 130 may selectively operate in some cases. That is, the final score calculation unit 150 may calculate the final score using only the score calculated by the growth potential evaluation unit 110 , or calculate the final score using only the score calculated by the risk possibility evaluation unit 130 . Alternatively, the final rating may be calculated using both the rating calculated by the growth potential evaluation unit 110 and the rating calculated by the risk probability evaluation unit 130 . For example, when the technology evaluation is performed in consideration of only the company's growth potential with respect to the technology to be evaluated, only the growth potential evaluation unit 110 operates and the risk possibility evaluation unit 130 does not operate, so that the final score calculation unit 150 ) can calculate the final score using only the growth potential score. As another example, when the technology evaluation is performed in consideration of only the risk possibility of the company in relation to the technology to be evaluated, only the risk evaluation unit 130 operates and the growth potential evaluation unit 110 does not operate, so that the final score calculation unit 150 can calculate the final score using only the risk probability score. As another example, when a technology evaluation is performed in consideration of both the company's growth potential and risk potential in relation to the technology to be evaluated, the growth potential evaluation unit 110 and the risk possibility evaluation unit 130 both operate, so that the final score calculation unit 150 ) can calculate the final score using both the growth potential score and the risk potential score. The grade of the evaluation target technology may be evaluated using the final grade calculated as described above. Hereinafter, with reference to FIGS. 1 to 4 , when the final score is calculated using only the growth potential score, when the final score is calculated using only the risk potential score, the final score is calculated using both the growth potential score and the risk potential score The calculation cases will be described separately.

First, a description will be given of a technology evaluation method using artificial intelligence that calculates a final score reflecting only growth potential with reference to FIGS. 1 and 2 .

Referring to FIGS. 1 and 2 , the expert growth point calculation unit 120 of the information processing device 100 receives first growth data GDATA_1 , which are evaluation values according to a technology evaluation index, and receives the first growth data GDATA_1 . By giving a growth weight to each evaluation item to calculate a weighted average value, it is possible to calculate an expert growth score for the growth potential according to the technology evaluation index (S210). That is, step S210 is a step of calculating the expert growth point by using the evaluation value evaluated by the evaluator (expert) for the evaluation items of the technical evaluation index, and the evaluation value includes the evaluation evaluated by the evaluator using quantitative data. Values and evaluation values qualitatively evaluated by the evaluator may be included. The technology evaluation index includes evaluation items for technology evaluation of the technology to be evaluated, for example, evaluation items for management competency, evaluation items for technology, evaluation items for marketability, and business feasibility. At least one of the evaluation items may be included. Table 1 shows an example of the evaluation items constituting the technical evaluation index and an example of the evaluation value evaluated by the evaluator, and the first growth data (GDATA_1) includes evaluation values for each evaluation item in Table 1 can do.

However, in the present invention, the first growth data GDATA_1 is not limited to the evaluation items of the technical evaluation index of Table 1, and evaluation items may be added or excluded as necessary.

The expert growth point calculation unit 120 may calculate the expert growth point by giving a growth weight for each evaluation item to the first growth data GDATA_1 and calculating a weighted average value. The growth weight is a value given to each evaluation item in order to evaluate the growth potential of the technology to be evaluated, and may be a preset value or a value input by the expert growth point calculation unit 120 .

The first growth potential evaluation model GM_1 of the information processing device 110 receives the second growth data GDATA_2, and the second growth data GDATA_2 is an artificial intelligence that is a rating for the growth potential according to the technology evaluation index. A growth point may be calculated (S220). That is, the first growth potential evaluation model GM_1 uses the second growth data GDATA_2 as an input variable (input variable) and the growth potential by the technology evaluation index as a target variable (output variable) using machine learning. It may be a trained model. The second growth data GDATA_2 may include quantitative data for evaluating at least one of the evaluation values and the technology evaluation index. In learning the first growth potential evaluation model GM_1 using machine learning, it can be learned by classifying it according to an industry or corporate situation. For example, the first growth potential evaluation model GM_1 may perform machine learning by classifying manufacturing and service industries, or performing machine learning by classifying start-up companies and general companies. Alternatively, the first growth potential evaluation model (GM_1) is more specifically subdivided into industries such as metal, mechanical, electrical and electronic, chemical fiber, environmental bio, other manufacturing, software development and construction, IT service, and engineering. Machine learning can also be performed by dividing it into a service field and a knowledge service field.

The second growth data GDATA_2 may include at least one evaluation value among evaluation values of the evaluation items of the technical evaluation index examined in relation to step S210 and quantitative data for evaluating the evaluation items of the technical evaluation index. . For example, the evaluation value of the evaluation item to be qualitatively evaluated among the second growth data GDATA_2 may be the evaluation value evaluated by the evaluator in step S210 as it is. In addition, the second growth data GDATA_2 may include quantitative data related to an evaluation item that can be evaluated quantitatively, or may include an evaluation value evaluated using the quantitative data. Since the evaluation value evaluated using quantitative data comes out the same evaluation value regardless of the evaluator, the evaluation value evaluated in step S210 may be included in the second growth data (GDATA_2) as it is in relation to evaluation items that can be evaluated quantitatively. Also, since the evaluator may make a mistake in evaluating using the quantitative data, quantitative data for evaluating the evaluation items may be included in the second growth data GDATA_2 .

The first growth potential evaluation model GM_1 may be an artificial intelligence model learned using machine learning, for example, a recurrent neural network (RNN), a convolutional neural network (CNN). Network) or an artificial neural network model such as a deep neural network (DNN). The first growth potential evaluation model (GM_1) combines the evaluation values of the above evaluation items under optimal conditions by learning using the existing technology evaluation contents and actual results (risk or growth) related to the technology evaluation target technology. It may be a model that can calculate the AI growth score. That is, in the first growth potential evaluation model (GM_1), the evaluation values previously evaluated by experts for the evaluation target technology, and whether the company that commercialized using the evaluation target technology has actually grown after the evaluation has caused risk or insolvency. It is possible to learn using the result values of whether or not it was done, and through this, it is possible to derive a relationship between the evaluation values of the evaluation items and the result values. For example, the first growth potential evaluation model GM_1 may set an AI growth weight for each evaluation item of the second growth data GDATA_2 by learning using machine learning, and evaluated in the second growth data GDATA_2 By giving the AI growth weight to each item and calculating a weighted average value, the AI growth point may be calculated. If the artificial intelligence growth weight and the growth weight of step S210 are the same, the expert growth point calculated in step S210 and the artificial intelligence growth point calculated in step S220 will have the same value, and the artificial intelligence growth weight and step S210 If the growth weights of are different, the expert growth point calculated in step S210 and the artificial intelligence growth point calculated in step S220 will have different values. That is, in step S220, the first growth potential evaluation model (GM_1) does not calculate the evaluation values of the items to be qualitatively evaluated among the technical evaluation indicators, but the evaluation values evaluated by the evaluator are unchanged for the items to be qualitatively evaluated. and combining these evaluation values and quantitatively calculated evaluation values under optimal conditions to calculate the AI growth score. For example, the quantitative data of the second growth data GDATA2 may include the number of years in the same industry of the manager, the number of commercialization cases, the number of patents, and the like. However, in the present invention, quantitative data among the second growth data GDATA_2 is not limited to raw data for the above variables, and quantitative data or evaluation items may be added, changed, or excluded as necessary.

The final score calculation unit 150 of the information processing apparatus 100 may combine the expert growth score and the artificial intelligence growth score to calculate a final score, which is a rating for growth potential (S230). Since the embodiment of FIG. 2 relates to a technology evaluation method using artificial intelligence that calculates a final score reflecting only growth potential, the final score calculation unit 150 combines the expert growth point and the artificial intelligence growth score to calculate the result is determined as the final score for the technology to be evaluated.

The final grade calculation unit 150 may calculate the final grade by combining the expert growth grade and the artificial intelligence growth grade by using a set combination ratio. For example, the final score calculation unit 150 may determine a result of combining the expert growth score and the artificial intelligence growth score in a ratio of 4:6 as the final score.

As another example, the final score calculation unit 150 may combine the expert growth point and the artificial intelligence growth point at an optimal ratio using a combined model learned using machine learning. For example, the coupling model may be an artificial neural network model such as a recurrent neural network (RNN), a convolutional neural network (CNN), or a deep neural network (DNN). The combined model combines the expert growth point and the artificial intelligence growth point in an optimal condition by learning using the actual result (risk or growth) related to the existing technology evaluation content and the technology evaluated evaluation target technology to obtain a final score It may be a model that can calculate .

Alternatively, the combined model may be learned using machine learning to calculate a score for high growth potential from the expert growth point and the artificial intelligence growth point and determine it as the score. That is, the combined model may be a model learned using machine learning with the expert growth point and the artificial intelligence growth point as input variables (input variables) and high growth potential as target variables (output variables). In this case, the final score calculated using the combined model may be an evaluation of the evaluation target technology from the viewpoint of high growth potential.

Next, a description will be given of a technology evaluation method using artificial intelligence that calculates a final score reflecting only the risk possibility with reference to FIGS. 1 to 3 .

1 to 3 , the expert risk score calculation unit 130 of the information processing device 100 receives first risk data WDATA_1 , which are evaluation values according to the technology evaluation index, and receives the first risk data WDATA_1 . By giving a risk weight to each evaluation item and calculating a weighted average value, it is possible to calculate an expert risk score for the risk possibility by the technical evaluation index (S310). That is, step S310 is a step of calculating the expert risk score using the evaluation values evaluated by the evaluator (expert) for the evaluation items of the technical evaluation index, and the evaluation value includes the evaluation evaluated by the evaluator using quantitative data. Values and evaluation values qualitatively evaluated by the evaluator may be included. The technology evaluation index includes evaluation items for technology evaluation of the technology to be evaluated, for example, evaluation items for management competency, evaluation items for technology, evaluation items for marketability, and business feasibility. At least one of the evaluation items may be included. The technology evaluation index may have the same evaluation items as the technology evaluation index described in relation to the embodiment of FIG. 2 or may have different evaluation items. For example, when the technology evaluation index of FIG. 3 and the technology evaluation index of FIG. 2 are the same, the first risk data WDATA_1 may be the same data as the first growth data GDATA_1 . Since the first risk data WDATA_1 may include evaluation items or variables similar to those of the first growth data GDATA_1 , the overlapping description below will be replaced with a description of the first growth data GDATA_1 .

The expert risk score calculating unit 130 may calculate the expert risk score by calculating a weighted average value by assigning a risk weight to the first risk data WDATA_1 for each evaluation item. The risk weight is a value assigned to each evaluation item in order to evaluate the risk potential of the technology to be evaluated, and may be a preset value or a value input by the expert risk score calculation unit 130 .

The first risk assessment model (WM_1) of the information processing device 110 receives the second risk data (WDATA_2), and from the second risk data (WDATA_2), an artificial intelligence that is a rating for the risk by the technology evaluation index. A risk score may be calculated (S320). That is, the first risk assessment model (WM_1) uses the second risk data (WDATA_2) as an input variable (input variable) and uses machine learning to use the risk potential by the technology evaluation index as a target variable (output variable). It may be a trained model. The second risk data WDATA_2 may include quantitative data capable of evaluating at least one of the evaluation values and the technical evaluation index. The first risk assessment model WM_1 can be learned by classifying it according to an industry or corporate situation when learning using machine learning. For example, the first risk assessment model WM_1 may perform machine learning by dividing the manufacturing industry and the service industry, or machine learning may be performed by dividing the start-up company and the general company. Alternatively, the first risk assessment model (WM_1) subdivides the industry in more detail, such as metal field, machinery field, electrical and electronic field, chemical fiber field, environmental bio field, other manufacturing field, software development and construction field, IT service field, engineering field Machine learning can also be performed by dividing it into a service field and a knowledge service field.

The second risk data GDATA_2 may include at least one evaluation value among evaluation values of the evaluation items of the technical evaluation index examined in relation to step S310 and quantitative data for evaluating the evaluation items of the technical evaluation index. . For example, as an evaluation value for an evaluation item to be qualitatively evaluated among the second risk data WDATA_2, the evaluation value evaluated by the evaluator in step S310 may be used as it is. In addition, the second risk data WDATA_2 may include quantitative data related to an evaluation item that can be evaluated quantitatively, or may include an evaluation value evaluated using the quantitative data. Since the evaluation value evaluated using quantitative data comes out the same evaluation value regardless of the evaluator, the evaluation value evaluated in step S310 in relation to evaluation items that can be evaluated quantitatively may be included in the second risk data (WDATA_2) as it is. Also, since the evaluator may make a mistake in evaluating using the quantitative data, quantitative data for evaluating the evaluation items may be included in the second risk data WDATA_2. The evaluation items related to the second risk data WDATA_2 may be the same as or different from the evaluation items related to the second growth data WDATA_2 . For example, when the technology evaluation index of FIG. 3 and the technology evaluation index of FIG. 2 are the same, the second risk data WDATA_2 may be the same data as the second growth data GDATA_2 . Since the second risk data WDATA_2 has similar evaluation items and variables to the second growth data GDATA_2 , the overlapping description below will be replaced with the description of the second growth data GDATA_2 .

The first risk assessment model WM_1 may be an artificial intelligence model learned using machine learning, for example, a recurrent neural network (RNN), a convolutional neural network (CNN). Network) or an artificial neural network model such as a deep neural network (DNN). The first risk assessment model (WM_1) combines the evaluation values of the above evaluation items under optimal conditions by learning using the actual results (risk or growth) related to the existing technology evaluation content and the technology evaluated evaluation target technology. It may be a model that can calculate an AI risk score. That is, the first risk possibility evaluation model (WM_1) is the evaluation values previously evaluated by experts for the evaluation target technology, and whether the company that commercialized using the evaluation target technology has actually grown after the evaluation has caused risk or insolvency. It is possible to learn using the result values of whether or not it was done, and through this, it is possible to derive a relationship between the evaluation values of the evaluation items and the result values.

For example, the first risk assessment model (WM_1) can set the artificial intelligence risk weight for each evaluation item of the second risk data (WDATA_2) by learning using machine learning, and evaluate the second risk data (WDATA_2) The artificial intelligence risk score may be calculated by calculating the weighted average value by giving the artificial intelligence risk weight to each item. If the artificial intelligence risk weight and the risk weight in step S310 are the same, the expert risk score calculated in step S310 and the artificial intelligence risk score calculated in step S320 will have the same value, and the artificial intelligence growth weight and step S310 If the growth weights of are different, the expert risk score calculated in step S310 and the artificial intelligence risk score calculated in step S320 will have different values. That is, in step S320, the first risk possibility evaluation model (WM_1) does not calculate the evaluation values of the items to be qualitatively evaluated among the technical evaluation indicators, but the evaluation values evaluated by the evaluator are unchanged for the items to be qualitatively evaluated. The AI risk score can be calculated by combining these evaluation values and quantitatively calculated evaluation values under optimal conditions.

The final score calculation unit 150 of the information processing device 100 may combine the expert risk score and the artificial intelligence risk score to calculate a final score, which is a score for risk possibility (S330). Since the embodiment of FIG. 3 relates to a technology evaluation method using artificial intelligence that calculates a final score reflecting only risk possibility, the final score calculation unit 150 combines the expert risk score and the artificial intelligence risk score. is determined as the final score for the technology to be evaluated.

The final score calculation unit 150 may calculate the final score by combining the expert risk score and the artificial intelligence risk score using a set combination ratio. For example, the final score calculation unit 150 may determine a result of combining the expert risk score and the artificial intelligence risk score in a ratio of 3:7 as the final score.

As another example, the final score calculation unit 150 may combine the expert growth point and the artificial intelligence growth point at an optimal ratio using a combined model learned using machine learning. For example, the coupling model may be an artificial neural network model such as a recurrent neural network (RNN), a convolutional neural network (CNN), or a deep neural network (DNN). The combined model combines the expert risk score and the artificial intelligence risk score in an optimal condition by learning using the actual results (risk or growth) related to the existing technology evaluation content and the technology evaluated evaluation target technology to obtain a final score It may be a model that can calculate .

Alternatively, the combined model may be learned using machine learning to calculate a score for high growth potential from the expert growth point and the artificial intelligence growth point and determine it as the score. That is, the combined model may be a model learned using machine learning with the expert growth point and the artificial intelligence growth point as input variables (input variables) and high growth potential as target variables (output variables). In this case, the final score calculated using the combined model may be an evaluation of the evaluation target technology from the viewpoint of high growth potential.

Next, a technology evaluation method using artificial intelligence for calculating a final score reflecting both growth potential and risk potential will be described with reference to FIGS. 1 to 4 .

1 to 4 , the growth potential evaluation unit 110 may operate as described with reference to FIG. 2 to transmit the expert growth point and the artificial intelligence growth point to the final score calculation unit 150 . Since the operation of the growth potential evaluation unit 110 has been described in detail with reference to FIG. 2 , the overlapping description will be replaced with the description of FIG. 2 . In addition, the risk assessment unit 120 may operate as described with reference to FIG. 3 to transmit the expert risk score and the artificial intelligence risk score to the final score calculation unit 150 . Since the operation of the risk assessment unit 120 has been described in detail with reference to FIG. 3 , the overlapping description will be replaced with the description of FIG. 3 .

The final score calculation unit 150 of the information processing device 100 combines the expert growth score, the artificial intelligence growth score, the expert risk score, and the artificial intelligence risk score to obtain a final score considering both growth potential and risk potential. can be calculated (S410).

The final score calculation unit 150 calculates the final score by combining the expert growth score, the AI growth score, the expert risk score, and the artificial intelligence risk score using a set combination ratio. have. For example, the final score calculation unit 150 combines the expert growth score, the artificial intelligence growth score, the expert risk score, and the artificial intelligence risk score in a ratio of 2:3:2:3 to the final score. It can be determined by rating.

Alternatively, the final score calculation unit 150 combines the expert growth score, the artificial intelligence growth score, the expert risk score, and the artificial intelligence risk score at an optimal ratio using a combined model learned using machine learning. may be The combined model can be learned by classifying it according to an industry or corporate situation in learning using machine learning. For example, the combined model may perform machine learning by dividing the manufacturing industry and the service industry, respectively, or by dividing the start-up company and the general company into machine learning. Alternatively, the combined model more specifically subdivides the industry, so that the metal field, the mechanical field, the electrical and electronic field, the chemical fiber field, the environmental bio field, the other manufacturing field, the software development construction field, the IT service field, the engineering service field, the knowledge service field Machine learning can also be performed by dividing For example, the coupling model may be an artificial neural network model such as a recurrent neural network (RNN), a convolutional neural network (CNN), or a deep neural network (DNN). The combined model is the expert growth point, the artificial intelligence growth point, the expert risk score, and the artificial intelligence by learning using the existing technology evaluation content and the actual result (risk or growth) related to the technology evaluated evaluation target technology. It may be a model that can calculate the final score by combining the risk score under optimal conditions.

Alternatively, one combined model learned using machine learning is learned using machine learning, combining the expert growth point and the artificial intelligence growth point to calculate a future growth potential score, and another learned using machine learning One combination model may calculate a risk probability score by combining the expert risk score and the artificial intelligence risk score. And, the final score calculation unit 150 calculates the final score by combining the future growth potential score and the risk probability score using a set combination ratio, or another combined model learned using machine learning. can be used to combine the future growth potential score and the risk potential score to calculate the final score.

5 is an embodiment of an information processing apparatus 500 for performing a technology evaluation method using artificial intelligence according to another embodiment according to the technical idea of the present invention, and FIG. 6 is the information processing apparatus 500 of FIG. 5 . This is a flowchart of a technology evaluation method using artificial intelligence that calculates the final score reflecting only the growth potential using 7 is a flowchart of a technology evaluation method using artificial intelligence that calculates a final score reflecting only risk possibility using the information processing device 500 of FIG. 5, and FIG. This is a flowchart of a technology evaluation method using artificial intelligence that calculates a final score that reflects both growth potential and risk potential.

Referring to FIG. 5 , the information processing device 500 may include at least one of a growth potential evaluation unit 510 , a risk potential evaluation unit 530 , and a final score calculation unit 550 . The information processing device 500 is a device that receives data, information, or signals, calculates, and then outputs them, and may be various devices such as a computer or a server. The growth potential evaluation unit 510 includes the expert growth evaluation unit 520, the first growth potential evaluation model (GM_1), the second growth potential evaluation model (GM_2), the third growth potential evaluation model (GM_3), and the fourth growth A possibility evaluation model (GM_4) may be included. The risk assessment unit 530 includes an expert risk assessment unit 540, a first risk assessment model (WM_1), a second risk assessment model (WM_2), a third risk assessment model (WM_3), and a fourth risk It may include a possibility evaluation model (WM_4). The operation of each component will be described in more detail below with reference to FIGS. 5 to 8 .

The growth potential evaluation unit 510 and the risk potential evaluation unit 530 may selectively operate in some cases. That is, the final score calculation unit 550 may calculate the final score using only the score calculated by the growth potential evaluation unit 510 , or calculate the final score using only the score calculated by the risk possibility evaluation unit 530 . Alternatively, the final rating may be calculated using both the rating calculated by the growth potential evaluation unit 510 and the rating calculated by the risk probability evaluation unit 530 . For example, when a technology evaluation is performed in consideration of only the company's growth potential in relation to the technology to be evaluated, only the growth potential evaluation unit 510 operates and the risk possibility evaluation unit 530 does not operate, so that the final score calculation unit 550 ) can calculate the final score using only the growth potential score. As another example, when the technology evaluation is performed in consideration of only the risk possibility of the company in relation to the technology to be evaluated, only the risk evaluation unit 530 operates and the growth potential evaluation unit 510 does not operate, so that the final score calculation unit 550 is not operated. can calculate the final score using only the risk probability score. As another example, when a technology evaluation is performed in consideration of both the company's growth potential and risk potential in relation to the technology to be evaluated, the growth potential evaluation unit 510 and the risk possibility evaluation unit 530 operate both, so that the final score calculation unit 550 ) can calculate the final score using both the growth potential score and the risk potential score. The grade of the evaluation target technology may be evaluated using the final grade calculated as described above. Hereinafter, with reference to FIGS. 5 to 8 , when the final score is calculated using only the growth potential score, when the final score is calculated using only the risk potential score, the final score is calculated using both the growth potential score and the risk potential score The calculation cases will be described separately.

First, a description will be given of a technology evaluation method using artificial intelligence that calculates a final score reflecting only growth potential with reference to FIGS. 5 and 6 .

5 and 6 , the expert growth point calculation unit 520 of the information processing device 500 receives first growth data GDATA_1 , which are evaluation values according to the technology evaluation index, and receives the first growth data GDATA_1 . By giving a growth weight to each evaluation item and calculating a weighted average value, it is possible to calculate an expert growth score for the growth potential according to the technology evaluation index (S610). Since step S610 is the same as step S210 of FIG. 2 , the overlapping content is replaced with the description of step S210 of FIG. 2 .

The first growth potential evaluation model GM_1 of the information processing device 510 inputs at least one of the evaluation values and second growth data GDATA_2 including quantitative data for evaluating the technology evaluation index. and, from the second growth data (GDATA_2), an AI growth score, which is a score for growth potential according to the technology evaluation index, may be calculated (S620). Since step S620 is the same as step S220 of FIG. 2 , the overlapping content is replaced with the description of step S220 of FIG. 2 .

The third growth potential evaluation model GM_3 of the information processing device 510 may calculate a future growth potential score, which is a score for high growth potential, from the expert growth score and the artificial intelligence growth score (S630). That is, the third growth potential evaluation model (GM_3) is a model trained using machine learning with the expert growth point and the artificial intelligence growth point as input variables (input variables) and high growth potential as target variables (output variables). can be The third growth potential evaluation model (GM_3) can be learned by classifying it according to the industry or corporate situation when learning using machine learning. For example, the third growth potential evaluation model (GM_3) may perform machine learning by dividing the manufacturing industry and the service industry, or machine learning may be performed by dividing the start-up company and the general company. Alternatively, the third growth potential evaluation model (GM_3) is more specifically subdivided into industries such as metal, mechanical, electrical and electronic, chemical fiber, environmental bio, other manufacturing, software development and construction, IT service, and engineering. Machine learning can also be performed by dividing it into a service field and a knowledge service field. For example, the third growth potential evaluation model (GM_3) may be an artificial neural network model such as a recurrent neural network (RNN), a convolutional neural network (CNN), or a deep neural network (DNN). can The third growth potential evaluation model (GM_3) optimizes the expert growth point and the artificial intelligence growth point by learning using the existing technology evaluation content and the actual result (risk or growth) related to the technology evaluation target technology. It may be a model capable of calculating the future growth potential score by combining the conditions. An embodiment in which the future growth potential rating is used as the final rating has been described in detail with reference to FIG. 2 , and the embodiment of FIG. 6 is related to an embodiment in which the future growth potential rating is not used as the final rating.

The third growth potential evaluation model (GM_3) is a model that learned high growth potential using machine learning, and is calculated by combining the expert growth point and the artificial intelligence growth point using the third growth potential evaluation model (GM_3). By performing the process of comparing the value and the objective function Y, the third growth potential evaluation model GM_3 may calculate the future growth potential score for the high growth potential.

; i기업의 t년도 매출액(

; Sales of company i in year t

; i기업의 t+1년도 매출액

; Sales of company i in year t+1

; t년도 산업 평균매출액)

; Industry average sales in year t)

Equation 1 is a new growth rate measurement concept that supplements the disadvantages of the conventional absolute and relative growth rates and takes only the advantages, and is the growth rate corrected for the scale effect from the absolute growth rate. That is, Equation 1 can measure the growth rate more accurately than the conventional absolute or relative growth rate measurement method by considering both the sales of the company and the average sales of the industry to which the technology to be evaluated belongs.

The second growth potential evaluation model (GM_2) of the information processing device 500 receives the third growth data (GDATA_3) for the evaluation items of the technological innovation capability index, and evaluates the second growth potential learned using machine learning. Using the model GM_2, it is possible to calculate a technological innovation capability score, which is a score for sustainable growth potential, from the third growth data GDATA_3 (S640). That is, the second growth potential evaluation model GM_2 is a model learned using machine learning with the third growth data GDATA_3 as an input variable (input variable) and the sustainable growth potential as a target variable (output variable). can The second growth potential evaluation model (GM_2) can be learned by classifying it according to an industry or corporate situation when learning using machine learning. For example, the second growth potential evaluation model (GM_2) may perform machine learning by dividing the manufacturing industry and the service industry, or machine learning may be performed by dividing the start-up company and the general company. Alternatively, the second growth potential evaluation model (GM_2) can be more specifically subdivided into industries such as metal, mechanical, electrical and electronic, chemical fiber, environmental bio, other manufacturing, software development and construction, IT service, and engineering. Machine learning can also be performed by dividing it into a service field and a knowledge service field.

The technology innovation capability index includes evaluation items that can evaluate whether there is a possibility of continuous growth when a company commercializes using the evaluation target technology. The technological innovation capability index may include evaluation items for the evaluation of the sustainable growth potential of the technology to be evaluated, for example, the ratio of technical manpower as an item against the company's infrastructure, the number of technology development cases as an item for corporate activities, and performance. The number of commercialization cases, etc., which are items for However, in the present invention, the items constituting the technological innovation capability indicator are not limited to the above cases, and the items may be changed as needed, and quantitative data for evaluating the items may be changed.

The second growth potential evaluation model GM_2 may be an artificial intelligence model learned using machine learning, for example, a recurrent neural network (RNN), a convolutional neural network (CNN). Network) or an artificial neural network model such as a deep neural network (DNN). The second growth potential evaluation model (GM_2) combines the evaluation values of the above evaluation items under optimal conditions by learning using the existing technology evaluation contents and actual results (risk or growth) related to the technology evaluation target technology. It may be a model capable of calculating the technological innovation competency score. That is, the second growth potential evaluation model (GM_2) determines whether the evaluation values previously evaluated for the items of the technological innovation competency index, and whether the company that commercialized using the evaluation target technology has actually grown after evaluation, risk or It is possible to learn by using the result values of whether or not insolvency has occurred, and through this, it is possible to derive a relationship between the evaluation values of the evaluation items and the result values. For example, the second growth potential evaluation model (GM_2) can set the AI innovation weight for each item rating calculated using the third growth data (GDATA_3) by learning using machine learning, and the third growth By giving the AI innovation weight to the score for each item calculated using the data GDATA_3 and calculating a weighted average value, the technological innovation competency score can be calculated.

The final score calculation unit 550 of the information processing device 510 may calculate a final score from the future growth potential score and the technological innovation capability score. For example, the final score calculation unit 550 may calculate the final score by combining the future growth potential score and the technological innovation capability score by using a set combination ratio.

Alternatively, as shown in FIG. 5 , the final score calculation unit 550 may output the technology business growth score calculated by the fourth growth potential evaluation model GM_4 as the final score. The fourth growth potential evaluation model GM_4 may calculate the technology business growth point from the future growth potential score and the technological innovation capability score ( S630 ). That is, the fourth growth potential evaluation model (GM_4) uses the future growth potential score and the technological innovation competency score as input variables (input variables) and uses the final result for growth potential as a target variable (output variable) to perform machine learning. It may be a model trained using The fourth growth potential evaluation model (GM_4) can be learned by classifying it according to the industry or corporate situation when learning using machine learning. For example, the fourth growth potential evaluation model (GM_4) may perform machine learning by dividing the manufacturing industry and the service industry, or machine learning may be performed by dividing the start-up company and the general company. Alternatively, the 4th growth potential evaluation model (GM_4) subdivides the industry in more detail, such as metal field, machinery field, electrical and electronic field, chemical fiber field, environmental bio field, other manufacturing field, software development and construction field, IT service field, engineering field Machine learning can also be performed by dividing it into a service field and a knowledge service field. For example, the fourth growth potential evaluation model (GM_4) may be an artificial neural network model such as a recurrent neural network (RNN), a convolutional neural network (CNN), or a deep neural network (DNN). can The fourth growth potential evaluation model (GM_4) optimizes the future growth potential score and the technological innovation competency score by learning using the existing technology evaluation content and actual results (risk or growth) related to the technology evaluation target technology. It may be a model that can calculate the technology business growth score by combining it under the condition of .

First, a description will be given of a technology evaluation method using artificial intelligence that calculates a final score reflecting only the risk possibility with reference to FIGS. 5 to 7 .

5 to 7 , the expert risk score calculation unit 540 of the information processing device 500 receives first risk data WDATA_1, which are evaluation values according to the technology evaluation index, and receives the first risk data WDATA_1 By giving a risk weight to each evaluation item and calculating a weighted average value, it is possible to calculate an expert risk score for the risk possibility by the technical evaluation index (S710). Since step S710 is the same as step S310 of FIG. 3 , the overlapping content is replaced with the description of step S310 of FIG. 3 .

The first risk evaluation model WM_1 of the information processing device 510 inputs at least one of the evaluation values and the second risk data WDATA_2 including quantitative data for evaluating the technology evaluation index. and, from the second risk data WDATA_2, it is possible to calculate an artificial intelligence risk score, which is a score for risk possibility by the technology evaluation index (S720). Since step S720 is the same as step S320 of FIG. 3 , the overlapping content is replaced with the description of step S320 of FIG. 3 .

The third risk assessment model WM_3 of the information processing device 510 may calculate a risk probability score, which is a rating for the risk possibility by the technology evaluation index, from the expert risk score and the artificial intelligence risk score (S730) ). That is, the third risk assessment model (WM_3) uses the expert risk score and the artificial intelligence risk score as input variables (input variables) and uses the risk potential by the technology evaluation index as a target variable (output variable) for machine learning. It may be a model trained using The third risk assessment model (WM_3) can be learned by classifying it according to the industry or corporate situation when learning using machine learning. For example, the first risk assessment model WM_1 may perform machine learning by dividing the manufacturing industry and the service industry, or machine learning may be performed by dividing the start-up company and the general company. Alternatively, the 3rd risk assessment model (WM_3) subdivides the industry in more detail, such as metal field, machinery field, electrical and electronic field, chemical fiber field, environmental bio field, other manufacturing field, software development and construction field, IT service field, engineering field Machine learning can also be performed by dividing it into a service field and a knowledge service field. For example, the third risk assessment model (WM_3) may be an artificial neural network model such as a recurrent neural network (RNN), a convolutional neural network (CNN), or a deep neural network (DNN). can The third risk assessment model (WM_3) optimizes the expert risk score and the artificial intelligence risk score by learning using the existing technology evaluation content and actual results (risk or growth) related to the technology evaluation target technology. It may be a model capable of calculating the risk probability score by combining conditions. An embodiment in which the risk likelihood score is used as the final score has been described in detail with reference to FIG. 3 , and the embodiment of FIG. 7 relates to an embodiment in which the risk likelihood score is not used as the final score.

The second risk assessment model (WM_2) of the information processing device 500 receives the third risk data (WDATA_3) for the evaluation items of the environmental risk index, and the second risk assessment model learned using machine learning It is possible to calculate an environmental risk score, which is a score for the risk possibility due to the environment, from the third risk data WDATA_3 using (WM_2) (S740). That is, the second risk assessment model (WM_2) uses the third risk data (WDATA_3) as an input variable (input variable) and the risk possibility due to the environment as a target variable (output variable) using machine learning. can be a model. The second risk assessment model (WM_2) can be learned by classifying it according to an industry or corporate situation when learning using machine learning. For example, the second risk assessment model WM_2 may perform machine learning by dividing the manufacturing industry and the service industry, or machine learning may be performed by dividing the start-up company and the general company. Alternatively, the second risk assessment model (WM_2) subdivides the industry in more detail, such as metal field, machinery field, electrical and electronic field, chemical fiber field, environmental bio field, other manufacturing field, software development and construction field, IT service field, engineering field Machine learning can also be performed by dividing it into a service field and a knowledge service field.

The environmental risk index includes evaluation items that can evaluate whether there is a possibility of risk due to the internal or external environment when a company commercializes the technology using the evaluation target technology. For example, the environmental risk index includes evaluation items for evaluating the possibility of risk by the internal environment or external environment, for example, corporate information related to the internal environment (whether or not to own the headquarters, etc.) and the company's It may include financial soundness information (R&D intensity, etc.) or macroeconomic information (small business production index, etc.) related to the external environment. However, in the present invention, the items constituting the environmental risk index are not limited to the above cases, and the items may be changed as needed, and quantitative data for evaluating the items may be changed.

The second risk assessment model WM_2 may be an artificial intelligence model learned using machine learning, for example, a recurrent neural network (RNN), a convolutional neural network (CNN). Network) or an artificial neural network model such as a deep neural network (DNN). The second risk assessment model (WM_2) combines the evaluation values of the above evaluation items under optimal conditions by learning using the actual results (risk or growth) related to the existing technology evaluation contents and the technology evaluated evaluation target technology. It may be a model that can calculate an environmental risk score. That is, the second risk assessment model (WM_2) determines whether the company that has commercialized using the evaluation target technology after the evaluation has actually grown or whether risk or insolvency has occurred after the evaluation values previously evaluated for the environmental risk index. It is possible to learn by using the results of For example, the second risk assessment model (WM_2) can set the artificial intelligence environment weight for the rating for each item calculated using the third risk data (WDATA_3) by learning using machine learning, and the third risk The environmental risk score may be calculated by calculating a weighted average value by giving the AI environment weight to the score for each item calculated using the data WDATA_3 .

The final score calculation unit 550 of the information processing device 510 may calculate a final score from the risk probability score and the environmental risk score. For example, the final score calculation unit 550 may calculate the final score by combining the risk probability score and the environmental risk score using a set combination ratio.

Alternatively, the final score calculation unit 550 may output the technology project risk score calculated by the fourth risk possibility evaluation model WM_4 as the final score, as shown in FIG. 5 . The fourth risk assessment model WM_4 may calculate the technology project risk score from the risk potential score and the environmental risk score (S730). That is, the fourth risk assessment model (WM_4) uses the risk probability score and the environmental risk score as input variables (input variables) and uses machine learning as the target variable (output variable) for the final result of risk potential. It may be a trained model. The fourth risk assessment model (WM_4) can be learned by classifying it according to an industry or corporate situation when learning using machine learning. For example, the fourth risk assessment model (WM_4) may perform machine learning by dividing the manufacturing industry and the service industry, or machine learning may be performed by dividing the start-up company and the general company. Alternatively, the 4th risk assessment model (WM_4) subdivides the industry in more detail and subdivides it into metal field, machinery field, electrical and electronic field, chemical fiber field, environmental bio field, other manufacturing field, software development and construction field, IT service field, and engineering field. Machine learning can also be performed by dividing it into a service field and a knowledge service field. For example, the fourth risk assessment model (WM_4) may be an artificial neural network model such as a recurrent neural network (RNN), a convolutional neural network (CNN), or a deep neural network (DNN). can The fourth risk assessment model (WM_4) uses the existing technology evaluation content and the actual result (risk or growth) related to the technology evaluated evaluation target technology to learn the risk probability score and the environmental risk score under optimal conditions. It may be a model capable of calculating the technology business growth score by combining them with

Next, a technology evaluation method using artificial intelligence for calculating a final score reflecting both growth potential and risk potential will be described with reference to FIGS. 5 to 8 .

5 to 8 , the growth potential evaluation unit 510 operates as described with reference to FIG. 6 to transmit the technology business growth score to the final score calculation unit 550 . Since the operation of the growth potential evaluation unit 510 has been described in detail with reference to FIG. 6 , the overlapping description will be replaced with the description of FIG. 6 . In addition, the risk assessment unit 520 may operate as described with reference to FIG. 7 to transmit the technology project risk score to the final score calculation unit 550 . Since the operation of the risk assessment unit 520 has been described in detail with reference to FIG. 7 , the overlapping description will be replaced with the description of FIG. 7 .

The final score calculation unit 550 of the information processing device 500 may combine the technology business growth score and the technology business risk score to calculate a final score in consideration of both growth potential and risk potential ( S810 ).

The final score calculation unit 550 may calculate the final score by combining the technology business growth point and the technology business risk score using a set combination ratio. For example, the final score calculation unit 550 may determine a result of combining the technology business growth score and the technology business risk score in a ratio of 5:5 as the final score.

Alternatively, the final score calculation unit 550 may combine the technology business growth score and the technology business risk score at an optimal ratio using a combined model learned using machine learning. The combined model can be learned by classifying it according to an industry or corporate situation in learning using machine learning. For example, the combined model may perform machine learning by dividing the manufacturing industry and the service industry, respectively, or by dividing the start-up company and the general company into machine learning. Alternatively, the combined model more specifically subdivides the industry, so that the metal field, the mechanical field, the electrical and electronic field, the chemical fiber field, the environmental bio field, the other manufacturing field, the software development construction field, the IT service field, the engineering service field, the knowledge service field Machine learning can also be performed by dividing For example, the coupling model may be an artificial neural network model such as a recurrent neural network (RNN), a convolutional neural network (CNN), or a deep neural network (DNN). The combined model combines the technology business growth score and the technology business risk score under optimal conditions by learning by using the actual results (risk or growth) related to the existing technology evaluation content and the technology evaluated evaluation target technology. It may be a model capable of calculating a rating.

As described above, the best embodiment has been disclosed in the drawings and the specification. Although specific terms are used herein, they are used only for the purpose of describing the present invention and are not used to limit the meaning or scope of the present invention described in the claims. Therefore, it will be understood by those skilled in the art that various modifications and equivalent other embodiments are possible therefrom. Accordingly, the true technical protection scope of the present invention should be defined by the technical spirit of the appended claims.

Claims (9)

In the information processing device, receiving first growth data, which are evaluation values according to the technology evaluation index, and giving a growth weight for each evaluation item to the first growth data, calculating an expert growth rating for growth potential according to the technology evaluation index step;
In the information processing device, receiving second growth data including quantitative data for evaluating at least one of the evaluation values and the technical evaluation index, and evaluating the first growth potential learned using machine learning calculating an artificial intelligence growth score, which is a score for growth potential according to the technology evaluation index, from the second growth data using a model;
The information processing device receives first risk data, which are evaluation values according to the technical evaluation index, and assigns a risk weight for each evaluation item to the first risk data to calculate an expert risk score for the risk possibility by the technology evaluation index to do;
In the information processing device, receiving second risk data including quantitative data for evaluating at least one of the evaluation values and the technical evaluation index, and first risk assessment learned using machine learning calculating an artificial intelligence risk score, which is a score for risk possibility by the technology evaluation index, from the second risk data using a model; and
Comprising, in the information processing device, calculating a final score by combining the expert growth score, the artificial intelligence growth score, the expert risk score, and the artificial intelligence risk score,
The first growth potential evaluation model,
The artificial intelligence growth weight is learned by using the technology evaluation content and the actual result related to the technology evaluated evaluation target technology, and the artificial intelligence growth point is calculated by giving the artificial intelligence growth weight to the second growth data for each evaluation item. and
The first risk assessment model is,
The artificial intelligence risk weight is learned using the technical evaluation contents and the actual result related to the technology evaluated evaluation target technology, and the artificial intelligence risk score is calculated by assigning the artificial intelligence risk weight to the second risk data for each evaluation item. A technology evaluation method using artificial intelligence, characterized in that The method according to claim 1, wherein the technology evaluation method using artificial intelligence comprises:
In the information processing device, the third growth data for the evaluation items of the technological innovation competency index is received, and the sustainable growth potential from the third growth data is evaluated using the second growth potential evaluation model learned using machine learning. calculating a technological innovation competency rating, which is a rating for and
In the information processing device, the third risk data for the evaluation items of the environmental risk indicator is input, and the risk possibility due to the environment from the third risk data using the second risk assessment model learned using machine learning Further comprising the step of calculating an environmental risk rating that is a rating for
Calculating the final score includes:
In the information processing device, calculating the final score by combining the expert growth score, the artificial intelligence growth score, the technological innovation capability score, the expert risk score, the artificial intelligence risk score, and the environmental risk score A technology evaluation method using artificial intelligence, characterized in that The method of claim 2, wherein calculating the final score comprises:
calculating, in the information processing device, a future growth potential score, which is a score for high growth potential, from the expert growth score and the artificial intelligence growth score using a third growth potential evaluation model learned using machine learning; and
calculating, in the information processing device, a technology business growth point from the future growth potential score and the technological innovation capability score by using a fourth growth potential evaluation model learned through machine learning;
In the information processing device, using a third risk assessment model learned using machine learning, a risk probability score is calculated from the expert risk score and the artificial intelligence risk score, which is a rating for the risk possibility by the technology evaluation index to do; and
calculating, in the information processing device, a technology business risk score from the risk likelihood score and the environmental risk score using a fourth risk assessment model learned through machine learning; and
and calculating, in the information processing device, the final score by combining the technology business growth score and the technology business risk score. In the information processing device, an expert on growth potential according to the technology evaluation index by receiving first growth data, which are evaluation values according to the technology evaluation index, and calculating a weighted average value by giving a growth weight for each evaluation item to the first growth data calculating a growth point;
In the information processing device, receiving second growth data including quantitative data for evaluating at least one of the evaluation values and the technical evaluation index, and evaluating the first growth potential learned using machine learning calculating an artificial intelligence growth score, which is a score for growth potential according to the technology evaluation index, from the second growth data using a model; and
Comprising, in the information processing device, combining the expert growth point and the artificial intelligence growth point to calculate a final score, which is a score for growth potential,
The first growth potential evaluation model,
The artificial intelligence growth weight is learned by using the technology evaluation content and the actual result related to the technology evaluated evaluation target technology, and the artificial intelligence growth point is calculated by giving the artificial intelligence growth weight to the second growth data for each evaluation item. A technology evaluation method using artificial intelligence, characterized in that The method of claim 4, wherein the technology evaluation method using artificial intelligence comprises:
In the information processing device, the third growth data for the evaluation items of the technological innovation competency index is received, and the sustainable growth potential from the third growth data is evaluated using the second growth potential evaluation model learned using machine learning. Further comprising the step of calculating a technological innovation competency score,
Calculating the final score includes:
and calculating, in the information processing device, the final score by combining the expert growth point, the artificial intelligence growth point, and the technological innovation capability point. The method of claim 5, wherein calculating the final score comprises:
calculating, in the information processing device, a future growth potential score, which is a score for high growth potential, from the expert growth score and the artificial intelligence growth score using a third growth potential evaluation model learned using machine learning; and
In the information processing device, using a fourth growth potential evaluation model learned through machine learning, calculating the final score from the future growth potential score and the technological innovation competency score. The technology evaluation method used. In the information processing device, an expert on the risk possibility by the technical evaluation index by receiving the first risk data, which are evaluation values according to the technical evaluation index, and calculating a weighted average value by giving a risk weight for each evaluation item to the first risk data calculating a risk score;
In the information processing device, receiving second risk data including quantitative data for evaluating at least one of the evaluation values and the technical evaluation index, and first risk assessment learned using machine learning calculating an artificial intelligence risk score, which is a score for risk possibility by the technology evaluation index, from the second risk data using a model; and
Comprising, in the information processing device, combining the expert risk score and the artificial intelligence risk score to calculate a final score, which is a score for risk possibility,
The first risk assessment model is,
The artificial intelligence risk weight is learned using the technical evaluation contents and the actual result related to the technology evaluated evaluation target technology, and the artificial intelligence risk score is calculated by assigning the artificial intelligence risk weight to the second risk data for each evaluation item. A technology evaluation method using artificial intelligence, characterized in that The method according to claim 7, wherein the technology evaluation method using artificial intelligence comprises:
In the information processing device, the third risk data for the evaluation items of the environmental risk indicator is input, and the risk possibility due to the environment from the third risk data using the second risk assessment model learned using machine learning Further comprising the step of calculating an environmental risk rating that is a rating for
Calculating the final score includes:
and calculating, in the information processing device, the final score by combining the expert risk score, the artificial intelligence risk score, and the environmental risk score. The method of claim 8, wherein calculating the final score comprises:
In the information processing device, using a third risk assessment model learned using machine learning, a risk probability score is calculated from the expert risk score and the artificial intelligence risk score, which is a rating for the risk possibility by the technology evaluation index to do; and
Technology using artificial intelligence, comprising the step of calculating, in the information processing device, the final score from the risk likelihood score and the environmental risk score using a fourth risk assessment model learned through machine learning Assessment Methods.

Images