CN116452052A - Patent value evaluation method, system, device and storage medium - Google Patents
Patent value evaluation method, system, device and storage medium Download PDFInfo
- Publication number
- CN116452052A CN116452052A CN202310422842.5A CN202310422842A CN116452052A CN 116452052 A CN116452052 A CN 116452052A CN 202310422842 A CN202310422842 A CN 202310422842A CN 116452052 A CN116452052 A CN 116452052A
- Authority
- CN
- China
- Prior art keywords
- value
- index
- weighting
- evaluation
- weights
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 56
- 238000003860 storage Methods 0.000 title claims abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 147
- 238000013210 evaluation model Methods 0.000 claims abstract description 32
- 238000004458 analytical method Methods 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims description 28
- 239000011159 matrix material Substances 0.000 claims description 19
- 238000010606 normalization Methods 0.000 claims description 15
- 238000009826 distribution Methods 0.000 claims description 11
- 238000007781 pre-processing Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 7
- 238000000556 factor analysis Methods 0.000 claims description 6
- 238000000513 principal component analysis Methods 0.000 claims description 6
- 238000007477 logistic regression Methods 0.000 claims description 5
- 230000009699 differential effect Effects 0.000 claims description 4
- 239000002131 composite material Substances 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 238000011157 data evaluation Methods 0.000 abstract description 2
- 238000012545 processing Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 15
- 239000013598 vector Substances 0.000 description 15
- 238000004422 calculation algorithm Methods 0.000 description 13
- 230000006870 function Effects 0.000 description 10
- 238000012423 maintenance Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 101000911753 Homo sapiens Protein FAM107B Proteins 0.000 description 4
- 102100026983 Protein FAM107B Human genes 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000006467 substitution reaction Methods 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000009792 diffusion process Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 101100070630 Arabidopsis thaliana HHT1 gene Proteins 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000005295 random walk Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 230000002195 synergetic effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/18—Legal services
- G06Q50/184—Intellectual property management
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Technology Law (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Software Systems (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a patent value evaluation method, a system, a device and a storage medium, and the embodiment of the invention acquires a patent index to be evaluated; the method starts from multiple dimensions and fully arranges the patent information; carrying out weight analysis on the patent indexes based on a preset weighting method by using a first evaluation model to obtain a plurality of groups of weights; corresponding multiple groups of weights are obtained through multiple weighting modes, so that the randomness of evaluation is avoided; comprehensively weighting a plurality of groups of weights based on game theory to obtain comprehensive value scores; the weights of the multiple groups are further comprehensively weighted, so that the weight difference is smaller, and better convergence of the evaluation result is realized; and carrying out value analysis on the patent index by using the second evaluation model to obtain the commercial value amount. In addition to score evaluation, value evaluation is combined, subjectivity of evaluation is avoided, and meanwhile accurate reference is provided for patent evaluation. The embodiment of the invention can reasonably and accurately evaluate the patent value, and can be widely applied to the technical field of data evaluation and processing.
Description
Technical Field
The invention relates to the technical field of data evaluation processing, in particular to a patent value evaluation method, a system, a device and a storage medium.
Background
As an important intangible asset, patents are increasingly valued by various industries, and assessment of quality and commercial value is also becoming an important issue. The traditional patent value evaluation method mainly comprises the following steps: (1) cost method: also called a reset cost method, which is an evaluation method based on the current market price and by using the cost of redeveloping or purchasing products with the same functions and purposes as those of the patent to be evaluated in the current time as a calculation standard, thereby determining the value of the patent to be evaluated; (2) market method: the method is also called as the current market price method or market comparison method, and is a method for evaluating by selecting trade conditions and prices of a plurality of similar patent technologies in technical markets as reference and comparison objects and properly adjusting the characteristics of the to-be-evaluated patent technologies through market investigation; (3) revenue method: the profit method, also called as the profit-present value method, is an evaluation method for converting the expected profits of the patent to be evaluated in each period of the rest economic life period into the evaluation reference daily present value by using the proper discount rate and adding up the evaluation reference daily present value to be used as the value of the patent; (4) physical option method: the patent is regarded as option, and the option of a management decision maker in the decision making of problems such as investment, production, product research and development and the like is considered, so that the option value of the decision making in the process of implementing the patent can be fully reflected, and the value of the patent technology can be evaluated more reasonably and accurately.
However, none of the above methods can utilize the information of the patent itself more comprehensively, and there is a certain subjectivity and randomness. In view of this, how to comprehensively perform reasonably accurate patent evaluation based on patent information is an urgent problem to be solved.
Disclosure of Invention
In view of the above, the embodiment of the invention provides a patent value evaluation method, a system, a device and a storage medium, which can reasonably and accurately evaluate the patent value.
In one aspect, an embodiment of the present invention provides a patent value evaluation method, including:
obtaining a patent index to be evaluated; the patent indexes comprise legal index sets, technical index sets and economic index sets;
carrying out weight analysis on the patent indexes based on a preset weighting method by using a first evaluation model to obtain a plurality of groups of weights; the preset weighting method comprises a supervised weighting method and an unsupervised weighting method;
comprehensively weighting a plurality of groups of weights based on game theory to obtain comprehensive value scores;
performing value analysis on the patent index by using a second evaluation model to obtain commercial value amount; the second evaluation model is generated based on the supervision model through the patent data training of the marked value money labels.
Optionally, obtaining the patent index to be evaluated includes:
constructing a patent index system based on a patent information database;
the patent index system is constructed based on legal dimension, technical dimension and economic dimension;
patent data to be evaluated are obtained, and the patent data are arranged on the basis of a patent index system to obtain patent indexes.
Optionally, the method further comprises:
carrying out data preprocessing on numerical indexes in the patent indexes;
wherein, the data preprocessing comprises a log-based bias distribution operation and a normalization operation.
Optionally, the unsupervised weighting method includes a director weighting method and an objective weighting method, and the weighting analysis is performed on the patent index based on a preset weighting method by using a first evaluation model to obtain multiple groups of weights, including:
according to the patent index;
subjective weighting is carried out by using an analytic hierarchy process to obtain subjective weighting;
and, respectively carrying out objective weighting by using an entropy weight TOPSIS method, a factor analysis weighting method, a principal component analysis weighting method and a CRITIC weighting method to obtain a plurality of groups of objective weights;
and performing supervised weighting by using a logistic regression weighting method to obtain the supervised weighting.
Optionally, comprehensively weighting the multiple groups of weights based on the game theory to obtain a comprehensive value score, including:
Performing arbitrary linear combination on the multiple groups of weights to obtain a first comprehensive weight;
optimizing weight coefficients of each group of weights through a game theory based on the first comprehensive weights; performing first derivative optimization according to the differential property of the matrix to obtain a target weight coefficient set;
determining a second comprehensive weight according to the target weight coefficient set and the combined weight; wherein the second comprehensive weight comprises the comprehensive weight of each index in the patent indexes;
and determining the comprehensive value score according to the target weight coefficient set and the weight.
Optionally, determining the composite value score according to the target weight coefficient set and the weight includes:
determining a comprehensive value score according to the target weight coefficient set and the weight through a comprehensive value score formula;
wherein, the comprehensive value score formula is:
wherein Score represents the integrated value Score, M represents the number of patent indexes, w i Comprehensive weight of the i-th index, v i The score obtained by the i-th index is shown.
Optionally, the method further comprises the step of generating a second assessment model based on the supervision model through the training of the patent data tagged with the value money tag, the step comprising:
acquiring patent case data of a preset term; the case data comprises case money value;
Correcting the patent case data based on the historical consumer commodity price index to obtain label data of the value amount;
patent data of patents corresponding to patent case data are obtained, and training data are obtained through arrangement based on preset patent indexes;
and inputting the training data and the label data into a pre-constructed supervision model for monetary assessment training, and adjusting the supervision model based on the training result to obtain a second assessment model.
In another aspect, an embodiment of the present invention provides a patent value evaluation system, including:
the first module is used for acquiring the patent index to be evaluated; the patent indexes comprise legal index sets, technical index sets and economic index sets;
the second module is used for carrying out weight analysis on the patent indexes based on a preset weighting method by utilizing the first evaluation model to obtain a plurality of groups of weights; the preset weighting method comprises a supervised weighting method and an unsupervised weighting method;
the third module is used for comprehensively weighting a plurality of groups of weights based on the game theory to obtain a comprehensive value score;
a fourth module, configured to perform value analysis on the patent index by using the second evaluation model, to obtain a commercial value amount; the second evaluation model is generated based on the supervision model through the patent data training of the marked value money labels.
In another aspect, an embodiment of the present invention provides a patent value evaluation apparatus, including a processor and a memory;
the memory is used for storing programs;
the processor executes a program to implement the method as before.
In another aspect, embodiments of the present invention provide a computer-readable storage medium storing a program for execution by a processor to perform a method as previously described.
Embodiments of the present invention also disclose a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The computer instructions may be read from a computer-readable storage medium by a processor of a computer device, and executed by the processor, to cause the computer device to perform the foregoing method.
The method comprises the steps of firstly, obtaining a patent index to be evaluated; the patent indexes comprise legal index sets, technical index sets and economic index sets; the embodiment of the invention is based on legal index sets, technical index sets and economic index sets, and fully arranges patent information from three dimensions; further, using a first evaluation model, carrying out weight analysis on the patent index based on a preset weighting method to obtain a plurality of groups of weights; the preset weighting method comprises a supervised weighting method and an unsupervised weighting method; corresponding multiple groups of weights are obtained through multiple weighting modes, so that the randomness of evaluation is avoided; comprehensively weighting a plurality of groups of weights based on game theory to obtain comprehensive value scores; the weights of the multiple groups are further comprehensively weighted, so that the weight difference is smaller, and better convergence of the evaluation result is realized; finally, carrying out value analysis on the patent index by using a second evaluation model to obtain commercial value amount; the second evaluation model is generated based on the supervision model through the patent data training of the marked value money labels. In addition to score evaluation, value evaluation is further combined, subjectivity of evaluation is avoided, and meanwhile accurate reference is provided for patent evaluation. The embodiment of the invention can reasonably and accurately evaluate the patent value.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a patent value evaluation method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an overall framework of a patent value evaluation method according to an embodiment of the present invention;
FIG. 3 is a schematic structural diagram of an image super-resolution model according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a PA spatial attention mechanism structure according to an embodiment of the present invention;
fig. 5 is a schematic diagram of an ASFT structure provided in an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
In one aspect, as shown in fig. 1, an embodiment of the present invention provides a patent value evaluation method, including:
s100, acquiring a patent index to be evaluated;
wherein, the patent index comprises a legal index set, a technical index set and an economic index set; it should be noted that, in some embodiments, a patent index system is constructed based on a patent information database; the patent index system is constructed based on legal dimension, technical dimension and economic dimension; patent data to be evaluated are obtained, and the patent data are arranged on the basis of a patent index system to obtain patent indexes.
Specifically, as shown in fig. 2, a patent index system is constructed based on a patent information database, and 28 indexes are all obtained. The method is divided into three index sets according to three dimensions of law, technology and economy, and specifically comprises the following steps: legal dimension has 9 indexes U 1 ＝{u 1,1 ,u 1,2 ,...,u 1,9 The technical dimension has 12 indexes U 2 ＝{u 2,1 ,u 2,2 ,...,u 2,12 7 indexes U in economic dimension 3 ＝{u 3,1 ,u 3,2 ,...,u 3,7 }. Legal index sets include whether PCT international patent, patent type, number of claims, number of independent claims, number of dependent claims, remaining effective time, time of approval, number of pages of description and number of drawings; the technical index set comprises the scale and the citation number of the patent in the same family &Ranking percentage, quiltQuantity of references&Ranking percentage, citation popularity, citation authority, pre-technology popularity, pre-technology novelty, current technology novelty, winning index, technical field scope, inventor number, and diffusion index, economic index set including litigation number&Ranking percentage, number of permissions&Ranking percentage, number of mortgage&Ranking percentage, number of transfers&Ranking percentage, priority number&Ranking percentage, number of invalidations, international market coverage scale. And (3) carrying out supplementary explanation on part of indexes:
heat of reference: after all patents are built into a reference network according to the reference relation, the PageRank algorithm is used for generating the reference heat of each patent. The idea of the PageRank algorithm is to give a directed graph, define a random walk model on the directed graph, namely a first-order Markov chain, and assume that the directed graph jumps to the next node according to the probability of the link relation and the like, and continue. PageRank represents a smooth distribution of this Markov chain. Let us set the transfer matrix M of the reference network. Node probability distribution vector R at any moment t The next time can be recursively derived:
R t+1 ＝MR t
assuming that probabilities of initial time being located at all nodes are equal, the probability vector R of initial time 0 The method comprises the following steps:
where n represents the number of nodes in the reference network and T represents the transpose.
Then as t approaches infinity, the limits of the probability distribution can be obtained:
the limit R of the probability vector is a stable distribution, and mr=r is satisfied. This probability value is the PageRank value of the algorithm. The higher the probability value of the corresponding node, the more important the patent node in the patent network, and the higher the heat of reference.
Further, because the nodes in the reference network are not necessarily fully connected, there may be unconnected nodes. Therefore, a random jump probability needs to be added, 1-alpha is the probability of random jump, and the probability of each node is equal. In contrast, α is the probability of pointing to a jump in accordance with the directed graph. The final plateau distribution R satisfies:
reference authority: after all patents are built into a citation network according to citation relations, an HITS algorithm is used for representing the Authority of citation of each patent by using an Authority value.
The HITS algorithm also measures its importance in the referencing network based on the cited and cited cases of the patent. The algorithm considers that in a patent citation network, if one patent has high importance, then other patents cited by that patent have high importance, while if the patent with high importance is cited by other patents, then other patents cited by that patent have high importance.
The number of patents pointed to in the HITS algorithm is defined as the Hub value, and the number of patents pointed to is defined as the Authority value. An Authority of high is an authoritative patent, which means that the quality of the patent is relatively high. Hub is a Hub patent, which itself integrates more proprietary technology.
The basic idea of the algorithm is that high authority patents can be cited by a plurality of high hub worth patents; a high pivot value patent will refer to a high authority patent, both of which are a synergistic relationship.
The Authority value a of the patent i can be obtained i It is the sum of the Hub values of all patents cited patent i:
wherein, arbitraryH of (2) j Representing Hub values of other patents that refer to patent i.
Hub value h of patent i It is the sum of the authentications of all patents cited by patent i:
wherein any one of a j The authentications values of other patents referring to patent i are represented.
Assume an initial state:
where n is the number of patent nodes in the reference network.
Iterating to convergence according to the following formula:
wherein the method comprises the steps ofRepresenting the Authority value obtained in the next iteration,/->Representing the Hub value obtained for the next iteration.
And always remain normalized, i.e. satisfy vectorsAndthe modulus of (2) is 1:
the resulting authenticability value vector converges to:
a＝(a 1 ,a 2 ,...,a n )
Wherein the method comprises the steps of
The Hub value vector converges to:
h＝(h 1 ,h 2 ,…,h n )
wherein the method comprises the steps of
Let adjacency matrix of patent citation network be A n×n Then, according to the nature of the adjacency matrix, there are:
h＝Aa
a＝A T h
substituting the cancellation h or the cancellation a into each other can obtain:
a＝(A T A)a
h＝(AA T )h
then a is (A) T A) The higher the value of the Authority value representing the page, the higher the Authority value of the page is represented by the element. h is (AA) T ) The higher the value of the Hub representing the page, the higher the central value of the page.
Heat of the pre-technology: the total cited number of patents cited in this patent is counted to represent the heat of the prior art.
The novelty of the prepositioning technology is as follows: statistics of the year novelty of the patent cited in this patent. Using forward reference with temporal weighting, each reference over 5 years is noted as 1, and each patent over 5 years is noted as 5/(1+year-invention year). The cited patents are then summed as a pre-technology novelty indicator for this patent.
Current technology novelty: statistics the year novelty of other patents that refer to this patent. A backward reference representation with time weighting is used, i.e., the time difference between the cited patent and the year of the patent.
Diffusion index: represented by the size of the IPC union of the other patents cited in this patent. The broader the distribution of the subsequent technology, the more original and heuristic it is.
International market coverage scale: the number of countries and regions covered by the patent family is counted, the market sizes of different countries and regions are weighted, and the GDP (global position system) duty ratio of each country and region in 2021 is adopted for weight assignment at present.
Other indexes, such as patent types, claim numbers, etc., can be obtained directly or indirectly based on the patent information, and therefore, the description thereof will be omitted.
S200, carrying out weight analysis on the patent index based on a preset weighting method by using a first evaluation model to obtain a plurality of groups of weights;
the preset weighting method comprises a supervised weighting method and an unsupervised weighting method;
it should be noted that, in some embodiments, the method further includes a step of performing data preprocessing on the numerical indicators in the patent indicators, where the data preprocessing includes a bias distribution operation based on a logarithmic method and a normalization operation.
Specifically, the reference times, the referenced times, the claim number and the like of the data preprocessing operation are distributed in a bias state before the weighting, and y=log (1+x) is used for transformation; other numerical indicators also perform normalization operations (including MinMax).
In some embodiments, the unsupervised weighting method includes a director weighting method and an objective weighting method, and the performing weight analysis on the patent index based on the preset weighting method by using the first evaluation model to obtain multiple sets of weights includes: according to the patent index; subjective weighting is carried out by using an analytic hierarchy process to obtain subjective weighting; and, respectively carrying out objective weighting by using an entropy weight TOPSIS method, a factor analysis weighting method, a principal component analysis weighting method and a CRITIC weighting method to obtain a plurality of groups of objective weights; and performing supervised weighting by using a logistic regression weighting method to obtain the supervised weighting.
Specifically, the weighting algorithm used includes a supervised weighting method and an unsupervised weighting method (including a subjective weighting method and an objective weighting method). The method comprises the following steps: the subjective weighting method has an analytic hierarchy process, the objective weighting method has an entropy weighting TOPSIS method (namely, the entropy weighting method is used for obtaining the weight firstly and then the TOPSIS method is used for researching the TOPSIS method), the factor analysis weighting method, the principal component analysis weighting method and the CRITIC weighting method, and the supervised weighting method is a logistic regression weighting method based on the 5-year patent maintenance information.
The analytic hierarchy process is that for m indexes { u } 1 ,u 2 ,...,u m Comparing the relative importance degree between every two indexes, wherein the corresponding evaluation scale values are shown in table 1:
TABLE 1
The evaluation scale value of index i to index j is filled in the ith row and jth column of the judgment matrix A, and obviously, the diagonal line of A is always 1, and two elements which are symmetrical in diagonal line are reciprocal, namely a ji ＝1/a ij 。
The idealized decision matrix A should satisfy the consistency condition, i.e., a ij ＝a ik a kj . A in the actual case needs to pass the consistency check to confirm that the consistency is approximately met.
Let A have a maximum eigenvalue of lambda max Then define the portability index CI:
smaller CI values demonstrate better consistency, ideally ci=0. The CI value is corrected by introducing a random consistency index RI that measures the magnitude of the consistency deviation caused by the random factor, considering that the deviation of consistency may be due to random reasons. RI index is directly obtained by table look-up 2:
TABLE 2
| Matrix order | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| RI | 0 | 0 | 0.58 | 0.90 | 1.12 | 1.24 | 1.32 | 1.41 | 1.45 | 1.49 |
The final adapted test statistic is CR:
when CR <0.1, the consistency of the judgment matrix is considered acceptable, otherwise appropriate corrections to the judgment matrix are required.
If the judgment matrix A passes the test, the maximum eigenvalue lambda of A max Corresponding feature vector ζ max The vector w= (w) is obtained after normalization operation 1 ,w 2 ,…,w m )，w i The weight value corresponding to the i-th index.
The entropy weight method is to evaluate the sample information of n patents by m indexes, and set the value obtained by the j patent on the i index as X ij 。
First, for each index, a normalization operation is performed, for example, a minimum value and maximum value normalization is adopted: let an index take the value { x }, on a series of patents j -where the minimum value is x min Maximum value of x max Then for any x j The result of the minimum and maximum normalization operation is:
thereby from the original data X ij Obtaining normalized data Y ij The value of the i-th index normalized by the j-th patent is shown.
The information entropy of the first index can be obtained as follows:
the value range of the information entropy is [0,1]. The larger e is, the larger the degree of dispersion of the index i is, and more information can be obtained. Therefore, the index should be given a higher weight. The weight values of the indexes are respectively set as follows:
The TOPSIS algorithm is used for correcting the entropy weight method result. For the evaluation index matrix Y subjected to the normalization operation, the element Y is ij The value of the i-th index normalized by the j-th patent is shown. A normalized matrix is constructed based on the matrix, and the aim is to make the column vector as a unit vector, so that the subsequent unified distance calculation is facilitated. Element in normalized matrix R:
the normalized matrix V of weights is then calculated from the normalized matrix of data:
V ij ＝w i R ij
the positive ideal solution (optimal target) and the negative ideal solution (worst target) are determined by the weight normalization values, respectively. The positive ideal solution is that each index takes the optimal value from the sample, and the negative ideal solution is that each index takes the worst value from the sample. The method comprises the following steps:
wherein:
and->Respectively representing the maximum value and the minimum value in the sample values corresponding to the index i.
The distance of each target to the positive ideal solution and the negative ideal solution is then calculated:
finally, the closeness of the current target to the ideal solution can be obtained:
wherein C is i The value range of (2) is [0,1 ]]The closer to the ideal solution, the more important the specification is. Finally to C i And (3) performing normalization operation:
the obtainedThe result of correcting the weight obtained by the entropy weight method by using the TOPSIS algorithm is used as the index weight of the model.
The purpose of factor analysis is to describe the links between multiple indicators with a few factors, which themselves do not necessarily have specific physical significance. Then, for each index, the accumulated contribution rate of the commonality factors is calculated, and the weight is determined according to the contribution rate.
In contrast, principal component analysis is obtained by linear weighting of the original features. The size of the index weight is determined according to the variance contribution rate of the principal component.
The CRITIC method combines the standard deviation method and the correlation coefficient method, based on two basic assumptions: the larger the standard deviation of the numerical values in the same index is, the larger the information quantity is, and the stronger positive correlation between the two indexes indicates that the conflict of the two indexes is lower.
Let the value of the j patent after normalization operation on the i index be x ij And the average value of the i index after normalization isThe standard deviation s of the i-th index is used to represent the information amount of the index itself:
the larger the standard deviation is, the larger the numerical difference of the index is, the more information can be reflected, the stronger the evaluation strength of the index is, and more weight should be allocated to the index.
The collision of the indicators is represented by a correlation coefficient. Let the correlation coefficient between index i and index j be r ij Then use:
to represent the degree of overall correlation between the index i itself and other indexes. R is R i The smaller the collision with other indexes, the more the same information is reflected, and the more the presented evaluation content is repeated, the evaluation strength of the index is weakened to a certain extent, and the weight allocated to the index is reduced.
Thus combining the information amount and the conflict, a CRITIC value is obtained:
C i the larger the i-th evaluation index is, the greater the effect in the whole evaluation index system, the more weight should be assigned to it. And thus, the weight value of the ith index is obtained through normalization operation:
s300, comprehensively weighting a plurality of groups of weights based on game theory to obtain comprehensive value scores;
it should be noted that, in some embodiments, the method includes: performing arbitrary linear combination on the multiple groups of weights to obtain a first comprehensive weight; optimizing weight coefficients of each group of weights through a game theory based on the first comprehensive weights; performing first derivative optimization according to the differential property of the matrix to obtain a target weight coefficient set; determining a second comprehensive weight according to the target weight coefficient set and the combined weight; wherein the second comprehensive weight comprises the comprehensive weight of each index in the patent indexes; and determining the comprehensive value score according to the target weight coefficient set and the weight.
In some embodiments, determining the composite value score according to the target weight coefficient set and the weights includes: determining a comprehensive value score according to the target weight coefficient set and the weight through a comprehensive value score formula; wherein, the comprehensive value score formula is:
wherein Score represents the total value Score of a given patent, M represents the number of patent indexes, w i Representing the weight of the ith index in a scoring system, v i The score obtained by the i-th index of the patent (value [0,1 ]])。
Specifically, according to the Nash equilibrium theory of game theory, for multiple sets of weight values, the model can minimize the deviation between the comprehensive weight and each initial weight, and seek to obtain an optimal combination state. The L methods are provided for calculating weights for M evaluation indexes, and the kth group of weight vectors is set as u k ＝(u k1 ,u k2 ,…,u kM ) The integrated weight vector obtained by arbitrary linear combination of the L different vectors is:
optimizing L weight coefficients alpha by using game theory idea k So that u is equal to each u k The dispersion of (2) is minimized, namely:
wherein alpha is j Weight coefficient vector representing the j-th set of weights, T representing the transpose, u i And u j Respectively representing the i-th and j-th sets of weight vectors.
Based on the differential properties of the matrix, the first derivative is de-optimized, translating into:
According to the above equation (alpha) 1 ,α 2 ,…,α L ) Then finally the comprehensive weight of the ith evaluation indexThe weight is as follows:
wherein alpha is k Weight coefficient representing kth group weight, u ki The weight of the kth item evaluation index is represented.
And then by the generated weight { w } 1 ,w 2 ,…,w 28 Obtaining the comprehensive value score of the patent:
wherein Score represents the comprehensive value Score, M represents the number of patent indexes, w i Representing the weight of the ith index in a scoring system, v i The score obtained by the patent at the ith index is shown. The score obtained is a percentile, and the higher the score, the higher the overall value.
S400, performing value analysis on the patent index by using a second evaluation model to obtain commercial value amount;
the second evaluation model is generated based on the supervision model through the training of the patent data marked with the value and monetary labels; it should be noted that, in some embodiments, the method further includes a step of generating a second evaluation model based on the supervision model through the patent data training of the value money label, and this step includes: acquiring patent case data of a preset term; the case data comprises case money value; correcting the patent case data based on the historical consumer commodity price index to obtain label data of the value amount; patent data of patents corresponding to patent case data are obtained, and training data are obtained through arrangement based on preset patent indexes; and inputting the training data and the label data into a pre-constructed supervision model for monetary assessment training, and adjusting the supervision model based on the training result to obtain a second assessment model.
Specifically, the supervision model may be an XGBoost model, and the patent commercial value evaluation model training is performed based on the XGBoost model, which is described in detail below. The index of patent value evaluation also relates to three dimensions of law, technology and economy, but the index of patent value evaluation is more focused on economic value because the selection of specific indexes is different from the index of comprehensive quality evaluation. The selection of specific indicators is shown in fig. 3.
Because of the complexity of the patent value evaluation problem, the modeling evaluation cannot be directly performed by adopting a linear model, and therefore, the patent value evaluation is modeled by adopting an XGBoost model. Model training was performed using court case data from the 1991-2021 section.
Carrying out data preprocessing operation before model training, wherein the reference times, the referenced times, the claim number and the like are distributed in a biased state, and transforming by using a logarithmic method; other numerical indexes also perform MinMax normalization operation.
Case monetary value is corrected with reference to the value of the past year consumer price index (Consumer Price Index, CPI) for a 30 year time span, unifying with 2021 purchasing power.
The amount between 100 yuan and 1 billion yuan was screened and treated with a base 10 logarithm. Patent information with the compensation amount greater than or equal to 100 yuan and less than or equal to 1 hundred million yuan is screened out, and the compensation amount is processed by taking the logarithm based on 10, so that the compensation amount is converted from long tail distribution to unimodal distribution which is approximate to normal, and the compensation amount is taken as a prediction target. And meanwhile, inputting the normalized index values of the corresponding patents into the model for training to obtain the corresponding patent value evaluation model.
And then select 23 indexes { v } 1 ,v 2 ,…,v 23 Inputting a training completion model to obtain a logarithmic value based on 10 of the commercial value of the patent, and obtaining the estimated specific amount by exponential reduction:
Price＝exp(f XGBoost (v 1 ,v 2 ,…,v 23 ))
in the XGBoost model for the regression experiment, the SHAP value visualization (shown in fig. 4) of each index can show that the SHAP values of the indexes such as the newly added index rd, forwardcitnumn, pagerank, weighed _ cited, weighed _cit are higher, and the contribution degree of the newly added indexes in the model is proved to be larger, so that the newly added indexes have a certain value and are improved compared with the original model.
In some embodiments, the implementation flow of the present invention is shown in fig. 5, and includes:
1. establishing a patent comprehensive value index system according to the indexes shown in fig. 2;
2. generating comprehensive value evaluation weights by a plurality of weighting methods, including a hierarchical analysis method, an entropy weighting method+TOPSIS, a factor analysis weighting method, a principal component analysis weighting method, a CRITIC weighting method and a logistic regression weighting method based on 5-year patent maintenance information;
3. obtaining comprehensive weights by using a comprehensive weighting method based on game theory, and using the comprehensive weights in a patent comprehensive value index system;
4. establishing a patent commercial value index system according to the indexes shown in fig. 3;
5. Training a patent commercial value evaluation model by using the patent commercial value index and the court case data of the 1991-2021 part of patent;
6. and inputting a patent comprehensive value evaluation model and a patent commercial value evaluation model according to the data of the patent given by the index set, and respectively generating a patent comprehensive value score and a patent commercial value amount.
In summary, compared with the prior art, the invention at least has the following beneficial effects: 1. the reference information is mined deeply. Considering influence of patents, a reference network is constructed according to specific reference information of the patents, nodes in the network represent the patents, and directed edges represent reference relations. And then adopting a graph analysis algorithm, including PageRank and HITS algorithms, to analyze the reference heat and the reference authority of the patent nodes in the reference network. Compared with the conventional model, only counting the number of references and the number of the references, the model digs out deeper information of patent references; 2. the true patent data is taken as reference, so that the method has persuasion and credibility. The real patent is used for 5 years to maintain information as a reference of the comprehensive quality fraction of the patent, and the real court case data of the patent is used as a reference of the patent value evaluation. There are studies showing that the maintenance information of the patent itself can be used as an evaluation criterion for whether it is a high quality patent. Investigation shows that the whole maintenance time of the Chinese patent is short, and about half of the patents are abandoned within 5 years after the patent is authorized, and the reason is probably that the purpose of applying the patent is to complete the task or that the economic value of the patent per se is not high for enterprises, and the power for continuing to pay annual fee maintenance is not available. Therefore, the maintenance condition of the patent within 5 years can be used as a judging standard of the comprehensive quality of the patent. Likewise, the court case data is generated in the actual judicial process, and the patent value can be truly represented through the amount obtained by weighing the court and related institutions; 3. in the XGBoost model aiming at the regression experiment, the SHAP value visualization (shown in fig. 4) of each index can show that the SHAP values of the newly added partial indexes are higher, and the contribution degree of the newly added partial indexes in the model is proved to be larger, so that the newly added indexes have a certain value, and compared with the original model which does not comprise the indexes, the newly added indexes are improved; 4. expert scoring is not needed, the difficulty of patent value evaluation is reduced, and subjectivity and randomness existing in the evaluation process are reduced.
In another aspect, an embodiment of the present invention provides a patent value evaluation system, including: the first module is used for acquiring the patent index to be evaluated; the patent indexes comprise legal index sets, technical index sets and economic index sets; the second module is used for carrying out weight analysis on the patent indexes based on a preset weighting method by utilizing the first evaluation model to obtain a plurality of groups of weights; the preset weighting method comprises a supervised weighting method and an unsupervised weighting method; the third module is used for comprehensively weighting a plurality of groups of weights based on the game theory to obtain a comprehensive value score; a fourth module, configured to perform value analysis on the patent index by using the second evaluation model, to obtain a commercial value amount; the second evaluation model is generated based on the supervision model through the patent data training of the marked value money labels.
The content of the method embodiment of the invention is suitable for the system embodiment, the specific function of the system embodiment is the same as that of the method embodiment, and the achieved beneficial effects are the same as those of the method.
Another aspect of the embodiment of the invention also provides a patent value evaluation device, which comprises a processor and a memory;
The memory is used for storing programs;
the processor executes a program to implement the method as before.
The content of the method embodiment of the invention is suitable for the device embodiment, the specific function of the device embodiment is the same as that of the method embodiment, and the achieved beneficial effects are the same as those of the method.
Another aspect of the embodiments of the present invention also provides a computer-readable storage medium storing a program that is executed by a processor to implement a method as before.
The content of the method embodiment of the invention is applicable to the computer readable storage medium embodiment, the functions of the computer readable storage medium embodiment are the same as those of the method embodiment, and the achieved beneficial effects are the same as those of the method.
Embodiments of the present invention also disclose a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The computer instructions may be read from a computer-readable storage medium by a processor of a computer device, and executed by the processor, to cause the computer device to perform the foregoing method.
In some alternative embodiments, the functions/acts noted in the block diagrams may occur out of the order noted in the operational illustrations. For example, two blocks shown in succession may in fact be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved. Furthermore, the embodiments presented and described in the flowcharts of the present invention are provided by way of example in order to provide a more thorough understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed, and in which sub-operations described as part of a larger operation are performed independently.
Furthermore, while the invention is described in the context of functional modules, it should be appreciated that, unless otherwise indicated, one or more of the functions and/or features may be integrated in a single physical device and/or software module or may be implemented in separate physical devices or software modules. It will also be appreciated that a detailed discussion of the actual implementation of each module is not necessary to an understanding of the present invention. Rather, the actual implementation of the various functional modules in the apparatus disclosed herein will be apparent to those skilled in the art from consideration of their attributes, functions and internal relationships. Accordingly, one of ordinary skill in the art can implement the invention as set forth in the claims without undue experimentation. It is also to be understood that the specific concepts disclosed are merely illustrative and are not intended to be limiting upon the scope of the invention, which is to be defined in the appended claims and their full scope of equivalents.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method of the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution apparatus, device, or apparatus, such as a computer-based apparatus, processor-containing apparatus, or other apparatus that can fetch the instructions from the instruction execution apparatus, device, or apparatus and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution apparatus, device, or apparatus.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium may even be paper or other suitable medium upon which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.
It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution device. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that: many changes, modifications, substitutions and variations may be made to the embodiments without departing from the spirit and principles of the invention, the scope of which is defined by the claims and their equivalents.
While the preferred embodiment of the present invention has been described in detail, the present invention is not limited to the embodiments, and those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention, and the equivalent modifications or substitutions are intended to be included in the scope of the present invention as defined in the appended claims.
Claims (10)
1. A method for evaluating patent value, comprising:
obtaining a patent index to be evaluated; the patent indexes comprise legal index sets, technical index sets and economic index sets;
carrying out weight analysis on the patent indexes based on a preset weighting method by using a first evaluation model to obtain a plurality of groups of weights; the preset weighting method comprises a supervised weighting method and an unsupervised weighting method;
comprehensively weighting a plurality of groups of weights based on game theory to obtain comprehensive value scores;
performing value analysis on the patent index by using a second evaluation model to obtain commercial value amount; the second evaluation model is generated based on a supervision model through patent data training of the marked value money labels.
2. The method for evaluating patent value according to claim 1, wherein the step of obtaining the patent index to be evaluated comprises:
constructing a patent index system based on a patent information database;
the patent index system is constructed based on legal dimension, technical dimension and economic dimension;
and acquiring the patent data to be evaluated, and sorting the patent data based on the patent index system to obtain the patent index.
3. The patent value evaluation method according to claim 1, characterized by further comprising:
carrying out data preprocessing on numerical indexes in the patent indexes;
wherein the data preprocessing comprises a log-based bias distribution operation and a normalization operation.
4. The patent value evaluation method according to claim 1, wherein the unsupervised weighting method includes a master weighting method and an objective weighting method, the weighting analysis is performed on the patent index based on a preset weighting method by using a first evaluation model to obtain a plurality of sets of weights, and the method comprises:
according to the patent index;
subjective weighting is carried out by using an analytic hierarchy process to obtain subjective weighting;
and, respectively carrying out objective weighting by using an entropy weight TOPSIS method, a factor analysis weighting method, a principal component analysis weighting method and a CRITIC weighting method to obtain a plurality of groups of objective weights;
and performing supervised weighting by using a logistic regression weighting method to obtain the supervised weighting.
5. The patent value evaluation method according to claim 1, wherein the comprehensively weighting a plurality of sets of weights based on game theory to obtain a comprehensive value score comprises:
Performing arbitrary linear combination on a plurality of groups of weights to obtain a first comprehensive weight;
optimizing weight coefficients of each group of weights through game theory based on the first comprehensive weights; performing first derivative optimization according to the differential property of the matrix to obtain a target weight coefficient set;
determining a second comprehensive weight by combining the weights according to the target weight coefficient set; wherein the second comprehensive weight comprises the comprehensive weight of each index in the patent indexes;
and determining the comprehensive value score according to the target weight coefficient set and the weight.
6. The patent value assessment method according to claim 5, wherein said determining a composite value score from said target weight coefficient set and said weights comprises:
determining a comprehensive value score according to the target weight coefficient set and the weight through a comprehensive value score formula;
wherein, the comprehensive value score formula is:
wherein Score represents the integrated value Score, M represents the number of patent indexes, w i Comprehensive weight of the i-th index, v i The score obtained by the i-th index is shown.
7. The patent value assessment method according to claim 1, further comprising the step of generating a second assessment model based on the supervision model through the training of the patent data with value money tags noted, the step comprising:
Acquiring patent case data of a preset term; the case data comprises case money value;
correcting the patent case data based on the historical consumer price index to obtain tag data of the value amount;
acquiring patent data of a patent corresponding to the patent case data, and sorting based on preset patent indexes to obtain training data;
and inputting the training data and the label data into a pre-constructed supervision model for monetary assessment training, and adjusting the supervision model based on a training result to obtain a second assessment model.
8. A patent value evaluation system, comprising:
the first module is used for acquiring the patent index to be evaluated; the patent indexes comprise legal index sets, technical index sets and economic index sets;
the second module is used for carrying out weight analysis on the patent indexes based on a preset weighting method by utilizing the first evaluation model to obtain a plurality of groups of weights; the preset weighting method comprises a supervised weighting method and an unsupervised weighting method;
the third module is used for comprehensively weighting a plurality of groups of weights based on game theory to obtain comprehensive value scores;
a fourth module, configured to perform value analysis on the patent index by using the second evaluation model, to obtain a commercial value amount; the second evaluation model is generated based on a supervision model through patent data training of the marked value money labels.
9. A patent value evaluation device comprises a processor and a memory;
the memory is used for storing programs;
the processor executing the program implements the method of any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that the storage medium stores a program that is executed by a processor to implement the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310422842.5A CN116452052A (en) | 2023-04-19 | 2023-04-19 | Patent value evaluation method, system, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310422842.5A CN116452052A (en) | 2023-04-19 | 2023-04-19 | Patent value evaluation method, system, device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116452052A true CN116452052A (en) | 2023-07-18 |
Family
ID=87133298
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310422842.5A Pending CN116452052A (en) | 2023-04-19 | 2023-04-19 | Patent value evaluation method, system, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116452052A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117314253A (en) * | 2023-10-13 | 2023-12-29 | 武汉索元数据信息有限公司 | Value evaluation method and device |
| CN117932285A (en) * | 2024-03-25 | 2024-04-26 | 清华大学 | Data importance assessment method and device based on different composition |
| CN118365008A (en) * | 2024-06-20 | 2024-07-19 | 青岛中投创新技术转移有限公司 | Patent management system based on blockchain technology |
-
2023
- 2023-04-19 CN CN202310422842.5A patent/CN116452052A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117314253A (en) * | 2023-10-13 | 2023-12-29 | 武汉索元数据信息有限公司 | Value evaluation method and device |
| CN117932285A (en) * | 2024-03-25 | 2024-04-26 | 清华大学 | Data importance assessment method and device based on different composition |
| CN118365008A (en) * | 2024-06-20 | 2024-07-19 | 青岛中投创新技术转移有限公司 | Patent management system based on blockchain technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | The role of text-extracted investor sentiment in Chinese stock price prediction with the enhancement of deep learning | |
| Caruso et al. | Cluster Analysis for mixed data: An application to credit risk evaluation | |
| Cao et al. | Exploration of stock index change prediction model based on the combination of principal component analysis and artificial neural network | |
| Singh et al. | Re-estimation and comparisons of alternative accounting based bankruptcy prediction models for Indian companies | |
| Min et al. | A binary classification method for bankruptcy prediction | |
| Brezigar-Masten et al. | CART-based selection of bankruptcy predictors for the logit model | |
| TW530234B (en) | Methods and systems for efficiently sampling portfolios for optimal underwriting | |
| KR100771718B1 (en) | Methods and systems for automated inferred valuation of credit scoring | |
| CN116452052A (en) | Patent value evaluation method, system, device and storage medium | |
| CZ20013132A3 (en) | Valuation prediction models in situations with missing inputs | |
| CN110009503A (en) | Finance product recommended method, device, computer equipment and storage medium | |
| Zhou et al. | Predicting pavement roughness using deep learning algorithms | |
| Cao et al. | Early warning of enterprise decline in a life cycle using neural networks and rough set theory | |
| Aguilar-Palacios et al. | Cold-start promotional sales forecasting through gradient boosted-based contrastive explanations | |
| Wichmann et al. | Auctioning risky conservation contracts | |
| Pechlivanidis et al. | Can intangible assets predict future performance? A deep learning approach | |
| Odeh et al. | A multi-objective approach for the prediction of loan defaults | |
| WO2021257610A1 (en) | Time series forecasting and visualization methods and systems | |
| Liu et al. | Probabilistic graph-based valuation model for measuring the relative patent value in a valuation scenario | |
| Zhu et al. | Risk assessment of biological asset mortgage loans of China’s new agricultural business entities | |
| Yang et al. | A carbon price hybrid forecasting model based on data multi-scale decomposition and machine learning | |
| Peng et al. | Attention-based CNN–LSTM for high-frequency multiple cryptocurrency trend prediction | |
| Wanke et al. | Endogenous and exogenous performance sources in Asian Banking: A hybrid stochastic Multi-Criteria Decision-Making approach based on sign decomposition and transfer entropy | |
| Ferguson et al. | Institution-driven comparative advantage, complex goods and organizational choice | |
| CN115841345B (en) | Cross-border big data intelligent analysis method, system and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |