Patent combination value analysis method, apparatus and device, and storage medium
Through semantic recognition and similarity calculation, the problem of low accuracy in patent portfolio value analysis is solved, and the accurate evaluation of patent portfolio value is achieved.
Patent Information
- Application Number
- CN202411547414.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-01
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The prior art has poor accuracy in patent portfolio value analysis, especially in the value analysis process of multiple patents, where the simple superposition method is inaccurate.
By obtaining the identification and text of the patent, semantic recognition is performed to obtain technical semantic features and interpret semantic features, calculate the similarity to determine the correction coefficient, and weighted correction of the patent value to form a combined value.
It improves the accuracy of patent portfolio value analysis, avoids the error caused by simple superposition, and provides a more accurate portfolio value evaluation.
Smart Images

Figure CN119067126A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of computer technology, and in particular to a method, device, equipment and storage medium for analyzing the value of a patent portfolio. Background Art
[0002] At present, the value of patents is generally analyzed manually. This analysis method relies on the experience and knowledge reserves of analysts and is highly subjective. In addition, especially when analyzing the value of multiple patents, simply adding up the value of each patent may lead to inaccurate analysis.
[0003] Therefore, current analytical solutions for patent portfolio value have poor accuracy. Summary of the invention
[0004] The present invention provides a method, device, equipment and storage medium for analyzing the value of a patent portfolio, which can improve the accuracy of the analysis of the value of a patent portfolio.
[0005] In order to achieve the above object, the present invention adopts the following technical scheme: In a first aspect, the present invention provides a method for analyzing the value of a patent portfolio, the method comprising: Obtaining patent identifiers of a plurality of patents to be analyzed, wherein the patent identifiers of the plurality of patents at least include a first identifier of a first patent and a second identifier of a second patent; According to the first identifier, a first value and a first patent text of the first patent are obtained, and according to the second identifier, a second value and a second patent text of the second patent are obtained; the first patent text includes a first claim text and a first specification text, and the second patent text includes a second claim text and a second specification text; Performing semantic recognition on the first claim text to obtain a first technical semantic feature, and performing semantic recognition on the second claim text to obtain a second technical semantic feature; Acquire a first interpretation semantic feature related to the first technical semantic feature from the first specification text, and acquire a second interpretation semantic feature related to the second technical semantic feature from the second specification text; determining a first correction coefficient according to a similarity between the first interpreted semantic feature and the second interpreted semantic feature; Determine the similarity between the first patent and the second patent according to the first specification text and the second specification text, and determine a second correction coefficient according to the similarity between the first patent and the second patent; Performing a weighted summation on the first correction coefficient and the second correction coefficient to obtain a combined correction coefficient; According to the combination correction coefficient, the sum of the first value and the second value is corrected to obtain the combined value of the first patent and the second patent.
[0006] Optionally, determining a first correction coefficient according to a similarity between the first interpreted semantic feature and the second interpreted semantic feature includes: Determine the relationship between the similarity between the first interpretation semantic feature and the second interpretation semantic feature and a first similarity threshold, and obtain a first determination result; If the first judgment result indicates that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is less than a first similarity threshold, determining that the first correction coefficient is less than 1; If the first judgment result indicates that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is equal to a first similarity threshold, determining that a first correction coefficient is equal to 1; If the first judgment result indicates that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is greater than a first similarity threshold, determining that a first correction coefficient is greater than 1; The first correction coefficient is positively correlated with the similarity between the first interpretation semantic feature and the second interpretation semantic feature.
[0007] Optionally, determining a second correction coefficient according to the similarity between the first patent and the second patent includes: Determine the relationship between the similarity between the first patent and the second patent and a second similarity threshold, and obtain a second determination result; If the second judgment result indicates that the similarity between the first patent and the second patent is greater than or equal to the second similarity threshold, then the first technical semantic feature and the second technical semantic feature are compared to obtain a first distinguishing semantic feature in the first technical semantic feature and a second distinguishing semantic feature in the second technical semantic feature, respectively, and the first technical semantic feature, the first distinguishing semantic feature, the second technical semantic feature, the second distinguishing semantic feature, and the technical field feature to which the first patent and the second patent belong are input into a pre-trained prediction model to obtain a probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship in the technical field corresponding to the technical field feature; a second correction coefficient is determined according to the probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship, wherein the second correction coefficient is positively correlated with the probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship, and the second correction coefficient is greater than or equal to 1; If the second judgment result indicates that the similarity between the first patent and the second patent is less than the second similarity threshold, it is determined that the second correction coefficient is less than 1, and the second correction coefficient is positively correlated with the similarity between the first patent and the second patent.
[0008] Optionally, the acquiring a first interpretation semantic feature related to the first technical semantic feature from the first specification text, and acquiring a second interpretation semantic feature related to the second technical semantic feature from the second specification text includes: Performing semantic recognition on the first specification text to obtain a first specification semantic feature, determining a first specification sub-semantic feature based on the first specification semantic feature and the first technical semantic feature, determining a first specification sub-text corresponding to the first specification sub-semantic feature, and performing semantic recognition on the first specification sub-text and a context text of the first specification sub-text to obtain the first interpretation semantic feature; Perform semantic recognition on the second specification text to obtain second specification semantic features, determine second specification sub-semantic features based on the second specification semantic features and the second technical semantic features, determine the second specification sub-text corresponding to the second specification sub-semantic features, perform semantic recognition on the second specification sub-text and the context text of the second specification sub-text to obtain the second interpretation semantic features.
[0009] Optionally, performing weighted summation on the first correction coefficient and the second correction coefficient to obtain a combined correction coefficient includes: Select a target option from a plurality of weight options, and determine a first weight of the first correction factor and a second weight of the second correction factor; A product of the first correction coefficient and the first weight is added to a product of the second correction coefficient and the second weight to obtain a combined correction coefficient.
[0010] In a second aspect, the present invention provides a device for analyzing the value of a patent portfolio, the device comprising: an acquisition module, configured to acquire patent identifiers of a plurality of patents to be analyzed, wherein the patent identifiers of the plurality of patents at least include a first identifier of a first patent and a second identifier of a second patent; acquire a first value and a first patent text of the first patent according to the first identifier, and acquire a second value and a second patent text of the second patent according to the second identifier; the first patent text includes a first claim text and a first specification text, and the second patent text includes a second claim text and a second specification text; a semantic recognition module, configured to perform semantic recognition on the first claim text to obtain a first technical semantic feature, and perform semantic recognition on the second claim text to obtain a second technical semantic feature; obtain a first interpretation semantic feature related to the first technical semantic feature from the first specification text, and obtain a second interpretation semantic feature related to the second technical semantic feature from the second specification text; A first determination module, configured to determine a first correction coefficient according to a similarity between the first interpretation semantic feature and the second interpretation semantic feature; a second determination module, configured to determine the similarity between the first patent and the second patent according to the first specification text and the second specification text, and determine a second correction coefficient according to the similarity between the first patent and the second patent; The analysis module is used to perform weighted summation of the first correction coefficient and the second correction coefficient to obtain a combined correction coefficient; and to correct the sum of the first value and the second value according to the combined correction coefficient to obtain a combined value of the first patent and the second patent.
[0011] Optionally, the first determination module is specifically used to determine the relationship between the similarity between the first interpreted semantic feature and the second interpreted semantic feature and a first similarity threshold to obtain a first judgment result; if the first judgment result indicates that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is less than the first similarity threshold, then the first correction coefficient is determined to be less than 1; if the first judgment result indicates that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is equal to the first similarity threshold, then the first correction coefficient is determined to be equal to 1; if the first judgment result indicates that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is greater than the first similarity threshold, then the first correction coefficient is determined to be greater than 1; wherein, the first correction coefficient is positively correlated with the similarity between the first interpreted semantic feature and the second interpreted semantic feature.
[0012] Optionally, the second determination module is specifically used to determine the relationship between the similarity between the first patent and the second patent and a second similarity threshold to obtain a second judgment result; if the second judgment result indicates that the similarity between the first patent and the second patent is greater than or equal to the second similarity threshold, then the first technical semantic feature and the second technical semantic feature are compared to obtain a first distinguishing semantic feature in the first technical semantic feature and a second distinguishing semantic feature in the second technical semantic feature, respectively, and the first technical semantic feature, the first distinguishing semantic feature, the second technical semantic feature, the second distinguishing semantic feature, and the technical field feature to which the first patent and the second patent belong are input into a pre-trained prediction model to obtain a probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship in the technical field corresponding to the technical field feature; according to the probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship, determine a second correction coefficient, wherein the second correction coefficient is positively correlated with the probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship, and the second correction coefficient is greater than or equal to 1; if the second judgment result indicates that the similarity between the first patent and the second patent is less than the second similarity threshold, determine that the second correction coefficient is less than 1, and the second correction coefficient is positively correlated with the similarity between the first patent and the second patent.
[0013] In a third aspect, the present invention provides a computing device, including a memory and a processor; One or more computer programs are stored in the memory, and the one or more computer programs include instructions; when the instructions are executed by the processor, the computing device executes the method as described in any one of the first aspects.
[0014] In a fourth aspect, the present invention provides a computer-readable storage medium for storing a computer program for executing the method as described in any one of the first aspects.
[0015] It can be seen from the above technical solution that the present invention has at least the following beneficial effects: The present invention provides a method for analyzing the value of a patent combination. The method can be executed by a processing device. The processing device can first obtain a first patent and a second patent to be analyzed, and then respectively perform semantic recognition on the first claim text of the first patent and the second claim text of the second patent to obtain a first technical semantic feature and a second technical semantic feature, and then obtain a first interpretation semantic feature related to the first technical semantic feature from the first specification text, and obtain a second interpretation semantic feature related to the second technical semantic feature from the second specification text; then determine the first correction coefficient based on the similarity between the first interpretation semantic feature and the second interpretation semantic feature; at the same time, the processing device determines the similarity between the first patent and the second patent according to the first specification text and the second specification text, and determines the second correction coefficient according to the similarity between the first patent and the second patent, and then weighted sums the first correction coefficient and the second correction coefficient to obtain a combination correction coefficient, and finally uses the combination correction coefficient to correct the sum of the first value and the second value to obtain the combination value of the first patent and the second patent. The method adjusts the combination value of patents from two levels, rather than simply superimposing the values of multiple patents, so that the value of the patent combination can be given more accurately.
[0016] It should be understood that the description of technical features, technical solutions, beneficial effects or similar language in the present invention does not imply that all features and advantages can be realized in any single embodiment. On the contrary, it is understood that the description of features or beneficial effects means that specific technical features, technical solutions or beneficial effects are included in at least one embodiment. Therefore, the description of technical features, technical solutions or beneficial effects in this specification does not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions and beneficial effects described in the present embodiment can also be combined in any appropriate manner. Those skilled in the art will understand that the embodiment can be realized without one or more specific technical features, technical solutions or beneficial effects of a specific embodiment. In other embodiments, additional technical features and beneficial effects can also be identified in specific embodiments that do not embody all embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 A flow chart of a method for analyzing the value of a patent portfolio provided by an embodiment of the present invention; Figure 2 A schematic diagram of a patent portfolio value analysis device provided by an embodiment of the present invention; Figure 3 A schematic diagram of a computing device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0018] The terms "first", "second", "third", etc. in the specification of the present invention and the accompanying drawings are used to distinguish different objects rather than to limit a specific order.
[0019] In the embodiments of the present invention, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.
[0020] In order to make the technical solution of the present invention clearer and easier to understand, the technical solution of the present invention is introduced below in conjunction with the accompanying drawings. Figure 1 As shown, this figure is a flow chart of a method for analyzing the value of a patent portfolio provided by an embodiment of the present invention. The method can be executed by a processing device, which can be a terminal or a server. Terminals include but are not limited to smart phones, tablet computers, laptops, personal digital assistants, or smart wearable devices. The server can be a cloud server, such as a central server in a central cloud computing cluster, or an edge server in an edge cloud computing cluster. Of course, the server can also be a server in a local data center. A local data center refers to a data center that is directly controlled by the user.
[0021] Specifically, the method includes: S101. A processing device obtains patent identifications of multiple patents to be analyzed.
[0022] The patent identifiers of the multiple patents include at least a first identifier of a first patent and a second identifier of a second patent.
[0023] In some embodiments, the processing device may provide a user with an interface supporting human-computer interaction, and the user may input patent identifiers of multiple patents to be analyzed on the human-computer interaction interface, and then the processing device obtains the patent identifiers of multiple patents to be analyzed input by the user. The patent identifier may be a patent application number, publication number, authorization number, etc., and the patent identifier is used to uniquely mark the patent to be analyzed.
[0024] For ease of description, multiple patents including a first patent and a second patent are taken as an example for introduction below, where the patent identifier of the first patent is the first identifier, and the patent identifier of the second patent is the second identifier.
[0025] S102. The processing device obtains a first value and a first patent text of the first patent according to the first identifier, and obtains a second value and a second patent text of the second patent according to the second identifier.
[0026] After the processing device obtains the patent identifier, it can search for the patent text corresponding to the patent identifier in the public database, wherein the public database includes all patent texts that have been in the public state.
[0027] It should be noted that the processing device obtaining the patent text based on the patent identification is only an example. In other embodiments, the user can also directly provide the patent text of the patent to be analyzed to the processing device instead of providing the patent identification. The patent text includes a first patent text and a second patent text. The first patent text includes a first claim text and a first specification text. The second patent text includes a second claim text and a second specification text.
[0028] In some embodiments, the processing device can obtain the patent value of each patent in real time. After the user inputs the first identifier and the second identifier, the first value of the first patent can be searched based on the first identifier, and the second value of the second patent can be searched based on the second identifier.
[0029] It should be noted that the present invention does not specifically limit the method for determining the patent value of a single patent. The purpose of the present invention is to determine the value of these patent combinations based on the value of a single patent.
[0030] S103: The processing device performs semantic recognition on the first claim text to obtain a first technical semantic feature, and performs semantic recognition on the second claim text to obtain a second technical semantic feature.
[0031] The first claim text may be the text of the solution corresponding to a claim in the first patent. For example, take the claims shown in Table 1 as an example: Table 1:
[0032] For example, if the first claim text is the text of the solution corresponding to the first claim of the first patent, then the first claim text is "A photographing method, characterized in that it includes: S1; S2 and S3"; for another example, if the first claim text is the text of the solution corresponding to the second claim of the first patent, then the first claim text is "A photographing method, characterized in that it includes: S1; S2 and S3; said S1 includes S11; S12".
[0033] The method for determining the second claim text is similar to that for the first claim text and will not be repeated here.
[0034] After the processing device determines the first claim text and the second claim text respectively, it can perform semantic recognition on the first claim text to obtain a first technical semantic feature, and perform semantic recognition on the second claim text to obtain a second technical semantic feature.
[0035] S104: The processing device obtains a first interpretation semantic feature related to the first technical semantic feature from the first specification text, and obtains a second interpretation semantic feature related to the second technical semantic feature from the second specification text.
[0036] The first interpretation semantic feature refers to the feature of the text in the first specification text used to interpret the first claim text, and the second interpretation semantic feature refers to the feature of the text in the second specification text used to interpret the second claim text. The first claim text corresponds to the first technical semantic feature, and the second claim text corresponds to the second technical semantic feature.
[0037] In some embodiments, the processing device can perform semantic recognition on the first specification text to obtain the first specification semantic feature, determine the first specification sub-semantic feature based on the first specification semantic feature and the first technical semantic feature, determine the first specification sub-text corresponding to the first specification sub-semantic feature, perform semantic recognition on the first specification sub-text and the context text of the first specification sub-text to obtain the first interpretation semantic feature; perform semantic recognition on the second specification text to obtain the second specification semantic feature, determine the second specification sub-semantic feature based on the second specification semantic feature and the second technical semantic feature, determine the second specification sub-text corresponding to the second specification sub-semantic feature, perform semantic recognition on the second specification sub-text and the context text of the second specification sub-text to obtain the second interpretation semantic feature.
[0038] The following is an introduction using an example in which a processing device determines a first interpreted semantic feature.
[0039] The first specification text may refer to a text of a certain paragraph in the specification of the first patent. The processing device identifies the first specification text to obtain a first specification semantic feature, and then the processing device may calculate the similarity between the first specification semantic feature and the first technical semantic feature. If the similarity between the first specification semantic feature and the first technical semantic feature is greater than or equal to a first preset threshold, the first specification semantic feature is determined as a first specification sub-semantic feature related to the first technical semantic feature; if the similarity between the first specification semantic feature and the first technical semantic feature is less than the first preset threshold, the search continues, that is, the text of the next paragraph in the specification is determined, and then the text of the next paragraph is semantically identified, and the specification semantic feature of the next paragraph is compared with the first technical semantic feature for similarity.
[0040] After obtaining the first specification sub-semantic feature, the processing device can determine the first specification sub-text corresponding to the first specification sub-semantic feature, and then determine the context text of the first specification sub-text, wherein the context text can be the text of the first m paragraphs and the text of the last n paragraphs of the first specification sub-text, wherein m and n are both positive integers. In some examples, m=2 and n=3.
[0041] Finally, the processing device performs semantic recognition on the first specification sub-text and the context text of the first specification sub-text to obtain a first interpretation semantic feature.
[0042] It should be noted that the processing device determines the second interpreted semantic feature in a similar manner, which will not be described in detail here.
[0043] S105: The processing device determines a first correction coefficient according to the similarity between the first interpreted semantic feature and the second interpreted semantic feature.
[0044] After obtaining the first interpreted semantic feature and the second interpreted semantic feature, the processing device may calculate the similarity between the first interpreted semantic feature and the second interpreted semantic feature, and then determine the first correction coefficient based on the similarity.
[0045] In some embodiments, the processing device can determine the relationship between the similarity between the first interpreted semantic feature and the second interpreted semantic feature and the first similarity threshold to obtain a first judgment result. If the first judgment result represents that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is less than the first similarity threshold, then it is determined that the first correction coefficient is less than 1; if the first judgment result represents that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is equal to the first similarity threshold, then it is determined that the first correction coefficient is equal to 1; if the first judgment result represents that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is greater than the first similarity threshold, then it is determined that the first correction coefficient is greater than 1; wherein the first correction coefficient is positively correlated with the similarity between the first interpreted semantic feature and the second interpreted semantic feature, that is, the greater the similarity between the first interpreted semantic feature and the second interpreted semantic feature, the greater the first correction coefficient. Specifically, the positive correlation can be determined by a linear fitting function, and the corresponding first correction coefficient is subsequently determined based on the fitted function.
[0046] In some examples, taking the first similarity threshold as 0.7 as an example, the positive correlation relationship can be referred to in the following formula:
[0047] in, To express the first correction coefficient, It should be noted that the above positive correlation is only an exemplary introduction.
[0048] In an embodiment of the present invention, the processing device determines whether there is a logical association or continuity between the first technical semantic feature and the second technical semantic feature based on the similarity between the first interpretation semantic feature and the second interpretation semantic feature, and then feeds back to the first correction coefficient to correct the value of the combination of the first patent and the second patent.
[0049] S106. The processing device determines the similarity between the first patent and the second patent according to the first specification text and the second specification text, and determines a second correction coefficient according to the similarity between the first patent and the second patent.
[0050] The processing device can also determine the similarity between the first patent and the second patent based on the first specification text and the second specification text. Since the specification text includes content that is not written into the claims for protection, it is determined based on the specification text whether the two patents are similar as a whole, and then the second correction coefficient is determined based on the similarity between the first patent and the second patent.
[0051] In some embodiments, the processing device may determine the relationship between the similarity between the first patent and the second patent and a second similarity threshold to obtain a second determination result.
[0052] If the second judgment result indicates that the similarity between the first patent and the second patent is greater than or equal to the second similarity threshold, the processing device compares the first technical semantic feature and the second technical semantic feature to obtain the first distinguishing semantic feature in the first technical semantic feature and the second distinguishing semantic feature in the second technical semantic feature respectively, and inputs the first technical semantic feature, the first distinguishing semantic feature, the second technical semantic feature, the second distinguishing semantic feature and the technical field feature to which the first patent and the second patent belong to a pre-trained prediction model to obtain a probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship in the technical field corresponding to the technical field feature; and determines a second correction coefficient based on the probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship, wherein the second correction coefficient is positively correlated with the probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship, and the second correction coefficient is greater than or equal to 1.
[0053] The above prediction model can be obtained by training based on sample data, and the sample data includes the first sample technical semantic feature of the first sample patent, the first sample distinguishing semantic feature, the second sample technical semantic feature of the second sample patent, the second sample distinguishing semantic feature, the sample technical field features to which the first sample patent and the second sample patent belong, and the sample probability value. The above prediction model can be a neural network model, and the neural network used can be a feedforward neural network (MLP, CNN, RBF network, etc.), a feedback neural network (such as RNN, LSTM, GRU, etc.), and other neural networks (such as GAN, Transformer, etc.).
[0054] If the second judgment result indicates that the similarity between the first patent and the second patent is less than a second similarity threshold, the processing device determines that the second correction coefficient is less than 1, and the second correction coefficient is positively correlated with the similarity between the first patent and the second patent.
[0055] In an embodiment of the present invention, the first distinguishing semantic feature refers to the feature of the text of the first authority text that is different from the second authority text, and the second distinguishing semantic feature refers to the feature of the text of the second authority text that is different from the first authority text. When the above-mentioned prediction model is used to predict the probability value of the existence of a complementary relationship between the first distinguishing semantic feature and the second distinguishing semantic feature in the same field. Among them, the complementary relationship means that the first distinguishing semantic feature and the second distinguishing semantic feature belong to a large feature set. For example, the text corresponding to the first distinguishing semantic feature is "The mechanical bearing connected between the keycap of the keyboard and the keyboard body is a blue shaft", and the text corresponding to the second distinguishing semantic feature is "The mechanical bearing connected between the keycap of the keyboard and the keyboard body is a red shaft".
[0056] In some examples, taking the second similarity threshold as 0.8 as an example, the positive correlation relationship can be referred to in the following formula:
[0057] in, To express the second correction factor, represents the similarity between the first patent and the second patent, The probability value indicating that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship, It should be noted that the above positive correlation is only an exemplary introduction.
[0058] In the embodiment of the present invention, the processing device feeds back to the second correction coefficient based on the complementary relationship between the patents, so as to correct the value of the combination of the first patent and the second patent.
[0059] S107: The processing device performs weighted summation on the first correction coefficient and the second correction coefficient to obtain a combined correction coefficient.
[0060] Exemplarily, the first correction coefficient corresponds to the first weight, and the second correction coefficient corresponds to the second weight. The first weight and the second weight may be user-defined weights or weights configured based on experience. The first weight and the second weight are not zero.
[0061] In some examples, the processing device may display multiple weight options to the user through a human-computer exchange interface for the user to select, and then determine the first weight of the first correction coefficient and the second weight of the second correction coefficient based on the target option selected by the user from the multiple weight options; add the product of the first correction coefficient and the first weight to the product of the second correction coefficient and the second weight to obtain a combined correction coefficient.
[0062] Among them, different weight options in the multiple weight options correspond to different values of the first weight and the second weight. The multiple weight options may include default options, associated options, supplementary options, etc. For example, if the target option is the default option, the first weight corresponding to the target option is 0.5 and the second weight is 0.5; for another example, if the target option is an associated option, the first weight corresponding to the target option is 0.7 and the second weight is 0.3; if the target option is a supplementary option, the first weight corresponding to the target option is 0.3 and the second weight is 0.7.
[0063] The processing device multiplies the first correction coefficient by the first weight, multiplies the second correction coefficient by the second weight, and then sums the multiplication results to obtain a combined correction coefficient.
[0064] S108. The processing device corrects the sum of the first value and the second value according to the combination correction coefficient to obtain the combined value of the first patent and the second patent.
[0065] After the processing device obtains the combined correction coefficient, it can use the combined correction coefficient to correct the sum of the first value and the second value to obtain the combined value of the first patent and the second patent.
[0066] Furthermore, the processing device may display the combined value of the first patent and the second patent. For example, the processing device displays the combined value of the first patent and the second patent through a human-computer interaction interface.
[0067] Based on the above description, the processing device can first obtain the first patent and the second patent to be analyzed, and then perform semantic recognition on the first claim text of the first patent and the second claim text of the second patent respectively to obtain the first technical semantic feature and the second technical semantic feature, and then obtain the first interpretation semantic feature related to the first technical semantic feature from the first specification text, and obtain the second interpretation semantic feature related to the second technical semantic feature from the second specification text; then determine the first correction coefficient based on the similarity between the first interpretation semantic feature and the second interpretation semantic feature; at the same time, the processing device determines the similarity between the first patent and the second patent according to the first specification text and the second specification text, and determines the second correction coefficient according to the similarity between the first patent and the second patent, and then weighted sums the first correction coefficient and the second correction coefficient to obtain the combined correction coefficient, and finally uses the combined correction coefficient to correct the sum of the first value and the second value to obtain the combined value of the first patent and the second patent. This method adjusts the combined value of patents from two levels, rather than simply superimposing the values of multiple patents, so that the value of the patent combination can be given more accurately.
[0068] In some embodiments, the processing device may further obtain first post-processing information of the first patent and second post-processing information of the second patent, respectively. The first post-processing information may be relevant information of the post-processing procedure for the first patent, such as the number of times the first patent has been invalidated, the number of claims of the first patent declared invalid, the number of times the first patent has been protected, etc. The second post-processing information may be relevant information of the post-processing procedure for the second patent, such as the number of times the second patent has been invalidated, the number of claims of the second patent declared invalid, the number of times the second patent has been protected, etc.
[0069] The processing device can also adjust the combined value of the first patent and the second patent based on the first post-processing information and the second post-processing information. Exemplarily, the greater the number of times a patent is invalidated, the smaller the number of claims of the patent pair that are declared invalid, and the more times the patent is protected, the higher the value of the patent. Therefore, the mapping relationship between the above-mentioned influencing factors and the post-processing correction coefficients can be preset. In some examples, the mapping relationship between the post-processing correction coefficients and the above-mentioned influencing factors can be represented by the following formula:
[0070]
[0071] in, is the intermediate coefficient, Indicates the number of times the first patent was invalidated, Indicates the number of claims of the first patent that have been declared invalid, Indicates the number of times the first patent is protected. Indicates the number of times the second patent was invalidated, Indicates the number of claims of the second patent that have been declared invalid, Indicates the number of times the second patent is protected. Represents the post-processing correction factor.
[0072] After the processing equipment obtains the post-processing correction coefficient, it can multiply the value of the patent portfolio by the post-processing correction coefficient to obtain the revised patent portfolio value.
[0073] In the embodiment of the present invention, the influence of the post-processing procedure of the patent on the patent value is taken into consideration, thereby further improving the accuracy of the patent portfolio value.
[0074] In some embodiments, the processing device may also obtain the number of first drawings of the first patent and the number of second drawings of the second patent respectively. The processing device may determine whether there are application scenario drawings in the drawings of the first patent, and determine whether there are application scenario drawings in the drawings of the second patent. If both exist, further determine the correlation between the first application scenario drawings in the first patent and the second application scenario drawings in the second patent, determine the drawing adjustment coefficient based on the correlation, and then use the drawing adjustment coefficient to further adjust the value of the above patent combination, thereby improving the accuracy of the analysis.
[0075] Taking the processing device processing the first patent as an example, the processing device can identify the number of characters included in the drawings in the first patent, and filter out the first target drawings whose number of characters is less than the preset character number threshold. Similarly, the processing device can also determine the second target drawings from the second patent.
[0076] Different situations are introduced below. If the first target illustration and the second target illustration each include one illustration, the similarity between the first target illustration and the second target illustration is directly calculated, and the similarity is used as the correlation between the first application scenario illustration and the second application scenario illustration.
[0077] As shown in the figure, the first target illustration includes multiple illustrations and the second target illustration includes one illustration, then the similarities between the first target illustration and the second target illustration are calculated respectively, and the maximum value of the calculated similarities is used as the correlation between the first application scenario illustration and the second application scenario illustration. The case where the first target illustration includes one illustration and the second target illustration includes multiple illustrations is similar.
[0078] As shown in the figure, the first target illustration and the second target illustration both include multiple illustrations, then the similarities of the first target illustration and the second target illustration are calculated respectively, and the maximum value of the calculated similarities is used as the correlation between the first application scenario illustration and the second application scenario illustration. Further, the similarity is used as the illustration adjustment coefficient.
[0079] In some embodiments, the processing device can also determine whether the first patent and the second patent include patents of the same family. If both the first patent and the second patent include patents of the same family, the first distribution region of the first patent's first patent and the second distribution region of the second patent's second patent's second patent are further obtained. Different regions have different protection conditions, so different regions can be quantitatively scored. In some examples, the regional scoring coefficient = regional basic score + regional adjustment score, where the regional basic score is determined by a pre-assigned value, and different regions have different corresponding regional basic scores. The regional adjustment score is determined by the number of litigation cases in the technical field corresponding to the patent in the region.
[0080] For example, the regional scoring coefficient is greater than 1 and less than 2, the regional basic score is greater than 0.5 and less than 1, and the regional adjustment score is greater than 0.5 and less than 1. For example, the regional basic score corresponding to region 1 is 0.8, and the regional basic score corresponding to region 2 is 0.9. The regional basic scores corresponding to different regions can be pre-stored in a table, and the regional basic scores corresponding to the regions can be determined by looking up the table; if the number of litigation cases is 0-100, the regional adjustment score is 0.5, the number of litigation cases is 100-1000, the regional adjustment score is 0.6, and the number of litigation cases is 1000-5000, and the regional adjustment score is 0.7. Different intervals of the number of litigation cases correspond to different regional adjustment scores.
[0081] After the processing equipment determines the regional scoring coefficient, it uses the regional scoring coefficient to further adjust the value of the above patent portfolio, thereby improving the accuracy of the analysis.
[0082] It should be noted that the processing equipment simultaneously uses the above-mentioned post-processing correction coefficient, illustration adjustment coefficient and regional scoring coefficient to further correct the combined value of the patent, and can also use the above-mentioned coefficients alone to further correct the combined value of the patent.
[0083] Combination of the above Figure 1 The analysis method of the patent portfolio value provided by the embodiment of the present invention is introduced in detail. The device and equipment provided by the embodiment of the present invention will be introduced in conjunction with the accompanying drawings.
[0084] like Figure 2 As shown, this figure is a schematic diagram of a patent portfolio value analysis device provided by an embodiment of the present invention, and the device includes: The acquisition module 201 is used to acquire patent identifiers of multiple patents to be analyzed, wherein the patent identifiers of the multiple patents at least include a first identifier of a first patent and a second identifier of a second patent; according to the first identifier, a first value and a first patent text of the first patent are acquired, and according to the second identifier, a second value and a second patent text of the second patent are acquired; the first patent text includes a first claim text and a first specification text, and the second patent text includes a second claim text and a second specification text; The semantic recognition module 202 is used to perform semantic recognition on the first claim text to obtain a first technical semantic feature, and perform semantic recognition on the second claim text to obtain a second technical semantic feature; obtain a first interpretation semantic feature related to the first technical semantic feature from the first specification text, and obtain a second interpretation semantic feature related to the second technical semantic feature from the second specification text; A first determination module 203, configured to determine a first correction coefficient according to a similarity between the first interpretation semantic feature and the second interpretation semantic feature; A second determination module 204 is used to determine the similarity between the first patent and the second patent according to the first specification text and the second specification text, and determine a second correction coefficient according to the similarity between the first patent and the second patent; The analysis module 205 is used to perform weighted summation of the first correction coefficient and the second correction coefficient to obtain a combined correction coefficient; and to correct the sum of the first value and the second value according to the combined correction coefficient to obtain a combined value of the first patent and the second patent.
[0085] Optionally, the first determination module 203 is specifically used to determine the relationship between the similarity between the first interpretation semantic feature and the second interpretation semantic feature and a first similarity threshold to obtain a first judgment result; if the first judgment result represents that the similarity between the first interpretation semantic feature and the second interpretation semantic feature is less than the first similarity threshold, then the first correction coefficient is determined to be less than 1; if the first judgment result represents that the similarity between the first interpretation semantic feature and the second interpretation semantic feature is equal to the first similarity threshold, then the first correction coefficient is determined to be equal to 1; if the first judgment result represents that the similarity between the first interpretation semantic feature and the second interpretation semantic feature is greater than the first similarity threshold, then the first correction coefficient is determined to be greater than 1; wherein, the first correction coefficient is positively correlated with the similarity between the first interpretation semantic feature and the second interpretation semantic feature.
[0086] Optionally, the second determination module 204 is specifically used to determine the relationship between the similarity between the first patent and the second patent and a second similarity threshold value to obtain a second determination result; if the second determination result indicates that the similarity between the first patent and the second patent is greater than or equal to the second similarity threshold value, then the first technical semantic feature and the second technical semantic feature are compared to obtain a first distinguishing semantic feature in the first technical semantic feature and a second distinguishing semantic feature in the second technical semantic feature, respectively, and the first technical semantic feature, the first distinguishing semantic feature, the second technical semantic feature, the second distinguishing semantic feature, and the technical field features to which the first patent and the second patent belong are input into a pre-determined The trained prediction model obtains the probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship in the technical field corresponding to the technical field feature; according to the probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship, a second correction coefficient is determined, wherein the second correction coefficient is positively correlated with the probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship, and the second correction coefficient is greater than or equal to 1; if the second judgment result represents that the similarity between the first patent and the second patent is less than the second similarity threshold, it is determined that the second correction coefficient is less than 1, and the second correction coefficient is positively correlated with the similarity between the first patent and the second patent.
[0087] Optionally, the semantic recognition module 202 is specifically used to perform semantic recognition on the first specification text to obtain a first specification semantic feature, determine a first specification sub-semantic feature based on the first specification semantic feature and the first technical semantic feature, determine the first specification sub-text corresponding to the first specification sub-semantic feature, perform semantic recognition on the first specification sub-text and the context text of the first specification sub-text to obtain the first interpretation semantic feature; perform semantic recognition on the second specification text to obtain a second specification semantic feature, determine a second specification sub-semantic feature based on the second specification semantic feature and the second technical semantic feature, determine the second specification sub-text corresponding to the second specification sub-semantic feature, perform semantic recognition on the second specification sub-text and the context text of the second specification sub-text to obtain the second interpretation semantic feature.
[0088] Optionally, the analysis module is specifically used to select a target option from multiple weight options, determine a first weight of the first correction coefficient and a second weight of the second correction coefficient; and add a product of the first correction coefficient and the first weight to a product of the second correction coefficient and the second weight to obtain a combined correction coefficient.
[0089] The patent portfolio value analysis device according to the embodiment of the present invention may correspond to the method described in the embodiment of the present invention, and the above-mentioned other operations and / or functions of each module / unit of the patent portfolio value analysis device are respectively to realize Figure 1 For the sake of brevity, the corresponding processes of each method in the illustrated embodiment are not described in detail here.
[0090] The embodiment of the present invention also provides a computing device. Figure 3 As shown, this figure is a schematic diagram of a computing device provided by an embodiment of the present invention, and the computing device 300 includes a bus 301, a processor 302, a communication interface 303 and a memory 304. The processor 302, the memory 304 and the communication interface 303 communicate with each other through the bus 301.
[0091] The bus 301 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.
[0092] The processor 302 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).
[0093] The communication interface 303 is used for communicating with the outside. For example, the communication interface 303 can be used for communicating with the terminal 102. The communication interface 303 is used to send the allocation value to the terminal 102 so that the terminal 102 presents the allocation value, thereby displaying the allocation result of the virtual object.
[0094] The memory 304 may include a volatile memory, such as a random access memory (RAM). The memory 304 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).
[0095] The memory 304 stores executable codes, and the processor 302 executes the executable codes to perform the aforementioned patent portfolio value analysis method.
[0096] Specifically, in implementing Figure 2 In the case of the embodiment shown, and Figure 2 When each module or unit of the patent portfolio value analysis device described in the embodiment is implemented by software, the execution Figure 2 The software or program code required for the functions of each module / unit in the invention may be partially or completely stored in the memory 304. The processor 302 executes the program code corresponding to each unit stored in the memory 304 to perform the aforementioned analysis method of the patent portfolio value.
[0097] The embodiment of the present invention also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that can be stored by a computing device or a data storage device such as a data center that contains one or more available media. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid-state hard disk). The computer-readable storage medium includes instructions that instruct the computing device to execute the above-mentioned patent portfolio value analysis method.
[0098] The embodiment of the present invention further provides a computer program product, which includes one or more computer instructions. When the computer instructions are loaded and executed on a computing device, the process or function described in the embodiment of the present invention is fully or partially generated.
[0099] The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer or data center to another website, computer or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means.
[0100] When the computer program product is executed by a computer, the computer executes any of the aforementioned methods for analyzing the value of a patent portfolio. The computer program product may be a software installation package, and when any of the aforementioned methods for analyzing the value of a patent portfolio is needed, the computer program product may be downloaded and executed on a computer.
[0101] The descriptions of the processes or structures corresponding to the above-mentioned figures have different emphases. For parts that are not described in detail in a certain process or structure, please refer to the relevant descriptions of other processes or structures.
[0102] The above description is only a specific implementation of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions within the technical scope disclosed by the present invention should be included in the protection scope of the present invention.
Claims
1. A method for analyzing the value of a patent portfolio, characterized in that: The method comprises: Obtaining patent identifiers of a plurality of patents to be analyzed, wherein the patent identifiers of the plurality of patents at least include a first identifier of a first patent and a second identifier of a second patent; According to the first identifier, a first value and a first patent text of the first patent are obtained, and according to the second identifier, a second value and a second patent text of the second patent are obtained; the first patent text includes a first claim text and a first specification text, and the second patent text includes a second claim text and a second specification text; Performing semantic recognition on the first claim text to obtain a first technical semantic feature, and performing semantic recognition on the second claim text to obtain a second technical semantic feature; Acquire a first interpretation semantic feature related to the first technical semantic feature from the first specification text, and acquire a second interpretation semantic feature related to the second technical semantic feature from the second specification text; determining a first correction coefficient according to a similarity between the first interpreted semantic feature and the second interpreted semantic feature; Determine the similarity between the first patent and the second patent according to the first specification text and the second specification text, and determine a second correction coefficient according to the similarity between the first patent and the second patent; Performing a weighted summation on the first correction coefficient and the second correction coefficient to obtain a combined correction coefficient; According to the combination correction coefficient, the sum of the first value and the second value is corrected to obtain the combined value of the first patent and the second patent.
2. The method according to claim 1, characterized in that The determining of a first correction coefficient according to the similarity between the first interpreted semantic feature and the second interpreted semantic feature includes: Determine the relationship between the similarity between the first interpretation semantic feature and the second interpretation semantic feature and a first similarity threshold, and obtain a first determination result; If the first judgment result indicates that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is less than a first similarity threshold, determining that the first correction coefficient is less than 1; If the first judgment result indicates that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is equal to a first similarity threshold, determining that a first correction coefficient is equal to 1; If the first judgment result indicates that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is greater than a first similarity threshold, determining that a first correction coefficient is greater than 1; The first correction coefficient is positively correlated with the similarity between the first interpretation semantic feature and the second interpretation semantic feature.
3. The method according to claim 1, characterized in that Determining a second correction coefficient according to the similarity between the first patent and the second patent includes: Determine the relationship between the similarity between the first patent and the second patent and a second similarity threshold, and obtain a second determination result; If the second judgment result indicates that the similarity between the first patent and the second patent is greater than or equal to the second similarity threshold, then the first technical semantic feature and the second technical semantic feature are compared to obtain a first distinguishing semantic feature in the first technical semantic feature and a second distinguishing semantic feature in the second technical semantic feature, respectively, and the first technical semantic feature, the first distinguishing semantic feature, the second technical semantic feature, the second distinguishing semantic feature, and the technical field feature to which the first patent and the second patent belong are input into a pre-trained prediction model to obtain a probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship in the technical field corresponding to the technical field feature; a second correction coefficient is determined according to the probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship, wherein the second correction coefficient is positively correlated with the probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship, and the second correction coefficient is greater than or equal to 1; If the second judgment result indicates that the similarity between the first patent and the second patent is less than the second similarity threshold, it is determined that the second correction coefficient is less than 1, and the second correction coefficient is positively correlated with the similarity between the first patent and the second patent.
4. The method according to claim 1, characterized in that: The step of acquiring a first interpretation semantic feature related to the first technical semantic feature from the first specification text, and acquiring a second interpretation semantic feature related to the second technical semantic feature from the second specification text includes: Performing semantic recognition on the first specification text to obtain a first specification semantic feature, determining a first specification sub-semantic feature based on the first specification semantic feature and the first technical semantic feature, determining a first specification sub-text corresponding to the first specification sub-semantic feature, and performing semantic recognition on the first specification sub-text and a context text of the first specification sub-text to obtain the first interpretation semantic feature; Perform semantic recognition on the second specification text to obtain second specification semantic features, determine second specification sub-semantic features based on the second specification semantic features and the second technical semantic features, determine the second specification sub-text corresponding to the second specification sub-semantic features, perform semantic recognition on the second specification sub-text and the context text of the second specification sub-text to obtain the second interpretation semantic features.
5. The method according to any one of claims 1 to 4, characterized in that: The step of performing weighted summation on the first correction coefficient and the second correction coefficient to obtain a combined correction coefficient includes: Select a target option from a plurality of weight options, and determine a first weight of the first correction factor and a second weight of the second correction factor; A product of the first correction coefficient and the first weight is added to a product of the second correction coefficient and the second weight to obtain a combined correction coefficient.
6. A device for analyzing the value of a patent portfolio, characterized in that: The device comprises: an acquisition module, configured to acquire patent identifiers of a plurality of patents to be analyzed, wherein the patent identifiers of the plurality of patents at least include a first identifier of a first patent and a second identifier of a second patent; acquire a first value and a first patent text of the first patent according to the first identifier, and acquire a second value and a second patent text of the second patent according to the second identifier; the first patent text includes a first claim text and a first specification text, and the second patent text includes a second claim text and a second specification text; a semantic recognition module, configured to perform semantic recognition on the first claim text to obtain a first technical semantic feature, and perform semantic recognition on the second claim text to obtain a second technical semantic feature; obtain a first interpretation semantic feature related to the first technical semantic feature from the first specification text, and obtain a second interpretation semantic feature related to the second technical semantic feature from the second specification text; A first determination module, configured to determine a first correction coefficient according to a similarity between the first interpretation semantic feature and the second interpretation semantic feature; a second determination module, configured to determine the similarity between the first patent and the second patent according to the first specification text and the second specification text, and determine a second correction coefficient according to the similarity between the first patent and the second patent; The analysis module is used to perform weighted summation of the first correction coefficient and the second correction coefficient to obtain a combined correction coefficient; and to correct the sum of the first value and the second value according to the combined correction coefficient to obtain a combined value of the first patent and the second patent.
7. The device according to claim 6, characterized in that The first determination module is specifically used to determine the relationship between the similarity between the first interpretation semantic feature and the second interpretation semantic feature and a first similarity threshold to obtain a first determination result; if the first determination result indicates that the similarity between the first interpretation semantic feature and the second interpretation semantic feature is less than the first similarity threshold, then determine that the first correction coefficient is less than 1; If the first judgment result indicates that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is equal to a first similarity threshold, determining that a first correction coefficient is equal to 1; If the first judgment result represents that the similarity between the first interpreted semantic feature and the second interpreted semantic feature is greater than a first similarity threshold, then a first correction coefficient is determined to be greater than 1; wherein the first correction coefficient is positively correlated with the similarity between the first interpreted semantic feature and the second interpreted semantic feature.
8. The device according to claim 6, characterized in that The second determination module is specifically used to determine the magnitude relationship between the similarity between the first patent and the second patent and a second similarity threshold value to obtain a second determination result; if the second determination result indicates that the similarity between the first patent and the second patent is greater than or equal to the second similarity threshold value, then the first technical semantic feature and the second technical semantic feature are compared to obtain a first distinguishing semantic feature in the first technical semantic feature and a second distinguishing semantic feature in the second technical semantic feature respectively, and the first technical semantic feature, the first distinguishing semantic feature, the second technical semantic feature, the second distinguishing semantic feature and the technical field feature to which the first patent and the second patent belong are input into a pre-trained prediction model to obtain a probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship in the technical field corresponding to the technical field feature; according to the probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship, a second correction coefficient is determined, wherein the second correction coefficient is positively correlated with the probability value that the first distinguishing semantic feature and the second distinguishing semantic feature have a complementary relationship, and the second correction coefficient is greater than or equal to 1; if the second determination result indicates that the similarity between the first patent and the second patent is less than the second similarity threshold value, the second correction coefficient is determined to be less than 1, and the second correction coefficient is positively correlated with the similarity between the first patent and the second patent.
9. A computing device, characterized in that including memory and processor; One or more computer programs are stored in the memory, and the one or more computer programs include instructions; when the instructions are executed by the processor, the computing device executes the method as claimed in any one of claims 1 to 5.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method according to any one of claims 1 to 5.
