Intelligent docking system for patent supply and demand relationship between innovation subjects

Abstract

The invention relates to an intelligent docking system for patent supply and demand relationships between innovation subjects, belongs to the technical field of intelligent recommendation, and solves the problems of poor flexibility, low processing efficiency, low accuracy and the like of an existing docking mode for the patent supply and demand relationships between the innovation subjects. The system comprises a to-be-converted patent determination module for summarizing technical problems of an innovative subject patent of a supplier / demander; taking the high-value patents under each technical problem category as patents to be converted; the multi-dimensional technical summary module is used for acquiring technical summary of each to-be-converted patent in different dimensions to obtain technical summary results of corresponding dimensions; the multi-dimensional candidate patent retrieval module is used for performing feature extraction and intelligent retrieval on the technical summary results of each to-be-converted patent in different dimensions to obtain a candidate patent set in the corresponding dimension; and the supply-demand relationship determination module is used for selecting the co-occurrence patent with the highest recommendation degree from the candidate patent set, and taking the patentee of the co-occurrence patent as an innovation main body of a demander / supplier to establish a docking relationship.

Description

Technical Field

[0001] The present invention relates to the field of intelligent recommendation technology, and in particular to an intelligent docking system for patent supply and demand relationships between innovative entities. Background Art

[0002] Patent transformation and application refers to the process of converting patent rights into actual productivity and economic benefits. Patent transformation and application are of great significance in promoting the practical application of technological innovation, optimizing the overall resource allocation of society, and promoting industrial upgrading. The primary task of patent transformation and application is to establish a communication bridge between the supplier and demander of patent technology through certain means, so as to effectively promote the transformation and application of patents, transform the advantages of the patent system into a powerful driving force for innovation and development, and help achieve high-level scientific and technological self-reliance. However, due to the information barriers between patent holders and potential users, a large number of valuable technological achievements are difficult to be discovered and effectively applied.

[0003] In the prior art, the relationship between the supplier and the demander of patent technology can be established in the following ways:

[0004] (1) Starting from the supplier's patents to be transformed, find the demand objects of the patents to be transformed;

[0005] (2) The demander conducts scientific and technological achievement transactions with the supplier through display and search on the e-commerce platform;

[0006] (3) Use data labels to determine the matching relationship between supply and demand. First, the technology needs of the demand side can be directly labeled by domain and keyword, and partners can be found based on this. Second, a preliminary matching library can be built based on manually edited matching training data, and the supply and demand matching library can be gradually improved by continuously optimizing the matching algorithm. Third, knowledge graph technology can be used to match demand information and supply information by building a supply and demand information graph.

[0007] In the above existing technologies, supply and demand information is generally provided by suppliers and demanders, and it is impossible to actively discover demand and automatically match it through big data. Therefore, in practical applications, it is limited by the source of data and has limited flexibility and processing efficiency. In addition, the accuracy of the matching method of existing supply and demand information is limited, which affects the accuracy of the intelligent docking results of patent supply and demand relationships between innovation entities. Summary of the invention

[0008] In view of the above analysis, the embodiments of the present invention aim to provide an intelligent docking system for patent supply and demand relationships between innovative entities, so as to solve the problems of poor flexibility, low processing efficiency and low accuracy in the docking method of patent supply and demand relationships between innovative entities in the existing patent transformation and application process.

[0009] The present invention discloses an intelligent docking system for patent supply and demand relationship between innovative entities, the system comprising:

[0010] The module for determining patents to be converted uses a large model to summarize the technical issues of patents disclosed by the innovative entities of one of the supply and demand sides, and classify the technical issues; the high-value patents under each technical issue category are respectively regarded as patents to be converted;

[0011] The multi-dimensional technical summary module uses a large model to obtain the technical summary of each patent to be converted in different dimensions, and obtains the technical summary results of the corresponding patent to be converted in the corresponding dimension after reviewing the technical summary;

[0012] The multi-dimensional candidate patent search module extracts features from the technical summary results of each patent to be converted in different dimensions, and performs intelligent searches based on the extracted technical features to obtain a candidate patent set of the corresponding patent to be converted in the corresponding dimension;

[0013] The supply and demand relationship determination module selects the co-occurring patent with the highest recommendation degree from the candidate patent sets of different dimensions for each patent to be converted; the patent owner of the co-occurring patent is used as the innovation subject of the other party between the supplier and the demander, and a corresponding relationship between the patent to be converted and the co-occurring patent under each technical problem category is established, forming a docking relationship between the innovation subjects of the supplier and the demander.

[0014] On the basis of the above scheme, the present invention also makes the following improvements:

[0015] Furthermore, the module for determining patents to be transformed includes: a technical problem summary unit for the innovation subject, a technical problem classification unit, and a high-value patent selection unit; among which,

[0016] The innovation subject technical problem summary unit receives patents disclosed by the innovation subject of one of the supply side and the demand side, and summarizes the technical problems of each patent using the big model;

[0017] The technical problem classification unit clusters technical problems based on clustering distribution to achieve classification of technical problems;

[0018] The high-value patent selection unit conducts value assessment on each patent under each technical problem category and selects high-value patents as patents to be converted.

[0019] Furthermore, the multi-dimensional technical summary module includes a preliminary technical summary unit and an output review unit for each dimension; wherein,

[0020] The preliminary technical summary unit of each dimension is used to summarize the technical content of the corresponding dimension in the patent to be converted based on the macro model of the corresponding dimension, and obtain the preliminary technical summary results of the patent to be converted in the corresponding dimension;

[0021] The output review unit of each dimension is used to output and review the preliminary technical summary results of the patent to be transformed in the corresponding dimension, and obtain the technical summary results of the patent to be transformed in the corresponding dimension.

[0022] Furthermore, in each dimension of the large model, various forms of interactive input templates of the corresponding dimension are provided;

[0023] When making a technical summary of the technical contents of the corresponding dimensions in the patent to be transformed, the technical contents of the corresponding dimensions in the patent to be transformed are placed in an interactive input template of the corresponding dimension as the input information of the big model of the corresponding dimension. After the big model of the corresponding dimension processes the input information, a preliminary technical summary result of the corresponding dimension is obtained.

[0024] Furthermore, the output review unit of each dimension is used to review whether the technical summary result of the corresponding dimension is technical related information of the patent to be converted.

[0025] If yes, the output is reviewed and approved, and the preliminary technical summary results of the corresponding dimensions are used as the technical summary results of the patent to be converted in the corresponding dimensions;

[0026] Otherwise, the output review fails, and the preliminary technical summary unit of the corresponding dimension is returned, and the interactive input template of the corresponding dimension is replaced. The technical content of the corresponding dimension in the patent to be converted is placed in the replaced interactive input template of the corresponding dimension, and again used as the input information of the large model of the corresponding dimension. The large model of the corresponding dimension processes the input information to obtain the preliminary technical summary result of the corresponding dimension, and then the output review unit of the corresponding dimension performs output review until the output review passes.

[0027] Furthermore, each dimension of the output audit unit has a built-in output audit controller and multiple output auditors of the corresponding dimension; wherein,

[0028] The output audit controller is used to control whether each output auditor is effective, and to control the logical relationship between multiple effective output auditors;

[0029] Each output reviewer, from its own review perspective, reviews whether the preliminary technical summary results of the corresponding dimension are technically relevant information for the patent to be converted.

[0030] Further, the output reviewer is one of a technical classifier, a tendency classifier, an output approval keyword whitelist, and an output approval keyword blacklist;

[0031] The output audit controller of each dimension adjusts the audit strength of the output audit unit of the corresponding dimension by controlling the effectiveness of different output auditors and controlling the logical relationship between the effective output auditors.

[0032] Furthermore, the multi-dimensional candidate patent search module includes: a technical feature extraction unit for each dimension, a search expression generation unit, and an intelligent search unit; wherein,

[0033] The technical feature extraction unit of each dimension extracts features from the technical summary results of the patent to be converted in the corresponding dimension, and obtains the technical features of the patent to be converted in the corresponding dimension;

[0034] The search expression generation unit comprehensively analyzes the technical features of the patent to be converted in different dimensions and obtains multiple intelligent search expressions of the patent to be converted in different dimensions;

[0035] The intelligent retrieval unit uses a variety of intelligent retrieval expressions of the patent to be converted in different dimensions to perform patent retrieval in the intelligent retrieval system to obtain a candidate patent set of the patent to be converted in different dimensions.

[0036] Furthermore, in the supply and demand relationship determination module, for each patent to be converted, the co-occurrence degree, average similarity and operation degree of the co-occurring patents between the candidate patent sets of different dimensions of the patent to be converted are obtained to determine the recommendation degree of each co-occurring patent; and the co-occurring patent with the highest recommendation degree is selected.

[0037] Furthermore, in the supply and demand relationship determination module, for each patent to be converted, execute:

[0038] Arrange the similarity of each candidate patent in the candidate patent set of different dimensions of the patent to be converted in descending order and assign corresponding similarity scores;

[0039] Select candidate patents that appear in at least two candidate patent sets of different dimensions as co-occurring patents, add up the similarity scores of each co-occurring patent in each candidate patent set, and normalize them to obtain the co-occurrence degree of each co-occurring patent;

[0040] The similarity scores of each co-occurring patent in each candidate patent set are averaged to obtain the average similarity of each co-occurring patent;

[0041] According to the number of different types of operating behaviors implemented by the patentee in each co-occurring patent, the operating degree of the corresponding co-occurring patent is obtained.

[0042] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0043] The patent supply and demand relationship intelligent docking system between innovative entities provided by the present invention can summarize and classify technical issues based on the relevant patents of either the supplier or the demander, thereby obtaining the patents to be converted corresponding to the corresponding technical issues, and by obtaining the co-occurring patents with the highest recommendation corresponding to the patents to be converted, the patent holders of the co-occurring patents are used as the innovation entities of the other party of the supplier and the demander, thereby establishing a corresponding relationship between the patents to be converted and the co-occurring patents under each technical problem category, and forming a docking relationship between the innovation entities of the supplier and the demander. The above method fundamentally solves the problem of patent conversion paths between supply-side innovation entities (such as universities or scientific research institutions) and demand-side innovation entities (such as enterprises), can effectively improve the patent conversion rate of universities or scientific research institutions, and guide universities or scientific research institutions to purposefully conduct research and development according to the needs of enterprises, and has the advantages of strong flexibility, high processing efficiency, and high accuracy.

[0044] In particular, the technical problems that the supplier's innovative entities need to solve in specific fields can be analyzed based on their existing patents, and based on this, the demand-side innovative entities can be found for their key patents. Similarly, the technical difficulties faced by the demand-side innovative entities can be analyzed based on their existing patents, and the supplier's innovative entities with relevant technologies in this field can be found.

[0045] In the present invention, the above-mentioned technical solutions can also be combined with each other to achieve more preferred combination solutions. Other features and advantages of the present invention will be described in the subsequent description, and some advantages can become obvious from the description, or can be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained through the contents particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are only used for the purpose of illustrating specific embodiments and are not to be considered as limiting the present invention. In the entire drawings, the same reference symbols represent the same components;

[0047] Figure 1 A schematic diagram of the structure of the intelligent docking system for patent supply and demand relationships between innovative entities provided by an embodiment of the present invention;

[0048] Figure 2 A schematic diagram of the structure of another intelligent docking system for patent supply and demand relationships between innovative entities provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0050] A specific embodiment of the present invention discloses an intelligent docking system for patent supply and demand relationship between innovative entities, the structural diagram of which is shown in FIG. Figure 1 and Figure 2 shown.

[0051] (1) Module for determining patents to be converted

[0052] The module for determining patents to be converted uses a large model to summarize the technical issues of patents disclosed by the innovative entities of one of the supply and demand sides, and classify the technical issues; high-value patents under each technical issue category are identified as patents to be converted.

[0053] It should be noted that this system aims to realize the intelligent docking of patent supply and demand relationships between innovative entities. Therefore, in the specific implementation process, either the supplier or the demander can determine the patent to be transformed, and seek a matching supplier or demander based on the patent to be transformed. Exemplarily, the supplier can determine the patent to be transformed, and seek a matching demander of the patent to be transformed based on the patent to be transformed determined by the supplier; or the demander can determine the patent to be transformed, and seek a matching supplier of the patent to be transformed based on the patent to be transformed determined by the demander. Therefore, whether starting from the supplier or the demander, the intelligent docking of patent supply and demand relationships between innovative entities can be realized.

[0054] Preferably, in this embodiment, the patent to be converted determination module includes: an innovation subject technical problem summary unit, a technical problem classification unit, and a high-value patent selection unit. The functions of each unit are described in detail as follows.

[0055] 1) Summary unit of technical issues of innovation subjects

[0056] The innovation subject technical problem summary unit receives patents disclosed by the innovation subject of one of the supplier and the demander, and summarizes the technical problems of each patent using a large model.

[0057] In the specific implementation process, the big model, especially the deep learning model, can analyze and understand the text information in the patent document through natural language processing (NLP) technology. These models can identify and extract key technical terms and concepts in the patent description, so as to automatically summarize the technical problems that each patent attempts to solve. For example, by analyzing the background technology part of the patent, the big model can identify the limitations of the prior art and the innovative points of the invention, and then summarize the technical problems of the patent. In this embodiment, the big model is preferably implemented by a generative artificial intelligence interaction system. The more common generative artificial intelligence interaction systems currently include ChatGPT, Wenxinyiyan or chatGLM. Before using the big model to automatically summarize and output the technical problems of the actual patent, it is also necessary to train the parameters of the big model through a large amount of training corpus to fit the corresponding relationship between input and output. In the big model used in the technical problem summary unit of the innovation subject, the background technology of many patents can be used as input information, and the corresponding summarized technical problems can be used as output information to construct training expectations and train the big model so that it can automatically summarize the technical problems of each patent.

[0058] In addition, restrictive conditions can be preset in the technical problem summary unit of the innovation subject, including but not limited to IPC classification, strategic emerging industry classification, national economic classification, etc. When the preset restrictive conditions are enabled, the technical problem summary unit of the innovation subject receives patents disclosed by the innovation subject of one of the supply and demand sides that meet the preset restrictive conditions, and uses the big model to automatically summarize the technical problems of each patent. By setting restrictive conditions, it is possible to focus on the patents of a certain angle of the innovation subject and determine the docking relationship between the supply and demand sides at the corresponding angle.

[0059] 2) Technical Problem Classification Unit

[0060] The technical problem classification unit clusters technical problems based on clustering distribution to achieve classification of technical problems.

[0061] In the process of classifying technical problems, clustering and diffusion can be used to classify technical problems. This method belongs to unsupervised learning and aims to divide objects in a data set into multiple categories or "clusters" so that the similarity of data points in the same cluster is as high as possible, while the similarity between different clusters is as low as possible. For example, K-Means clustering can be used for clustering.

[0062] 3) High-value patent selection unit

[0063] The high-value patent selection unit conducts value assessment on each patent under each technical problem category and selects high-value patents as patents to be converted.

[0064] During the specific implementation process, the criteria for value assessment can be adaptively set according to the specific situation. For example, the criteria for value assessment can be quantifiable evaluation indicators such as legal status, number of citations, number of operations, etc., or it can be the weighted fusion result of several evaluation indicators. These evaluation indicators can help innovative entities identify patents with higher commercial potential and market value. For example, patents with a large number of citations usually mean that they have a greater technological influence, while patents with a large number of operations may indicate that they have higher value in practical applications. Sort each patent in order from high to low according to the value assessment results, and treat several patents with higher value assessment results as high-value patents, and treat them as patents to be converted.

[0065] (2) Multi-dimensional technical summary module

[0066] The multi-dimensional technical summary module uses a large model to obtain the technical summary of each patent to be converted in different dimensions, and obtains the technical summary results of the corresponding patent to be converted in the corresponding dimension after reviewing the technical summary.

[0067] Specifically, in this embodiment, the dimensions involved include technical problem dimension, technical means dimension and technical effect dimension. Therefore, in the multi-dimensional technical summary module, the technical summary of the patent to be converted is performed in the technical problem dimension, technical means dimension and technical effect dimension respectively, so as to obtain the technical summary results of the patent to be converted in the technical problem dimension, technical means dimension and technical effect dimension.

[0068] Preferably, in this embodiment, the multi-dimensional technology summary module includes a preliminary technology summary unit and an output review unit for each dimension. The functions of each unit are specifically introduced below.

[0069] 1) Preliminary technical summary unit for each dimension

[0070] The preliminary technical summary unit of each dimension is used to make a technical summary of the technical content of the corresponding dimension in the patent to be transformed based on the big model of the corresponding dimension, and obtain the preliminary technical summary results of the patent to be transformed in the corresponding dimension.

[0071] Preferably, in this embodiment, the technical contents of different dimensions in the patent to be converted are not the same. Specifically, the technical contents of the technical problem dimension include the background technology of the patent to be converted, the technical contents of the technical means dimension include the claims of the patent to be converted, and the technical contents of the technical effect dimension include the beneficial effects of the patent to be converted.

[0072] During the specific implementation process, large models of different dimensions may be implemented in the form of models of the same or different categories. In this embodiment, the large model is preferably implemented by a generative artificial intelligence interaction system. Currently, the more common generative artificial intelligence interaction systems include ChatGPT, Wenxinyiyan or chatGLM. Before using the large model to predict the output of the actual preliminary technical summary results, it is necessary to train the parameters of the large model through a large amount of training corpus to fit the correspondence between input and output. During the training process, large models of different dimensions need to use training expectations of corresponding dimensions for model training. For example, the large model of the technical problem dimension needs to use the background technology in multiple patent applications as training expectations for training; the large model of the technical means dimension needs to use the claims in multiple patent applications as training expectations for training; the large model of the technical effect dimension uses the beneficial effects in multiple patent applications as training expectations for training.

[0073] At the same time, in this embodiment, a variety of interactive input templates of the corresponding dimension are provided in the big model of each dimension; when making a technical summary of the technical content of the corresponding dimension in the patent to be transformed, the technical content of the corresponding dimension in the patent to be transformed is placed in an interactive input template of the corresponding dimension as the input information of the big model of the corresponding dimension, and the big model of the corresponding dimension processes the input information to obtain a preliminary technical summary result of the corresponding dimension.

[0074] In a certain example, the following method can be adopted to set the functions of the preliminary technical summary unit of the technical problem dimension, the technical means dimension and the technical effect dimension respectively.

[0075] a) Preliminary technical summary unit of technical problem dimension

[0076] The preliminary technical summary unit of the technical problem dimension is used to perform a technical summary of the background technology in the patent to be converted based on the large model of the technical problem dimension, and obtain the preliminary technical summary results of the patent to be converted in the technical problem dimension.

[0077] It should be noted that the background technology of the patent to be converted contains technical problems. In order to ensure the accuracy of subsequent feature extraction and intelligent retrieval and reduce the workload of subsequent processes, it is necessary to summarize the technical problems of the background technology.

[0078] Preferably, in the preliminary technical summary unit of the technical problem dimension, a large model of the technical problem dimension is built in. Various forms of interactive input templates of the technical problem dimension are provided in the large model of the technical problem dimension. By setting various forms of interactive input templates of the technical problem dimension, the output results of the large model of the technical problem dimension can be controlled to a certain extent. Exemplarily, an interactive input template of the technical problem dimension can be expressed as: summarizing the technical problems of "background technology". Here, "background technology" refers to the background technology in the patent to be converted. In addition, in other interactive input templates of the technical problem dimension, further limitations can be made on the basis of the above-mentioned interactive input templates in terms of the number of words in the technical problem summary, the included keywords, etc., so as to obtain various forms of interactive input templates of the technical problem dimension.

[0079] During the specific implementation process, the background technology of the patent to be converted is placed in an interactive input template in the technical problem dimension as input information of the big model of the technical problem dimension. After the big model of the technical problem dimension processes the input information, a preliminary technical summary result of the technical problem dimension is obtained.

[0080] b) Preliminary technical summary unit of technical means dimension

[0081] The preliminary technical summary unit of the technical means dimension is used to make a technical summary of the claims in the patent to be transformed based on the big model of the technical means dimension, and obtain the preliminary technical summary results of the patent to be transformed in the technical means dimension.

[0082] Preferably, in the preliminary technical summary unit of the technical means dimension, a large model of the technical means dimension is built in. Various forms of interactive input templates of the technical means dimension are provided in the large model of the technical means dimension. By setting various forms of interactive input templates of the technical means dimension, the output results of the large model of the technical means dimension can be controlled to a certain extent. Exemplarily, an interactive input template of the technical means dimension can be expressed as: summarizing the technical means of the "claims". Here, "claims" refers to the claims in the patent to be converted. In addition, in other interactive input templates of the technical means dimension, further limitations can be made on the basis of the above-mentioned interactive input templates in terms of the number of words in the technical means summary, the included keywords, etc., so as to obtain various forms of interactive input templates of the technical means dimension.

[0083] During the specific implementation process, the claims of the patent to be converted are placed in an interactive input template in the technical means dimension as input information of the big model of the technical means dimension. The big model of the technical means dimension processes the input information to obtain the preliminary technical summary results of the technical means dimension.

[0084] c) Preliminary technical summary unit of technical effect dimension

[0085] The preliminary technical summary unit of the technical effect dimension is used to make a technical summary of the beneficial effects in the patent to be transformed based on the large model of the technical effect dimension, and obtain the preliminary technical summary results of the patent to be transformed in the technical effect dimension.

[0086] It should be noted that, in the specific implementation process, the beneficial effects of the patent to be transformed (and / or specific phrases containing vocabulary representing technical effects) can be summarized into technical effects to obtain preliminary technical summary results of the patent to be transformed in the technical effect dimension.

[0087] Preferably, in the preliminary technical summary unit of the technical effect dimension, a large model of the technical effect dimension is built in. Various forms of interactive input templates of the technical effect dimension are provided in the large model of the technical effect dimension. By setting various forms of interactive input templates of the technical effect dimension, the output results of the large model of the technical effect dimension can be controlled to a certain extent. Exemplarily, an interactive input template of the technical effect dimension can be expressed as: summarizing the technical effects of "beneficial effects", where "beneficial effects" refer to the beneficial effects in the patent to be converted. In addition, in other interactive input templates of the technical effect dimension, further limitations can be made on the basis of the above-mentioned interactive input templates in terms of the number of words in the technical effect summary, the included keywords, etc., so as to obtain various forms of interactive input templates of the technical effect dimension.

[0088] During the specific implementation process, the beneficial effects of the patent to be converted are placed in an interactive input template in the technical effect dimension as input information of the big model of the technical effect dimension. After the big model of the technical effect dimension processes the input information, a preliminary technical summary result of the technical effect dimension is obtained.

[0089] 2) Output review unit for each dimension

[0090] The output review unit of each dimension is used to output and review the preliminary technical summary results of the patent to be transformed in the corresponding dimension, and obtain the technical summary results of the patent to be transformed in the corresponding dimension.

[0091] Preferably, in this embodiment, the output review unit of each dimension is used to review whether the technical summary result of the corresponding dimension is the technical-related information of the patent to be converted. If so, the output review is passed, and the preliminary technical summary result of the corresponding dimension is used as the technical summary result of the patent to be converted in the corresponding dimension; otherwise, the output review fails, and the preliminary technical summary unit of the corresponding dimension is returned, and the interactive input template of the corresponding dimension is replaced, and the technical content of the corresponding dimension in the patent to be converted is placed in the replaced interactive input template of the corresponding dimension, and again used as the input information of the large model of the corresponding dimension. The large model of the corresponding dimension processes the input information to obtain the preliminary technical summary result of the corresponding dimension, and then the output review unit of the corresponding dimension performs output review until the output review is passed.

[0092] During the specific implementation process, the output review unit of each dimension has a built-in output review controller and multiple output reviewers of the corresponding dimension; wherein the output review controller is used to control whether each output reviewer is effective and to control the logical relationship between the multiple effective output reviewers. Each output reviewer, from its own review perspective, reviews whether the preliminary technical summary results of the corresponding dimension are technical-related information of the patent to be converted. Exemplarily, the output reviewer can be one of a technology classifier, a tendency classifier, an output approval keyword whitelist, and an output approval keyword blacklist. During the specific implementation process, the output review controller adjusts the review intensity of the output review unit of the corresponding dimension by controlling the effectiveness of different output reviewers and controlling the logical relationship between the effective output reviewers.

[0093] Below, the technical classifier, tendency classifier, output approval keyword whitelist and output approval keyword blacklist are explained in detail.

[0094] Preferably, in this embodiment, both the technical classifier and the tendency classifier can be obtained based on neural network training. Among them, the technical classifier is a binary classification judgment of "yes / no is it a technical statement" for the preliminary technical summary results. And the tendency classifier is a binary classification judgment of "yes / no contains bad tendencies" for the preliminary technical summary results. Exemplarily, the technical classifier and the tendency classifier can be obtained based on a general CNN neural network model. In the specific implementation process, for the technical classifier, a large amount of input information and its technical labels can be made in advance. Exemplarily, 0 and 1 labels can be designed, where 1 represents that the input information is a technical statement and 0 represents that the input information is not a technical statement, thereby forming a technical training sample set. And based on the technical training sample set, the neural network corresponding to the technical classifier is trained for multiple rounds. After the training is passed, a technical classifier that can be used for binary classification judgment of actual technical statements is obtained. Similar to the technical classifier, the tendency classifier can also make a large amount of input information and its tendency labels in advance during the training process. Exemplarily, 0 and 1 labels can be designed, where 1 represents that the input information does not contain bad tendencies and 0 represents that the input information contains bad tendencies, thereby forming a tendency training sample set. Based on the tendency training sample set, the neural network corresponding to the tendency classifier is trained for multiple rounds. After the training, a tendency classifier that can be used for actual tendency binary classification is obtained. In the specific implementation process, in order to improve the efficiency of training and model application, the same technical classifier and tendency classifier can be selected for output audit units of different dimensions.

[0095] Preferably, considering that the output approval keyword whitelist and output approval keyword blacklist of different dimensions have different focuses, it is necessary to specifically set the output approval keyword whitelist and output approval keyword blacklist of different dimensions according to the content of the patent to be converted. Exemplarily, the output approval keyword whitelist of the technical problem dimension may include the core vocabulary in the technical problem solved by the patent to be converted. The output approval keyword blacklist of the technical problem dimension may include the relevant vocabulary of common technical problems that the patent to be converted does not involve in the relevant technical field. The output approval keyword whitelist of the technical means dimension may include the core vocabulary in the technical means of the patent to be converted. The output approval keyword blacklist of the technical means dimension may include the relevant vocabulary of common technical means that the patent to be converted does not involve in the relevant technical field. In addition, the output approval keyword whitelist of the technical effect dimension may include the core vocabulary in the technical effect of the patent to be converted. The output approval keyword blacklist of the technical effect dimension may include the relevant vocabulary of common technical effects that the patent to be converted does not involve in the relevant technical field.

[0096] Based on the above description, it can be known that for each output auditor, it is possible to audit whether the technical summary results of the corresponding dimension are technical related information of the patent to be converted from the corresponding audit perspective. That is, for the technical classifier, it can be judged whether the preliminary technical summary results are technical statements. For example, 1 is a technical statement, that is, technical related information of the patent to be converted; 0 is not a technical statement, that is, it is not technical related information of the patent to be converted. For the tendency classifier, it can be judged whether the preliminary technical summary results contain bad tendencies. 1 is not containing bad tendencies, that is, technical related information of the patent to be converted; 0 is containing bad tendencies, that is, it is not technical related information of the patent to be converted. For the output approval keyword whitelist, it can be judged whether the preliminary technical summary results contain words in the output approval keyword whitelist. For example, 1 is included, that is, technical related information of the patent to be converted; 0 is not included, that is, it is not technical related information of the patent to be converted. For the output approval keyword blacklist, it can be judged whether the preliminary technical summary results contain words in the output approval keyword blacklist. For example, 1 is not included, that is, technical related information of the patent to be converted; 0 is included, that is, it is not technical related information of the patent to be converted.

[0097] The output audit controller can control different output auditors to be effective, and only the effective output auditor can audit whether the technical summary results of the corresponding dimension are technical related information of the patent to be converted. In addition, the output audit controller can also adjust the audit strength of the output audit unit of the corresponding dimension by controlling the logical relationship between the effective output auditors. Exemplarily, when the output audit controller controls the above four output auditors to be effective and the logical relationship is "and", only when the judgment results of the above four output auditors are all 1, the output audit passes; otherwise, the output audit fails. When the output audit controller controls the above four output auditors to be effective and the logical relationship is "or", the judgment result of any output auditor is 1, and the output audit passes; only when the judgment results of the four output auditors are all 0, the output audit fails. Therefore, the purpose of adjusting the audit strength of the output audit unit can be achieved by controlling the effectiveness and logical relationship of the output auditor by the output audit controller.

[0098] (3) Multi-dimensional candidate patent search module

[0099] The multi-dimensional candidate patent search module extracts features from the technical summary results of each patent to be converted in different dimensions, and performs intelligent searches based on the extracted technical features to obtain a set of candidate patents for the corresponding dimensions of the patents to be converted.

[0100] That is, the multi-dimensional candidate patent search module extracts features from the technical summary results of the patents to be transformed in the technical problem dimension, technical means dimension, and technical effect dimension, and performs intelligent searches based on the extracted technical features to obtain a set of candidate patents for the patents to be transformed in the technical problem dimension, technical means dimension, and technical effect dimension.

[0101] Preferably, the multi-dimensional candidate patent search module includes: a technical feature extraction unit for each dimension, a search expression generation unit, and an intelligent search unit. The specific functions are described as follows.

[0102] 1) Technical feature extraction unit for each dimension

[0103] The technical feature extraction unit of each dimension extracts features from the technical summary results of the patent to be transformed in the corresponding dimension, and obtains the technical features of the patent to be transformed in the corresponding dimension.

[0104] Specifically, in the technical feature extraction unit of the technical problem dimension, feature extraction is performed on the technical summary results of the patent to be transformed in the technical problem dimension to obtain the technical features of the patent to be transformed in the technical problem dimension. In the technical feature extraction unit of the technical means dimension, feature extraction is performed on the technical summary results of the patent to be transformed in the technical means dimension to obtain the technical features of the patent to be transformed in the technical means dimension. In the technical feature extraction unit of the technical effect dimension, feature extraction is performed on the technical summary results of the patent to be transformed in the technical effect dimension to obtain the technical features of the patent to be transformed in the technical effect dimension.

[0105] Preferably, in this embodiment, the TF-IDF algorithm can be combined with a key dictionary to extract features from the technical summary results of the patent to be transformed in the technical problem dimension, technical means dimension and technical effect dimension, so as to obtain the technical features of the patent to be transformed in the technical problem dimension, technical means dimension and technical effect dimension; or, feature extraction is performed on the technical summary results of the patent to be transformed in the technical problem dimension, technical means dimension and technical effect dimension, and after obtaining the corresponding feature extraction results, the corresponding feature extraction results are also associated and inferred with hypernym elements to summarize the technical feature hypernyms of the corresponding feature extraction results, and the feature extraction results and technical feature hypernyms of the patent to be transformed in the technical problem dimension, technical means dimension and technical effect dimension are respectively summarized to form the technical features of the patent to be transformed in the technical problem dimension, technical means dimension and technical effect dimension. It should be noted that in this embodiment, the technical features of the patent to be transformed in the technical problem dimension, technical means dimension and technical effect dimension are embodied in the form of feature words.

[0106] In the specific implementation process, a deep learning model, such as a convolutional neural network (CNN), can be used to associate and infer hypernym elements of each technical feature extraction result to obtain the technical feature hypernym of the corresponding feature extraction result. Through the above deep learning model, intelligent retrieval based on the summarized technical feature hypernyms can effectively identify certain patents with different patent expressions but the same hypernym concepts (that is, the same essential content).

[0107] 2) Retrieval expression generation unit

[0108] The search expression generation unit comprehensively analyzes the technical features of the patents to be converted in different dimensions and obtains a variety of intelligent search expressions for the patents to be converted in different dimensions.

[0109] Preferably, the search expression generation unit can combine the format requirements of the intelligent search engine for the search expression and use the following method to obtain the intelligent search expression of the patent to be converted in the technical problem dimension, technical means dimension and technical effect dimension:

[0110] The technical features of the patent to be converted that appear simultaneously in the technical problem dimension, technical means dimension and technical effect dimension are regarded as key technical features;

[0111] Determine the similarity of multiple technical features of the patent to be converted in the technical problem dimension, and take multiple technical features whose similarity exceeds a certain similarity threshold as similar features in the technical problem dimension, and connect them with an or relationship;

[0112] Determine the similarity of multiple technical features in the technical means dimension of the patent to be converted, and take multiple technical features whose similarity exceeds a certain similarity threshold as similar features in the technical means dimension, and connect them with an or relationship;

[0113] Determine the similarity of multiple technical features of the patent to be converted in the technical effect dimension, and take multiple technical features whose similarity exceeds a certain similarity threshold as similar features in the technical effect dimension, and connect them with an or relationship;

[0114] Use key technical features as necessary keywords for retrieval, and connect similar features of the technical problem dimension connected by the relationship of "or", and the remaining technical features of the technical problem dimension using multiple forms of logical relationships to form multiple intelligent retrieval expressions of the patent to be converted in the technical problem dimension;

[0115] Use key technical features as necessary keywords for retrieval, and connect similar features in the technical means dimension connected by the relationship of "or", and the remaining technical features in the technical means dimension using a variety of different forms of logical relationships to form a variety of intelligent retrieval expressions in the technical means dimension of the patent to be converted;

[0116] The key technical features are used as necessary keywords for retrieval, and similar features in the technical effect dimension connected by the relationship of "or" and the remaining technical features in the technical effect dimension are connected by a variety of different forms of logical relationships to form a variety of intelligent retrieval expressions for the patent to be converted in the technical effect dimension.

[0117] Exemplarily, the various forms of logical relationship connections used here can be logical relationships such as and, or, etc., thereby enriching the form of intelligent search expressions and being closer to the actual search habits of technicians. In the above manner, a variety of intelligent search expressions can be formed for the patents to be converted in the dimensions of technical problems, technical means, and technical effects, which can ensure that the intelligent search expressions contain key technical features and the richness of the retrieved patents.

[0118] 3) Intelligent retrieval unit

[0119] The intelligent retrieval unit uses a variety of intelligent retrieval expressions of the patent to be converted in different dimensions to perform patent retrieval in the intelligent retrieval system to obtain a candidate patent set of the patent to be converted in different dimensions.

[0120] Preferably, in the intelligent retrieval unit, a plurality of intelligent retrieval expressions of the patent to be transformed in the technical problem dimension, technical means dimension and technical effect dimension are used to perform patent retrieval in the intelligent retrieval system, and the retrieved similar patents of the patent to be transformed in the technical problem dimension, technical means dimension and technical effect dimension are used as the candidate patent sets of the patent to be transformed in the technical problem dimension, technical means dimension and technical effect dimension.

[0121] Specifically, in the intelligent search unit, for each dimension, the candidate patent set of the patent to be converted in the corresponding dimension is obtained in the following way:

[0122] Use each intelligent search expression of the patent to be converted in the corresponding dimension to perform patent search in the intelligent search system respectively, and obtain M related patents with the highest similarity corresponding to each intelligent search expression.

[0123] The related patents that appear repeatedly in the top M related patents with similarity corresponding to different types of intelligent search expressions are taken as the preliminary candidate patents of the corresponding dimension, and the top N preliminary candidate patents with similarity to the technical content of the patent to be converted in the corresponding dimension are selected from them as the candidate patent set of the patent to be converted in the corresponding dimension.

[0124] In the specific implementation process, for the technical problem dimension, technical means dimension and technical effect dimension, by performing the above operations respectively, the candidate patent sets of the patent to be converted in the technical problem dimension, technical means dimension and technical effect dimension can be obtained. Exemplarily, M is greater than N. In the specific implementation process, considering that the number of patent conversion and application objects finally selected is small, or even only one. The value of N should not be too large. Exemplarily, N is 50. The value of N can also be set adaptively according to the number of preliminary candidate patents.

[0125] (3) Supply and demand relationship determination module

[0126] The supply and demand relationship determination module selects the co-occurring patent with the highest recommendation degree from the candidate patent sets of different dimensions for each patent to be converted; the patent owner of the co-occurring patent is used as the innovation subject of the other party between the supplier and the demander, and a corresponding relationship between the patent to be converted and the co-occurring patent under each technical problem category is established, forming a docking relationship between the innovation subjects of the supplier and the demander.

[0127] In the supply and demand relationship determination module, for each patent to be converted, the co-occurrence degree, average similarity and operation degree of the co-occurring patents between the candidate patent sets of different dimensions of the patent to be converted are obtained to determine the recommendation degree of each co-occurring patent; and the co-occurring patent with the highest recommendation degree is selected.

[0128] In the supply and demand relationship determination module, the candidate patent sets of patents to be converted in different dimensions are processed as follows:

[0129] 1) Calculate the co-occurrence DC of co-occurring patents

[0130] The similarity of each candidate patent in the candidate patent set of different dimensions of the patent to be converted (i.e., the similarity of the technical content of each candidate patent with the patent to be converted in the corresponding dimension) is arranged in descending order, and a corresponding similarity score is assigned. For example, in each candidate patent set, the similarity score of the candidate patent with the highest similarity is 50 points, and so on, the similarity score of the candidate patent with the lowest similarity is 1 point.

[0131] Candidate patents that appear in at least two candidate patent sets in different dimensions are selected as co-occurring patents. The similarity scores of each co-occurring patent in each candidate patent set are summed up and normalized to obtain the co-occurrence degree DC of each co-occurring patent.

[0132] Exemplarily, the maximum value of the sum of similarity scores of co-occurring patents in each candidate patent set is 150, and is normalized to a value of [1,100], thereby obtaining the co-occurrence degree of each co-occurring patent.

[0133] 2) Calculate the average similarity DS of co-occurring patents

[0134] The similarity scores of each co-occurring patent in each candidate patent set are averaged to obtain the average similarity of each co-occurring patent.

[0135] It should be noted that only co-occurring patents are included in the calculation. That is, if a candidate patent only appears in two candidate patent sets, only the average value of its similarity in the two candidate patent sets is calculated.

[0136] The maximum value of the similarity score is 100, and the maximum value of the average similarity DS is also 100.

[0137] 3) Calculate the operational degree DO of co-occurring patents

[0138] According to the number of different types of operating behaviors implemented by the patentee in each co-occurring patent, the operating degree of the corresponding co-occurring patent is obtained.

[0139] Specifically, different types of operations implemented by patent owners include: transfer, license, and pledge. The weights of the number of transfers, licenses, and pledges are set to a, b, and c, respectively, and a+b+c=1. For each co-occurring patent, assume that the number of transfers is N1, the number of licenses is N2, and the number of pledges is N3. At this time, a*N1+b*N2+c*N3 is normalized to a value of [1,100] to obtain the operation degree DO of the current co-occurring patent.

[0140] 4) Calculate the recommendation degree R of co-occurring patents

[0141] Assume that the weights of co-occurrence, average similarity, and operation degree are α, β, and γ respectively, and satisfy α+β+γ=1.

[0142] The recommendation degree R of each co-occurring patent is calculated according to the formula: R = α*DC + β*DS + γ*DO.

[0143] After that, the co-occurring patents with the highest recommendation can be selected.

[0144] Subsequently, the patent holder of the co-occurring patent can be used as the other innovation subject of the supplier and the demander, and a corresponding relationship between the patent to be transformed and the co-occurring patent under each technical problem category can be established, thereby forming a docking relationship between the innovation subjects of the supplier and the demander.

[0145] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0146] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. An intelligent docking system for patent supply and demand relationships between innovative entities, characterized by: The system comprises: The module for determining patents to be converted uses a large model to summarize the technical issues of patents disclosed by the innovative entities of one of the supply and demand sides, and classify the technical issues; the high-value patents under each technical issue category are respectively regarded as patents to be converted; The multi-dimensional technical summary module uses a large model to obtain the technical summary of each patent to be converted in different dimensions, and obtains the technical summary results of the corresponding patent to be converted in the corresponding dimension after reviewing the technical summary; The multi-dimensional candidate patent search module extracts features from the technical summary results of each patent to be converted in different dimensions, and performs intelligent searches based on the extracted technical features to obtain a candidate patent set of the corresponding patent to be converted in the corresponding dimension; The supply and demand relationship determination module selects the co-occurring patent with the highest recommendation degree from the candidate patent sets of different dimensions for each patent to be converted; the patent owner of the co-occurring patent is used as the innovation subject of the other party between the supplier and the demander, and a corresponding relationship between the patent to be converted and the co-occurring patent under each technical problem category is established, forming a docking relationship between the innovation subjects of the supplier and the demander.

2. The patent supply and demand intelligent docking system between innovation entities according to claim 1 is characterized in that: The module for determining patents to be converted includes: a unit for summarizing technical issues of the innovation subject, a unit for classifying technical issues, and a unit for selecting high-value patents; among them, The innovation subject technical problem summary unit receives patents disclosed by the innovation subject of one of the supply side and the demand side, and summarizes the technical problems of each patent using the big model; The technical problem classification unit clusters technical problems based on clustering distribution to achieve classification of technical problems; The high-value patent selection unit conducts value assessment on each patent under each technical problem category and selects high-value patents as patents to be converted.

3. The patent supply and demand intelligent docking system between innovation entities according to claim 1 is characterized in that: The multi-dimensional technical summary module includes a preliminary technical summary unit and an output review unit for each dimension; The preliminary technical summary unit of each dimension is used to summarize the technical content of the corresponding dimension in the patent to be converted based on the macro model of the corresponding dimension, and obtain the preliminary technical summary results of the patent to be converted in the corresponding dimension; The output review unit of each dimension is used to output and review the preliminary technical summary results of the patent to be transformed in the corresponding dimension, and obtain the technical summary results of the patent to be transformed in the corresponding dimension.

4. The patent supply and demand intelligent docking system between innovation entities according to claim 3 is characterized in that: In each dimension of the large model, there are various forms of interactive input templates of the corresponding dimension; When making a technical summary of the technical contents of the corresponding dimensions in the patent to be transformed, the technical contents of the corresponding dimensions in the patent to be transformed are placed in an interactive input template of the corresponding dimension as the input information of the big model of the corresponding dimension. After the big model of the corresponding dimension processes the input information, a preliminary technical summary result of the corresponding dimension is obtained.

5. The patent supply and demand intelligent docking system between innovation entities according to claim 3 is characterized in that: The output review unit of each dimension is used to review whether the technical summary result of the corresponding dimension is the technical related information of the patent to be converted. If yes, the output is reviewed and approved, and the preliminary technical summary results of the corresponding dimensions are used as the technical summary results of the patent to be converted in the corresponding dimensions; Otherwise, the output review fails, and the preliminary technical summary unit of the corresponding dimension is returned, and the interactive input template of the corresponding dimension is replaced. The technical content of the corresponding dimension in the patent to be converted is placed in the replaced interactive input template of the corresponding dimension, and again used as the input information of the large model of the corresponding dimension. The large model of the corresponding dimension processes the input information to obtain the preliminary technical summary result of the corresponding dimension, and then the output review unit of the corresponding dimension performs output review until the output review passes.

6. The patent supply and demand intelligent docking system between innovation entities according to claim 5 is characterized in that: Each dimension of the output audit unit has a built-in output audit controller of the corresponding dimension and multiple output auditors; among them, The output audit controller is used to control whether each output auditor is effective, and to control the logical relationship between multiple effective output auditors; Each output reviewer, from its own review perspective, reviews whether the preliminary technical summary results of the corresponding dimension are technically relevant information for the patent to be converted.

7. The patent supply and demand intelligent docking system between innovation entities according to claim 6 is characterized in that: The output reviewer is one of a technical classifier, a tendency classifier, an output approval keyword whitelist, and an output approval keyword blacklist; The output audit controller of each dimension adjusts the audit strength of the output audit unit of the corresponding dimension by controlling the effectiveness of different output auditors and controlling the logical relationship between the effective output auditors.

8. The patent supply and demand intelligent docking system between innovation entities according to any one of claims 1 to 7, characterized in that: The multi-dimensional candidate patent search module includes: a technical feature extraction unit for each dimension, a search expression generation unit, and an intelligent search unit; The technical feature extraction unit of each dimension extracts features from the technical summary results of the patent to be converted in the corresponding dimension, and obtains the technical features of the patent to be converted in the corresponding dimension; The search expression generation unit comprehensively analyzes the technical features of the patent to be converted in different dimensions and obtains multiple intelligent search expressions of the patent to be converted in different dimensions; The intelligent retrieval unit uses a variety of intelligent retrieval expressions of the patent to be converted in different dimensions to perform patent retrieval in the intelligent retrieval system to obtain a candidate patent set of the patent to be converted in different dimensions.

9. The patent supply and demand intelligent docking system between innovation entities according to claim 8 is characterized in that: In the supply and demand relationship determination module, for each patent to be converted, the co-occurrence degree, average similarity and operation degree of the co-occurring patents between the candidate patent sets of different dimensions of the patent to be converted are obtained to determine the recommendation degree of each co-occurring patent; and the co-occurring patent with the highest recommendation degree is selected.

10. The patent supply and demand intelligent docking system between innovation entities according to claim 9 is characterized in that: In the supply and demand relationship determination module, for each patent to be converted, execute: Arrange the similarity of each candidate patent in the candidate patent set of different dimensions of the patent to be converted in descending order and assign corresponding similarity scores; Select candidate patents that appear in at least two candidate patent sets of different dimensions as co-occurring patents, add up the similarity scores of each co-occurring patent in each candidate patent set, and normalize them to obtain the co-occurrence degree of each co-occurring patent; The similarity scores of each co-occurring patent in each candidate patent set are averaged to obtain the average similarity of each co-occurring patent; According to the number of different types of operating behaviors implemented by the patentee in each co-occurring patent, the operating degree of the corresponding co-occurring patent is obtained.

Images