Intellectual property service intelligent recommendation method and system and electronic equipment

Abstract

The invention provides an intellectual property service intelligent recommendation method and system and electronic equipment, and the method comprises the steps: carrying out the feature extraction of an invention technology material submitted by a user, obtaining a technical feature vector, carrying out the data preprocessing, constructing a technical feature matrix based on the technical feature vector, and constructing a human body immune model. Inputting the technical characteristic matrix as an antigen into a human immune model, generating an initial technical antibody, obtaining an initial technical antibody candidate set based on the initial technical antibody, performing clone mutation operation on the initial technical antibody candidate set based on two-channel evaluation, obtaining a mutation technical antibody candidate set, and obtaining a mutation technical antibody candidate set; carrying out three-property optimization screening on the variation technology antibody candidate set to obtain an immune technology antibody set; carrying out simulation examination verification on the immune technology antibody set based on GAN to obtain a technology optimization strategy; optimizing the immune technology antibody set based on the technology optimization strategy to generate a corresponding intellectual property service strategy; and a more refined service recommendation strategy is provided for intellectual property service recommendation.

Description

Technical Field

[0001] The present invention relates to the technical field of intellectual property services, and in particular to an intelligent recommendation method, system and electronic equipment for intellectual property services. Background Art

[0002] Against the backdrop of increasingly fierce market competition and continuous technological advancements, intellectual property has become a strategic resource for companies to build core competitiveness. However, with the increasing complexity of technological innovation and the explosive growth of patent data, traditional intellectual property service models have certain limitations in this context.

[0003] On the one hand, traditional methods rely on simple keyword matching or TF-IDF weight calculation when extracting innovative points, and are unable to capture the deep semantic associations of the technical materials of the invention, such as the implicit logic of the technical principles and the characteristics of cross-domain technology integration. For example, when extracting features for "battery life prediction method based on deep learning", traditional models may only extract isolated keywords such as "battery", "life", and "deep learning", while ignoring potential core innovations such as "electrochemical impedance spectroscopy feature fusion" and "transfer learning cross-scenario adaptation". This will cause subsequent patent searches and novelty assessments to miss key technical differences, resulting in the risk of misjudgment.

[0004] On the other hand, traditional methods of judging and analyzing the innovative points of intellectual property mostly focus on the "three characteristics" at the patent legal level, namely novelty, creativity and practicality, but lack quantitative analysis of the commercialization potential of technology, such as market exclusivity and industrialization costs.

[0005] Therefore, existing intellectual property service methods seem to be unable to cope with complex and changing market demands. It is difficult to accurately identify high-value technical points in customers' invention technologies and to formulate forward-looking protection strategies for the above high-value technical points. Summary of the Invention

[0006] In view of the above-mentioned problems, in combination with the first aspect of the present invention, an embodiment of the present invention provides an intelligent recommendation method for intellectual property services, the method comprising: Extract features from the invention technical materials submitted by users, obtain technical feature vectors, perform data preprocessing on the technical feature vectors, and construct a technical feature matrix based on the technical feature vectors; Constructing a human immune model, inputting the technical feature matrix as an antigen into the human immune model, generating initial technical antibodies, and obtaining an initial technical antibody candidate set based on the initial technical antibodies, wherein the initial technical antibodies represent the innovation points corresponding to the inventive technical materials; Performing a cloning mutation operation on the initial technical antibody candidate set based on a dual-channel evaluation to obtain a mutated technical antibody candidate set. The dual-channel evaluation includes a technical channel and a commercial channel. The technical channel is represented by calculating the novelty score of the invention technical materials submitted by the user, and the commercial channel is represented by calculating the commercial value index of the invention technical materials submitted by the user; Performing three-property optimization screening on the candidate set of variant technology antibodies to obtain an immune technology antibody set, wherein the three-property optimization screening is performed on the candidate set of variant technology antibodies to evaluate novelty, practicality, and creativity; Conducting simulated review and verification of the immune technology antibody set based on GAN to obtain technology optimization strategies. The simulated review and verification is represented by conducting internal simulated technical review of the invention technical materials. Based on the technical optimization strategy, the immune technology antibody set is optimized to generate the corresponding intellectual property service strategy.

[0007] In another aspect, an embodiment of the present invention further provides an intellectual property service intelligent recommendation system, comprising: A collection module, which is used to obtain the invention technical materials submitted by the user and obtain the technical feature vectors corresponding to the invention technical materials; a processing module, the processing module being used to perform data preprocessing on the technical feature vector and construct a technical feature matrix based on the technical feature vector; A generation module, which is used to construct a human immune model and input a technical feature matrix into the human immune model to obtain an initial technical antibody candidate set based on the initial technical antibodies, wherein the initial technical antibodies represent the innovation points corresponding to the inventive technical materials; A cloning module, which is used to perform a cloning mutation operation on the initial technical antibody candidate set through a dual-channel evaluation to obtain a mutated technical antibody candidate set; A screening and review module is used to perform three-property optimization screening on the candidate set of variant technology antibodies to obtain an immune technology antibody set, and to perform simulation review and verification on the immune technology antibody set based on GAN to obtain a technology optimization strategy; The service module is used to optimize the immune technology antibody set according to the technology optimization strategy and generate a corresponding intellectual property service strategy.

[0008] In another aspect, an embodiment of the present invention further provides an electronic device, including: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform any method of the embodiments of the present invention.

[0009] Based on the above aspects, the embodiment of the present application inputs the innovative points extracted from the invention technical materials submitted by the customer as antigens into the human immunological model, matches the antigens with corresponding initial technical antibodies based on the above model, and obtains a candidate set of mutant technical antibodies by cloning and mutating the initial technical antibodies. The candidate set of mutant technical antibodies is subjected to improved three-property screening to generate an immune technology antibody set, so that the generated immune technology antibody set is more in line with the core innovative points of the invention technical materials. At the same time, by introducing the advanced three-property screening and internal simulation review mechanism, the commercialization potential of the customer's invention technology is analyzed and evaluated, and based on the above content, customers are provided with services that better meet customer needs. In summary, this intelligent recommendation method for intellectual property services improves the technical limitations of the extraction of innovative points that cannot capture the deep semantic associations of the invention technical materials and the overly one-sided judgment and analysis of the innovative points of intellectual property rights, and provides a more refined service recommendation strategy for intellectual property service recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 The present invention provides an embodiment of an intelligent recommendation method for intellectual property services.

[0011] Figure 2 Schematic diagram of an intelligent recommendation system for intellectual property services provided by an embodiment of the present invention.

[0012] Figure 3 is a schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0013] The present invention will be described in detail below with reference to the accompanying drawings. Figure 1 This is a schematic diagram of the execution flow of an intelligent recommendation method for intellectual property services provided by an embodiment of the present invention. The intelligent recommendation method for intellectual property services is introduced in detail below.

[0014] Step S1: extract features from the invention technical materials submitted by the user to obtain a technical feature vector, perform data preprocessing on the technical feature vector, and construct a technical feature matrix based on the technical feature vector.

[0015] In this embodiment, step S1 includes: Step S1-1, extracting features from the invention technical materials submitted by the user.

[0016] Specifically, feature extraction is performed on the text content in the invention technical materials submitted by users, and conflicts in synonyms in the text content are eliminated to construct a "problem, solution, effect" triple. For example, a certain invention technical material contains the text "adopting a multi-layer graphene electrode structure and improving the lithium ion transmission efficiency through gradient pore design, so that the battery cycle life is extended to 1200 times." After feature extraction of the above text, the triple {structure: multi-layer graphene electrode, function: gradient pore design, effect: the lithium ion transmission efficiency is improved so that the battery cycle life is not less than 1200 times} is generated.

[0017] The drawing content of the invention technical materials is converted into a vector structure, and the geometric parameters in the drawing content are extracted to establish a structure-function mapping relationship. For example, a thermos cup manufacturer proposed a double-layer vacuum structure solution by improving the design drawings to address the problem of insufficient thermal insulation performance of the traditional single-layer stainless steel cup body. The improved drawings contain three-layer components, of which the outer layer is 0.3mm stainless steel with a surface polishing accuracy not exceeding 0.8μm, the middle layer is a 0.1mm vacuum layer with a vacuum degree not exceeding 0.001Pa, the interlayer spacing is 5mm, and the inner layer is a 0.3mm copper-plated layer with a surface roughness of 1.2μm. The above three-layer structure is analyzed by vectorization, and the key parameters are extracted at the same time to establish the corresponding structure-function mapping relationship. The structure-function mapping relationship is expressed as {middle layer, suppresses heat convection, and the insulation time is increased from 2 hours to 6 hours}, {inner layer, diffuse reflection improves heat reflection efficiency by 15% and reduces heat radiation loss} and {outer layer, increases friction and reduces the risk of the thermos cup slipping}.

[0018] The tabular contents in technical materials are stored in a structured manner and numerically normalized, and unstructured parameters are quantified into standard parameters.

[0019] Step S1-2: convert the "problem, solution, effect" triple, the structure-function mapping relationship and the standard parameters into multiple technical feature vectors, and map the generated multiple technical feature vectors to the same feature space, and construct a technical feature matrix based on the above technical feature vectors.

[0020] Specifically, the "problem, solution, effect" triple is converted into a semantic technical feature vector, the structure-function mapping relationship is converted into a physical technical feature vector, the standard parameters are converted into a numerical technical feature vector, and the semantic technical feature vector, physical technical feature vector and numerical technical feature vector are mapped to the same feature space to form a technical feature matrix.

[0021] Step S2: construct a human immune model, input the technical feature matrix as an antigen into the human immune model, generate initial technical antibodies, and obtain an initial technical antibody candidate set based on the initial technical antibodies. The initial technical antibodies represent the innovation points corresponding to the invention technical materials.

[0022] In this embodiment, step S2 includes: Step S2-1: Obtain a preset antibody library and perform antibody matching on the antibody library based on the input antigen.

[0023] Specifically, the antibody library includes three sub-antibody libraries: basic antibody library, case antibody library and dynamic antibody library. The basic antibody library is constructed based on the international patent classification system. The basic antibody library is used to store the technical feature word embedding vectors extracted from the patent claims in each technical field, and appends the IPC classification code to achieve accurate matching of technical fields. For example, the technical feature word embedding vector of a G06N classification can be expressed as a keyword combination {Grey Wolf Algorithm, Neural Network, Deep Learning Model}. The case antibody library includes complete technical solutions of historical authorized patents and stores case feature vectors.

[0024] For example, the structure-parameter combination formed by the gradient pore electrode and the fractal flow channel structure and its corresponding parameters is marked with legal status labels. The legal status labels include authorized, invalid and in litigation. The authorized label is given a weight of 1.2, the invalid label is given a weight of 0.8, and the in litigation label is given a weight of 1.0. The dynamic antibody library aggregates the technology hotspots of the past three years in real time, quantifies the technology evolution trend through LSTM, and attaches the technology life cycle curve to make three-stage judgments on germination, growth and maturity. For example, the average annual growth rate of technological innovation related to solid-state battery electrolytes is 21%, which determines that the technological innovation of solid-state battery electrolytes is in its infancy. The matching weights of the three sub-antibody libraries mentioned above, namely, the basic antibody library, the case antibody library and the dynamic antibody library, account for 40%, 35% and 25% respectively.

[0025] Furthermore, the antigen is matched with the antibody library in parallel. The parallel antibody matching is performed by simultaneously matching the antigen with the three sub-antibody libraries in the antibody library, calculating the similarity of the antigen to the basic antibody library in the basic antibody library, obtaining the basic antibody library similarity data, screening antibodies in the same technical field based on the basic antibody library similarity data, calculating the case antibody library similarity of the antigen in the case antibody library, obtaining the case antibody library similarity data, and screening successful case antibodies in combination with the case legal status label, calculating the dynamic antibody library similarity in the dynamic antibody library, obtaining the dynamic antibody library similarity data, and screening antibodies in the budding or growth stage in combination with the life cycle curve.

[0026] Step S2-2: Obtain initial technical antibodies with similarity, and output the top 30 initial technical antibodies ranked by similarity as the initial technical antibody candidate set.

[0027] Specifically, based on the obtained basic antibody library similarity data, case antibody library similarity data, and dynamic antibody library similarity data, and combined with the matching weight ratios of the three sub-antibody libraries of the dynamic antibody library, the initial technical antibody similarity is calculated to obtain the initial technical antibody similarity result.

[0028] For example, the technical feature matrix corresponding to a new lithium battery electrode design is input into the human immune model as an antigen. The antigen is matched with antibodies in the basic antibody library. The overlap between the three keywords "electrode", "porosity" and "fractal flow channel" in the antigen and the matching antibodies is calculated. The co-occurrence rate of the technical feature words is 0.72 based on the overlap. At the same time, the IPC subclass matching degree between the antigen and the matching antibody is calculated to be 0.95. The basic library similarity calculation is performed on the co-occurrence rate of the technical feature words and the IPC subclass matching degree. The basic library similarity calculation is expressed as the basic library similarity. , where 0.6 is the weight of the co-occurrence rate of technical feature words, 0.4 is the weight of the IPC subclass matching degree, and the antigen is matched with the antibody in the case antibody library. Taking the antigen as the benchmark, the relative difference ratio between the case antibody and the antigen is calculated. At this time, the expansion rate and cycle life of the antigen are 6% and 1500 times respectively, and the expansion rate and cycle life of the antibody matched in the case technical antibody library are 4% and 1000 times respectively, and the legal status label of the matched antibody is authorized. The relative difference ratio of the expansion rate and cycle life is calculated respectively, that is, the relative difference ratio of the expansion rate , relative difference ratio of cycle life , calculate the average relative difference ratio of expansion rate and cycle life , the average relative difference ratio is combined with the corresponding weight of the legal status label to obtain the case antibody library similarity The antigen is matched with the antibody in the dynamic antibody library. The matching antibody is the technical hotspot "solid-state battery interface optimization". The dot product between the antigen and the matching antibody is 0.85. The dot product is the dot product of the antigen feature vector and the matching antibody vector. The technical hotspot coefficient is 1.15. The technical hotspot coefficient is that the popularity of this technical hotspot has increased by 15% in the past three months. The dynamic library similarity is calculated based on the dot product between the antigen and the matching antibody and the technical hotspot coefficient. The initial technical antibody similarity is calculated based on the basic antibody library similarity, case antibody library similarity, and dynamic antibody library similarity obtained by the above calculations. .

[0029] Furthermore, the initial technical antibody similarities are sorted in descending order of similarity, and the top 30 initial technical antibodies are output as the initial technical antibody candidate set.

[0030] Step S3, performing a cloning mutation operation on the initial technical antibody candidate set based on a dual-channel evaluation to obtain a mutated technical antibody candidate set, wherein the dual-channel evaluation includes a technical channel and a commercial channel, wherein the technical channel is represented by calculating the novelty score of the invention technical material, and the commercial channel is represented by calculating the commercial value index of the invention technical material submitted by the user.

[0031] In this embodiment, the technology channel calculates the difference between the initial technology antibody and the existing patents based on the Jaccard-TFIDF algorithm by comparing the existing patents with similar technology feature vectors in the external patent search library in real time. The Jaccard-TFIDF calculation can be expressed as: ; Obtain the difference results, introduce the time attenuation factor into the difference results, and generate a novelty score. For example, the initial technical antibody candidate set corresponding to a new lithium battery electrode design scheme is compared with the external patent library in real time. The comparison patent is found to have technical features similar to the initial technical antibody candidate set, such as {electrode porosity 10%, linear flow channel, expansion rate 5%}, and the application time of the comparison patent is 3 years different from that of the initial technical antibody. Therefore, the attenuation factor corresponding to the comparison patent is 0.85. The difference between the comparison patent and the initial technical antibody candidate set is calculated by Jaccard-TFIDF. , calculate the novelty score based on the difference result and the attenuation factor .

[0032] In this embodiment, the commercial channel constructs a dual-dimensional index system of practicality and creativity, which includes a practicality dimension and a creativity dimension. The practicality dimension includes two dimensional attributes: implementation feasibility and cost controllability. The implementation feasibility score and the cost controllability score are calculated, and the practicality score is calculated based on the practicality scoring formula. The practicality scoring formula can be expressed as: ; above Expressed as a practicality score, Expressed as implementation feasibility score, Expressed as the cost controllability score, Expressed as the implementation feasibility score threshold, Cost controllability score threshold, It is expressed as a domain sensitivity coefficient. For example, if the supply chain maturity score of the initial technology antibody corresponding to a new lithium battery electrode design is 1.2 and the production line compatibility score is 0.8, then the initial technology antibody , the above 0.6 is the supply chain maturity score weight, 0.4 is the production line compatibility score weight, the initial technology antibody corresponding to the marginal cost score is 1.18, the patent maintenance cost score is 1.0, then the initial technology antibody , the above 0.7 is the marginal cost score weight of output, 0.3 is the patent maintenance cost score weight, the initial technology antibody corresponds to is 1.2, is 1.3, is 3.0, then the initial technical antibody .

[0033] The creativity dimension includes two dimensional attributes: technological breakthrough and market exclusivity. The technological breakthrough score and market exclusivity score are calculated based on the creativity scoring formula. The creativity scoring formula can be expressed as: ; above Expressed as a creativity score, Expressed as a technology breakthrough score, Expressed as a market exclusivity score, Expressed as the technology breakthrough score threshold, Market exclusivity score threshold, It is expressed as the field competition intensity coefficient. For example, the patent citation network centrality score of the initial technology antibody corresponding to a new lithium battery electrode design solution is 1.5, and the technology gap filling score is 0.65. The patent citation network centrality is expressed as the calculated betweenness centrality BC, and the technology gap filling score is expressed as the percentage of blank areas under the IPC classification. , the above 0.6 is the patent citation network centrality score weight, 0.4 is the technology gap filling score weight, the initial technology antibody corresponding to the claim coverage score is 0.85, and the circumvention difficulty score is 0.85. The claim coverage is expressed as the proportion of technology nodes covered by the protection scope, and the circumvention difficulty is expressed as the proportion of alternative technology paths. The initial technology antibody has a claim coverage score of 0.85 and a circumvention difficulty score of 0.85. , the above 0.7 is the claim coverage score weight, 0.3 is the circumvention difficulty score weight, the initial technical antibody corresponds to is 1.5, is 1.0, β is 2.5, then the initial technical antibody .

[0034] Furthermore, the practicality score and the creativity score are combined with the practicality score weight and the creativity score weight to synthesize them. The practicality score weight and the creativity score weight are 0.7 and 0.3 respectively, to obtain the commercial value index. For example, the practicality score and the creativity score of the initial technical antibody corresponding to the above-mentioned new lithium battery electrode design scheme are 0.80 and 0.69 respectively. Then the commercial value index of the initial technical antibody = 0.80×0.7+0.69×0.3=0.767.

[0035] In this embodiment, the comprehensive score of each initial technology antibody in the initial technology antibody candidate set is calculated based on the novelty score and the commercial value index. For example, the novelty score and commercial value index of the initial technology antibody corresponding to the above-mentioned novel lithium battery electrode design scheme are 0.323 and 0.767 respectively, so the comprehensive score of the initial technology antibody is , screen the top 10% of the initial technical antibodies in the comprehensive score as cloning mothers for cloning and amplification, generate multiple groups of clone copies, and guide the mutation of the clone copies based on lateral substitution mutation, cross-domain combination mutation and trend-guided mutation to obtain a candidate set of mutant technical antibodies. For example, the technical features corresponding to the initial technical antibody are that the lithium battery electrode uses a graphene material with a cost coefficient of 1.18 and an expansion rate of 4%. Based on lateral substitution mutation, three mutant technical antibodies are generated: the lithium battery electrode uses a carbon nanotube material with a cost coefficient of 1.10 and an expansion rate of 4.2%, the lithium battery electrode uses a silicon-carbon composite material with a cost coefficient of 1.05 and an expansion rate of 5.0%, and the lithium battery electrode uses a transition metal oxide material with a cost coefficient of 0.95 and an expansion rate of 6.5%. Based on cross-domain combination mutation, a mutant technical antibody is generated that adds a porosity dynamic control algorithm based on a gradient pore structure. Based on trend-guided mutation, a mutant technical antibody is generated that adjusts the pore conductive network distribution and adapts to the high nickel material system.

[0036] Step S4, performing three-property optimization screening on the candidate set of variant technology antibodies to obtain an immune technology antibody set, wherein the three-property optimization screening is represented by evaluating and screening the candidate set of variant technology antibodies for novelty, practicality, and creativity.

[0037] Specifically, the three-quality optimization screening includes novelty dimension screening, practicality dimension screening, and creativity dimension screening.

[0038] In this embodiment, the novelty dimension screening is represented by comparing the differences between the variant technical antibodies in the candidate set of variant technical antibodies and the initial technical antibodies, determining whether the variant technical antibodies break through the technical boundaries of the invention technical materials, and eliminating the variant technical antibodies that exceed the technical boundaries of the invention technical materials; In this embodiment, the practicality dimension screening is represented by verifying whether the variant technical antibodies in the candidate set of variant technical antibodies have technical feature vectors that deviate from the theoretical assumptions, and if so, the variant technical antibodies are eliminated.

[0039] In this embodiment, the creativity dimension screening is represented by screening the top ten variant technology antibodies in terms of commercial value index in the candidate set of variant technology antibodies.

[0040] Step S5: Based on GAN, simulate the review and verification of the immune technology antibody set to obtain the technology optimization strategy. The simulated review and verification is represented by an internal simulated technical review of the invention technology materials.

[0041] In this embodiment, the immune technology antibody set is input into GAN, and the creativity, practicality and novelty of the immune technology antibody set are evaluated based on GAN, a simulated examination opinion is generated, and a technical optimization strategy is generated based on the simulated examination opinion.

[0042] Step S6: Optimize the immune technology antibody set based on the technology optimization strategy and generate a corresponding intellectual property service strategy.

[0043] In this embodiment, the immune technology antibodies contained in the immune technology antibody set are optimized according to the technical optimization strategy, and the optimized immune technology antibodies are divided into service types. Based on the results of the service type division, a corresponding intellectual property service strategy is generated. Users can provide differentiated intellectual property services to customers based on the intellectual property service strategy.

[0044] Figure 2 A schematic diagram of an intelligent recommendation system for intellectual property services provided by some embodiments of the present application that can implement the ideas of the present application is shown. The following is a detailed introduction to the intelligent recommendation system for intellectual property services.

[0045] Specifically, an intellectual property service intelligent recommendation system includes: A collection module, which is used to obtain the invention technical materials submitted by the user and obtain the technical feature vectors corresponding to the invention technical materials; a processing module, the processing module being used to perform data preprocessing on the technical feature vector and construct a technical feature matrix based on the technical feature vector; A generation module, which is used to construct a human immune model and input a technical feature matrix into the human immune model to obtain an initial technical antibody candidate set based on the initial technical antibodies, wherein the initial technical antibodies represent the innovation points corresponding to the inventive technical materials; A cloning module, which is used to perform a cloning mutation operation on the initial technical antibody candidate set through a dual-channel evaluation to obtain a mutated technical antibody candidate set; A screening and review module is used to perform three-property optimization screening on the candidate set of variant technology antibodies to obtain an immune technology antibody set, and to perform simulation review and verification on the immune technology antibody set based on GAN to obtain a technology optimization strategy; The service module is used to optimize the immune technology antibody set according to the technology optimization strategy and generate a corresponding intellectual property service strategy.

[0046] The specific usage and function of this embodiment are described below: First, feature extraction is performed on the invention technical materials submitted by the user to obtain technical feature vectors, and data preprocessing is performed on the technical feature vectors. A technical feature matrix is ​​constructed based on the technical feature vectors. Then, a human immune model is constructed, and the technical feature matrix is ​​input into the human immune model as an antigen to generate initial technical antibodies. The invention technical materials submitted by the user are more comprehensively matched with the technical features associated with them. An initial technical antibody candidate set is obtained based on the initial technical antibodies. Then, a clone mutation operation is performed on the initial technical antibody candidate set based on a dual-channel evaluation to further expand the technical features associated with the invention technical materials, making the user's invention more novel, practical and creative, and obtaining a variant technical antibody candidate set. Then, the variant technical antibody candidate set is subjected to three-property optimization screening to eliminate variant technical antibodies with too low correlation with the user's invention materials, and obtain an immune technical antibody set. Next, the immune technical antibody set is simulated and reviewed and verified based on GAN, and the technical features corresponding to the generated immune technical antibodies are verified to ensure the rationality and feasibility of the generated immune technical antibodies, obtain a technical optimization strategy, and finally optimize the immune technical antibody set based on the technical optimization strategy to generate a corresponding intellectual property service strategy.

[0047] Figure 3 1 is a schematic diagram of an electronic device provided by an embodiment of the present invention. The electronic device is described in detail below.

[0048] An embodiment of the present invention further provides an electronic device, including: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method proposed in the embodiment of the present invention.

[0049] The following is a detailed introduction to the various components of electronic equipment: The term "processor" is the control center of an electronic device and can be a single processor or a collective term for multiple processing elements. For example, the processor can be one or more central processing units (CPUs), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the first embodiment of the present invention, such as one or more digital signal processors (DSPs) or one or more field programmable gate arrays (FPGAs).

[0050] The processor can execute various functions of the electronic device by running or executing software programs stored in the memory and calling data stored in the memory.

[0051] The memory is used to store the software program for executing the solution of the present invention, and the execution is controlled by the processor. The specific implementation method can refer to the above method embodiment and will not be repeated here.

[0052] The memory may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, a random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only (CD-ROM), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto. The memory may be integrated with the processor or exist independently and be coupled to the processor via an interface circuit of the electronic device, and this is not specifically limited in the embodiments of the present invention.

[0053] The above embodiments can be implemented in whole or in part via software, hardware (e.g., circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When loaded or executed on a computer, the processes or functions described in accordance with the embodiments of the present invention are fully or partially performed. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wireless communication (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium accessible by a computer, or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0054] It should be understood that the term "and / or" as used herein simply describes an association between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent the existence of A alone, the existence of both A and B, or the existence of B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the related objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0055] It should be understood that in the embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0056] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A method for intelligent recommendation of intellectual property services, characterized in that: The method comprises: Extract features from the invention technical materials submitted by users, obtain technical feature vectors, perform data preprocessing on the technical feature vectors, and construct a technical feature matrix based on the technical feature vectors; Constructing a human immune model, inputting the technical feature matrix as an antigen into the human immune model, generating initial technical antibodies, and obtaining an initial technical antibody candidate set based on the initial technical antibodies, wherein the initial technical antibodies represent the innovation points corresponding to the inventive technical materials; Performing a cloning mutation operation on the initial technical antibody candidate set based on a dual-channel evaluation to obtain a mutated technical antibody candidate set. The dual-channel evaluation includes a technical channel and a commercial channel. The technical channel is represented by calculating the novelty score of the invention technical materials submitted by the user, and the commercial channel is represented by calculating the commercial value index of the invention technical materials submitted by the user; Performing three-property optimization screening on the candidate set of variant technology antibodies to obtain an immune technology antibody set, wherein the three-property optimization screening is performed on the candidate set of variant technology antibodies to evaluate novelty, practicality, and creativity; Conducting simulated review and verification of the immune technology antibody set based on GAN to obtain technology optimization strategies. The simulated review and verification is represented by conducting internal simulated technical review of the invention technical materials. Based on the technical optimization strategy, the immune technology antibody set is optimized to generate the corresponding intellectual property service strategy.

2. The method for intelligent recommendation of intellectual property services according to claim 1, characterized in that: The method of constructing a human immune model, inputting the technical feature matrix as an antigen into the human immune model, generating initial technical antibodies, and obtaining an initial technical antibody candidate set based on the initial technical antibodies includes: Obtain a preset antibody library, perform antibody matching on the antibody library based on the input antigen, obtain initial technical antibodies with similarity, and output the top 30 initial technical antibodies with similarity as the initial technical antibody candidate set.

3. The method for intelligent recommendation of intellectual property services according to claim 1, characterized in that: The step of performing a cloning mutation operation on the initial technical antibody candidate set based on the dual-channel evaluation to obtain a mutated technical antibody candidate set includes: The technology channel compares existing patents with similar technology feature vectors in an external patent search library in real time, calculates the difference between the initial technology antibody and the existing patent based on the Jaccard-TFIDF algorithm, obtains the difference result, introduces the time attenuation factor into the difference result, and generates a novelty score; The commercial channel constructs a dual-dimensional index system of practicality and creativity, which includes practicality dimension and creativity dimension, calculates the practicality score and creativity score of the invention technical material, and synthesizes the practicality score and creativity score to obtain a commercial value index; Based on the novelty score and commercial value index, a comprehensive score of each initial technology antibody in the initial technology antibody candidate set is calculated. The initial technology antibodies ranked in the top 10% in the comprehensive score are selected as cloning mothers for cloning and amplification to generate multiple groups of clone copies. The clone copies are mutated based on horizontal substitution mutation, cross-domain combination mutation and trend-guided mutation to obtain a candidate set of mutated technology antibodies.

4. The method for intelligent recommendation of intellectual property services according to claim 3, characterized in that: The method further comprises: The practicality dimension includes two dimensional attributes: implementation feasibility and cost controllability; The creativity dimension includes two dimensional attributes: technological breakthrough and market exclusivity.

5. The method for intelligent recommendation of intellectual property services according to claim 1, characterized in that: The method of performing three-property optimization screening on the candidate set of variant technology antibodies to obtain the immune technology antibody set includes: The three-dimensional optimization screening includes novelty dimension screening, practicality dimension screening and creativity dimension screening; The novelty dimension screening is represented by comparing the differences between the variant technical antibodies in the candidate set of variant technical antibodies and the initial technical antibodies, determining whether the variant technical antibodies break through the technical boundaries of the invention technical materials, and eliminating the variant technical antibodies that exceed the technical boundaries of the invention technical materials; The practicality dimension screening is to verify whether there are technical feature vectors of variant technical antibodies in the candidate set of variant technical antibodies that deviate from the theoretical assumptions, and if so, the variant technical antibodies are eliminated; The creativity dimension screening is represented by screening the top ten variant technology antibodies ranked by commercial value index in the candidate set of variant technology antibodies.

6. The method for intelligent recommendation of intellectual property services according to claim 1, characterized in that: The GAN-based simulation review and verification of the immune technology antibody set to obtain technology optimization strategies includes: The immune technology antibody set is input into GAN, and the creativity, practicality and novelty of the immune technology antibody set are evaluated based on GAN, simulated examination opinions are generated, and technical optimization strategies are generated based on the simulated examination opinions.

7. The method for intelligent recommendation of intellectual property services according to claim 1, characterized in that: The technology optimization strategy is based on which the immune technology antibody set is optimized to generate the corresponding intellectual property service strategy, including: The immune technology antibodies contained in the immune technology antibody set are optimized according to the technical optimization strategy, the optimized immune technology antibodies are classified into service types, the service type classification results are obtained, and the corresponding intellectual property service strategy is generated based on the service type classification results. Users can provide customers with differentiated intellectual property services based on the intellectual property service strategy.

8. The method for intelligent recommendation of intellectual property services according to claim 1, characterized in that: The feature extraction of the invention technical materials submitted by the user to obtain the technical feature vector, and the data preprocessing of the technical feature vector, and the construction of the technical feature matrix based on the technical feature vector include: Feature extraction is performed on the text content in user-submitted invention and technical materials. Synonym conflicts within the text content are eliminated, and a "problem, solution, effect" triple is constructed. Drawing content in the invention and technical materials is converted into a vector structure, and geometric parameters in the drawing content are extracted to establish a structure-function mapping relationship. Table content in the technical materials is structured and numerically normalized, quantifying unstructured parameters into standard parameters. The "problem, solution, effect" triples, the structure-function mapping relationship, and the standard parameters are converted into multiple technical feature vectors, and the generated multiple technical feature vectors are mapped to the same feature space. A technical feature matrix is ​​constructed based on the above technical feature vectors.

9. An intellectual property service intelligent recommendation system, characterized in that: include: A collection module, which is used to obtain the invention technical materials submitted by the user and obtain the technical feature vectors corresponding to the invention technical materials; a processing module, the processing module being used to perform data preprocessing on the technical feature vector and construct a technical feature matrix based on the technical feature vector; A generation module, which is used to construct a human immune model and input a technical feature matrix into the human immune model to obtain an initial technical antibody candidate set based on the initial technical antibodies, wherein the initial technical antibodies represent the innovation points corresponding to the inventive technical materials; A cloning module, which is used to perform a cloning mutation operation on the initial technical antibody candidate set through a dual-channel evaluation to obtain a mutated technical antibody candidate set; A screening and review module is used to perform three-property optimization screening on the candidate set of variant technology antibodies to obtain an immune technology antibody set, and to perform simulation review and verification on the immune technology antibody set based on GAN to obtain a technology optimization strategy; The service module is used to optimize the immune technology antibody set according to the technology optimization strategy and generate a corresponding intellectual property service strategy.

10. An electronic device, characterized in that: include: at least one processor; And, a memory communicatively connected to at least one processor; wherein the memory stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor to enable the at least one processor to perform the method according to any one of claims 1-8.

Images