Scientific and technological achievement transformation intelligent aid decision-making system based on large model

Abstract

The invention relates to the technical field of scientific and technological achievement conversion, and particularly discloses a scientific and technological achievement conversion intelligent aid decision-making system based on a large model, which obtains innovation subjects, innovation activities, innovation resources and industry / technical fields input by a user and inputs the innovation subjects, the innovation activities, the innovation resources and the industry / technical fields into a scientific and technological achievement identifier to obtain scientific and technological achievement type labels, and meanwhile, sends the scientific and technological achievement type labels to the scientific and technological achievement identifier. Determining factors and influence factors input by a user are obtained, then the scientific and technological achievement type labels, the determining factors and the influence factors are input into a scientific and technological achievement conversion decision-making module based on a large model to obtain a scientific and technological achievement conversion auxiliary decision-making result, and the scientific and technological achievement conversion auxiliary decision-making result is a standard element conforming to the scientific and technological achievements. Therefore, the related information of the scientific and technological achievements can be comprehensively integrated, and the scientific and reasonable auxiliary decision-making result is generated in combination with the standard elements, so that the decision-making accuracy and efficiency are improved, and the compliance and market adaptability of the scientific and technological achievements are ensured.

Description

Technical Field

[0001] The present application relates to the technical field of transformation of scientific and technological achievements, and more specifically, to an intelligent auxiliary decision-making system for transformation of scientific and technological achievements based on a large model. Background Art

[0002] In today's era of technological innovation-driven development, the transformation of scientific and technological achievements plays a vital role in promoting economic growth, enhancing industrial competitiveness and promoting social progress. The transformation of scientific and technological achievements refers to the process of transforming scientific research results into practical applications, which involves multiple complex links and factors.

[0003] Although the existing scientific and technological achievement transformation system can help users evaluate the transformation potential of scientific and technological achievements to a certain extent, it still has significant shortcomings. On the one hand, the successful transformation of scientific and technological achievements not only depends on their innovation and technological advancement, but also requires comprehensive consideration of market demand, financial support, social collaboration and other factors. The existing system is usually based on simple analysis and can only process information of some dimensions. It is difficult to comprehensively evaluate the various attributes of scientific and technological achievements and their interrelationships, resulting in inaccurate decision-making. On the other hand, the existing system often ignores whether scientific and technological achievements comply with various specifications such as international standards, national standards, and industry standards, which may cause risks and market access barriers, affecting the actual application and promotion of scientific and technological achievements.

[0004] Therefore, an optimized decision-making support scheme for the transformation of scientific and technological achievements is desired. Summary of the invention The present application provides an intelligent decision-making support system for the transformation of scientific and technological achievements based on a large model, which can comprehensively integrate relevant information of scientific and technological achievements, and generate scientific and reasonable decision-making support results in combination with standard elements, which not only improves the accuracy and efficiency of decision-making, but also ensures the compliance and market adaptability of scientific and technological achievements.

[0005] According to one aspect of the present application, there is provided an intelligent decision-making support system for the transformation of scientific and technological achievements based on a large model, including: a user input data acquisition module, used to acquire the innovation subject, innovation activity, innovation resource and industry / technical field input by the user; a scientific and technological achievement type identification module, used to input the innovation subject, the innovation activity, the innovation resource and the industry / technical field into a scientific and technological achievement identifier to obtain a scientific and technological achievement type label; a determining factor and influencing factor acquisition module, used to acquire the determining factors and influencing factors input by the user, the determining factors include the transformation subject, the transformation method, the transformation scope and the transformation object, and the influencing factors include the government-side influencing factors, the market-side influencing factors and the social-side influencing factors; a scientific and technological achievement transformation decision-making support module, used to input the scientific and technological achievement type label, the determining factors and the influencing factors into a scientific and technological achievement transformation decision-making module based on a large model to obtain a scientific and technological achievement transformation decision-making support result, wherein the scientific and technological achievement transformation decision-making support result is the standard element that the scientific and technological achievement meets.

[0006] In the above-mentioned intelligent decision-making support system for the transformation of scientific and technological achievements based on a big model, the government-side influencing factors include a text description of policy intensity and a text description of governance level, the market-side influencing factors include a text description of technological progress and a text description of product competition, and the social-side influencing factors include a text description of intermediary services and a text description of multi-party collaboration.

[0007] In the above-mentioned intelligent decision-making support system for transformation of scientific and technological achievements based on large models, the standard elements include international standards, national standards, industry standards, local standards, group standards and enterprise standards.

[0008] In the above-mentioned intelligent auxiliary decision-making system for the transformation of scientific and technological achievements based on a large model, the scientific and technological achievement type identification module includes: a semantic encoding unit for key information on the transformation of scientific and technological achievements, which is used to use a semantic encoder based on the Bert model to semantically encode the innovation subject, the innovation activity, the innovation resource and the industry / technical field to obtain a semantic encoding vector of key information on the transformation of scientific and technological achievements; and a scientific and technological achievement type label determination unit, which is used to input the semantic encoding vector of key information on the transformation of scientific and technological achievements into a scientific and technological achievement identifier based on a classifier to obtain the scientific and technological achievement type label.

[0009] In the above-mentioned intelligent auxiliary decision-making system for transformation of scientific and technological achievements based on a large model, the auxiliary decision-making module for transformation of scientific and technological achievements includes: a scientific and technological achievement type label unique-hot encoding unit, which is used to perform unique-hot encoding on the scientific and technological achievement type label to obtain a unique-hot encoding vector of the scientific and technological achievement type label; a determinant-influencing factor semantic embedding encoding unit, which is used to perform semantic embedding encoding on the determinant and the influencing factor to obtain a set of determinant sentence granularity semantic embedding encoding vectors and a set of influencing factor sentence granularity semantic embedding encoding vectors; a related factor set merging unit, which is used to perform set merging on the set of the determinant sentence granularity semantic embedding encoding vectors and the set of the influencing factor sentence granularity semantic embedding encoding vectors to obtain a set of related factor sentence granularity semantic embedding encoding vectors; a scientific and technological achievement-related factor interaction unit, which is used to perform scientific and technological achievement-related factor decision anchoring semantic deep interaction analysis on the set of the scientific and technological achievement type label unique-hot encoding vector and the set of the related factor sentence granularity semantic embedding encoding vector to obtain a scientific and technological achievement transformation semantic response encoding vector; a scientific and technological achievement transformation auxiliary decision-making result generation unit, which is used to obtain the scientific and technological achievement transformation auxiliary decision-making result based on the scientific and technological achievement transformation semantic response encoding vector.

[0010] In the above-mentioned intelligent auxiliary decision-making system for the transformation of scientific and technological achievements based on a large model, the scientific and technological achievement-related factor interaction unit includes: a scientific and technological achievement-related factor semantic response subunit, which is used to calculate the semantic response decision anchoring features between the scientific and technological achievement type label unique hot encoding vector and each relevant factor sentence granularity semantic embedding encoding vector in the set of relevant factor sentence granularity semantic embedding encoding vectors to obtain a set of scientific and technological achievement-related factor local feature response anchor encoding matrices; a scientific and technological achievement-related factor splicing factor calculation subunit, which is used to determine the scientific and technological achievement-related factor local feature response anchor encoding matrix based on the feature distribution of each scientific and technological achievement-related factor local feature response anchor encoding matrix in the set of scientific and technological achievement-related factor local feature response anchor encoding matrix. The decision anchor adaptive splicing factors of each scientific and technological achievement-related factor local feature response anchor coding matrix are used to obtain a set of scientific and technological achievement-related factor decision anchor adaptive splicing factors; the scientific and technological achievement-related factor weighting processing subunit is used to perform weighting processing on the set of scientific and technological achievement-related factor decision anchor adaptive splicing factors based on the Softmax function to obtain a set of scientific and technological achievement-related factor decision anchor adaptive splicing weight factors; the scientific and technological achievement-transformation feature fusion subunit is used to fuse the set of scientific and technological achievement-related factor local feature response anchor coding matrices based on the set of scientific and technological achievement-related factor decision anchor adaptive splicing weight factors to obtain the scientific and technological achievement transformation semantic response coding vector.

[0011] In the above-mentioned intelligent auxiliary decision-making system for the transformation of scientific and technological achievements based on a large model, the scientific and technological achievement-related factor semantic response subunit is used to: perform deep implicit feature extraction based on fully connected coding on the one-hot encoding vector of the scientific and technological achievement type label to obtain the scientific and technological achievement type label deep implicit encoding vector; perform deep implicit feature extraction based on fully connected coding on each relevant factor sentence granularity semantic embedding encoding vector in the set of relevant factor sentence granularity semantic embedding encoding vectors to obtain a set of relevant factor sentence granularity semantic deep implicit encoding vectors; perform semantic response decision anchoring on each relevant factor sentence granularity semantic deep implicit encoding vector in the set of the scientific and technological achievement type label deep implicit encoding vector and the relevant factor sentence granularity semantic deep implicit encoding vector to obtain a set of the scientific and technological achievement-related factor local feature response anchor encoding matrices.

[0012] In the above-mentioned intelligent auxiliary decision-making system for the transformation of scientific and technological achievements based on a large model, the scientific and technological achievement-related factor splicing factor calculation subunit is used to: obtain the scientific and technological achievement-related factor decision anchor adaptive splicing factor corresponding to the scientific and technological achievement-related factor local feature response anchor coding matrix based on the maximum value, mean, variance and difference magnification coefficient of the scientific and technological achievement-related factor local feature response anchor coding matrix and the scientific and technological achievement-related factor drift coefficient; wherein the difference magnification coefficient is the number of eigenvalues ​​of the scientific and technological achievement-related factor local feature response anchor coding matrix; the scientific and technological achievement-related factor drift coefficient is obtained by performing weak to strong interpretable generalization processing on the scientific and technological achievement-related factor local feature response anchor coding matrix.

[0013] In the above-mentioned intelligent auxiliary decision-making system for the transformation of scientific and technological achievements based on a large model, the scientific and technological achievement transformation auxiliary decision-making result generation unit is used to: input the scientific and technological achievement transformation semantic response coding vector into a classifier-based transformation decision engine to obtain the scientific and technological achievement transformation auxiliary decision-making result.

[0014] The application provides an intelligent decision-making support system for the transformation of scientific and technological achievements based on a large model, which obtains the innovation subject, innovation activity, innovation resource and industry / technical field input by the user and inputs them into the scientific and technological achievement identifier to obtain the scientific and technological achievement type label, and at the same time, obtains the determining factors and influencing factors input by the user, and then inputs the scientific and technological achievement type label, determining factors and influencing factors into the scientific and technological achievement transformation decision-making module based on the large model to obtain the scientific and technological achievement transformation decision-making support result, and the scientific and technological achievement transformation decision-making support result is the standard element that the scientific and technological achievement meets. In this way, the relevant information of the scientific and technological achievements can be fully integrated, and scientific and reasonable decision-making support results can be generated in combination with the standard elements, which not only improves the accuracy and efficiency of decision-making, but also ensures the compliance and market adaptability of scientific and technological achievements. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0016] Figure 1 This is a schematic block diagram of an intelligent decision-making support system for transformation of scientific and technological achievements based on deep learning and a large model according to an embodiment of the present application.

[0017] Figure 2 This is a data flow diagram of the deep learning-based and large-model-based intelligent decision-making support system for the transformation of scientific and technological achievements in an embodiment of the present application.

[0018] Figure 3 This is a schematic block diagram of a scientific and technological achievement type identification module in a deep learning-based and large-model-based intelligent auxiliary decision-making system for scientific and technological achievement transformation in an embodiment of the present application.

[0019] Figure 4 This is a schematic block diagram of a scientific and technological achievement transformation decision-making assistance module in a deep learning-based and large-model-based intelligent decision-making assistance system for scientific and technological achievement transformation in an embodiment of the present application.

[0020] Figure 5 This is a schematic block diagram of the scientific and technological achievement-related factor interaction unit in the deep learning-based and large-model-based intelligent auxiliary decision-making system for the transformation of scientific and technological achievements in an embodiment of the present application. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0022] Based on this, this application proposes an intelligent decision-making support system for the transformation of scientific and technological achievements based on a large model, which can comprehensively integrate relevant information of scientific and technological achievements, and combine standard elements to generate scientific and reasonable decision-making support results, which not only improves the accuracy and efficiency of decision-making, but also ensures the compliance and market adaptability of scientific and technological achievements. Figure 1 This is a schematic block diagram of an intelligent decision-making support system for transformation of scientific and technological achievements based on deep learning and a large model according to an embodiment of the present application. Figure 2 This is a data flow diagram of the intelligent decision-making support system for the transformation of scientific and technological achievements based on deep learning and large models in the embodiment of the present application. Figure 1 and Figure 2As shown, the intelligent auxiliary decision-making system for the transformation of scientific and technological achievements based on a large model includes: a user input data acquisition module 10, which is used to obtain the innovation subject, innovation activity, innovation resource and industry / technical field input by the user; a scientific and technological achievement type identification module 20, which is used to input the innovation subject, the innovation activity, the innovation resource and the industry / technical field into the scientific and technological achievement identifier to obtain a scientific and technological achievement type label; a determining factor and influencing factor acquisition module 30, which is used to obtain the determining factors and influencing factors input by the user, the determining factors include the transformation subject, the transformation method, the transformation scope and the transformation object, and the influencing factors include the government-side influencing factors, the market-side influencing factors and the social-side influencing factors; a scientific and technological achievement transformation auxiliary decision-making module 40, which is used to input the scientific and technological achievement type label, the determining factors and the influencing factors into the scientific and technological achievement transformation decision-making module based on the large model to obtain the scientific and technological achievement transformation auxiliary decision-making result, and the scientific and technological achievement transformation auxiliary decision-making result is the standard element that the scientific and technological achievement meets.

[0023] Exemplarily, in the user input data acquisition module 10, the innovation subject, innovation activity, innovation resource and industry / technical field input by the user are obtained. It should be understood that in the process of transforming scientific and technological achievements into practical applications, it is crucial to obtain the information of the innovation subject, innovation activity, innovation resource and industry / technical field input by the user. These data not only help the system to accurately identify the type of scientific and technological achievements, but also provide the necessary background and basis for subsequent analysis. Specifically, the innovation subject defines the creator or owner of the scientific and technological achievements, which can be a university, enterprise or scientific research institution, etc., and its nature and ability directly affect the application orientation and transformation potential of the achievements; the innovation activity indicates the research and development direction of the achievements, covering basic research, applied research and experimental development and other aspects, which helps to determine the technological advancement and market demand matching of scientific and technological achievements; the innovation resources reflect the funds, equipment and other material conditions invested in the research and development process, which together determine the scale and technical content of scientific and technological achievements; and the industry / technical field defines the specific category to which the scientific and technological achievements belong, and scientific and technological achievements in different fields have different characteristics and evaluation criteria.

[0024] In a specific embodiment, a user-friendly interface can be designed to allow users to easily enter relevant data. This may include filling out online forms, uploading documents, or directly entering text descriptions in dialog boxes. It is important to ensure that the collected data is accurate and as detailed and comprehensive as possible to facilitate subsequent processing and analysis. For example, when filling out information about "innovation subjects", users may need to provide the name, type (such as colleges and universities, enterprises, etc.), main research direction, etc. of the subject; for "innovation activities", it may be necessary to explain the specific content, goals and stage of the project; as for "innovation resources", it involves details such as funding sources, budget arrangements, and equipment or materials used; finally, "industry / technical field" requires clear indication of which industry or technical field the scientific and technological achievements belong to, such as information technology, bioengineering, etc.

[0025] Exemplarily, in the scientific and technological achievement type identification module 20, the innovation subject, the innovation activity, the innovation resource and the industry / technical field are input into the scientific and technological achievement identifier to obtain the scientific and technological achievement type label. It should be understood that the accurate type of scientific and technological achievements is helpful to provide scientific decision-making support, optimize resource allocation, guide transformation strategies, and provide clear development directions and actionable insights for subjects such as universities, enterprises, and institutes, thereby accelerating the commercialization process of scientific and technological achievements and enhancing market competitiveness. However, the type of scientific and technological achievements is not determined by a single factor, but the result of the interaction of multiple factors such as innovation subjects, innovation activities, innovation resources and industry / technical fields. Specifically, the nature and ability of the innovation subject affect the application orientation and transformation ability of the results. Enterprises tend to apply results, and universities and institutes may focus more on theoretical or cutting-edge technological achievements. The content of the innovation activity directly reflects the research and development direction of the results, whether it is basic research, applied research or experimental development, which determines the nature of the results. The investment of innovation resources, such as the amount of funds, the advanced level of equipment, etc., will affect the scale and technical content of the results. The industry / technical field limits the scope of the results, and the results in different fields have different characteristics and evaluation criteria.

[0026] In one embodiment, Figure 3 This is a schematic block diagram of a scientific and technological achievement type identification module in a scientific and technological achievement transformation intelligent auxiliary decision-making system based on deep learning and a large model in an embodiment of the present application. Figure 3As shown, the scientific and technological achievement type identification module 20 includes: a scientific and technological achievement conversion key information semantic encoding unit 21, which is used to use a semantic encoder based on the Bert model to semantically encode the innovation subject, the innovation activity, the innovation resource and the industry / technical field to obtain a semantic encoding vector of the scientific and technological achievement conversion key information; a scientific and technological achievement type label determination unit 22, which is used to input the scientific and technological achievement conversion key information semantic encoding vector into a classifier-based scientific and technological achievement identifier to obtain the scientific and technological achievement type label.

[0027] Exemplarily, in the semantic encoding unit 21 of key information for conversion of scientific and technological achievements, a semantic encoder based on the Bert model is used to semantically encode the innovation subject, the innovation activity, the innovation resources and the industry / technical field to utilize the bidirectional Transformer architecture of the Bert model, which can simultaneously consider the preceding and following information of the text, thereby deeply understanding the contextual semantics between each text and obtaining the semantic encoding vector of key information for conversion of scientific and technological achievements.

[0028] Exemplarily, in the scientific and technological achievement type label determination unit 22, the semantic coding vector of the key information of the scientific and technological achievement conversion is input into the scientific and technological achievement identifier based on the classifier to capture the characteristic pattern of the corresponding category by using the powerful climbing discrimination ability of the classifier, and obtain the scientific and technological achievement type label. Among them, the classifier has the ability to extract the characteristic pattern related to the type of scientific and technological achievement from the complex semantic coding vector by learning a large amount of training data. It should be understood that the scientific and technological achievement type label is an identifier for classifying scientific and technological achievements, which can help us better understand the essential characteristics of scientific and technological achievements and their applicable scenarios. For example, the scientific and technological achievement type label may include theoretical technology, common technology, proprietary technology and general technology. Among them, theoretical technology mainly refers to the results of basic scientific research, which often have high academic value but are far from practical application; common technology is the core technology that can be widely used in multiple fields, and this type of technology is usually of great significance to promoting industry progress; proprietary technology refers to the unique technical achievements in a specific field, which often have high commercial value and technical barriers; and general technology covers more common technical achievements, which are widely used in daily production and life.

[0029] Exemplarily, in the determinant influencing factor acquisition module 30, the determinant and influencing factors input by the user are obtained, the determinant includes the transformation subject, transformation method, transformation scope and transformation object, and the influencing factors include government-side influencing factors, market-side influencing factors and social-side influencing factors. In one embodiment, the government-side influencing factors include a text description of policy intensity and a text description of governance level, the market-side influencing factors include a text description of technological progress and a text description of product competition, and the social-side influencing factors include a text description of intermediary services and a text description of multi-party collaboration. It should be understood that obtaining the determinant and influencing factors input by the user, including the transformation subject, transformation method, transformation scope and transformation object, as well as the government-side influencing factors, market-side influencing factors and social-side influencing factors, is crucial for the transformation process of scientific and technological achievements. This information not only helps the system to fully understand the specific situation of scientific and technological achievements in practical applications, but also provides an important basis for formulating scientific and reasonable auxiliary decision-making results.

[0030] First of all, the determining factors are the key elements that directly affect whether scientific and technological achievements can be successfully transformed. The subject of transformation refers to the specific organization or individual responsible for promoting the transformation of scientific and technological achievements into actual products or services, such as enterprises, universities or scientific research institutions. Different transformation subjects have different resources and capabilities, which will significantly affect the effect and efficiency of the transformation. The transformation method refers to the specific path of how scientific and technological achievements go from the laboratory to the market, such as technology transfer, cooperative development or independent entrepreneurship. Different transformation methods are suitable for different types of scientific and technological achievements, and have different requirements for resource allocation and time cycles. The scope of transformation describes the application field of scientific and technological achievements and their potential market coverage, and clarifies their applicability and market demand. The object of transformation is the target group or industry for the final application of scientific and technological achievements, which determines the actual application scenario of scientific and technological achievements and the matching degree of market demand.

[0031] Secondly, the influencing factors cover the impact of the external environment on the transformation of scientific and technological achievements. Government-side influencing factors include policy strength and governance level, such as tax incentives and subsidy policies. These policy measures can directly promote or hinder the commercialization process of scientific and technological achievements. Market-side influencing factors involve technological progress and product competition. Factors such as the market demand for new technologies and the risk of technological substitution will affect the market acceptance and commercial value of scientific and technological achievements. Social-side influencing factors include intermediary services and multi-party collaboration. Good intermediary services can accelerate the transformation of scientific and technological achievements, while multi-party collaboration can help integrate resources from all parties and improve the success rate of transformation.

[0032] Exemplarily, in the scientific and technological achievement transformation auxiliary decision module 40, the scientific and technological achievement type label, the determining factors and the influencing factors are input into the scientific and technological achievement transformation decision module based on the large model to obtain the scientific and technological achievement transformation auxiliary decision result, and the scientific and technological achievement transformation auxiliary decision result is the standard element that the scientific and technological achievement meets. In one embodiment, the standard elements include international standards, national standards, industry standards, local standards, group standards and enterprise standards. It should be understood that by following various standard elements, it is possible to ensure the universality and compatibility of scientific and technological achievements in different scopes and levels around the world and domestically, ensure quality and safety, regulate market order, promote technical exchanges, cooperation and innovation, and enhance competitiveness, thereby achieving efficient transformation and wide application.

[0033] Based on this, in the technology achievement transformation auxiliary decision module, the technical concept of this application is to use data analysis and processing technology based on large models and artificial intelligence to perform unique-hot encoding on the technology achievement type label, and at the same time, semantically embed the determining factors and the influencing factors and merge the relevant factor sentence granularity, so as to automatically generate the technology achievement transformation auxiliary decision result according to the decision anchor semantic response representation between the unique-hot encoding features of the technology achievement type label and the semantic embedding features of the relevant factor sentence granularity. In this way, the various attributes of the technology achievement and its related internal and external factors can be fully integrated, not only the basic classification information of the technology achievement is captured, but also its complex semantic relationship is deeply understood, and the transformation potential of the technology achievement can be comprehensively evaluated in multiple dimensions, providing more accurate auxiliary decision results.

[0034] In one embodiment, Figure 4 This is a schematic block diagram of a scientific and technological achievement transformation auxiliary decision-making module in a scientific and technological achievement transformation intelligent auxiliary decision-making system based on deep learning and a large model according to an embodiment of the present application. Figure 4As shown, the scientific and technological achievement transformation auxiliary decision module 40 includes: a scientific and technological achievement type label unique hot encoding unit 41, which is used to perform unique hot encoding on the scientific and technological achievement type label to obtain a scientific and technological achievement type label unique hot encoding vector; a determining factor and influencing factor semantic embedding encoding unit 42, which is used to perform semantic embedding encoding on the determining factor and the influencing factor to obtain a set of determining factor sentence granularity semantic embedding encoding vectors and a set of influencing factor sentence granularity semantic embedding encoding vectors; a related factor set merging unit 43, which is used to perform set merging on the set of determining factor sentence granularity semantic embedding encoding vectors and the set of influencing factor sentence granularity semantic embedding encoding vectors to obtain a set of related factor sentence granularity semantic embedding encoding vectors; a scientific and technological achievement-related factor interaction unit 44, which is used to perform scientific and technological achievement-related factor decision anchoring semantic deep interaction analysis on the set of the scientific and technological achievement type label unique hot encoding vector and the set of the related factor sentence granularity semantic embedding encoding vector to obtain a scientific and technological achievement transformation semantic response encoding vector; a scientific and technological achievement transformation auxiliary decision result generation unit 45, which is used to obtain the scientific and technological achievement transformation auxiliary decision result based on the scientific and technological achievement transformation semantic response encoding vector.

[0035] Exemplarily, in the scientific and technological achievement type label unique hot encoding unit 41, the scientific and technological achievement type label is unique hot encoded to obtain a scientific and technological achievement type label unique hot encoding vector. It should be understood that considering that the scientific and technological achievement type label is usually discrete category data, such as theoretical technology, common technology, proprietary technology, general technology, etc. Therefore, in order to convert this discrete category information into a numerical vector form that is easy for the model to understand and process, so that the model can effectively analyze and calculate it, the present application performs unique hot encoding on the scientific and technological achievement type label to obtain a scientific and technological achievement type label unique hot encoding vector. That is, unique hot encoding maps each category to a unique vector, with only one position in the vector being 1 and the rest being 0, so that various types of scientific and technological achievements can be clearly distinguished, and subsequent processing of the model provides a clear and clear feature representation.

[0036] Exemplarily, in the determinant-influencing factor semantic embedding coding unit 42, the determinant and the influencing factor are semantically embedded to obtain a set of determinant sentence granularity semantic embedding coding vectors and a set of influencing factor sentence granularity semantic embedding coding vectors. It should be understood that, considering that the determinant and the influencing factor usually contain a large number of natural language descriptions, these descriptions not only contain factual information, but also contain rich semantic relationships. Based on this, in order to capture and refine the semantic information between the contexts in each factor description sentence, in the technical solution of the present application, the determinant and the influencing factor are semantically embedded to obtain a set of determinant sentence granularity semantic embedding coding vectors and a set of influencing factor sentence granularity semantic embedding coding vectors. In this way, the semantic characteristics of each factor can be accurately reflected, thereby providing an accurate basis for generating auxiliary decision results. In particular, in a specific example of the present application, the complex semantic information in the text can be learned and represented by using a semantic encoder including an embedding layer to semantically encode and process each of the determinants and the influencing factors, so that the model can understand the true meaning of the text and obtain a set of determinant sentence-granularity semantic embedding coding vectors and a set of influencing factor sentence-granularity semantic embedding coding vectors with rich semantic representation capabilities.

[0037] Exemplarily, in the relevant factor set merging unit 43, the set of the sentence granularity semantic embedding coding vectors of the determining factors and the set of the sentence granularity semantic embedding coding vectors of the influencing factors are set merged to obtain a set of sentence granularity semantic embedding coding vectors of the relevant factors. It should be understood that the determining factors and influencing factors have effects on the transformation of scientific and technological achievements from different angles. Determining factors such as transformation subjects, methods, scopes and objects directly define the specific context of the transformation of scientific and technological achievements; influencing factors include various factors on the market and social ends, which shape the external environment of transformation. Therefore, in order to integrate all information related to the transformation of scientific and technological achievements, provide a comprehensive data basis for subsequent analysis, and avoid the one-sided understanding of the transformation of scientific and technological achievements due to information dispersion, this application integrates the information of different dimensions into a unified vector representation by merging the set of the sentence granularity semantic embedding coding vectors of the determining factors and the set of the sentence granularity semantic embedding coding vectors of the influencing factors, so as to obtain a set of sentence granularity semantic embedding coding vectors of the relevant factors, so that the model can consider multiple factors at the same time and provide a more comprehensive understanding and analysis.

[0038] Exemplarily, in the scientific and technological achievement-related factor interaction unit 44, the scientific and technological achievement-related factor decision anchoring semantic deep interaction analysis is performed on the set of the scientific and technological achievement type label unique hot encoding vector and the related factor sentence granularity semantic embedding encoding vector to obtain the scientific and technological achievement transformation semantic response encoding vector. It should be understood that considering that the transformation of scientific and technological achievements is a complex process affected by the combined effects of multiple factors, there is a close and hidden complex relationship between the scientific and technological achievement type label and many related factors (determining factors and influencing factors). Different types of scientific and technological achievements have different responses and reliance on various related factors in the transformation process. For example, basic research scientific and technological achievements may rely more on the professional capabilities of university scientific research teams, while applied technology achievements may pay more attention to market demand, technology transformation capabilities and product competition. Based on this, in order to dig out the deep semantic association between the type of scientific and technological achievements and related factors, and comprehensively reflect the comprehensive impact of these factors on the transformation of scientific and technological achievements, this application conducts a scientific and technological achievement-related factor decision anchoring semantic deep interactive analysis on the set of the scientific and technological achievement type label one-hot encoding vector and the related factor sentence granularity semantic embedding encoding vector to accurately capture the overall and detailed relationship, comprehensively and accurately reflect the overall picture of the transformation, and obtain the scientific and technological achievement transformation semantic response encoding vector.

[0039] In one embodiment, Figure 5 This is a schematic block diagram of the scientific and technological achievements-related factors interaction unit in the intelligent auxiliary decision-making system for scientific and technological achievements transformation based on deep learning and large models in the embodiment of the present application. Figure 5As shown, the scientific and technological achievements-related factors interaction unit 44 includes: a scientific and technological achievements-related factors semantic response subunit 441, which is used to calculate the semantic response decision anchoring features between the scientific and technological achievements type label unique hot encoding vector and each relevant factor sentence granularity semantic embedding encoding vector in the set of relevant factor sentence granularity semantic embedding encoding vectors to obtain a set of scientific and technological achievements-related factors local feature response anchor encoding matrices; a scientific and technological achievements-related factors splicing factor calculation subunit 442, which is used to determine the scientific and technological achievements-related factors based on the feature distribution of each scientific and technological achievements-related factors local feature response anchor encoding matrix in the set of scientific and technological achievements-related factors local feature response anchor encoding matrix. The decision anchor adaptive splicing factors of the element local feature response anchor coding matrix are used to obtain a set of scientific and technological achievement-related factor decision anchor adaptive splicing factors; the scientific and technological achievement-related factor weighting processing subunit 443 is used to perform weighting processing on the set of scientific and technological achievement-related factor decision anchor adaptive splicing factors based on the Softmax function to obtain a set of scientific and technological achievement-related factor decision anchor adaptive splicing weight factors; the scientific and technological achievement-transformation feature fusion subunit 444 is used to fuse the set of scientific and technological achievement-related factor local feature response anchor coding matrices based on the set of scientific and technological achievement-related factor decision anchor adaptive splicing weight factors to obtain the scientific and technological achievement transformation semantic response coding vector.

[0040] In one embodiment, the scientific and technological achievement-related factor semantic response subunit 441 is used to: perform deep implicit feature extraction based on fully connected encoding on the unique hot encoding vector of the scientific and technological achievement type label to obtain the deep implicit encoding vector of the scientific and technological achievement type label. Specifically, the process can be expressed by the formula: in, is the one-hot encoding vector of the scientific and technological achievement type label, is matrix multiplication, and are the scientific and technological achievement type label weight matrix and the scientific and technological achievement type label bias vector respectively, is the activation function, is the deep implicit coding vector of the scientific and technological achievement type label; The deep implicit feature extraction based on the fully connected coding is performed on each relevant factor sentence granular semantic embedding coding vector in the set of relevant factor sentence granular semantic embedding coding vectors to obtain a set of relevant factor sentence granular semantic deep implicit coding vectors. Specifically, the process can be expressed by the formula: in, is the set of the relevant factor sentence granularity semantic embedding encoding vectors, , , and are the first, second, and third in the set of the relevant factor sentence granularity semantic embedding encoding vectors. and The relevant factor sentence-level semantic embedding encoding vector, and are the relevant factor weight matrix and the relevant factor bias vector respectively, is the first in the set of relevant factor sentence-level semantic depth implicit encoding vectors The semantic depth latent encoding vector of sentence granularity is constructed based on the relevant factors.

[0041] Semantic response decision anchoring is performed on each relevant factor sentence granularity semantic depth implicit coding vector in the set of the scientific and technological achievement type label deep implicit coding vector and the relevant factor sentence granularity semantic depth implicit coding vector to obtain a set of the scientific and technological achievement-related factor local feature response anchor coding matrix. Specifically, the process can be expressed by the formula: in, yes The transposed vector of yes Length, It is the first in the set of local feature response anchor coding matrices of scientific and technological achievements-related factors. A scientific and technological achievement-related factor local feature response anchor coding matrix.

[0042] Specifically, first, the deep implicit feature extraction based on fully connected encoding is performed on the one-hot encoding vector of the scientific and technological achievement type label, with the aim of converting the originally discrete category information into a deep implicit encoding vector with rich semantic representation capabilities. After the scientific and technological achievement type label is one-hot encoded, it is mapped by a fully connected neural network to generate a deep implicit encoding vector of the scientific and technological achievement type label. In this process, the network weights learn the global feature representation, while the nonlinear activation function captures the nonlinear interaction relationship in the original features. This processing method not only enhances the semantic expression ability of the scientific and technological achievement type, but also lays the foundation for the subsequent deep integration with other related factors.

[0043] Next, the deep implicit feature extraction based on fully connected coding is performed on each relevant factor sentence granular semantic embedding coding vector in the set of relevant factor sentence granular semantic embedding coding vectors. This step aims to extract the deep semantic information of each factor from a large number of natural language descriptions. For example, relevant factors may include influencing factors on the government side (such as the strength of policy support), influencing factors on the market side (such as the degree of technological progress), and influencing factors on the social side (such as the quality of intermediary services). Each factor contains detailed text descriptions, which not only provide factual information but also contain rich contextual relationships. Through the fully connected neural network, the system is able to convert these descriptions into a set of relevant factor sentence granular semantic deep implicit coding vectors with high representation capabilities. This approach enables the model to understand the true meaning of the text and provide an accurate basis for generating auxiliary decision-making results.

[0044] Subsequently, the system anchors the semantic response decision of each relevant factor sentence granularity semantic deep implicit coding vector in the set of the scientific and technological achievement type label deep implicit coding vector and the relevant factor sentence granularity semantic deep implicit coding vector. The core of this process is to establish a deep semantic association between the scientific and technological achievement type and the relevant factors. Specifically, the system calculates the semantic response features between the scientific and technological achievement type and each relevant factor to form a series of local feature response anchor coding matrices. These matrices not only represent the interactive information between the scientific and technological achievement type and the relevant factors, but also play a bridging role in global-local semantic alignment. Through these local feature response anchor coding matrices, the system can capture the complex dependency between the scientific and technological achievement type and the relevant factors. These local feature response anchor coding matrices are equivalent to the building blocks that constitute the final decision anchor. They are combined and spliced ​​in a specific way to form a complete decision anchor deep implicit representation. This process ensures the dynamic response and precise matching between the scientific and technological achievement type and the relevant factors, thereby providing a scientific and reasonable auxiliary decision result for the transformation of scientific and technological achievements.

[0045] In one embodiment, the scientific and technological achievements-related factors splicing factor calculation subunit 442 is used to: obtain the scientific and technological achievements-related factors decision anchor adaptive splicing factor corresponding to the local feature response anchor coding matrix of the scientific and technological achievements-related factors based on the maximum value, mean, variance and difference magnification coefficient of the local feature response anchor coding matrix of the scientific and technological achievements-related factors and the drift coefficient of the scientific and technological achievements-related factors; wherein the difference magnification coefficient is the number of eigenvalues ​​of the local feature response anchor coding matrix of the scientific and technological achievements-related factors; the drift coefficient of the scientific and technological achievements-related factors is obtained by performing weak to strong interpretable generalization processing on the local feature response anchor coding matrix of the scientific and technological achievements-related factors. Specifically, the process can be expressed by the formula: in, express The variance of express The mean of For calculation The number of eigenvalues ​​of express The difference amplification factor, express The maximum value of Indicates the drift coefficient of scientific and technological achievements-related factors, It is an intermediate state transition. yes No. eigenvalues, It is the first in the set of adaptive splicing factors of scientific and technological achievements-related factors decision anchor A scientific and technological achievement-related factor decision anchor adaptive splicing factor.

[0046] It should be understood that based on the feature distribution of the anchor coding matrix of the local feature response of scientific and technological achievements-related factors, the adaptive splicing factors of the decision anchors of the respective scientific and technological achievements-related factors are calculated. The core of this step is to quantify the importance and influence of each local feature response anchor coding matrix of scientific and technological achievements-related factors in the overall decision. Specifically, the system analyzes the statistical characteristics of the maximum value, mean, variance, etc. of each local feature response anchor coding matrix of scientific and technological achievements-related factors, and combines the difference amplification coefficient and drift coefficient to determine the adaptive splicing factors of the decision anchors of scientific and technological achievements-related factors. The difference amplification coefficient reflects the number of eigenvalues ​​in the matrix and is used to enhance the significance of local features; while the drift coefficient of scientific and technological achievements-related factors enhances the global dominance of local features by interpreting and generalizing the anchor coding matrix of the local feature response of scientific and technological achievements-related factors from weak to strong. In this way, the system can extract the most representative features from complex data and provide accurate weight basis for subsequent feature fusion. Finally, the system uses these decision anchor adaptive splicing factors to combine and splice the local feature response anchor coding matrix of scientific and technological achievements-related factors to form a complete deep implicit representation of the decision anchor of scientific and technological achievements-related factors. This combination method not only takes into account the independent contribution of each local feature, but also achieves global optimization through weight adjustment. In this way, the system not only improves the scientificity and accuracy of scientific and technological achievement transformation decision-making, but also ensures the reasonable weight distribution of various relevant factors at different stages.

[0047] In particular, for the scientific and technological achievements-related factors decision anchor adaptive splicing factor, the scientific and technological achievements-related factors drift coefficient , the input feature set distribution of the anchor coding matrix of the local feature response of the scientific and technological achievements-related factors transitions from the state of weak overall interpretability of the mean to strong local interpretability of the maximum value, through the drift coefficient of the scientific and technological achievements-related factors The weak to strong interpretable generalization is used to enhance the global dominance of local features.

[0048] Specifically, As an intermediate state transition representation from weak interpretability to strong interpretability, each eigenvalue of the anchor coding matrix is ​​responded to by the local characteristics of scientific and technological achievements-related factors , taking it as the importance score of the input scientific and technological achievements-related factors local feature response anchor coding matrix for the global smooth state transition, to analyze the intermediate state transition The importance score weights relative to the global state transition are globally controlled to achieve interpretable generalized inference of the weight-dependency of the adaptive splicing factors of the scientific and technological achievement-related factor decision anchors.

[0049] Exemplarily, in the scientific and technological achievements-related factors weighted processing subunit 443, the set of scientific and technological achievements-related factors decision anchor adaptive splicing factors is weighted based on the Softmax function to convert the importance of each scientific and technological achievements-related factors local feature response anchor coding matrix into a probability distribution form, so that the system can better understand and weigh the impact of different factors, thereby providing more accurate and explainable auxiliary decision support. Among them, the role of the Softmax function is to convert a set of original numerical values ​​into a probability distribution, in which each value is mapped between 0 and 1, and the sum of all values ​​is 1. This processing method not only makes the importance of each factor intuitive and easy to understand, but also can amplify the significant differences of local features through the characteristics of the exponential function, and enhance the distinguishing ability of the feature distribution. Specifically, the process can be expressed by the formula: in, is the normalization function, It is the first in the set of adaptive splicing weight factors of scientific and technological achievements-related factors decision anchor A scientific and technological achievement-related factor decision anchor adaptive splicing weight factor.

[0050] Exemplarily, in the scientific and technological achievement-conversion feature fusion subunit 444, a set of scientific and technological achievement-related factor local feature response anchor coding matrices is fused based on a set of scientific and technological achievement-related factor decision anchor adaptive splicing weight factors, and finally a scientific and technological achievement conversion semantic response coding vector is generated to comprehensively consider the importance of each local feature and their mutual relationship, so as to form a comprehensive and accurate representation vector, thereby providing scientific and reasonable auxiliary decision support for scientific and technological achievement conversion. Specifically, the process can be expressed as follows: in, is the semantic response coding matrix of scientific and technological achievement transformation, is the reshaping operation, It is the semantic response encoding vector of the scientific and technological achievement transformation.

[0051] Exemplarily, in the scientific and technological achievement transformation auxiliary decision result generating unit 45, the scientific and technological achievement transformation auxiliary decision result is obtained based on the scientific and technological achievement transformation semantic response coding vector. In one embodiment, the scientific and technological achievement transformation auxiliary decision result generating unit is used to: input the scientific and technological achievement transformation semantic response coding vector into a transformation decision engine based on a classifier to obtain the scientific and technological achievement transformation auxiliary decision result. That is, the scientific and technological achievement transformation semantic response coding vector obtained by decision anchoring interaction using the set of the scientific and technological achievement type label unique hot coding vector and the relevant factor sentence granularity semantic embedding coding vector is classified and processed, so as to automatically generate the scientific and technological achievement transformation auxiliary decision result. It should be understood that the classifier can learn the mapping relationship between different feature patterns and decision results after specific training. Specifically, the scientific and technological achievement transformation semantic response coding vector integrates the deep interactive information of the scientific and technological achievement type, relevant determinants and influencing factors, and this information is presented in the form of a complex vector. The classifier can process this structured numerical vector data, and analyze the information contained in the vector through its internal algorithm and learned knowledge, so as to make a decision judgment on the transformation of scientific and technological achievements. In this way, direct guidance and suggestions can be provided for the actual operation of scientific and technological achievements transformation. For example, for innovative achievements in a specific industry, combined with factors such as market scope, the industry standards and national standards that should be followed can be determined to provide precise guidance for the compliance of achievement transformation.

[0052] In summary, according to the embodiment of the present application, the intelligent auxiliary decision-making system for the transformation of scientific and technological achievements based on deep learning and large models is explained, which obtains the innovation subject, innovation activity, innovation resource and industry / technical field input by the user and inputs them into the scientific and technological achievement identifier to obtain the scientific and technological achievement type label, and at the same time, obtains the determining factors and influencing factors input by the user, and then inputs the scientific and technological achievement type label, determining factors and influencing factors into the scientific and technological achievement transformation decision module based on the large model to obtain the scientific and technological achievement transformation auxiliary decision-making result, and the scientific and technological achievement transformation auxiliary decision-making result is the standard element that the scientific and technological achievements meet. In this way, the relevant information of scientific and technological achievements can be fully integrated, and scientific and reasonable auxiliary decision-making results can be generated in combination with standard elements, which not only improves the accuracy and efficiency of decision-making, but also ensures the compliance and market adaptability of scientific and technological achievements.

[0053] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0054] It should be understood that the specific examples in this article are only intended to help those skilled in the art better understand the embodiments of the present application, rather than to limit the scope of the embodiments of the present application.

[0055] It should also be understood that in the various embodiments of the present application, the size of the serial number of each process 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 application.

[0056] It should also be understood that the various implementation methods described in this specification can be implemented individually or in combination, and the embodiments of the present application are not limited to this.

[0057] Unless otherwise stated, all technical and scientific terms used in the embodiments of the present application are the same as the meanings generally understood by those skilled in the art of the technical field of the present application. The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the scope of the present application. The term "and / or" used in the present application includes any and all combinations of one or more related listed items. The singular forms of "a kind of", "above" and "the" used in the embodiments of the present application and the appended claims are also intended to include majority forms, unless the context clearly indicates other meanings. In addition, the terms "first", "second" etc. are only used for description purposes and cannot be understood as indicating or suggesting relative importance.

[0058] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0059] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0060] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0061] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0062] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent decision-making support system for the transformation of scientific and technological achievements based on a large model, characterized in that: include: A user input data acquisition module is used to acquire the innovation subject, innovation activity, innovation resource and industry / technical field input by the user; A scientific and technological achievement type identification module, used for inputting the innovation subject, the innovation activity, the innovation resource and the industry / technical field into a scientific and technological achievement identifier to obtain a scientific and technological achievement type label; A determining factor and influencing factor acquisition module is used to acquire determining factors and influencing factors input by a user, wherein the determining factors include a transformation subject, a transformation method, a transformation scope, and a transformation object, and the influencing factors include government-side influencing factors, market-side influencing factors, and social-side influencing factors; The scientific and technological achievement transformation decision-making assistance module is used to input the scientific and technological achievement type label, the determining factors and the influencing factors into the scientific and technological achievement transformation decision-making module based on the big model to obtain the scientific and technological achievement transformation decision-making assistance results, and the scientific and technological achievement transformation decision-making assistance results are the standard elements that the scientific and technological achievements meet.

2. According to the large model-based intelligent decision-making system for transformation of scientific and technological achievements according to claim 1, it is characterized in that: The influencing factors on the government side include a text description of policy strength and a text description of governance level; the influencing factors on the market side include a text description of technological progress and a text description of product competition; and the influencing factors on the society side include a text description of intermediary services and a text description of multi-party collaboration.

3. The intelligent decision-making support system for transformation of scientific and technological achievements based on a large model according to claim 2 is characterized in that: The standard elements include international standards, national standards, industry standards, local standards, group standards and enterprise standards.

4. The intelligent decision-making support system for transformation of scientific and technological achievements based on a large model according to claim 3 is characterized in that: The scientific and technological achievement type identification module includes: A semantic coding unit for key information on conversion of scientific and technological achievements, used for semantically coding the innovation subject, the innovation activity, the innovation resource and the industry / technical field using a semantic encoder based on a Bert model to obtain a semantic coding vector for key information on conversion of scientific and technological achievements; The scientific and technological achievement type label determination unit is used to input the semantic coding vector of the key information of the scientific and technological achievement conversion into a scientific and technological achievement identifier based on a classifier to obtain the scientific and technological achievement type label.

5. According to claim 4, the intelligent auxiliary decision-making system for transformation of scientific and technological achievements based on a large model is characterized in that: The scientific and technological achievement transformation decision-making auxiliary module includes: A scientific and technological achievement type label one-hot encoding unit, used for one-hot encoding the scientific and technological achievement type label to obtain a scientific and technological achievement type label one-hot encoding vector; A determining factor and influencing factor semantic embedding coding unit, used for performing semantic embedding coding on the determining factor and the influencing factor to obtain a set of determining factor sentence granularity semantic embedding coding vectors and a set of influencing factor sentence granularity semantic embedding coding vectors; A related factor set merging unit, used for merging the set of the determining factor sentence granularity semantic embedding coding vectors and the set of the influencing factor sentence granularity semantic embedding coding vectors to obtain a set of related factor sentence granularity semantic embedding coding vectors; A scientific and technological achievement-related factor interaction unit, used for performing a scientific and technological achievement-related factor decision anchoring semantic deep interaction analysis on the set of the scientific and technological achievement type label one-hot encoding vector and the related factor sentence granularity semantic embedding encoding vector to obtain a scientific and technological achievement transformation semantic response encoding vector; The scientific and technological achievement transformation auxiliary decision-making result generating unit is used to obtain the scientific and technological achievement transformation auxiliary decision-making result based on the scientific and technological achievement transformation semantic response coding vector.

6. The intelligent decision-making support system for transformation of scientific and technological achievements based on a large model according to claim 5 is characterized in that: The scientific and technological achievements-related factors interaction unit includes: A scientific and technological achievement-related factor semantic response subunit, used for calculating the semantic response decision anchoring features between the scientific and technological achievement type label one-hot encoding vector and each relevant factor sentence granularity semantic embedding encoding vector in the set of relevant factor sentence granularity semantic embedding encoding vectors to obtain a set of scientific and technological achievement-related factor local feature response anchoring encoding matrices; A scientific and technological achievement-related factor splicing factor calculation subunit is used to determine the decision anchor adaptive splicing factor of each scientific and technological achievement-related factor local feature response anchor coding matrix based on the feature distribution of each scientific and technological achievement-related factor local feature response anchor coding matrix in the set of the scientific and technological achievement-related factor local feature response anchor coding matrix to obtain a set of scientific and technological achievement-related factor decision anchor adaptive splicing factors; A scientific and technological achievement-related factor weighting processing subunit, used for performing weighting processing on the set of scientific and technological achievement-related factor decision anchor adaptive splicing factors based on a Softmax function to obtain a set of scientific and technological achievement-related factor decision anchor adaptive splicing weighting factors; The scientific and technological achievement-transformation feature fusion subunit is used to fuse the set of scientific and technological achievement-related factor local feature response anchor coding matrices based on the set of adaptive splicing weight factors of the scientific and technological achievement-related factor decision anchors to obtain the scientific and technological achievement transformation semantic response coding vector.

7. The intelligent decision-making support system for transformation of scientific and technological achievements based on a large model according to claim 6 is characterized in that: The scientific and technological achievements-related factors semantic response subunit is used to: Performing deep implicit feature extraction based on fully connected encoding on the one-hot encoding vector of the scientific and technological achievement type label to obtain a deep implicit encoding vector of the scientific and technological achievement type label; Performing deep implicit feature extraction based on fully connected coding on each relevant factor sentence granular semantic embedding coding vector in the set of relevant factor sentence granular semantic embedding coding vectors to obtain a set of relevant factor sentence granular semantic deep implicit coding vectors; Semantic response decision anchoring is performed on each relevant factor sentence granularity semantic deep implicit coding vector in the set of the scientific and technological achievement type label deep implicit coding vector and the relevant factor sentence granularity semantic deep implicit coding vector to obtain a set of the scientific and technological achievement-related factor local feature response anchor coding matrices.

8. The intelligent decision-making support system for transformation of scientific and technological achievements based on a large model according to claim 7 is characterized in that: The scientific and technological achievement-related factor splicing factor calculation subunit is used for: Based on the maximum value, mean value, variance and difference magnification coefficient of the local feature response anchor coding matrix of the scientific and technological achievements-related factors and the drift coefficient of the scientific and technological achievements-related factors, the scientific and technological achievements-related factors decision anchor adaptive splicing factor corresponding to the local feature response anchor coding matrix of the scientific and technological achievements-related factors is obtained; The difference amplification coefficient is the number of eigenvalues ​​of the anchor coding matrix of the local characteristic response of the scientific and technological achievements-related factors; The scientific and technological achievements-related factors drift coefficient is obtained by performing weak to strong interpretable generalization processing on the local feature response anchor coding matrix of the scientific and technological achievements-related factors.

9. The intelligent decision-making support system for transformation of scientific and technological achievements based on a large model according to claim 8 is characterized in that: The scientific and technological achievement transformation auxiliary decision-making result generating unit is used to: input the scientific and technological achievement transformation semantic response coding vector into a classifier-based transformation decision engine to obtain the scientific and technological achievement transformation auxiliary decision-making result.

Images