Science and technology project management method and system based on BTA fusion model

Abstract

The invention provides a science and technology project management method based on a BTA fusion model, and the method comprises the steps: carrying out the preprocessing of a science and technology project declaration, extracting the content of the science and technology project declaration through an identification rule, and obtaining all kinds of project information in the science and technology project declaration; identifying innovation elements of project information in the science and technology project declaration through a pre-created BTA fusion model; performing retrieval according to the innovation elements, and determining an innovation level of the science and technology project declaration through a retrieval result; wherein the pre-created BTA fusion model is created based on a small convolutional neural network model and a large language model. According to the method, the key content in the science and technology project declaration is recognized and extracted by combining the recognition rule with the BTA fusion model, the accuracy is improved for retrieving the prior art, and meanwhile, necessary support is provided for avoiding repeated scientific research route research and evaluating the innovation level of scientific research projects.

Description

Technical Field

[0001] The present invention relates to the technical field of scientific and technological innovation management, and in particular to a scientific and technological project management method and system based on a BTA fusion model. Background Art

[0002] Existing scientific research management systems are often developed relatively early, resulting in limited module reproducibility and portability. This leads to significant R&D costs when management methods are updated and business functions and processes are adjusted. This makes iteration inflexible, especially for complex functions.

[0003] Furthermore, with the increasing number of scientific research projects, large scientific research enterprises often face numerous challenges in project management. Two key challenges are particularly prominent. First, how to exchange technical findings between different research projects and minimize duplication of research paths. Second, how to evaluate the level of innovation in research projects and provide the necessary support for projects with high scientific value. Summary of the Invention

[0004] In order to solve the problems of how to achieve the exchange of technical achievements between different scientific research projects in the existing technology, avoid repeated scientific research route research as much as possible, and how to evaluate the degree of innovation of scientific research projects to better provide necessary support for scientific research innovation projects with higher scientific research value, the present invention proposes a scientific and technological project management method based on the BTA fusion model, including:

[0005] Pre-process the science and technology project application form, extract the content of the science and technology project application form through identification rules, and obtain various project information in the science and technology project application form;

[0006] Identify innovative elements of the project information in the science and technology project application through a pre-created BTA fusion model;

[0007] Conduct a search based on the innovation elements and determine the innovation level of the science and technology project application based on the search results;

[0008] The pre-created BTA fusion model is created based on a small convolutional neural network model and a large language model.

[0009] Preferably, the creation of the BTA fusion model includes:

[0010] Based on the innovative content in the project information, the innovative points in the innovative content are classified and marked and summarized into a fixed sentence format to form a training sample;

[0011] Based on the training samples, a small convolutional neural network model as the main logical judgment model and a large language model as the auxiliary experience judgment model are trained to obtain the BTA fusion model.

[0012] Preferably, the small convolutional neural network model includes:

[0013] Add the TextCNN convolutional neural network model with the Attention mechanism and the BERT pre-trained language model;

[0014] The TextCNN convolutional neural network model with the Attention mechanism is used to capture the innovative content in the project information;

[0015] The BERT pre-trained language model is used to understand unlabeled project information.

[0016] Preferably, before identifying the innovative elements of the project information in the science and technology project application using the pre-created BTA fusion model, the following steps are included:

[0017] A similarity judgment is performed based on the various types of project information. When the number of categories with similar project information exceeds a preset threshold, it is determined that the similarity exceeds the standard. The technology project application with the similarity exceeding the standard is marked as a similar application, and an alarm for similarity exceeding the standard is issued.

[0018] Preferably, the searching based on the innovation elements and determining the innovation level of the science and technology project application through the search results include:

[0019] Searching based on each innovation point in the innovation element and the technical field corresponding to the innovation point to obtain search results corresponding to the innovation point;

[0020] Compare all the innovative points in the science and technology project application to determine whether the innovative point is the same as the search result corresponding to the innovative point. If they are the same, modify the innovative point into a technical element; otherwise, modify the innovative point into a technical development content;

[0021] The innovation level is determined according to the quantity of the technological development contents and the contribution of the technological development contents.

[0022] Preferably, determining the innovation level according to the amount of the technological development contents and the contribution of the technological development contents includes:

[0023] Determine the degree of innovation based on matching the technological development content with pre-set templates;

[0024] Determine the contribution of the current technological development content based on the degree of innovation;

[0025] The innovation level is determined based on the sum of the contributions of all technological development contents in the current science and technology project application.

[0026] Preferably, the determination of the degree of innovation based on matching the technological development content with a pre-set template includes:

[0027] The technology development content is matched according to a pre-set template. If the same template exists, the innovation degree is calculated according to the template; otherwise, a new template is set according to the technology development content.

[0028] Based on the same inventive concept, the present invention also provides a technology project management system based on the BTA fusion model, the system comprising:

[0029] The decision analysis module is used to pre-process the science and technology project application form and extract the content of the science and technology project application form through identification rules to obtain various project information in the science and technology project application form;

[0030] An element identification module is used to identify innovative elements of the project information in the science and technology project application form through a pre-created BTA fusion model;

[0031] A level determination module, configured to search based on the innovation elements and determine the innovation level of the science and technology project application based on the search results;

[0032] The pre-created BTA fusion model is created based on a small convolutional neural network model and a large language model.

[0033] Preferably, the element identification module includes:

[0034] The BTA fusion model creation unit is used to form training samples based on the innovative content in the project information by classifying and marking the innovative points in the innovative content and summarizing them into a fixed sentence format; based on the training samples, a small convolutional neural network model as the main logical judgment model and a large language model as the auxiliary experience judgment model are trained to obtain a BTA fusion model.

[0035] Preferably, the small convolutional neural network model includes:

[0036] Add the TextCNN convolutional neural network model with the Attention mechanism and the BERT pre-trained language model;

[0037] The TextCNN convolutional neural network model with the Attention mechanism is used to capture the innovative content in the project information;

[0038] The BERT pre-trained language model is used to understand unlabeled project information.

[0039] Preferably, the system further comprises:

[0040] Similarity judgment module, used for

[0041] A similarity judgment is performed based on the various types of project information. When the number of categories with similar project information exceeds a preset threshold, it is determined that the similarity exceeds the standard. The technology project application with the similarity exceeding the standard is marked as a similar application, and an alarm for similarity exceeding the standard is issued.

[0042] Preferably, the level determination module includes:

[0043] A search unit, configured to search based on each innovation point in the innovation element and the technical field corresponding to the innovation point, and obtain search results corresponding to the innovation point;

[0044] an innovation point judgment unit, configured to compare all innovation points in the science and technology project application, and judge whether the innovation point is the same as the search result corresponding to the innovation point; if they are the same, modify the innovation point into a technical element; otherwise, modify the innovation point into a technical development content;

[0045] The innovation level determination unit is used to determine the innovation level according to the quantity of the technology development contents and the contribution of the technology development contents.

[0046] Preferably, the innovation level determination unit includes:

[0047] The innovation degree determination subunit is used to determine the innovation degree based on matching the technology development content with a pre-set template;

[0048] a contribution determination subunit, configured to determine the contribution of the current technological development content based on the degree of innovation;

[0049] The innovation level determination sub-unit is used to determine the innovation level based on the sum of the contributions of all technological development contents in the current science and technology project application.

[0050] In another aspect, the present application further provides an electronic device, comprising: at least one processor and a memory; the memory and the processor are connected via a bus;

[0051] The memory is used to store one or more programs;

[0052] When the one or more programs are executed by the at least one processor, the above-mentioned scientific and technological project management method based on the BTA fusion model is implemented.

[0053] On the other hand, the present application also provides a readable storage medium on which an execution program is stored. When the execution program is executed, a technology project management method based on the BTA fusion model as described above is implemented.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] The present invention provides a technology project management method based on a BTA fusion model, comprising: preprocessing technology project applications and extracting the contents of the technology project applications through identification rules to obtain various types of project information in the technology project applications; identifying innovative elements of the project information in the technology project applications through a pre-created BTA fusion model; searching based on the innovative elements, and determining the innovation level of the technology project applications through the search results; wherein the pre-created BTA fusion model is created based on a small convolutional neural network model and a large language model. The present invention identifies and extracts key content in technology project applications through identification rules combined with a BTA fusion model, thereby improving the accuracy of searching for existing technologies and providing necessary support for avoiding duplication of scientific research routes and evaluating the innovation level of scientific research projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A flowchart of a technology project management method based on the BTA fusion model provided by the present invention;

[0057] Figure 2 A schematic diagram of an electronic device for a technology project management method based on a BTA fusion model provided by the present invention. DETAILED DESCRIPTION

[0058] The present invention proposes a technology project management method based on the BTA fusion model. The specific embodiments of the present invention are further described in detail below with reference to the accompanying drawings.

[0059] Example 1:

[0060] A technology project management method based on the BTA fusion model, such as Figure 1 Shown, including:

[0061] Step 1: Pre-process the science and technology project application form and extract the content of the science and technology project application form through recognition rules to obtain various project information in the science and technology project application form;

[0062] Step 2: Identify the innovative elements of the project information in the science and technology project application through the pre-created BTA fusion model;

[0063] Step 3: Search based on the innovation elements and determine the innovation level of the science and technology project application based on the search results;

[0064] The pre-created BTA fusion model is created based on a small convolutional neural network model and a large language model.

[0065] Step 1: Pre-process the science and technology project application form and extract the content of the science and technology project application form through identification rules to obtain various project information in the science and technology project application form, including:

[0066] A user uploads a technology project application to the technology project management system, which pre-processes the application and converts its content into data text within the system. The content of the application is extracted based on the text recognition parameters configured within the system. The text recognition parameters contain a large number of "title-content" recognition rules. These rules can be used to extract content from the data text and locate it, enabling more accurate identification of the specific location of innovative content within the application. The extracted content is stored in the technology project management system's database for subsequent statistical processing. "Title-content" recognition rules can be created and obtained during the expert review process. Alternatively, the system can prompt application writers to follow predetermined requirements, agreeing on the format and location of innovative content, thereby eliminating the need to search for innovative content. This content recognition function accurately captures various project information within the application and allows for preliminary classification of the application based on the project information. Furthermore, various pre-defined factor categories are assigned to the application based on the project information, facilitating subsequent decision-making and statistical analysis.

[0067] In order to achieve more objective, accurate, and efficient expert review of science and technology project applications, the present invention combines the BTA (BERT-TextCNN-Attention) fusion model. The creation of the BTA fusion model includes:

[0068] Based on the innovative content in the project information, the innovative points in the innovative content are classified and marked and summarized into a fixed sentence format to form a training sample;

[0069] Based on the training samples, a small convolutional neural network model as the main logical judgment model and a large language model as the auxiliary experience judgment model are trained to obtain the BTA fusion model.

[0070] Specifically, when identifying innovative content within project information, the document needs to be annotated. This annotation process is used during model training. This is done manually, with experts in the relevant fields reviewing existing science and technology project proposals and annotating the relevant innovative statements according to their classification. This annotation process uses fixed phrases to summarize the innovations, such as increasing the output power of the M23 motor by 12W and reducing the friction between the motor rotor and bearing by 0.05N. Each annotated science and technology project proposal serves as a data sample. The BTA fusion model is trained on a sufficient number of data samples to achieve its intended purpose, resulting in a trained BTA fusion model. This trained BTA fusion model can identify innovative elements within the project information, including: innovation points, technical fields, and technical elements. It also determines the location of the innovation points, the specific description of the innovation points, and the level of technological innovation. An excellent model can identify multiple innovation points within a project, with the text corresponding to each innovation point being highly targeted. The level of technological creation is divided according to the degree of innovation. If it is a completely new field, the level of technological creation is high. If the degree of improvement on existing mature technologies is relatively small, the level of technological creation is low. As an indicator for evaluating scientific and technological project applications, it is used to reflect the impact of technological creation on human technology as a whole. It can be divided into 7 levels: 1: Parameter adjustment of existing structures and models; 2: Structural adjustment of existing structures and models; 3: Structural improvement of existing structural models to produce corresponding effects; 4: Structural changes to existing structural models to solve corresponding problems or contradictions; 5: Improvement of the system; 6: Improvement of scientific principles, or adoption of new principles to realize the original functions; 7: Adoption of new discoveries to realize new functions.

[0071] The BTA fusion model uses a small neural network model as the main logical judgment model to improve the accuracy and reliability of innovative content recognition; it uses a large language model as an auxiliary experience judgment model to improve the coverage and comprehensiveness of innovative content recognition when a small convolutional neural network model has difficulty in recognizing innovative content.

[0072] The small convolutional neural network model includes:

[0073] Add the TextCNN convolutional neural network model with the Attention mechanism and the BERT pre-trained language model;

[0074] The TextCNN convolutional neural network model with the Attention mechanism is used to capture the innovative content in the project information;

[0075] The BERT pre-trained language model is used to understand unlabeled project information.

[0076] Specifically, the small convolutional neural network model is based on the TextCNN convolutional neural network model, and the Attention mechanism is added to enhance the ability to focus on the input text as a whole, and better capture the content in the text that is closely related to the innovative content; in addition, in order to enhance the ability to recognize unlabeled text and ensure the stability of the model, a pre-trained language model is set at the model input position to understand the unlabeled text corpus and realize the understanding and expression of the common language. The pre-trained language model is represented by the BERT model. The large language model uses the HFL (Hugging Face Llama) large language model to recognize innovative texts. By defining the role, task, and output format of the HFL model, the HFL model can better complete the given task. The present invention combines the BTA fusion model to more efficiently identify the content of scientific and technological innovation projects and classify various projects.

[0077] Step 2, before identifying the innovative elements of the project information in the science and technology project application using the pre-created BTA fusion model, specifically includes:

[0078] A similarity judgment is performed based on the various types of project information. When the number of categories with similar project information exceeds a preset threshold, it is determined that the similarity exceeds the standard. The technology project application with the similarity exceeding the standard is marked as a similar application, and an alarm for similarity exceeding the standard is issued.

[0079] Specifically, after the user uploads the project application form to the science and technology project management system, the text content directly related to the scientific research and innovation project in the application form is extracted first. Such as the level of the project (national, provincial, etc.), project name, technical field, technical branch, etc. If the number of similarities in these contents exceeds a predetermined numerical value, two similar applications are marked as similar applications, and a similar application form alarm is prompted, and the application is output to relevant technical experts for manual evaluation. Thereby, the expert review resources are utilized in a targeted manner, the fatigue caused by blindly searching for similar applications is avoided, and the submission of the same project application form is avoided as much as possible. The present invention adopts the method of integrating existing scientific and technological innovation projects, supervises the direction of scientific and technological research and development and the degree of scientific and technological research and development, compares the text content of the project submitted by contrast, finds the same project inside the system, thereby avoids repeated research and development between projects, and reduces unnecessary waste of resources.

[0080] Step 3, searching based on the innovation elements and determining the innovation level of the science and technology project application based on the search results, specifically includes:

[0081] Searching based on each innovation point in the innovation element and the technical field corresponding to the innovation point to obtain search results corresponding to the innovation point;

[0082] Compare all the innovative points in the science and technology project application to determine whether the innovative point is the same as the search result corresponding to the innovative point. If they are the same, modify the innovative point into a technical element; otherwise, modify the innovative point into a technical development content;

[0083] The innovation level is determined according to the quantity of the technological development contents and the contribution of the technological development contents.

[0084] Specifically, the search scope is narrowed when searching for identical existing documents based on the technical fields within the innovation elements. The technical elements within the innovation elements serve as prerequisite technologies for implementing the technical solutions in the science and technology project proposals. The innovative points within the innovation elements are a collection of technical elements that contribute to existing academic fields. Each innovative point corresponds to one or more search results. During the review process, the review experts compare the search results with the innovation points to determine whether they are identical. If the innovation point is identical to the search result, the innovation point is changed to a technical element, and the next innovation point is determined. If the innovation point is different from the search result, the innovation point is changed to technological development content, which is confirmed to be truly innovative after comparison with the search results. Once all innovation points have been evaluated, the number of technological development contents in the science and technology project proposals is counted and used as a key parameter for evaluating the innovation level. By identifying the innovative point content in the project text, the content is obtained and used to search for projects with similarities to external literature. This avoids duplication of research and development between projects and reduces unnecessary waste of resources.

[0085] Determining the innovation level based on the amount of the technological development content and the contribution of the technological development content includes:

[0086] Determine the degree of innovation based on matching the technological development content with pre-set templates;

[0087] Determine the contribution of the current technological development content based on the degree of innovation;

[0088] The innovation level is determined based on the sum of the contributions of all technological development contents in the current science and technology project application.

[0089] The determination of the degree of innovation based on matching the technological development content with a pre-set template includes:

[0090] The technology development content is matched according to a pre-set template. If the same template exists, the innovation degree is calculated according to the template; otherwise, a new template is set according to the technology development content.

[0091] Specifically, the innovation level is determined by the technological development content and the corresponding contribution level of the technological development content. In an application, the value determined by adding up the contribution levels corresponding to all technological development contents is the innovation level. As an indicator, the innovation level is used to assist in the evaluation of applications and predict the award-winning status of scientific and technological project applications, and to give projects with higher innovation levels higher priority and weight. Therefore, in subsequent decision-making analysis, projects with higher innovation levels are given priority. This reduces the risk of higher-quality scientific and technological innovation projects being missed due to human error. The expert review process, which is highly subjective, is transformed into a more logical judgment of similarities and differences, avoiding the interference of personal subjectivity in the review results.

[0092] The present invention uses a technology project management system to evaluate and predict technology project applications. However, the evaluation and prediction results are not output as final results. Instead, they are output after a more detailed and accurate evaluation of the project by review experts. On the one hand, the technology project management system provides prompts to review experts to avoid bias in the review results caused by personal subjective intentions. On the other hand, the actual output results are used to make reverse corrections, making the technology project management system and review experts complement each other, achieving more accurate and efficient management of scientific and technological innovation projects.

[0093] Specifically, there are two types of reverse corrections. One is feedback from literature comparisons. After comparing the innovation point with the comparative literature, if they are the same, the result is recorded, indicating that the search method is correct or a good search method. If the search is incorrect, the keywords or search method are changed to search for comparative literature. The other is the identification of the innovation point. In the process of comparing the innovation point with the comparative literature to see if they are the same, the review experts will also verify whether the innovation point has been correctly identified and feedback the results to the fusion model.

[0094] The Science and Technology Innovation Management System also features various micro-applications, which call upon multiple modules within the system as needed, forming corresponding module interfaces. These micro-applications implement specialized functions, which can access data from various workstations as much as possible. These micro-applications include: basic service micro-applications, expert review micro-applications, expert database and expert management micro-applications, and risk management micro-applications.

[0095] Existing expert reviews rely heavily on the subjective opinions of the reviewers, which can significantly impact the review results. The practice of completely replacing expert reviewers with artificial intelligence is immature, and the results are unlikely to convince the reviewees. This invention provides objective review support for reviewers, reducing the significant impact of their subjective opinions on the review results and alleviating their workload. It also fully leverages the accuracy and objectivity of manual review, ensuring the humanization and accuracy of scientific and technological project management.

[0096] Example 2:

[0097] The present invention based on the same inventive concept also provides a technology project management system based on the BTA fusion model, the system comprising:

[0098] The decision analysis module is used to pre-process the science and technology project application form and extract the content of the science and technology project application form through identification rules to obtain various project information in the science and technology project application form;

[0099] An element identification module is used to identify innovative elements of the project information in the science and technology project application form through a pre-created BTA fusion model;

[0100] A level determination module, configured to search based on the innovation elements and determine the innovation level of the science and technology project application based on the search results;

[0101] The pre-created BTA fusion model is created based on a small convolutional neural network model and a large language model.

[0102] The element identification module includes:

[0103] The BTA fusion model creation unit is used to form training samples based on the innovative content in the project information by classifying and marking the innovative points in the innovative content and summarizing them into a fixed sentence format; based on the training samples, a small convolutional neural network model as the main logical judgment model and a large language model as the auxiliary experience judgment model are trained to obtain a BTA fusion model.

[0104] The small convolutional neural network model includes:

[0105] Add the TextCNN convolutional neural network model with the Attention mechanism and the BERT pre-trained language model;

[0106] The TextCNN convolutional neural network model with the Attention mechanism is used to capture the innovative content in the project information;

[0107] The BERT pre-trained language model is used to understand unlabeled project information.

[0108] The system further comprises:

[0109] Similarity judgment module, used for

[0110] A similarity judgment is performed based on the various types of project information. When the number of categories with similar project information exceeds a preset threshold, it is determined that the similarity exceeds the standard. The technology project application with the similarity exceeding the standard is marked as a similar application, and an alarm for similarity exceeding the standard is issued.

[0111] Level determination module, including:

[0112] A search unit, configured to search based on each innovation point in the innovation element and the technical field corresponding to the innovation point, and obtain search results corresponding to the innovation point;

[0113] an innovation point judgment unit, configured to compare all innovation points in the science and technology project application, and judge whether the innovation point is the same as the search result corresponding to the innovation point; if they are the same, modify the innovation point into a technical element; otherwise, modify the innovation point into a technical development content;

[0114] The innovation level determination unit is used to determine the innovation level according to the quantity of the technology development contents and the contribution of the technology development contents.

[0115] The innovation level determination unit includes:

[0116] The innovation degree determination subunit is used to determine the innovation degree based on matching the technology development content with a pre-set template;

[0117] a contribution determination subunit, configured to determine the contribution of the current technological development content based on the degree of innovation;

[0118] The innovation level determination sub-unit is used to determine the innovation level based on the sum of the contributions of all technological development contents in the current science and technology project application.

[0119] This invention provides a technology project management system based on the BTA fusion model, which integrates existing technology innovation projects and monitors the direction and level of R&D. By comparing the textual content of submitted projects, the system identifies identical projects within the system. Furthermore, by identifying innovative content within the project texts, the system can retrieve this information and use it to search for projects that share similarities with external literature. This prevents duplication of R&D between projects and reduces unnecessary resource waste.

[0120] Example 3:

[0121] like Figure 2As shown, the present invention also provides an electronic device, which may be a computer, a single-chip microcomputer, a smart mobile device, or the like. The electronic device in this embodiment may include a processor, a memory, a transceiver component, and the like. The memory, processor, and transceiver component are connected via a bus; the memory may be used to store an execution program, which may include instructions; and the processor may be used to execute the instructions stored in the memory. The memory may also be used to store data, which may be accessed and / or modified during the execution of the instructions.

[0122] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the storage medium to implement the corresponding method flow or corresponding function, so as to realize the steps of a scientific and technological project management method based on the BTA fusion model in the above embodiment.

[0123] Example 4:

[0124] Based on the same inventive concept, the present invention also provides a readable storage medium, specifically an electronic device readable storage medium (Memory), which is a memory device in an electronic device for storing programs and data. It can be understood that the storage medium here can include both built-in storage media in the electronic device and, of course, extended storage media supported by the electronic device. The storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space. These instructions can be one or more execution programs (including program codes). It should be noted that the storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk storage. The processor loads and executes one or more instructions stored in the storage medium, which can implement the steps of a technology project management method based on the BTA fusion model in the above embodiment.

[0125] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0126] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0127] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0128] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0129] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are included in the scope of the claims of the present invention to be approved.

Claims

1. A technology project management method based on the BTA fusion model, characterized by: include: Pre-process the science and technology project application form, extract the content of the science and technology project application form through identification rules, and obtain various project information in the science and technology project application form; Identify innovative elements of the project information in the science and technology project application through a pre-created BTA fusion model; Conduct a search based on the innovation elements and determine the innovation level of the science and technology project application based on the search results; The pre-created BTA fusion model is created based on a small convolutional neural network model and a large language model.

2. The method according to claim 1, characterized in that The creation of the BTA fusion model includes: Based on the innovative content in the project information, the innovative points in the innovative content are classified and marked and summarized into a fixed sentence format to form a training sample; Based on the training samples, a small convolutional neural network model as the main logical judgment model and a large language model as the auxiliary experience judgment model are trained to obtain the BTA fusion model.

3. The method according to claim 2, characterized in that The small convolutional neural network model includes: Add the TextCNN convolutional neural network model with the Attention mechanism and the BERT pre-trained language model; The TextCNN convolutional neural network model with the Attention mechanism is used to capture the innovative content in the project information; The BERT pre-trained language model is used to understand unlabeled project information.

4. The method according to claim 1, wherein Before identifying the innovative elements of the project information in the science and technology project application through the pre-created BTA fusion model, the following steps are included: A similarity judgment is performed based on the various types of project information. When the number of categories with similar project information exceeds a preset threshold, it is determined that the similarity exceeds the standard. The technology project application with the similarity exceeding the standard is marked as a similar application, and an alarm for similarity exceeding the standard is issued.

5. The method according to claim 1, wherein The search is performed based on the innovation elements, and the innovation level of the science and technology project application is determined based on the search results, including: Searching based on each innovation point in the innovation element and the technical field corresponding to the innovation point to obtain search results corresponding to the innovation point; Compare all the innovative points in the science and technology project application to determine whether the innovative point is the same as the search result corresponding to the innovative point. If they are the same, modify the innovative point into a technical element; otherwise, modify the innovative point into a technical development content; The innovation level is determined according to the quantity of the technological development contents and the contribution of the technological development contents.

6. The method according to claim 5, characterized in that Determining the innovation level based on the amount of the technological development content and the contribution of the technological development content includes: Determine the degree of innovation based on matching the technological development content with pre-set templates; Determine the contribution of the current technological development content based on the degree of innovation; The innovation level is determined based on the sum of the contributions of all technological development contents in the current science and technology project application.

7. The method according to claim 6, characterized in that The determination of the degree of innovation based on matching the technological development content with a pre-set template includes: The technology development content is matched according to a pre-set template. If the same template exists, the innovation degree is calculated according to the template; otherwise, a new template is set according to the technology development content.

8. A technology project management system based on the BTA fusion model, characterized in that: The system comprises: The decision analysis module is used to pre-process the science and technology project application form and extract the content of the science and technology project application form through identification rules to obtain various project information in the science and technology project application form; An element identification module is used to identify innovative elements of the project information in the science and technology project application form through a pre-created BTA fusion model; A level determination module, configured to search based on the innovation elements and determine the innovation level of the science and technology project application based on the search results; The pre-created BTA fusion model is created based on a small convolutional neural network model and a large language model.

9. An electronic device, characterized in that: include: at least one processor and memory; The memory and the processor are connected via a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, a scientific and technological project management method based on a BTA fusion model as described in any one of claims 1 to 7 is implemented.

10. A readable storage medium, characterized in that: An execution program is stored thereon, and when the execution program is executed, a technology project management method based on a BTA fusion model as described in any one of claims 1 to 7 is implemented.

Images