An AI-based scientific and innovative material intelligent duplicate checking method and system

Abstract

Embodiments of the present specification relate to the field of information technology, in particular to an AI-based scientific and innovative material intelligent duplicate checking method and system. The method comprises the steps of: reading historical scientific research project materials, generating deconstruction information of the scientific research project materials, the deconstruction information comprising a plurality of technical target descriptions, technical stack images and subject information abstracts; establishing a quick check table of the historical scientific research project materials according to the deconstruction information; reading scientific and innovative materials to be checked, extracting technical targets, technical abstracts and subject information of the scientific and innovative materials; respectively using the technical targets, technical abstracts and subject information of the scientific and innovative materials to compare with the quick check table to obtain all matching rows; generating technical target duplication, technical abstract duplication and subject information duplication according to all matching rows; and generating a duplicate checking rate of the scientific and innovative materials to be checked according to the technical target duplication, technical abstract duplication and subject information duplication.

Description

TECHNICAL FIELD

[0001] Embodiments of the present specification relate to the field of information technology, in particular to an AI-based intelligent duplicate checking method and system for scientific and innovative materials. BACKGROUND

[0002] With the development of science and technology and the increasing frequency of scientific research activities, duplicate checking of scientific research project materials has become an important link to ensure scientific research integrity. Traditional duplicate checking methods mainly rely on text similarity comparison. This method is less sensitive to changes in written expression and is difficult to effectively identify scientific research materials that have been rewritten but still contain substantial duplicate content. Especially in the field of scientific innovation, scientific research projects often involve complex technology stacks, unique technical goals, and diverse subject information, which further increases the difficulty of duplicate checking.

[0003] Artificial intelligence (AI) technology has made significant progress in natural language processing and pattern recognition, providing new solutions to improve the accuracy and efficiency of scientific research material duplicate checking. AI-based methods can deeply understand the actual content of scientific research materials. However, there is currently a lack of corresponding research on duplicate checking of scientific and innovative materials. SUMMARY

[0004] Embodiments of the present specification describe an AI-based intelligent duplicate checking method and system for scientific and innovative materials.

[0005] In a first aspect, the embodiments of the present specification provide an AI-based intelligent duplicate checking method for scientific and innovative materials, comprising the steps of:

[0006] reading historical scientific research project materials, generating deconstruction information of the scientific research project materials, the deconstruction information including a number of technical goal descriptions, technology stack portraits, and subject information abstracts;

[0007] establishing a quick check table of historical scientific research project materials according to the deconstruction information, each row of the quick check table recording the identification and deconstruction information of a historical scientific research project material;

[0008] reading scientific and innovative materials to be checked, extracting the technical goals, technical abstracts, and subject information of the scientific and innovative materials;

[0009] respectively comparing the technical goals, technical abstracts, and subject information of the scientific and innovative materials to be checked with the quick check table to obtain all matching rows;

[0010] generating technical goal duplication, technical abstract duplication, and subject information duplication according to all matching rows;

[0011] generating a duplicate checking rate of the scientific and innovative materials to be checked according to the technical goal duplication, technical abstract duplication, and subject information duplication.

[0012] In a second aspect, the embodiments of the present specification provide an AI-based scientific and innovative material intelligent duplicate checking system, characterized in that it comprises:

[0013] a reading module that reads historical scientific research project materials, generates deconstruction information of the scientific research project materials, and the deconstruction information comprises a plurality of technical target descriptions, technical stack images and subject information abstracts;

[0014] a first generating module that establishes a quick check table of historical scientific research project materials according to the deconstruction information, and each row of the quick check table records the deconstruction information of a historical scientific research project material;

[0015] an extracting module that reads scientific and innovative materials to be checked, and extracts technical targets, technical abstracts and subject information of the scientific and innovative materials;

[0016] a comparing module that respectively compares the technical targets, technical abstracts and subject information of the scientific and innovative materials with the quick check table to obtain all matching rows;

[0017] a second generating module that respectively generates technical target duplication, technical abstract duplication and subject information duplication according to all matching rows;

[0018] a result module that generates a duplicate checking rate of the scientific and innovative materials to be checked according to the technical target duplication, technical abstract duplication and subject information duplication.

[0019] In a third aspect, the embodiments of the present specification provide an electronic device comprising a processor and a memory;

[0020] The processor is connected to the memory;

[0021] The memory is configured to store executable program codes;

[0022] The processor runs a program corresponding to the executable program codes by reading the executable program codes stored in the memory, so as to execute the method of any one of the above aspects.

[0023] In a fourth aspect, the embodiments of the present specification provide a computer readable storage medium having a computer program stored thereon, and the computer program is executed by a processor to implement the method of any one of the above aspects.

[0024] In a fifth aspect, the embodiments of the present specification provide a computer program product comprising a computer program, and the computer program is executed by a processor to implement the method of any one of the above aspects.

[0025] The technical solutions provided by some embodiments of the present specification have at least the following beneficial effects:

[0026] In the plurality of embodiments of the present specification, the AI-based scientific and innovative material intelligent duplicate checking method and system provided by the present application can calculate the scientific and innovative material duplicate checking rate by deconstructing and comparing the technical target, technical stack image and subject information in the scientific research project material, can deeply understand the core content of the material, effectively identify the content that is substantially repeated even after rewriting, and significantly improve the accuracy of duplicate checking. The AI-based scientific and innovative material intelligent duplicate checking method not only depends on the text similarity, but also needs to consider the technical target, technical abstract and subject information in multiple dimensions. Through the pre-configured matching strategy, the originality of the scientific and innovative material to be checked can be more effectively evaluated. By weighting the matching results, the duplicate checking rate of the scientific and innovative material to be checked is automatically generated, which provides a scientific basis for scientific research management personnel and helps to make more fair and transparent decisions in scientific research project approval, review and other links.

[0027] Other features and advantages of the plurality of embodiments of the present specification will be further disclosed in the following specific embodiments and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly explain the technical solutions in the embodiments of the present specification, the drawings required in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present specification, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0029] Figure 1 The scientific and innovative material intelligent duplicate checking schematic diagram provided by the embodiments of the present specification.

[0030] Figure 2 The scientific and innovative material intelligent duplicate checking method flowchart provided by the embodiments of the present specification.

[0031] Figure 3 The search matching degree schematic diagram provided by the embodiments of the present specification.

[0032] Figure 4 The scientific and innovative material intelligent duplicate checking system schematic diagram provided by the embodiments of the present specification.

[0033] Figure 5 The electronic device schematic diagram provided by the embodiments of the present specification. DETAILED DESCRIPTION

[0034] The technical solutions of the embodiments of the present specification will be explained and described below in combination with the drawings of the embodiments of the present specification. However, the following embodiments are only preferred embodiments of the present specification, not all. Based on the embodiments in the embodiments, other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present specification.

[0035] The terms "first", "second", "third", and the like in the description of the specification and claims in the specification and the above drawings are used to distinguish different objects, and are not used to describe a specific order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to the process, method, product, or device.

[0036] In the following description, the appearance of terms such as "inner", "outer", "upper", "lower", "left", "right", etc. is only for the convenience of describing the embodiments and simplifying the description, and does not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present specification.

[0037] The data involved in the present application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data comply with relevant laws, regulations and standards of relevant countries and regions.

[0038] Before introducing the technical solutions recorded in the present specification, the application scenarios of the technical solutions and related technologies are introduced.

[0039] The originality of scientific research materials is one of the important standards for evaluating their value. With the development of science and technology, the number of scientific research projects has increased rapidly, and it is particularly important to ensure the uniqueness and innovation of each scientific research material. Therefore, scientific research material duplication checking has become an important part of scientific research activities. The significance of scientific research material duplication checking is to prevent academic misconduct and protect intellectual property rights, such as plagiarism, plagiarism, etc., and to promote scientific research integrity and improve the quality of scientific research materials. Through effective duplication checking means, the authenticity and uniqueness of scientific research achievements can be guaranteed, and the fairness and transparency of the scientific community can be maintained. For example, an innovative technology company plans to apply for an annual technology evaluation award for its newly developed environmentally friendly battery technology. In order to ensure the uniqueness and innovation of the technology, and during the review process, the relevant scientific research materials of the environmentally friendly battery technology need to be checked for duplication. The duplication checking results provide strong data support for the review experts, which helps to improve the fairness and transparency of the review.

[0040] Many current duplication checking technologies mainly rely on direct comparison of text, and assess similarity by calculating the repetition rate of words between documents. Although this method can identify completely or partially copied and pasted content to some extent, its limitation lies in focusing only on the matching of the text level, without being able to check in depth on the semantic dimension. Therefore, the present specification provides an AI-based scientific and technological material intelligent duplication checking method and system, please refer to the attachedFigure 1 First, historical scientific research project materials are read and deconstruction information is generated, including technical target description, technology stack image, and subject information summary; then, a quick detection table is established according to these information. Next, the scientific research materials to be checked are analyzed, the technical target, technical summary and subject information are extracted, and compared with the information in the quick detection table to calculate the repetition rate, i.e. the duplication rate.

[0041] The present specification first provides an AI-based scientific research material intelligent duplication checking method, please refer to the attached Figure 2 , including the steps of:

[0042] Step S1) reading historical scientific research project materials, generating deconstruction information of the scientific research project materials, the deconstruction information including a plurality of technical target descriptions, technology stack images and subject information summaries.

[0043] The technical target description aims to capture the fundamental purpose of the scientific research project and the expected effect. In order to generate such a description, a large language model is used to extract the focused problem description, related technology description and effect description in the scientific research project materials. These descriptions will be fitted into a preset technical target description prompt template, which requires the description to start with the focused problem, pass through the related technology means, and finally achieve a certain effect, so as to obtain a clear description of the technical target of the scientific research project.

[0044] The technology stack image identifies and records the set of technical tools used in the scientific research project. The system reads a preset set of technical tools, which contains the technical names and descriptions of various technical tools. These technical tools are compared with the scientific research project materials to determine the matching degree between them. If the matching degree of a certain technical tool exceeds the preset threshold, it will be included in the technology stack of the project, and the corresponding use description will be generated. According to the technical tools contained in the technology stack and their use descriptions, the technology stack image is formed.

[0045] The subject information summary includes the summary of key information such as the person in charge, participants, research institutions, etc. of the scientific research project, which is used to quickly understand the background of the scientific research project.

[0046] The main purpose of step S1) is to extract key information that is helpful to the subsequent duplication checking process, i.e. technical target description, technology stack image and subject information summary, through in-depth analysis of historical scientific research project materials. These information together constitute the "fingerprint" of the scientific research project materials, so that the subsequent duplication checking work can be more accurately performed.

[0047] For example, a scientific research project named "Intelligent City Traffic Management System", the following is an example of the deconstruction information of the scientific research project materials:

[0048] Technical Objective Description: Aims to address urban traffic congestion and its environmental impact by improving traffic management efficiency through advanced data analysis techniques and AI algorithms.

[0049] Technical Stack Image: Covers a range of technical tools from data collection to model training, such as Python, Apache Kafka, and TensorFlow.

[0050] Subject Information Summary: Implemented by a team of experts from a university's intelligent transportation research center.

[0051] Among them, the step of generating the technical objective description of the scientific research project material includes:

[0052] Indicate the pre-accessed large language model to extract the focused problem description, related technology description, and effect description of the scientific research project material;

[0053] Fit the focused problem description, related technology description, and effect description into the preset technical objective description prompt template, and submit it to the large language model, and obtain the pursuit description according to the response of the large language model, the technical objective description prompt template includes a description sentence that takes the focused problem description as the starting point, the related technology description as the approach, and the effect description as the result.

[0054] The process of generating the technical objective description of the scientific research project material involves using the pre-accessed large language model to analyze and extract the core elements of the scientific research project, and generating an easy-to-understand and compare technical objective description through a specific template. The system will instruct the large language model to extract three core elements from the scientific research project material.

[0055] Focused problem description, that is, what is the main problem or challenge that the scientific research project aims to solve;

[0056] Related technology description, what technical means or methods are used in the process of trying to solve the problem;

[0057] Effect description, the expected results or impact achieved through the above technologies and methods.

[0058] Put the three extracted core elements into the preset technical objective description prompt template, the purpose of the template design is to ensure that the final generated technical objective description can clearly convey the description sentence that takes the focused problem description as the starting point, the related technology description as the approach, and the effect description as the result. After filling in the template, it is submitted to the large language model again to obtain a more fluent, accurate and logical technical objective description. This step utilizes the powerful text processing capabilities of the large language model to ensure that the final output technical objective description not only contains the necessary information, but also expresses more naturally and professionally.

[0059] The technical objectives of the "Smart City Traffic Management System" research project are as follows,

[0060] Focus problem description: Current urban traffic congestion is severe, leading to increased energy waste and environmental pollution;

[0061] Related technology description: Utilize big data analysis, Internet of Things (IoT) devices, and artificial intelligence algorithms to optimize traffic flow management;

[0062] Effect description: Expected to reduce urban center area traffic congestion time by 30%, reduce carbon emissions, and improve citizen travel efficiency.

[0063] The steps to generate the technology stack image of the research project material include:

[0064] Read the pre-set technology tool set, which records the technical name and technical description of the technology tool;

[0065] Compare the technical name and technical description in the technology tool set with the research project material respectively, and obtain the matching degree of the technology tool and the research project material;

[0066] Include technology tools with a matching degree higher than the pre-set threshold in the technology stack, and generate the usage description of the technology stack according to the research project material;

[0067] Generate the technology stack image according to the technical name of the technology tool included in the technology stack and the usage description.

[0068] The process of generating the technology stack image of the research project material is a systematic method for identifying key technologies used in the research project and building a detailed technology stack image based on this information.

[0069] Load the pre-set technology tool set, which contains detailed information of various technology tools, including but not limited to technical name and technical description.

[0070] The system compares each technical name and technical description in the technology tool set with the research project material, evaluates the relevance or matching degree between each technology tool and the current research project, and the matching degree can be calculated by text similarity algorithm, keyword matching, etc.

[0071] According to the obtained matching degree, all technology tools are screened, and the technology tools with a matching degree higher than the pre-set threshold are included in the technology stack of the project. Based on the research project material, generate detailed usage description for the technology tools, explaining how the technology tools are applied in the project. Integrate all technology tools included in the technology stack and their usage description to form a technology stack image that fully reflects the technical architecture relied on by the research project.

[0072] The technology stack of the "Smart City Traffic Management System" research project is as follows,

[0073] Technology toolset reading: Load a technology toolset containing various technical tools, such as the Python programming language, the Apache Kafka message queue system, the TensorFlow machine learning framework, etc.

[0074] Comparison and determination of matching degree: Compare these technology names and technology descriptions with the materials of the research project, and find that the project mainly uses the following technologies: Python is used for writing data analysis scripts, Apache Kafka is used for processing real-time data streams from multiple sensors, and TensorFlow is used for training deep learning models to predict traffic flow.

[0075] Incorporate technology stack and generate description: For each technology tool with a matching degree exceeding the preset threshold, incorporate it into the technology stack and generate a detailed technology stack usage description based on the project requirements. For example: Python is used as the core programming language to implement data cleaning, conversion, and visualization functions. Apache Kafka builds a distributed message delivery system to ensure efficient processing of data streams from thousands of vehicle sensors. TensorFlow is applied to build neural network models to predict future traffic conditions and propose optimal path recommendations.

[0076] Step S2) Establish a quick detection table of historical research project materials based on the deconstruction information. Each row of the quick detection table records the identification and deconstruction information of a historical research project material.

[0077] Specifically, the method for establishing a quick detection table of historical research project materials includes:

[0078] Open a cache area in the high-speed storage space;

[0079] Use the identification and deconstruction information of the historical research project materials to establish a KV value pair, and store the KV value pair in the cache area;

[0080] Persist the content stored in the cache area at a set period.

[0081] A dedicated cache area is opened in the high-speed storage space of the system. The cache area is used to temporarily store the identification and deconstruction information of the historical research project materials that are about to be processed. Using the cache area can significantly improve the speed of data reading and writing, thereby speeding up the entire duplicate detection process.

[0082] For each historical scientific research project material, the system creates a key-value (KV) pair. The "key" is usually a unique identifier of the scientific research project, such as a project number or name; while the "value" is the deconstruction information of the project. These KV value pairs are stored in the previously opened cache area for subsequent quick access and query. In this way, each scientific research project is simplified into a form that is easy to manage and compare.

[0083] To prevent data loss and ensure long-term preservation, the system will periodically persist the contents in the cache area to more stable storage media, such as hard disk or database. This ensures that even if system failure or other unexpected situations occur, the processed data will not be lost and can be reloaded into the cache for further processing when needed.

[0084] Through the above steps, the system can effectively build a quick check table containing all historical scientific research project material identification and its deconstruction information. The quick check table not only supports quick query and matching operations, but also ensures data security and reliability through regular data persistence.

[0085] Step S3) Read the scientific and technological materials to be checked, and extract the technical target, technical summary and main information of the scientific and technological materials.

[0086] The system needs to obtain the scientific and technological materials to be checked, identify and extract the technical target of the project from the scientific and technological materials. The technical target usually refers to the core problem, target or expected effect of the scientific research project. Natural language processing technology is used to analyze the text content, identify the focused problem description, related technology description and effect description, and generate the final technical target description through the preset technical target description prompt template.

[0087] Extract the technical summary from the scientific and technological materials. The technical summary summarizes the technical means, tools and methods used in the scientific research project, including but not limited to identifying specific technical names mentioned in the project, technical description and application method. This process also relies on natural language processing technology to accurately capture technology-related information and form a concise and clear technology stack image.

[0088] Extract the main information in the scientific and technological materials. The main information usually contains the background information of the scientific research project, such as the person in charge, participants, research institutions, etc., which helps to understand the background and implementation environment of the scientific research project, and is very important for comprehensive evaluation of the uniqueness and originality of the scientific research project.

[0089] The system can effectively extract necessary information from the to-be-duplicated scientific and technological materials, including technical targets, technical summaries, and main body information. The extracted information will be used to compare with the quick check table of historical scientific research project materials to calculate the duplication rate and judge the novelty and uniqueness of the scientific and technological materials.

[0090] Step S4) Compare the technical target, technical summary, and main body information of the scientific and technological material with the quick check table respectively to obtain all matched rows.

[0091] Specifically, please refer to the attached Figure 3 The step of obtaining all matched rows includes:

[0092] Compare the technical target of the scientific and technological material with the technical target description of the row in the quick check table to obtain a technical target matching degree.

[0093] Compare the technical summary of the scientific and technological material with the technical stack image of the row in the quick check table to obtain a technical matching degree.

[0094] Compare the main body information of the scientific and technological material with the main body information summary of the row in the quick check table to obtain a main body matching degree.

[0095] Generate a first coefficient and a second coefficient with an initial value of 1.

[0096] When the technical target matching degree is higher than a preset first threshold, update the value of the first coefficient to be greater than 1; when the technical target matching degree is not higher than the preset first threshold, update the value of the first coefficient to be less than 1.

[0097] When the main body matching degree is higher than a preset second threshold, update the value of the second coefficient to be less than 1; when the main body matching degree is not higher than the preset second threshold, update the value of the second coefficient to be greater than 1.

[0098] According to the technical matching degree, the first coefficient, and the second coefficient, obtain a search matching degree of the scientific and technological material and the row.

[0099] When the search matching degree is higher than a preset reference threshold, the row is taken as a matched row.

[0100] Compare the technical target of the to-be-duplicated scientific and technological material with the "technical target description" of the historical scientific research project of each row record in the quick check table to obtain a technical target matching degree representing the similarity between the two. Compare the technical summary of the to-be-duplicated scientific and technological material with the "technical stack image" of the corresponding row in the quick check table to evaluate the similarity of the two in technical application and thus obtain a technical matching degree.

[0101] Comparing the subject information of the to-be-duplicated innovative material with the "subject information summary" of the corresponding row in the quick detection table helps to identify possible repetitive research work performed by the same or related team, resulting in a subject matching degree.

[0102] If the technical target matching degree is higher than the preset first threshold value, it is considered that the technical target has high similarity, and the value of the first coefficient is updated to be greater than 1; otherwise, if the technical target matching degree is not high, the value of the first coefficient is reduced to be less than 1.

[0103] For the subject matching degree, if it is higher than the preset second threshold value, the value of the second coefficient is updated to be less than 1, which means that the high similarity of the subject information may indicate low uniqueness; otherwise, the value of the second coefficient is greater than 1.

[0104] Combining the matching degrees of the above three dimensions and the adjusted first coefficient and second coefficient, the system calculates a comprehensive search matching degree. This matching degree reflects the overall similarity between the to-be-duplicated innovative material and each historical scientific research project in the quick detection table. Finally, if the search matching degree of a certain historical scientific research project exceeds the preset reference threshold value, the item is considered to have significant duplication or similarity with the to-be-duplicated innovative material, and is marked as a matched row.

[0105] Suppose the to-be-duplicated innovative material is project X, and its deconstruction information is as follows:

[0106] Technical target: Solve the problem of insufficient medical resources in remote areas, improve the diagnosis accuracy rate by 20% through a remote medical system. Technical summary: Real-time audio and video communication based on WebRTC, React Native development of mobile application, AWS Lambda serverless computing. Subject information: Remote medical department of a hospital, Dr. Wang's team.

[0107] Step S4) Specific operation example, technical target matching degree calculation: The technical target of the to-be-checked material (project X) is "to solve the problem of insufficient medical resources in remote areas, and improve the diagnosis accuracy rate by 20% through a remote medical system". The technical target description of the historical project (ProjectID_C) is "to provide convenient medical services for residents in remote areas and improve the diagnosis accuracy rate by 20%". The matching degree is calculated: through a text similarity algorithm, it is found that the focused problem and effect description of the two are highly consistent, and the technical target matching degree is 0.9. Technical matching degree calculation: The technical summary of the to-be-checked material (project X) is React Native (mobile terminal), WebRTC (real-time communication), and AWS Lambda (serverless computing).

[0108] ProjectID_C: React Native, WebRTC, AWS Lambda. Matching degree calculation: the technical tools are completely consistent, and the technical matching degree is 1.0. Subject matching degree calculation: the subject information of the material to be investigated (Project X): the remote medical department of a hospital, Dr. Wang's team. The subject information summary of the historical project (ProjectID_C): the remote medical department of a hospital, Dr. Wang and others.

[0109] Matching degree calculation: the subject information (institution and responsible person) is completely consistent, and the subject matching degree is 1.0. Coefficient adjustment: initial value: first coefficient = 1.0, second coefficient = 1.0. Rule includes that the technical target matching degree is higher than the first threshold value (assuming 0.8): update the first coefficient to 1.2. The subject matching degree is higher than the second threshold value (assuming 0.8): update the second coefficient to 0.8. After adjustment: first coefficient = 1.2 (technical target matching degree 0.9 > 0.8), second coefficient = 0.8 (subject matching degree 1.0 > 0.8).

[0110] The search matching degree calculation specifically includes the formula: search matching degree = technical matching degree x first coefficient x second coefficient. After substituting the numerical values, the search matching degree = 1.0 x 1.2 x 0.8 = 0.96. Determine whether it matches, the preset reference threshold value is assumed to be 0.9. Result: search matching degree 0.96 > 0.9, the final result is matching behavior ProjectID_C, the historical project is highly similar to the material to be investigated.

[0111] Key logic explanation, technical target matching degree dominance includes that if the technical target matching degree is high (such as 0.9), the first coefficient is amplified (1.2), emphasizing the importance of target similarity. Subject information suppression repetition includes that if the subject matching degree is high (such as 1.0), the second coefficient is reduced (0.8), suppressing the excessively high matching degree caused by repeated submission of the same team. Technical matching degree is the core, including the complete consistency of the technology stack (1.0) is the key factor of the final matching.

[0112] Step S5) According to all the matched rows, respectively generate technical target repetition degree, technical summary repetition degree and subject information repetition degree.

[0113] The specific steps include:

[0114] According to the maximum value of the technical target matching degree of the scientific and creative material and all the matched rows, obtain the initial value of the technical target repetition degree;

[0115] According to the maximum value of the technical matching degree of the scientific and creative material and all the matched rows, obtain the initial value of the technical summary repetition degree;

[0116] According to the maximum value of the subject information matching degree of the scientific and technological material and all matching rows, obtain the subject information repetition degree;

[0117] When the subject information repetition degree is higher than the preset reference value, the initial value of the technical target repetition degree is reduced as the final technical target repetition degree, and the initial value of the technical summary repetition degree is reduced as the final technical target repetition degree.

[0118] Step S5) Further calculate specific repetition degree indexes according to all matching rows obtained in step S4). These repetition degree indexes help to quantify the similarity degree between the scientific and technological material to be checked and the historical scientific research project.

[0119] The system will check all the rows determined to be matched in step S4), and find the maximum value of the technical target matching degree from them. This maximum value reflects the highest similarity in technical targets between the scientific and technological material to be checked and the most similar historical scientific research project. The maximum value is used as the initial value of the technical target repetition degree.

[0120] The system will find the maximum value of the technical matching degree from all the matching rows to represent the highest similarity in technical implementation between the scientific and technological material to be checked and the closest historical scientific research project. This maximum value is used as the initial value of the technical summary repetition degree.

[0121] The system will also extract the maximum value of the subject matching degree from all the matching rows, which represents the highest similarity in research team or subject between the scientific and technological material to be checked and a historical scientific research project. This maximum value directly becomes the subject information repetition degree.

[0122] If the subject information repetition degree is higher than the preset reference value, it means that the scientific and technological material to be checked has a high similarity in research subject with some historical scientific research projects. In order to reduce the risk of over-high repetition evaluation caused by the same subject, the initial values of the technical target repetition degree and the technical summary repetition degree are adjusted downward to generate the final technical target repetition degree and the technical summary repetition degree.

[0123] Step S6) According to the technical target repetition degree, the technical summary repetition degree and the subject information repetition degree, generate the duplication rate of the scientific and technological material to be checked.

[0124] The specific steps include:

[0125] Obtain the search matching degree of the row corresponding to the maximum value of the technical target matching degree and the technical matching degree, respectively;

[0126] Obtain the weighted average value of the technical target repetition degree and the technical summary repetition degree by taking the search matching degree as the weight;

[0127] The quotient of the preset constant and the subject information repetition degree is calculated, and the quotient is multiplied by the weighted average value to obtain the duplication rate of the scientific and technological material to be checked.

[0128] Step S6) is to generate the overall duplication rate of the scientific and technological material to be checked according to the technical target repetition degree, the technical abstract repetition degree and the subject information repetition degree calculated before.

[0129] The row corresponding to the maximum value of the technical target matching degree and the maximum value of the technical matching degree are determined, and the retrieval matching degrees of these rows are obtained. The retrieval matching degree reflects the similarity of the scientific and technological material to be checked and the historical scientific research project in the comprehensive level.

[0130] The technical target repetition degree and the technical abstract repetition degree are calculated by using the retrieval matching degree obtained above as a weight. If a historical scientific research project is very similar to the scientific and technological material to be checked in the technical target or the technical implementation, the related repetition degree of this project will have a greater proportion in the final weighted average value.

[0131] The quotient of the preset constant and the subject information repetition degree is calculated. The "preset constant" is a fixed value for adjusting the proportion relationship of the overall calculation. The quotient is multiplied by the weighted average value calculated before to obtain the final duplication rate, which further considers the influence of the repetition of the research subject on the overall duplication result. If the subject information repetition degree is high, it means that there may be a high risk of duplication, thereby affecting the final duplication rate.

[0132] On the other hand, in another embodiment, the scheme described in the present embodiment also calculates the text similarity and the text similarity based on synonyms, and calculates the final duplication rate according to the text similarity, the text similarity based on synonyms, the technical target repetition degree, the technical abstract repetition degree and the subject information repetition degree.

[0133] On the other hand, the present specification provides an AI-based intelligent scientific and technological material duplication checking system, please refer to the attached Figure 4 , comprising:

[0134] The reading module 100 reads the historical scientific research project materials and generates the deconstruction information of the scientific research project materials, wherein the deconstruction information includes a plurality of technical target descriptions, technical stack images and subject information abstracts.

[0135] The first generation module 200 establishes a quick detection table of the historical scientific research project materials according to the deconstruction information, and each row of the quick detection table records the deconstruction information of a historical scientific research project material.

[0136] The extraction module 300 reads the scientific and technological material to be checked and extracts the technical target, the technical abstract and the subject information of the scientific and technological material.

[0137] The comparison module 400 compares the technical target, the technical abstract and the main body information of the scientific and innovative material with the quick detection table respectively, and obtains all matching rows;

[0138] The second generation module 500 generates the technical target repetition degree, the technical abstract repetition degree and the main body information repetition degree according to the all matching rows;

[0139] The result module 600 generates the duplication checking rate of the scientific and innovative material to be checked according to the technical target repetition degree, the technical abstract repetition degree and the main body information repetition degree.

[0140] Please refer to Figure 5 The electronic device 1100 shown in the embodiment of the present specification provides a structural schematic diagram of an electronic device.

[0141] As Figure 5 shown, the electronic device 1100 can include at least one processor 1101, at least one network interface 1104, a user interface 1103, a memory 1105 and at least one communication bus 1102. The communication bus 1102 can be used to realize the connection and communication of the above-mentioned components. The user interface 1103 can include a key, and the optional user interface can also include a standard wired interface, a wireless interface. The network interface 1104 can include but is not limited to a Bluetooth module, an NFC module, a Wi-Fi module, etc. The processor 1101 can include one or more processing cores. The processor 1101 connects various parts in the entire electronic device 1100 through various interfaces and lines, executes various functions of the routing device and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 1105, and calling data stored in the memory 1105. Optionally, the processor 1101 can be realized by at least one of DSP, FPGA and PLA. The processor 1101 can integrate CPU, GPU and modem, etc. The CPU mainly processes operating systems, user interfaces and application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; and the modem is used for processing wireless communication.

[0142] It can be understood that the above-mentioned modem can also not be integrated into the processor 1101, but be realized by a separate chip.

[0143] The memory 1105 can include a RAM and also can include a ROM. Optionally, the memory 1105 includes a non-transitory computer-readable medium. The memory 1105 can be used to store instructions, programs, codes, code sets, or instruction sets. The memory 1105 can include a program storage area and a data storage area, where the program storage area can store the instructions for implementing the operating system, the instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), the instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area can store the data involved in the above-mentioned various method embodiments, etc. The memory 1105 can also be at least one storage device located away from the aforementioned processor 1101. The memory 1105, as a computer storage medium, can include an operating system, a network communication module, a user interface module, and an application program. The processor 1101 can be used to invoke the application program stored in the memory 1105 and execute the method in the above-mentioned various embodiments.

[0144] The embodiments of the present specification also provide a computer-readable storage medium, which stores instructions, and when the instructions run on a computer or a processor, the computer or the processor executes the steps in the above-mentioned embodiments. The various constituent modules of the above-mentioned electronic device, if realized in the form of software function units and sold or used as independent products, can be stored in the computer-readable storage medium.

[0145] The embodiments of the present specification also provide a computer program product, which includes a computer program, and the computer program is executed by a processor to realize the steps in the above-mentioned embodiments.

[0146] The technical features in the embodiments and the implementation forms can be combined arbitrarily without conflict.

[0147] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes a plurality of computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present specification are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in or transmitted by a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through a wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. integrated with a plurality of available media. The available media can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a digital versatile disc (DVD)), or a semiconductor medium (for example, a solid state disk (SSD)) and the like.

[0148] When implemented by hardware or firmware, the foregoing method processes are programmed into a hardware circuit to obtain a corresponding hardware circuit structure, and the corresponding functions are implemented. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is an integrated circuit whose logic function is determined by the user programming the device. A digital system is "integrated" on a PLD by the designer himself programming, without having to ask the chip manufacturer to design and manufacture a special integrated circuit chip. Moreover, instead of manually making integrated circuit chips, such programming is now mostly implemented using "logic compiler" software, which is similar to the software compiler used when developing programs, and the original code to be compiled must also be written in a specific programming language, which is called a hardware description language (HDL), and there are many HDLs. Those skilled in the art should also understand that only the method processes need to be logically programmed in the above-mentioned several hardware description languages and programmed into integrated circuits to easily obtain hardware circuits that implement the logical method processes.

[0149] The above-described embodiments are merely exemplary diagnostic description rather than limitations on the scope of the present specification. Any modifications, equivalent replacements, and improvements made to the technical solutions of the present specification by those of ordinary skill in the art, without departing from the design spirit of the present specification, shall fall within the protective scope of the claims of the present specification.

Claims

1. An AI-based scientific and innovative material intelligent duplicate checking method, characterized in that, The method comprises the steps of: reading historical scientific research project materials, generating deconstruction information of the scientific research project materials, the deconstruction information including a plurality of technical target descriptions, technical stack images, and subject information summaries, the subject information summaries including summaries of the principal investigator, participants, and research institutions of the scientific research project, for quickly understanding the background of the scientific research project; establishing a quick detection table of the historical scientific research project materials according to the deconstruction information, each row of the quick detection table recording an identification and deconstruction information of a historical scientific research project material; reading the scientific and creative materials to be checked for duplication, extracting the technical target, technical summary, and subject information of the scientific and creative materials; comparing the technical target, technical summary, and subject information of the scientific and creative materials with the quick detection table respectively to obtain all matching rows; generating a technical target duplication degree, a technical summary duplication degree, and a subject information duplication degree according to the all matching rows; generating a duplication rate of the scientific and creative materials to be checked for duplication according to the technical target duplication degree, the technical summary duplication degree, and the subject information duplication degree; the step of generating the technical target description of the scientific research project material comprises: indicating a pre-accessed large language model to extract the focused problem description, involved technology description, and effect description of the scientific research project material; fitting the focused problem description, involved technology description, and effect description into a preset technical target description prompt template, and submitting to the large language model to obtain a technical target description according to the response of the large language model, the technical target description prompt template including description sentences starting with the focused problem description, involving the technology description as the way, and the effect description as the result.

2. The AI-based scientific and creative material intelligent duplication checking method according to claim 1, wherein the step of generating the technical stack image of the scientific research project material comprises: reading a preset technical tool set, the technical tool set recording technical names and technical descriptions of technical tools; comparing the technical names and technical descriptions in the technical tool set with the scientific research project material respectively to obtain a matching degree of the technical tools with the scientific research project material; including the technical tools with a matching degree higher than a preset threshold in a technical stack, and generating a use description of the technical stack according to the scientific research project material; generating a technical stack image according to the technical names of the technical tools included in the technical stack and the use description.

3. The AI-based scientific and creative material intelligent duplication checking method according to any one of claims 1 to 2, wherein the step of comparing the technical target, technical summary, and subject information of the scientific and creative materials with the quick detection table respectively to obtain all matching rows comprises: comparing the technical target of the scientific and creative materials with the technical target description of the row in the quick detection table to obtain a technical target matching degree; comparing the technical summary of the scientific and creative materials with the technical stack image of the row in the quick detection table to obtain a technical matching degree; comparing the subject information of the scientific and creative materials with the subject information summary of the row in the quick detection table to obtain a subject matching degree; generating a first coefficient and a second coefficient with an initial value of 1; When the technical target matching degree is higher than the preset first threshold, the value of the first coefficient is greater than 1, and when the technical target matching degree is not higher than the preset first threshold, the value of the first coefficient is less than 1; When the subject matching degree is higher than the preset second threshold, the value of the second coefficient is less than 1, and when the subject matching degree is not higher than the preset second threshold, the value of the second coefficient is greater than 1; According to the technical matching degree, the first coefficient and the second coefficient, the retrieval matching degree of the scientific and innovative material and the line is obtained; When the retrieval matching degree is higher than the preset reference threshold, the line is taken as a matched line.

4. The AI-based scientific and innovative material intelligent duplicate checking method according to claim 3, wherein the step of generating a technical target duplication degree, a technical abstract duplication degree and a subject information duplication degree according to all matched lines comprises: According to the maximum value of the technical target matching degree of the scientific and innovative material and all matched lines, an initial value of the technical target duplication degree is obtained; According to the maximum value of the technical matching degree of the scientific and innovative material and all matched lines, an initial value of the technical abstract duplication degree is obtained; According to the maximum value of the subject matching degree of the scientific and innovative material and all matched lines, a subject information duplication degree is obtained; When the subject information duplication degree is higher than the preset reference value, the initial value of the technical target duplication degree is reduced as the final technical target duplication degree, and the initial value of the technical abstract duplication degree is reduced as the final technical target duplication degree.

5. The AI-based scientific and innovative material intelligent duplicate checking method according to claim 4, wherein the step of generating a duplicate checking rate of the scientific and innovative material to be checked according to the technical target duplication degree, the technical abstract duplication degree and the subject information duplication degree comprises: The retrieval matching degrees of the lines corresponding to the maximum values of the technical target matching degree and the technical matching degree are obtained respectively; The weighted average value of the technical target duplication degree and the technical abstract duplication degree is obtained by taking the retrieval matching degrees as weights; The quotient of a preset constant and the subject information duplication degree is calculated, and the quotient multiplied by the weighted average value is taken as the duplicate checking rate of the scientific and innovative material to be checked.

6. The AI-based scientific and innovative material intelligent duplicate checking method according to any one of claims 1 to 2, wherein the method for establishing a fast checking table of historical scientific research project materials comprises: A cache area is opened in a high-speed storage space; A KV value pair is established using the identification and deconstruction information of the historical scientific research project materials, and the KV value pair is stored in the cache area; The content stored in the cache area is persisted at a set period. Comprise: A reading module reads historical scientific research project materials and generates deconstruction information of the scientific research project materials, the deconstruction information comprising a plurality of technical target descriptions, technical stack images and subject information abstracts, and the subject information abstracts comprising summaries of the person in charge, the participants and the research institution of the scientific research project, for quickly understanding the background of the scientific research project; A first generation module establishes a fast checking table of historical scientific research project materials according to the deconstruction information, and each line of the fast checking table records the deconstruction information of a historical scientific research project material.

7. An AI-based scientific and innovative material intelligent duplicate checking system, characterized in that, ​ ​ ​ The extraction module reads the scientific and technological materials to be checked for duplication, extracts the technical target, technical abstract and subject information of the scientific and technological materials; The comparison module respectively compares the technical target, technical abstract and subject information of the scientific and technological materials with the quick detection table to obtain all matched rows; The second generation module respectively generates the technical target repetition degree, technical abstract repetition degree and subject information repetition degree according to the all matched rows; The result module generates the duplication checking rate of the scientific and technological materials to be checked for duplication according to the technical target repetition degree, technical abstract repetition degree and subject information repetition degree; The step of generating the technical target description of the scientific research project material comprises: Indicating a pre-accessed large language model to extract the focused problem description, related technology description and effect description of the scientific research project material; The focused problem description, related technology description and effect description are fitted into a preset technical target description prompt template, and submitted to the large language model, and a technical target description is obtained according to the response of the large language model, and the technical target description prompt template comprises a description sentence describing the focused problem description as the starting point, the related technology description as the approach and the effect description as the result.

8. An electronic device, comprising: comprise a processor and a memory; The processor is connected with the memory; The memory is used for storing executable program codes; The processor runs the program corresponding to the executable program codes by reading the executable program codes stored in the memory, so as to execute the method according to any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the method according to any one of claims 1-6.

Images