Scientific achievement data analysis method, system, equipment and medium

Abstract

The invention relates to the technical field of medical data processing, and discloses a scientific research achievement data analysis method, system, equipment and medium, hospital scientific research accurate recommendation is realized by integrating multi-source data, and the method has the following effects: (1) medical field hotspot trend data is intelligently analyzed based on authoritative hospital data; (2) matching the target hospital resource data with the scientific research trend data to generate a practical scientific research direction suggestion; and (3) the recommendation direction is refined in combination with an individual research background, and scientific researchers are assisted to locate high-potential subjects. According to the method, cross-hospital data, hospital resources and personal data are integrated, suggestions are ensured to be able to fall through multi-level matching, the scientific research planning efficiency is remarkably improved through automatic analysis, and the problems that an existing method is insufficient in analysis depth and low in intelligent level are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of scientific research achievement data analysis, and in particular to a scientific research achievement data analysis method, system, equipment and medium. Background Art

[0002] Current scientific research data analysis systems face the following key issues: First, data analysis capabilities are weak. Existing functions only provide basic statistics and visualization, making it difficult to establish correlations between scientific research results and the actual needs of hospital scientific research data analysis (such as research trends and improving disciplinary competitiveness). Furthermore, they are unable to conduct correlation analysis and in-depth mining of multidimensional data such as papers, patents, research projects, and scientific and technological awards, resulting in limited guiding value of the analysis results for hospital management. Second, intelligent services are lacking. The system lacks artificial intelligence technology support (such as natural language processing and machine learning), fails to establish predictive models based on disciplinary research trends, and lacks personalized recommendation functions for researchers (such as intelligent push notifications for project and award applications, and quantitative assessments of patent applications and the potential for transformation of scientific research results). This makes it difficult to provide data support and precise guidance for hospital scientific research strategic decision-making, disciplinary development planning, and the transfer and industrialization of scientific and technological achievements. These issues have seriously hindered the improvement of scientific research management efficiency and the progress of scientific research innovation.

[0003] Therefore, a new method for analyzing scientific research results data is urgently needed. Summary of the Invention

[0004] The present invention provides a scientific research achievement data analysis method, system, equipment and medium to solve the defects of existing scientific research achievement data analysis methods, such as insufficient analysis depth, low intelligence level, and difficulty in meeting the complex needs of scientific research management.

[0005] The present invention provides a method for analyzing scientific research results data, comprising:

[0006] Obtain scientific research results text data from multiple other hospitals, resource text data from the target hospital, and personal scientific research text data. The multiple other hospitals include several hospitals that rank high in various medical fields.

[0007] Based on the scientific research results text data of multiple target hospitals, obtain the scientific research trend data in various medical fields;

[0008] Match the target hospital's resource text data with research trend data in various medical fields to obtain recommendations on the target hospital's research direction;

[0009] Match the individual scientific research text data of the target hospital with the scientific research direction recommendation suggestions of the target hospital to obtain the scientific research direction recommendation suggestions of individuals in the target hospital.

[0010] According to a scientific research achievement data analysis method provided by the present invention, the scientific research achievement text data includes at least one of the following text data or any combination thereof: papers, patents, scientific and technological awards, scientific research projects, and achievement transformation.

[0011] According to a scientific research achievement data analysis method provided by the present invention, the scientific research trend data of various medical fields is obtained based on the scientific research achievement text data of multiple target hospitals, including:

[0012] Preprocess the scientific research results text data of multiple target hospitals;

[0013] Obtain text features based on the pre-processed scientific research results text data of multiple target hospitals;

[0014] According to the text features, the topic vocabulary is obtained through the topic model;

[0015] Based on the subject vocabulary, keywords are obtained through unsupervised keyword extraction;

[0016] Filter and cluster keywords to obtain research trend data in various medical fields;

[0017] Visualize scientific research trend data in various medical fields.

[0018] According to a scientific research achievement data analysis method provided by the present invention, preprocessing includes at least one of the following or any combination thereof: text cleaning, word segmentation, and removal of stop words.

[0019] According to a scientific research achievement data analysis method provided by the present invention, the topic model is an LDA (Latent Dirichlet Allocation) or NMF (Non-negative Matrix Factorization) topic model.

[0020] According to a scientific research achievement data analysis method provided by the present invention, the unsupervised keyword extraction method is TextRank or YAKE!.

[0021] According to a scientific research achievement data analysis method provided by the present invention, resource text data includes at least one of the following text data or any combination thereof: advantageous field data, scientific research resources, scientific researcher data, and patient data.

[0022] According to a scientific research achievement data analysis method provided by the present invention, the resource text data of the target hospital is matched with the scientific research trend data of various medical fields to obtain the scientific research direction recommendation of the target hospital, including:

[0023] Calculate the matching degree of scientific research directions based on the target hospital's advantageous field data and scientific research trend data in various medical fields;

[0024] Calculate resource matching based on the target hospital's scientific research resources and research trend data in various medical fields;

[0025] Calculate the matching degree of scientific research trends based on the target hospital's scientific research personnel data and the scientific research trend data in various medical fields;

[0026] Based on the research direction matching calculation results, resource matching calculation results, and research trend matching calculation results of the target hospital corresponding to the research trends in various medical fields, the comprehensive research matching degree of the target hospital corresponding to the research trends in various medical fields is obtained;

[0027] Based on the comprehensive scientific research matching degree of the target hospital and the scientific research trends in various medical fields, we can obtain the recommended scientific research direction of the target hospital.

[0028] According to a scientific research achievement data analysis method provided by the present invention, the scientific research direction matching degree calculation is performed based on the target hospital's advantageous field data and the scientific research trend data of various medical fields, including:

[0029] Based on the scientific research resources of the target hospital and the scientific research trend data of various medical fields, the cosine similarity between each scientific research direction of the target hospital and the scientific research trends of various medical fields is calculated to obtain the scientific research direction matching calculation results of the target hospital corresponding to the scientific research trends of various medical fields.

[0030] According to a scientific research achievement data analysis method provided by the present invention, the resource matching degree calculation is performed based on the scientific research resources of the target hospital and the scientific research trend data of various medical fields, including:

[0031] Using preset rules, we determine whether the research trends in various medical fields fall within the key disciplines supported by the target hospital, and select research directions that fall within the key disciplines supported by the target hospital;

[0032] Based on the scientific research resources of the target hospital, the funding support ranking of the scientific research directions within the key discipline support scope of the target hospital is obtained;

[0033] Based on the number of scientific research directions within the scope of support of the key disciplines of the target hospital and the ranking of funding support for scientific research directions within the scope of support of the key disciplines of the target hospital, the first expression is used to obtain the resource matching calculation results of the scientific research trends of the target hospital in various medical fields.

[0034] According to a scientific research achievement data analysis method provided by the present invention, the first expression is:

[0035]

[0036] In the first expression, SS represents the resource matching degree of the target hospital corresponding to the scientific research trends in various medical fields, r represents the funding support ranking of the scientific research directions within the key discipline support scope of the target hospital, and N represents the number of scientific research directions within the key discipline support scope of the target hospital.

[0037] According to a scientific research achievement data analysis method provided by the present invention, the scientific research trend matching degree calculation is performed based on the scientific research personnel data of the target hospital and the scientific research trend data of various medical fields, including:

[0038] Based on the target hospital's scientific researcher data, we can obtain the target hospital's corresponding paper data in various medical fields;

[0039] According to the paper data of the target hospital in each medical field, the second expression is used to obtain the trend growth index of the target hospital in each medical field;

[0040] The trend growth index of the target hospital corresponding to each medical field is standardized to obtain the scientific research trend matching calculation results of the target hospital corresponding to each medical field.

[0041] According to a scientific research achievement data analysis method provided by the present invention, the second expression is:

[0042]

[0043] In the second expression, TGI i represents the trend growth index of the target hospital in the medical field i, C i,t represents the number of papers published in the target hospital in the corresponding medical field i in the previous year t, t = 1, 2, ..., T, t = T represents the latest year, w t represents the time weight, for example, the linear weight w t =t / T or index weight w t =2 t-T , if C i,t-1 =0, then ignore the year.

[0044] According to a scientific research achievement data analysis method provided by the present invention, the trend growth index of the target hospital corresponding to each medical field is standardized to obtain the scientific research trend matching calculation result of the target hospital corresponding to each medical field, including:

[0045] The third expression is used to standardize the trend growth index of the target hospital corresponding to each medical field, and the calculation results of the scientific research trend matching degree of the target hospital corresponding to each medical field are obtained.

[0046] According to a scientific research achievement data analysis method provided by the present invention, the third expression is:

[0047]

[0048] In the third expression, TS i Indicates the matching degree of the target hospital’s research trend in medical field i, TGI i It represents the trend growth index of the target hospital corresponding to medical field i, max(TGI) represents the maximum value of the trend growth index of the target hospital corresponding to each medical field, and min(TGI) represents the minimum value of the trend growth index of the target hospital corresponding to each medical field.

[0049] According to a scientific research achievement data analysis method provided by the present invention, the scientific research achievement data analysis method provided by the present invention calculates the scientific research trend matching degree based on the scientific research personnel data of the target hospital and the scientific research trend data of various medical fields, including:

[0050] Based on the target hospital's scientific researcher data, we can obtain the target hospital's corresponding paper data in various medical fields;

[0051] According to the paper data of the target hospital in various medical fields, the fourth expression is used to obtain the scientific research trend matching calculation results of the target hospital in various medical fields.

[0052] According to a scientific research achievement data analysis method provided by the present invention, the fourth expression is:

[0053]

[0054] In the fourth expression, TS′ i represents the matching degree of the target hospital’s research trend in medical field i, C i,t represents the number of papers published in medical field i in the previous t year, t = 1, 2, ..., T, t = T represents the latest year, max(∑C) represents the maximum value of the total number of papers published in each medical field by the target hospital within T years, and min(∑C) represents the minimum value of the total number of papers published in each medical field by the target hospital within T years.

[0055] According to a scientific research achievement data analysis method provided by the present invention, the comprehensive scientific research matching degree of the target hospital corresponding to the scientific research trends in various medical fields is obtained based on the scientific research direction matching degree calculation results, resource matching degree calculation results, and scientific research trend matching degree calculation results of the target hospital corresponding to the scientific research trends in various medical fields, including:

[0056] Based on the calculation results of the scientific research direction matching degree, resource matching degree, and scientific research trend matching degree of the target hospital corresponding to the scientific research trends in various medical fields, the fifth expression is used to obtain the comprehensive scientific research matching degree of the target hospital corresponding to the scientific research trends in various medical fields. The fifth expression is:

[0057] Z = α·scientific research direction matching degree + β·resource matching degree + γ·scientific research trend matching degree,

[0058] In the fifth expression, α, β, and γ represent weight parameters.

[0059] According to a scientific research achievement data analysis method provided by the present invention, the target hospital's scientific research direction recommendation is obtained based on the comprehensive scientific research matching degree of the target hospital's scientific research trends in various medical fields, including:

[0060] According to the comprehensive scientific research matching degree of the target hospital's corresponding scientific research trends in various medical fields, the scientific research trends in various medical fields are sorted in descending order, and the top N scientific research trends are obtained as the scientific research recommendation directions for the target hospital; or

[0061] A preset threshold is set, and the comprehensive scientific research matching degree of the target hospital's scientific research trends in various medical fields is compared with the preset threshold. The scientific research trend greater than the preset threshold is used as the recommended scientific research direction of the target hospital, wherein the preset threshold is set manually in advance or through the permutation test method.

[0062] According to a scientific research achievement data analysis method provided by the present invention, the resource text data of the target hospital is matched with the scientific research trend data of various medical fields to obtain a scientific research direction recommendation for the target hospital, and further includes:

[0063] Obtain the number of papers published and the number of cases corresponding to the research trends in various medical fields of the target hospital;

[0064] Based on the number of papers published and the number of cases corresponding to the target hospital's scientific research trends in various medical fields, time series analysis is performed using the moving average method or exponential smoothing method, combined with derivatives and second-order derivatives, to determine whether the target hospital's scientific research trends in various medical fields are stable.

[0065] According to a scientific research achievement data analysis method provided by the present invention, the resource text data of the target hospital is matched with the scientific research trend data of various medical fields to obtain a scientific research direction recommendation for the target hospital, and further includes:

[0066] Obtain historical funding data on research trends in various medical fields for the target hospital;

[0067] Based on the historical funding data of the target hospital's scientific research trends in various medical fields, the funding forecasting model is used to predict the funding expenditure trends of the target hospital's scientific research trends in various medical fields.

[0068] According to a scientific research achievement data analysis method provided by the present invention, the individual scientific research text data of a target hospital is matched with the scientific research direction recommendation suggestions of the target hospital to obtain the scientific research direction recommendation suggestions for individuals in the target hospital, including:

[0069] Obtain project application information for the target hospital's recommended research directions;

[0070] Build user profiles of individuals in the target hospital based on the individual scientific research text data of the target hospital;

[0071] Based on the project application information of the corresponding target hospital's recommended scientific research direction and the user portraits of individuals in the target hospital, the matching degree between individuals in the target hospital and the scientific research projects involved in the scientific research recommended directions of each target hospital is calculated to obtain recommended scientific research directions for individuals in the target hospital.

[0072] According to a scientific research achievement data analysis method provided by the present invention, project application information includes any one of the following or any combination thereof: basic project information (project name, funding agency, project type, application deadline, etc.), project application conditions (professional title requirements, research direction, institutional requirements, etc.), and project keywords (project research field, keywords, etc.).

[0073] According to a scientific research achievement data analysis method provided by the present invention, personal scientific research text data includes any of the following text data or any combination thereof: professional title, age, education background, major, research direction, personal scientific research achievement text data (papers published, projects participated in, awards received, etc.).

[0074] According to a method for analyzing scientific research achievement data provided by the present invention, obtaining project application information corresponding to the target hospital's recommended scientific research direction includes:

[0075] Encode and vectorize the project application information of the scientific research recommended direction of the corresponding target hospital.

[0076] According to a scientific research achievement data analysis method provided by the present invention, constructing a user profile of an individual in a target hospital based on the individual scientific research text data of the target hospital includes:

[0077] Encode and vectorize the individual scientific research text data of the target hospital;

[0078] Based on the encoded and vectorized personal scientific research text data, a multi-layer perceptron is used to obtain the embedded representation of the user portrait of individuals in the target hospital as the historical research vector of individuals in the target hospital.

[0079] According to a scientific research achievement data analysis method provided by the present invention, the matching degree between individuals in the target hospital and the scientific research projects involved in the scientific research recommended directions of each target hospital is calculated based on the project application information of the scientific research recommended directions of the target hospital and the user profiles of individuals in the target hospital, and the scientific research direction recommendation suggestions for individuals in the target hospital are obtained, including:

[0080] Based on the project application information of the corresponding target hospital's recommended scientific research direction and the user portraits of individuals in the target hospital, the maximum similarity between the historical research vectors of individuals in the target hospital and the scientific research projects involved in the scientific research recommended directions of each target hospital is calculated to obtain recommended scientific research directions for individuals in the target hospital.

[0081] According to a scientific research achievement data analysis method provided by the present invention, the individual scientific research text data of the target hospital is matched with the scientific research direction recommendation suggestions of the target hospital to obtain the scientific research direction recommendation suggestions for individuals in the target hospital, and further includes:

[0082] Obtain scientific and technological award data for scientific research projects related to scientific research trends in various medical fields, and obtain scientific and technological award data for scientific research recommended directions for individuals in the corresponding target hospitals.

[0083] A method for analyzing scientific research results data according to the present invention further includes:

[0084] Obtain the registered papers on the research trends of various medical fields in the target hospital;

[0085] Based on the target hospital's filed papers on research trends in various medical fields, use the first preset rule to determine whether it meets the patent application conditions;

[0086] When there is a registered paper that meets the patent application conditions, the second preset rule is used to obtain the patent classification number of the registered paper and the judgment result of whether the patent classification number meets the classification number requirements for rapid pre-examination or priority examination.

[0087] The present invention also provides a scientific research achievement data analysis system, comprising:

[0088] The data acquisition module is used to obtain the scientific research results text data of multiple other hospitals, the resource text data of the target hospital, and personal scientific research text data, wherein the multiple other hospitals include multiple hospitals that are ranked high in various medical fields;

[0089] The scientific research trend analysis module is used to obtain scientific research trend data in various medical fields based on the scientific research results text data of multiple target hospitals;

[0090] The first research direction recommendation module is used to match the target hospital's resource text data with the research trend data in various medical fields to obtain research direction recommendations for the target hospital;

[0091] The second research direction recommendation module is used to match the individual research text data of the target hospital with the research direction recommendation suggestions of the target hospital to obtain the research direction recommendation suggestions of individuals in the target hospital.

[0092] The present invention also provides an electronic device, comprising a processor and a memory storing a computer program, wherein the processor implements any of the above-mentioned scientific research results data analysis methods when executing the computer program.

[0093] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, it implements any of the above-mentioned scientific research results data analysis methods.

[0094] The present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute any of the above-mentioned scientific research results data analysis methods.

[0095] The present invention provides a scientific research achievement data analysis method, system, device, and medium that integrate multi-source heterogeneous data with intelligent analysis technology to achieve accurate recommendations for hospital research directions and optimize resource allocation, with at least the following beneficial effects:

[0096] (1) Intelligent analysis of scientific research trends: Based on the authoritative scientific research results text data of top-ranked hospitals, it automatically identifies the scientific research hotspots and development trends in various medical fields, and provides objective, data-driven discipline development references for target hospitals.

[0097] (2) Hospital-level scientific research direction recommendation: Dynamically match the existing resources of the target hospital (such as discipline foundation, equipment, and funding) with the scientific research trends in the field to generate scientific research direction recommendations that meet the actual conditions of the hospital and avoid resource mismatch.

[0098] (3) Adaptation of personal research direction: Combining the hospital's recommended direction with the researcher's personal research background (such as published papers and ongoing projects), the recommended direction is further refined to help researchers quickly identify high-potential topics.

[0099] The present invention provides a scientific research results data analysis method, system, equipment and medium, which integrate cross-institutional scientific research results, hospital resources and personal data, break the information island, and ensure that the suggestions can be implemented through multi-level matching (field trends → hospital resources → personal capabilities). Automated analysis replaces manual research, significantly shortens the scientific research planning cycle, and can solve the defects of existing scientific research results data analysis methods such as insufficient analysis depth, low intelligence level, and difficulty in meeting the complex needs of scientific research management. BRIEF DESCRIPTION OF THE DRAWINGS

[0100] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0101] Figure 1 This is a flow chart of a method for analyzing scientific research results data provided by the present invention.

[0102] Figure 2 This is a structural diagram of a scientific research achievement data analysis system provided by the present invention.

[0103] Figure 3 This is a schematic structural diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0104] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the drawings in the present invention. Obviously, the embodiments described are part of the embodiments of the present invention, not all of the embodiments, and they should not be understood as limitations on the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In the description of the present invention, it should be understood that the terms used are only for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0105] Figure 1 The flowchart of the scientific research achievement data analysis method provided by the present invention is shown in FIG. The execution subject of the scientific research achievement data analysis method provided by the present invention can be any applicable terminal side device or network side device, such as a scientific research achievement data analysis device.

[0106] See also Figure 1 The present invention provides a method for analyzing scientific research results data, which may include:

[0107] S110. Obtaining multiple text data on scientific research results from other hospitals, resource text data from the target hospital, and personal scientific research text data, wherein the multiple other hospitals include multiple hospitals that rank high in various medical fields.

[0108] In one embodiment, the plurality of hospitals ranked high in various medical fields may be the top ten hospitals ranked in various medical fields.

[0109] In one embodiment, the research results text data of other hospitals, the resource text data of the target hospital, and the personal research text data can all be obtained from the database of the corresponding hospital. The research results text data includes: data on papers, patents, scientific and technological awards, scientific research projects, and results transformation. The resource text data includes: data on advantageous fields, scientific research resources, scientific research personnel data, and patient data. The personal research text data includes: professional title, age, education, major, research direction, and personal research results text data (papers published, projects participated in, awards received, etc.).

[0110] S120. Obtain research trend data in various medical fields based on the scientific research results text data of multiple target hospitals.

[0111] In one embodiment, S120 may include:

[0112] Preprocessing the scientific research results text data of multiple target hospitals, wherein the preprocessing includes at least one of the following or any combination thereof: text cleaning, word segmentation, and stop word removal;

[0113] Obtain text features based on the pre-processed scientific research results text data of multiple target hospitals;

[0114] According to the text features, topic vocabulary is obtained through a topic model. In this embodiment, the topic model is an LDA (Latent Dirichlet Allocation) or NMF (Non-negative Matrix Factorization) topic model;

[0115] Based on the subject vocabulary, keywords are obtained through an unsupervised keyword extraction method. In this embodiment, the unsupervised keyword extraction method is TextRank or YAKE!;

[0116] Filter and cluster keywords to obtain research trend data in various medical fields;

[0117] Visualize scientific research trend data in various medical fields, for example, construct a pie chart showing the proportion of research directions in various medical fields, etc.

[0118] S130: Match the resource text data of the target hospital with the scientific research trend data of various medical fields to obtain a recommendation on the scientific research direction of the target hospital.

[0119] In one embodiment, S130 may include:

[0120] S1301. Calculate the matching degree of scientific research directions based on the target hospital's advantageous field data and the scientific research trend data of various medical fields. In this embodiment, S1301 calculates the cosine similarity between the target hospital's various scientific research directions and the scientific research trends of various medical fields based on the target hospital's scientific research resources and the scientific research trend data of various medical fields, and obtains the calculation result of the matching degree of scientific research directions of the target hospital corresponding to the scientific research trends of various medical fields.

[0121] S1302. Calculate resource matching based on the target hospital's scientific research resources and scientific research trend data in various medical fields.

[0122] In one embodiment, S1302 may include:

[0123] Using preset rules, it is determined whether the scientific research trends in various medical fields fall within the key discipline support scope of the target hospital, and the scientific research directions that fall within the key discipline support scope of the target hospital are screened out. The key discipline support scope of the target hospital can be manually set in advance. When the disciplines involved in the scientific research trends in the medical field exist within the key discipline support scope, it is determined to be a scientific research direction that falls within the key discipline support scope of the target hospital;

[0124] According to the scientific research resources of the target hospital, the funding support ranking of the scientific research directions within the target hospital's key discipline support scope is obtained. Specifically, the funding of the scientific research directions involved in the target hospital's key discipline support scope is sorted in descending order to obtain the funding support ranking of the scientific research directions involved in the target hospital's key discipline support scope, and then the funding support ranking of the scientific research directions within the target hospital's key discipline support scope is obtained;

[0125] Based on the number of scientific research directions within the target hospital's key discipline support scope and the ranking of funding support for these research directions within the target hospital's key discipline support scope, the first expression is used to calculate the resource matching degree of the target hospital's scientific research trends in various medical fields.

[0126] The first expression is:

[0127]

[0128] In the first expression, SS represents the resource matching degree of the target hospital corresponding to the scientific research trends in various medical fields, r represents the funding support ranking of the scientific research directions within the key discipline support scope of the target hospital, and N represents the number of scientific research directions within the key discipline support scope of the target hospital.

[0129] S1303. Calculate the scientific research trend matching degree based on the scientific research personnel data of the target hospital and the scientific research trend data of various medical fields.

[0130] In one embodiment, S1303 may include:

[0131] Based on the target hospital's scientific researcher data, we can obtain the target hospital's corresponding paper data in various medical fields;

[0132] According to the target hospital's corresponding paper data in various medical fields, the second expression is used to obtain the target hospital's corresponding trend growth index in various medical fields, where the second expression is:

[0133]

[0134] In the second expression, TGI i represents the trend growth index of the target hospital in the medical field i, C i,t represents the number of papers published in the target hospital in the corresponding medical field i in the previous year t, t = 1, 2, ..., T, t = T represents the latest year, w t represents the time weight, for example, the linear weight w t =t / T or index weight w t =2 t-T , if C i,t-f1 =0, then ignore the year;

[0135] The third expression is used to normalize the trend growth index of the target hospital in each medical field to obtain the scientific research trend matching calculation results of the target hospital in each medical field. The third expression is:

[0136]

[0137] In the third expression, TS i Indicates the matching degree of the target hospital’s research trend in medical field i, TGI i It represents the trend growth index of the target hospital corresponding to medical field i, max(TGI) represents the maximum value of the trend growth index of the target hospital corresponding to each medical field, and min(TGI) represents the minimum value of the trend growth index of the target hospital corresponding to each medical field.

[0138] In another embodiment, when the trend growth index is difficult to calculate, S1303 can be implemented in the following manner:

[0139] Based on the target hospital's scientific researcher data, we can obtain the target hospital's corresponding paper data in various medical fields;

[0140] According to the target hospital's corresponding paper data in various medical fields, the fourth expression is used to obtain the target hospital's corresponding scientific research trend matching results in various medical fields. The fourth expression is:

[0141]

[0142] In the fourth expression, TS′ i represents the matching degree of the target hospital’s research trend in medical field i, C i,t represents the number of papers published in medical field i in the previous t year, t = 1, 2, ..., T, t = T represents the latest year, max(∑C) represents the maximum value of the total number of papers published in each medical field by the target hospital within T years, and min(∑C) represents the minimum value of the total number of papers published in each medical field by the target hospital within T years.

[0143] S1304. Based on the research direction matching calculation results, resource matching calculation results, and research trend matching calculation results of the target hospital corresponding to the research trends in various medical fields, a comprehensive research matching degree of the target hospital corresponding to the research trends in various medical fields is obtained.

[0144] In one embodiment, S1304 may include:

[0145] Based on the calculation results of the scientific research direction matching degree, resource matching degree, and scientific research trend matching degree of the target hospital corresponding to the scientific research trends in various medical fields, the fifth expression is used to obtain the comprehensive scientific research matching degree of the target hospital corresponding to the scientific research trends in various medical fields. The fifth expression is:

[0146] Z = α·scientific research direction matching degree + β·resource matching degree + γ·scientific research trend matching degree,

[0147] In the fifth expression, α, β, and γ represent weight parameters.

[0148] S1305. Based on the comprehensive scientific research matching degree of the target hospital and the scientific research trends in various medical fields, obtain the target hospital's scientific research direction recommendation.

[0149] In one embodiment, S1305 may include:

[0150] According to the comprehensive scientific research matching degree of the target hospital's corresponding scientific research trends in various medical fields, the scientific research trends in various medical fields are sorted in descending order, and the top N scientific research trends are obtained as the scientific research recommendation directions for the target hospital; or

[0151] A preset threshold is set, and the comprehensive scientific research matching degree of the target hospital's scientific research trends in various medical fields is compared with the preset threshold, and the scientific research trend greater than the preset threshold is used as the target hospital's scientific research recommendation direction, wherein the preset threshold is set manually in advance or set by the permutation test method. The permutation test method is set specifically to randomly sample the data of researchers and each scientific research direction from the scientific research results text data, and then calculate the weighted matching degree according to the above formula, repeat this degree 1000 times, and obtain the distribution of matching degree under random conditions (i.e., low matching degree), and determine the judgment threshold with high matching degree by calculating the 95% confidence interval of the distribution.

[0152] In addition, the recommendation results can be dynamically adjusted based on the feedback and research results of researchers. The machine learning model is introduced to optimize the recommended model parameters (such as α, β, and γ weight parameters) based on the feedback of researchers. The specific strategy is to establish a loss function for the recommendation results based on the feedback of researchers (good or bad): for the direction with a matching degree ranking of k, if the feedback of researchers is good, the loss is If the feedback is bad, the loss is The total loss function is obtained by adding up the losses of all results. The machine learning model calculates the optimal parameters by minimizing the loss function.

[0153] In one embodiment, S130 may further include:

[0154] Obtain the number of papers published and the number of cases corresponding to the research trends in various medical fields of the target hospital;

[0155] Based on the number of papers published and the number of cases corresponding to the target hospital's scientific research trends in various medical fields, a time series analysis is performed using the moving average method or the exponential smoothing method, combined with derivatives and second-order derivatives, to determine whether the target hospital's scientific research trends in various medical fields are stable. Specifically, a time series analysis is performed on the number of papers published in each scientific research direction or the number of cases and samples involved in the research, and the annual growth rate and trend direction are calculated. The derivatives and second-order derivatives are used to determine whether there is a significant growth or decrease trend in certain research fields (the positive or negative value of the derivative is used to determine whether the direction is popular, and the positive or negative value of the second-order derivative is used to determine whether the field is growing rapidly). The moving average method or exponential smoothing method in time series analysis is used to determine whether the trend is stable.

[0156] In one embodiment, S130 may further include:

[0157] Obtain historical funding data on research trends in various medical fields for the target hospital;

[0158] Based on the historical funding data of the target hospital's scientific research trends in various medical fields, the funding forecasting model is used to predict the funding expenditure trends of the target hospital's scientific research trends in various medical fields.

[0159] In one embodiment, this can be achieved by:

[0160] ① Obtain funding data for various research projects from the hospital's research department for the past five years or a user-specified time period. The data format should include information such as project name, project type, amount, and year / quarter. Categorize the data by project type (e.g., clinical research, basic research, Beaver research, etc.). Ensure that the data is arranged by year or quarter to form time series data. Preprocess the data (missing value handling, outlier detection and handling, normalization, and transformation). Fill missing values using the mean, median, or linear interpolation. For short time periods with a high missingness rate, delete relevant data points. Identify outliers using the Z-score method (Z > ±3 is considered abnormal). After identifying outliers, choose to remove or correct them based on business needs. Convert funding amounts to logarithmic form to eliminate scale differences. ② Perform trend decomposition using the STL (Season and Trend decomposition using LOESS) method to decompose the time series data into three components: trend (long-term change direction); seasonality (cyclical fluctuation); and residual (random fluctuation). For scientific research projects of the same type, trend curves can be fitted separately to analyze their long-term change direction (increasing / decreasing). ③ Estimate trends and seasonality based on scientific research projects of the same type. Seasonal analysis: Analyze whether there are obvious cyclical fluctuations in the data (such as quarterly or annual seasonal patterns), and record the changing characteristics of seasonal components (amplitude, frequency, etc.). ④ Use a time series prediction model (to model trends and residuals. For data with obvious seasonality, consider adding seasonal terms for modeling. ⑤ Use historical data to train the model and evaluate its prediction performance through cross-validation (such as mean absolute error MAE, root mean square error RMSE, etc.). Based on historical trends and trends of scientific research projects of the same type, predict the funding expenditure trend of future scientific research projects. Taking into account the uncertainty of the trend, uncertainty information such as confidence intervals (such as ±1σ or ±2σ) can be added to the trend prediction results. Combined with the prediction results, analyze the funding expenditure trend of each scientific research project in the next few years, determine whether the trend is stable (increasing, decreasing or stable), and generate a report.

[0161] S140. Match the individual scientific research text data of the target hospital with the scientific research direction recommendation suggestions of the target hospital to obtain the scientific research direction recommendation suggestions of individuals in the target hospital.

[0162] In one embodiment, S140 may include:

[0163] Obtain project application information of the scientific research recommended direction of the corresponding target hospital, and encode and vectorize the project application information of the scientific research recommended direction of the corresponding target hospital, wherein the project application information includes any one of the following or any combination thereof: basic project information (project name, funding agency, project type, application deadline, etc.), project application conditions (professional title requirements, research direction, institutional requirements, etc.), project keywords (project research field, keywords, etc.);

[0164] Based on the personal scientific research text data of the target hospital, a user profile of the individuals in the target hospital is constructed. In this embodiment, the personal scientific research text data of the target hospital can be encoded and vectorized. Then, based on the encoded and vectorized personal scientific research text data, a multi-layer perceptron is used to obtain an embedded representation of the user profile of the individuals in the target hospital, which serves as the historical research vector of the individuals in the target hospital.

[0165] According to the project application information of the scientific research recommended direction of the target hospital and the user portrait of the individual in the target hospital, the matching degree between the individual in the target hospital and the scientific research projects involved in the scientific research recommended direction of each target hospital is calculated to obtain the scientific research direction recommendation suggestion for the individual in the target hospital. In this embodiment, according to the project application information of the scientific research recommended direction of the target hospital and the user portrait of the individual in the target hospital, the maximum similarity between the historical research vector of the individual in the target hospital and the scientific research projects involved in the scientific research recommended direction of each target hospital is calculated to obtain the scientific research direction recommendation suggestion for the individual in the target hospital, wherein the expression of the maximum similarity is:

[0166] Sim(u,p)=max(s1,s2,…,s i ),

[0167]

[0168] Where Sim(u,p) represents the historical research vector of individual u in the target hospital The vector v of the scientific research projects p related to the target hospital's scientific research recommendation direction (p) The maximum similarity of Represents the historical research vector of individual u in the target hospital The vector v of the scientific research projects p involved in the target hospital's recommended scientific research direction (p) The cosine similarity of .

[0169] In one embodiment, the scientific research achievement data analysis method provided by the present invention may further include:

[0170] Obtain the registered papers on the research trends of various medical fields in the target hospital;

[0171] Based on the target hospital's filed papers on research trends in various medical fields, use the first preset rule to determine whether it meets the patent application conditions;

[0172] When there is a registered paper that meets the patent application conditions, the second preset rule is used to obtain the patent classification number of the registered paper and the judgment result of whether the patent classification number meets the classification number requirements for rapid pre-examination or priority examination.

[0173] Specifically, this can be achieved through the following steps:

[0174] ① Based on the filed draft papers (text data) and historical patent data (including patent text, classification number, examination results, etc.), use a large language model for preprocessing to extract keywords and patent themes. ② Perform keyword matching and patent condition judgment (build a keyword library: extract patent-related keywords from historical patent data. Use domain knowledge or expert rules to supplement the keyword library. Matching judgment: calculate the similarity between the keywords of the draft paper and the patent keyword library (such as cosine similarity). Set a threshold to determine whether it meets the patent application conditions). ③ Recommend patent classification numbers (train a classification model: use historical patent data, take the patent text as input, and the classification number as a label to train a text classification model. Recommend classification numbers: input the draft paper into the trained model to recommend its possible patent classification number.) Perform rapid pre-examination or priority examination judgment (rule matching: determine whether the predicted classification number meets the conditions based on the published classification number requirements for rapid pre-examination or priority examination). ④ Predict the application success rate (train a regression model: use historical patent data, with the patent text, classification number, applicant information (including unit, title, etc.) as features, and the application success rate (such as whether it is authorized) as a label to train the regression model. Predict the success rate: input the draft of the paper and related information into the model to predict the application success rate). The final output of the model is as follows: Whether it meets the patent application conditions: Yes / No. Patent classification number: For example, G06N3 / 04. Whether it meets the requirements for accelerated pre-examination or priority examination: Yes / No. Application success rate: For example, 85%.

[0175] In one embodiment, the scientific research achievement data analysis method provided by the present invention may further include:

[0176] Match existing or registered scientific research results (such as patents...) with the business types of existing transformation companies, and provide transformation suggestions including the success rate of transformation of similar results, the matching degree between business and company, and the information required for transformation.

[0177] Specifically, this can be achieved through the following steps:

[0178] ① Researchers' research achievements obtained from the hospital's research department: patent title, patent number, patent type, technical field, abstract, claims, legal status, and researcher information: research direction, affiliation, title, contact information, etc. Data on transfer companies is obtained from the comprehensive service platform, including basic company information: company name, business type, industry, scale, contact information, etc. Transfer demand: The company's demand for technology transfer (e.g., technical field, patent type, collaboration model, etc.). Historical transfer data: The company's past transfer cases (e.g., transfer success rate, collaboration model, transfer cycle, etc.). External data includes market data: technology transfer market trends, popular fields, and policy support. Legal and policy data: intellectual property-related laws and regulations, technology transfer policies, etc. ② Research achievement characteristics: Technical field vectorization: Convert text data such as technical field and patent abstracts into vectors (similar to the previous method). Patent type encoding: Convert patent type (e.g., invention patent, utility model patent) into a numerical code. Legal status encoding: Convert legal status (e.g., authorized, under review) into a numerical code. Business type vectorization: Convert text data such as company business type and industry field into vectors. Company size coding: Convert company size (such as small, medium, large) into a numerical code. ③ Matching features, technical field matching: Calculate the similarity between scientific research results and the technical fields of the company's business type. (Operation method: For example: Calculate similarity based on patent classification numbers, count the number of patents owned by both parties under the same patent classification number, and calculate the technical similarity formula: Where P0 is the total number of patents held by both parties under the same classification, and PA and PB are the total number of patents for scientific research achievements and company business, respectively. A higher ratio indicates a higher degree of technical field match; cosine similarity can also be calculated based on similarity from text analysis. Patent Type Matching: Determines whether the company has demand for a specific patent type (extracts the technical features and application scenarios of scientific research achievements, extracts the technical requirements of the company's business, uses semantic analysis to extract keywords, uses the AO structure (Agent-Object) in natural language processing to match features with requirements, and calculates the degree of match between technical features and requirements). Historical Conversion Success Rate: Calculates the conversion success rate of similar achievements based on the company's historical conversion data. Matching Calculation: Uses cosine similarity to calculate the matching degree between scientific research achievements and the company's business type (same as the previous method). ④ Conversion Recommendation Generation, Conversion Success Rate of Similar Achievements: Calculates the conversion success rate of similar achievements based on the company's historical conversion data. Business-Company Matching: Outputs a matching score between scientific research achievements and the company's business type. Required Conversion Materials: Generates a list of required conversion materials (such as patent certificates, technical specifications, market analysis reports, etc.) based on the company's needs. The rule engine defines a rule base for handling specific scenarios (e.g., patents with a legal status of "under review" require additional explanation). The rule engine can dynamically generate conversion recommendations based on research results and company needs. ⑤ Model training: Data annotation uses historical conversion data as training data, annotating the matching results between research results and company information (e.g., successful conversion to positive examples, unconverted examples to negative examples).

[0179] The present invention provides a scientific research achievement data analysis method, system, device, and medium that integrate multi-source heterogeneous data with intelligent analysis technology to achieve accurate recommendations for hospital research directions and optimize resource allocation, with at least the following beneficial effects:

[0180] (1) Intelligent analysis of scientific research trends: Based on the authoritative scientific research results text data of top-ranked hospitals, it automatically identifies the scientific research hotspots and development trends in various medical fields, and provides objective, data-driven discipline development references for target hospitals.

[0181] (2) Hospital-level scientific research direction recommendation: Dynamically match the existing resources of the target hospital (such as discipline foundation, equipment, and funding) with the scientific research trends in the field to generate scientific research direction recommendations that meet the actual conditions of the hospital and avoid resource mismatch.

[0182] (3) Adaptation of personal research direction: Combining the hospital's recommended direction with the researcher's personal research background (such as published papers and ongoing projects), the recommended direction is further refined to help researchers quickly identify high-potential topics.

[0183] The present invention provides a scientific research results data analysis method, system, equipment and medium, which integrate cross-institutional scientific research results, hospital resources and personal data, break the information island, and ensure that the suggestions can be implemented through multi-level matching (field trends → hospital resources → personal capabilities). Automated analysis replaces manual research, significantly shortens the scientific research planning cycle, and can solve the defects of existing scientific research results data analysis methods such as insufficient analysis depth, low intelligence level, and difficulty in meeting the complex needs of scientific research management.

[0184] The scientific research achievement data analysis system provided by the present invention is described below. The scientific research achievement data analysis system described below and the scientific research achievement data analysis method described above can be referenced to each other.

[0185] See also Figure 2 The present invention provides a scientific research achievement data analysis system, which may include:

[0186] The data acquisition module is used to obtain the scientific research results text data of multiple other hospitals, the resource text data of the target hospital, and personal scientific research text data, wherein the multiple other hospitals include multiple hospitals that are ranked high in various medical fields;

[0187] The scientific research trend analysis module is used to obtain scientific research trend data in various medical fields based on the scientific research results text data of multiple target hospitals;

[0188] The first research direction recommendation module is used to match the target hospital's resource text data with the research trend data in various medical fields to obtain research direction recommendations for the target hospital;

[0189] The second research direction recommendation module is used to match the individual research text data of the target hospital with the research direction recommendation suggestions of the target hospital to obtain the research direction recommendation suggestions of individuals in the target hospital.

[0190] Figure 3 An example of a physical structure diagram of an electronic device is shown below. Figure 3 As shown, the electronic device may include: a processor 810, a communications interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communications interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to perform the following steps:

[0191] Obtain scientific research results text data from multiple other hospitals, resource text data from the target hospital, and personal scientific research text data. The multiple other hospitals include several hospitals that rank high in various medical fields.

[0192] Based on the scientific research results text data of multiple target hospitals, obtain the scientific research trend data in various medical fields;

[0193] Match the target hospital's resource text data with research trend data in various medical fields to obtain recommendations on the target hospital's research direction;

[0194] Match the individual scientific research text data of the target hospital with the scientific research direction recommendation suggestions of the target hospital to obtain the scientific research direction recommendation suggestions of individuals in the target hospital.

[0195] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0196] In another aspect, the present invention further provides a computer program product, comprising a computer program, which may be stored on a non-transitory computer-readable storage medium, and wherein when the computer program is executed by a processor, the computer is capable of performing the following steps:

[0197] Obtain scientific research results text data from multiple other hospitals, resource text data from the target hospital, and personal scientific research text data. The multiple other hospitals include several hospitals that rank high in various medical fields.

[0198] Based on the scientific research results text data of multiple target hospitals, obtain the scientific research trend data in various medical fields;

[0199] Match the target hospital's resource text data with research trend data in various medical fields to obtain recommendations on the target hospital's research direction;

[0200] Match the individual scientific research text data of the target hospital with the scientific research direction recommendation suggestions of the target hospital to obtain the scientific research direction recommendation suggestions of individuals in the target hospital.

[0201] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is configured to execute the following steps when executed by a processor:

[0202] Obtain scientific research results text data from multiple other hospitals, resource text data from the target hospital, and personal scientific research text data. The multiple other hospitals include several hospitals that rank high in various medical fields.

[0203] Based on the scientific research results text data of multiple target hospitals, obtain the scientific research trend data in various medical fields;

[0204] Match the target hospital's resource text data with research trend data in various medical fields to obtain recommendations on the target hospital's research direction;

[0205] Match the individual scientific research text data of the target hospital with the scientific research direction recommendation suggestions of the target hospital to obtain the scientific research direction recommendation suggestions of individuals in the target hospital.

[0206] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0207] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0208] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for analyzing scientific research results data, characterized in that: include: Obtain scientific research results text data from multiple other hospitals, resource text data from the target hospital, and personal scientific research text data; Based on the scientific research results text data of multiple target hospitals, obtain the scientific research trend data in various medical fields; Match the target hospital's resource text data with research trend data in various medical fields to obtain recommendations on the target hospital's research direction; Match the individual scientific research text data of the target hospital with the scientific research direction recommendation suggestions of the target hospital to obtain the scientific research direction recommendation suggestions of individuals in the target hospital.

2. The scientific research achievement data analysis method according to claim 1, characterized in that: Scientific research results text data includes at least one of the following text data or any combination thereof: papers, patents, scientific and technological awards, scientific research projects, and achievement transformation; resource text data includes at least one of the following text data or any combination thereof: advantageous field data, scientific research resources, scientific researcher data, and patient data; personal scientific research text data includes any one of the following text data or any combination thereof: professional title, age, education background, major, research direction, and personal scientific research results text data; and / or, Based on the scientific research results text data of multiple target hospitals, scientific research trend data in various medical fields are obtained, including: Preprocess the scientific research results text data of multiple target hospitals; Obtain text features based on the pre-processed scientific research results text data of multiple target hospitals; According to the text features, the topic vocabulary is obtained through the topic model; Based on the subject vocabulary, keywords are obtained through unsupervised keyword extraction; Filter and cluster keywords to obtain research trend data in various medical fields; Visualize scientific research trend data in various medical fields.

3. The scientific research achievement data analysis method according to claim 2, characterized in that: The target hospital's resource text data is matched with the research trend data in various medical fields to obtain the target hospital's research direction recommendations, including: Calculate the matching degree of scientific research directions based on the target hospital's advantageous field data and scientific research trend data in various medical fields; Calculate resource matching based on the target hospital's scientific research resource data and scientific research trend data in various medical fields; Calculate the matching degree of scientific research trends based on the target hospital's scientific research personnel data and the scientific research trend data in various medical fields; Based on the research direction matching calculation results, resource matching calculation results, and research trend matching calculation results of the target hospital corresponding to the research trends in various medical fields, the comprehensive research matching degree of the target hospital corresponding to the research trends in various medical fields is obtained; Based on the comprehensive scientific research matching degree of the target hospital and the scientific research trends in various medical fields, we can obtain the recommended scientific research direction of the target hospital.

4. The scientific research achievement data analysis method according to claim 3, characterized in that: The research direction matching calculation is performed based on the target hospital's advantageous field data and the research trend data in various medical fields, including: Based on the scientific research resources of the target hospital and the scientific research trend data of various medical fields, the cosine similarity between each scientific research direction of the target hospital and the scientific research trends of various medical fields is calculated to obtain the scientific research direction matching calculation results of the target hospital corresponding to the scientific research trends of various medical fields; and / or, The resource matching calculation is performed based on the target hospital's scientific research resources and the scientific research trend data in various medical fields, including: Using preset rules, we determine whether the research trends in various medical fields fall within the key disciplines supported by the target hospital, and select research directions that fall within the key disciplines supported by the target hospital; Based on the scientific research resources of the target hospital, the funding support ranking of the scientific research directions within the key discipline support scope of the target hospital is obtained; Based on the number of scientific research directions within the scope of key disciplines supported by the target hospital and the ranking of funding support for scientific research directions within the scope of key disciplines supported by the target hospital, the resource matching calculation results of the target hospital's scientific research trends in various medical fields are obtained using the first expression, where the first expression is: In the first expression, SS represents the resource matching degree of the target hospital corresponding to the scientific research trends in various medical fields, r represents the funding support ranking of the scientific research directions within the key discipline support scope of the target hospital, and N represents the number of scientific research directions within the key discipline support scope of the target hospital.

5. The method for analyzing scientific research results data according to claim 4, characterized in that: The research trend matching calculation is performed based on the target hospital's scientific research personnel data and the research trend data in various medical fields, including: Based on the target hospital's scientific researcher data, we can obtain the target hospital's corresponding paper data in various medical fields; According to the target hospital's corresponding paper data in various medical fields, the second expression is used to obtain the target hospital's corresponding trend growth index in various medical fields, where the second expression is: In the second expression, TGI i represents the trend growth index of the target hospital in the medical field i, C i,t represents the number of papers published in the target hospital in the corresponding medical field i in the previous year t, t = 1, 2, ..., T, t = T represents the latest year, w t represents the time weight, if C i,t-1 =0, then ignore the year; The third expression is used to normalize the trend growth index of the target hospital in each medical field to obtain the scientific research trend matching calculation results of the target hospital in each medical field. The third expression is: In the third expression, TS i Indicates the matching degree of the target hospital’s research trend in medical field i, TGI i represents the trend growth index of the target hospital in medical field i, max(TGI) represents the maximum value of the trend growth index of the target hospital in each medical field, and min(TGI) represents the minimum value of the trend growth index of the target hospital in each medical field; or The research trend matching calculation is performed based on the target hospital's scientific research personnel data and the research trend data in various medical fields, including: Based on the target hospital's scientific researcher data, we can obtain the target hospital's corresponding paper data in various medical fields; According to the target hospital's corresponding paper data in various medical fields, the fourth expression is used to obtain the target hospital's corresponding scientific research trend matching results in various medical fields. The fourth expression is: In the fourth expression, TS′ i represents the matching degree of the target hospital’s research trend in medical field i, C i,t represents the number of papers published in medical field i in the previous t year, t = 1, 2, ..., T, t = T represents the latest year, max(∑C) represents the maximum value of the total number of papers published in each medical field by the target hospital within T years, and min(∑C) represents the minimum value of the total number of papers published in each medical field by the target hospital within T years.

6. The method for analyzing scientific research results data according to claim 5, characterized in that: The above-mentioned calculation results of the scientific research direction matching degree, resource matching degree, and scientific research trend matching degree of the target hospital corresponding to the scientific research trends in various medical fields are used to obtain the comprehensive scientific research matching degree of the target hospital corresponding to the scientific research trends in various medical fields, including: Based on the calculation results of the scientific research direction matching degree, resource matching degree, and scientific research trend matching degree of the target hospital corresponding to the scientific research trends in various medical fields, the fifth expression is used to obtain the comprehensive scientific research matching degree of the target hospital corresponding to the scientific research trends in various medical fields. The fifth expression is: Z = α·scientific research direction matching degree + β·resource matching degree + γ·scientific research trend matching degree, In the fifth expression, α, β, and γ represent weight parameters; Based on the comprehensive scientific research matching degree of the target hospital's corresponding scientific research trends in various medical fields, the target hospital's scientific research direction recommendations are obtained, including: According to the comprehensive scientific research matching degree of the target hospital's corresponding scientific research trends in various medical fields, the scientific research trends in various medical fields are sorted in descending order, and the top N scientific research trends are obtained as the scientific research recommendation directions for the target hospital; or A preset threshold is set, and the comprehensive scientific research matching degree of the target hospital's scientific research trends in various medical fields is compared with the preset threshold. The scientific research trend greater than the preset threshold is used as the recommended scientific research direction of the target hospital, wherein the preset threshold is set manually in advance or through the permutation test method.

7. The method for analyzing scientific research results data according to any one of claims 1 to 6, characterized in that: The individual scientific research text data of the target hospital is matched with the scientific research direction recommendation suggestions of the target hospital to obtain the scientific research direction recommendation suggestions of individuals in the target hospital, including: Obtain project application information for the target hospital's recommended research directions; Build user profiles of individuals in the target hospital based on the individual scientific research text data of the target hospital; Based on the project application information of the target hospital's recommended research direction and the user profiles of individuals in the target hospital, the matching degree between individuals in the target hospital and the research projects involved in the recommended research directions of each target hospital is calculated to obtain the recommended research direction suggestions for individuals in the target hospital; The acquisition of project application information corresponding to the target hospital's recommended research direction includes: Encode and vectorize the project application information of the target hospital's recommended scientific research direction; The user profiles of individuals in the target hospital are constructed based on the individual scientific research text data of the target hospital, including: Encode and vectorize the individual scientific research text data of the target hospital; Based on the encoded and vectorized personal scientific research text data, a multi-layer perceptron is used to obtain the embedded representation of the user portrait of individuals in the target hospital, which serves as the historical research vector of individuals in the target hospital. According to the project application information of the target hospital's recommended scientific research direction and the user profile of the individual in the target hospital, the matching degree between the individual in the target hospital and the scientific research projects involved in the scientific research recommended direction of each target hospital is calculated, and the scientific research direction recommendation suggestions for the individual in the target hospital are obtained, including: Based on the project application information of the corresponding target hospital's recommended scientific research direction and the user portraits of individuals in the target hospital, the maximum similarity between the historical research vectors of individuals in the target hospital and the scientific research projects involved in the scientific research recommended directions of each target hospital is calculated to obtain recommended scientific research directions for individuals in the target hospital.

8. A scientific research results data analysis system, characterized in that: include: The data acquisition module is used to obtain the scientific research results text data of multiple other hospitals, the resource text data of the target hospital, and personal scientific research text data, wherein the multiple other hospitals include multiple hospitals that are ranked high in various medical fields; The scientific research trend analysis module is used to obtain scientific research trend data in various medical fields based on the scientific research results text data of multiple target hospitals; The first research direction recommendation module is used to match the target hospital's resource text data with the research trend data in various medical fields to obtain research direction recommendations for the target hospital; The second research direction recommendation module is used to match the individual research text data of the target hospital with the research direction recommendation suggestions of the target hospital to obtain the research direction recommendation suggestions of individuals in the target hospital.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the scientific research results data analysis method as described in any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the scientific research achievement data analysis method as described in any one of claims 1 to 7.

Images