A method, apparatus, storage medium, and electronic device for recommending scientific research resources.
By constructing a logical rule matrix to analyze the scientific research resource recommendation system, the problem of irrelevant recommendation results in existing technologies is solved, personalized scientific research resource recommendations are realized, and the accuracy and practicality of the recommendations are improved.
Patent Information
- Application Number
- CN202611328748.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-31
- Publication Date
- 2026-11-17
AI Technical Summary
Existing methods for recommending scientific research resources rely on keyword matching, resulting in poor relevance and usability of the recommendation results.
By acquiring internal and external academic data of the target audience, a logical rule matrix is constructed for assessment. The assessment shortcomings and the shortcomings of research directions deviating from academic hotspots are analyzed, and personalized resource recommendations are made.
It achieves relevance and practicality in the recommendation of scientific research resources, avoids blind recommendations, and improves the accuracy of recommendation results.
Smart Images

Figure CN123116079A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of information technology, and in particular to a method, apparatus, storage medium and electronic device for recommending scientific research resources. Background Technology
[0002] With the continuous development of big data technology, university research management is gradually transforming from a traditional information-based model to an intelligent one. Most existing research management systems only provide basic data collection and report visualization functions, lacking intelligent recommendation capabilities for research projects, topics, and other research resources.
[0003] Currently, research resource recommendations typically rely on simple keyword matching. However, this method relies solely on user-input keywords for retrieval and recommendations, making the entire process somewhat arbitrary and resulting in poor relevance and usability of the recommendations. Summary of the Invention
[0004] In view of this, this application provides a method, apparatus, storage medium and electronic device for recommending scientific research resources, which mainly improves the relevance and practicality of the scientific research resource recommendation results.
[0005] According to a first aspect of this application, a method for recommending scientific research resources is provided, the method comprising: Acquire internal and external academic data related to the target object; The internal academic data is input into the logical rule matrix corresponding to the assessment system document to obtain the assessment results for the target object. Based on the assessment results, analyze the assessment shortcomings of the target object; Based on the internal academic data and the external academic data, the academic research feature vector of the target object and the hot spot feature vector corresponding to the external academic hot spots are determined respectively. Based on the hotspot feature vector and the academic research feature vector, determine whether the target object has a shortcoming factor that deviates from the academic hotspot; If such factors exist, then resource recommendations will be made for the target object based on the shortcomings of the research direction deviating from academic hotspots and the shortcomings of the assessment.
[0006] According to a second aspect of this application, a scientific research resource recommendation device is provided, the device comprising: The acquisition unit is used to acquire internal and external academic data related to the target object. The assessment unit is used to input the internal academic data into the logical rule matrix corresponding to the assessment system document to obtain the assessment result corresponding to the target object. The analysis unit is used to analyze the assessment shortcomings of the target object based on the assessment results. The determining unit is used to determine the academic research feature vector of the target object and the hot spot feature vector corresponding to the external academic hot spot based on the internal academic data and the external academic data, respectively. The determination unit is used to determine, based on the hotspot feature vector and the academic research feature vector, whether the target object has a shortcoming factor that deviates from the academic hotspot in its research direction; The recommendation unit is used to recommend resources to the target object based on the shortcomings factors of the research direction deviating from academic hotspots and the assessment shortcomings factors, if they exist.
[0007] According to a third aspect of this application, a storage medium is provided that stores a computer program thereon, which, when executed by a processor, implements the above-described method for recommending scientific research resources.
[0008] According to a fourth aspect of this application, an electronic device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the program to implement the above-described method for recommending scientific research resources.
[0009] By employing the above technical solution, this application provides a research resource recommendation method, apparatus, storage medium, and electronic device. Compared with existing technologies, this method inputs internal academic data into the logical rule matrix corresponding to the assessment system document to determine the assessment results of the target object. By analyzing the assessment results, it identifies the target object's assessment weaknesses. Simultaneously, based on internal and external academic data, it determines the target object's academic research feature vector and academic hotspot feature vector, respectively. Based on these vectors, it analyzes whether the target object has a weakness factor where its research direction deviates from academic hotspots. If so, it recommends resources to the target object based on this weakness factor and the assessment weakness factor. Therefore, this application can achieve personalized research resource recommendations based on the target object's assessment weakness factor and the weakness factor where its research direction deviates from academic hotspots, ensuring the relevance and practicality of the recommendation results and avoiding blind recommendations.
[0010] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description
[0011] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A flowchart illustrating a scientific research resource recommendation method provided in an embodiment of this application is shown. Figure 2 This illustration shows a structural schematic diagram of a scientific research resource recommendation device provided in an embodiment of this application; Figure 3 A schematic diagram of another scientific research resource recommendation device provided in an embodiment of this application is shown. Detailed Implementation
[0012] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.
[0013] Existing methods for recommending scientific research resources rely solely on keywords entered by users, resulting in a somewhat arbitrary recommendation process and poor relevance and usability of the results.
[0014] To address the above problems, embodiments of the present invention provide a method for recommending scientific research resources, such as... Figure 1 As shown, the method includes: Step 10: Obtain internal and external academic data related to the target object.
[0015] The target audience can specifically include teachers, colleges, and schools. Internal academic data refers to academic data within the institution related to the target audience, specifically including project data, funding execution information, achievement data, and basic information. External academic data refers to academic data from outside the institution related to the target audience, specifically including literature abstracts and application guide texts. It should be noted that the internal and external academic data in this embodiment of the invention are not limited to the above-listed examples and may also include other data.
[0016] The embodiments of this invention are primarily applicable to scientific research resource recommendation scenarios. The executing entity of these embodiments is a device or equipment capable of recommending scientific research resources, specifically located on a server side.
[0017] Regarding the specific processes for collecting internal and external academic data, such as Figure 2 As shown, it specifically includes: Step 11: Obtain project data, funding execution information, achievement data, and basic information related to the target object from the internal system, and identify the project data, funding execution information, achievement data, and basic information as the internal academic data.
[0018] The project data includes project level and source, with project levels specifically including national, provincial / ministerial, and municipal levels, and project sources including natural sciences and humanities / social sciences. Funding execution information includes total appropriation amount, total expenditure amount, annual appropriation amount, annual support amount, and funding execution rate. Outcome information includes published papers and journals, and the types of papers and journals. When the target audience is a teacher, basic information includes the teacher's professional title, graduating institution, previous educational experience, and work experience; when the target audience is a college, basic information includes academic disciplines and faculty information; when the target audience is a university, basic information includes the type of university (e.g., comprehensive, medical, science and engineering), and academic disciplines. It should be noted that the project data, funding execution information, outcome data, and basic information in this embodiment of the invention are not limited to the above-listed examples and may include other information.
[0019] In this embodiment of the invention, project data, funding execution information, achievement data and basic information related to the target object can be extracted in real time from the university's scientific research management system through API interface or ETL tool, thereby obtaining the target object's internal academic data.
[0020] Step 12: Based on the attribute information of the target object, obtain the literature abstract and application guide text related to the target object from the external system, and identify the literature abstract and the application guide text as the external learning data.
[0021] Specifically, the attribute information can be the subject information of the target object or other required information.
[0022] In this embodiment of the invention, based on the subject information or other requirements of the target object, relevant literature abstracts and application guide texts are obtained from external academic websites (such as Web of Science, arXiv, etc.), thereby acquiring external academic data of the target object. It should be noted that the external academic data in this embodiment of the invention is not limited to literature abstracts and application guide texts, and may also include other text data.
[0023] After acquiring the internal and external academic data of the target object, we can perform subject modeling on the internal and external academic data of the target object, that is, build a text database of the target object, map and fuse the internal and external academic data of the target object, and thus generate a 360-degree dynamic digital portrait of the target object.
[0024] By fusing internal and external academic data, this invention can accurately construct a profile of the target object, thereby improving the accuracy of subsequent analysis.
[0025] Step 20: Input the internal academic data into the logical rule matrix corresponding to the assessment system document to obtain the assessment results corresponding to the target object.
[0026] When the target is a teacher or college, the assessment system document can be an assessment document issued by the school to the teacher or college; when the target is the school, the assessment system document can be an assessment document issued by the state to the school.
[0027] In this embodiment of the invention, when assessing a target object using an assessment system document, the assessment system document needs to be converted into a corresponding logical rule matrix. The method for this process includes: inputting the assessment system document and a preset prompt word template into a large logical rule matrix generation model to obtain the logical rule matrix.
[0028] The preset prompt word template records the input and output information of the logical rule matrix, as well as the programming language used by the code module corresponding to the logical rule matrix. The input information of the logical rule matrix specifically consists of key indicators and their corresponding indicator values, such as a project level of provincial or ministerial level. The output information of the logical rule matrix specifically consists of the specific score for that key indicator, such as a score of 10 points for a national-level project, 8 points for a provincial or ministerial-level project, and 5 points for a municipal-level project. The programming languages include Python, C++, Java, etc.
[0029] When generating the logical rule matrix, the logical rule matrix generation model reads the key indicators and their corresponding score divisions from the assessment system document, and generates the logical rule matrix based on the key indicators, score divisions, and preset prompt word templates. Specifically, the logical rule matrix generation model can be an LLM (Local Level Model).
[0030] After converting the assessment system document into a logical rule matrix, the internal academic data is input into the logical rule matrix corresponding to the assessment system document to obtain the scores corresponding to each key indicator; based on the scores corresponding to each key indicator, the total assessment score of the target object is determined; based on the total assessment score, the assessment result corresponding to the target object is determined. Step 30: Based on the assessment results, analyze the assessment shortcomings of the target object.
[0031] In this embodiment of the invention, if the target object's total assessment score is lower than the preset assessment score, then based on the scores corresponding to each key indicator, a target key indicator whose score is lower than the preset score is determined from among the key indicators; and based on the target key indicator, the assessment weakness factor corresponding to the target object is determined. The preset assessment score and preset score can be set according to actual business needs, and this embodiment of the invention does not impose specific limitations on them.
[0032] For example, based on the assessment results, the analysis of the target group's assessment shortcomings revealed that the rate of receipt of horizontal funding was low and the output of interdisciplinary papers was low.
[0033] It should be noted that, in the embodiments of the present invention, assessment weakness factor analysis can be initiated not only when the total assessment score is lower than the preset assessment score, but also when the assessment score meets the standard, that is, when the assessment score is greater than or equal to the preset assessment score, so as to make resource recommendations for the target object's weaknesses in the future.
[0034] Step 40: Based on the internal academic data and the external academic data, determine the academic research feature vector of the target object and the hotspot feature vector corresponding to the external academic hotspots, respectively.
[0035] In this embodiment of the invention, an academic research feature vector of the target object is determined based on project data and achievement data from the target object's internal academic data. Specifically, this process includes: extracting project keywords and project summaries, as well as achievement keywords and achievement summaries, from the project data and achievement data in the internal academic data; concatenating the project keywords and project summaries and then performing text encoding to obtain the project feature vector of the target object; concatenating the achievement keywords and achievement summaries and then performing text encoding to obtain the achievement feature vector of the target object; and weightedly fusing the project feature vector and the achievement feature vector to obtain the academic research feature vector.
[0036] Specifically, for each project, project keywords and project abstracts are obtained and concatenated. The concatenated keywords and abstracts are then input into a text encoder for encoding, resulting in a project vector. If the target object has multiple projects, the project vectors of these projects are weighted or averaged to obtain an overall project vector, i.e., the project feature vector. Similarly, for each achievement, achievement keywords and achievement abstracts are obtained and concatenated. The concatenated keywords and abstracts are then input into a text encoder for encoding, resulting in an achievement vector. If the target object has multiple achievements, the achievement vectors of these achievements are weighted to obtain an overall achievement vector, i.e., the achievement feature vector. During weighting, high-level journals or invention patents have a higher weight than ordinary papers or software copyrights, and recent achievements have a higher weight. Finally, the project feature vector and achievement feature vector are weighted and fused to obtain the academic research feature vector of the target object. The weights corresponding to the project feature vector and achievement feature vector can be adjusted according to actual business needs.
[0037] Simultaneously, based on external academic data of the target object, an academic hotspot feature vector is determined. This process specifically includes: obtaining literature abstracts and application guide texts within a preset time window from the external academic data; text encoding the literature abstracts and application guide texts within the preset time window to obtain a literature hotspot feature vector and a guideline-oriented feature vector; and weighted fusion of the literature hotspot feature vector and the guideline-oriented feature vector to obtain the hotspot feature vector corresponding to the external academic hotspot.
[0038] The preset time window can be set according to actual business needs, such as obtaining literature abstracts and application guide texts from the past 3 years.
[0039] Specifically, the main text, key independent research directions, and proposed scientific problems to be solved are extracted from the application guidelines. These sections are then concatenated and input into a text encoder to obtain a guideline-oriented vector. If multiple application guidelines exist, their guideline-oriented vectors are weighted to obtain a guideline-oriented feature vector, with national-level guidelines having a higher weight than provincial / ministerial-level guidelines. Simultaneously, each literature abstract is input into the text encoder to obtain a single-article abstract vector. All abstract vectors within a time window are then time-weighted and aggregated to obtain a literature hotspot feature vector, where newer abstracts have higher weights, and highly cited or high-impact abstracts can be assigned even higher weights. Finally, the guideline-oriented feature vector and the literature hotspot feature vector are weighted and fused to obtain an external academic hotspot feature vector. The weights of the guideline-oriented feature vector and the literature hotspot feature vector can be adjusted according to actual business needs.
[0040] Step 50: Based on the hotspot feature vector and the academic research feature vector, determine whether the target object has a shortcoming factor that causes its research direction to deviate from the academic hotspot.
[0041] In this embodiment of the invention, after determining the hot topic feature vector and the academic research feature vector, the cosine similarity between the hot topic feature vector and the academic research feature vector is calculated. Based on the cosine similarity, it is determined whether the target object has a weakness factor that causes its research direction to deviate from academic hot topics. Specifically, if the cosine similarity is greater than a preset similarity threshold, it is determined that the target object does not have a weakness factor that causes its research direction to deviate from academic hot topics; if the cosine similarity is less than or equal to the preset similarity threshold, it is determined that the target object has a weakness factor that causes its research direction to deviate from academic hot topics.
[0042] Step 60: If present, then based on the shortcomings factors of the research direction deviating from academic hotspots and the assessment shortcomings factors, resource recommendations are made for the target object.
[0043] In this embodiment of the invention, based on the shortcomings of research direction deviating from academic hotspots and assessment shortcomings, internal and external resources are queried to make targeted recommendations for the target objects. Internal resources include on-campus cultivation funds, key laboratories, etc., while external resources include off-campus paper journals, off-campus laboratory open projects, off-campus horizontal projects, etc.
[0044] When recommending resources, a resource feature vector is determined for each resource, including internal and external resources. Based on the shortcoming factor that the research direction deviates from academic hotspots, the resource feature vectors, hotspot feature vectors, and academic research feature vectors are used to determine the degree to which each resource is close to the external academic hotspots and the degree of fit between each resource and the target object. The friendliness of each resource to the assessment shortcoming factor is determined. Based on the degree to which each resource is close to the external academic hotspots, the degree of fit between each resource and the target object, and the friendliness, a recommendation score is determined for each resource. Based on the recommendation scores of each resource, resources are recommended to the target object.
[0045] Specifically, firstly, all internal and external resources are vectorized. For example, the funding guidelines and key support directions in the application documents for internal cultivation funds or projects are extracted and encoded into corresponding resource feature vectors. Similarly, the research directions and publicly available results of key laboratory projects are extracted and encoded into corresponding resource feature vectors. Furthermore, for external journal articles, abstracts are extracted and encoded into resource feature vectors. Then, based on the resource feature vectors, hot topic feature vectors, and academic research feature vectors corresponding to each resource, the first cosine similarity between the resource feature vector and the hot topic feature vector is calculated. Based on this first cosine similarity, the degree to which each resource aligns with external academic hot topics is determined; higher similarity indicates greater alignment and a higher score. Simultaneously, based on the second cosine similarity between the resource feature vectors and academic research feature vectors, the degree to which each resource fits the target object is determined; higher similarity indicates greater fit and a higher score.
[0046] Next, if the resource is an internal cultivation fund, key laboratory, open project from an external laboratory, or external collaborative project, it is necessary to determine its suitability for assessment weaknesses. For example, if a teacher has a low number of successfully applied projects, then for this weakness, it is necessary to evaluate which of the following resources—internal cultivation funds, key laboratories, open projects from external laboratories, and external collaborative projects—can enable the teacher to quickly produce results. For instance, the scores could be ranked from highest to lowest as: internal cultivation funds, open projects from external laboratories, internal laboratory projects, general external collaborative projects, and other internal projects. If the resource is an external journal article, then it is not necessary to determine its suitability for assessment weaknesses.
[0047] Furthermore, based on the resource's relevance to external academic hot topics, its relevance to the target audience, and its user-friendliness, a recommendation score is determined for each resource. Higher relevance, relevance, and user-friendliness result in a higher score, and vice versa. Finally, based on the recommendation scores, the resources are ranked, and appropriate internal and external resources are recommended to the target audience according to the ranking results.
[0048] It should be noted that target resources with recommendation scores greater than a certain value can also be selected, and then a portion of the text of the target resource can be extracted and input into a large model for sorting, with the large model performing the sorting and recommendation.
[0049] This invention provides a scientific research resource recommendation method that analyzes internal and external academic data related to a target object to determine the target object's assessment weaknesses and whether the target object has weaknesses in its research direction that deviates from academic hotspots. This method enables personalized recommendations of scientific research resources based on the target object's assessment weaknesses and weaknesses in its research direction that deviate from academic hotspots, thereby ensuring the relevance and practicality of the recommendation results and avoiding blind recommendations.
[0050] Furthermore, as Figure 1 and Figure 2 The specific implementation of the method shown in this embodiment provides a scientific research resource recommendation device, such as... Figure 3 As shown, the device includes: an acquisition unit 101, an assessment unit 102, an analysis unit 103, a determination unit 104, a judgment unit 105, and a recommendation unit 106.
[0051] The acquisition unit 101 can be used to acquire internal academic data and external academic data related to the target object.
[0052] The assessment unit 102 can be used to input the internal academic data into the logical rule matrix corresponding to the assessment system document to obtain the assessment result corresponding to the target object.
[0053] The analysis unit 103 can be used to analyze the assessment shortcomings of the target object based on the assessment results.
[0054] The determining unit 104 can be used to determine the academic research feature vector of the target object and the hot spot feature vector corresponding to the external academic hot spot based on the internal academic data and the external academic data, respectively.
[0055] The determination unit 105 can be used to determine, based on the hotspot feature vector and the academic research feature vector, whether the target object has a shortcoming factor that deviates from the academic hotspot in its research direction.
[0056] The recommendation unit 106 can be used to recommend resources to the target object based on the shortcomings factors of the research direction deviating from academic hotspots and the assessment shortcomings factors, if they exist.
[0057] In some embodiments, the acquisition unit 101 may be specifically used to acquire project data, funding execution information, achievement data, and basic information related to the target object from an internal system; determine the project data, funding execution information, achievement data, and basic information as the internal academic data; acquire literature abstracts and application guide texts related to the target object from an external system based on the attribute information of the target object; and determine the literature abstracts and application guide texts as the external learning data.
[0058] In some embodiments, the apparatus further includes a conversion unit.
[0059] The conversion unit can be used to input the assessment system document and the preset prompt word template into the logic rule matrix generation model to obtain the logic rule matrix. The preset prompt word template records the input and output information of the logic rule matrix, as well as the programming language used by the code module corresponding to the logic rule matrix. When generating the logic rule matrix, the logic rule matrix generation model reads each key indicator and its corresponding score division in the assessment system document, and generates the logic rule matrix based on the key indicators, the score division and the preset prompt word template.
[0060] In some embodiments, the assessment unit 102 may be specifically used to input the internal academic data into the logical rule matrix corresponding to the assessment system document to obtain the scores corresponding to each key indicator; determine the total assessment score of the target object based on the scores corresponding to each key indicator; and determine the assessment result corresponding to the target object based on the total assessment score. In some embodiments, the analysis unit 103 may be specifically used to determine, based on the scores corresponding to each key indicator, the target key indicator whose score is lower than the preset score if the total assessment score is lower than the preset assessment score; and determine the assessment weakness factor corresponding to the target object based on the target key indicator.
[0061] In some embodiments, the determining unit 104 may be specifically configured to: extract project keywords and project abstracts, as well as achievement keywords and achievement abstracts, based on project data and achievement data in the internal academic data; concatenate the project keywords and project abstracts and perform text encoding to obtain the project feature vector of the target object; concatenate the achievement keywords and achievement abstracts and perform text encoding to obtain the achievement feature vector of the target object; perform weighted fusion of the project feature vector and the achievement feature vector to obtain the academic research feature vector; obtain literature abstracts and application guide texts within a preset time window from the external academic data; perform text encoding on the literature abstracts and application guide texts within the preset time window to obtain literature hotspot feature vectors and guideline-oriented feature vectors; and perform weighted fusion of the literature hotspot feature vectors and the guideline-oriented feature vectors to obtain the hotspot feature vector corresponding to the external academic hotspot.
[0062] In some embodiments, the determination unit 105 may be specifically used to calculate the cosine similarity between the hotspot feature vector and the academic research feature vector; based on the cosine similarity, determine whether the target object has a shortcoming factor that deviates from the academic hotspot in its research direction.
[0063] In some embodiments, the recommendation unit 106 may be specifically used to determine the resource feature vector corresponding to each resource, wherein each resource includes internal resources and external resources; based on the shortcoming factor that the research direction deviates from the academic hotspot, using the resource feature vector corresponding to each resource, the hotspot feature vector, and the academic research feature vector, to determine the degree to which each resource is close to the external academic hotspot, and the degree of fit between each resource and the target object; to determine the friendliness of each resource to the assessment shortcoming factor; based on the degree to which each resource is close to the external academic hotspot, the degree of fit between each resource and the target object, and the friendliness, to determine the recommendation score of each resource; and to recommend resources to the target object according to the recommendation score of each resource.
[0064] It should be noted that other corresponding descriptions of the functional units involved in the scientific research resource recommendation device provided in this embodiment can be found in [reference needed]. Figure 1 and Figure 2 The corresponding description in [the document] will not be repeated here.
[0065] Based on the above, Figure 1 and Figure 2 Accordingly, this embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the above-described method. Figure 1 and Figure 2 The method for recommending scientific research resources is shown.
[0066] Based on this understanding, the technical solution of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as CD-ROM, USB flash drive, mobile hard drive, etc.) and includes several instructions to cause an electronic device (such as personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of this application.
[0067] Based on the above, Figure 1 and Figure 2 The method shown, and Figure 3 To achieve the above objectives, the present application also provides an electronic device, specifically a personal computer, tablet computer, server, or other network device, as shown in the virtual device embodiment. This device includes a storage medium and a processor; the storage medium stores a computer program; the processor executes the computer program to achieve the above-described objectives. Figure 1 and Figure 2 The method for recommending scientific research resources is shown.
[0068] Optionally, the aforementioned physical devices may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Wi-Fi interfaces), etc.
[0069] Those skilled in the art will understand that the physical device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine certain components, or have different component arrangements.
[0070] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the aforementioned physical device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software in the information processing physical device.
[0071] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platform, or it can be implemented by hardware.
[0072] This invention analyzes internal and external academic data related to the target object to determine the target object's assessment weaknesses and whether the target object has weaknesses in its research direction that deviates from academic hotspots. It can make personalized recommendations of scientific research resources based on the target object's assessment weaknesses and weaknesses in its research direction that deviate from academic hotspots, thereby ensuring the relevance and practicality of the recommendation results and avoiding blind recommendations.
[0073] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this application. Those skilled in the art will understand that the modules in the apparatus of the embodiment can be distributed within the apparatus of the embodiment as described, or can be modified to be located in one or more apparatuses different from this embodiment. The modules of the above-described embodiment can be combined into one module, or further divided into multiple sub-modules.
[0074] The serial numbers in this application are for descriptive purposes only and do not represent the superiority or inferiority of any particular implementation scenario. The above disclosures are merely a few specific implementation scenarios of this application; however, this application is not limited thereto, and any variations conceived by those skilled in the art should fall within the protection scope of this application.
Claims
1. A method for recommending scientific research resources, characterized in that, include: Acquire internal and external academic data related to the target object; The internal academic data is input into the logical rule matrix corresponding to the assessment system document to obtain the assessment results for the target object. Based on the assessment results, analyze the assessment shortcomings of the target object; Based on the internal academic data and the external academic data, the academic research feature vector of the target object and the hot spot feature vector corresponding to the external academic hot spots are determined respectively. Based on the hotspot feature vector and the academic research feature vector, determine whether the target object has a shortcoming factor that deviates from the academic hotspot; If such factors exist, then resource recommendations will be made for the target object based on the shortcomings of the research direction deviating from academic hotspots and the shortcomings of the assessment.
2. The method according to claim 1, characterized in that, The acquisition of internal and external academic data related to the target object includes: Obtain project data, funding execution information, results data, and basic information related to the target object from the internal system; The project data, the funding execution information, the achievement data, and the basic information are identified as the internal academic data. Based on the attribute information of the target object, obtain literature abstracts and application guide texts related to the target object from external systems; The literature abstract and the application guide text are identified as the external learning data.
3. The method according to claim 1, characterized in that, Before inputting the internal academic data into the logical rule matrix corresponding to the assessment system document to obtain the assessment result corresponding to the target object, the method further includes: The assessment system document and the preset prompt word template are input into the logic rule matrix generation model to obtain the logic rule matrix. The preset prompt word template records the input and output information of the logic rule matrix, as well as the programming language used by the code module corresponding to the logic rule matrix. When generating the logic rule matrix, the logic rule matrix generation model reads each key indicator and its corresponding score division in the assessment system document, and generates the logic rule matrix based on the key indicators, the score division and the preset prompt word template.
4. The method according to claim 1, characterized in that, The step of inputting the internal academic data into the logical rule matrix corresponding to the assessment system document to obtain the assessment result corresponding to the target object includes: The internal academic data is input into the logical rule matrix corresponding to the assessment system document to obtain the scores corresponding to each key indicator. Based on the scores corresponding to each of the key indicators, the total assessment score of the target object is determined; Based on the overall assessment score, the assessment result corresponding to the target object is determined; The step of analyzing the performance shortcomings of the target object based on the assessment results includes: If the total assessment score is lower than the preset assessment score, then based on the scores corresponding to each key indicator, a target key indicator whose score is lower than the preset score is determined from among the key indicators. Based on the key performance indicators, determine the performance bottleneck factors corresponding to the target object.
5. The method according to claim 1, characterized in that, The step of determining the academic research feature vector of the target object and the hotspot feature vector corresponding to external academic hotspots based on the internal academic data and the external academic data includes: Based on the project data and achievement data in the aforementioned internal academic data, project keywords and project abstracts, as well as achievement keywords and achievement abstracts, are extracted respectively. The project keywords and the project summary are concatenated and then encoded into text to obtain the project feature vector of the target object; The keyword and abstract of the results are concatenated and then encoded into text to obtain the feature vector of the target object; The project feature vector and the achievement feature vector are weighted and fused to obtain the academic research feature vector; Obtain literature abstracts and application guide texts within a preset time window from the external academic data; Text encoding is performed on the literature abstracts and application guide texts within the preset time window to obtain literature hotspot feature vectors and guide guidance feature vectors; The hotspot feature vector of the literature and the guide-oriented feature vector are weighted and fused to obtain the hotspot feature vector corresponding to the external academic hotspot.
6. The method according to claim 1, characterized in that, The determination of whether the target object has a weakness factor that deviates from academic hotspots based on the hotspot feature vector and the academic research feature vector includes: Calculate the cosine similarity between the hotspot feature vector and the academic research feature vector; Based on the cosine similarity, it is determined whether the target object has any shortcomings in its research direction that deviates from academic hot topics.
7. The method according to claim 5, characterized in that, The resource recommendation for the target object based on the shortcomings factors of the research direction deviating from academic hotspots and the assessment shortcomings factors includes: Each resource is assigned a corresponding resource feature vector, wherein each resource includes internal resources and external resources; Based on the shortcomings of the research direction deviating from academic hotspots, the resource feature vectors corresponding to each resource, the hotspot feature vectors, and the academic research feature vectors are used to determine the degree to which each resource is close to the external academic hotspots, and the degree of fit between each resource and the target object. Determine the degree to which each resource is relevant to the assessment weakness factor; Based on the degree to which each resource is relevant to the external academic hot topics, the degree to which each resource is relevant to the target audience, and the degree of friendliness, a recommendation score is determined for each resource; Based on the recommendation scores of each resource, resource recommendations are made for the target object.
8. A scientific research resource recommendation device, characterized in that, include: The acquisition unit is used to acquire internal and external academic data related to the target object. The assessment unit is used to input the internal academic data into the logical rule matrix corresponding to the assessment system document to obtain the assessment result corresponding to the target object. The analysis unit is used to analyze the assessment shortcomings of the target object based on the assessment results. The determining unit is used to determine the academic research feature vector of the target object and the hot spot feature vector corresponding to the external academic hot spot based on the internal academic data and the external academic data, respectively. The determination unit is used to determine, based on the hotspot feature vector and the academic research feature vector, whether the target object has a shortcoming factor that deviates from the academic hotspot in its research direction; The recommendation unit is used to recommend resources to the target object based on the shortcomings factors of the research direction deviating from academic hotspots and the assessment shortcomings factors, if they exist.
9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.
10. An electronic device comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.
