Scholar recommendation method and device based on journal paper big data and electronic equipment
By using the scholar relationship network and core journal quantity sorting in scholar recommendations, the problem of ignoring scholar relationships in the existing technology is solved, and the accurate recommendation of scholars and the efficient utilization of academic resources is achieved.
Patent Information
- Application Number
- CN202411002829.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-25
- Publication Date
- 2024-12-17
AI Technical Summary
The existing technology ignores the relationship between scholars in the recommendation of scholars, making it difficult to achieve accurate recommendations, resulting in the lack of academic resources and the scarcity of manuscripts of top experts.
By obtaining the subjects, hot topics and topic selection directions entered by the user, filtering the target articles from the preset article database, using the preset scholar relationship network to determine other authors who have cooperative relationships with the author, and comprehensively sort them according to the core journal quantity. Scholar recommendations.
It fully considers the relationship between scholars, expands the search scope of scholars, realizes accurate recommendations from scholars, and improves the efficiency of academic resources and the academic level of journals.
Smart Images
Figure CN119149684A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of information technology, and in particular to a scholar recommendation method, device and electronic equipment based on journal article big data. Background Art
[0002] In today's digital age, the development of academic journals faces the challenges of information explosion and knowledge fragmentation. The lack of big data analysis in journal editorial board changes, invitations to reviewers, and invitations to authors has led to a lack of academic resources, insufficient high-quality manuscript sources, a scarcity of manuscripts from top experts, and a small number of editorial boards and reviewers in the journal editorial department, and their low level of activity. These problems have seriously restricted the high-quality development of journals. In order to improve the academic level and influence of journals, it is particularly important to accurately recommend scholars.
[0003] At present, scholars are usually recommended based on their academic research interests and professional expertise. However, in this recommendation method, each scholar is considered in isolation, which ignores the relationship between each scholar, making it difficult to achieve accurate recommendation of scholars. Summary of the invention
[0004] In view of this, the present application provides a scholar recommendation method, device and electronic device based on journal article big data, the main purpose of which is to achieve accurate recommendation of scholars.
[0005] According to the first aspect of the present application, a scholar recommendation method based on journal article big data is provided, the method comprising:
[0006] Obtain the disciplines, hot topics and topic directions input by the user;
[0007] Based on the subject, the hot topic and the topic direction, select target articles from a preset article database;
[0008] Based on the author of the target article and the preset scholar network, determine other authors who have a cooperative relationship with the author;
[0009] Based on the core journal volume of the author and the core journal volume of the other authors, comprehensively rank the author and the other authors to obtain a ranking result;
[0010] Based on the ranking results, scholars are recommended to the user.
[0011] According to a second aspect of the present application, a scholar recommendation device based on journal article big data is provided, the device comprising:
[0012] An acquisition unit, used to acquire the subject, hot topic and topic direction input by the user;
[0013] A screening unit, configured to screen target articles from a preset article database based on the subject, the hot topic and the topic direction;
[0014] A determination unit, configured to determine other authors having a cooperative relationship with the author based on the author corresponding to the target article and a preset scholar relationship network;
[0015] A ranking unit, used for comprehensively ranking the author and the other authors based on the core journal volume of the author and the core journal volume of the other authors, to obtain a ranking result;
[0016] A recommendation unit is used to recommend scholars to the user based on the ranking result.
[0017] According to the third aspect of the present application, a storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned scholar recommendation method based on journal article big data is implemented.
[0018] According to the fourth aspect of the present 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 when the processor executes the program, the above-mentioned scholar recommendation method based on journal article big data is implemented.
[0019] By means of the above technical solution, the present application provides a scholar recommendation method, device and electronic device based on journal article big data, which can screen out target articles from a preset article database based on the subject, hot topic and topic direction input by the user, and determine other authors with cooperative relationships with the author based on the author corresponding to the target article and the preset scholar relationship network, and finally recommend scholars to users based on the author corresponding to the target article and other authors. It can be seen from this that when recommending scholars, the present application can use the preset scholar relationship network to determine other scholars with cooperative relationships with relevant scholars, so as to fully consider the correlation between scholars, expand the search scope of scholars, and thus achieve accurate recommendation of scholars.
[0020] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0022] Figure 1 A schematic diagram of a flow chart of a scholar recommendation method based on journal article big data provided in an embodiment of the present application is shown;
[0023] Figure 2 A schematic diagram showing a user selecting a subject provided by an embodiment of the present application is shown;
[0024] Figure 3 A schematic diagram of hot topics of a subject selected by a user provided in an embodiment of the present application is shown;
[0025] Figure 4 A schematic diagram showing a user selecting a hot topic provided by an embodiment of the present application is shown;
[0026] Figure 5 A schematic diagram showing the topic selection direction of a hot spot selected by a user provided in an embodiment of the present application is shown;
[0027] Figure 6 A schematic diagram showing a user selecting a topic direction provided by an embodiment of the present application;
[0028] Figure 7 A schematic diagram of scholar recommendation provided by an embodiment of the present application is shown;
[0029] Figure 8 A schematic diagram of a process of another scholar recommendation method based on journal article big data provided in an embodiment of the present application is shown;
[0030] Fig. 9 A schematic diagram of the structure of a scholar recommendation device based on journal article big data provided in an embodiment of the present application is shown. DETAILED DESCRIPTION
[0031] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.
[0032] The existing technology considers each scholar in isolation and ignores the connections between scholars, making it difficult to achieve accurate recommendation of scholars.
[0033] In order to solve the above problems, the embodiment of the present invention provides a scholar recommendation method based on journal article big data, such as Figure 1 As shown, the method includes:
[0034] Step 101: Obtain the subject, hot topic and topic direction input by the user.
[0035] Among them, the disciplines include radio, telecommunications, computer hardware, computer software, the Internet, automation, etc. Hot topics are composed of different keywords, such as publishing industry, publishing power, and high-quality development. The topic direction refers to the research direction and theme, such as the connotation, path and challenge research of high-quality development of the publishing industry.
[0036] The embodiments of the present invention are mainly applicable to the accurate recommendation of journal editors, reviewers and commissioned authors. The execution subject of the embodiments of the present invention is a device or equipment, such as a server, that can accurately recommend journal editors, reviewers and commissioned authors. In the first embodiment, the specific implementation process of the journal editor recommendation scenario and the commissioned author recommendation scenario will be specifically introduced. In the second embodiment, the specific implementation process of the reviewer recommendation scenario will be specifically introduced.
[0037] For the embodiment of the present invention, in order to achieve accurate recommendations from journal editors, review experts, and commissioned authors, the embodiment of the present invention will collect papers, journals, and other articles from a recent period of time, such as collecting articles published in the past 10 years, and then classify and store the collected articles according to disciplines, hot topics, and topic directions. For this process, the method includes: collecting each article within a preset time period; classifying the collected articles according to disciplines to obtain each article under different disciplines; classifying each article under any of the different disciplines according to hot topics to obtain each article under different hot topics; classifying each article under any of the different hot topics according to topic directions to obtain each article under different topic directions. Among them, articles include papers, journals, and the like.
[0038] When the collected articles are classified according to disciplines, the first feature vectors of the collected articles are extracted based on the abstract information of the collected articles; the first feature vectors are input into a preset discipline classification model for discipline classification to obtain the discipline categories corresponding to the collected articles; the collected articles are divided based on the discipline categories to obtain the articles under the different disciplines. Specifically, the preset discipline classification model can be a random forest, support vector, naive Bayes, logistic regression model, etc.
[0039] Specifically, word segmentation is performed on the abstract information of each collected article to obtain the individual word segments corresponding to the abstract information. Then, some words without practical significance are removed, such as "of", "is", "therefore", etc., and only the words with practical significance are retained. Next, text cleaning is carried out, that is, some useless noise information is removed, such as numbers, punctuation marks, HTML tags, picture information, etc. Further, using word vector models such as Bag of Words (BOW), term frequency (TF), term frequency-inverse document frequency (TF-IDF), etc., the word vectors corresponding to each word are determined. If the dimension of the word vector is relatively high, feature dimensionality reduction algorithms such as principal component analysis (PCA), linear discriminant analysis (LDA), etc. can be used to reduce the dimension of the high-dimensional word vector and reduce the scale of the feature space. Further, in order to reduce the computational complexity in the process of subject classification, feature selection can also be performed to extract representative words in the abstract, such as selecting features related to the research field of the article and removing irrelevant features. Finally, the word vectors of the selected words are horizontally concatenated to obtain the first feature vector of each article.
[0040] Further, the extracted first feature vector is input into a preset subject classification model for subject classification to obtain the subject category corresponding to each article. Then, according to the subject category, each article is classified to obtain each article under different subjects.
[0041] When classifying each article under any subject according to the hot topic selection, based on the keywords of each article under the any subject, the second feature vector of each article under the any subject is extracted; the second feature vector is input into a preset hot topic classification model for hot topic classification to obtain the hot topic selection category corresponding to each article under the any subject; based on the hot topic selection category, each article under the any subject is divided to obtain each article under different hot topic selections. Among them, the preset hot topic classification model can specifically be a random forest, support vector, naive Bayes, logistic regression model, etc.
[0042] Specifically, for each article in any subject, determine the keywords involved in each article, and use the word vector model Bag of Words (BOW), term frequency (TF), term frequency-inverse document frequency (TF-IDF), etc. to determine the word vector of each keyword, and then horizontally splice the word vectors of each keyword to obtain the second feature vector of each article, and then input the second feature vector into the preset hot spot classification model for hot spot classification to obtain the hot topic of each article, and finally classify each article in any subject according to the hot topic to obtain each article under different hot topics. In addition, there is another classification method, which is to determine the number of identical or similar keywords in any two articles, classify the articles with the same or similar keywords reaching a certain value into one category, and determine the hot topics of this category of articles based on the identical or similar keywords.
[0043] When the articles under any hot topic are classified according to the topic direction, the target section text of each article under any hot topic is input into the preset topic model to predict the topic direction, and the predicted topic direction corresponding to each article under any hot topic is obtained; based on the predicted topic direction, each article under any hot topic is divided to obtain each article under the different topic directions. Among them, the preset topic model can specifically be an LDA topic model. The target paragraph text specifically includes an introduction, abstract, summary, etc.
[0044] Specifically, for each article under any topic direction, the target paragraph text is segmented and stop words are removed to obtain the processed segmented words. Then, the document bag of words model is used to convert the target paragraph text into a text vector, in which each vector element represents a word. Then, the text vector corresponding to each article is input into the LDA topic model to obtain the topic distribution of each article. Finally, the topic direction of each article is determined through the topic distribution. In addition, a machine learning algorithm can also be used to build a classification model. In specific applications, the text vector is input into the classification model for topic direction classification.
[0045] In this way, the collected articles can be classified and stored according to disciplines, hot topics and topic directions.
[0046] In order to achieve accurate recommendations for journal editors, reviewers, and commissioned authors, in addition to classifying and storing the collected articles according to disciplines, hot topics, and topic directions, the embodiment of the present invention also needs to build a preset scholar relationship network. For this process, the method includes: determining the published authors of each article under any of the hot topics; if any published author has published articles with other published authors or is in the same institution, then determining that there is a cooperative relationship between any of the published authors and the other published authors; based on the published authors and the other published authors, determining network nodes; based on the cooperative relationship, determining network edges; and building a preset scholar relationship network corresponding to any of the hot topics based on the network nodes and the network edges.
[0047] Specifically, since the collected articles have been classified, each article under any hot topic can be determined. When constructing a preset scholar relationship network, based on the authors of each article under any hot topic, it is determined whether any two authors have published articles together or are in the same institution. If any two authors have published articles together or are in the same institution, it is determined that there is a cooperative relationship between any two authors. After that, all the determined authors are used as network nodes respectively. If there is a cooperative relationship between any two authors, a line is connected between the corresponding network nodes of any two authors, thereby constructing a preset scholar relationship network corresponding to any hot topic. Through this preset scholar relationship network, the association relationship between scholars, that is, the cooperative relationship, can be determined.
[0048] After completing article classification and network construction, when it is necessary to recommend commissioned authors or journal editors, the user will enter the corresponding disciplines, hot topics and topic directions in the system interface.
[0049] Step 102: based on the subject, the hot topic and the topic direction, select target articles from a preset article database.
[0050] Among them, there are multiple target articles, and the target articles are articles that meet the user's subject requirements, hot topic requirements and topic direction requirements.
[0051] For the embodiment of the present invention, in order to filter out the target articles, step 102 specifically includes: based on the subject, determining the first article under the subject from the preset article database; based on the hot topic, determining the second article under the hot topic from the first article; based on the topic direction, determining the target article under the topic direction from the second article. Figure 2-6 As shown, the system can filter out target articles that meet the requirements based on the disciplines, hot topics and topic directions input by the user.
[0052] Step 103: Based on the author corresponding to the target article and a preset scholar relationship network, determine other authors who have a cooperative relationship with the author.
[0053] For the embodiment of the present invention, in order to expand the scope of scholar search, it is necessary to determine other authors who have a cooperative relationship with the author of the target article, and step 103 specifically includes: determining the target node corresponding to the author in the preset scholar relationship network; based on the target node, determining other nodes connected to the target node in the preset scholar relationship network; determining the authors corresponding to the other nodes as other authors who have a cooperative relationship with the author.
[0054] For example, the authors corresponding to the target article include author A, author B, and author C. The preset scholar relationship network corresponding to the hot topic input by the user is queried to determine the other nodes connected to the target nodes where author A, author B, and author C are located, thereby determining the connection relationship between the authors, such as author A is connected to author D, author B is connected to author A, author E, and author F, and author C is connected to author E. It can be determined that there is a cooperative relationship between author A and author D, there is a cooperative relationship between author B and author A, there is a cooperative relationship between author E and author F, there is a cooperative relationship between author C and author E, and author D, author E, and author F are other authors that have a cooperative relationship with author A, author B, and author C.
[0055] Step 104: Based on the core journal volume of the author and the core journal volume of the other authors, comprehensively rank the author and the other authors to obtain a ranking result.
[0056] Following the above example, author A, author B, author C, author D, author E, and author F are sorted according to the number of core journals to obtain the sorting results.
[0057] Step 105: Based on the ranking results, recommend scholars to the user.
[0058] According to the embodiment of the present invention, authors (scholars) ranked within a preset range can be recommended to the user based on the ranking results, such as recommending the top 5 authors to the user and displaying the titles of the authors' articles together, such as Figure 7 shown.
[0059] The embodiment of the present invention provides a scholar recommendation method based on journal article big data, which can use the preset scholar relationship network to determine other scholars who have cooperative relationships with relevant scholars, thereby fully considering the correlation between scholars, expanding the search scope of scholars, and further realizing accurate recommendation of scholars.
[0060] Further, as a refinement and extension of the specific implementation manner of the above embodiment, in order to fully illustrate the implementation manner of this embodiment, this embodiment also provides another method for recommending scholars based on big data of journal papers, as follows Figure 8 shown, this method includes:
[0061] Step 201, obtain the article to be reviewed.
[0062] Among them, the article to be reviewed is an article that needs to be recommended for review, specifically including papers, journals, etc.
[0063] In the embodiment of the present invention, not only can journal editorial board members and solicited authors be recommended, but also reviewers can be recommended for the article to be reviewed. When specifically recommending reviewers, the user can upload the article to be reviewed through the system.
[0064] Step 202, based on the abstract information, keywords and target paragraph text of the article to be reviewed, determine the review discipline, review hot topic and review topic direction corresponding to the article to be reviewed.
[0065] For the embodiment of the present invention, based on the abstract information of the article to be reviewed, the third feature vector of the article to be reviewed is extracted, and then the third feature vector of the article to be reviewed is input into a preset subject classification model for subject classification to obtain the review discipline corresponding to the article to be reviewed. Specifically, the abstract information of the article to be reviewed is segmented to obtain each segmented word corresponding to the abstract information, and then some words without practical significance are removed, such as "of", "is", "therefore", etc., and only words with practical significance are retained. Then, text cleaning is performed, that is, some useless noise information is removed, such as numbers, punctuation marks, HTML tags, picture information, etc. Further, using word vector models such as Bag of Words (BOW), term frequency (TF), term frequency-inverse document frequency (TF-IDF), etc., the word vector corresponding to each word is determined, and the feature vectors corresponding to each word are horizontally spliced to obtain the third feature vector of the article to be reviewed, and then it is input into a preset subject classification model for subject classification to obtain the review discipline corresponding to the article to be reviewed.
[0066] Furthermore, based on the keywords of the article to be reviewed, the fourth feature vector of the article to be reviewed is extracted, and then the fourth feature vector is input into the preset hot spot classification model for hot spot classification to obtain the hot topic selection corresponding to the article to be reviewed. Specifically, the word vector model Bag of Words (BOW), term frequency (TF), term frequency-inverse document frequency (TF-IDF), etc. is used to determine the word vector of each keyword, and then the word vectors of each keyword are horizontally spliced to obtain the fourth feature vector of the article to be reviewed, and then the fourth feature vector is input into the preset hot spot classification model for hot spot classification to obtain the hot topic selection corresponding to the article to be reviewed.
[0067] Furthermore, the target paragraph text of the article to be reviewed is input into the preset topic model to predict the topic direction, and the review topic direction corresponding to the article to be reviewed is obtained. Specifically, the target paragraph text is segmented and stop words are removed to obtain each processed word segmentation. Then, the document bag of words model is used to convert the target paragraph text into a text vector, in which each vector element represents a word. Then, the text vector corresponding to the article to be reviewed is input into the LDA topic model to obtain the topic distribution of the article to be reviewed, and finally the topic direction of the article to be reviewed is determined through the topic distribution.
[0068] Step 203: based on the review discipline, the review hot topic and the review topic direction, select relevant articles from the preset article database.
[0069] For the embodiment of the present invention, after determining the review discipline, review hot topic and review topic direction corresponding to the article to be reviewed, based on the review discipline, the third article under the review discipline is determined from the preset article database, and then based on the review hot topic, the fourth article under the review hot topic is determined from the third article, and finally based on the review topic direction, the relevant articles under the review topic direction are determined from the fourth article.
[0070] Step 204: Based on the review authors corresponding to the relevant articles and the preset scholar relationship network, determine other review authors who have a cooperative relationship with the review author.
[0071] For the embodiment of the present invention, in order to expand the scope of scholar search, it is necessary to determine other review authors who have a cooperative relationship with the review authors of the relevant articles. Specifically, the review node corresponding to the review author in the corresponding preset scholar relationship network is determined, and then based on the review node, other review nodes connected to the review node in the preset scholar relationship network are determined, and finally the review authors corresponding to the other review nodes are determined as other review authors.
[0072] For example, the review authors corresponding to the relevant articles include review author A, review author B, and review author C. The preset scholar relationship network corresponding to the hot topic input by the user is queried to determine the other review nodes connected to the review nodes where review author A, review author B, and review author C are located, thereby determining the connection relationship between the review authors, such as review author A is connected to review author D, review author B is connected to review author A, review author E, and review author F, and review author C is connected to review author E. It can be determined that there is a cooperative relationship between review author A and review author D, there is a cooperative relationship between review author B and review author A, there is a cooperative relationship between review author E and review author F, there is a cooperative relationship between review author C and review author E, and review author D, review author E, and review author F are other review authors that have a cooperative relationship with review author A, review author B, and review author C.
[0073] Step 205: Make scholar recommendations based on the review author and the other review authors.
[0074] For the embodiment of the present invention, after determining the review author and other review authors, the review author and other review authors can be comprehensively ranked based on the core journal volume of the review author and the core journal volume of the other review author to obtain a ranking result, and finally, based on the ranking result, the review author whose ranking is within a preset range is recommended for the article to be reviewed. (Scholar)
[0075] Another scholar recommendation method based on journal article big data provided by an embodiment of the present invention can use a preset scholar relationship network to determine other scholars who have a cooperative relationship with the relevant scholars, thereby fully considering the relationship between scholars, expanding the search scope of scholars, and further realizing accurate recommendation of scholars.
[0076] Further, as Figure 1 and Figure 8 The specific implementation of the method shown in the embodiment provides a scholar recommendation device based on journal article big data, such as Fig. 9 As shown, the device includes: an acquisition unit 31, a screening unit 32, a determination unit 33, a sorting unit 34 and a recommendation unit 35.
[0077] The acquisition unit 31 may be used to acquire the subject, hot topic and topic direction input by the user.
[0078] The screening unit 32 may be used to screen target articles from a preset article database based on the subject, the hot topic and the topic direction.
[0079] The determination unit 33 may be configured to determine other authors having a cooperative relationship with the author based on the author corresponding to the target article and a preset scholar relationship network.
[0080] The ranking unit 34 may be used to comprehensively rank the author and the other authors based on the core journal volume of the author and the core journal volume of the other authors to obtain a ranking result.
[0081] The recommendation unit 35 may be configured to recommend scholars to the user based on the ranking result.
[0082] In some embodiments, the screening unit 32 can be specifically used to determine the first article under the subject from the preset article database based on the subject; based on the hot topic, determine the second article under the hot topic from the first article; based on the topic direction, determine the target article under the topic direction from the second article.
[0083] In some embodiments, the determination unit 33 can be specifically used to determine the target node corresponding to the author in the preset scholar relationship network; based on the target node, determine other nodes connected to the target node in the preset scholar relationship network; and determine the authors corresponding to other nodes as other authors who have a cooperative relationship with the author.
[0084] In some embodiments, the device further comprises: a collection unit and a classification unit.
[0085] The collecting unit may be used to collect articles within a preset time period.
[0086] The classification unit can be used to classify the collected articles according to subjects to obtain articles under different subjects.
[0087] The classification unit may also be used to classify each article under any of the different disciplines according to hot topics to obtain each article under different hot topics.
[0088] The classification unit may also be used to classify each article under any of the different hot topics according to the topic direction to obtain each article under different topic directions.
[0089] In some embodiments, the classification unit can be specifically used to extract the first feature vector of each collected article based on the abstract information of each collected article; input the first feature vector into a preset subject classification model for subject classification to obtain the subject category corresponding to each collected article; divide the collected articles based on the subject category to obtain each article under the different subjects.
[0090] In some embodiments, the classification unit can also be specifically used to extract a second feature vector of each article under any subject based on the keywords of each article under any subject; input the second feature vector into a preset hot spot classification model for hot spot classification to obtain the hot topic category corresponding to each article under any subject; divide each article under any subject based on the hot topic category to obtain each article under different hot topics.
[0091] In some embodiments, the classification unit can also be specifically used to input the target paragraph text of each article under any hot topic into a preset topic model to predict the topic direction, and obtain the predicted topic direction corresponding to each article under any hot topic; based on the predicted topic direction, the articles under any hot topic are divided to obtain the articles under different topic directions.
[0092] In some embodiments, the device further comprises: a construction unit.
[0093] The construction unit can be used to determine the published authors of each article under any of the hot topics; if any published author has published articles with other published authors or is in the same institution, it is determined that there is a cooperative relationship between the any published author and the other published authors; based on the published authors and the other published authors, network nodes are determined; based on the cooperative relationship, network edges are determined; based on the network nodes and the network edges, a preset scholar relationship network corresponding to any of the hot topics is constructed.
[0094] In some embodiments, the acquisition unit 31 can also be used to acquire articles to be reviewed.
[0095] The determination unit 33 may also be used to determine the review discipline, review hot topic and review topic direction corresponding to the article to be reviewed based on the abstract information, keywords and target paragraph text of the article to be reviewed.
[0096] The screening unit 32 may also be used to screen out relevant articles from the preset article database based on the review discipline, the review hot topic and the review topic direction.
[0097] The determining unit 33 may also be used to determine other review authors having a cooperative relationship with the review author based on the review authors corresponding to the relevant articles and the preset scholar relationship network.
[0098] The recommendation unit 35 may also be used to make scholar recommendations based on the review author and the other review authors.
[0099] It should be noted that for other corresponding descriptions of the functional units involved in the scholar recommendation device based on journal article big data provided by the embodiment of the present invention, please refer to Figure 1 and Figure 8 The corresponding description in will not be repeated here.
[0100] Based on the above Figure 1 and Figure 8 The method shown in the embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned Figure 1 and Figure 8 The scholar recommendation method based on journal article big data is shown.
[0101] Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, USB flash drive, mobile hard disk, etc.), including a number of instructions for enabling an electronic device (which can be a personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of the present application.
[0102] Based on the above Figure 1 and Figure 8 The method shown, and Fig. 9 In order to achieve the above-mentioned purpose, the embodiment of the present application also provides an electronic device, which can be a personal computer, a tablet computer, a server, or other network equipment, etc. The device includes a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to achieve the above-mentioned Figure 1 and Figure 8 The scholar recommendation method based on journal article big data is shown.
[0103] Optionally, the above-mentioned physical device may also include a user interface, a network interface, a camera, a radio frequency (RF) circuit, a sensor, an audio circuit, a WI-FI module, etc. The user interface may include a display, an input unit such as a keyboard, etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface), etc.
[0104] Those skilled in the art will appreciate that the above-mentioned physical device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or a combination of certain components, or different arrangements of components.
[0105] 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 above-mentioned physical device, and supports the operation of the information processing program and other software and / or programs. The network communication module is used to realize the communication between the components inside the storage medium, and the communication with other hardware and software in the information processing physical device.
[0106] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform, or by hardware.
[0107] The embodiment of the present invention can utilize the preset scholar relationship network to determine other scholars who have cooperative relationships with relevant scholars, thereby being able to fully consider the associations between scholars, expand the search scope of scholars, and further achieve accurate recommendation of scholars.
[0108] Those skilled in the art will appreciate that the accompanying drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily necessary for implementing the present application. Those skilled in the art will appreciate that the modules in the devices in the implementation scenario can be distributed in the devices of the implementation scenario according to the description of the implementation scenario, or can be changed accordingly and located in one or more devices different from the present implementation scenario. The modules of the above-mentioned implementation scenario can be combined into one module, or can be further split into multiple submodules.
[0109] The above serial numbers of this application are only for description and do not represent the advantages and disadvantages of the implementation scenarios. The above disclosure is only a few specific implementation scenarios of this application, but this application is not limited to them, and any changes that can be thought of by technicians in this field should fall within the scope of protection of this application.
Claims
1. A scholar recommendation method based on journal article big data, characterized in that: include: Obtain the disciplines, hot topics and topic directions input by the user; Based on the subject, the hot topic and the topic direction, select target articles from a preset article database; Based on the author of the target article and the preset scholar network, determine other authors who have a cooperative relationship with the author; Based on the core journal volume of the author and the core journal volume of the other authors, comprehensively rank the author and the other authors to obtain a ranking result; Based on the ranking results, scholars are recommended to the user.
2. The method according to claim 1, characterized in that Based on the discipline, the hot topic and the topic direction, target articles are screened from the preset article database, including: Based on the subject, determining a first article under the subject from the preset article database; Based on the hot topic, determining a second article under the hot topic from the first article; Based on the topic selection direction, a target article under the topic selection direction is determined from the second articles.
3. The method according to claim 1, characterized in that Based on the author of the target article and the preset scholar network, other authors who have a cooperative relationship with the author are determined, including: Determine the target node corresponding to the author in the preset scholar relationship network; Based on the target node, determining other nodes in the preset scholar relationship network connected to the target node; The authors corresponding to the other nodes are determined as other authors having a cooperative relationship with the author.
4. The method according to claim 1, characterized in that: Before obtaining the subject, hot topic and topic direction input by the user, the method further includes: Collect individual articles within a preset time period; Classify the collected articles according to subjects to obtain articles under different subjects; Classify each article under any of the different subjects according to hot topics to obtain each article under different hot topics; Each article under any of the different hot topics is classified according to the topic direction to obtain each article under different topic directions.
5. The method according to claim 4, characterized in that The collected articles are classified according to subjects to obtain articles under different subjects, including: Extracting first feature vectors of the collected articles based on the collected summary information of the articles; Inputting the first feature vector into a preset subject classification model for subject classification, and obtaining subject categories corresponding to the collected articles; Divide the collected articles based on the subject categories to obtain articles under different subjects; Classify each article under any of the different disciplines according to hot topics to obtain each article under different hot topics, including: Extracting a second feature vector of each article under any subject based on the keywords of each article under any subject; Inputting the second feature vector into a preset hot topic classification model for hot topic classification, and obtaining the hot topic category corresponding to each article under any subject; Divide the articles under any subject based on the hot topic categories to obtain articles under different hot topics; Classify each article under any of the different hot topics according to the topic direction to obtain each article under different topic directions, including: Input the target paragraph text of each article under any hot topic into a preset topic model to predict the topic direction, and obtain the predicted topic direction corresponding to each article under any hot topic; The articles under any hot topic are divided based on the predicted topic direction to obtain the articles under the different topic directions.
6. The method according to claim 4, characterized in that The method further comprises: Determine the authors of each article under any of the hot topics; If any published author has published articles with other published authors or is in the same institution, it is determined that there is a cooperative relationship between any published author and the other published authors; Determining a network node based on the publishing author and the other publishing authors; Based on the cooperative relationship, determine the network edge; A preset scholar relationship network corresponding to any hot topic is constructed based on the network nodes and the network edges.
7. The method according to claim 1, characterized in that The method comprises: Get the articles to be reviewed; Based on the abstract information, keywords and target paragraph text of the article to be reviewed, determine the review discipline, review hot topic and review topic direction corresponding to the article to be reviewed; Based on the review discipline, the review hot topic and the review topic direction, select relevant articles from the preset article database; Based on the review authors corresponding to the relevant articles and the preset scholar network, determine other review authors who have a cooperative relationship with the review author; Based on the review authors and the other review authors, scholar recommendations are made.
8. A scholar recommendation device based on journal article big data, characterized in that: include: An acquisition unit, used to acquire the subject, hot topic and topic direction input by the user; A screening unit, configured to screen target articles from a preset article database based on the subject, the hot topic and the topic direction; A determination unit, configured to determine other authors having a cooperative relationship with the author based on the author corresponding to the target article and a preset scholar relationship network; A ranking unit, used for comprehensively ranking the author and the other authors based on the core journal volume of the author and the core journal volume of the other authors, to obtain a ranking result; A recommendation unit is used to recommend scholars to the user based on the ranking result.
9. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.
10. An electronic device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.