Discipline hotspot index calculation method and device, storage medium and computer equipment

Abstract

This application discloses a method, apparatus, storage medium, and computer equipment for calculating a subject hotspot index. The method includes: obtaining the subject, keywords, publication year, and citation count of documents; for documents belonging to the same subject, removing non-subject-related words from the corresponding keywords and performing deduplication to obtain subject keywords; for documents corresponding to the same subject keywords, calculating the average annual citation frequency per document based on the year difference between the publication year and the publication year, the total number of documents, and the total number of citations, and calculating the publication growth rate based on the number of documents published in the earliest and latest publication years, respectively; for multiple subject keywords corresponding to a subject, calculating the hotspot index of any subject keyword within the subject based on the total number of documents corresponding to the subject keywords, the average annual citation frequency per document, and the publication growth rate. By using composite calculation indicators, subject hotspots are evaluated based on the publication volume, increase, and citation value of documents, providing a more comprehensive assessment.

Description

TECHNICAL FIELD

[0001] The present application relates to the academic hot spot data mining technical field, and particularly to a subject hot spot index calculation method and device, a storage medium and a computer device. BACKGROUND

[0002] At present, the newly published academic literature is in a large amount every year, and academic information is increasingly rich, therefore, how to find the research hot spot of the subject field in a large amount of academic literature has become a problem concerned by academic researchers.

[0003] Generally, scholars find relevant literature by searching keywords related to the research field, and the keywords are usually key concepts extracted from the title, abstract and text of the literature, which play a role in describing, retrieving, understanding and standardizing the content of the literature, and improving the visibility and citation of the literature. The current judgment of the attention degree of the keywords generally uses the cited number, download number and other indicators of the literature related to the keywords to evaluate the "hotness" of the keywords.

[0004] However, such a direct statistical single indicator method has its one-sidedness, and only through a single original evaluation indicator of all the literature in the research field, the front hot spot of a research field cannot be comprehensively evaluated. SUMMARY

[0005] Therefore, the present application provides a subject hot spot index calculation method and device, a storage medium and a computer device, the subject, keywords, publication year and citation number of the literature are obtained; for the literature belonging to the same subject, the non-subject related words in the keywords corresponding to the literature are removed, and then the de-duplication processing is performed to obtain subject keywords; for the literature corresponding to the same subject keyword, the annual average literature cited frequency is calculated according to the start and end year difference of the publication year of the literature, the total number of literature and the total citation number, and the publication growth rate is calculated according to the number of literature published in the earliest and latest publication years; for the multiple subject keywords corresponding to the subject, the hot spot index of any subject keyword in the subject is calculated according to the total number of literature corresponding to the subject keyword, the annual average literature cited frequency and the publication growth rate. The subject hot spot is evaluated according to the publication volume, increment and citation value by composite calculation index, which is more comprehensive.

[0006] According to one aspect of the present application, a subject hot spot index calculation method is provided, which comprises:

[0007] obtaining the subject, keywords, publication year and citation number of the literature;

[0008] The non-discipline-related words in the keywords corresponding to the literature are removed, and the discipline keywords are obtained by removing the repeated literature;

[0009] The literature corresponding to the same discipline keyword is processed according to the difference between the start and end years of the publication year, the total number of articles and the total number of citations of the literature, the annual article average literature cited frequency is calculated, and the publication growth rate is calculated according to the number of articles published in the earliest publication year and the latest publication year.

[0010] For a plurality of discipline keywords corresponding to a discipline, the total number of articles, the annual article average literature cited frequency and the publication growth rate of the discipline keyword are calculated, and the hot spot index of any discipline keyword in the discipline is calculated.

[0011] Optionally, the annual article average literature cited frequency is calculated according to the difference between the start and end years of the publication year, the total number of articles and the total number of citations of the literature, including:

[0012] According to the publication year of the literature, the difference between the start and end years of the publication year is determined.

[0013] According to the annual article average literature cited frequency calculation formula, the annual article average literature cited frequency in the difference between the start and end years of the publication year is calculated, wherein the annual article average literature cited frequency calculation formula is:

[0014]

[0015] CPT i The annual article average literature cited frequency of the discipline keyword i is C i The total number of citations of the literature corresponding to the discipline keyword i in the difference between the start and end years of the publication year is P i The total number of articles of the literature corresponding to the discipline keyword i in the difference between the start and end years of the publication year is T i The difference between the start and end years of the publication year of the literature corresponding to the discipline keyword i in the discipline.

[0016] Optionally, the publication growth rate is calculated according to the number of articles published in the earliest publication year and the latest publication year, including:

[0017] According to the publication growth rate calculation formula, the publication growth rate of the literature published in the latest publication year compared with the number of articles published in the earliest publication year in the difference between the start and end years of the publication year is calculated, wherein the publication growth rate calculation formula is:

[0018]

[0019] PG i The publication growth rate of the discipline keyword i is P endThe number of papers published in the latest publication year among the papers corresponding to the subject keyword, P begin The number of papers published in the earliest publication year among the papers corresponding to the subject keyword.

[0020] Optionally, the hot spot index of any subject keyword in the subject is calculated according to the total number of papers corresponding to the subject keyword, the annual paper citation frequency, and the publication growth rate, including:

[0021] Select any one of the total number of papers, the annual paper citation frequency, and the publication growth rate as the target hot spot index, and obtain the hot spot index value of each subject keyword for the target hot spot index corresponding to the subject;

[0022] In the hot spot index value, any subject keyword is selected as a target subject keyword, the first difference value between the hot spot index value of the target subject keyword and the minimum hot spot index value is calculated, and the second difference value between the maximum hot spot index value and the minimum hot spot index value is calculated. According to the ratio of the first difference value to the second difference value, the linear normalization value of the target hot spot index corresponding to the target subject keyword is obtained;

[0023] The linear normalization values of each target hot spot index corresponding to the target subject keyword are summed to obtain the hot spot index of the target subject keyword in the subject.

[0024] Optionally, the non-subject related words in the keywords corresponding to the literature are removed and the de-duplication processing is performed to obtain the subject keywords, including:

[0025] The keywords corresponding to the literature are matched based on the preset subject non-related word library, and the matched keywords are removed. Based on the non-repeated keywords after removal, the subject keywords are obtained.

[0026] Optionally, the establishment method of the subject non-related word library is:

[0027] Obtain the literature sample, the publication journal name of the literature sample, the belonging two-level keyword sample of the literature sample, the belonging subject of the keyword sample, and the subject relevance identification situation corresponding to the keyword sample;

[0028] According to the subject relevance identification result, the non-subject related keyword sample is determined as the non-subject related keyword sample, and the non-subject related keyword library is established based on the non-subject related keyword.

[0029] Optionally, the subject includes two-level subject specialty and one-level subject category, the one-level subject category corresponds to a plurality of two-level subject specialties, and the method further includes:

[0030] obtaining a literature sample, a publication journal name of the literature sample, a two-level discipline specialty to which the literature sample belongs, and a one-level discipline category corresponding to the two-level discipline specialty;

[0031] For the literature samples with the same two-level discipline specialty, after the publication journal names corresponding to the literature samples are processed for deduplication, a discipline-related journal information library is established according to the deduplicated publication journal names, the two-level discipline specialty and the one-level discipline category corresponding to the two-level discipline specialty, wherein the discipline-related journal information library includes discipline-related journal names corresponding to different two-level discipline specialties under different one-level discipline categories;

[0032] Correspondingly, the keywords corresponding to the literature are matched based on the preset non-discipline-related vocabulary library, and the matched keywords are removed, and discipline keywords are obtained based on the non-repeated keywords after removal, including:

[0033] obtaining a publication journal name of the literature, performing journal name matching on the publication journal name corresponding to the literature based on the discipline-related journal information library, and obtaining initial discipline keywords according to the keywords corresponding to the literature matched successfully;

[0034] The initial discipline keywords are matched based on the preset non-discipline-related vocabulary library, and the matched initial discipline keywords are removed, and discipline keywords are obtained based on the non-repeated initial discipline keywords after removal.

[0035] According to another aspect of the present application, a discipline hot spot index calculation device is provided, which comprises:

[0036] The literature information acquisition module is configured to obtain the discipline to which the literature belongs, the keywords, the publication year and the number of citations of the literature;

[0037] The discipline keyword filtering module is configured to, for the literature belonging to the same discipline, remove the non-discipline-related vocabulary in the keywords corresponding to the literature and perform deduplication to obtain discipline keywords;

[0038] The keyword index calculation module is configured to, for the literature corresponding to the same discipline keyword, calculate the annual average literature cited frequency according to the start and end year difference of the publication year of the literature, the total number of literature and the total number of citations, and calculate the publication growth rate according to the number of literature published in the earliest publication year and the latest publication year, respectively;

[0039] The discipline hot spot representation module is configured to, for a plurality of discipline keywords corresponding to a discipline, calculate the hot spot index of any discipline keyword in the discipline according to the total number of literature corresponding to the discipline keyword, the annual average literature cited frequency and the publication growth rate.

[0040] Optionally, the keyword indicator calculation module is further configured to:

[0041] determine a publication year start-end difference between an earliest publication year and a latest publication year according to the publication years corresponding to the documents;

[0042] calculate an annual average document citation frequency within the publication year start-end difference according to an annual average document citation frequency calculation formula, wherein the annual average document citation frequency calculation formula is:

[0043]

[0044] CPT i is an annual average document citation frequency of a subject keyword i, C i is a total number of citations of the documents corresponding to the subject keyword i within the publication year start-end difference, P i is a total number of documents of the documents corresponding to the subject keyword i within the publication year start-end difference, T i is a publication year start-end difference of the documents corresponding to the subject keyword i in the subject.

[0045] Optionally, the keyword indicator calculation module is further configured to:

[0046] calculate a publication growth rate of the documents published in the latest publication year compared to the number of documents published in the earliest publication year within the publication year start-end difference according to a publication growth rate calculation formula, wherein the publication growth rate calculation formula is:

[0047]

[0048] PG i is a publication growth rate of a subject keyword i, P end is a number of documents published in the latest publication year among the documents corresponding to the subject keyword, P begin is a number of documents published in the earliest publication year among the documents corresponding to the subject keyword.

[0049] Optionally, the subject hotspot representation module is further configured to:

[0050] select any one of the total number of documents, the annual average document citation frequency and the publication growth rate as a target hotspot indicator, and obtain a hotspot indicator value of each subject keyword for the target hotspot indicator for a plurality of subject keywords corresponding to the subject.

[0051] In the hotspot index values, any discipline keyword is selected as a target discipline keyword, a first difference value between the hotspot index value of the target discipline keyword and the minimum hotspot index value is calculated, and a second difference value between the maximum hotspot index value and the minimum hotspot index value is calculated, a linear normalization value of a target hotspot index corresponding to the target discipline keyword is obtained according to a ratio of the first difference value to the second difference value.

[0052] The linear normalization values of the target hotspot indexes corresponding to the target discipline keyword are summed to obtain a hotspot index of the target discipline keyword in the discipline.

[0053] Optionally, the discipline keyword filtering module is further configured to:

[0054] The keywords corresponding to the literature are matched based on a preset discipline irrelevant vocabulary library, and the matched keywords are removed, and discipline keywords are obtained based on the non-repeated keywords after the removal.

[0055] Optionally, the apparatus further comprises a vocabulary establishing module configured to:

[0056] The keyword samples, the disciplines to which the keyword samples belong, and the discipline relevance recognition conditions corresponding to the keyword samples are obtained.

[0057] For the keyword samples belonging to the same discipline, non-discipline relevant keywords are determined according to the discipline relevance recognition results, and a discipline irrelevant vocabulary library is established based on the non-discipline relevant keywords.

[0058] Optionally, the discipline includes a secondary discipline major and a primary discipline category, the primary discipline category corresponds to a plurality of secondary discipline majors, and the vocabulary establishing module is further configured to:

[0059] The literature samples, the publication journal names of the literature samples, the secondary discipline majors to which the literature samples belong, and the primary discipline categories corresponding to the secondary discipline majors are obtained.

[0060] For the literature samples having the same secondary discipline major, the publication journal names corresponding to the literature samples are processed for deduplication, and a discipline related journal information library is established according to the secondary discipline major and the primary discipline category corresponding to the secondary discipline major, and the publication journal names after the deduplication processing, wherein the discipline related journal information library includes discipline related journal names corresponding to different secondary discipline majors under different primary discipline categories.

[0061] Optionally, the discipline keyword filtering module is further configured to:

[0062] obtaining a publication journal name of the literature, performing journal name matching on the publication journal name corresponding to the literature based on a subject-related journal information base, and obtaining initial subject keywords according to keywords corresponding to the literature that are successfully matched;

[0063] performing keyword matching on the initial subject keywords based on a preset subject non-related keyword base, and eliminating the initial subject keywords that are successfully matched, and obtaining subject keywords based on the non-repeated initial subject keywords after elimination.

[0064] According to another aspect of the present application, a storage medium having a computer program stored thereon is provided, and the program is executed by a processor to implement the subject hotspot index calculation method.

[0065] According to another aspect of the present application, a computer device is provided, which includes a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, and the processor implements the subject hotspot index calculation method when executing the program.

[0066] Through the above technical solutions, the subject hotspot index calculation method and device, storage medium, and computer device provided by the present application filter keywords, improve data quality, and comprehensively evaluate subject heat through calculation of total number of literatures, annual average number of literatures, cited frequency of literatures, and publication growth rate, realize a hotspot evaluation mode that takes into account academic level, research increment mutation, and research volume, and the subject hotspot keywords evaluated according to the composite calculation index can effectively help scholars evaluate the degree of attention to content, theme, and research field in 168 subjects, assist in subject hotspot discovery and research trend analysis, and have important guiding significance for research topic selection, research development, and finding of partners.

[0067] The above description is only a summary of the technical solutions of the present application, and in order to more clearly understand the technical means of the present application, the present application can be implemented according to the content of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0068] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The schematic 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:

[0069] Figure 1 A flowchart of a subject hotspot index calculation method provided by an embodiment of the present application is shown;

[0070] Figure 2 A flowchart of another subject hotspot index calculation method provided by an embodiment of the present application is shown.

[0071] Figure 3 A flowchart of another method for calculating a discipline hot spot index is shown;

[0072] Figure 4 A flowchart of another method for calculating a discipline hot spot index is shown;

[0073] Figure 5 A structural diagram of a discipline hot spot index calculation device is shown;

[0074] Figure 6 A structural diagram of another discipline hot spot index calculation device is shown. DETAILED DESCRIPTION

[0075] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0076] A discipline hot spot index calculation method is provided in the present embodiment, as shown in Figure 1 The method comprises:

[0077] Step 101, obtaining the discipline, keywords, publication year and citation times of the literature.

[0078] In the above embodiments of the present application, the original data of the published literature is extracted from the literature database. The literature database, such as "China Knowledge Network", can obtain literature materials from other online websites and resource stations that can query literature, and then extract some basic information of the literature from the literature materials, including the discipline, keywords, publication year and citation times, etc. Specifically, the keywords of the literature can be obtained directly according to the keywords listed in the literature, and other information such as the discipline, publication year and citation times can be obtained according to the literature related information marked in the literature database. In particular, the discipline can be obtained at the same time, including the first discipline category and the second discipline specialty. Through more detailed professional classification, the discipline hot spot can be more focused on the field. In addition, in order to ensure the timeliness of the hot spot evaluation, an evaluation period can be set, and then the literature materials published within the aforementioned evaluation period are obtained, for example, the literature materials in the past five years or the past three years are obtained, so as to ensure that the recent hot spots in the academic field can be evaluated, and the precision of the hot spot evaluation is improved.

[0079] Step 102, for the literature belonging to the same discipline, the non-discipline related words in the keywords corresponding to the literature are removed, and then the de-duplication processing is performed, to obtain the discipline keywords.

[0080] Next, for the literature belonging to the same discipline, for example, the keywords extracted from all the literature under the first-level discipline category of "economics", non-discipline-related words are removed to improve the accuracy of the basic data. The aforementioned non-discipline-related words are, for example, project application, experiment, optimization, thought, meaning, and other words unrelated to the discipline or too general. After filtering out invalid keywords, based on the remaining non-repeated keywords, the discipline keywords are obtained. For example, for the "art" discipline, the extracted discipline keywords can be: geometric figure expression, expression method, organizational relationship, and color characteristics.

[0081] Step 103, for the literature corresponding to the same discipline keyword, according to the difference between the start and end years of the publication year of the literature, the total number of articles and the total number of citations, the annual article average literature cited frequency is calculated, and according to the number of articles published in the earliest publication year and the latest publication year, the publication growth rate is calculated.

[0082] Next, for the literature corresponding to the same discipline keyword, for example, the keyword "geometric figure expression" is obtained in a total of 400 articles, at this time, in the aforementioned 400 articles, the earliest publication year of the literature is 1986, the latest publication year of the literature is 2022, that is, the difference between the start and end years of the publication year is 37 years, and the total number of citations in the 400 articles is 267 times. According to the difference between the start and end years of the publication year (37 years), the total number of citations in the 400 articles (total number of articles) published in the 37 years, the annual article average literature cited frequency of "geometric figure expression" is calculated, and according to the 67 articles published in 1986 and the 123 articles published in 2022, the publication growth rate of "geometric figure expression" is calculated. By calculating the annual article average literature cited frequency and the publication growth rate of the discipline keyword, the research increment mutation is considered while measuring the academic level, and the horizontal comparison index is realized.

[0083] Step 104, for the multiple discipline keywords corresponding to the discipline, according to the total number of articles of the discipline keywords, the annual article average literature cited frequency and the publication growth rate, the hot spot index of any discipline keyword in the discipline is calculated.

[0084] Further, for example, when calculating the hot spot index of the "geometric pattern expression" subject keyword under the "art" subject, the total number of documents, the annual average document citation frequency, and the publication growth rate of each subject keyword (geometric pattern expression, expression method, organizational relationship, and color characteristics) under the "art" subject can be used to calculate the hot spot index of the "geometric pattern expression" subject keyword in the "art" subject. Based on the hot spot index corresponding to each subject keyword, the top hot spot index corresponding to the hot spot keyword in all hot spot indexes can be determined, and academic researchers can conduct deeper academic research based on the hot spot keyword. The aforementioned preset number can be 2 or 3 or other numbers, which can be set at any time according to different analysis requirements.

[0085] By applying the technical solution of the embodiment, the subject, keyword, publication year, and citation frequency of the document are obtained. For documents belonging to the same subject, the non-subject-related keywords in the corresponding keywords of the documents are removed, and the subject keywords are obtained after de-duplication processing. For documents corresponding to the same subject keyword, the annual average document citation frequency is calculated based on the start and end year difference of the publication year of the documents, the total number of documents, and the total citation frequency. The publication growth rate is calculated based on the number of documents published in the earliest and latest publication years. For multiple subject keywords corresponding to the subject, the hot spot index of any subject keyword in the subject is calculated based on the total number of documents corresponding to the subject keyword, the annual average document citation frequency, and the publication growth rate. By filtering the keywords, the data quality is improved. By calculating the total number of documents, the annual average document citation frequency, and the publication growth rate, the subject hot spot is comprehensively evaluated, which can assist in subject hot spot discovery and research trend analysis while considering academic level, research increment mutation, and research volume.

[0086] Further, as a refinement and expansion of the above embodiment, in order to fully describe the specific implementation process of the embodiment, another subject hot spot index calculation method is provided, as shown in Figure 2 The method comprises the following steps:

[0087] Step 201, obtaining the subject, keyword, publication year, and citation frequency of the document.

[0088] Step 202, for documents belonging to the same subject, performing keyword matching on the corresponding keywords of the documents based on a preset subject non-related vocabulary library, and removing the matched keywords. Based on the non-repeated keywords after removal, the subject keywords are obtained.

[0089] Optionally, the method for establishing the discipline-unrelated vocabulary library comprises the following steps: obtaining keyword samples, disciplines to which the keyword samples belong, and discipline-relatedness identification conditions corresponding to the keyword samples; determining discipline-unrelated keywords according to the discipline-relatedness identification results for keyword samples belonging to the same discipline; and establishing the discipline-unrelated vocabulary library based on the discipline-unrelated keywords.

[0090] In the above embodiments of the present application, original information of published literature is extracted from a literature database, the original information including a discipline to which the literature belongs, keywords, a publication year, and a number of citations, and meanwhile, other information such as a two-level discipline professional code to which the literature belongs, a two-level discipline professional name to which the literature belongs, and a literature download frequency can also be obtained. Then, for literature belonging to the same discipline, keyword matching is performed on keywords corresponding to the literature according to a preset discipline-unrelated vocabulary library, and keywords that match successfully are removed. Then, discipline keywords are obtained based on the removed non-repeated keywords. Through the above process, effective keywords related to a discipline are quickly extracted from massive literature data, and the problem of information overload is effectively reduced. In particular, for the establishment of the discipline-unrelated vocabulary library, keyword samples, disciplines to which the keyword samples belong, and discipline-relatedness identification conditions corresponding to the keyword samples can be obtained in advance. For example, the keyword samples obtained are “contemporization”, “geometric image expression”, and “color characteristics”. After semantic analysis, the discipline-relatedness identification condition of the keyword “contemporization” is non-discipline-related, and then the keyword “contemporization” is added to the discipline-unrelated vocabulary library. Finally, the discipline-unrelated vocabulary library containing non-discipline-related keywords is obtained. Through the collection and arrangement of generalization keywords, commonly used keywords, and stop words, and the generation of the discipline-unrelated vocabulary library, keywords can be effectively filtered, and the accuracy of hot spot judgment is improved.

[0091] In step 203, for literature corresponding to the same discipline keyword, a publication year start-end year difference between the earliest publication year and the latest publication year is determined according to a publication year corresponding to the literature.

[0092] In step 204, an annual article average literature citation frequency in the publication year start-end year difference is calculated according to an annual article average literature citation frequency calculation formula, wherein the annual article average literature citation frequency calculation formula is:

[0093]

[0094] CPT i is the annual article average literature citation frequency of the discipline keyword i, C i is the total number of citations of the literature corresponding to the discipline keyword i in the publication year start-end year difference, P i is the total number of articles of the literature corresponding to the discipline keyword i in the publication year start-end year difference, T iThe difference between the end year and the start year of the publication year of the literature corresponding to the discipline keyword i.

[0095] In step 205, the publication growth rate of the literature published in the latest publication year compared to the number of articles published in the earliest publication year is calculated according to the publication growth rate calculation formula within the difference between the start year and the end year of the publication year, wherein the publication growth rate calculation formula is:

[0096]

[0097] PG i The publication growth rate of the discipline keyword i is P end The number of articles published in the latest publication year in the literature corresponding to the discipline keyword is P begin The number of articles published in the earliest publication year in the literature corresponding to the discipline keyword.

[0098] Then, for the literature corresponding to the same discipline keyword, the difference between the start year and the end year of the publication year is determined according to the publication year of the literature, and then the annual average literature citation frequency of the discipline keyword is calculated according to the annual average literature citation frequency calculation formula. Specifically, by comparing the total number of articles and the total number of citations in the difference between the start year and the end year of the publication of the literature, the annual average literature citation frequency is obtained, and the number of articles published in the earliest publication year and the latest publication year is calculated. The publication growth rate is calculated by studying the volume (total number of articles), increment (publication growth rate), and academic level (annual average literature citation frequency), which can evaluate the "hotness" of the discipline keyword horizontally.

[0099] Let the number of discipline keywords be K N The number of articles in each discipline keyword in the discipline is P i 0≤i<K N The total number of citations of the literature is C i The difference between the start year and the end year of the publication year of the literature is T i The number of articles published in the earliest publication year is P begin The number of articles published in the latest publication year is P end Then:

[0100] The annual average literature citation frequency of the discipline keyword i CPT i is:

[0101]

[0102] The publication growth rate of the discipline keyword i is:

[0103]

[0104] In particular, T i= max j＝1,P Y ij - min j＝1,P Y ij , where Y ij is the publication year of the ith keyword in the subject j, max j＝1,P Y ij is the maximum publication year, and min j＝1,P Y ij is the minimum publication year.

[0105] In step 206, any one of the total number of documents, the annual average document citation frequency, and the publication growth rate is selected as a target hotspot indicator, and the hotspot indicator values of the keywords corresponding to the subject are obtained for the target hotspot indicator.

[0106] In step 207, any keyword is selected as a target keyword in the hotspot indicator values, and a first difference between the hotspot indicator value of the target keyword and the minimum hotspot indicator value and a second difference between the maximum hotspot indicator value and the minimum hotspot indicator value are calculated.

[0107] In step 208, a linear normalization value of the target hotspot indicator corresponding to the target keyword is obtained according to the ratio of the first difference to the second difference.

[0108] In step 209, the linear normalization values of the target hotspot indicators corresponding to the target keyword are summed to obtain the hotspot index of the target keyword in the subject.

[0109] Then, the total number of documents of each keyword in the subject, the average citation frequency of each keyword in the subject, and the publication growth rate of each keyword in the subject are normalized and summed to obtain the hotspot index of the keyword. The hotspot index of the keyword is obtained by linear normalization of each evaluation indicator. Since the frequency, citation frequency, and time distribution of the keyword in the literature are considered, the difference in keyword frequency between different subjects can be eliminated, and the hotspot index of the keyword reflecting the importance of the keyword in the subject can be obtained.

[0110] Specifically, any one of the total number of documents, the annual average document citation frequency, and the publication growth rate is selected as a target hotspot indicator.

[0111] When the target hotspot indicator is the total number of documents, the linear normalization value of the total number of documents of each keyword in the subject, i.e., the hotspot indicator value of the target hotspot indicator (total number of documents), is sigmoid(P i ), and:

[0112]

[0113] wherein 0≤i N , P i is the number of documents in the subject field (hotspot indicator value of the target subject keyword) of the ith subject keyword, min(P i ) is the minimum number of documents in the subject field (minimum hotspot indicator value) of each subject keyword, max(P i ) is the maximum number of documents in the subject field (maximum hotspot indicator value) of each subject keyword, P i -min(P i ) is the first difference value, and max(P i )-min(P i ) is the second difference value.

[0114] Similarly, when the target hotspot indicator is the annual average document citation frequency, let the linear normalized value corresponding to the annual average document citation frequency of each subject keyword in the subject field be sigmoid(CPT i ), then:

[0115]

[0116] wherein 0≤i N , CPT i is the annual average document citation frequency of the ith subject keyword in the subject field, min(CPT i ) is the minimum annual average document citation frequency of each subject keyword in the subject field, and max(CPT i ) is the maximum annual average document citation frequency of each subject keyword in the subject field.

[0117] When the target hotspot indicator is the publication growth rate, let the linear normalized value corresponding to the publication growth rate of each subject keyword in the subject field be sigmoid(PG i ), then:

[0118]

[0119] wherein 0≤i N , PG i is the publication growth rate of the ith keyword in the subject field, min(PG i ) is the minimum publication growth rate of each keyword in the subject field, and max(PG i ) is the maximum publication growth rate of each keyword in the subject field.

[0120] Let the hotspot index of each subject keyword in the subject field be S i , then:

[0121] S i = sigmoid(P i ) + sigmoid(CPT i ) + sigmoid(PG i ).

[0122] In particular, the annual average download frequency of the obtained download frequency can also be calculated based on the calculation method of the linear normalized value of the annual average document cited frequency, which is used to compare the reference value of the document, and then the linear normalized value of the annual average download frequency is added to the final calculation result of the hot spot index S i . In order to enrich the comparison index. Specifically, according to the annual average download frequency calculation formula, the frequency of each document downloaded in the difference between the start and end years of the publication year is calculated, wherein the aforementioned annual average download frequency calculation formula is:

[0123]

[0124] Correspondingly, DPT i is the annual average download frequency (download per paper per year) of the subject keyword i, D i is the total number of downloads of the literature corresponding to the subject keyword i within the difference between the start and end years of the publication year, P i is the total number of papers of the literature corresponding to the subject keyword i within the difference between the start and end years of the publication year, and T i is the difference between the start and end years of the publication year of the literature corresponding to the subject keyword i in the subject.

[0125] At this time, the hot spot index S i of the subject keyword i in the subject can be calculated by the following formula:

[0126] S i = sigmoid(P i ) + sigmoid(CPT i ) + sigmoid(PG i ) + sigmoid(DPT i ),

[0127] wherein,

[0128] By applying the technical solution of the embodiment, the original data of published literature is extracted from the literature database, a subject non-related vocabulary library is established, non-academic related vocabulary, generalized words, common words, stop words, etc. are removed, subject keywords are obtained, and each evaluation index of the subject keywords is respectively counted, including: calculating the total number of papers (P, papers), the citation per paper per year (CPT, citation per paper per year), and the paper growth (PG, paper growth). At the same time, the linear normalization formula is used to comprehensively calculate each evaluation index, so as to finally obtain the hot spot index of the subject keywords. By comparing the numerical values of the hot spot indexes, scholars can quickly compare the research fields and the degree of attention to the problems, master the academic dynamics and trends, promote academic exchanges and innovation, and comprehensively consider the factors such as the occurrence frequency of keywords in literature, the citation frequency, the time distribution, and the growth rate.

[0129] Further, as a refinement and expansion of the above embodiment, in order to completely describe the specific implementation process of the embodiment, another method for calculating the subject hot spot index is provided, as shown in Figure 3 The method comprises:

[0130] Step 301: obtaining a literature sample, a publication journal name of the literature sample, a two-level subject specialty to which the literature sample belongs, and a one-level subject category corresponding to the two-level subject specialty.

[0131] Step 302: for the literature samples with the same two-level subject specialty, after the publication journal names corresponding to the literature samples are processed for deduplication, a subject-related journal information library is established according to the deduplicated publication journal names, the two-level subject specialty, and the one-level subject category corresponding to the two-level subject specialty, wherein the subject-related journal information library comprises subject-related journal names corresponding to different two-level subject specialties under different one-level subject categories.

[0132] In the above embodiment of the application, the literature sample, the publication journal name of the literature sample, the two-level subject specialty to which the literature sample belongs, and the one-level subject category corresponding to the two-level subject specialty are obtained. For example, the two-level subject specialty is “visual communication”, and the one-level subject category corresponding to the two-level subject specialty is “art”. By obtaining the subject to which the literature belongs (two-level specialty and one-level subject category) and the publication journal name of the literature, a subject-related journal information library can be established based on the publication journals corresponding to the subject. By querying the journal information in the subject-related journal information library, non-subject-related journals can be screened.

[0133] The method for establishing the subject-related journal information base comprises the following steps: (1) collecting the literature samples of the same secondary subject specialty; (2) removing the duplicate journal names corresponding to the literature samples; (3) classifying the journal names according to the secondary subject specialty and the primary subject category; and (4) obtaining the subject-related journal information base containing the subject-related journal names corresponding to the different secondary subject specialties under the different primary subject categories.

[0134] Step 303: obtaining the subject, keywords, publication year and citation times of the literature, and obtaining the publication journal names of the literature belonging to the same subject, wherein the subject comprises a secondary subject specialty and a primary subject category, and the primary subject category corresponds to a plurality of secondary subject specialties.

[0135] Step 304: performing journal name matching on the publication journal names of the literature based on the subject-related journal information base, and obtaining the initial subject keywords according to the keywords of the literature matched successfully.

[0136] Step 305: performing vocabulary matching on the initial subject keywords based on the preset subject-unrelated vocabulary base, and removing the initial subject keywords matched successfully, and obtaining the subject keywords based on the non-repeated initial subject keywords after removal.

[0137] Then, the subject, keywords, publication year and citation times of the literature are obtained, the publication journal names of the literature belonging to the same subject are obtained, the journal name matching is performed on the publication journal names of the literature based on the subject-related journal information base, the initial subject keywords are obtained according to the keywords of the literature matched successfully, the vocabulary matching is performed on the initial subject keywords based on the preset subject-unrelated vocabulary base, the initial subject keywords matched successfully are removed, and the subject keywords are obtained based on the non-repeated initial subject keywords after removal. The subject keywords obtained by filtering the non-subject-related journals and then filtering the non-subject-related vocabulary further improve the accuracy of the “original data” and have more reference value for evaluating the hot index.

[0138] Step 306: for the literature corresponding to the same subject keyword, the annual article average literature cited frequency is calculated according to the start and end year difference of the publication year of the literature, the total number of articles and the total citation times, and the publication growth rate is calculated according to the number of articles published in the earliest publication year and the latest publication year.

[0139] Step 307, for the plurality of subject keywords corresponding to the subject, according to the total number of literature corresponding to the subject keyword, the annual average literature cited frequency and the publication growth rate, the hot spot index of any subject keyword in the subject is calculated.

[0140] Then, for the literature corresponding to the same subject keyword, the annual average literature cited frequency is calculated according to the start and end year difference of the publication year corresponding to the literature, the total number of literature and the total number of citations, and the publication growth rate is calculated according to the number of literature published in the earliest publication year and the latest publication year, and for the plurality of subject keywords corresponding to the subject, according to the total number of literature corresponding to the subject keyword, the annual average literature cited frequency and the publication growth rate, the hot spot index of any subject keyword in the subject is calculated.

[0141] The current way of directly counting a single indicator to evaluate the subject hot spot has its one-sidedness: first, the target keyword and the quality of the related literature are not guaranteed, and non-academic words may be mixed, and the subject classification of the literature is not clear enough, so the obtained indicators are not accurate and comparable horizontally within and between subjects; second, the cited amount, the download amount and even the journal impact factor are all indicators that need time accumulation to reflect the level, so they cannot reflect the incremental mutation characteristics and volume characteristics of research hot spots.

[0142] By applying the technical solution of the embodiment, based on the subject related journal information database, non-subject related journals are filtered, and at the same time, based on the subject non-related vocabulary database, non-subject related words are filtered, ensuring the quality and relevance of the keywords. Then, the total number of literature, the annual average literature cited frequency and the publication growth rate are calculated to compound the indicators, which not only considers the academic level, but also measures the research increment mutation and research volume, and can directly compare the hot spot indicators horizontally.

[0143] In a specific embodiment, as shown in Figure 4 The original information of the literature is extracted from the literature database, including the subject, the keyword, the publication year and the number of citations, then the non-subject related words are filtered by using the subject non-related vocabulary database, and the non-professional related journal literature is filtered by using the subject related journal information database, to obtain the subject keyword, then the total number of literature of the subject keyword, the annual average literature cited frequency and the publication growth rate are calculated, and finally the hot spot index is calculated by using the linear normalization formula. By filtering the keywords and subject related literature journals, the data quality is improved, and the composite indicators solve the problem of considering academic level, research increment mutation and research volume. In addition, the hot spot index obtained by the normalization formula can be compared horizontally within and between subjects, which can comprehensively represent the keyword hotness, and assist in subject hot spot discovery and research trend analysis.

[0144] Further, asFigure 1 The embodiments of the present application provide a subject hot spot index calculation device, as shown in the specific implementation of the method. Figure 5 The device comprises:

[0145] A document information acquisition module 401 is configured to acquire the subject, keywords, publication year and citation times of a document;

[0146] A subject keyword filtering module 402 is configured to, for the documents belonging to the same subject, eliminate the non-subject related words in the keywords corresponding to the documents and then perform a deduplication process to obtain subject keywords;

[0147] A keyword index calculation module 403 is configured to, for the documents corresponding to the same subject keyword, calculate the annual average document citation frequency according to the start-end year difference of the publication year, the total number of documents and the total citation times of the documents, and calculate the publication growth rate according to the number of documents published in the earliest publication year and the latest publication year respectively;

[0148] A subject hot spot representation module 404 is configured to, for the plurality of subject keywords corresponding to a subject, calculate the hot spot index of any subject keyword in the subject according to the total number of documents, the annual average document citation frequency and the publication growth rate of the subject keyword.

[0149] Optionally, the keyword index calculation module 403 is further configured to:

[0150] determine the start-end year difference of the publication year between the earliest publication year and the latest publication year according to the publication year of the documents;

[0151] calculate the annual average document citation frequency within the start-end year difference of the publication year according to the annual average document citation frequency calculation formula, wherein the annual average document citation frequency calculation formula is:

[0152]

[0153] CPT i CPT is the annual average document citation frequency of the subject keyword i, C i CPT is the total citation times of the documents corresponding to the subject keyword i within the start-end year difference of the publication year, P i CPT is the total number of documents of the documents corresponding to the subject keyword i within the start-end year difference of the publication year, T i CPT is the start-end year difference of the publication year of the documents corresponding to the subject keyword i in the subject.

[0154] Optionally, the keyword index calculation module 403 is further configured to:

[0155] According to a publication growth rate calculation formula, a publication growth rate of literature published in the latest publication year compared to the number of articles published in the earliest publication year in the publication year difference is calculated, and the publication growth rate calculation formula is:

[0156]

[0157] PG i is the publication growth rate of the subject keyword i, P end is the number of articles published in the latest publication year in the literature corresponding to the subject keyword, P begin is the number of articles published in the earliest publication year in the literature corresponding to the subject keyword.

[0158] Optionally, the subject hotspot characterization module 404 is also used to:

[0159] Select any one of the total number of articles, the annual average article citation frequency, and the publication growth rate as a target hotspot indicator, and obtain the hotspot indicator value of each subject keyword for the target hotspot indicator for multiple subject keywords corresponding to the subject;

[0160] In the hotspot indicator value, select any subject keyword as a target subject keyword, calculate the first difference value between the hotspot indicator value of the target subject keyword and the minimum hotspot indicator value, and the second difference value between the maximum hotspot indicator value and the minimum hotspot indicator value, and obtain the linear normalization value of the target hotspot indicator corresponding to the target subject keyword according to the ratio of the first difference value to the second difference value;

[0161] Sum the linear normalization values of each target hotspot indicator corresponding to the target subject keyword to obtain the hotspot index of the target subject keyword in the subject.

[0162] Optionally, the subject keyword filtering module 402 is also used to:

[0163] Perform vocabulary matching on the keywords corresponding to the literature based on a preset subject non-relevant vocabulary library, and remove the matched keywords, and obtain subject keywords based on the non-repeated keywords after removal.

[0164] Optionally, the subject keyword filtering module 402 is also used to:

[0165] Obtain the publication journal name of the literature, perform journal name matching on the publication journal name corresponding to the literature based on a subject related journal information library, and obtain initial subject keywords according to the keywords corresponding to the literature matched successfully;

[0166] The initial subject key words are subjected to vocabulary matching based on a preset subject non-relevant vocabulary library, and the initial subject key words with successful matching are removed. The subject key words are obtained based on the non-repeated initial subject key words after removal.

[0167] Further, the embodiment of the present application provides another subject hot spot index calculation device, as shown in the figure, the device comprises: Figure 6

[0168] The document information acquisition module 501 is configured to acquire the subject, key words, publication year and citation times of the document;

[0169] The subject key word filtering module 502 is configured to remove the non-subject relevant vocabulary in the key words corresponding to the documents belonging to the same subject, and then perform deduplication processing to obtain the subject key words;

[0170] The key word index calculation module 503 is configured to calculate the annual average document citation frequency according to the start and end year difference of the publication year, the total number of documents and the total citation times of the documents corresponding to the same subject key word, and calculate the publication growth rate according to the number of documents published in the earliest publication year and the latest publication year respectively;

[0171] The subject hot spot representation module 504 is configured to calculate the hot spot index of any subject key word in the subject according to the total number of documents, the annual average document citation frequency and the publication growth rate of the subject key word corresponding to the subject key word.

[0172] The filtering vocabulary library establishment module 505 is configured to acquire the key word sample, the subject of the key word sample and the subject relevance identification condition corresponding to the key word sample, determine the non-subject relevant vocabulary according to the subject relevance identification result for the key word sample belonging to the same subject, and establish the subject non-relevant vocabulary library based on the non-subject relevant vocabulary.

[0173] Optionally, the key word index calculation module 502 is further configured to:

[0174] determine the start and end year difference of the publication year between the earliest publication year and the latest publication year according to the publication year of the document;

[0175] calculate the annual average document citation frequency within the start and end year difference of the publication year according to the annual average document citation frequency calculation formula, wherein the annual average document citation frequency calculation formula is:

[0176]

[0177] CPT i is the annual average document citation frequency of the subject key word i, and C​i is the total number of citations of the literature corresponding to the discipline keyword i within the range from the publication year to the end year, P i is the total number of articles of the literature corresponding to the discipline keyword i within the range from the publication year to the end year, T i is the range from the publication year to the end year of the literature corresponding to the discipline keyword i in the discipline.

[0178] Optionally, the keyword indicator calculation module 502 is further configured to:

[0179] According to the publication growth rate calculation formula, the publication growth rate of the literature published in the latest publication year compared with the number of articles of the literature published in the earliest publication year within the range from the publication year to the end year is calculated, wherein the publication growth rate calculation formula is:

[0180]

[0181] PG i is the publication growth rate of the discipline keyword i, P end is the number of articles of the literature published in the latest publication year in the literature corresponding to the discipline keyword, P begin is the number of articles of the literature published in the earliest publication year in the literature corresponding to the discipline keyword.

[0182] Optionally, the discipline hotspot characterization module 504 is further configured to:

[0183] Select any one of the total number of articles, the annual average article citation frequency, and the publication growth rate as a target hotspot indicator, and obtain the hotspot indicator value of each discipline keyword for the target hotspot indicator for the plurality of discipline keywords corresponding to the discipline;

[0184] In the hotspot indicator value, select any discipline keyword as a target discipline keyword, calculate the first difference value between the hotspot indicator value of the target discipline keyword and the minimum hotspot indicator value, and the second difference value between the maximum hotspot indicator value and the minimum hotspot indicator value, and obtain the linear normalization value of the target hotspot indicator corresponding to the target discipline keyword according to the ratio of the first difference value to the second difference value;

[0185] Sum the linear normalization values of each target hotspot indicator corresponding to the target discipline keyword to obtain the hotspot index of the target discipline keyword in the discipline.

[0186] Optionally, the discipline keyword filtering module 502 is further configured to:

[0187] Perform vocabulary matching on the keywords corresponding to the literature based on a preset discipline irrelevant vocabulary library, and remove the matched keywords, and obtain the discipline keyword based on the non-repeated keywords after removal.

[0188] Optionally, the discipline includes a secondary discipline major and a primary discipline category, the primary discipline category corresponds to a plurality of secondary discipline majors, and the filter library establishing module 505 is further configured to:

[0189] obtain literature samples, publication journal names of the literature samples, secondary discipline majors to which the literature samples belong, and primary discipline categories corresponding to the secondary discipline majors;

[0190] For the literature samples with the same secondary discipline major, after the publication journal names corresponding to the literature samples are processed for deduplication, a discipline-related journal information library is established according to each publication journal name after the deduplication processing, the secondary discipline major, and the primary discipline category corresponding to the secondary discipline major, wherein the discipline-related journal information library includes discipline-related journal names corresponding to different secondary discipline majors under different primary discipline categories.

[0191] Optionally, the discipline keyword filtering module 502 is further configured to:

[0192] obtain publication journal names of the literature, perform journal name matching on the publication journal names corresponding to the literature based on the discipline-related journal information library, obtain initial discipline keywords according to keywords corresponding to the literature for which the matching is successful, and

[0193] perform vocabulary matching on the initial discipline keywords based on a preset discipline non-related vocabulary library, and remove the initial discipline keywords for which the matching is successful, obtain discipline keywords based on the non-repeated initial discipline keywords after the removal.

[0194] It should be noted that other corresponding descriptions of each functional unit involved in the discipline hotspot index calculation device provided in the embodiments of the present application can be referred to the corresponding descriptions in the method, which will not be described here. Figures 1 to 3

[0195] Based on the above method as shown in Figures 1 to 3 , correspondingly, the embodiments of the present application further provide a storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the discipline hotspot index calculation method as shown in Figures 1 to 3

[0196] Based on such understanding, the technical solutions 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, a U disk, a mobile hard disk, etc.) and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in various implementation scenarios of the present application.

[0197] Based on the above method as shown in Figures 1 to 3 , and​​Figure 5 and 6 In order to achieve the above object, the embodiments of the present application further provide a computer device, which can be a personal computer, a server, a network device, etc., comprising a storage medium and a processor; the storage medium is used for storing a computer program; the processor is used for executing the computer program to realize the above-mentioned subject hot index calculation method as shown in Figures 1 to 3

[0198] Optionally, the computer device can further comprise 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 can comprise a display, an input unit such as a keyboard, etc. The optional user interface can further comprise a USB interface, a card reader interface, etc. The network interface can optionally comprise a standard wired interface, a wireless interface (such as a Bluetooth interface, a WI-FI interface), etc.

[0199] Those skilled in the art can understand that the structure of the computer device provided by the embodiments of the present application does not constitute a limitation on the computer device, and the computer device can comprise more or fewer components, or combine certain components, or different component arrangements.

[0200] The storage medium can further comprise an operating system and a network communication module. The operating system is a program for managing and saving computer device hardware and software resources, and supports the running of information processing programs and other software and / or programs. The network communication module is used for realizing the communication between the components in the storage medium, and the communication with other hardware and software in the entity device.

[0201] ​Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be realized by means of software and necessary general hardware platforms, or by hardware. The subject of the literature, the keywords, the publication year, and the number of citations are obtained; for the literature belonging to the same discipline, the non-discipline-related keywords in the literature are removed, and the de-duplication processing is performed to obtain the discipline keywords; for the literature corresponding to the same discipline keyword, the annual article average literature cited frequency is calculated according to the difference between the publication year and the end year, the total number of articles, and the total number of citations; and the publication growth rate is calculated according to the number of articles published in the earliest and latest publication years; for the multiple discipline keywords corresponding to the discipline, the hot spot index of any discipline keyword in the discipline is calculated according to the total number of articles, the annual article average literature cited frequency, and the publication growth rate of the discipline keyword. By filtering the keywords, the data quality is improved, and by calculating the total number of articles, the annual article average literature cited frequency, and the publication growth rate, the composite calculation index is calculated, which solves the problem of considering academic level, research increment mutation, and research volume. The evaluated discipline hot spot keywords can assist in discipline hot spot discovery and research trend analysis.

[0202] Those skilled in the art can understand that the drawings are only schematic diagrams of preferred implementation scenarios, and the modules or processes in the drawings are not necessarily required for implementing the present application. Those skilled in the art can understand that the modules in the devices in the implementation scenarios can be distributed in the devices in the implementation scenarios according to the description of the implementation scenarios, or can be changed and located in one or more devices different from the implementation scenarios. The modules of the above implementation scenarios can be combined into one module, or can be further split into multiple sub-modules.

[0203] The above application numbers 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 the present application, but the present application is not limited thereto, and any changes that can be thought of by those skilled in the art should fall within the protection scope of the present application.

Claims

1. A method for calculating a subject hotspot index, characterized in that, The method includes: Obtain the subject matter, keywords, publication year, and citation count of the literature; For literature belonging to the same discipline, after removing non-discipline-related words from the keywords corresponding to the literature, deduplication is performed to obtain discipline keywords; For literature corresponding to the same subject keywords, the average annual citation frequency is calculated based on the difference between the start and end years of publication, the total number of articles, and the total number of citations. The publication growth rate is calculated based on the number of articles published in the earliest and latest years, respectively. For multiple subject keywords corresponding to a subject, the hotspot index of any subject keyword in the subject is calculated based on the total number of documents corresponding to the subject keyword, the average annual citation frequency of documents, and the publication growth rate.

2. The method according to claim 1, characterized in that, The calculation of the average annual citation frequency per document, based on the year difference between the publication year and the publication date of the document, the total number of documents, and the total number of citations, includes: Based on the publication year of the document, determine the year difference between the earliest and latest publication years; The average annual citation frequency of documents is calculated according to the formula for calculating the average annual citation frequency of documents within the year difference of the publication year. The formula for calculating the average annual citation frequency of documents is as follows: CPT i C represents the average annual citation frequency of documents for subject keyword i. i P represents the total number of citations of the literature corresponding to subject keyword i within the year difference from the publication year. i T represents the total number of documents corresponding to subject keyword i within the year difference between the publication year and the publication date. i This represents the difference between the start and end years of publication of the literature corresponding to the subject keyword i within that subject.

3. The method according to claim 1, characterized in that, The calculation of the publication growth rate based on the number of articles published in the earliest and latest years, respectively, includes: According to the publication growth rate calculation formula, the publication growth rate is calculated by comparing the number of articles published in the latest publication year with the number of articles published in the earliest publication year within the specified year-end difference. The publication growth rate calculation formula is as follows: PG i P represents the publication growth rate of subject keyword i. end P represents the number of documents published in the latest year among those corresponding to the subject keywords. begin T represents the number of articles published in the earliest year among the literature corresponding to the subject keywords. i This represents the difference between the start and end years of publication of the literature corresponding to the subject keyword i within that subject.

4. The method according to claim 1, characterized in that, For each subject-specific keyword, based on the total number of articles corresponding to that keyword, the average annual citation frequency, and the publication growth rate, a hotspot index for each keyword within that subject-specific subject is calculated, including: Select any one of the following as the target hot spot indicator: total number of articles, average annual citation frequency per article, and publication growth rate. For multiple subject keywords corresponding to the subject, obtain the hot spot indicator value of each subject keyword for the target hot spot indicator. Among the hot spot index values, any subject keyword is selected as the target subject keyword. The first difference between the hot spot index value of the target subject keyword and the minimum hot spot index value, and the second difference between the maximum hot spot index value and the minimum hot spot index value are calculated. Based on the ratio of the first difference to the second difference, the linear normalized value of the target hot spot index corresponding to the target subject keyword is obtained. The linear normalized values ​​of each target hotspot index corresponding to the target subject keyword are summed to obtain the hotspot index of the target subject keyword in the subject.

5. The method according to any one of claims 1 to 4, characterized in that, After removing non-discipline-related words from the keywords corresponding to the literature, deduplication is performed to obtain subject-specific keywords, including: Based on a pre-defined subject-related vocabulary database, the keywords corresponding to the document are matched, and the successfully matched keywords are removed. Based on the removed non-repeating keywords, subject keywords are obtained.

6. The method according to claim 5, characterized in that, The method for establishing the subject-irrelevant vocabulary database is as follows: Obtain keyword samples, the disciplines to which the keyword samples belong, and the discipline relevance assessment of the keyword samples; For keyword samples belonging to the same discipline, keyword samples that are not related to the discipline based on the discipline relevance assessment results are identified as non-discipline-related words, and a discipline-related non-discipline-related word database is established based on these non-discipline-related words.

7. The method according to claim 6, characterized in that, The discipline includes second-level disciplines and first-level discipline categories, wherein the first-level discipline category corresponds to multiple second-level disciplines. The method further includes: Obtain the literature sample, the name of the journal in which the literature sample was published, the second-level discipline of the literature sample, and the first-level discipline category corresponding to the second-level discipline; For literature samples with the same second-level discipline, after deduplication of the journal names corresponding to the literature samples, a discipline-related journal information database is established based on the deduplicated journal names, the second-level discipline, and the first-level discipline category corresponding to the second-level discipline. The discipline-related journal information database includes the discipline-related journal names corresponding to different second-level disciplines under different first-level discipline categories. Accordingly, the process involves performing word matching on the keywords corresponding to the document based on a preset subject-relevant vocabulary database, removing successfully matched keywords, and obtaining subject-specific keywords based on the removed non-repeating keywords, including: Obtain the journal name of the published document, perform journal name matching on the journal name corresponding to the document based on the subject-related journal information database, and obtain the initial subject keywords based on the keywords corresponding to the successfully matched document; The initial subject keywords are matched based on a pre-defined non-relevant subject vocabulary database, and the successfully matched initial subject keywords are removed. Based on the removed non-repeating initial subject keywords, the subject keywords are obtained.

8. A device for calculating a subject hotspot index, characterized in that, The device includes: The document information retrieval module is used to retrieve the subject matter, keywords, publication year, and citation count of a document. The subject keyword filtering module is used to remove non-subject-related words from the keywords of documents belonging to the same subject and then perform deduplication to obtain subject keywords. The keyword index calculation module is used to calculate the average annual citation frequency of documents corresponding to the same subject keywords, based on the difference between the start and end years of publication, the total number of documents, and the total number of citations. It also calculates the publication growth rate based on the number of documents published in the earliest and latest publication years, respectively. The subject hotspot characterization module is used to calculate the hotspot index of any subject keyword in a subject based on the total number of documents corresponding to the subject keyword, the average annual citation frequency of documents, and the publication growth rate.

9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for calculating the subject hotspot index as described in any one of claims 1 to 7.

10. A computer 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 for calculating the subject hotspot index as described in any one of claims 1 to 7.

Images