Method for mining interdisciplinary degree of scholars based on academic conference data

Abstract

The invention discloses a method for mining the interdisciplinary degree of scholars based on academic conference data. The method comprises the following steps: determining an evaluation time period and an interdisciplinary degree index; the academic conference interdisciplinary degree index comprises the conference subject diversity of scholars, the subject difference between conferences, the subject diversity of papers and the subject difference between papers; acquiring academic conference data of the scholars, wherein the academic conference data of the scholars comprises conferences participated by the scholars in the evaluation time period and papers published on the related conferences; calculating each interdisciplinary degree index based on the academic conference data of the scholars to obtain an index value corresponding to each interdisciplinary degree index; performing equalization processing on each index value to obtain an interdisciplinary evaluation value; and determining the interdisciplinary degree of the scholar in the evaluation time period based on the interdisciplinary evaluation value. According to the method, the trend change of interdisciplinary scientific research of scholars can be tracked, so that the interdisciplinary level of current domestic academic scientific research is known.

Description

Technical Field

[0001] The present invention belongs to the field of interdisciplinary research, and in particular relates to a method for mining the interdisciplinary degree of scholars based on academic conference data. Background Art

[0002] Interdisciplinary phenomena are common in modern academic research. On the one hand, interdisciplinary research has become an inevitable trend because the development of a single discipline has entered a bottleneck period. On the other hand, national policy orientation requires strengthening the cross-disciplinary integration and promoting the development of new engineering and new science. An academic conference is a conference with the theme of promoting scientific development, academic exchanges, research projects and other academic topics. Scholars participate in various academic conferences to display or exchange academic achievements and progress. However, the existing technology lacks methods that can mine interdisciplinary information in academic conference data, and thus cannot provide support for the in-depth development of interdisciplinary research. Summary of the invention

[0003] The purpose of the present invention is to provide a method for mining the interdisciplinary degree of scholars based on academic conference data to solve the problems existing in the above-mentioned prior art.

[0004] To achieve the above object, the present invention provides a method for mining scholars' interdisciplinary degree based on academic conference data, comprising:

[0005] Determine the evaluation period and interdisciplinary indicators; the interdisciplinary indicators of academic conferences include the diversity of academic conference disciplines, the disciplinary differences between conferences, the diversity of academic papers, and the disciplinary differences between academic papers;

[0006] Obtaining academic conference data of scholars, wherein the academic conference data of scholars includes conferences attended by scholars during the evaluation period and papers published at relevant conferences;

[0007] Calculate each of the interdisciplinary degree indicators based on the academic conference data of the scholars to obtain the indicator value corresponding to each of the interdisciplinary degree indicators;

[0008] Each indicator value is averaged to obtain the interdisciplinary evaluation value;

[0009] The interdisciplinary degree of the scholar during the evaluation period is determined based on the interdisciplinary evaluation value.

[0010] Optionally, obtaining the academic conference data of scholars specifically includes:

[0011] Obtain the scholar's paper list and construct two sets of four-tuple information based on the paper list information: the first set is author, publication relationship, paper and time; the second set is conference, acceptance relationship, paper and time;

[0012] Based on the four-tuple information, a knowledge graph of conference data is constructed around scholars;

[0013] Conferences attended by scholars and papers published at relevant conferences during the evaluation period are extracted from the knowledge graph to obtain academic conference data of scholars.

[0014] Optionally, the process of calculating the index value of the conference subject diversity specifically includes:

[0015] Constructing a paper collection based on the papers corresponding to each conference in the academic conference data of the scholars, extracting keyword information of each paper in the paper collection, and constructing a keyword collection;

[0016] Inputting the keyword information in the keyword set into the keyword subject classification model for classification prediction to obtain subject information corresponding to each keyword information; wherein the keyword subject classification model is constructed based on the RNN recurrent neural network model;

[0017] Determine the subject distribution data of the conference based on the subject information corresponding to each keyword information;

[0018] Calculate the disciplinary diversity index of the corresponding conference based on the disciplinary distribution data combined with the Shannon-Wiener Index;

[0019] The disciplinary diversity index of academic conferences attended by scholars during the evaluation period is calculated based on the disciplinary diversity index of each conference.

[0020] Optionally, the training optimization process of the keyword subject classification model specifically includes:

[0021] Acquire training data, wherein the training data includes keyword training information and corresponding manually annotated subject classification labels;

[0022] Construct an initial keyword subject classification model,

[0023] Extracting a hyperparameter combination included in the initial keyword subject classification model;

[0024] Based on the training data, the initial keyword subject classification model corresponding to each of the hyperparameter combinations is trained, and the training results are evaluated, and the evaluation results are used as the evaluation scores corresponding to each of the hyperparameter combinations;

[0025] Generating a sparrow population for sparrows using each of the hyperparameter combinations;

[0026] Based on the evaluation scores corresponding to the hyperparameter combinations and in combination with the sparrow optimization algorithm, parameter optimization is performed in the sparrow population to obtain the optimal hyperparameter combination;

[0027] Constructing an optimized keyword subject classification model based on the optimal hyperparameter combination;

[0028] The keyword subject classification task is performed based on the optimized keyword subject classification model.

[0029] Optionally, the process of calculating the index value of the subject differences between the conferences specifically includes:

[0030] Obtain the subject distribution data corresponding to the papers included in each conference in the academic conference data of the scholar;

[0031] Calculate the differences in subject distribution between conferences based on the subject distribution data combined with KL divergence;

[0032] Computed disciplinary distribution heterogeneity measures the disciplinary heterogeneity between conferences that scholars attended during the evaluation period.

[0033] Optionally, the calculation process of the index value of the paper subject diversity specifically includes:

[0034] Obtain keyword information of papers published by scholars during the evaluation period;

[0035] Calculate the subject diversity index of the paper based on the keyword information combined with the Shannon-Wiener index;

[0036] The subject diversity index of the papers published by scholars at academic conferences is calculated based on the paper subject diversity index.

[0037] Optionally, the process of calculating the index value of the subject differences between the papers specifically includes:

[0038] Obtain the subject distribution data of each paper published by scholars during the evaluation period;

[0039] Based on the subject distribution data of each paper and the cosine similarity, the differences in the subject distribution of the papers are calculated;

[0040] Based on the differences in the subject distribution of papers, the subject difference index between papers published by scholars at academic conferences is calculated.

[0041] Optionally, the index values ​​are averaged, and the specific calculation formula is:

[0042]

[0043] Where H is the interdisciplinary evaluation value of scholars, and its value range is from 0 to 1. The closer it is to 1, the higher the degree of interdisciplinary scholars. i are the indicator values ​​corresponding to each interdisciplinary degree indicator.

[0044] The technical effects of the present invention are:

[0045] The present invention is based on the academic conference data information in which scholars participate, and evaluates the degree of interdisciplinarity of scholars within any specified time period. As academic research progresses, the degree of interdisciplinarity of scholars will show a constantly changing process. Therefore, the method adopted in this patent has a strong practical application value, and can track the trend changes of scholars' interdisciplinary scientific research, thereby understanding the current interdisciplinarity level of domestic academic research. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0047] Figure 1 The present invention is a flowchart of a method implementation in an embodiment of the present invention.

[0048] Figure 2 Schematic diagram of the RNN recurrent neural network model training process in an embodiment of the present invention;

[0049] Figure 3 4 is a flowchart of the hyperparameter optimization process in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] Various exemplary embodiments of the present invention will now be described in detail. This detailed description should not be considered as limiting the present invention, but should be understood as a more detailed description of certain aspects, features, and embodiments of the present invention.

[0051] It should be understood that the terms described in the present invention are only for describing special embodiments and are not intended to limit the present invention. In addition, for the numerical range in the present invention, it should be understood that each intermediate value between the upper and lower limits of the scope is also specifically disclosed. Each smaller range between the intermediate value in any stated value or stated range and any other stated value or intermediate value in the described range is also included in the present invention. The upper and lower limits of these smaller ranges can be independently included or excluded in the scope.

[0052] It will be apparent to those skilled in the art that various modifications and variations may be made to the specific embodiments of the present invention description without departing from the scope or spirit of the present invention. Other embodiments derived from the present invention description will be apparent to those skilled in the art. The present application description and examples are exemplary only.

[0053] The words “include,” “including,” “have,” “contain,” etc. used in this article are open-ended terms, meaning including but not limited to.

[0054] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0055] Embodiment 1

[0056] like Figure 1-Figure 3 As shown, this embodiment provides a method for mining the degree of interdisciplinarity of scholars based on academic conference data, including: determining an evaluation time period and an interdisciplinarity indicator; the academic conference interdisciplinarity indicator includes the conference disciplinary diversity of scholars, the disciplinary differences between conferences, the disciplinary diversity of papers, and the disciplinary differences between papers; obtaining the academic conference data of scholars, the academic conference data of scholars includes the conferences attended by scholars during the evaluation time period and the papers published at relevant conferences; calculating each of the interdisciplinarity indicators based on the academic conference data of scholars to obtain the indicator value corresponding to each of the interdisciplinarity indicators; averaging each indicator value to obtain an interdisciplinary evaluation value; determining the degree of interdisciplinarity of scholars during the evaluation time period based on the interdisciplinary evaluation value.

[0057] This embodiment proposes a method for studying the interdisciplinary level of scholars based on academic conference data mining, taking academic conferences as the research entry point, analyzing the interdisciplinary nature of academic conferences, and forming a comprehensive assessment of the interdisciplinary level of scholars by mining and analyzing the information of academic conferences attended by scholars.

[0058] This embodiment mainly includes the following steps: 1. Determine the interdisciplinary degree index of academic conferences; 2. Collect academic conference data of scholars; 3. Calculate the interdisciplinary degree index based on the academic conference data; 4. Evaluate the interdisciplinary degree of scholars.

[0059] Step 1: Since the academic research directions and topics of scholars will evolve and change over time, this embodiment focuses on the interdisciplinarity of scholars within a specific time period. Academic conferences are an important place for scholars to participate in academic discussions and publish academic achievements. Academic conferences are characterized by short cycles and high frequencies. Therefore, this embodiment evaluates the interdisciplinarity of scholars based on academic conference data.

[0060] The interdisciplinary degree indicators selected in this embodiment include the following:

[0061] 1. Diversity of disciplines in academic conferences attended by scholars 1 :Academic conferences include special academic conferences and comprehensive academic conferences. The positioning of academic conferences determines their interdisciplinary degree. Through data analysis of the disciplinary diversity of academic conferences attended by scholars, we can reflect the scholars' participation in interdisciplinary academic activities.

[0062] 2. Disciplinary differences among academic conferences attended by scholars2 :Scholars will attend multiple academic conferences within a period of time. There are differences in themes and directions between academic conferences. By analyzing the disciplinary differences between the academic conferences attended by scholars, we can reflect the differences in the disciplines in which scholars participate. The greater the difference, the higher the degree of interdisciplinaryness of the scholars.

[0063] 3. Diversity of disciplines of papers published by scholars at academic conferences 3 : Papers are the results of scholars' academic research. The disciplinary diversity of conference papers published by the institute directly reflects the scholars' interdisciplinary level and whether scientific research results have been formed in interdisciplinary research, not just staying in the research stage, but having a certain research depth.

[0064] 4. Disciplinary differences among papers published by scholars at academic conferences 4 :It is not comprehensive enough to only analyze the disciplinary diversity of the papers published at the conference. This indicator conducts a horizontal comparative analysis of the disciplinary differences between the papers published by scholars, which can obtain the breadth of scholars' interdisciplinary knowledge.

[0065] Step 2: This embodiment searches for the scholar's name in a search engine, enters the scholar's homepage, obtains the scholar's paper list, and extracts information such as the paper author, paper name, conference name, publication time, and paper URL according to the paper list format, thereby forming two groups of four-tuple information: the first group is the author, publication relationship, paper and time, and the second group is the conference, acceptance relationship, paper and time. On this basis, a knowledge graph of conference data is constructed around the scholar.

[0066] By specifying a time period, conferences attended by scholars within the corresponding time period and papers published at relevant conferences are extracted from the constructed knowledge graph.

[0067] Step 3: After obtaining the conferences that scholars attend within a specified time period and the papers published at the conferences, this embodiment calculates the indicators based on this.

[0068] 1. Diversity of disciplines in academic conferences attended by scholars 1 :

[0069] This embodiment calculates the diversity of academic conferences by statistically analyzing the conference papers received in the past three years. First, the conference website is found according to the conference name, and the papers included in the conference in the past three years are obtained to form a paper set. Keyword information is extracted from all the papers in the set to form a keyword set, as follows:

[0070] W i ={w i1 ,w i2 ,w i3 …w ij …win}

[0071] Among them, W i is the keyword distribution of the i-th conference, w ij is the jth keyword of the i-th meeting.

[0072] K i ={k i1 ,k i2 ,k i3 …k ij …k in}

[0073] Among them, K i is the distribution of the number of keywords in the i-th conference, k ij is the number of the jth keyword in the i-th meeting.

[0074] In this embodiment, the training data set is formed by manually annotating the disciplines to which the keywords belong. in, For keyword w ij A word vector of length R, y ij For the corresponding subject, the RNN recurrent neural network model is trained through the training data to obtain the keyword subject classification model. The input is the jth keyword word vector, and the output is a probability vector y with a length of m after the sofmax function. ij , m is the number of subjects, and the subject to which keyword j belongs is:

[0075] s ij =findSubject(S,y ij )

[0076] Among them, findSubject is based on the subject library S and the probability vector y ij Obtain subject information.

[0077] The training and optimization process of the keyword subject classification model includes:

[0078] Construct an initial keyword subject classification model,

[0079] Extracting a hyperparameter combination included in the initial keyword subject classification model;

[0080] The training data set is divided into training set, test set and validation set according to the preset ratio; the training set is used for model training and learning, the test set is used to finally evaluate the generalization ability of the model, and the validation set is used to adjust and optimize the model parameters during the training process.

[0081] Based on the training set, the keyword subject classification model corresponding to each of the hyperparameter combinations is trained, and the performance of the keyword subject classification model corresponding to each of the hyperparameter combinations is verified and evaluated through the verification set, and the evaluation result is used as the evaluation score corresponding to each of the hyperparameter combinations;

[0082] Loss function calculation: The cross-entropy loss function is used. Cross-entropy is a measurement method for evaluating the difference between two probability distributions. The performance of model training is evaluated by comparing the difference between the probability distribution predicted by the model and the true probability distribution.

[0083] Hyperparameter optimization:

[0084] Setting the value range, population size and number of iterations of the hyperparameter combination, and using each of the hyperparameter combinations as sparrows to generate a sparrow population;

[0085] Based on the sparrow optimization algorithm, each hyperparameter combination in the sparrow population is iteratively updated multiple times. In each iterative update process, the evaluation score corresponding to each hyperparameter combination is calculated, and the sparrow population is updated based on the evaluation score corresponding to each hyperparameter combination until the number of iterations reaches a preset value and the optimal hyperparameter combination is output.

[0086] The Sparrow Search Algorithm (SSA) is an optimization algorithm inspired by the behavior of sparrow groups in nature. It finds solutions to optimization problems by simulating the foraging behavior and escaping from predators of sparrows.

[0087] Construct an optimized keyword subject classification model based on the optimal hyperparameter combination;

[0088] The keyword subject classification task is performed based on the optimized keyword subject classification model.

[0089] By analyzing all the keywords of the conference, we can get the subject distribution of the conference:

[0090] S i ={s i1 ,s i2 …,s im}

[0091] Among them, S i is the subject distribution of the i-th conference, and m is the total number of subjects.

[0092] Q i = {q i1 ,q i2 …,q im}

[0093] Among them, Qi The number of disciplines in the i-th conference, q i2 For subjects i2 The number of

[0094] After obtaining the subject distribution of the conference, this embodiment uses the Shannon-Wiener index to calculate the subject diversity of the conference, as follows:

[0095]

[0096] Among them, H 1i is the disciplinary diversity index of the i-th conference, m is the total number of disciplines, and p ij is the proportion of the jth discipline in conference i, which is calculated as follows:

[0097]

[0098] After obtaining the disciplinary diversity index of the i-th conference, the disciplinary diversity of the conferences that scholars attended during this period is calculated as follows:

[0099]

[0100] Among them, H 1 is the conference disciplinary diversity indicator, H 1i is the disciplinary diversity index of the i-th conference, C is the total number of conferences, and w i is the weight of the i-th meeting, calculated as:

[0101]

[0102] Among them, c i is the number of times the scholar attended the i-th conference.

[0103] 2. Disciplinary differences among academic conferences attended by scholars 2 :

[0104] By analyzing the papers collected in the conference in the past three years, we can get the following disciplinary distribution of each paper in the conference:

[0105] G i = {g i1 ,g i2 …,g iT}

[0106] Among them, G i represents the subject distribution of papers published in the i-th conference, g i2 represents the subject distribution of the second paper, and T is the total number of papers.

[0107] In order to compare the differences in subject distribution between conferences, this embodiment uses KL divergence to calculate the differences, and the calculation formula is as follows:

[0108]

[0109] Among them, f i , f t is the probability density of the subject distribution of conference i, t. In this embodiment, the kernel density function is used to estimate the probability density function, where f i The calculation formula is as follows:

[0110]

[0111] Among them, f i represents the probability density function estimate, h is the bandwidth, h d is the volume of a multidimensional cube with dimension d, K is the kernel density function, and its formula is as follows:

[0112]

[0113] Since the KL divergence ranges from 0 to infinity, it needs to be normalized. The processing formula is as follows:

[0114]

[0115] After calculating the differences in the conference discipline distribution, the overall differences are calculated as follows:

[0116]

[0117] Among them, H 2 is the conference discipline difference index, D it is the disciplinary difference index between the i-th conference and the t-th conference, C is the total number of conferences, and w it is the weight, calculated as:

[0118]

[0119] Among them, c i ,c t is the number of times the scholar attended the i-th and t-th conferences. The more times the scholar attends the conferences, the greater the weight.

[0120] 3. Diversity of disciplines of papers published by scholars at academic conferences 3

[0121] In the selected time period, obtain the papers published by scholars at various conferences, obtain the keyword information of the papers, and obtain the subject distribution of each paper through the keyword information, as follows: B k = {b k1 ,bk2 …,b km}

[0122] Among them, B k is the subject distribution of the kth paper, and m is the total number of subjects.

[0123] G k = {g k1 ,g k2 …,g km}

[0124] Among them, G k The number of disciplines of the kth paper, g k2 For subject b k2 The number of

[0125] After obtaining the subject distribution of the papers, this embodiment uses the Shannon-Wiener index to calculate the subject diversity of the papers, as follows:

[0126]

[0127] Among them, H 3k is the subject diversity index of the kth paper, m is the total number of subjects, d kj is the proportion of the jth discipline in paper k, which is calculated as follows:

[0128]

[0129] After obtaining the disciplinary diversity index of the kth paper, the disciplinary diversity of the papers published by scholars during this period is calculated as follows:

[0130]

[0131] Among them, H 3 is the subject diversity index of papers, H 3k is the subject diversity index of the kth paper, and M is the total number of papers

[0132] 4. Disciplinary differences among papers published by scholars at academic conferences 4 :

[0133] By analyzing the papers published by scholars in the selected period, the subject distribution of each paper in the conference can be obtained as follows:

[0134] O k ={o k1 ,o k2 …,o km}

[0135] Among them, O k represents the subject distribution of the kth paper, o k2It represents the number of the second category disciplines, and m is the total number of disciplines.

[0136] In order to compare the differences in the subject distribution of papers, this embodiment calculates H based on cosine similarity. 4 , and its calculation formula is as follows:

[0137]

[0138] Among them, Cos(O j ,O k ) is the cosine similarity of the subject distribution between paper j and paper k, and M is the total number of documents.

[0139] Step 4: After the calculation of the indicators is completed, the degree of interdisciplinarity of the scholars in the selected period is evaluated. The evaluation method is to accumulate and average the individual indicators obtained in step 3, as follows:

[0140]

[0141] Among them, H is the evaluation of the scholar's interdisciplinarity, and its value ranges from 0 to 1. The closer it is to 1, the higher the scholar's interdisciplinarity.

[0142] This embodiment evaluates the interdisciplinarity of scholars within any specified time period based on the academic conference data information in which scholars participated. As academic research progresses, the interdisciplinarity of scholars will change continuously. Therefore, the method adopted in this embodiment has a strong practical application value and can track the trend changes of scholars' interdisciplinary scientific research, thereby understanding the current interdisciplinarity level of domestic academic research.

[0143] The above is only a preferred specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for mining scholars' interdisciplinary degree based on academic conference data, characterized by: include: Determine the evaluation period and interdisciplinary indicators; the interdisciplinary indicators of academic conferences include the diversity of academic conference disciplines, the disciplinary differences between conferences, the diversity of academic papers, and the disciplinary differences between academic papers; Obtaining academic conference data of scholars, wherein the academic conference data of scholars includes conferences attended by scholars during the evaluation period and papers published at relevant conferences; Calculate each of the interdisciplinary degree indicators based on the academic conference data of the scholars to obtain the indicator value corresponding to each of the interdisciplinary degree indicators; Each indicator value is averaged to obtain the interdisciplinary evaluation value; The interdisciplinary degree of the scholar during the evaluation period is determined based on the interdisciplinary evaluation value.

2. The method for mining scholars' interdisciplinary degree based on academic conference data according to claim 1 is characterized by: The acquisition of scholar academic conference data specifically includes: Obtain the scholar's paper list and construct two sets of four-tuple information based on the paper list information: the first set is author, publication relationship, paper and time; the second set is conference, acceptance relationship, paper and time; Based on the four-tuple information, a knowledge graph of conference data is constructed around scholars; Conferences attended by scholars during the evaluation period and papers published at relevant conferences are extracted from the knowledge graph to obtain academic conference data of scholars.

3. The method for mining scholars' interdisciplinary degree based on academic conference data according to claim 1 is characterized by: The process of calculating the index value of the conference discipline diversity specifically includes: Constructing a paper collection based on the papers corresponding to each conference in the academic conference data of the scholars, extracting keyword information of each paper in the paper collection, and constructing a keyword collection; Inputting the keyword information in the keyword set into the keyword subject classification model for classification prediction to obtain subject information corresponding to each keyword information; wherein the keyword subject classification model is constructed based on the RNN recurrent neural network model; Determine the subject distribution data of the conference based on the subject information corresponding to each keyword information; Calculate the disciplinary diversity index of the corresponding conference based on the disciplinary distribution data combined with the Shannon-Wiener Index; The disciplinary diversity index of academic conferences attended by scholars during the evaluation period is calculated based on the disciplinary diversity index of each conference.

4. The method for mining scholars' interdisciplinary degree based on academic conference data according to claim 3 is characterized in that: The training optimization process of the keyword subject classification model specifically includes: Acquire training data, wherein the training data includes keyword training information and corresponding manually annotated subject classification labels; Construct an initial keyword subject classification model, Extracting a hyperparameter combination included in the initial keyword subject classification model; Based on the training data, the initial keyword subject classification model corresponding to each of the hyperparameter combinations is trained, and the training results are evaluated, and the evaluation results are used as the evaluation scores corresponding to each of the hyperparameter combinations; Generating a sparrow population for sparrows using each of the hyperparameter combinations; Based on the evaluation scores corresponding to the hyperparameter combinations and in combination with the sparrow optimization algorithm, parameter optimization is performed in the sparrow population to obtain the optimal hyperparameter combination; Constructing an optimized keyword subject classification model based on the optimal hyperparameter combination; The keyword subject classification task is performed based on the optimized keyword subject classification model.

5. The method for mining scholars' interdisciplinary degree based on academic conference data according to claim 1 is characterized in that: The process of calculating the index value of the subject differences between the conferences specifically includes: Obtain the subject distribution data corresponding to the papers included in each conference in the academic conference data of the scholar; Calculate the differences in subject distribution between conferences based on the subject distribution data combined with KL divergence; Computed disciplinary distribution heterogeneity measures the disciplinary heterogeneity between conferences that scholars attended during the evaluation period.

6. The method for mining scholars' interdisciplinary degree based on academic conference data according to claim 1 is characterized in that: The calculation process of the index value of the paper subject diversity specifically includes: Obtain keyword information of papers published by scholars during the evaluation period; Calculate the subject diversity index of the paper based on the keyword information combined with the Shannon-Wiener index; The subject diversity index of the papers published by scholars at academic conferences is calculated based on the paper subject diversity index.

7. The method for mining scholars' interdisciplinary degree based on academic conference data according to claim 1 is characterized by: The calculation process of the index value of the subject differences between the papers specifically includes: Obtain the subject distribution data of each paper published by scholars during the evaluation period; Based on the subject distribution data of each paper and the cosine similarity, the differences in the subject distribution of the papers are calculated; Based on the differences in the subject distribution of papers, the subject difference index between papers published by scholars at academic conferences is calculated.

8. The method for mining scholars' interdisciplinary degree based on academic conference data according to claim 1 is characterized by: The specific calculation formula for performing mean value processing on each index value is as follows: Where H is the interdisciplinary evaluation value of scholars, and its value range is from 0 to 1. The closer it is to 1, the higher the degree of interdisciplinary scholars. i are the indicator values ​​corresponding to each interdisciplinary degree indicator.

Images