A multi-perspective scientific and technological data profiling system and method thereof

Abstract

The present invention discloses a multi-perspective scientific and technological data portrait system and its method. The scientific and technological indicator integration evaluation module constructs a scientific and technological indicator portrait to realize horizontal / vertical comparative analysis of scientific and technological competitiveness in multiple dimensions; the scientific and technological hot spot label annotation module constructs a scientific and technological hot spot portrait to display the scientific and technological hot spots in the future period of time in the form of a label word cloud; the scientific and technological map construction and analysis module constructs a scientific and technological map portrait to determine the potential relationships between scientific and technological entities, present the evolution of the entire life cycle of scientific and technological entities, and display the spatio-temporal associations between scientific and technological entities; the scientific and technological competition and cooperation situation perception module constructs a scientific and technological field portrait to monitor the latest trends in key scientific and technological fields and reveal the competition and cooperation situation of resources in key fields; the scientific and technological portrait multi-dimensional display module is used for the comprehensive display of the scientific and technological indicator portrait, the scientific and technological hot spot portrait, the scientific and technological map portrait, and the scientific and technological field portrait. The present invention realizes the perspective-based analysis of the scientific and technological overview and game research and judgment, and improves the efficiency of scientific and technological big data analysis.

Description

Technical Field

[0001] The present invention relates to the field of science and technology big data portraits, and specifically to a multi-perspective science and technology data portrait system and method thereof. Background Art

[0002] Science and technology data portraits are regarded as an effective means to integrate and analyze multi-source science and technology information, present the overall picture of science and technology strength, and provide valuable insights. By digitally analyzing the widely collected multi-source science and technology data, science and technology data portraits can achieve a comprehensive assessment and comparison of science and technology strength, revealing the advantages and disadvantages in different fields. At the same time, using tagging technology to gain insights into the characteristics of science and technology strength helps to discover important laws and trends hidden in the data. Currently, the existing publicly available science and technology indicator systems provide important data support and analysis perspectives for constructing science and technology data portraits. In the basic idea of constructing science and technology data portraits, it is necessary to review and comprehensively analyze the existing publicly available science and technology big data indicator systems, obtain relevant data through comprehensive analysis, and construct science and technology digital portraits that meet actual needs through tagging and selective screening of the data. However, the types of science and technology data are numerous, the sources are extensive, and the data volume is huge. How to effectively integrate big data processing, data mining, and artificial intelligence technologies to establish science and technology data portraits has become an important challenge. In particular, how to organically combine domain knowledge, data science technologies, and decision analysis methods to achieve cross-domain and cross-disciplinary cross-analysis and understanding of science and technology data from different perspectives has become an important challenge. To sum up, there is currently a lack of research on a multi-perspective science and technology data portrait method and system in the related field, that is, an integrated solution for multi-perspective science and technology data portrait modeling, analysis, and application based on the establishment of a complete picture of the data. Summary of the Invention

[0003] The purpose of the present invention is to propose a multi-perspective science and technology data portrait system and method thereof to depict the overall picture of science and technology strength from multiple perspectives and provide fine-grained science and technology value insight support for the field.

[0004] The technical solution for achieving the purpose of the present invention is as follows: A multi-perspective science and technology data portrait system includes a science and technology indicator integrated evaluation module, a science and technology hot spot label annotation module, a science and technology map construction and analysis module, a science and technology competition and cooperation situation perception module, and a science and technology portrait multi-dimensional display module, wherein:

[0005] The science and technology indicator integrated evaluation module is used to construct a science and technology indicator portrait. A three-level science and technology indicator system is established from the dimensions of science and technology input, output, and benefit. The relative importance of each level of indicators is determined by the "Delphi method", and geometric analysis is used for indicator quantification and ranking to achieve horizontal / vertical comparative analysis of multi-dimensional science and technology competitiveness.

[0006] The technology hot topic tag annotation module is used to construct a technology hot topic portrait. It extracts technology keywords from a large amount of technology text data in parallel through the TextRank algorithm, establishes a technology theme word library through the similarity mapping between technology keywords and the vector encoding of technology theme words, constructs a technology hot topic label prediction model and trains it based on the technology theme word library to display the technology hot topics in the future for a period of time in the form of a label word cloud.

[0007] The technology map construction and analysis module is used to construct a technology map portrait. By means of knowledge graph modeling and technology entity relationship extraction, the connections between technology entities are established. Through technology map association analysis, technology map time series analysis, and technology map space-time analysis, the potential relationships between technology entities are discovered, the evolution of the entire life cycle of technology entities is presented, and the space-time associations between technology entities are displayed.

[0008] The technology competition and cooperation situation awareness module is used to construct a portrait of the technology field. By analyzing and extracting the core view content from think tank texts and monitoring the latest trends in key technology fields, and by constructing a trade competition and cooperation situation network to calculate the import dependence and trade stability index, the competition and cooperation situation of resources in key fields is revealed.

[0009] The multi-dimensional display module of technology portraits provides a unified user portal for the comprehensive display of technology index portraits, technology hot topic portraits, technology map portraits, and technology field portraits.

[0010] Furthermore, the technology index integrated evaluation module includes a technology index evaluation system construction module, a multi-dimensional technology competitiveness index ranking module, and a technology index portrait construction module, where:

[0011] The technology index evaluation system construction module consists of five first-level indicators: the technology development environment index , the technology development resource index , the scientific and technological innovation output index , the technology and industrial security index , and the technology comprehensive contribution index ; The first-level indicators can be correspondingly decomposed into multiple second-level indicators, expressed as ; The second-level indicators consist of multiple directly quantifiable third-level indicators, expressed as ;

[0012] The index system is quantified by formula (1a);

[0013] (1a);

[0014] s.t.

[0015] (1b);

[0016] Among them, i, j, and k respectively represent the serial numbers of the first-level indicators, second-level indicators, and third-level indicators. and represent the maximum and minimum values of the third-level indicator ; and represent the weight coefficients of the corresponding third-level indicator and the second-level indicator ; represents the number of second-level indicators included in the first-level indicator , represents the number of third-level indicators included in the second-level indicator ;

[0017] The multi-dimensional science and technology competitiveness index ranking module realizes the ranking of the science and technology competitiveness index from 5 indicator dimensions: the science and technology development environment index, the science and technology development resource index, the scientific and technological innovation output index, the science and technology and industrial security index, and the science and technology comprehensive contribution index.

[0018] The science and technology indicator portrait construction module, from 5 indicator dimensions: the science and technology development environment index, the science and technology development resource index, the scientific and technological innovation output index, the science and technology and industrial security index, and the science and technology comprehensive contribution index, uses a method combining charts and numerical analysis to conduct horizontal comparisons of the rankings of the first-level evaluation indicators of different evaluation objects within the same time period, and conducts horizontal / vertical comparisons of different indicator dimensions of different evaluation objects at different times through the first-level indicators to "drill down" to the second-level and third-level indicators.

[0019] Furthermore, the solution process of the weight coefficient of the second-level indicator and the weight coefficient of the third-level indicator is as follows:

[0020] First, use the "Delphi method" to determine the relative importance between any two indicators at the corresponding level , respectively represent equally important, slightly important, important, relatively important, very important compared to ; then construct the relative importance comparison matrix , and use the geometric analysis method of formula (2) to calculate the indicator weights;

[0021] (2);

[0022] Among them, represents the weight coefficient of any indicator at each level, and D represents the indicator dimension at the corresponding level.

[0023] Furthermore, the technology indicator portrait construction module displays the ranking change trend of different indicator values of the same evaluation object over time through a line chart, displays the ranking comparison of different evaluation objects in the same indicator dimension through a radar chart, and displays the change situation of the evaluation object in different indicator dimensions in the recent several time periods through a heat map.

[0024] Furthermore, the technology hot topic label annotation module includes a technology theme thesaurus construction module, a technology hot topic label prediction module, and a technology hot topic word cloud annotation module, where:

[0025] The technology theme thesaurus construction module: First, taking papers, patents, think tanks, and reports as the technology text data sources, parallelly constructs a technology text keyword relationship graph G. The technology text keyword relationship graph G takes technology keywords as nodes and the co-occurrence relationship of technology keywords in the text as edges, where the co-occurrence relationship establishes a relationship edge only when the co-occurrence frequency of two technology keywords in the same text is greater than the threshold ; then uses the TextRank algorithm to rank the technology keywords in the technology text keyword relationship graph G to generate a set of (technology keyword, score); then, uses Word2Vec to perform word vector encoding on the technology keywords in the technology text keyword relationship graph G, calculates the similarity between the technology keyword and the technology theme word, and replaces the technology keyword with the technology theme word with the largest similarity to achieve standardization processing; finally, accumulates and aggregates the scores of different technology keywords with the same technology theme word, and takes the top K theme words with the highest scores as technology hot topic labels to form a technology theme thesaurus;

[0026] The technology hot topic label prediction module is used to predict the labels that may become the technology hot topics of the evaluation object in the future. The technology hot topic label prediction model is cascaded by three layers of LSTM networks, and finally adds a fully connected layer with 32 nodes with a SoftMax activation function. The input is the embedding encoding sequence of technology theme words in the previous m stages, and the output is the probability of technology theme words in the next stage; uses the technology theme thesaurus constructed by papers, patents, think tanks, and report technology texts to construct a training sample set {([embedding encoding sequence], technology theme word)}, and fine-tunes the network structure of the technology hot topic label prediction model in an incremental training manner, and optimizes the training loss of the sample through a discount coefficient to improve the model performance, where is the serial number of the time period to which the sample belongs, and Q is the number of time period divisions;

[0027] The technology hot topic word cloud annotation module adjusts the size and color of the labels according to the technology hot topic label probability values output by the technology hot topic label prediction model to form a cloud-shaped image with technology theme words as visual elements, showing the technology field hot topics of the evaluation object in the future time period.

[0028] Furthermore, the technology map construction and analysis module includes a technology map construction module and a technology knowledge association mining module, where:

[0029] The technology map construction module first integrates three types of unstructured technology text data, namely technology projects, technology bulletins, and technology funds. It uses the BERT pre-trained model to encode the input technology text, extracts the semantic information of the text, and generates a sequence of semantic feature vectors for each word , where P is the number of words in the text; then, it uses a bidirectional LSTM network to further process the output of the BERT pre-trained model , captures the context-dependent relationships between words, and obtains an enhanced sequence of semantic feature vectors , and, takes the output of the bidirectional LSTM network as the input of a conditional random field to identify the boundaries and types of technology entities in the text, and filters out technology entities and their attribute information according to the types, where technology entities include technology organizations, technology funds, technology projects, and technology talents; then, extracts technology entity relationships, and the technology entity relationships are determined by the co-occurrence relationships of technology entities in the text and the similarity of semantic feature vectors between technology entities, that is, if the co-occurrence frequency of two technology entities in the same context exceeds a threshold and the similarity of semantic feature vectors exceeds a threshold , then there is a relationship edge between the two technology entities; finally, stores the identified technology entities and technology entity relationships in a graph database or a relational database to construct a complete technology map;

[0030] The technology knowledge association mining module discovers potential relationships between technology entities and events from the technology map through three methods: technology map association analysis, technology map time series analysis, and technology map spatio-temporal analysis. Among them, technology map association analysis forms a subgraph of the object of interest by querying multi-hop relationships in the technology map, and obtains hidden association relationships between technology entities by viewing other entities directly and indirectly associated with the technology entity; technology map time series analysis dynamically presents through the existence time of technology entities and the start and end times of technology entity relationships, as well as the playback of the technology map state at any moment within the start and end time range, presenting the full life cycle evolution process of technology entities; technology map spatio-temporal analysis associates and displays the technology map to a three-dimensional earth according to the geographical location information of technology entities, showing the spatio-temporal association relationships between technology entities.

[0031] Furthermore, the technology competition and cooperation situation awareness module includes a key area trend monitoring module, a key area resource competition and cooperation situation analysis module, and a key area portrait construction module, where:

[0032] The Key Technology Field Trend Monitoring Module collects think tank reports in key fields to form a think tank text library, and uses open-source large language models to automatically extract the content framework, abstracts, and core viewpoints of think tank bulletins. It displays the regional distribution of key fields in the form of a map, and shows the core viewpoints of the latest think tank bulletins in key fields when the mouse hovers over them.

[0033] The Key Technology Field Resource Competition and Cooperation Situation Analysis Module constructs a trade competition and cooperation situation network based on resource reserves and production and trade exchanges. In the trade competition and cooperation situation network, it shows the degree of dependence on trade imports by calculating the import dependence, and measures the stability of trade by calculating the trade stability index, i.e., the network structure entropy, to analyze the trade competition and cooperation situation of the industrial chain and supply chain.

[0034] The Key Technology Field Portrait Construction Module analyzes the trade competition and cooperation situation network of important mineral resources involved in the integrated circuit and large-capacity battery industries, forms a resource portrait based on the production and reserve data of mineral resource demands, shows the distribution of mineral resources, and analyzes the potential competition and cooperation relationships of key minerals.

[0035] A multi-perspective method for scientific and technological data portrait, based on the above-mentioned multi-perspective scientific and technological data portrait system, realizes multi-perspective scientific and technological data portrait, specifically:

[0036] Use the Science and Technology Index Integrated Evaluation Module to construct a science and technology index portrait. Establish a three-level science and technology index system from the dimensions of science and technology input, output, and benefits. Determine the relative importance of each level of indicators through the "Delphi method", and use geometric analysis to quantify and rank the indicators to achieve horizontal / vertical comparative analysis of multi-dimensional science and technology competitiveness.

[0037] Use the Science and Technology Hotspot Tagging Module to construct a science and technology hotspot portrait. Parallelly extract science and technology keywords from scientific and technological text data through the TextRank algorithm, establish a science and technology thesaurus through the similarity mapping of science and technology keywords and science and technology theme word vector coding, construct a science and technology hotspot tag prediction model and train it based on the science and technology thesaurus sample set, and display the science and technology hotspots in the form of a tag word cloud in the future period.

[0038] Use the Science and Technology Atlas Construction and Analysis Module to construct a science and technology atlas portrait. Establish connections between scientific and technological entities through knowledge graph modeling and scientific and technological entity relationship extraction. Determine the potential relationships between scientific and technological entities through scientific and technological atlas association analysis, scientific and technological atlas time series analysis, and scientific and technological atlas spatio-temporal analysis, present the evolution of the entire life cycle of scientific and technological entities, and display the spatio-temporal associations between scientific and technological entities.

[0039] Construct a portrait of the technology field using the technology competition and cooperation situation awareness module, extract the core viewpoint content by analyzing think tank texts, monitor the latest trends in key technology fields, and calculate the import dependence and trade stability index by constructing a trade competition and cooperation situation network to reveal the competition and cooperation situation of resources in key fields;

[0040] Use the technology portrait multi-dimensional display module to unify the user portal for the comprehensive display of technology indicator portraits, technology hot spot portraits, technology map portraits, and technology field portraits.

[0041] A computer device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the multi-perspective technology data portrait method is implemented to achieve a multi-perspective technology data portrait.

[0042] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the multi-perspective technology data portrait method is implemented to achieve a multi-perspective technology data portrait.

[0043] Compared with the prior art, the significant advantages of the present invention are as follows: 1) A multi-perspective technology data portrait method is proposed, expanding the breadth and depth of the analysis of the technology development trend. 2) A technology data portrait value insight system is provided, realizing perspective-based technology overview analysis and game research and judgment from the application perspective, and improving the efficiency of technology big data analysis. 3) Based on the above effects, the present invention can achieve the rapid integration and portrait presentation of multi-modal technology big data in engineering applications, and has strong operability and practical value. Brief Description of the Drawings

[0044] Figure 1 It is a structural diagram of a multi-perspective technology data portrait system of the present invention.

[0045] Figure 2 It is a structural diagram of a technology indicator evaluation system.

[0046] Figure 3 It is a flow chart of technology indicator quantification.

[0047] Figure 4 It is a flow chart of constructing a technology thesaurus.

[0048] Figure 5 It is an architecture diagram of a technology hot spot prediction model.

[0049] Figure 6 It is a structural diagram of a technology map construction and analysis module.

[0050] Figure 7 It is a flow chart of a technology competition and cooperation situation awareness module. Detailed Embodiments

[0051] In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0052] By integrating technologies such as scientific and technological data source integration technology, scientific and technological hotspot annotation technology, scientific and technological map analysis technology, and scientific and technological competition and cooperation situation perception technology, the present invention realizes the construction and quantitative analysis of scientific and technological data portraits, provides valuable insights, and the overall architecture is as Figure 1 shown.

[0053] A multi-perspective scientific and technological data portrait system includes: a scientific and technological indicator integration and evaluation module, a scientific and technological hotspot label annotation module, a scientific and technological map construction and analysis module, a scientific and technological competition and cooperation situation perception module, and a scientific and technological portrait multi-dimensional display module. The composition and functions of each module will be described in detail below with reference to the attached Figures 2 - 7 drawings.

[0054] The scientific and technological indicator integration and evaluation module is used for the construction of scientific and technological indicator portraits. A three-level scientific and technological indicator system is established from the dimensions of scientific and technological input, output, and benefits. The relative importance of the indicators is determined through the "Delphi method", and geometric analysis is used to quantitatively rank the indicators at each level, realizing horizontal / vertical comparative analysis of scientific and technological competitiveness in different indicator dimensions. Refer to Figures 2 - 3 , and the specific implementation steps are as follows:

[0055] (1) Construction of the scientific and technological indicator evaluation system

[0056] Refer to Figure 2 , the three-level scientific and technological indicator system covers five first-level indicators including the scientific and technological development environment index , the scientific and technological development resource index , the scientific and technological innovation output index , the science and technology and industrial security index , and the scientific and technological comprehensive contribution index ; the first-level indicators can be correspondingly decomposed into multiple second-level indicators, expressed as ; the second-level indicators are composed of multiple directly quantifiable third-level indicators, expressed as , as shown in Table 1:

[0057] Table 1 Three-level scientific and technological indicator system table

[0058]

[0059] The scientific and technological index system determines the relative importance of indicators using the "Delphi method", constructs a comparison matrix of the relative importance of indicators, determines the weight coefficients of indicators at each level through geometric analysis, and calculates the indicator values at each level through reverse weighting of the third-level indicators → second-level indicators → first-level indicators, as shown in Equation (1a).

[0060] (1a)

[0061] s.t.

[0062] (1b)

[0063] Among them, i, j, and k respectively represent the serial numbers of the first-level indicators, second-level indicators, and third-level indicators, and represent the maximum and minimum values of the third-level indicator ; and represent the weight coefficients of the corresponding third-level indicator and the second-level indicator ; represents the number of second-level indicators contained in the first-level indicator , represents the number of third-level indicators contained in the second-level indicator .

[0064] The solution process of the weight coefficient of the second-level indicator and the weight coefficient of the third-level indicator is as follows: First, use the "Delphi method" to determine the relative importance between of any two indicators at each level , respectively represent that the indicator is equally important, slightly important, important, relatively important, very important compared to ; then construct a comparison matrix of the relative importance of indicators and calculate using the geometric analysis method of formula (2).

[0065] (2)

[0066] Among them, represents the weight coefficient of any indicator at each level, and D represents the indicator dimension of the corresponding level.

[0067] (2)Multidimensional scientific and technological competitiveness index ranking

[0068] The technological competitiveness index refers to the comprehensive strength of the evaluation object in the technological field, reflecting its levels in aspects such as technological innovation ability, technological achievement transformation ability, technological resource allocation efficiency, and the quality of the technological talent team. This system realizes the ranking of the technological competitiveness index from 5 index dimensions: the technological development environment index, the technological development resource index, the technological innovation output index, the technology and industry security index, and the comprehensive technological contribution index.

[0069] Refer to Figure 3 , the calculation and ranking of the technological competitiveness index include the following steps:

[0070] ① Set the set of objects to be evaluated , set the starting year of the evaluation and the ending year ; Initialize variables , , , to 0;

[0071] ② , take an evaluation object from C to start calculating the index value;

[0072] ③ Judge whether all evaluation objects in C have been traversed. If so, go to step ⑩; otherwise, go to step ④;

[0073] ④ Judge whether the index of the current evaluation object has been calculated. If so, go to step ③; otherwise, go to step ⑤;

[0074] ⑤ Judge , whether it is 0. If so, go to step ⑥; otherwise, go to step ⑦;

[0075] ⑥ Obtain the relative importance of the evaluation indicators by expert scoring, and use the geometric analysis method of formula (2) to determine the weights of the secondary indicators and tertiary indicators , ;

[0076] ⑦ Judge , whether it is 0. If so, go to step ⑧; otherwise, go to step ⑨;

[0077] ⑧ Traverse the set C and calculate , ;

[0078] ⑨ Use formula (1a) to calculate the value of the current evaluated primary indicator;

[0079] ⑩ Rank all evaluated objects according to the primary indicator to obtain the rankings of the evaluated objects in the 5 dimensions of technological development environment, technological development resources, technological innovation output, technology and industry security, and comprehensive technological contribution during the time period t;

[0080] ⑪ Judgment , if yes, go to step ⑬, otherwise go to step ⑫;

[0081] ⑫ , go to step ③;

[0082] ⑬ End the evaluation.

[0083] (3) Construction of science and technology index portrait

[0084] The construction of the science and technology index portrait is based on the three-level science and technology index system, and realizes the horizontal / vertical comparative analysis of the science and technology development status and trend from five index dimensions: the science and technology development environment index, the science and technology development resource index, the science and technology innovation output index, the science and technology and industrial security index, and the science and technology comprehensive contribution index, in the form of a combination of charts and numerical analysis.

[0085] Specifically, for the horizontal comparison of the science and technology development status and trend, a horizontal comparative analysis is carried out by ranking the science and technology development environment index, the science and technology development resource index, the science and technology innovation output index, the science and technology and industrial security index, and the science and technology comprehensive contribution index of different evaluation objects (i.e., the main bodies of science and technology evaluation) within the same time period;

[0086] For the vertical comparison of the science and technology development status and trend, a vertical comparative analysis is carried out by ranking the three-level, two-level, and one-level indicators and rankings of a specific evaluation object in different time periods, and the science and technology development status of the evaluation object in different periods is revealed by calculating the ranking growth rate of different indicators in different time periods.

[0087] Further, the ranking change trend of different index values of the same evaluation object over time is displayed through a line chart, the ranking comparison of different evaluation objects in the same index dimension is displayed through a radar chart, and the change of the evaluation object in the index dimensions such as the science and technology development environment, the science and technology development resources, and the science and technology innovation output in recent time periods is displayed through a heat map.

[0088] Further, the horizontal / vertical comparison of different index dimensions of different evaluation objects in different time periods is carried out by "drilling down" from the first-level indicators to the second-level indicators and the third-level indicators, revealing the advantages and disadvantages of different evaluation objects in different index dimensions in science and technology development.

[0089] The science and technology hot spot label annotation module is used to construct a science and technology hot spot portrait, extract science and technology keywords from a large amount of science and technology text data, and establish a science and technology thesaurus. Further, by constructing a science and technology hot spot label prediction model, the accurate prediction and trend analysis of the science and technology development direction are realized in the form of a label word cloud. Refer to Figures 4 - 5 , the specific implementation steps are as follows:

[0090] (1) Construction of science and technology thesaurus

[0091] Reference Figure 4 The construction of the scientific and technological thesaurus uses a large amount of scientific and technological text data from papers, patents, think tanks, and reports as data sources. By analyzing scientific and technological keywords from a large amount of scientific and technological text data, scoring the scientific and technological keywords, and mapping them to the scientific and technological thesaurus for standardization processing, the top K scientific and technological keywords with the cumulative score ranking are taken as the hot scientific and technological labels during the evaluation object time period. The specific implementation steps are as follows:

[0092] 1) Collect the scientific and technological text data during the evaluation time period of the object to be evaluated as the data source for scientific and technological hot spot analysis, and form a text set , where N represents the number of texts. For any parallelly perform word segmentation and stop word removal processing to form ;

[0093] 2) Randomly divide into M batches to form . The size of the 1st to M - 1th batches is , and the size of the Mth batch is ;

[0094] 3) For any , use scientific and technological keywords as nodes , and based on their co-occurrence relationship in the text as the edge R, construct a subgraph . A relationship edge is established only when the co-occurrence frequency of two scientific and technological keywords in the same text is greater than the threshold . Merge different subgraphs through the same nodes to obtain the scientific and technological text keyword relationship graph G. Use the TextRank algorithm to rank the scientific and technological keywords in the scientific and technological text keyword relationship graph G to form a (scientific and technological keyword, score) set , where is the score value of the TextRank for the scientific and technological keyword , and E is the number of scientific and technological keywords;

[0095] 4) For any scientific and technological keyword , use Word2Vec for word vector encoding (Word Embedding), and calculate the cosine similarity between the scientific and technological keyword and all scientific and technological thesaurus word vector encodings. Replace the scientific and technological keyword with the scientific and technological thesaurus word with the maximum similarity to achieve standardization processing. Accumulate the scores of different scientific and technological keywords for the same scientific and technological thesaurus word as the score of the scientific and technological thesaurus word to form a (scientific and technological thesaurus word, score) set , where H is the number of scientific and technological theme words. In particular, the scientific and technological theme word list is a full set of scientific and technological theme words constructed by domain experts and can be obtained open source.

[0096] 5) Sort the scientific and technological theme words from largest to smallest according to the score value, and select the top K scientific and technological theme words as the scientific and technological hot topic labels of the object to be evaluated within the time period. The scientific and technological theme word library is formed by calculating the scientific and technological hot topic labels in all historical time periods.

[0097] (2) Prediction of scientific and technological hot topic labels

[0098] Refer to Figure 5 , the scientific and technological hot topic label prediction model is cascaded by three layers of LSTM networks, and a fully connected layer with 32 nodes with a SoftMax activation function is added at the end to predict the probability of scientific and technological hot topic labels (scientific and technological theme words) in the future time period. The scientific and technological hot topic label prediction model uses the scientific and technological theme word library constructed from paper, patent, think tank and report text data to generate training samples , fine-tune the neural network structure parameters in an incremental training manner, and optimize the sample training loss through a discount coefficient to improve the model performance. The specific implementation steps are as follows:

[0099] 1) Organize the embedded encodings of historical scientific and technological theme words into a long sequence in chronological order of time periods;

[0100] 2) Use a sliding window with a length of m time periods to divide this long sequence, and the sliding window moves backward by one time period each time. The embedded encodings of scientific and technological theme words in m time periods within the sliding window form an embedded encoding sequence, and the scientific and technological theme word in the next time period immediately following the sliding window is used as the label of this subsequence, thus generating a training sample set with a quantity of , where Q represents the number of time period divisions;

[0101] 3) The scientific and technological hot topic label prediction model first uses the historical sample set for preliminary training. As time goes by, new samples are continuously generated. The model continuously fine-tunes the neural network structure parameters through incremental training. At the same time, a discount coefficient is used to scale the sample training loss to improve the role of new samples in model parameter adjustment, where is the serial number of the time period to which the sample belongs, and Q is the number of time period divisions;

[0102] 4) Use the embedded encoding sequence of scientific and technological theme words in the last time period to input the trained scientific and technological hot topic label prediction model, predict the probability of scientific and technological hot topic labels (scientific and technological theme words) in the future time period, and select the top K labels with the highest probability values as the scientific and technological hot topics of the evaluation object in the future time period.

[0103] (3)Technology Hotspot Word Cloud Annotation

[0104] The technology word cloud is distinguished by the size and color of the technology hotspot label probability value, showing the technology field hotspots of the evaluation object in the future time period. Specifically, according to the size of the technology hotspot label probability value output by the technology hotspot label prediction model and the adjustment of the label color, a cloud-shaped image with technology theme words as visual elements is formed.

[0105] The technology map construction and analysis module is used to construct a technology map portrait. By building a knowledge graph model and extracting technology entity relationships, potential connections between technology entities are established. Through technology map association analysis, technology map time series analysis, and technology map spatio-temporal analysis, hidden association relationships between technology entities are discovered, the evolution of the full life cycle of technology entities is presented, and the spatio-temporal association relationships between technology entities are shown. Refer to Figure 6 , and the specific implementation steps are as follows:

[0106] (1)Technology Map Construction

[0107] 1) Technology data integration. The sources of the technology data include three types of unstructured technology text data: technology projects, technology bulletins, and technology funds. The technology data integration first cleans the errors, false information, etc. from the Internet; then removes punctuation marks, special characters, and stop words without practical meaning; finally, performs word segmentation, stemming, and part-of-speech tagging processing.

[0108] 2) Technology entity recognition. The technology entity is the representation of a thing in the technology map, including attribute information such as the name, alias, description information, type, existence time (start and end time), and geographical location of technology organizations, technology funds, technology projects, and technology talents. The implementation steps of the technology entity recognition are as follows:

[0109] a) Use the BERT pre-trained model to encode the input technology text, extract the semantic information of the text, and obtain the semantic feature vector sequence of each word , where is the semantic feature vector of word j, and P is the number of words.

[0110] b) Optionally, use a bidirectional LSTM network (BiLSTM) to further process the output of BERT , capture the context-dependent relationships between words, and output an enhanced semantic feature vector sequence .

[0111] c) Use the output of the BiLSTM network as the input of the conditional random field to identify the boundaries and types of technology entities in the text. Further screen out technology entity types and attribute information such as technology organizations, technology funds, technology projects, and technology talents.

[0112] 3) Extraction of scientific and technological relationships. The scientific and technological relationships are the representations in the scientific and technological knowledge graph of the connections between scientific and technological entities, including the starting and ending points of the relationships between scientific and technological entities and the starting and ending times of the relationships. Among them, the starting and ending times of the relationships between scientific and technological entities are represented as the intersection of the existence times of the starting and ending scientific and technological entities. The extraction of the scientific and technological entity relationships is determined based on the extraction of scientific and technological entity information, through the co-occurrence relationship of scientific and technological entities in scientific and technological texts and the similarity (cosine similarity) of the semantic feature vectors between scientific and technological entities, that is, if the co-occurrence frequency of two scientific and technological entities in the same context exceeds the threshold and the similarity of the semantic feature vectors exceeds the threshold , then there is a relationship edge between the two scientific and technological entities.

[0113] 4) Generation of scientific and technological knowledge graph. Use graph databases (Neo4j, ArangoDB) or relational databases (MySQL, PostgreSQL) to store the identified scientific and technological entities and relationships. Use knowledge graph reasoning technology to perform reasoning between scientific and technological entities and relationships to further enrich the knowledge graph. Use visualization tools (Gephi, Graphviz) to display the constructed knowledge graph.

[0114] (2) Mining of scientific and technological knowledge associations

[0115] The mining of scientific and technological knowledge associations refers to the process of discovering the association relationships between new scientific and technological entities and events from the scientific and technological knowledge graph, so as to reveal the knowledge structure, development trend and innovation opportunities in the scientific and technological field. Referring to Figure 6 , the mining of scientific and technological knowledge associations includes three methods: association analysis of scientific and technological knowledge graph, temporal analysis of scientific and technological knowledge graph, and spatio-temporal analysis of scientific and technological knowledge graph:

[0116] 1) Association analysis of scientific and technological knowledge graph. By querying the multi-hop relationships of the scientific and technological knowledge graph, a sub-graph of the object of interest is formed, and by viewing the other entities directly and indirectly associated with the scientific and technological entities, the hidden association relationships between the scientific and technological entities are discovered;

[0117] 2) Temporal analysis of scientific and technological knowledge graph. Dynamically present scientific and technological entities and association relationships through the existence times of scientific and technological entities in the scientific and technological knowledge graph and the starting and ending times of scientific and technological entity relationships, support the playback of the state of the scientific and technological knowledge graph at any moment within the starting and ending time range, and present the evolution process of the full life cycle of scientific and technological entities;

[0118] 3) Spatio-temporal analysis of scientific and technological knowledge graph. By associating and displaying the scientific and technological knowledge graph to the three-dimensional earth according to the geographical location information of scientific and technological entities, display the spatio-temporal association relationships between scientific and technological entities, and improve the breadth and depth of scientific and technological knowledge graph analysis.

[0119] The above-mentioned technology competition and cooperation situation awareness module is used to construct a portrait of the technology field. By analyzing think tank texts, the core view content is extracted, and the latest trends in key technology fields are monitored. By constructing a trade competition and cooperation situation network, including the core indicators of trade objects, the definition and calculation of network core indicators, the competition and cooperation situation of resources in key fields is revealed. Refer to Figure 7 , the specific implementation steps are as follows:

[0120] (1) Monitoring of the trends in key technology fields

[0121] By collecting think tank reports in key fields, a think tank text library is formed. Through open-source large language models, the content framework, abstract, and core views of think tank bulletins are automatically extracted. The regional distribution of key fields is displayed in the form of a map, and the core view content of the latest think tank bulletins in key fields is shown when the mouse hovers.

[0122] (2) Analysis of the competition and cooperation situation in key technology fields

[0123] The competition and cooperation situation in the key technology fields refers to the current situation of trade competition and cooperation in the industrial chain and supply chain in a specific field.

[0124] By constructing a trade competition and cooperation situation network with resource reserve and production situation and trade relationship data, the following key core indicators are displayed in the trade competition and cooperation situation network:

[0125] Core indicators of trade objects: Import dependence = Import volume / (Internal production volume + Import volume),

[0126] Network core indicator: Trade stability index

[0127] (3)

[0128] Among them is the proportion of the imports of trade object j in the total imports of all objects, and L is the number of trade objects. The trade stability index is used to measure the complexity and stability of the entire trade network.

[0129] (3) Construction of the portrait of key fields

[0130] The portrait of the field is constructed by analyzing the important mineral resources involved in the integrated circuit and large-capacity battery industries in key emerging industries, constructing a trade competition and cooperation situation network, and forming a resource portrait based on the production volume, reserve data, core indicators of trade objects, and network indicators for the demand of these mineral resources, so as to display the distribution and trade situation of mineral resources and analyze the potential competition and cooperation relationships of key minerals.

[0131] The above-mentioned multi-dimensional display module of technology portraits provides a unified user portal for the comprehensive display of a total of four portraits, namely technology indicator portraits, technology hot spot portraits, technology map portraits, and technology field portraits.

[0132] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0133] The above-described embodiments only represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A multi-perspective scientific and technological data portrait system, characterized in that: It includes the technology indicator integrated evaluation module, technology hotspot labeling module, technology map construction and analysis module, technology competition and cooperation situation awareness module, and technology portrait multi-dimensional display module, among which: The technology indicator integrated evaluation module is used to construct a technology indicator portrait, establish a three-level technology indicator system from the dimensions of technology input, output and benefit, determine the relative importance of indicators at each level through the "Delphi method", and use the geometric analysis method to quantify and rank indicators, so as to achieve multi-dimensional horizontal / vertical comparative analysis of technology competitiveness; The technology hotspot labeling module is used to construct a technology hotspot portrait, extract technology keywords from technology text data in parallel through the TextRank algorithm, establish a technology theme word library through the similarity mapping of technology keywords and technology theme word vector encoding, and construct a technology hotspot label prediction model and build a sample set training based on the technology theme word library to display the technology hotspots in the future in the form of a label word cloud; The technology graph construction and analysis module is used to construct a technology graph portrait, establish connections between technology entities through knowledge graph modeling and technology entity relationship extraction, determine the potential relationship between technology entities through technology graph association analysis, technology graph time series analysis, and technology graph spatiotemporal analysis, present the evolution of the entire life cycle of technology entities, and display the spatiotemporal relationship between technology entities; The technology competition and cooperation situation awareness module is used to build a portrait of the technology field, extract core viewpoints by analyzing think tank texts, monitor the latest trends in key technology fields, and calculate import dependence and trade stability index by building a trade competition and cooperation situation network to reveal the competition and cooperation situation of resources in key fields; The multi-dimensional display module of science and technology portraits provides a unified user portal for comprehensive display of science and technology indicator portraits, science and technology hot spot portraits, science and technology map portraits, and science and technology field portraits; The technology competition and cooperation situation awareness module includes a key field trend monitoring module, a key field resource competition and cooperation situation analysis module and a key field portrait construction module.

2. The multi-perspective scientific and technological data portrait system according to claim 1 is characterized in that: The science and technology indicator integrated evaluation module includes a science and technology indicator evaluation system construction module, a multi-dimensional science and technology competitiveness index ranking module and a science and technology indicator portrait construction module, among which: The construction module of the science and technology indicator evaluation system is composed of the science and technology development environment index , Science and Technology Development Resource Index , Science and Technology Innovation Output Index , Science and Technology and Industrial Security Index , Comprehensive Science and Technology Contribution Index There are 5 first-level indicators in total; the first-level indicators can be decomposed into multiple second-level indicators, expressed as ; The second-level indicators are composed of multiple directly quantifiable third-level indicators, expressed as ; The indicator system is quantified by formula (1a); （1a）； st （1b）； Among them, i, j, and k represent the serial numbers of the first-level indicator, the second-level indicator, and the third-level indicator respectively; and Indicates the third level indicator The maximum and minimum values ​​of ; and Indicates the corresponding three-level indicators and secondary indicators The weight coefficient of Indicates the first level indicator Secondary indicators included The number of Represents secondary index The three-level indicators included the number of The multi-dimensional science and technology competitiveness index ranking module realizes the science and technology competitiveness index ranking from five indicator dimensions: science and technology development environment index, science and technology development resource index, science and technology innovation output index, science and technology and industrial security index, and science and technology comprehensive contribution index; The science and technology indicator portrait construction module uses charts and numerical analysis methods to conduct a horizontal comparison of the rankings of first-level evaluation indicators of different evaluation objects in the same time period from five indicator dimensions, namely, science and technology development environment index, science and technology development resource index, science and technology innovation output index, science and technology and industrial security index, and science and technology comprehensive contribution index. It also conducts horizontal / vertical comparison of different indicator dimensions of different evaluation objects in different time periods by "drilling down" from the first-level indicators to the second-level and third-level indicators.

3. The multi-perspective scientific and technological data portrait system according to claim 2 is characterized in that: Secondary index weight coefficient and weight coefficient of the third-level indicators The solution process is: First, the "Delphi method" is used to determine any two indicators at the corresponding level. and The relative importance of , Respectively Compared to Equally important, slightly important, important, relatively important, very important; then construct a comparison matrix of the relative importance of indicators , the geometric analysis method of formula (2) is used to calculate the index weight; （2）； in, Indicates any indicator at each level The weight coefficient of , D represents the indicator dimension of the corresponding level.

4. The multi-perspective scientific and technological data portrait system according to claim 2 is characterized in that: The science and technology indicator portrait construction module uses a line graph to display the ranking change trend of different indicator values ​​of the same evaluation object over time, uses a radar chart to display the ranking comparison of different evaluation objects in the same indicator dimension, and uses a heat map to display the changes in different indicator dimensions of the evaluation object in recent time periods.

5. The multi-perspective scientific and technological data portrait system according to claim 1 is characterized in that: The technology hotspot labeling module includes a technology theme word library construction module, a technology hotspot label prediction module and a technology hotspot word cloud labeling module, wherein: The technology thesaurus construction module first uses papers, patents, think tanks, and reports as data sources for technology texts, and constructs a technology text keyword relationship graph G in parallel. The technology text keyword relationship graph G uses technology keywords as nodes and the co-occurrence relationship of technology keywords in the text as edges. The co-occurrence relationship is only valid if the co-occurrence frequency of two technology keywords in the same text is greater than a threshold. A relationship edge is established when the word vector is generated; then the TextRank algorithm is used to rank the technology keywords in the technology text keyword relationship graph G to generate a (technology keyword, score) set; then, Word2Vec is used to encode the technology keywords in the technology text keyword relationship graph G, calculate the similarity between the technology keywords and the technology theme words, and replace the technology keywords with the technology theme words with the greatest similarity to achieve standardization; finally, the scores of different technology keywords with the same technology theme words are accumulated and aggregated, and the technology theme words with the top K scores are taken as technology hotspot tags to form a technology theme word library; The technology hotspot label prediction module is used to predict the labels that may become technology hotspots of the evaluated object in the future. The technology hotspot label prediction model consists of a three-layer LSTM network cascade, and a 32-node fully connected layer with a SoftMax activation function is added at the end. The input is the embedded coding sequence of the technology theme words in the previous m stages, and the output is the probability of the technology theme words in the next stage; the technology theme word library constructed by the technology texts of papers, patents, think tanks, and reports is used to construct the training sample set , fine-tuning the network structure of the technology hotspot label prediction model through incremental training, Optimize sample training loss to improve model performance, where is the serial number of the time period to which the sample belongs, and Q is the number of time period divisions; The technology hot word cloud annotation module adjusts the size and color of the label according to the technology hot label probability value output by the technology hot label prediction model, forming a cloud-like image with technology keywords as visual elements, showing the technology hot spots of the evaluation object in the future time period.

6. The multi-perspective scientific and technological data portrait system according to claim 1 is characterized in that: The technology map construction and analysis module includes a technology map construction module and a technology knowledge association mining module, wherein: The technology graph construction module first integrates three types of unstructured technology text data: technology projects, technology bulletins, and technology funds. It uses the BERT pre-training model to encode the input technology text, extract the semantic information of the text, and generate a semantic feature vector sequence for each word. , P is the number of words in the text; then, the output of the BERT pre-trained model is further processed using a bidirectional LSTM network , capturing the contextual dependencies between words and obtaining an enhanced semantic feature vector sequence , and the output of the bidirectional LSTM network As the input of the conditional random field, the boundaries and types of scientific and technological entities in the text are identified, and scientific and technological entities and attribute information are screened out according to the type, where scientific and technological entities include scientific and technological organizations, scientific and technological funds, scientific and technological projects, and scientific and technological talents; then, the scientific and technological entity relationships are extracted. The scientific and technological entity relationships are determined by the co-occurrence relationship of scientific and technological entities in the text and the similarity of the semantic feature vectors between scientific and technological entities. That is, if the frequency of co-occurrence of two scientific and technological entities in the same context exceeds the threshold And the similarity of semantic feature vectors exceeds the threshold , then there is a relationship edge between the two scientific and technological entities; finally, the identified scientific and technological entities and scientific and technological entity relationships are stored in a graph database or a relational database to build a complete scientific and technological graph; The science and technology knowledge association mining module discovers the potential relationship between science and technology entities and events from the science and technology graph through three methods: science and technology graph association analysis, science and technology graph time series analysis, and science and technology graph time-space analysis. Among them, the science and technology graph association analysis queries the multi-hop relationship of the science and technology graph to form a subgraph of the object of interest, and obtains the hidden association relationship between science and technology entities by checking other entities directly and indirectly related to the science and technology entity; the science and technology graph time series analysis dynamically presents the existence time of the science and technology entity and the start and end time of the science and technology entity relationship, and plays back the science and technology graph state at any time within the start and end time range, presenting the evolution process of the entire life cycle of the science and technology entity; the science and technology graph time-space analysis displays the time-space association relationship between science and technology entities by associating the science and technology graph with the geographical location information of the science and technology entity and displaying it on the three-dimensional earth.

7. The multi-perspective scientific and technological data portrait system according to claim 1 is characterized in that: In the technology competition and cooperation situation awareness module, The key science and technology field trend monitoring module collects think tank reports in key fields to form a think tank text library. It uses an open source large language model to automatically extract the content framework, summary and core viewpoints of think tank express reports, and displays the regional distribution of key fields in the form of a map. Hovering the mouse will display the core viewpoints of the latest think tank express reports in key fields. The module for analyzing resource competition and cooperation in key science and technology fields builds a trade competition and cooperation network based on resource reserves and production and trade exchanges. In the trade competition and cooperation network, the import dependence is calculated to show the degree of dependence on trade imports. The stability of trade is measured by calculating the trade stability index, i.e., the network structure entropy, to analyze the trade competition and cooperation situation of the industrial chain and supply chain. The module for constructing portraits of key scientific and technological fields conducts a network analysis of the competition and cooperation in the trade of important mineral resources involved in the integrated circuit and large-capacity battery industries, forms a resource portrait based on the output and reserve data of mineral resource demand, and displays the distribution of mineral resources in order to analyze the potential competition and cooperation between key minerals.

8. A multi-perspective scientific and technological data profiling method, characterized in that: Based on the multi-perspective scientific and technological data portrait system described in any one of claims 1 to 7, a multi-perspective scientific and technological data portrait is realized, specifically: The science and technology indicator integrated evaluation module is used to build a science and technology indicator portrait, and a three-level science and technology indicator system is established from the dimensions of science and technology input, output and benefit. The relative importance of indicators at each level is determined through the "Delphi method", and the geometric analysis method is used to quantify and rank the indicators, so as to achieve multi-dimensional horizontal / vertical comparative analysis of science and technology competitiveness; Use the technology hotspot labeling module to build a technology hotspot portrait, extract technology keywords from technology text data in parallel through the TextRank algorithm, establish a technology theme word library through the similarity mapping of technology keywords and technology theme word vector encoding, and build a technology hotspot label prediction model and build a sample set training based on the technology theme word library to display the technology hotspots in the future in the form of a label word cloud; Use the technology graph construction analysis module to build a technology graph portrait. Through knowledge graph modeling and technology entity relationship extraction, establish the connection between technology entities. Through technology graph association analysis, technology graph time series analysis, and technology graph spatiotemporal analysis, determine the potential relationship between technology entities, present the evolution of the entire life cycle of technology entities, and show the spatiotemporal relationship between technology entities. Use the technology competition and cooperation situation awareness module to build a portrait of the technology field, extract the core content by analyzing think tank texts, monitor the latest trends in key technology fields, and calculate the import dependence and trade stability index by building a trade competition and cooperation situation network to reveal the competition and cooperation situation of resources in key fields; The multi-dimensional display module of science and technology portraits is used to unify the user portal for the comprehensive display of science and technology indicator portraits, science and technology hotspot portraits, science and technology map portraits, and science and technology field portraits.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the multi-perspective scientific and technological data portrait method of claim 8 is implemented to realize multi-perspective scientific and technological data portrait.

10. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the multi-perspective scientific and technological data portrait method of claim 8 is implemented to realize multi-perspective scientific and technological data portrait.

Images