Key core technology identification method and device based on science and technology information, computer equipment, storage medium and program product

Abstract

The invention provides a key core technology identification method and device based on science and technology information, computer equipment, a storage medium and a program product, and the method comprises the steps: obtaining an SAO triple corresponding to each text data in a science and technology information data set of a target field, obtaining at least one technical cluster of the target domain based on each SAO triple; wherein the scientific and technological information data set comprises text data corresponding to scientific and technological achievements disclosed in a target field; and calculating in-degree centrality, out-degree centrality, link strength and network density of the technical cluster based on the SAO triple corresponding to each piece of text data in each technical cluster, and obtaining the key core degree of the technical cluster based on the in-degree centrality, out-degree centrality, link strength and network density of the technical cluster.

Description

Technical Field

[0001] The present invention relates to the field of data mining technology, and in particular to a method, device, computer equipment, storage medium and program product for identifying key core technologies based on scientific and technological information. Background Art

[0002] The demand for key core technologies and technology foresight are extremely important components in the process of a country or industry formulating medium- and long-term science and technology development plans, and play a supporting role in strategic decision-making. With the rapid development of global science and technology, how to accurately grasp the development trend of science and technology, predict future development directions, and determine key technology fields in order to combine national development needs, seek benefits and avoid harm, and win future development advantages is also a focus of attention for governments and industries. At present, the identification, demand analysis and foresight of key core technologies have been paid attention to by more and more countries and have become a new technology strategic management tool. In the article "Research Progress on Key Core Technology Identification Methods" proposed by Zhao Jian et al., it is summarized that the existing key core technology identification mainly includes four types: methods based on expert experience, methods based on patent indicators, methods based on patent relationship networks, and methods based on text mining.

[0003] The tobacco industry is an important part of the national economy, and scientific and technological innovation plays a vital supporting role in the high-quality development and modernization of the tobacco industry. At present, the tobacco industry faces both opportunities and challenges in terms of scientific and technological development. Under this background, research on key core technology needs and technology forecasts at the industry level is conducted, especially in terms of determining key core technology areas, sorting out technical shortcomings, and forecasting future key technologies. This plays an important role in decision-making support for the formulation of strategic plans and overall layout related to scientific and technological innovation in the industry.

[0004] However, the current technical information in the tobacco field is scattered and complex, and multiple technical fields are cross-integrated. It is difficult to deeply understand the content and grasp the technical development trend by relying only on the information in the catalog of scientific and technological literature. In addition, there are large differences among scholars in the academic community when constructing the indicator system, making it difficult to form a unified standard. The lack of scientific judgment standards also leads to subjectivity, one-sidedness and uncertainty in the identification of key core technologies in the tobacco industry, making it difficult to objectively and comprehensively identify key core technologies. Therefore, it is particularly important to study and establish key core technology identification applicable to the tobacco industry.

[0005] In order to solve the above problems, people have been seeking an ideal technical solution. Summary of the invention

[0006] Based on this, it is necessary to provide a key core technology identification method, device, computer equipment, storage medium and program product based on scientific and technological information to address the above technical problems.

[0007] In order to achieve the above-mentioned object, the present invention provides a key core technology identification method based on scientific and technological information in a first aspect, comprising the following steps: Acquire SAO triples corresponding to each text data in the target field science and technology information data set, and acquire at least one technology cluster in the target field based on each SAO triple; wherein the science and technology information data set includes text data corresponding to the scientific and technological achievements disclosed in the target field; Calculate the in-degree centrality, out-degree centrality, link strength and network density of each technology cluster based on the SAO triple corresponding to each text data in each technology cluster; The key core degree of the technology cluster is obtained based on the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster.

[0008] In an optional embodiment of the first aspect, calculating the in-degree centrality, out-degree centrality, link strength and network density of each technology cluster based on the SAO triple of each technology cluster includes: Obtaining the in-degree centrality of the technology cluster based on the in-degree centrality of each node and the number of nodes in the SAO triple of the technology cluster; wherein the in-degree centrality is used to characterize the importance of the technology cluster; The out-degree centrality of the technology cluster is obtained based on the out-degree centrality of each node and the number of nodes in the SAO triple of the technology cluster; wherein the out-degree centrality is used to characterize the sociability of the technology cluster; Based on the number of other nodes connected to each node in the SAO triplet of the technology cluster, the link strength of the technology cluster is obtained; wherein the link strength is used to characterize the number of links of each node in the technology cluster; Based on the connection relationship between each node and other nodes in the SAO triple of the technology cluster, the network density of the technology cluster is obtained; wherein the network density is used to characterize the closeness between each node and other nodes in the technology cluster.

[0009] In an optional embodiment of the first aspect, obtaining the critical core degree of the technology cluster based on the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster includes: Based on the preset contribution weights of in-degree centrality, out-degree centrality, link strength and network density, the in-degree centrality, out-degree centrality, link strength and network density are weighted and fused respectively to obtain the key core score of the technology cluster.

[0010] In an optional embodiment of the first aspect, obtaining the SAO triples corresponding to each text data in the target domain scientific and technological information dataset includes: Obtain the SAO triples corresponding to each text data in the target field science and technology information dataset, including: When the target field scientific and technological information dataset is a paper text, a text mining method is used to extract SAO triples of the paper text, and SAO screening is performed based on paper keywords to obtain SAO triples of each paper; wherein the paper keywords are the paper keywords in the field paper keyword vocabulary, and the field paper keyword vocabulary is constructed by extracting the paper keywords in the paper text bibliography and extracting the paper keywords in the paper abstract.

[0011] In an optional embodiment of the first aspect, SAO triples corresponding to each text data in a target field scientific and technological information dataset are obtained, including: when the target field scientific and technological information dataset is a patent text, the CRF model is used to extract the SAO triples of the patent text, and SAO screening is performed based on patent keywords generated by a large language model to obtain SAO triples for each patent.

[0012] In an optional embodiment of the first aspect, acquiring at least one technology cluster in the target field based on each SAO triplet includes: The SAO triples corresponding to each text data are semantically represented based on the Bert model, and the synonyms of SAO triples are merged based on KMeans clustering; A technology association network is constructed based on the association relationship of SAO triplets, and technology clusters are divided based on the graph theory community detection method.

[0013] In order to achieve the above object, the second aspect of the present invention provides a technology identification method based on scientific and technological information, comprising the following steps: Acquire SAO triples corresponding to each text data in the target field science and technology information data set, and acquire at least one technology cluster in the target field based on each SAO triple; wherein the science and technology information data set includes text data corresponding to the scientific and technological achievements disclosed in the target field; Calculate the temporal novelty, innovative novelty, growth, technological impact, and social impact of the technology cluster based on the bibliographic data of each text data in each technology cluster; Among them, the time novelty is used to characterize the average disclosure time of the scientific and technological achievements corresponding to the technology cluster; the innovation novelty is used to characterize the total number of citations of the scientific and technological achievements corresponding to the technology cluster; the growth is used to characterize the average growth rate of the scientific and technological achievements corresponding to the technology cluster; the technical impact is used to characterize the technical citations of the scientific and technological achievements corresponding to the technology cluster; the social impact is used to characterize the total number of R&D personnel of the scientific and technological achievements corresponding to the technology cluster; the degree of emergence of the technology cluster is obtained based on the time novelty, innovation novelty, growth, technical impact and social impact of the technology cluster; Calculate the in-degree centrality, out-degree centrality, link strength and network density of each technology cluster based on the SAO triple corresponding to each text data in each technology cluster; The key core degree of the technology cluster is obtained based on the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster.

[0014] In order to achieve the above-mentioned object, the third aspect of the present invention provides a key core technology identification device based on scientific and technological information, comprising: A technology cluster acquisition module is used to acquire SAO triples corresponding to each text data in the target field science and technology information data set, and acquire at least one technology cluster in the target field based on each SAO triple; wherein the science and technology information data set includes text data corresponding to the scientific and technological achievements disclosed in the target field; A technology cluster indicator acquisition module, used to calculate the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster based on the SAO triple corresponding to each text data in each technology cluster; The technology cluster key core degree acquisition module is used to acquire the key core degree of the technology cluster based on the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster.

[0015] In order to achieve the above-mentioned purpose, the fourth aspect of the present invention provides a computer device, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; the memory is used to store computer programs; the processor is used to implement the key core technology identification method based on scientific and technological information as described in the first aspect when executing the program stored in the memory, or to implement the technology identification method based on scientific and technological information as described in the second aspect.

[0016] In order to achieve the above-mentioned objectives, the fifth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it implements the key core technology identification method based on scientific and technological information as described in the first aspect, or is used to implement the technology identification method based on scientific and technological information as described in the second aspect.

[0017] In order to achieve the above-mentioned objectives, the sixth aspect of the present invention provides a computer program product, including a computer program, which, when executed by a processor, implements the key core technology identification method based on scientific and technological information as described in the first aspect, or is used to implement the technology identification method based on scientific and technological information as described in the second aspect.

[0018] The beneficial effects of the present invention are: The present invention performs technical association mining on each text data in the target field scientific and technological information data set, obtains the SAO triples corresponding to each text data in the target field scientific and technological information data set, obtains at least one technology cluster in the target field based on each SAO triple, calculates the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster based on the SAO triples corresponding to each text data in each technology cluster, and obtains the key core degree of the technology cluster based on the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster. The above scheme can accurately identify technical key points and innovation points.

[0019] Furthermore, at least one technology cluster is obtained based on the connection between technologies, and a corresponding indicator system is designed for emerging technologies - temporal novelty, innovative novelty, growth, technological impact, and social impact. Then, based on the text data in each technology cluster, each technical indicator of the technology cluster is calculated, and based on the various technical indicators, the degree of emergence of the technology cluster is obtained, and finally the technical key points and innovation points are accurately identified, thereby providing a scientific basis and decision-making support for the technological development of the tobacco industry.

[0020] Furthermore, the improved Subject-Action-Object (SAO) method can be used to deeply analyze text data through the construction of SAO association networks, explore the intricate connections between technologies, and provide macro and micro perspectives for scientific and technological development.

[0021] Furthermore, in the data preprocessing stage, key core technologies such as synonymous merging, semantic association analysis and named entity recognition were used to ensure high accuracy and depth of information extraction. At the same time, a large language model was used to generate patent keywords, further enhancing the professionalism and accuracy of the SAO method. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 1 is a flow chart of the key core technology identification method based on scientific and technological information according to Embodiment 1 of the present invention; Figure 2 It is a schematic diagram of performing technical association mining on each text data in the target field scientific and technological information data set in Embodiment 1 of the present invention; Figure 3 is a flow chart of the emerging technology identification method based on scientific and technological information according to Embodiment 2 of the present invention; Figure 4 It is a structural schematic diagram of a key core technology identification device based on scientific and technological information of the present invention; Figure 5 It is a schematic diagram of the structure of the computer device of the present invention. DETAILED DESCRIPTION

[0023] The technical solution of the present invention is further described in detail below through specific implementation methods.

[0024] It will be understood by those skilled in the art that, unless specifically stated, the singular forms "one", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the terms "including" and "comprising" used in the embodiments of the present application mean that the corresponding features can be implemented as the presented features, information, data, steps, operations, elements and / or components, but do not exclude implementation as other features, information, data, steps, operations, elements, components and / or combinations thereof supported by the technical field. It should be understood that when we say that an element is "connected" or "coupled" to another element, the one element may be directly connected or coupled to the other element, or may refer to the one element and the other element establishing a connection relationship through an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein indicates at least one of the items defined by the term, for example, "A and / or B" may be implemented as "A", or as "B", or as "A and B".

[0025] In order to make the objectives, technical solutions and advantages of the present application more clear, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0026] Example 1 The tobacco industry is an important part of the national economy, and scientific and technological innovation plays a vital supporting role in the high-quality development and modernization of the tobacco industry. At present, the tobacco industry faces both opportunities and challenges in terms of scientific and technological development. Under this background, research on key core technology needs and technology forecasts at the industry level is conducted, especially in terms of determining key core technology areas, sorting out technical shortcomings, and forecasting future key technologies. This plays an important role in decision-making support for the formulation of strategic plans and overall layout related to scientific and technological innovation in the industry.

[0027] The embodiment of this application takes the tobacco field as an example to provide a specific implementation method of the key core technology identification method based on scientific and technological information.

[0028] Figure 1 A flowchart of a key core technology identification method based on scientific and technological information provided in an embodiment of the present application, the execution subject of the method can be a device terminal or a server, etc. Figure 1 As shown, the key core technology identification method based on scientific and technological information provided in the embodiment of the present application includes the following steps: Step S1, obtaining SAO triples corresponding to each text data in the target field scientific and technological information dataset, and obtaining at least one technology cluster in the target field based on each SAO triple.

[0029] Among them, the tobacco field science and technology information dataset includes text data corresponding to the scientific and technological achievements disclosed in the target field, such as paper texts, patent texts, fund project texts, book texts, and research report texts. This embodiment mainly constructs a science and technology information dataset based on domestic and foreign patent texts in the tobacco industry, where the patent text at least includes patent abstract text information. Specifically, the patent abstract text information includes: patentee, application date, inventor, IPC classification number, comparative documents, and specification abstract, etc.

[0030] This embodiment mainly constructs a scientific and technological information dataset based on domestic and foreign patent texts in the tobacco industry.

[0031] Specifically, Figure 2 The figure is a flowchart of technical association mining in an example of an embodiment of the present application.

[0032] Step 101, such as Figure 2 As shown, the CRF model is used to extract the SAO triples of the patent abstract text, and SAO screening is performed based on the patent keywords generated by the large language model to obtain the SAO triples of each patent abstract text.

[0033] Among them, CRF (Conditional Random Fields) is a statistical model for sequence labeling and structured prediction tasks, which is used to extract SAO (Subject-Predicate-Object) triples.

[0034] The advantages of the SAO structure are mainly reflected in the following aspects: 1) Revealing deep semantic relationships and extracting high-dimensional information. Compared with ordinary keyword or topic extraction, the SAO structure can provide richer information dimensions and reveal deep semantic relationships in the text. For example, when processing patent documents, "using A to prepare B" (Subject: "A", Action: "Preparation", Object: "B") can reveal the application and output relationship between A and B, rather than simply identifying the technical action of "preparation", and provide a more comprehensive technical description through the combination of subject, predicate and object.

[0035] 2) Accurate information association. The SAO structure can more accurately identify related technologies and their functions or application scenarios. This is very important in the division of technology clusters and the identification of key core technologies and core technologies, and can help researchers identify similarities and differences between technologies.

[0036] 3) Support complex query and analysis. In technology trend analysis or competitive intelligence analysis, researchers can construct complex queries based on the SAO structure to identify the actions and objects of specific subjects, thereby discovering potential technological innovations and technology routes.

[0037] 4) Enhanced semantic similarity calculation: When using semantic similarity to divide technology clusters, the SAO structure can provide more semantic feature points to help the calculation model identify subtle semantic connections between different documents, thereby improving the accuracy and efficiency of technology cluster division.

[0038] Therefore, this application extracts the SAO (Subject-Predicate-Object) triples in the patent abstract text and uses them as a basis to carry out in-depth technical identification work.

[0039] Specifically, in this embodiment, the specific steps of using the CRF model to label and extract SAO triples include: 1) Data preparation: Perform natural language processing preprocessing steps such as word segmentation, part-of-speech tagging, and named entity recognition on the patent abstract text to collect a text dataset containing SAO information.

[0040] 2) Feature Engineering: Select appropriate features for the CRF model, such as part of speech, entity type, contextual vocabulary, etc. These features help capture the information of SAO relations.

[0041] 3) Labeled data: Combine manual annotation with existing large language models to add SAO tags to the words in each sentence and generate feature vectors for each word and tag.

[0042] 4) Model training: Use the labeled dataset to train the conditional CRF model so that the CRF model learns to predict the SAO label of each word based on the input features and context.

[0043] 5) Predicting SAO relations: Run the trained CRF model on the unlabeled patent abstract text to predict the SAO label for each vocabulary.

[0044] Furthermore, some post-processing rules can be applied to the generated SAO triples to improve accuracy and consistency, for example, verifying the legitimacy of the SAO triples according to grammatical rules.

[0045] After SAO processing, the SAO triplet of each patent is obtained. The following table shows the SAO triplet of the patent.

[0046] Table 1 Patented SAO triples Patent ID Patent Application Number SAO Triplet 0 WO2015129098A1 [[('cigarette rawmaterials', 'beentreated with', 'alkali'), ('flavorcomponents', 'weretrapped in', 'closedspace')]] 1 BR112020011222A2 ['None', [('cartridge', 'include','compartment')], 'None', [('heater', 'have', 'sinusoidal -shaped member')], [('absorbent material', 'surround', 'sinusoidally shapedmember')], 'None'] 2 AU2019339246A1 [[('people', 'stop', 'smoking')], 'None',[('splitting section','separate', 'inhalation section'),('splitting section','separate', 'organicmatter combustionsection')], [('external atmosphericair', 'enter', 'inhalation section')],[('smoke', 'beinhaled', 'combustion')], 'None'] Step 102: semantically represent the SAO triples corresponding to each patent based on the Bert model, and merge synonyms of the SAO triples based on KMeans clustering.

[0047] It is understandable that BERT (Bidirectional Encoder Representations from Transformers) is a powerful language representation model that uses the Transformer architecture and undergoes large-scale unsupervised training to enable it to capture contextual semantic features. Representing SAO entities as high-dimensional vectors with the help of the BERT model is conducive to similarity calculation.

[0048] Specifically, the BERT model encodes the input text, where [CLS] represents the first token of a sentence and [SEP] represents the sentence separator. Through the processing of Token Embedding, Segment Embedding, and Position Embedding layers, BERT converts the text into a vector of fixed dimension and considers the contextual information of the sentence. Finally, a high-dimensional vector representation of each SAO entity is obtained, and then the cosine similarity is used to calculate the similarity of SAO phrases.

[0049] It should be noted that the text representation of SAO triples has inconsistent word representations, such as aerosol-generating article and aerosol-generating device, so synonym merging is required. The KMeans algorithm is a classic clustering algorithm, and its main use is to find the most representative data points, namely cluster centers, among a large number of high-dimensional data points. These cluster centers represent data of different categories, thereby achieving data classification. In this embodiment, the KMeans algorithm is used to cluster SAO entities to identify and merge synonyms.

[0050] Specifically, this embodiment uses the BERT model to represent SAO entities as high-dimensional vectors, so that each entity can compare their semantic similarities by calculating vector similarity, and then applies the KMeans algorithm to cluster these high-dimensional vectors into multiple cluster centers, each cluster center representing a class of similar entities. This process is iterated continuously until the cluster center no longer changes, and the similarity between each entity and each cluster center is calculated to determine whether to classify the entity into a specific cluster. If the similarity between the entity and a cluster center exceeds a predetermined threshold, it will be classified into the cluster. After being processed by this method, similar SAO entities can be classified into one category, realizing the synonym merging function, and increasing the density of the SAO network.

[0051] Step 103: construct a technology association network based on the association relationship of the SAO triples, and divide the technology clusters based on the graph theory community detection method.

[0052] Specifically, before dividing the technology clusters, based on the association relationship of SAO triples, NetworkX in Python is used to build a technology association network, and the associated SAO triples are connected through edges to form a graphical network structure. Then, the Louvain algorithm is applied to divide the technology clusters.

[0053] Among them, NetworkX is a Python library for creating, manipulating, and studying complex network structures. It is suitable for graph theory and network analysis, and provides rich graph data structures and many commonly used algorithms.

[0054] The Louvain algorithm is a community detection algorithm based on graph theory, which can identify communities or clusters formed by closely related nodes in the graph. The optimization goal of community or cluster division is to maximize the modularity of the entire community network. Iteratively build communities and observe the modularity of the entire community network. When the community modularity no longer changes, it means that the optimal community division is completed. It can be understood that in specific applications, each technology cluster can be regarded as a community containing similar SAO entities. The Louvain algorithm is used to cluster SAO entities with high correlation in the same technology cluster, thereby realizing the deep and fine-grained division and organization of text and semantic information, which is helpful for subsequent technology identification and analysis.

[0055] Step S2, such as Figure 1 As shown, the in-degree centrality, out-degree centrality, link strength and network density of each technology cluster are calculated based on the SAO triple corresponding to each text data in each technology cluster.

[0056] It can be understood that key core technologies refer to technologies that play an important role and are indispensable in specific technological fields, and have characteristics such as stability, centrality, and importance.

[0057] According to the research of domestic and foreign scholars, as well as the properties and characteristics of SAO itself, this embodiment uses in-degree centrality, out-degree centrality, link strength and network density as important features for identifying key core technologies.

[0058] Centrality As an index, centrality can reveal the position of a node in the center of the entire SAO triple network. In a directed graph, centrality can be divided into in-degree centrality and out-degree centrality, which can describe the popularity and sociability of a node respectively.

[0059] Introducing the concept of centrality in the SAO network can help us understand the semantic meaning of each node more clearly. If a noun node has a higher closeness centrality, then this node is very likely to be a key technology, and if a verb node has a higher closeness centrality, then this verb node is very likely to be a key step or function of a certain technology.

[0060] Similarly, if a noun node has a high in-degree centrality, then the node is an object node, which is likely to be a useful technology in the field, or an important indicator for measuring the performance of a certain technology. If a noun node has a high out-degree centrality, then the noun node will be a topic node, and is likely to be a general technology in the relevant field, or a solution to some technical problems.

[0061] In this embodiment, the in-degree centrality is used to characterize the importance of the technology cluster; the in-degree centrality of the technology cluster is obtained based on the in-degree centrality of each node in the SAO triple of the technology cluster and the number of nodes. The calculation formula of the in-degree centrality is: (1) in, C in Representation Node m The in-degree is close to the centrality, It means node m To Node l The number of relationships between .

[0062] In this embodiment, the out-degree centrality is used to characterize the sociability of the technology cluster; the out-degree centrality of the technology cluster is obtained based on the out-degree centrality of each node in the SAO triple of the technology cluster and the number of nodes. The out-degree centrality calculation formula is shown in 2: (2) in, C out Representation Node m The out-degree is close to the centrality, which means that the node m To Node l The number of relationships between .

[0063] Link Strength In this embodiment, the link strength indicates the number of links of a node. Specifically in the SAO network, the link strength indicates the number of relationships between nouns and verbs, revealing the influence of a node in the SAO network. The higher the link strength, the greater the influence of the node in the network and the more reliable the information.

[0064] In this embodiment, the link strength of the technology cluster is obtained based on the number of other nodes connected to each node in the SAO triplet of the technology cluster; the formula of the link strength is shown in calculation formula 3: (3) in, Degree It means that the node n m The number of other noun or verb nodes connected.

[0065] Network density In this embodiment, the network density represents the connection strength between a node and other nodes in the SAO network. If the density of a cluster is 1, it means that all nodes in the cluster are connected to all other nodes. If the density of a cluster is 0, it will not be connected to any node. If the density of a cluster is high, it means that this technology has been widely verified.

[0066] In this embodiment, based on the connection relationship between each node and other nodes in the SAO triplet of the technology cluster, the network density of the technology cluster is obtained; wherein the network density is used to characterize the closeness between each node in the technology cluster and other nodes. The calculation formula of the network density is shown in 4: (4) in, Density is the network density, k is the number of nodes, and is the number of nodes n m , n l The amount of relationship between .

[0067] Step S3, obtaining the critical core degree of the technology cluster based on the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster.

[0068] Specifically, based on the preset contribution weights of in-degree centrality, out-degree centrality, link strength and network density, the in-degree centrality, out-degree centrality, link strength and network density are weighted and fused respectively to obtain the key core score of the technology cluster.

[0069] Furthermore, in order to facilitate the comprehensive comparison of indicators and key core scoring, the in-degree centrality, out-degree centrality and link strength are standardized, and the indicators are weighted according to the dimension level, that is, the weight distribution rules are preset to obtain the key core technology score. The formula is as follows: (5).

[0070] in,c 1 、c 2 、c 3 、c 4 Reset the weights for the preset weight assignment rules.

[0071] It can be understood that the key core score is the key core degree of the technology cluster. The higher the key core score, the higher the key core degree of the technology cluster.

[0072] Of course, in other embodiments, after the key core score of each technology cluster is obtained, the key core level of the technology cluster is obtained according to the key core score of each technology cluster.

[0073] For example, the key core scores are divided into multiple intervals in descending order, where the interval with the highest score is the first key core level, and the second interval is the second key core level. Then, the key core level is determined by judging which interval the key core scores of each technology cluster fall into.

[0074] Verification: Based on the emerging technology identification scheme in the previous part, the domestic patent technology of e-cigarettes in the tobacco field in the past 10 years is used as the verification set, and the key core points in the domestic patent technology of e-cigarettes are calculated based on the in-degree centrality, out-degree centrality, link strength and network density.

[0075] It should be noted that after 0-1 standardization, the score range of each indicator is [0,1]. The closer the score is to 1, the higher the characteristics of the technology are compared with other technologies.

[0076] Table 2 lists the 10 highest scoring technology clusters.

[0077] Table 2 Top 10 technology clusters by patent key core scores cluster Cluster Density Cluster Centrality Cluster size normalized score Cluster coreness score C55 0.004 1.000 1.000 2.000 C99 0.021 0.097 0.145 0.242 C545 0.022 0.096 0.130 0.226 C193 0.024 0.081 0.119 0.200 C41 0.026 0.076 0.108 0.184 C425 0.036 0.053 0.079 0.132 C1206 0.038 0.046 0.075 0.121 C502 0.038 0.047 0.070 0.117 C561 0.044 0.041 0.061 0.103 C35 0.050 0.038 0.054 0.092 Table 3 lists the 5 representative SAO triple information corresponding to these 10 technology clusters. Among them, the representative triples are selected according to the in-degree and occurrence frequency of the SAO nodes to assist in the positioning of all triple information.

[0078] Table 3 Representative SAO triples of the top 10 technology clusters with key core scores cluster Represents a triple C55 Atomizer|||Including|||Liquid storage component; Atomizer|||Including|||Base assembly; Atomizer|||Including|||Liquid storage assembly; Atomizer|||Including|||Atomizing element C99 Atomizer|||convenient|||smoke oil; atomizer|||appears|||leakage; oil storage bottle|||contains|||smoke oil; oil guide|||absorbs|||smoke oil; oil reservoir|||transports|||smoke oil C545 Electronic cigarette liquid|||composition|||flavor composition; electronic cigarette liquid|||composition|||flavor composition; electronic cigarette liquid|||includes|||cough and phlegm relieving effect; electronic cigarette liquid|||has|||lung moistening and cough relieving effect; electronic cigarette liquid|||includes|||Chinese herbal medicine extract C193 Technical solution|||Prevent|||Transport process; Technical solution|||Prevent|||Transport process; Technical solution|||Enhance|||Oil-proof performance; Technical solution|||Has|||Simple structure; Technical solution|||Avoid|||Aerosol generation substrate C41 The cartridge|||reduces|||the oil; the cartridge|||includes|||the airway; the cartridge|||has|||seal; the cartridge|||includes|||the cartridge bin; the cartridge|||includes|||the oil tank C425 atomization|||makes|||smoke; atomization|||accumulation|||buffer device; atomization|||uses|||damping tube; aerosol|||has|||particle size distribution; atomization|||closes|||gas path C1206 Electronic cigarette oil|||has|||health care effect; Electronic cigarette oil|||has|||health care effect; Honeysuckle extract|||has|||heat-clearing and detoxifying effect; Electronic cigarette oil|||brings|||sweet and greasy feeling; Electronic cigarette oil|||contains|||cordyceps sinensis C502 Electronic cigarette atomizer|||including||atomization unit; cigarette cartridge atomizer|||having|||T-shaped airway; electronic cigarette atomizer|||including|||mouthpiece device; electronic cigarette atomization core|||including|||oil-conducting non-woven fabric; electronic cigarette atomization core|||including|||electronic cigarette oil tank C561 The controller|||receives|||atomizer; the controller|||is used for|||electrical signal; the second controller|||has|||a vapor supply system; the controller|||is in|||working mode; the controller|||includes|||a storage module C35 Atomizer|||with||suction nozzle;front end|||with||suction nozzle;capsule|||with||suction nozzle;sliding cover|||covering|||suction nozzle;suction nozzle|||with|||suction opening It can be seen that technology cluster C55 has the highest centrality and cluster size, and is located at the core of the entire technology cluster network. At the same time, the smaller cluster density indicates that the network is sparse, which is consistent with the characteristics of emerging technologies. Based on the representative SAO triple information of technology cluster C55, the SAO triple information and patent literature are read to supplement the information, and the technology cluster is summarized as an electronic cigarette atomizer. The SAO information includes atomizers and related components, smoke channels, ultrasonic oscillators, etc.

[0079] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0080] Example 2 This embodiment is different from Embodiment 1 in that: a technology identification method based on scientific and technological information is provided.

[0081] Specifically, Figure 3 As shown, the technology identification method based on scientific and technological information, in addition to the steps of the key core technology identification method based on scientific and technological information described in Example 1, also includes: Step S4, after obtaining the technology cluster, calculating the temporal novelty, innovative novelty, growth, technological impact, and social impact of the technology cluster based on the bibliographic data of each text data in each technology cluster; Among them, the time novelty is used to characterize the average disclosure time of the scientific and technological achievements corresponding to the technology cluster; the innovation novelty is used to characterize the total number of citations of the scientific and technological achievements corresponding to the technology cluster; the growth is used to characterize the average growth rate of the scientific and technological achievements corresponding to the technology cluster; the technical impact is used to characterize the technical citations of the scientific and technological achievements corresponding to the technology cluster; the social impact is used to characterize the total number of R&D personnel of the scientific and technological achievements corresponding to the technology cluster; the degree of emergence of the technology cluster is obtained based on the time novelty, innovation novelty, growth, technical impact and social impact of the technology cluster; Step S5, calculating the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster based on the SAO triples corresponding to each text data in each technology cluster; Step S6, obtaining the critical core degree of the technology cluster based on the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster.

[0082] It is understandable that emerging technology (ET) was first proposed by the Emerging Technology Management Research Group of the Wharton School of the University of Pennsylvania. It refers to relatively fast-growing fundamental innovative technologies generated in the process of knowledge production, which have the potential to influence future economic and social development. Emerging technology is the source of power for technological renewal and innovation, which requires not only being "new" in time or space, but also "emerging" and "developing".

[0083] According to the research of domestic and foreign scholars, novelty, growth and impact are the key features for identifying emerging technologies. Therefore, this embodiment constructs the identification indicators of emerging technologies around these three features.

[0084] (1.1) Novelty Novelty is mainly considered from two dimensions: time and innovation, namely, temporal novelty and innovative novelty.

[0085] Among them, temporal novelty ( TN ) considers the time when the technology appeared. The later it appeared, the more it is an emerging technology. The calculation formula is shown in 6, where t i Indicates patents in technology clusters i The application year, n Represents the number of patents in a technology cluster.

[0086] (6) Innovation and novelty ( IN ) considers the content of the technology and the technical value reflected in the in-depth text content. The more patents a technology cites, the more knowledge it incorporates and the greater its innovation. The calculation formula is shown in 7, where C i Indicates patents in technology clusters i The number of cited patents, n Represents the number of patents in a technology cluster.

[0087] (7) (1.2) Growth "Emerging" represents the budding growth of new technologies, which grow faster than other technologies in the same field and are relative. The growth characteristics are mainly measured from the trend of technology clusters over time, using the average growth rate indicator. The calculation formula is shown in Figure 8, where N TIndicates the time period divided, which is the time period of the technology cluster j The number of patents is the number of patents in a technology cluster over time. j-1 The number of patents.

[0088] (8) (1.3) Impact The significant impact of emerging technologies is reflected in both technical and social aspects. They will not only affect future technological trends, but also affect the entire industry and even people's lifestyles. Therefore, this embodiment measures the impact characteristics from two aspects: technical and social impact.

[0089] The number of technology citations can reflect the impact of technology ( TI ), this embodiment uses relative citation counts for correction, that is, the number of patent citations divided by the average number of citations of all patents granted in the same year, to eliminate the impact of patent age on citation frequency counts. The calculation formula is shown in 9, where B i,t Indicates patents in technology clusters i The number of citations, represents the average number of citations of all patents granted in the same year, n Represents the number of patents in a technology cluster.

[0090] (9) The number of researchers conducting research on emerging technologies can reflect social impact (SI). This embodiment uses the number of technology inventors to measure this feature. The calculation formula is shown in 10, where N p represents the number of patent holders, n Represents the number of patents in a technology cluster.

[0091] (10) Step S3, and obtaining the degree of emergence of the technology cluster based on the temporal novelty, innovative novelty, growth, technological impact, and social impact of the technology cluster.

[0092] Specifically, according to the first preset weight allocation rule, the temporal novelty, innovative novelty, growth, technological impact, and social impact of the patent are weighted and integrated to obtain the emerging score of the technology cluster.

[0093] In the specific implementation, in order to facilitate the comprehensive comparison of indicators and emerging scores, this embodiment performs a 0-1 standardization process on the above five indicators, and weights the indicators according to the dimension level, that is, obtains the first preset weight allocation rule, and then obtains the patent emerging score EScore1 according to the first preset weight allocation rule, and the formula is as follows: (11) in, a 1 、a 2 、a 3 、a 4 、a 5 The weight is reset for the first preset weight allocation rule.

[0094] It can be understood that the emerging score is the degree of emergence of the technology cluster. The higher the emerging score, the higher the degree of emergence of the technology cluster.

[0095] Of course, in other embodiments, after the emerging scores of each technology cluster are obtained, the emerging levels of the technology clusters are obtained according to the emerging scores of each technology cluster.

[0096] For example, the emerging scores are divided into multiple intervals in descending order, where the interval with the highest score is the first emerging level, and the second interval is the second emerging level. Then, the emerging level is determined by judging which interval the emerging scores of each technology cluster fall into.

[0097] Verification: Based on the emerging technology identification scheme in the previous part, the domestic patent technology of e-cigarettes in the tobacco field in the past 10 years is used as the verification set. The emerging technologies in e-cigarette technology are calculated based on the numerical values ​​of 5 indicators in the three dimensions of novelty, growth and influence, as shown in Table 4. Table 4 lists the 10 technology clusters with the highest scores.

[0098] It should be noted that after 0-1 standardization, the score range of each indicator is [0,1]. The closer the score is to 1, the higher the characteristics of the technology are compared with other technologies.

[0099] Table 4 Top 10 emerging technology clusters by score cluster Temporal novelty (TN) Innovation Novelty (IN) Growth (R) Technology Impact (TI) Social Impact (SI) Emerging Score (EScore) C55 0.627 1.000 0.588 1.000 1.000 2.402 C445 0.793 0.003 0.997 0.006 0.007 1.402 C434 0.635 0.008 0.855 0.008 0.007 1.184 C142 0.733 0.021 0.631 0.017 0.031 1.032 C625 0.739 0.051 0.569 0.056 0.050 1.017 C394 0.637 0.015 0.636 0.033 0.030 0.994 C35 0.627 0.049 0.577 0.046 0.066 0.971 C35 0.607 0.130 0.461 0.132 0.141 0.966 It can be seen that technology cluster C55 has the highest emerging score, with the highest scores in terms of innovation impact, technical impact and social impact. It also highlights its emerging characteristics in the dimension of time novelty, and its score in terms of growth is at a medium level, indicating that the development of this technology cluster has reached a certain stage, accumulated a lot of influence, and is currently receiving widespread attention, with a sustained and stable output of patents. Technology cluster C2 has a relatively fast growth feature.

[0100] Table 5 lists the 5 representative SAO triple information corresponding to these 10 technology clusters. Among them, the representative triples are selected according to the in-degree and occurrence frequency of the SAO nodes to assist in the positioning of all triple information.

[0101] Table 5 Representative SAO triples of the top 10 emerging scoring technology clusters cluster Represents SAO triple C55 Atomizer|||Including|||Liquid storage component; Atomizer|||Including|||Base assembly; Atomizer|||Including|||Liquid storage assembly; Atomizer|||Including|||Atomizing element C445 Atomization carrier|||Damage|||Inside of electronic cigarette; Atomization liquid carrier|||Including|||Second contact part; Atomization medium|||Meet|||Safety standard; Oil-blocking body|||Adsorb|||Atomization medium C434 The cartridge|||has|||a second part; the winding|||includes|||a third part; the first part|||includes|||a first material; the second part|||exposes|||a storage compartment; the heating core|||includes|||a second part C142 Aerosol generator|||including|||surface acoustic wave nebulizer; aerosol generator|||including|||surface acoustic wave nebulizer; ceramic nebulizer|||including|||heat-generating medium; glass nebulizer|||having|||smooth surface; surface acoustic wave nebulizer|||including|||active surface C625 Aerosol generator|||including|||gas-electric hybrid joint;Aerosol generator|||including|||gas-electric hybrid joint;Solvent reflux capture device|||increase|||substance concentration;Solvent reflux capture device|||save|||solvent usage C394 Connecting hole|||Return|||Oil-blocking silicone;Connecting hole|||Return|||Oil-blocking silicone;Atomizing upper seat|||Located at|||Horizontal through hole;Connecting hole|||Higher than|||Condensation chamber;E-liquid bottle|||Has|||First oil storage chamber C35 Atomizer|||with||suction nozzle;front end|||with||suction nozzle;capsule|||with||suction nozzle;sliding cover|||covering|||suction nozzle;suction nozzle|||with|||suction opening C35 Atomizer|||convenient|||smoke oil; atomizer|||appears|||leakage; oil storage bottle|||contains|||smoke oil; oil guide|||absorbs|||smoke oil; oil reservoir|||transports|||smoke oil C41 The cartridge|||reduces|||the oil; the cartridge|||includes|||the airway; the cartridge|||has|||seal; the cartridge|||includes|||the cartridge bin; the cartridge|||includes|||the oil tank C257 Atomizer|||Including|||Atomizing shell; Single chip microcomputer|||Adjustment|||Atomizing wire; Atomizing oil|||Infiltration|||Oil absorbing cotton; Atomizing sheet|||Including|||Third side; Air energy|||Passing|||Atomizing sheet Based on the representative SAO triplet information positioning of technology cluster C55, the SAO triplet information and patent literature were read to supplement the information, and the technology cluster was summarized as electronic cigarette atomization technology. The SAO information includes the atomizer covering multiple key components such as liquid storage parts, base components, liquid delivery elements, sensor elements, smoke channels, ultrasonic oscillation plates, etc.

[0102] According to the representative SAO triple information of technology cluster C445, it has high temporal novelty and growth rate, and is a technology point that has received attention recently, so its technical impact and social impact are still relatively low. After reading the SAO triple information and patent literature for information supplement, the technology cluster is summarized as solid atomization medium optimization technology. SAO information includes the structure and material of solid atomization medium, as well as storage space design.

[0103] Example 3 The difference between this embodiment and embodiment 1 is that this embodiment mainly constructs a scientific and technological information data set based on domestic and foreign paper data of the tobacco industry, wherein the paper data at least includes paper bibliographic information. Specifically, the bibliographic information refers to the basic information of the document, for example, the basic information of the paper includes publication time, title, author, keywords, author organization, document source and abstract, etc.

[0104] Specifically, Figure 1 As shown, the key technology identification method based on scientific and technological information includes: Step S1, performing technical association mining on each paper data in the tobacco field scientific and technological information data set to obtain at least one technology cluster.

[0105] It should be noted that paper abstracts are usually filled with a lot of background information, and because they contain multiple lengthy sentences, the SAO structure is difficult to understand intuitively. This not only hinders the accurate calculation of SAO structural similarity, but also affects the accuracy of technology cluster division.

[0106] To solve this problem, this embodiment extracts keywords from the front part of the paper, and automatically extracts paper keywords from the paper abstract as a supplement, thereby constructing a keyword vocabulary for the technical field, namely, the field paper keyword vocabulary.

[0107] like Figure 2 As shown, the specific steps of step S1 are: The SAO triples of the paper text are extracted by using text mining methods, and SAO screening is performed based on the paper keywords to obtain the SAO triples of each paper; among which, the paper keywords are the paper keywords in the field paper keyword vocabulary; Table 6 shows the SAO information after paper screening. It can be seen that the SAO structure becomes more concise and clear after keyword filtering. This not only retains the key technical points to a large extent, but also clearly shows the relationship between the technical points.

[0108] Table 6 SAO information after paper screening Paper ID Title SAO Field 0 Synthesis and Antibacterial Activity of Symmetrical Gemini Quaternary Ammonium Salts Containing Hydronopyl [["Methylhydrogenolpylamine","synthesis","gemini quaternary ammonium salt"],["Methylhydrogenolpylamine","synthesis","symmetric"], ["ethyl","synthesis","symmetric"], ["diethylhydrogenolpylamine","synthesis","symmetric"], ["hydrogenolpyl","synthesis","gemini quaternary ammonium salt"],["compound","has","inhibitory effect"], ["compound","is","ethyl"], ["compound","contains","ethyl"]] 1 Hyperspectral inversion model of pectin under salt and physical damage stress in wheat [["Physical damage", "become","wheat yield"], ["Physical damage", "become", "wheat"],["Physical damage", "lead to", "nutrients"], ["Pectin","composition", "main component of plant cell wall"], ["Linear regression", "establish", "pectin"]] 2 Screening of compound biocontrol bacteria against tobacco gray mold [["Preventive and control effect", "belongs to", "Pseudomonas"],["Strain","belongs to", "Pseudomonas"]] The SAO triples corresponding to each paper are semantically represented based on the Bert model, and the synonyms of SAO triples are merged based on KMeans clustering; A technology association network is constructed based on the association relationship of SAO triplets, and technology clusters are divided based on the graph theory community detection method.

[0109] Then, at least one technology cluster in the target field is obtained based on each SAO triple, specifically including: The SAO triples corresponding to each text data are semantically represented based on the Bert model, and the synonyms of SAO triples are merged based on KMeans clustering; A technology association network is constructed based on the association relationship of SAO triplets, and technology clusters are divided based on the graph theory community detection method.

[0110] Step S2, calculating the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster based on the SAO triples corresponding to each text data in each technology cluster.

[0111] Step S3, obtaining the critical core degree of the technology cluster based on the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster.

[0112] The specific contents of step S2 and step S3 can be referred to in Example 1, and will not be described in detail here.

[0113] Verification: Specifically, foreign papers in the tobacco field in the past 10 years (including conference papers and journal papers) were used as the verification data set for verification. Table 2 lists the 10 technology clusters with the highest scores.

[0114] Table 7 Top 10 technology clusters by key core scores of papers cluster Cluster Density Cluster Centrality Cluster size normalized score Cluster coreness score C16 0.002 1.000 1.000 2.000 C62 0.007 0.171 0.197 0.368 C462 0.009 0.182 0.175 0.357 C23 0.012 0.149 0.111 0.260 C4 0.016 0.070 0.083 0.154 C756 0.023 0.053 0.059 0.113 C226 0.033 0.039 0.041 0.080 C386 0.042 0.030 0.033 0.063 C625 0.044 0.029 0.031 0.061 C188 0.040 0.026 0.031 0.058 It can be seen that technology cluster C16 has the highest centrality and cluster size, and is located at the core of the entire research topic cluster network; the smaller cluster density indicates that the network is sparse and belongs to a new cutting-edge technology.

[0115] Similarly, if Figure 3 As shown, this embodiment also identifies emerging technologies based on scientific and technological information.

[0116] Step S4, based on at least one paper text in each technology cluster, obtain the temporal novelty, growth, and social impact of the paper.

[0117] Due to the limitations of citation and cited data in paper data, this embodiment retains the attributes of emerging technologies in the three dimensions of novelty, growth and impact when analyzing paper data, and reduces the number of indicators, that is, evaluating novelty attributes from the perspective of temporal novelty, growth rate to evaluate growth attributes, and social impact to evaluate impact attributes. Specifically, the temporal novelty, growth, and social impact of the paper refer to the aforementioned method for obtaining the temporal novelty, innovative novelty, growth, technological impact, and social impact of the patent. It only requires that the information related to the patent be modified to the information related to the paper, such as modifying the patent application time to the paper publication time, and modifying the patent inventor to the paper author, etc., which will not be repeated here.

[0118] Step S5, obtaining the degree of emergence of the technology cluster based on the temporal novelty, growth, and social impact of the technology cluster.

[0119] The specific steps are: according to the second preset weight allocation rule, the temporal novelty, growth, and social impact of the paper are weighted and integrated to obtain the emerging score of the technology cluster.

[0120] Similarly, in order to facilitate comprehensive comparison of indicators and emerging scores, this embodiment performs a 0-1 standardization process on the above-mentioned time novelty, growth, and social impact, and weights the indicators according to the dimension level, that is, obtains the second preset weight allocation rule, and then according to the second preset weight allocation rule, weights the time novelty, growth, and social impact of the paper to obtain the emerging score EScore2 of the technology cluster. The formula is as follows: .

[0121] in, b 1 、b 2 、b 3 The weight is reset for the second preset weight allocation rule.

[0122] Verification: Based on the emerging technology identification scheme in the previous part, domestic papers (including journals and conference papers) in the tobacco field in the past 10 years were used as the verification data set for verification. Table 8 lists the 10 technology clusters with the highest scores.

[0123] Table 8 Top 10 emerging technology clusters by score cluster Novelty (TN) Growth (R) Impact (SI) Emerging Score (EScore) C16 0.555 0.284 1.000 1.839 C1222 0.613 0.972 0.031 1.616 C635 0.566 1.000 0.013 1.579 C1677 0.923 0.607 0.004 1.534 C62 0.560 0.272 0.616 1.447 C33 0.633 0.779 0.012 1.424 C882 0.577 0.807 0.011 1.395 C2252 0.657 0.714 0.010 1.380 C154 0.638 0.717 0.022 1.377 C222 0.643 0.723 0.004 1.370 Similarly, after 0-1 standardization, the score range of each indicator is [0,1]. The closer the score is to 1, the higher the characteristics of the technology are compared with other technologies.

[0124] Table 9 lists the 5 representative SAO triple information corresponding to these 10 technology clusters. Among them, the representative triples are selected according to the in-degree and occurrence frequency of the SAO nodes to assist in the positioning of all triple information.

[0125] It can be seen that technology cluster C16 has the highest emerging score, the highest score in terms of technological impact, and also highlights its emerging characteristics in the dimension of time novelty, but its emerging characteristics are not obvious in the dimension of growth, indicating that this technology cluster has more papers output under the condition of more social attention. This is also consistent with the fact that technology cluster 16 has a smaller cluster density in the aforementioned identification of key core technologies.

[0126] Example 4 Based on the same inventive concept, the embodiment of the present application also provides a key core technology identification device based on scientific and technological information for implementing the key core technology identification method based on scientific and technological information involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more embodiments of the key core technology identification device based on scientific and technological information provided below can refer to the limitations of the key core technology identification method based on scientific and technological information in the above embodiment 1, and will not be repeated here.

[0127] Specifically, the key core technology identification device based on scientific and technological information is as follows: Figure 4 As shown, including: A technology cluster acquisition module is used to acquire SAO triples corresponding to each text data in the target field science and technology information data set, and acquire at least one technology cluster in the target field based on each SAO triple; wherein the science and technology information data set includes text data corresponding to the scientific and technological achievements disclosed in the target field; A technology cluster indicator acquisition module, used to calculate the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster based on the SAO triple corresponding to each text data in each technology cluster; The technology cluster key core degree acquisition module is used to acquire the key core degree of the technology cluster based on the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster.

[0128] Example 5 Based on the above embodiment, this embodiment provides a computer device, which may be a terminal, and its internal structure diagram may be as follows: Figure 5As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a key core technology identification method based on scientific and technological information described in Example 1, or a technology identification method based on scientific and technological information described in Example 2, is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0129] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0130] Example 6 Based on the above embodiments, this embodiment provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, it implements a key core technology identification method based on scientific and technological information as described in Example 1 or a technology identification method based on scientific and technological information as described in Example 2.

[0131] Example 7 Based on the above embodiments, this embodiment provides a computer program product, including a computer program, which, when executed by a processor, implements a key core technology identification method based on scientific and technological information as described in Example 1 or a technology identification method based on scientific and technological information as described in Example 2.

[0132] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0133] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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, they should be considered to be within the scope of this specification.

[0134] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

[0135] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or some technical features can be replaced by equivalents without departing from the spirit of the technical solution of the present invention, which should be included in the scope of the technical solution for protection of the present invention.

Claims

1. A key core technology identification method based on scientific and technological information, characterized in that: The following steps are involved: Acquire SAO triples corresponding to each text data in the target field science and technology information data set, and acquire at least one technology cluster in the target field based on each SAO triple; wherein the science and technology information data set includes text data corresponding to the scientific and technological achievements disclosed in the target field; Calculate the in-degree centrality, out-degree centrality, link strength and network density of each technology cluster based on the SAO triple corresponding to each text data in each technology cluster; The key core degree of the technology cluster is obtained based on the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster.

2. According to claim 1, a key core technology identification method based on scientific and technological information is characterized in that: Based on the SAO triples of each technology cluster, the in-degree centrality, out-degree centrality, link strength and network density of each technology cluster are calculated, including: Obtaining the in-degree centrality of the technology cluster based on the in-degree centrality of each node and the number of nodes in the SAO triple of the technology cluster; wherein the in-degree centrality is used to characterize the importance of the technology cluster; The out-degree centrality of the technology cluster is obtained based on the out-degree centrality of each node and the number of nodes in the SAO triple of the technology cluster; wherein the out-degree centrality is used to characterize the sociability of the technology cluster; Based on the number of other nodes connected to each node in the SAO triplet of the technology cluster, the link strength of the technology cluster is obtained; wherein the link strength is used to characterize the number of links of each node in the technology cluster; Based on the connection relationship between each node and other nodes in the SAO triple of the technology cluster, the network density of the technology cluster is obtained; wherein the network density is used to characterize the closeness between each node and other nodes in the technology cluster.

3. A key core technology identification method based on scientific and technological information according to claim 1 or 2, characterized in that: The key core degree of the technology cluster is obtained based on the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster, including: Based on the preset contribution weights of in-degree centrality, out-degree centrality, link strength and network density, the in-degree centrality, out-degree centrality, link strength and network density are weighted and fused respectively to obtain the key core score of the technology cluster.

4. The key core technology identification method based on scientific and technological information according to claim 3 is characterized in that: Obtain the SAO triples corresponding to each text data in the target field science and technology information dataset, including: When the target field scientific and technological information dataset is a patent text, the CRF model is used to extract the SAO triples of the patent text, and the SAO screening is performed based on the patent keywords generated by the large language model to obtain the SAO triples of each patent; When the target field scientific and technological information dataset is a paper text, a text mining method is used to extract SAO triples of the paper text, and SAO screening is performed based on paper keywords to obtain SAO triples of each paper; wherein the paper keywords are the paper keywords in the field paper keyword vocabulary, and the field paper keyword vocabulary is constructed by extracting the paper keywords in the paper text bibliography and extracting the paper keywords in the paper abstract.

5. The key core technology identification method based on scientific and technological information according to claim 4 is characterized in that: At least one technology cluster in the target field is obtained based on each SAO triple, including: The SAO triples corresponding to each text data are semantically represented based on the Bert model, and the synonyms of SAO triples are merged based on KMeans clustering; A technology association network is constructed based on the association relationship of SAO triplets, and technology clusters are divided based on the graph theory community detection method.

6. A technology identification method based on scientific and technological information, characterized in that: The following steps are involved: Acquire SAO triples corresponding to each text data in the target field science and technology information data set, and acquire at least one technology cluster in the target field based on each SAO triple; wherein the science and technology information data set includes text data corresponding to the scientific and technological achievements disclosed in the target field; Calculate the temporal novelty, innovative novelty, growth, technological impact, and social impact of the technology cluster based on the bibliographic data of each text data in each technology cluster; Among them, the time novelty is used to characterize the average disclosure time of the scientific and technological achievements corresponding to the technology cluster; the innovation novelty is used to characterize the total number of citations of the scientific and technological achievements corresponding to the technology cluster; the growth is used to characterize the average growth rate of the scientific and technological achievements corresponding to the technology cluster; the technical impact is used to characterize the technical citations of the scientific and technological achievements corresponding to the technology cluster; the social impact is used to characterize the total number of R&D personnel of the scientific and technological achievements corresponding to the technology cluster; the degree of emergence of the technology cluster is obtained based on the time novelty, innovation novelty, growth, technical impact and social impact of the technology cluster; Calculate the in-degree centrality, out-degree centrality, link strength and network density of each technology cluster based on the SAO triple corresponding to each text data in each technology cluster; The key core degree of the technology cluster is obtained based on the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster.

7. A key core technology identification device based on scientific and technological information, characterized in that: include: A technology cluster acquisition module is used to acquire SAO triples corresponding to each text data in the target field science and technology information data set, and acquire at least one technology cluster in the target field based on each SAO triple; wherein the science and technology information data set includes text data corresponding to the scientific and technological achievements disclosed in the target field; A technology cluster indicator acquisition module, used to calculate the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster based on the SAO triple corresponding to each text data in each technology cluster; The technology cluster key core degree acquisition module is used to acquire the key core degree of the technology cluster based on the in-degree centrality, out-degree centrality, link strength and network density of the technology cluster.

8. A computer device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; A processor, when used to execute a program stored in a memory, implements the key core technology identification method based on scientific and technological information as described in any one of claims 1 to 5, or implements the technology identification method based on scientific and technological information as described in claim 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the key core technology identification method based on scientific and technological information as described in any one of claims 1 to 5, or implements the technology identification method based on scientific and technological information as described in claim 6.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, it implements the key core technology identification method based on scientific and technological information as described in any one of claims 1 to 5, or implements the technology identification method based on scientific and technological information as described in claim 6.

Images