Electrochemical energy storage technology identification method and device based on patent text analysis

Abstract

The invention provides an electrochemical energy storage technology identification method and device based on patent text analysis, and relates to the technical field of data processing.The method comprises the steps that a patent text database is obtained, and the patent text database comprises patent texts related to electrochemical energy storage; through a Dirichlet distribution topic model, topic distribution corresponding to the patent texts is obtained, and topics included in the patent texts serve as technical features of electrochemical energy storage; obtaining the number of applications of the patent texts in a plurality of patent institutions, determining the average marginal effect of each technical feature based on the subject distribution and the number of applications of the patent texts, and determining a target technical feature based on the average marginal effect; and clustering the target technical features to obtain an electrochemical energy storage technology identification result, the electrochemical energy storage technology identification result comprising a technical theme obtained by clustering the target technical features. And objectivity and timeliness of an electrochemical energy storage technology identification result can be improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an electrochemical energy storage technology identification method and device based on patent text analysis. Background Art

[0002] Although new energy sources such as photovoltaics and wind power can reduce carbon emissions, their intermittent and instability pose challenges to the power system. Electrochemical energy storage technology can effectively store and release energy to ensure the stable operation of the power system. Although electrochemical energy storage technology has developed rapidly, it still faces the problem of uncertainty in technical route decisions. Therefore, establishing an effective electrochemical energy storage technology identification method can screen a large number of electrochemical energy storage technologies and provide scientific decision-making basis for enterprises and research institutions.

[0003] In the existing technology, the identification method of electrochemical energy storage technology mainly adopts qualitative analysis methods. This kind of method relies on the experience and knowledge of experts, and uses tools such as Delphi method, hierarchical analysis method, scenario analysis method, etc. to identify and evaluate key technologies. First, a team of experts in the field is formed to design an evaluation index system, and then expert opinions are collected through questionnaires or seminars. Finally, conclusions are drawn based on the comprehensive expert scores. Due to the complete reliance on the expert's personal knowledge reserves and experience judgment, in an environment of rapid technological iteration, it is often difficult for experts to grasp the latest technological trends in a timely and comprehensive manner, resulting in poor objectivity and timeliness of the evaluation results. Summary of the invention

[0004] The present invention provides an electrochemical energy storage technology identification method and device based on patent text analysis, so as to solve the defects of poor objectivity and timeliness of the results of electrochemical energy storage technology identification using qualitative analysis methods in the prior art, and to improve the objectivity and timeliness of the electrochemical energy storage technology identification results.

[0005] The present invention provides an electrochemical energy storage technology identification method based on patent text analysis, comprising: Obtaining a patent text database, wherein the patent text database includes patent texts related to electrochemical energy storage; The topic distribution corresponding to the patent text is obtained by using the Dirichlet distribution topic model, and the topics included in each of the patent texts are used as technical features of electrochemical energy storage; Obtain the number of applications for the patent text in multiple patent institutions, determine the average marginal effect of each of the technical features based on the subject distribution of the patent text and the number of applications, and determine the target technical feature based on the average marginal effect; The target technical features are clustered to obtain an electrochemical energy storage technology identification result, wherein the electrochemical energy storage technology identification result includes a technical theme obtained by clustering the target technical features.

[0006] According to a method for identifying electrochemical energy storage technology based on patent text analysis provided by the present invention, the topic distribution corresponding to the patent text is obtained by using the Dirichlet distribution topic model, including: The patent text is sampled and labeled using the Dirichlet distribution topic model, wherein the label is a Derwent manual code; Sampling the topic distribution of each of the labels through the Dirichlet distribution topic model to obtain the topic distribution corresponding to each topic and the label of the patent text; The Dirichlet distribution topic model is used to sample the topic word distribution for each of the topics to obtain the topic word distribution corresponding to each topic word and the topic of the patent text.

[0007] According to a method for identifying electrochemical energy storage technology based on patent text analysis provided by the present invention, the average marginal effect of each of the technical features is determined based on the subject distribution of the patent text and the number of applications, including: Constructing a fitting relationship between the technical features in the patent text and the number of applications through an ordered logistic regression model; The average marginal effect of each of the technical features in each of the patent texts is calculated based on the fitting relationship.

[0008] According to a method for identifying electrochemical energy storage technology based on patent text analysis provided by the present invention, the target technology features are determined based on the average marginal effect, including: The technical feature whose average marginal effect in the patent text with the lowest number of applications is lower than a preset threshold is selected as the target technical feature.

[0009] According to a method for identifying electrochemical energy storage technology based on patent text analysis provided by the present invention, clustering the target technology features to obtain electrochemical energy storage technology identification results includes: Obtaining a similarity matrix corresponding to the target technical features, wherein the similarity matrix includes similarities between the target technical features; Based on the similarity matrix, a clustering result is determined based on a neighbor propagation algorithm, and each cluster in the clustering result is used as a technical theme; The electrochemical energy storage technology identification result is determined based on the technical subject.

[0010] According to a method for identifying electrochemical energy storage technology based on patent text analysis provided by the present invention, after determining the electrochemical energy storage technology identification result based on the technical subject, the method comprises: Performing a technology life cycle analysis on the technical features included in the technical subject to obtain the technical maturity of each of the technical features in the technical subject; Based on the average value of the technical maturity of each of the technical features included in the technical subject and each of the technical features included in the technical subject, an evaluation result of each of the technical subjects in the electrochemical energy storage technology identification result is generated.

[0011] The present invention also provides an electrochemical energy storage technology identification device based on patent text analysis, comprising: A patent text acquisition module, used to acquire a patent text database, wherein the patent text database includes patent texts related to electrochemical energy storage; A text topic mining module, used to obtain the topic distribution corresponding to the patent text through the Dirichlet distribution topic model, and use the topics included in each of the patent texts as the technical features of electrochemical energy storage; A technical feature extraction module is used to obtain the number of applications of the patent text in multiple patent institutions, determine the average marginal effect of each of the technical features based on the subject distribution of the patent text and the number of applications, and determine the target technical feature based on the average marginal effect; The technical feature clustering module is used to cluster the target technical features to obtain an electrochemical energy storage technology identification result, wherein the electrochemical energy storage technology identification result includes a technical theme obtained by clustering the target technical features.

[0012] The present invention also provides an electronic 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 electrochemical energy storage technology identification method based on patent text analysis as described in any one of the above is implemented.

[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described electrochemical energy storage technology identification methods based on patent text analysis.

[0014] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned electrochemical energy storage technology identification methods based on patent text analysis.

[0015] The present invention provides an electrochemical energy storage technology identification method and device based on patent text analysis, the method comprising: obtaining a patent text database, the patent text database including patent texts related to electrochemical energy storage; obtaining a topic distribution corresponding to the patent text through a Dirichlet distribution topic model, and using the topics included in each patent text as a technical feature of electrochemical energy storage; obtaining the number of applications for the patent text in multiple patent institutions, determining the average marginal effect of each technical feature based on the topic distribution and the number of applications of the patent text, and determining the target technical feature based on the average marginal effect; clustering the target technical feature to obtain an electrochemical energy storage technology identification result, the electrochemical energy storage technology identification result including a technical theme obtained by clustering the target technical feature.

[0016] In this way, by analyzing the patent texts related to electrochemical energy storage, using the Dirichlet distribution topic model, we can mine the technical information of the patent texts and extract the technical features. Then, in the face of the inconsistent content level of the recognition results, we select the target technical features based on the size of the patent family to eliminate the technical features with broad content or low value, and finally cluster the selected target technical features into technical themes. Since patent applications are forward-looking, the recognition of electrochemical energy storage technology based on a large amount of patent text analysis can improve the objectivity and timeliness of the recognition results and provide decision makers with accurate reference information. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a flow chart of the electrochemical energy storage technology identification method based on patent text analysis provided by the present invention.

[0019] Figure 2 It is a probability graph of the Dirichlet distribution topic model in the electrochemical energy storage technology identification method based on patent text analysis provided by the present invention.

[0020] Figure 3 It is a structural schematic diagram of the electrochemical energy storage technology identification device based on patent text analysis provided by the present invention.

[0021] Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] Combine the following Figure 1-Figure 2 The present invention describes the electrochemical energy storage technology identification method based on patent text analysis. Figure 1 As shown, the electrochemical energy storage technology identification method based on patent text analysis includes the following steps: S110, obtaining a patent text database, where the patent text database includes patent texts related to electrochemical energy storage; S120, obtaining the topic distribution corresponding to the patent text through the Dirichlet distribution topic model, and taking the topics included in each patent text as the technical features of electrochemical energy storage; S130. Obtain the number of applications for patent texts in multiple patent institutions, determine the average marginal effect of each technical feature based on the subject distribution and application number of the patent text, and determine the target technical feature based on the average marginal effect; S140, clustering the target technical features to obtain an electrochemical energy storage technology identification result, wherein the electrochemical energy storage technology identification result includes a technical theme obtained by clustering the target technical features.

[0024] The electrochemical energy storage technology identification method based on patent text analysis provided by the present invention analyzes patent texts related to electrochemical energy storage, uses the Dirichlet distribution topic model to mine the technical information of the patent text and extract technical features, and then faces the problem of inconsistent content levels of the identification results, selects target technical features based on the size of the patent family to eliminate technical features with broad content or low value, and finally clusters the selected target technical features into technical themes. Since patent applications are forward-looking, the identification of electrochemical energy storage technology based on a large amount of patent text analysis can improve the objectivity and timeliness of the identification results and provide decision makers with accurate reference information.

[0025] By searching and collecting patents related to electrochemical energy storage in a patent database, a patent text database used in the method provided by the present invention is constructed. In a possible implementation, patents related to electrochemical energy storage can be searched in the Derwent Patent Intelligence Database (DerwentInnovations Index) to obtain a patent text database used in the method provided by the present invention. The Derwent Patent Intelligence Database is a global patent intelligence and scientific and technological intelligence agency that can more comprehensively reflect the development trends of international science and technology. The Derwent database provides detailed classification information, including patent families, Derwent classification codes and manual codes, which are more detailed than conventional patent classification systems (such as the International Patent Classification IPC). Collect patent text data related to "electrochemical energy storage" and collect patent data since the 21st century using "DC=(X16)" as the search formula. After preprocessing, the data is formed into a high-quality patent text database by fields (including patent titles, abstracts, Derwent manual codes, application details, patent family information, etc.).

[0026] Through the Dirichlet distribution topic model, the patent text in the patent text database is mined for technical information. The Derwent manual code can effectively index a patent. In the method provided by the present invention, the Derwent manual code is embedded into the Dirichlet distribution topic model to constrain the model output results and obtain directional label information. Specifically, through the Dirichlet distribution topic model, the topic distribution corresponding to the patent text is obtained, including: The Dirichlet distribution topic model is used to sample the label distribution of patent texts, and the labels are Derwent's manual codes; The Dirichlet distribution topic model is used to sample the topic distribution of each label to obtain the topic distribution corresponding to each topic and the label of the patent text; The Dirichlet distribution topic model is used to sample the topic word distribution for each topic, and the topic word distribution corresponding to each topic word and the topic of the patent text is obtained.

[0027] In the method provided by the present invention, a label dimension is introduced into the hierarchical structure of document-topic-vocabulary in the Dirichlet distribution topic model to form a four-level structure of document-label-topic-vocabulary.

[0028] The specific process of forming the four-level structure of document-label-topic-word is as follows: (1) For each patent text , sampling label distribution ,in Is the length The vector and The elements are ; (2) For each label , sampled topic distribution ; (3) For each topic , sampling keyword distribution

[0029] (4) For each document Each word in The generation of Sampling a label , and then from the topic distribution Sampling a potential topic , and finally from the distribution of subject words Sample words .

[0030] The above process is as follows Figure 2 As shown, the black nodes represent nodes whose values ​​can be directly observed. Through the method provided by the present invention, the joint distribution of documents and topics can be obtained. And the distribution of subject terms , estimated by the Markov Chain Monte Carlo sampling method.

[0031] In the above process and Figure 2 In the above equation, D represents the number of patent texts in the patent text database, d represents one of the patent texts, represents the number of words in the patent text d, L represents the number of all tags in the patent text database, For one of the tags, represents all the tags that document d has, represents the label distribution of document d, Representation Tags The number of potential topics, Indicates the document tag The topic distribution of Representation Tags The distribution of keywords for the kth topic, Represents topic distribution The hyperparameters of the prior Dirichlet distribution, Represents the distribution of subject words The hyperparameters of the prior Dirichlet distribution, represents the observed words, z represents each observed word potential themes.

[0032] The method provided by the present invention uses label information to control each patent text to extract only the topics associated with the label it has, and uses the patent classification system as label information. It can be different from traditional unsupervised text mining technology, and the topics mined from the patent text are limited to a specific technical scope, so that the output topic results contain both the technical field information of the patent classification label and the technical content of the data segmentation technology field (electrochemical energy storage technology) that is more detailed and more in line with the patent classification label at a finer level. Patent classification information provides a macro-technical field framework, limits the scope of analysis, and reduces noise interference; text semantic information mines fine-grained technical features in patents and captures deep semantic associations. The combination of the two realizes multi-level technical feature extraction from macro to micro, which not only improves the accuracy of technical information mining, but also enhances the ability to identify cross-domain technologies and adapt to complex technology ecosystems.

[0033] The topic output by the model is used as a technical feature, and its specific technical information is explained through its subject word distribution. At the same time, the method provided by the present invention represents each patent text as a document-topic distribution (i.e., the proportion of each technical feature in the document, and the sum of the proportion of each topic in all documents is called the popularity of the topic), thereby obtaining the technical feature vector of each patent.

[0034] Generally speaking, patents that are deployed in multiple patent institutions have higher technical value. In the method provided by the present invention, the number of applications for patent texts in multiple patent institutions is obtained, and based on the subject distribution and application number of the patent texts, the average marginal effect of each technical feature is determined, which specifically includes: The fitted relationship between the technical features in the patent text and the number of applications was constructed through an ordered logistic regression model; The average marginal effect of each technical feature in each patent text is calculated based on the fitting relationship.

[0035] Specifically, the number of applications from major patent application agencies (such as the State Intellectual Property Office of China, the European Patent Office, the Japan Patent Office, the Korean Intellectual Property Office, and the United States Patent and Trademark Office) with a large number of patent applications in the patent text database can be obtained. Taking five patent application agencies as an example, the number of applications for each patent in the patent text database in the five patent agencies is counted to obtain an integer ranging from 0 to 5: 0 is obtained if no application is made in any of the patent agencies, 1 is obtained if an application is made in one of the patent agencies, 2 is obtained if an application is made in two patent agencies, and so on. The obtained integers are encoded to obtain multiple categories, for example, 0 and 1 are encoded as category 0, 2 is encoded as category 1, 3 is encoded as category 2, and 4 and 5 are encoded as category 3. In this way, the patent technology value measurement index can be constructed as a four-category ordered component. The higher the value, the more patent application agencies the patent has been applied for, and the higher the technology value.

[0036] Based on the subject distribution of the patent text and the number of applications for the patent text, a fitting relationship between the technical features in the patent text and the number of applications is constructed. The construction of the fitting relationship can be constructed through an ordered logistic regression (OLR) model. In other words, the number of applications index = OLR (patent technical feature vector), and the relationship between the distribution of technical features of the patent text and the number of applications of the patent text can be expressed by the formula: ; in, represents the classification threshold of application quantity category j, represents the i-th item of the patent technical feature vector, M is the total number of technical features, express The regression coefficient of the corresponding technical feature, Y represents the patent technology value measurement index, and P represents the probability.

[0037] According to the results of the ordered logistic regression model fitting, the average marginal effect (AME) of each technical feature can be calculated. The process is as follows: ; ; Where N is the total number of patent texts, It is for Patent text Calculate the marginal effects. is the average marginal effect of the kth technical feature.

[0038] The marginal effect of a technical feature calculated for a certain patent text reflects the impact of the technical feature on the number of applications for the patent text (reflecting the technical value of the patent). The marginal effects of the technical feature calculated for all patent texts are averaged to obtain the average marginal effect of the technical feature.

[0039] Determine the target technology characteristics based on the average marginal effect, including: The technical features whose average marginal effects in the patent texts with the lowest number of applications are lower than the preset threshold are selected as target technical features.

[0040] Specifically, the average marginal effect of category 0 (representing the category with the lowest number of applications) can be selected as the basis for selecting the target technical features, that is, when the component of the selected technical features increases by a unit, the change value of the probability that the patent belongs to category 0 (that is, the patent technology value measurement index is 0 or 1). Obviously, when the value of a certain technical feature is negative, the higher the component of the technical feature contained in the patent, the more it can reduce the probability of belonging to category 0, that is, the probability of belonging to high technology value increases. Therefore, if the average marginal effect of the technical feature in the category representing the lowest technology value is negative, it indicates that it has a positive correlation with improving the technology value of the patent, and the larger the absolute value of the value, the greater the influence of the technical feature. Finally, all technical features with an average marginal effect less than 0 (at a significance level of 0.05) in the category representing the lowest technology value are selected as the key technical features (that is, the target technical features) of electrochemical energy storage technology, and the absolute value of the average marginal effect of these features is taken as the numerical basis for calculating the technology value in the subsequent period.

[0041] The technical value of patented technology is quantified by constructing a new technical value measurement index, and an influence model between technical value and technical features is established based on an ordered logistic regression model. The method provided by the present invention innovatively combines technical value measurement with statistical modeling, which can effectively identify key technical features that have a greater impact on technical value and provide a scientific basis for key technology screening.

[0042] The target technical features selected through the above steps have clear meanings and accurately point to technical content with high technical value. In order to support technical management decisions, the method provided by the present invention further clusters and condenses the target technical features into more representative meso-technical themes in order to refine electrochemical energy storage relationship technology.

[0043] Cluster the target technology features to obtain the electrochemical energy storage technology identification results, including: Obtaining a similarity matrix corresponding to the target technical features, wherein the similarity matrix includes similarities between the target technical features; Based on the similarity matrix, the clustering results are determined based on the nearest neighbor propagation algorithm, and each cluster in the clustering results is used as a technical theme; Electrochemical energy storage technology identification results are determined based on technology themes.

[0044] First, we construct a The similarity matrix ,in For the The top 15 key words of the project's technical features, each element Indicates technical features and The Jaccard coefficient of: ; Then, an Affinity Propagation (AP) algorithm based on the similarity matrix is ​​constructed to cluster the technical features. An iterative message passing mechanism is used to propagate the two pieces of information, “responsibility value” and “availability”, between data points to determine the cluster center of each data point. is the responsibility value, indicating the data point Consider data points as the fitness of its cluster center. is the availability, indicating the data point As a data point The fitness of the cluster centers is updated in the following iterations until convergence.

[0045] ; ; .

[0046] According to the final value, choose the largest As the cluster center, the neighboring points transmit information to each other, continuously iterate the responsibility matrix and the available matrix, and finally adaptively obtain the number of clusters and the cluster center according to the calculation results of the two matrices, so as to realize the clustering of technical features into meso-scale technical themes.

[0047] In patent theme clustering, the confirmation of theme number and clustering robustness have always been difficult problems. In the method provided by the present invention, the nearest neighbor propagation clustering algorithm is used, and the number of clusters does not need to be specified in advance. Instead, all data points are regarded as potential cluster centers, and clustering is achieved through message passing, thereby avoiding the subjectivity of the hyperparameter of the decision-making number of clusters. At the same time, the Jaccard coefficient is used instead of the Euclidean distance metric, which provides higher robustness.

[0048] The Jaccard coefficient of the technical feature theme keyword set is used as the similarity measure to construct a similarity matrix, and the number and center of clusters are adaptively determined by the nearest neighbor propagation algorithm to organize the technical features into moderately specific and strategically oriented technical themes. The method provided by the present invention does not need to preset the number of clusters, thus avoiding subjectivity, and has higher robustness when processing non-Euclidean spatial data, and can better adapt to the complexity and diversity of technical features. More representative meso-technical themes are extracted through clustering results to improve the interpretability and practicality of decision support.

[0049] Through the above clustering operation, the key technology themes are finally obtained. For each key technology theme, the specific contents of the top 15 theme keywords with the highest content in the theme are combined to identify the key electrochemical energy storage technologies represented by each technical theme.

[0050] In a possible implementation, the technical maturity of each technical topic can be further obtained, and the technical maturity of the technical topic and the number of applications corresponding to the technical topic (reflecting the technical value of the technical characteristics) can be combined to evaluate each technical topic to provide more reference information for technical decision makers. That is, after determining the electrochemical energy storage technology identification results based on the technical topic, it includes: Conduct technology life cycle analysis on the technical features included in the technical theme to obtain the technical maturity of each technical feature in the technical theme; Based on the average value of the technical maturity of each technical feature included in the technical subject and each technical feature included in the technical subject, an evaluation result of each technical subject in the electrochemical energy storage technology identification result is generated.

[0051] Specifically, the method provided by the present invention models the identified electrochemical energy storage target technology characteristics through the Gompertz growth S curve in the technology life cycle analysis, which is expressed as:

[0052] In the formula, is the growth saturation value, is the parameter that controls the growth rate, t For the year, Represents the popularity of the target technical features. Based on the growth data of the popularity of all target technical features, fit the S curve and estimate the parameters. (i.e., the ratio of the situation in the most recent year to the maximum value of the predicted limit) as the technical maturity of the technical feature. Finally, each technical theme uses the mean maturity of all technical features in its cluster as a quantitative indicator of its technical maturity.

[0053] , where m represents the number of technical features included in the technical subject.

[0054] This method measures the application breadth and development depth of a technical feature by measuring the technical value and technical maturity of key technical features. The technical value and technical maturity of technical features belonging to the same topic are averaged, and the average value is used as the technical value and technical maturity of the identified key electrochemical energy storage technology. The technical value and technical maturity measurement indicators of each technology are normalized and used as the vertical and horizontal axes to construct a key technology evaluation matrix. According to the standards of high and low technical value (taking the approximate middle value of the distribution range of technical value values, i.e. 0.15) and early, middle and late stages of technology development (taking technical maturity 0.25 and 0.75), the development types of key technologies are qualitatively divided. Finally, the key technologies of early and middle stages of electrochemical energy storage with high technical value are obtained.

[0055] The Gompertz growth curve is used to model the technology life cycle, and the maturity of the technology theme is quantified by the mean within the cluster; a key technology evaluation matrix is ​​constructed through technology value and technology maturity, and ultimately early and mid-term key technologies with high technology value are identified. The laws of different technologies in development stages and market value are revealed, providing clear guidance for technology management.

[0056] The electrochemical energy storage technology identification device based on patent text analysis provided by the present invention is described below. The electrochemical energy storage technology identification device based on patent text analysis described below and the electrochemical energy storage technology identification method based on patent text analysis described above can be referenced to each other. Figure 3 As shown, the electrochemical energy storage technology identification device based on patent text analysis provided by the present invention includes: A patent text acquisition module 310 is used to acquire a patent text database, wherein the patent text database includes patent texts related to electrochemical energy storage; A text topic mining module 320 is used to obtain the topic distribution corresponding to the patent text through the Dirichlet distribution topic model, and use the topics included in each patent text as the technical features of electrochemical energy storage; A technical feature extraction module 330 is used to obtain the number of applications of a patent text in multiple patent institutions, determine the average marginal effect of each technical feature based on the subject distribution and the number of applications of the patent text, and determine the target technical feature based on the average marginal effect; The technical feature clustering module 340 is used to cluster the target technical features to obtain an electrochemical energy storage technology identification result, and the electrochemical energy storage technology identification result includes a technical theme obtained by clustering the target technical features.

[0057] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430 and a communication bus 440, wherein the processor 410, the communication interface 420 and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the electrochemical energy storage technology identification method based on patent text analysis, the method comprising: obtaining a patent text database, the patent text database including patent texts related to electrochemical energy storage; obtaining the topic distribution corresponding to the patent text through the Dirichlet distribution topic model, and taking the topics included in each patent text as the technical features of electrochemical energy storage; obtaining the number of applications of the patent text in multiple patent institutions, determining the average marginal effect of each technical feature based on the topic distribution of the patent text and the number of applications, and determining the target technical feature based on the average marginal effect; clustering the target technical feature to obtain the electrochemical energy storage technology identification result, the electrochemical energy storage technology identification result including the technical theme obtained by clustering the target technical feature.

[0058] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0059] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the electrochemical energy storage technology identification method based on patent text analysis provided by the above methods, the method comprising: obtaining a patent text database, the patent text database including patent texts related to electrochemical energy storage; obtaining the topic distribution corresponding to the patent text through a Dirichlet distribution topic model, and taking the topics included in each patent text as the technical features of electrochemical energy storage; obtaining the number of applications for the patent text in multiple patent institutions, determining the average marginal effect of each technical feature based on the topic distribution and the number of applications of the patent text, and determining the target technical feature based on the average marginal effect; clustering the target technical feature to obtain an electrochemical energy storage technology identification result, the electrochemical energy storage technology identification result including the technical theme obtained by clustering the target technical feature.

[0060] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the electrochemical energy storage technology identification method based on patent text analysis provided by the above methods, the method comprising: obtaining a patent text database, the patent text database including patent texts related to electrochemical energy storage; obtaining a topic distribution corresponding to the patent text through a Dirichlet distribution topic model, and taking the topics included in each patent text as a technical feature of electrochemical energy storage; obtaining the number of applications for the patent text in multiple patent institutions, determining the average marginal effect of each technical feature based on the topic distribution and the number of applications of the patent text, and determining the target technical feature based on the average marginal effect; clustering the target technical feature to obtain an electrochemical energy storage technology identification result, the electrochemical energy storage technology identification result including the technical theme obtained by clustering the target technical feature.

[0061] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0062] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0063] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying electrochemical energy storage technology based on patent text analysis, characterized in that: include: Obtaining a patent text database, wherein the patent text database includes patent texts related to electrochemical energy storage; The topic distribution corresponding to the patent text is obtained by using the Dirichlet distribution topic model, and the topics included in each of the patent texts are used as technical features of electrochemical energy storage; Obtain the number of applications for the patent text in multiple patent institutions, determine the average marginal effect of each of the technical features based on the subject distribution of the patent text and the number of applications, and determine the target technical feature based on the average marginal effect; The target technical features are clustered to obtain an electrochemical energy storage technology identification result, wherein the electrochemical energy storage technology identification result includes a technical theme obtained by clustering the target technical features.

2. The electrochemical energy storage technology identification method based on patent text analysis according to claim 1 is characterized in that: The topic distribution corresponding to the patent text is obtained by using the Dirichlet distribution topic model, including: The patent text is sampled and labeled using the Dirichlet distribution topic model, wherein the label is a Derwent manual code; Sampling the topic distribution of each of the labels through the Dirichlet distribution topic model to obtain the topic distribution corresponding to each topic and the label of the patent text; The Dirichlet distribution topic model is used to sample the topic word distribution for each of the topics to obtain the topic word distribution corresponding to each topic word and the topic of the patent text.

3. The electrochemical energy storage technology identification method based on patent text analysis according to claim 1 is characterized in that: The determining of the average marginal effect of each of the technical features based on the subject distribution of the patent text and the number of applications includes: Constructing a fitting relationship between the technical features in the patent text and the number of applications through an ordered logistic regression model; The average marginal effect of each of the technical features in each of the patent texts is calculated based on the fitting relationship.

4. The electrochemical energy storage technology identification method based on patent text analysis according to claim 1 is characterized in that: Determining the target technical features based on the average marginal effect includes: The technical feature whose average marginal effect in the patent text with the lowest number of applications is lower than a preset threshold is selected as the target technical feature.

5. The electrochemical energy storage technology identification method based on patent text analysis according to claim 2 is characterized in that: The clustering of the target technology features to obtain electrochemical energy storage technology identification results includes: Obtaining a similarity matrix corresponding to the target technical features, wherein the similarity matrix includes similarities between the target technical features; Based on the similarity matrix, a clustering result is determined based on a neighbor propagation algorithm, and each cluster in the clustering result is used as a technical theme; The electrochemical energy storage technology identification result is determined based on the technical subject.

6. The electrochemical energy storage technology identification method based on patent text analysis according to claim 5 is characterized in that: After determining the electrochemical energy storage technology identification result based on the technical subject, the method includes: Performing a technology life cycle analysis on the technical features included in the technical subject to obtain the technical maturity of each of the technical features in the technical subject; Based on the average value of the technical maturity of each of the technical features included in the technical subject and each of the technical features included in the technical subject, an evaluation result of each of the technical subjects in the electrochemical energy storage technology identification result is generated.

7. An electrochemical energy storage technology identification device based on patent text analysis, characterized in that: include: A patent text acquisition module, used to acquire a patent text database, wherein the patent text database includes patent texts related to electrochemical energy storage; A text topic mining module, used to obtain the topic distribution corresponding to the patent text through the Dirichlet distribution topic model, and use the topics included in each of the patent texts as the technical features of electrochemical energy storage; A technical feature extraction module is used to obtain the number of applications of the patent text in multiple patent institutions, determine the average marginal effect of each of the technical features based on the subject distribution of the patent text and the number of applications, and determine the target technical feature based on the average marginal effect; The technical feature clustering module is used to cluster the target technical features to obtain an electrochemical energy storage technology identification result, wherein the electrochemical energy storage technology identification result includes a technical theme obtained by clustering the target technical features.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the electrochemical energy storage technology identification method based on patent text analysis as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the electrochemical energy storage technology identification method based on patent text analysis as described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the electrochemical energy storage technology identification method based on patent text analysis as described in any one of claims 1 to 6 is implemented.

Images