Patent retrieval method, device and equipment based on big data and storage medium

Abstract

The invention relates to a patent retrieval method, device and equipment based on big data and a storage medium, and relates to the technical field of computers.The method comprises the steps that keywords input by a user are obtained and expanded to obtain an expanded keyword set, and a preliminary retrieval result set is determined according to the keyword set, determining a relevance score matrix according to the text vector representation of each patent in the preliminary retrieval result set, determining a high-relevance patent set according to the relevance score matrix and an application date time sequence of each patent in the high-relevance patent set, and determining the application date time sequence of each patent in the high-relevance patent set according to the high-relevance patent set and the application date time sequence corresponding to the high-relevance patent set. And determining a technical trend prediction result of the high-relevance patent set according to the high-relevance patent set and the relevance score matrix, and generating a retrieval result set for the user according to the high-relevance patent set, the technical trend prediction result corresponding to the high-relevance patent set and the relevance score matrix. According to the method, the accuracy and comprehensiveness of retrieval can be improved.

Description

Technical Field

[0001] The present application relates to the field of big data technology, and in particular to a patent search method, device, equipment and storage medium based on big data. Background Art

[0002] Currently, existing patent search methods mainly rely on keyword matching and classification number search, and users search for relevant patents by entering specific terms.

[0003] However, existing search methods usually cannot effectively capture the correlation between patents and technology development trends. Keyword searches are easily affected by synonyms and polysemous words, and users may miss important related patents. At the same time, existing search methods lack the use of big data analysis, which makes the search process unable to fully explore potential technological innovations and market needs, reducing the comprehensiveness and accuracy of the search. Summary of the invention

[0004] In response to the technical problems existing in the prior art, this application provides a patent search method, device, equipment and storage medium based on big data that can improve the comprehensiveness and accuracy of patent searches.

[0005] In the first aspect, the present application provides a patent search method based on big data, the method comprising:

[0006] Acquire and expand the keywords input by the user to obtain an expanded keyword set;

[0007] Determine a preliminary search result set according to the keyword set;

[0008] Determine a correlation score matrix based on the text vector representation of each patent in the preliminary search result set; the elements in the correlation score matrix represent the correlation between each patent and other patents;

[0009] Determine a highly correlated patent set and a time series of application dates of each patent in the highly correlated patent set according to the correlation score matrix;

[0010] Determine the technology trend prediction result of the highly correlated patent set according to the highly correlated patent set and the corresponding application date time series, and the correlation score matrix;

[0011] A search result set is generated for the user based on the highly correlated patent set and the corresponding technology trend prediction results, as well as the correlation score matrix.

[0012] In a second aspect, the present application also provides a patent search device based on big data, the device comprising:

[0013] The vocabulary expansion module is used to obtain and expand the keywords input by the user to obtain an expanded keyword set;

[0014] A preliminary search module, used to determine a preliminary search result set based on the keyword set;

[0015] A correlation building module, used for determining a correlation score matrix based on the text vector representation of each patent in the preliminary search result set; the elements in the correlation score matrix represent the correlation between each patent and other patents;

[0016] A correlation screening module, used to determine a highly correlated patent set and a time series of application dates of each patent in the highly correlated patent set according to the correlation score matrix;

[0017] A technology trend module, used to determine the technology trend prediction result of the highly correlated patent set according to the highly correlated patent set and the corresponding application date time series, and the correlation score matrix;

[0018] A target search module is used to generate a search result set for the user based on the highly correlated patent set and the corresponding technology trend prediction results, as well as the correlation score matrix.

[0019] On the third aspect, the present application also proposes an electronic device, comprising: a memory for storing a computer software program; a processor for reading and executing the computer software program, thereby implementing a patent retrieval method based on big data as described above.

[0020] In a fourth aspect, the present application also proposes a non-transitory computer-readable storage medium, in which the computer software program is stored. When the computer software program is executed by a processor, it implements a patent retrieval method based on big data as described above.

[0021] The above technical solution provided by the embodiment of the present application has the following advantages compared with the prior art:

[0022] This application comprehensively considers the patent citation relationship and content similarity, making the correlation score between patents more comprehensive and able to identify potential technical intersections and innovations. Combined with the application time series of highly correlated patents and adjusted trend forecasts, it can provide insights into future technological developments and help users grasp industry dynamics and market demand. Through big data analysis capabilities, patent search results can be updated and adjusted in real time to adapt to the rapidly changing market and technological environment, ensuring that users obtain the latest technical information and improving the comprehensiveness and accuracy of patent searches. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1A flowchart of a patent search method based on big data provided for this application;

[0024] Figure 2 A schematic diagram of the structure of a patent search device based on big data provided for this application;

[0025] Figure 3 A schematic diagram of the hardware structure of a possible electronic device provided in this application. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0027] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, "plurality" means two or more, unless otherwise clearly and specifically defined.

[0028] In the description of the present application, the term "for example" is used to mean "used as an example, illustration or description". Any embodiment described as "for example" in the present application is not necessarily to be interpreted as being more preferred or advantageous than other embodiments. In order to enable any technician in the field to implement and use the present application, the following description is given. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present application can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid unnecessary details to obscure the description of the present application. Therefore, the present application is not intended to be limited to the embodiments shown, but is consistent with the widest range of principles and features disclosed in the present application.

[0029] See also Figure 1 , provides a flowchart of a patent search method based on big data of the present application, comprising the following steps:

[0030] Step 101: Acquire and expand the keywords input by the user to obtain an expanded keyword set.

[0031] The keywords input by the user refer to the initial search terms provided by the user when conducting a patent search. These terms are usually technical fields, products, problems or other related concepts that the user is interested in. Assuming that the keyword input by the user is "smart home", it is a broad term covering multiple technical fields.

[0032] The main purpose of keyword expansion is to capture all possible terms and concepts that are more relevant to the user's intent, thereby improving the comprehensiveness and accuracy of the retrieval. In some embodiments, synonyms of the user-entered keywords can be found. For example, synonyms of "smart home" may include "smart home", "smart home appliances", etc. Terms related to the user's keywords can be introduced, which may come from technical literature, industry standards, or professional vocabulary. For example, terms related to "smart home" include "Internet of Things", "automated system", etc. The context of the user's input can be analyzed, and natural language processing technology can be used to identify potential related keywords. In this way, other related words can be intelligently recommended based on the context of the user's query.

[0033] After the expansion process, a richer keyword set is formed, which includes the user's initial input and all related expanded words. For example, for the input "smart home", the expanded keyword set may include: "smart home", "smart family", "smart home appliances", etc.

[0034] Step 102: Determine a preliminary search result set based on the keyword set.

[0035] The expanded keyword set can include the keywords entered by the user and their synonyms, related terms, etc. This set is the basis of the retrieval process. Assume that the expanded keyword set is "smart home", "Internet of Things", "home automation", etc.

[0036] In some embodiments, the system uses a keyword set to search for patent documents, usually in the following ways: Keyword matching: Search the patent database for all patent documents containing the keywords in the set. This may include the title, abstract, and text of the patent. Classification number search: Map keywords to the relevant International Patent Classification (IPC) or other classification systems to obtain relevant patents.

[0037] In some embodiments, the search system can match qualified patent documents according to keywords and collect these documents as a preliminary search result set. After obtaining the preliminary results, the system can perform deduplication processing to ensure that each patent document appears only once. At the same time, the results are preliminarily sorted according to the matching degree or other criteria.

[0038] The final preliminary search result set contains patent documents that are highly relevant to the keywords entered by the user, and may include detailed information of multiple patents. For example, for the keyword "smart home", the preliminary search result set may contain multiple patents related to the subject, such as: Patent A: Technology about smart home control systems. Patent B: Involving the application of the Internet of Things in home automation. Patent C: Patent describing the interconnection of smart home appliances.

[0039] In summary, through the expanded keyword set, the preliminary search result set of this application can more comprehensively cover the technical fields of interest to users and improve relevance. The preliminary result set provides basic data for subsequent content similarity calculation and correlation analysis, ensuring the effectiveness and accuracy of subsequent steps. It can effectively transform user needs into actual patent information, laying a good foundation for further analysis and optimization.

[0040] Step 103: Determine a relevance score matrix based on the text vector representation of each patent in the preliminary search result set.

[0041] The elements in the correlation score matrix represent the correlation between each patent and other patents.

[0042] In some embodiments, step 103 may include:

[0043] According to the text vector representation of each patent, the content similarity between each patent and other patents is calculated to obtain a content similarity score between each patent and other patents;

[0044] According to the citation relationship between each two patents, the citation relationship score between each patent and other patents is obtained;

[0045] Determine the relevance score of each of the patents with other patents based on the content similarity score and citation relationship score of each of the patents;

[0046] According to each of the correlation scores, the correlation score matrix is ​​constructed.

[0047] In some embodiments, the content similarity score may be determined as follows:

[0048] Obtaining the modulus of the first vector representation of the patent;

[0049] Obtain the magnitude of the second vector representation of the other patents;

[0050] The content similarity score between each of the patents and other patents is determined based on a first product of the first vector and the second vector, and a second product of the modulus represented by the first vector and the modulus represented by the second vector.

[0051] In some embodiments, the content similarity score may be expressed as:

[0052]

[0053] in, is the content similarity score between the i-th patent and the j-th patent, V i is the first vector representation corresponding to the i-th patent, V j is the second vector representation corresponding to the j-th patent, ||V i ||、||V j || are the magnitude represented by the first vector and the magnitude represented by the second vector respectively.

[0054] In the specific implementation, The higher the score, the more similar the contents of the two patents are. i and V j It is usually generated by natural language processing technology (such as TF-IDF, Word2Vec or BERT, etc.), which can capture the semantic information in patent documents. i ·V j The dot product of two vectors is calculated, reflecting their similarity in semantic space. If two vectors point in similar directions (similar content), the dot product value will be larger. Use modulus for normalization, ||V i ||、||V j The product of || can eliminate the impact of different vector lengths and ensure that the similarity score only reflects the similarity in content rather than the difference in vector length.

[0055] The value of is usually between -1 and 1, with a score of 1 indicating that the two patents are exactly the same (pointing in the same direction in the vector space). A score of 0 indicates that the two patents have no similarity (the vectors are orthogonal). A score of -1 indicates that the two patents are completely opposite in content. By calculating the content similarity score, patents with similar technical content can be effectively identified, providing an important basis for subsequent correlation analysis. Patents with high similarity scores can help users find technical fields and innovations related to their queries, and improve the accuracy and effectiveness of patent retrieval. During the search process, through the content similarity score, the system can recommend other patents with similar content to the patents of interest to users, thereby expanding the user's horizons and promoting in-depth understanding and research of technology.

[0056] In summary, the calculation method of the content similarity score of this application can effectively quantify the similarity between patents and provide important data support for subsequent correlation analysis and search result optimization.

[0057] In some embodiments, the citation relationship score may be expressed as:

[0058]

[0059] Among them, C ij is the patent citation data, count(C ij ) represents the number of times the i-th patent cites the j-th patent.

[0060] In the specific implementation, represents the score of the i-th patent citing the j-th patent, reflecting the strength of the citation relationship between the two patents. ij Usually includes information such as patent number, citation time, citation type, etc. count(C ij )The higher the value, the greater the dependence and citation intensity of the i-th patent on the j-th patent.

[0061] if This indicates that the two are technically related. This means that the direct relationship between the two is weak or non-existent. By calculating the citation relationship score, patents that are technically interdependent can be effectively identified, which is of great significance for understanding the context of technological development and the evolution of innovation. Patents with high citation scores are generally considered to be more influential in a certain technical field, which can help users quickly identify key and representative patents.

[0062] During the patent search process, citation relationship scores can provide users with information about the technological innovation chain, help them identify important patents in related fields and their technical background, and enhance their understanding of technological development.

[0063] Through this process, the calculation of citation relationship scores can provide important data support for subsequent correlation analysis and patent recommendations, and improve the comprehensiveness and accuracy of patent retrieval.

[0064] In some embodiments, the correlation score of each of the patents with other patents is expressed as:

[0065]

[0066] Among them, R ij is the correlation score between the ith patent and the jth patent, is the content similarity score between the ith patent and the ith patent, is the citation relationship score between the ith patent and the ith patent, α is the first weight, and β is the second weight.

[0067] In a specific implementation, the first weight and the second weight can be set based on domain knowledge, user needs, or through historical data analysis. According to specific retrieval objectives and technical backgrounds, users or researchers can adjust these two parameters to better reflect the actual situation of technical associations.

[0068] High score R ij A high score indicates that there is a strong correlation between the i-th patent and the j-th patent, which means that the technologies are complementary, the citation relationship is close, or the content is highly similar. A low score indicates that the correlation between the two is weak, which means that they have no direct relationship in terms of technical content or application field.

[0069] By combining content similarity and citation relationships, the relevance score provides a comprehensive perspective to understand the technical connections between patents. Patents with high relevance scores can be recommended first in subsequent search results, helping users quickly find relevant technical innovations and solutions. In patent search and analysis, the relevance score can provide users with the context of technological development and the evolution of innovation, supporting users to make more informed technical decisions.

[0070] In summary, the calculation of the relevance score of this application provides important data support for subsequent search optimization and patent recommendation, and enhances the accuracy and effectiveness of patent search.

[0071] Step 104: Determine a highly correlated patent set and a time series of application dates of each patent in the highly correlated patent set based on the correlation score matrix.

[0072] Among them, the application date time series of each patent in the highly correlated patent set refers to the collection of all application dates related to a specific patent, usually arranged in chronological order. This sequence reflects the application activities and development history of patents in a certain technical field.

[0073] In some embodiments, step 104 may include:

[0074] Sorting all the correlation scores in the correlation score matrix to obtain a corresponding first sorting result;

[0075] Determining a target relevance score according to the first sorting result and a first threshold;

[0076] All patents corresponding to the target relevance score are determined as the high-relevance patent set.

[0077] The relevance score matrix is ​​a two-dimensional matrix, where each element R ij Represents the correlation score between the i-th patent and the j-th patent. Each row and column in the matrix corresponds to a different patent.

[0078] In some embodiments, all scores in the matrix can be extracted to form a one-dimensional list. This list can be sorted from high to low according to the value of the score. This process can use a standard sorting algorithm (such as quick sort, merge sort, etc.). After sorting, a new list can be obtained, called the first sorting result. The elements in this list are arranged from high to low according to the relevance score.

[0079] In some embodiments, a first threshold may be set based on actual application requirements or historical data analysis. This threshold is used to screen out patents with higher correlation scores, and is usually determined based on experience or data analysis. The correlation score in the first sorting result may be compared with the threshold. All patents with scores higher than or equal to the threshold are considered target patents. Based on the target correlation score, all patents corresponding to the target score are determined. These patents constitute a highly correlated patent set.

[0080] For example, if the first sorting result is [0.95, 0.85, 0.80, 0.75, 0.70] and the first threshold is set to 0.8, the highly correlated patent set will include patents with scores of 0.95, 0.85 and 0.80.

[0081] Through this process, users can quickly screen out patents that are highly relevant to their query or research topic, improving the accuracy of the search. Highly relevant patent sets provide important basic data for subsequent technical analysis, market evaluation, and R&D decisions, helping users grasp technology trends and market opportunities. In multiple fields such as patent search, technical analysis, and competitor research, this process can help researchers and companies effectively identify key and representative patents, providing support for technological innovation and market strategies.

[0082] In summary, the sorting of the correlation score matrix and the determination of the highly correlated patent set in this application provide users with more accurate and relevant patent information, and enhance the practicality and effectiveness of patent retrieval.

[0083] Step 105: Determine the technology trend prediction result of the highly correlated patent set based on the highly correlated patent set and its corresponding application date time series, as well as the correlation score matrix.

[0084] In some embodiments, step 105 may include:

[0085] Constructing a weight function for determining the technology trend prediction result according to the correlation score matrix and the highly correlated patent set;

[0086] The technology trend prediction result of the highly correlated patent set is determined according to the application date time series and the weight function.

[0087] In some embodiments, the technology trend prediction results of the highly correlated patent set can be expressed as:

[0088]

[0089] in, is the result of technology trend forecast, R a is a highly correlated patent set, T is a highly correlated patent set R a The patent application time series, f(A,R a ) is the weight function of .

[0090] In the specific implementation, is the result of technology trend prediction. This result reflects the technology development trend based on the highly correlated patent set. T is the patent application time series of the highly correlated patent set, which contains the application dates of all relevant patents. This series is used to capture the time-varying characteristics of technology development. a is a highly relevant patent set, that is, a patent set that is highly relevant to user needs and is screened out in the previous steps. f(A,R a ) is a weight function used to adjust and weight technology trends. It dynamically adjusts the impact of technology trends based on the relevance score matrix and the characteristics of highly relevance patent sets.

[0091] In some embodiments, the contribution of each patent to the technology trend can be determined by analyzing the content similarity and citation relationship of each patent in the highly correlated patent set. The weight function can be linear or nonlinear, depending on the specific application requirements.

[0092] Specifically, firstly, the application dates of all patents in the highly correlated patent set are extracted to form an application time series. A weight function can be applied to the application time series to adjust the technology trend prediction results. This function involves statistical analysis of the time series, weighted average and other processing methods.

[0093] It provides predictions on the future development of specific technology fields, helping users understand the potential trends of technology evolution. The technology trend forecast results can provide a basis for the company's R&D strategy, market layout and investment decisions, and enhance the foresight of future technology development. In the fields of technology research, market analysis, intellectual property management, etc., technology trend forecast results can help users grasp industry dynamics, identify technology opportunities, and support more effective decision-making.

[0094] In summary, the technology trend prediction results of the highly correlated patent set of this application can provide users with in-depth technology insights and trend analysis, and enhance the effectiveness of patent retrieval and technological innovation.

[0095] Step 106: Generate a search result set for the user based on the highly correlated patent set and its corresponding technology trend prediction results, as well as the correlation score matrix.

[0096] In some embodiments, step 106 may include:

[0097] According to the correlation score of each patent in the highly correlated patent set, the patents are sorted to obtain a corresponding second sorting result;

[0098] According to the inverse matrix of the correlation score and the technology trend prediction result, the second sorting result is adjusted to obtain a corresponding third sorting result;

[0099] According to the third ranking result and the second threshold, a plurality of target patents are determined from the highly correlated patent set, and the search result set including the plurality of target patents is obtained.

[0100] In some embodiments, all patents can be extracted into a list according to the relevance score of each patent in the highly relevance patent set. The list is sorted from high to low according to the score to form a second sorting result. This process can use a standard sorting algorithm.

[0101] It can be understood that the inverse correlation score matrix is ​​the inverse processing of the original correlation score matrix. It is obtained by calculation or other mathematical methods, and reflects the reverse correlation between patents. The scores of patents can be dynamically adjusted by combining the values ​​in the inverse correlation score matrix and the technology trend prediction results. The specific method can be to ensure that the ranking results can reflect the current technology trends and reverse correlations through weighted average or direct multiplication.

[0102] In some embodiments, a second threshold may be set according to specific application requirements or historical data analysis. This threshold is used to filter out patents with higher scores. The patent scores in the third sorting result may be compared with the second threshold. All patents with scores higher than or equal to the threshold are considered target patents. Based on the target patents, a search result set is determined. This result set will include all eligible target patents and be provided to the user as the final search result.

[0103] Through this process, users can quickly identify patents that are highly relevant to their research or business needs, improving the accuracy and effectiveness of the search. The search result set provides important basic data for subsequent technical analysis, market evaluation and R&D decisions, helping users grasp technology trends and market opportunities.

[0104] In many fields such as patent search, technology analysis and competitor research, this process can help researchers and companies effectively identify key and representative patents to support technological innovation and market strategy.

[0105] In summary, this application can help researchers and companies effectively identify key and representative patents in multiple fields such as patent search, technology analysis and competitor research, and support technological innovation and market strategy.

[0106] See also Figure 2 , Figure 2 A schematic diagram of the structure of a patent search device based on big data provided for this application.

[0107] like Figure 2 As shown, a patent search device based on big data proposed in the embodiment of the present application includes:

[0108] The vocabulary expansion module 201 is used to obtain and expand the keywords input by the user to obtain an expanded keyword set;

[0109] A preliminary search module 202, used to determine a preliminary search result set according to the keyword set;

[0110] The association building module 203 is used to determine a correlation score matrix based on the text vector representation of each patent in the preliminary search result set; the elements in the correlation score matrix represent the correlation between each patent and other patents;

[0111] A correlation screening module 204 is used to determine a highly correlated patent set and an application date time series of each patent in the highly correlated patent set according to the correlation score matrix;

[0112] A technology trend module 205, for determining a technology trend prediction result of the highly correlated patent set according to the highly correlated patent set and the corresponding application date time series, and the correlation score matrix;

[0113] The target search module 206 is used to generate a search result set for the user based on the highly correlated patent set and the corresponding technology trend prediction results, and the correlation score matrix.

[0114] See also Figure 3 , Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present application. Figure 3 As shown, an embodiment of the present application provides an electronic device, including a processor 301, a communication interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communication interface 302, and the memory 303 communicate with each other through the communication bus 304.

[0115] Memory 303, used for storing computer programs;

[0116] In one embodiment of the present application, the processor 301 is used to execute the program stored in the memory 303 to implement the patent search method based on big data provided by any of the above method embodiments, including:

[0117] Acquire and expand the keywords input by the user to obtain an expanded keyword set;

[0118] Determine a preliminary search result set according to the keyword set;

[0119] Determine a correlation score matrix based on the text vector representation of each patent in the preliminary search result set; the elements in the correlation score matrix represent the correlation between each patent and other patents;

[0120] Determine a highly correlated patent set and a time series of application dates of each patent in the highly correlated patent set according to the correlation score matrix;

[0121] Determine the technology trend prediction result of the highly correlated patent set according to the highly correlated patent set and the corresponding application date time series, and the correlation score matrix;

[0122] A search result set is generated for the user based on the highly correlated patent set and the corresponding technology trend prediction results, as well as the correlation score matrix.

[0123] The present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0124] Acquire and expand the keywords input by the user to obtain an expanded keyword set;

[0125] Determine a preliminary search result set according to the keyword set;

[0126] Determine a correlation score matrix based on the text vector representation of each patent in the preliminary search result set; the elements in the correlation score matrix represent the correlation between each patent and other patents;

[0127] Determine a highly correlated patent set and a time series of application dates of each patent in the highly correlated patent set according to the correlation score matrix;

[0128] Determine the technology trend prediction result of the highly correlated patent set according to the highly correlated patent set and the corresponding application date time series, and the correlation score matrix;

[0129] A search result set is generated for the user based on the highly correlated patent set and the corresponding technology trend prediction results, as well as the correlation score matrix.

[0130] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0131] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, devices, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0132] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (apparatus), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0133] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0134] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0135] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0136] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A patent search method based on big data, characterized in that: The method comprises: Acquire and expand the keywords input by the user to obtain an expanded keyword set; Determine a preliminary search result set according to the keyword set; Determine a correlation score matrix based on the text vector representation of each patent in the preliminary search result set; the elements in the correlation score matrix represent the correlation between each patent and other patents; Determine a highly correlated patent set and a time series of application dates of each patent in the highly correlated patent set according to the correlation score matrix; Determine the technology trend prediction result of the highly correlated patent set according to the highly correlated patent set and the corresponding application date time series, and the correlation score matrix; A search result set is generated for the user based on the highly correlated patent set and the corresponding technology trend prediction results, as well as the correlation score matrix.

2. The patent search method based on big data according to claim 1 is characterized in that: Determining a relevance score matrix according to the text vector representation of each patent in the preliminary search result set includes: According to the text vector representation of each patent, the content similarity between each patent and other patents is calculated to obtain a content similarity score between each patent and other patents; According to the citation relationship between each two patents, the citation relationship score between each patent and other patents is obtained; Determine the relevance score of each of the patents with other patents based on the content similarity score and citation relationship score of each of the patents; According to each of the correlation scores, the correlation score matrix is ​​constructed.

3. The patent search method based on big data according to claim 2 is characterized in that: The method of calculating the content similarity between each patent and other patents based on the text vector representation of each patent to obtain the content similarity score between each patent and other patents includes: Obtaining the modulus of the first vector representation of the patent; Obtain the magnitude of the second vector representation of the other patents; Determine a content similarity score between each of the patents and other patents according to a first product of the first vector and the second vector, and a second product of the modulus represented by the first vector and the modulus represented by the second vector; Among them, the content similarity score is expressed as: in, is the content similarity score between the i-th patent and the j-th patent, V i is the first vector representation corresponding to the i-th patent, V j is the second vector representation corresponding to the j-th patent, ||V i ||、||V i || are the magnitude represented by the first vector and the magnitude represented by the second vector respectively.

4. The patent search method based on big data according to claim 3 is characterized in that: The correlation score of each of the patents to other patents is expressed as: Among them, R ij is the correlation score between the ith patent and the jth patent, is the content similarity score between the i-th patent and the j-th patent, is the citation relationship score between the ith patent and the jth patent, α is the first weight, and β is the second weight.

5. The patent search method based on big data according to claim 4 is characterized in that: Determining a highly correlated patent set according to the correlation score matrix includes: Sorting all the correlation scores in the correlation score matrix to obtain a corresponding first sorting result; Determining a target relevance score according to the first sorting result and a first threshold; All patents corresponding to the target relevance score are determined as the high-relevance patent set.

6. The patent search method based on big data according to claim 5 is characterized in that: Determining the technology trend prediction result of the highly correlated patent set according to the highly correlated patent set and the corresponding application date time series, and the correlation score matrix, includes: Constructing a weight function for determining the technology trend prediction result according to the correlation score matrix and the highly correlated patent set; The technology trend prediction result of the highly correlated patent set is determined according to the application date time series and the weight function.

7. The patent search method based on big data according to claim 6 is characterized in that: The search result set generated for the user based on the highly correlated patent set and the corresponding technology trend prediction results, and the correlation score matrix, includes: According to the correlation score of each patent in the highly correlated patent set, the patents are sorted to obtain a corresponding second sorting result; According to the inverse matrix of the correlation score and the technology trend prediction result, the second sorting result is adjusted to obtain a corresponding third sorting result; According to the third ranking result and the second threshold, a plurality of target patents are determined from the highly correlated patent set, and the search result set including the plurality of target patents is obtained.

8. A patent search device based on big data, characterized in that: The device comprises: The vocabulary expansion module is used to obtain and expand the keywords input by the user to obtain an expanded keyword set; A preliminary search module, used to determine a preliminary search result set based on the keyword set; A correlation building module, used for determining a correlation score matrix based on the text vector representation of each patent in the preliminary search result set; the elements in the correlation score matrix represent the correlation between each patent and other patents; A correlation screening module, used to determine a highly correlated patent set and a time series of application dates of each patent in the highly correlated patent set according to the correlation score matrix; A technology trend module, used to determine the technology trend prediction result of the highly correlated patent set according to the highly correlated patent set and the corresponding application date time series, and the correlation score matrix; A target search module is used to generate a search result set for the user based on the highly correlated patent set and the corresponding technology trend prediction results, as well as the correlation score matrix.

9. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor, used to read and execute the computer software program to implement a patent search method based on big data as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, wherein the computer software program is stored, wherein: When the computer software program is executed by a processor, a patent search method based on big data is implemented as described in any one of claims 1 to 7.

Images