A method and system for searching key technologies in patents and documents

Abstract

The present invention provides a method and system for searching key technologies in patents and documents, wherein the method comprises: obtaining a professional category to be searched; subject extraction of documents and patent databases according to the professional category to obtain document subject groups and patent subject groups; patent strength judgment of each subject in the patent subject group to obtain a score of each patent subject; document strength judgment of each subject in the document subject group to obtain a score of each document subject; and determining the key technologies under the professional category according to the scores of each patent subject and each document subject. The present invention can find the key technologies in a field more comprehensively and accurately by analyzing documents and patents at the same time, and determining the key technologies under the professional category based on the scores of patent and document subjects, thereby greatly reducing the time for manual search.

Description

Technical Field

[0001] The present invention relates to the field of data retrieval technology, and in particular to a method and system for retrieving key technologies in patents and documents. Background Art

[0002] When researchers or scientific and technological workers want to understand the key technologies in a field, they need to retrieve a large number of patents and documents, and then extract useful information from the data in the documents and patents, but this often takes a lot of time. The patent with application number KR1020180094337 provides a patent analysis system, including: a patent data collector for collecting patent data in the target technology field; an evaluation index calculation unit for calculating patent evaluation indicators (patent indicators) from the collected patent data; a SOM generation unit for generating a self-organizing map (SOM) from the collected patent data; and a visualization unit for visualizing the self-organizing map. The patent with application number CN202011351495.4 relates to a patent information prediction system based on data mining, including a data screening module, a data mining module, a data analysis module and a result upload module electrically connected in sequence. Among them: the data screening module filters out key data information associated with keyword needs from massive big data, the data mining module performs data mining on key data information based on predetermined rules, the data analysis module analyzes the mining results to obtain keyword demand analysis results, and the result uploading module uploads the analysis results to the service platform for display.

[0003] It can be seen from this that the existing patent search method only searches the patent database, which has the problems of incomplete search and low search accuracy. Summary of the invention

[0004] In view of the deficiencies of the prior art, the purpose of the present invention is to provide a method and system for searching key technologies in patents and documents to solve the problem of low retrieval accuracy of existing patent retrieval methods.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A method for searching key technologies in patents and documents, comprising:

[0007] Get the professional category to be searched;

[0008] Performing subject extraction on the literature and patent databases according to the professional categories to obtain literature subject groups and patent subject groups;

[0009] Determine the patent strength of each subject in the patent subject group to obtain a score for each patent subject;

[0010] Conducting a literature strength judgment on each theme in the literature theme group to obtain a score for each literature theme;

[0011] The key technology under the professional category is determined according to the score of each patent topic and the score of each literature topic.

[0012] Preferably, the step of judging the patent strength of each subject in the patent subject group to obtain a score for each patent subject includes:

[0013] Add the number of patents contained in each theme in the patent theme group to obtain the total number of patents contained in each theme;

[0014] Calculate the average number of words in the first claims of all patents in all themes, and sum the number of corresponding patents in each theme whose first claim number is higher than the average number of words to obtain a first total;

[0015] Calculate the numerical average of the publication date minus the application date of all patents in all themes, and sum the number of corresponding patents in each theme whose numerical value of the publication date minus the application date of each patent is higher than the numerical average to obtain a second total;

[0016] Calculating a first average of the number of patent families of all patents in all themes, and summing the number of corresponding patents in each theme whose number of patent families of each patent is higher than the first average to obtain a third total;

[0017] Calculating a second average of the number of cited patents for all patents in all themes, and summing the number of corresponding patents in each theme whose number of cited patents is higher than the second average to obtain a fourth total;

[0018] Calculating a third average of the number of cited patents of all family patents in all themes, and summing the number of corresponding patents in each theme whose number of cited patents of the family patents of each patent is higher than the third average to obtain a fifth total;

[0019] Calculate the average maintenance period of validity of all patents in all subjects, and sum the number of corresponding patents in each subject whose maintenance period of validity of each patent is higher than the average maintenance period to obtain a sixth total;

[0020] Obtain the validity of each patent in each subject, and sum up the number of valid patents in each subject to get the seventh total;

[0021] The score of each patent topic is obtained according to the corresponding total number of patents, the first total number, the second total number, the third total number, the fourth total number, the fifth total number, the sixth total number and the seventh total number in each topic.

[0022] Preferably, judging the document strength of each subject in the document subject group to obtain a score for each document subject includes:

[0023] Add up the total number of all documents in each topic to get the total number of documents in each topic;

[0024] Calculate a fourth average of the number of all document units in all topics, and sum the number of corresponding document units in all topics whose number of document units is higher than the fourth average to obtain an eighth total;

[0025] Calculate the fifth average of the comprehensive impact factors of the source journals of all documents in all topics, and sum up the number of corresponding documents whose comprehensive impact factors of the source journals of each document in all topics are higher than the fifth average to obtain the ninth total;

[0026] Using a professional dictionary to screen the keywords of each document in each subject of the document to obtain the sixth average value of the number of keywords retained, and summing the number of corresponding documents in all subjects whose number of keywords retained in each document is higher than the sixth average value to obtain the tenth total number;

[0027] The score of each literature topic is obtained based on the total number, eighth total number, ninth total number and tenth total number of corresponding documents in each topic.

[0028] Preferably, determining the key technology under the professional category according to the score of each patent theme and the score of each literature theme includes:

[0029] Arrange each patent topic and each literature topic in descending order of score;

[0030] Select the topics with preset values ​​before ranking patents and documents to obtain the screened patents and documents;

[0031] The probabilities of the same keywords in the topics of the screened patents and documents are added up respectively, and the corresponding probabilities of different keywords are retained;

[0032] Draw a topic heat map based on the probability of keywords;

[0033] The key technologies under the professional category are determined according to the subject heat map.

[0034] Preferably, the score of each patent theme is obtained according to the corresponding total number of patents, the first total number, the second total number, the third total number, the fourth total number, the fifth total number, the sixth total number and the seventh total number in each theme, including:

[0035] The corresponding total number of patents, the first total number, the second total number, the third total number, the fourth total number, the fifth total number, the sixth total number and the seventh total number in each topic are multiplied by the preset weight coefficient and summed to obtain the score of each patent topic.

[0036] Preferably, the score of each document theme is obtained according to the total number, the eighth total number, the ninth total number and the tenth total number of the corresponding documents in each theme, including:

[0037] The total number, the eighth total number, the ninth total number and the tenth total number of the corresponding documents in each topic are multiplied by the preset weight coefficient and summed up to obtain the score of each document topic.

[0038] The present invention also provides a system for searching key technologies in patents and documents, which is characterized by comprising:

[0039] A search term acquisition module is used to obtain the professional category to be searched;

[0040] A subject extraction module, used to extract subjects from the literature and patent databases according to the professional categories to obtain literature subject groups and patent subject groups;

[0041] A patent strength judgment module, used to judge the patent strength of each subject in the patent subject group to obtain a score for each patent subject;

[0042] A document strength judgment module is used to judge the document strength of each topic in the document topic group to obtain a score for each document topic;

[0043] The key technology determination module is used to determine the key technology under the professional category according to the score of each patent topic and the score of each literature topic.

[0044] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0045] The present invention provides a method and system for searching key technologies in patents and documents, wherein the method comprises: obtaining a professional category to be searched; subject extraction of documents and patent databases according to the professional category to obtain document subject groups and patent subject groups; patent strength judgment of each subject in the patent subject group to obtain a score of each patent subject; document strength judgment of each subject in the document subject group to obtain a score of each document subject; and determining the key technologies under the professional category according to the scores of each patent subject and each document subject. The present invention can find the key technologies in a field more comprehensively and accurately by analyzing documents and patents at the same time, and determining the key technologies under the professional category based on the scores of patent and document subjects, thereby greatly reducing the time for manual search. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0047] Figure 1 A flow chart of a method for searching key technologies in patents and documents provided by the present invention;

[0048] Figure 2 A schematic diagram of a device for searching key technologies in patents and documents provided by the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0050] The purpose of the present invention is to provide a method and system for searching key technologies in patents and documents to solve the problem of low search accuracy of existing patent search methods.

[0051] To achieve the above object, the present invention provides the following solutions:

[0052] See also Figure 1 , a method for searching key technologies in patents and documents, comprising:

[0053] Step 100: Obtain the professional category to be searched;

[0054] Step 200: extracting topics from the literature and patent databases according to the professional categories to obtain literature topic groups and patent topic groups;

[0055] In practical applications, the present invention performs big data topic extraction on document and patent data respectively, obtains the perplexity curve and consistency curve corresponding to the document and patent data, obtains the optimal number of topics corresponding to the document and patent from the perplexity curve and the consistency curve respectively, and then, based on the selected optimal number of topics, performs big data topic extraction on the document and patent data respectively to obtain the topic groups corresponding to the document and patent data, and finally, based on the professional category to be retrieved, filters the topics unrelated to the analyzed profession to obtain the final document topic groups and patent topic groups.

[0056] Step 300: Determine the patent strength of each subject in the patent subject group to obtain a score for each patent subject; further step 300 includes:

[0057] Add the number of patents contained in each theme in the patent theme group to obtain the total number of patents contained in each theme;

[0058] Calculate the average number of words in the first claims of all patents in all themes, and sum the number of corresponding patents in each theme whose first claim number is higher than the average number of words to obtain a first total;

[0059] Calculate the numerical average of the publication date minus the application date of all patents in all themes, and sum the number of corresponding patents in each theme whose numerical value of the publication date minus the application date of each patent is higher than the numerical average to obtain a second total;

[0060] Calculating a first average of the number of patent families of all patents in all themes, and summing the number of corresponding patents in each theme whose number of patent families of each patent is higher than the first average to obtain a third total;

[0061] Calculating a second average of the number of cited patents for all patents in all themes, and summing the number of corresponding patents in each theme whose number of cited patents is higher than the second average to obtain a fourth total;

[0062] Calculating a third average of the number of cited patents of all family patents in all themes, and summing the number of corresponding patents in each theme whose number of cited patents of the family patents of each patent is higher than the third average to obtain a fifth total;

[0063] Calculate the average maintenance period of validity of all patents in all subjects, and sum the number of corresponding patents in each subject whose maintenance period of validity of each patent is higher than the average maintenance period to obtain a sixth total;

[0064] Obtain the validity of each patent in each subject, and sum up the number of valid patents in each subject to get the seventh total;

[0065] The score of each patent topic is obtained according to the corresponding total number of patents, the first total number, the second total number, the third total number, the fourth total number, the fifth total number, the sixth total number and the seventh total number in each topic.

[0066] Specifically, the corresponding total number of patents, the first total number, the second total number, the third total number, the fourth total number, the fifth total number, the sixth total number and the seventh total number in each topic are multiplied by the preset weight coefficient and summed to obtain the score of each patent topic.

[0067] It should be noted that the present invention can also obtain the score of each patent subject according to any one or more of the total number of patents, the first total number, the second total number, the third total number, the fourth total number, the fifth total number, the sixth total number and the seventh total number.

[0068] Step 400: Performing document strength judgment on each subject in the document subject group to obtain a score for each document subject; further, step 400 includes:

[0069] Add up the total number of all documents in each topic to get the total number of documents in each topic;

[0070] Calculate a fourth average of the number of all document units in all topics, and sum the number of corresponding document units in all topics whose number of document units is higher than the fourth average to obtain an eighth total;

[0071] Calculate the fifth average of the comprehensive impact factors of the source journals of all documents in all topics, and sum up the number of corresponding documents whose comprehensive impact factors of the source journals of each document in all topics are higher than the fifth average to obtain the ninth total;

[0072] Using a professional dictionary to screen the keywords of each document in each subject of the document to obtain the sixth average value of the number of keywords retained, and summing the number of corresponding documents in all subjects whose number of keywords retained in each document is higher than the sixth average value to obtain the tenth total number;

[0073] The score of each literature topic is obtained based on the total number, eighth total number, ninth total number and tenth total number of corresponding documents in each topic.

[0074] Specifically, the total number, the eighth total number, the ninth total number and the tenth total number of the corresponding documents in each topic are multiplied by the preset weight coefficient and summed to obtain the score of each document topic.

[0075] It should be noted that the present invention can also obtain the score of each document topic according to any one or more of the total number of documents, the eighth total number, the ninth total number and the tenth total number.

[0076] Step 500: Determine the key technology under the professional category according to the score of each patent topic and the score of each document topic.

[0077] Furthermore, step 500 includes:

[0078] Arrange each patent topic and each literature topic in descending order of score;

[0079] Select the topics with preset values ​​before ranking patents and documents to obtain the screened patents and documents;

[0080] The probabilities of the same keywords in the topics of the screened patents and documents are added up respectively, and the corresponding probabilities of different keywords are retained;

[0081] Draw a topic heat map based on the probability of keywords;

[0082] The key technologies under the professional category are determined according to the subject heat map.

[0083] The present invention analyzes documents and patents simultaneously, and determines the key technologies under professional categories based on the scores of patent and document topics. It can find the key technologies in a field more comprehensively and accurately, greatly reducing the time for manual search.

[0084] The present invention also provides a system for searching key technologies in patents and documents, which is characterized by comprising:

[0085] A search term acquisition module is used to obtain the professional category to be searched;

[0086] A subject extraction module, used to extract subjects from the literature and patent databases according to the professional categories to obtain literature subject groups and patent subject groups;

[0087] A patent strength judgment module, used to judge the patent strength of each subject in the patent subject group to obtain a score for each patent subject;

[0088] A document strength judgment module is used to judge the document strength of each topic in the document topic group to obtain a score for each document topic;

[0089] The key technology determination module is used to determine the key technology under the professional category according to the score of each patent topic and the score of each literature topic.

[0090] See also Figure 2 The present invention also provides a device for searching key technologies in patents and documents, including a contemporary document and patent database, a data preprocessing system, a big data processing system, a document and patent subject analysis system, and a key technology analysis system.

[0091] The following is a further description of the device for searching key technologies in patents and documents of the present invention in conjunction with specific embodiments:

[0092] The present invention first needs to establish a Chinese patent database and a Chinese literature database, wherein the data types of the patent database include the patent serial number, title, abstract, applicant, publication number, publication date, application number, application date, patent type, publication country, first claim, number of claims, word count of the first claim, cited patent, cited patent, family citation, family citation, patent validity. The data types of the Chinese literature database include source library, title, author, unit, literature source, keyword, abstract, time, year.

[0093] The data preprocessing system is used to preprocess the established literature and patent database. Its specific functions are as follows:

[0094] ① Establish a professional dictionary containing professional or technical keywords corresponding to the key technologies to be searched.

[0095] ②Perform Chinese word segmentation on the abstracts of literature data and patent data.

[0096] ③Build a dictionary to remove stop words.

[0097] The big data processing system is used to process the literature and patent data after data preprocessing. Its specific functions are as follows:

[0098] ① Perform big data topic extraction on the processed literature and patent data respectively, obtain the perplexity curve and consistency curve corresponding to the literature and patent data, and obtain the optimal number of topics corresponding to the literature and patent from the perplexity curve and consistency curve respectively.

[0099] ② Based on the selected optimal number of topics, big data topic extraction is performed on the literature and patent data respectively to obtain the topic groups corresponding to the literature and patent data respectively.

[0100] ③Filter topics that are not related to the analysis profession.

[0101] The subject analysis system is used to make corresponding patent strength judgments on patents and documents.

[0102] The number of patents in each subject of the patent data is obtained by adding up the total number of patents in each subject to obtain the total number of patents in each subject a.

[0103] The number of words in the first claim of each patent in each subject of the patent data is averaged, and the number of words above the average is recorded as 0, and the number below the average is recorded as 1. Then, the sum of the number of words of the first claim of all patents in each subject with a word count of 1 is obtained to obtain a total b.

[0104] The publication date of each patent in each subject of the patent data is subtracted from the patent application date. The average of the values ​​of the publication date of all patents in all subjects minus the patent application date is recorded as 1 if it is higher than the average, and as 0 if it is lower than the average. Then, the sum of the values ​​of the publication date of all patents in each subject minus the patent application date is 1 to obtain a total number c.

[0105] The number of patent families for each patent in each subject of the patent data is averaged, with those above the average being recorded as 1 and those below the average being recorded as 0. Then, the sum of the number of patent families for all patents in each subject with a value of 1 is taken to obtain a total d.

[0106] The number of cited patents for each patent in each topic of the patent data is averaged, with those above the average recorded as 1 and those below the average recorded as 0. Then, the number of cited patents for all patents in each topic with the number of cited patents being 1 is summed to obtain a total number e.

[0107] The number of family patents cited for each patent in each topic of the patent data is averaged, with those above the average recorded as 1 and those below the average recorded as 0. Then, the sum of the number of family patents cited for all patents in each topic with the number of citations being 1 is obtained to obtain a total number f.

[0108] The validity of each patent in each subject of the patent data, where a valid patent is recorded as 1 and an invalid patent is recorded as 0. Then, the total number of patents with a validity of 1 in each subject is counted.

[0109] The number of years of validity of each patent in each subject of the patent data is averaged, with those above the average recorded as 1 and those below the average recorded as 0. Then, the sum of the years of validity of all patents in each subject with the number being 1 is taken to obtain a total number g.

[0110] The parameters a, b, c, d, e, f, g, and h are given different weights, and then the parameters are multiplied by the weight coefficients and summed to obtain the final score of each topic, and each topic is sorted from high to low according to the final score. In the present invention, the weight coefficients of patent data are as follows:

[0111] Serial number parameter Weight coefficient 1 a 0.1 2 b 0.1 3 c 0.1 4 d 0.2 5 e 0.2 6 f 0.1 7 g 0.2

[0112] On the Judgment of Literature Strength Corresponding to Literature Data Themes

[0113] The number of documents in each topic of the literature data is obtained by adding up the total number of documents in each topic to obtain the total number of documents containing papers in each topic a.

[0114] The number of document units in each document in each topic of the literature data is averaged across all topics, with those above the average recorded as 1 and those below the average recorded as 0. Then, the sum of all document units with a value of 1 in each topic is taken to obtain a total number b.

[0115] The comprehensive impact factor of each source journal of each document in each topic of the literature data is averaged. The value above the average is recorded as 1, and the value below the average is recorded as 0. Then, the sum of the comprehensive impact factors of all source journals of the literature in each topic with a value of 1 is taken to obtain a total number c.

[0116] According to the professional dictionary in the above data preprocessing system, the keywords of each document in each subject of the literature data are screened, the keywords in the professional dictionary are retained and the number of retained keywords is counted, the keywords in the professional dictionary are retained for all documents in all subjects and the number of retained keywords is averaged, those above the average are recorded as 1, and those below the average are recorded as 0, then the keywords in the professional dictionary are retained for all documents in each subject and the number of retained keywords is summed up to 1 to obtain a total number d.

[0117] The parameters a, b, c, and d are given different weights, and then the parameters are multiplied by the weight coefficients to obtain the final score of each topic, and each topic is sorted from high to low according to the final score. In the present invention, the weight coefficients of the literature data are shown in the following table.

[0118] Weight coefficient table of literature data

[0119] Serial number parameter Weight coefficient 1 a 0.3 2 b 0.1 3 c 0.3 4 d 0.3

[0120] The key technology analysis system is used to find out the key technologies under professional categories. Its specific functions are as follows:

[0121] ① Statistics are collected on the topics of the screened high-intensity patents and documents, and the topic intensity of documents and patents is arranged in descending order.

[0122] ② Calculate the heat values ​​of the above topics and draw a topic heat map. Select the top 10 topics in patents and documents, add the probabilities of the same keywords in patent and document topics, retain the probabilities of different keywords, and draw a topic heat map based on the probability of keywords.

[0123] ③ According to the topic heat map, find out the topics that are similar or identical in the patents and documents. The topic or topics are the key technologies under the corresponding professional category.

[0124] The present invention adopts a big data method, by establishing a database of documents and patents at the same time, a data preprocessing system, a big data processing system, a document and patent subject analysis system and a key technology analysis system, the useful information of documents and patents is converted into quantitative indicators, and then the key technologies in the documents and patents are found through the quantitative indicators, so as to quickly find the key technology or key technology group corresponding to the field or profession in a large amount of document and patent data, so that scientific researchers or people in related fields can quickly understand the research direction and research methods in the field.

[0125] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0126] The present invention provides a method and system for searching key technologies in patents and documents, wherein the method comprises: obtaining a professional category to be searched; subject extraction of documents and patent databases according to the professional category to obtain document subject groups and patent subject groups; patent strength judgment of each subject in the patent subject group to obtain a score of each patent subject; document strength judgment of each subject in the document subject group to obtain a score of each document subject; and determining the key technologies under the professional category according to the scores of each patent subject and each document subject. The present invention can find the key technologies in a field more comprehensively and accurately by analyzing documents and patents at the same time, and determining the key technologies under the professional category based on the scores of patent and document subjects, thereby greatly reducing the time for manual search.

[0127] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0128] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for searching key technologies in patents and documents, characterized in that: include: Get the professional category to be searched; Performing subject extraction on the literature and patent databases according to the professional categories to obtain literature subject groups and patent subject groups; Determine the patent strength of each subject in the patent subject group to obtain a score for each patent subject; The patent strength judgment of each subject in the patent subject group to obtain a score for each patent subject includes: Add the number of patents contained in each theme in the patent theme group to obtain the total number of patents contained in each theme; Calculate the average number of words in the first claims of all patents in all themes, and sum the number of corresponding patents in each theme whose first claim number is higher than the average number of words to obtain a first total; Calculate the numerical average of the publication date minus the application date of all patents in all themes, and sum the number of corresponding patents in each theme whose numerical value of the publication date minus the application date of each patent is higher than the numerical average to obtain a second total; Calculating a first average of the number of patent families of all patents in all themes, and summing the number of corresponding patents of each patent in each theme whose number of patent families is higher than the first average to obtain a third total; Calculating a second average of the number of cited patents of all patents in all themes, and summing the number of corresponding patents in each theme whose number of cited patents is higher than the second average to obtain a fourth total; Calculating a third average of the number of cited patents of all family patents in all themes, and summing the number of corresponding patents in each theme whose number of cited patents of the family patents of each patent is higher than the third average to obtain a fifth total; Calculate the average maintenance period of validity of all patents in all subjects, and sum the number of corresponding patents in each subject whose maintenance period of validity of each patent is higher than the average maintenance period to obtain a sixth total; Obtain the validity of each patent in each subject, and sum up the number of valid patents in each subject to get the seventh total; The score of each patent theme is obtained according to the corresponding total number of patents, first total number, second total number, third total number, fourth total number, fifth total number, sixth total number and seventh total number in each theme; Conducting a literature strength judgment on each theme in the literature theme group to obtain a score for each literature theme; The step of judging the document strength of each subject in the document subject group to obtain a score for each document subject includes: Add up the total number of all documents in each topic to get the total number of documents in each topic; Calculate a fourth average of the number of all document units in all topics, and sum the number of corresponding document units in all topics whose number of document units is higher than the fourth average to obtain an eighth total; Calculate the fifth average of the comprehensive impact factors of the source journals of all documents in all topics, and sum up the number of corresponding documents whose comprehensive impact factors of the source journals of each document in all topics are higher than the fifth average to obtain the ninth total; Using a professional dictionary to screen the keywords of each document in each subject of the document to obtain the sixth average value of the number of keywords retained, and summing the number of corresponding documents in all subjects whose number of keywords retained in each document is higher than the sixth average value to obtain the tenth total number; The score of each literature theme is obtained according to the total number, eighth total number, ninth total number and tenth total number of the corresponding literature in each theme; Determine the key technology under the professional category according to the score of each patent subject and the score of each literature subject; The key technologies under the professional category are determined according to the scores of each patent subject and the scores of each literature subject, including: Arrange each patent topic and each literature topic in descending order of score; Select the topics with preset values ​​before ranking patents and documents to obtain the screened patents and documents; The probabilities of the same keywords in the topics of the screened patents and documents are added up respectively, and the corresponding probabilities of different keywords are retained; Draw a topic heat map based on the probability of keywords; The key technologies under the professional category are determined according to the subject heat map.

2. A method for searching key technologies in patents and documents according to claim 1, characterized in that: The score of each patent theme is obtained according to the total number of patents, the first total number, the second total number, the third total number, the fourth total number, the fifth total number, the sixth total number and the seventh total number in each theme, including: The corresponding total number of patents, the first total number, the second total number, the third total number, the fourth total number, the fifth total number, the sixth total number and the seventh total number in each topic are multiplied by the preset weight coefficient and summed to obtain the score of each patent topic.

3. A method for searching key technologies in patents and documents according to claim 2, characterized in that: The score of each document theme is obtained according to the total number, the eighth total number, the ninth total number and the tenth total number of the corresponding documents in each theme, including: The total number, the eighth total number, the ninth total number and the tenth total number of the corresponding documents in each topic are multiplied by the preset weight coefficient and summed up to obtain the score of each document topic.

4. A system for searching key technologies in patents and documents, characterized in that: include: A search term acquisition module is used to obtain the professional category to be searched; A subject extraction module, used to extract subjects from the literature and patent databases according to the professional categories to obtain literature subject groups and patent subject groups; A patent strength judgment module, used to judge the patent strength of each subject in the patent subject group to obtain a score for each patent subject; The patent strength judgment of each subject in the patent subject group to obtain a score for each patent subject includes: Add the number of patents contained in each theme in the patent theme group to obtain the total number of patents contained in each theme; Calculate the average number of words in the first claims of all patents in all themes, and sum the number of corresponding patents in each theme whose first claim number is higher than the average number of words to obtain a first total; Calculate the numerical average of the publication date minus the application date of all patents in all themes, and sum the number of corresponding patents in each theme whose numerical value of the publication date minus the application date of each patent is higher than the numerical average to obtain a second total; Calculating a first average of the number of patent families of all patents in all themes, and summing the number of corresponding patents of each patent in each theme whose number of patent families is higher than the first average to obtain a third total; Calculating a second average of the number of cited patents of all patents in all themes, and summing the number of corresponding patents in each theme whose number of cited patents is higher than the second average to obtain a fourth total; Calculating a third average of the number of cited patents of all family patents in all themes, and summing the number of corresponding patents in each theme whose number of cited patents of the family patents of each patent is higher than the third average to obtain a fifth total; Calculate the average maintenance period of validity of all patents in all subjects, and sum the number of corresponding patents in each subject whose maintenance period of validity of each patent is higher than the average maintenance period to obtain a sixth total; Obtain the validity of each patent in each subject, and sum up the number of valid patents in each subject to get the seventh total; The score of each patent theme is obtained according to the corresponding total number of patents, first total number, second total number, third total number, fourth total number, fifth total number, sixth total number and seventh total number in each theme; A document strength judgment module is used to judge the document strength of each topic in the document topic group to obtain a score for each document topic; The step of judging the document strength of each subject in the document subject group to obtain a score for each document subject includes: Add up the total number of all documents in each topic to get the total number of documents in each topic; Calculate a fourth average of the number of all document units in all topics, and sum the number of corresponding document units in all topics whose number of document units is higher than the fourth average to obtain an eighth total; Calculate the fifth average of the comprehensive impact factors of the source journals of all documents in all topics, and sum up the number of corresponding documents whose comprehensive impact factors of the source journals of each document in all topics are higher than the fifth average to obtain the ninth total; Using a professional dictionary to screen the keywords of each document in each subject of the document to obtain the sixth average value of the number of keywords retained, and summing the number of corresponding documents in all subjects whose number of keywords retained in each document is higher than the sixth average value to obtain the tenth total number; The score of each literature theme is obtained according to the total number, eighth total number, ninth total number and tenth total number of the corresponding literature in each theme; A key technology determination module, used to determine the key technology under the professional category according to the score of each patent subject and the score of each literature subject; The key technologies under the professional category are determined according to the scores of each patent subject and the scores of each literature subject, including: Arrange each patent topic and each literature topic in descending order of score; Select the topics with preset values ​​before ranking patents and documents to obtain the screened patents and documents; The probabilities of the same keywords in the topics of the screened patents and documents are added up respectively, and the corresponding probabilities of different keywords are retained; Draw a topic heat map based on the probability of keywords; The key technologies under the professional category are determined according to the subject heat map.

Images