Intelligent capturing method and system for key information of traditional Chinese medicine text

Abstract

The invention provides a traditional Chinese medicine text key information intelligent capturing method and system. The method comprises the following steps: receiving a traditional Chinese medicine text data set, and preprocessing each traditional Chinese medicine text data in the traditional Chinese medicine text data set; performing word feature extraction on each traditional Chinese medicine text data by using a word vector model to obtain a first word data set, and performing semantic analysis on each word data in the first word data set to obtain a simplified second word data set; inputting the second word data set and the traditional Chinese medicine knowledge graph into a pre-constructed deep learning model, wherein the deep learning model outputs a plurality of pieces of prescription information; and repeating the above process, completing the processing of all traditional Chinese medicine text data, and carrying out integration processing on the obtained prescription information. According to the scheme, the key information can be efficiently and accurately extracted from the traditional Chinese medicine text.

Description

Technical Field

[0001] The present invention relates to the technical field of content extraction, and in particular to a method and system for intelligently capturing key information of traditional Chinese medicine texts. Background Art

[0002] As the traditional medicine of the Chinese nation, Chinese medicine has a long history and rich cultural connotations. With the development of information technology, a large number of Chinese medicine documents are digitally stored. However, these documents are scattered and in various forms. Traditional manual methods are difficult to extract key information from them efficiently and accurately, and cannot meet the needs of modern Chinese medicine research and application. Therefore, there is an urgent need for an intelligent solution that can automatically, efficiently and accurately extract key information from Chinese medicine texts. Summary of the invention

[0003] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a method, system, electronic device, computer storage medium and computer program product for intelligently capturing key information of traditional Chinese medicine texts.

[0004] The present invention provides a method for intelligently capturing key information of traditional Chinese medicine texts, comprising the following steps: Receiving a traditional Chinese medicine text data set, and preprocessing each traditional Chinese medicine text data in the traditional Chinese medicine text data set; Use a word vector model to extract word features from each of the TCM text data to obtain a first word data set, and perform semantic analysis on each word data in the first word data set to obtain a streamlined second word data set; Inputting the second word dataset and the TCM knowledge graph into a pre-built deep learning model, and the deep learning model outputs a number of prescription information; The above process is repeated to complete the processing of all the TCM text data, and the obtained prescription information is integrated.

[0005] Optionally, the preprocessing of each TCM text data in the TCM text data set includes: The text of each TCM text data is standardized, including word segmentation, removal of stop words, etc.; then, the redundant data in each TCM text data after the standardized processing is removed.

[0006] Optionally, a word vector model is used to extract word features from each of the TCM text data to obtain a first word data set, and a semantic analysis is performed on each word data in the first word data set to obtain a streamlined second word data set, including: Use the word vector model to extract word features from each of the TCM text data to obtain the corresponding first word data sets; Each word data in the first word data set is restored to the corresponding position of the original traditional Chinese medicine text data, and the word data is semantically analyzed based on the context sentence corresponding to the corresponding position using the first general large language model. If the semantic analysis result shows that it belongs to prescription information, the word data is retained, otherwise it is deleted, thereby obtaining a streamlined second word data set.

[0007] Optionally, the second word dataset and the TCM knowledge graph are input into a pre-built deep learning model, and the deep learning model outputs a number of prescription information, including: Determine the text generation year information corresponding to each second word data in the second word data set, input the text generation year information and the traditional Chinese medicine knowledge graph into a second general large language model, and the second general large language model outputs a traditional Chinese medicine knowledge sub-graph matching the text generation year information; The second word dataset and the traditional Chinese medicine knowledge sub-graph are input into a pre-built deep learning model, and the deep learning model outputs a number of prescription information.

[0008] Optionally, the step of inputting the text generation time information and the TCM knowledge graph into a second general large language model, and the second general large language model outputting a TCM knowledge sub-graph matching the text generation time information, further includes: When the output prescription information reaches a preset number, the execution of the method is suspended, and the output corresponding prescription information is manually verified based on the traditional Chinese medicine text data, and the error rate is statistically obtained; A correction value is determined according to the error rate, and the text generation year information is downward expanded based on the correction value. The text generation year information and the traditional Chinese medicine knowledge graph after downward expansion are input into a second general large language model, and the second general large language model outputs the traditional Chinese medicine knowledge sub-graph that matches the text generation year information.

[0009] Optionally, the calculation formula of the correction value is as follows: ; In the formula, is the correction value, in years; is a constant; is the error rate.

[0010] The present invention also provides a system for intelligently capturing key information of traditional Chinese medicine texts, comprising a processing module and a storage module, wherein a computer program is stored in the storage module, and the processing module executes the computer program to implement the following method steps: Receiving a traditional Chinese medicine text data set, and preprocessing each traditional Chinese medicine text data in the traditional Chinese medicine text data set; Use a word vector model to extract word features from each of the TCM text data to obtain a first word data set, and perform semantic analysis on each word data in the first word data set to obtain a streamlined second word data set; Inputting the second word dataset and the TCM knowledge graph into a pre-built deep learning model, and the deep learning model outputs a number of prescription information; The above process is repeated to complete the processing of all the TCM text data, and the obtained prescription information is integrated.

[0011] The present invention also provides an electronic device, comprising: a memory storing executable program code; a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the method as described in any of the preceding items.

[0012] The present invention also provides a computer storage medium, on which a computer program is stored. When the computer program is executed by a processor, any of the above methods is executed.

[0013] The present invention also provides a computer program product, which includes computer code. When the computer code is executed by a processor of an electronic device, the method described in any one of the above items is implemented.

[0014] The beneficial effects of the present invention are that the scheme of the present invention can efficiently and accurately extract key information from traditional Chinese medicine texts, realize automated processing of traditional Chinese medicine texts, improve the efficiency and accuracy of information extraction, and is conducive to promoting the development of traditional Chinese medicine research and application. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 It is a flow chart of a method for intelligently capturing key information of traditional Chinese medicine texts disclosed in an embodiment of the present invention.

[0017] Figure 2 It is a schematic diagram of a process of obtaining a simplified second word data set disclosed in an embodiment of the present invention.

[0018] Figure 3The present invention discloses a structural diagram of a system for intelligently capturing key information of traditional Chinese medicine texts according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0020] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. 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.

[0021] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0022] In the description of the present invention, it should be noted that if the terms "upper", "lower", "inner", "outer", etc. appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the drawings, or the orientation or position relationship in which the product of the invention is usually placed when in use. It is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. In addition, if the terms "first", "second", etc. appear, they are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0023] like Figure 1 As shown, a method for intelligently capturing key information of a traditional Chinese medicine text according to an embodiment of the present invention comprises the following steps: Receiving a traditional Chinese medicine text data set, and preprocessing each traditional Chinese medicine text data in the traditional Chinese medicine text data set; Use a word vector model to extract word features from each of the TCM text data to obtain a first word data set, and perform semantic analysis on each word data in the first word data set to obtain a streamlined second word data set; Inputting the second word dataset and the TCM knowledge graph into a pre-built deep learning model, and the deep learning model outputs a number of prescription information; The above process is repeated to complete the processing of all the TCM text data, and the obtained prescription information is integrated.

[0024] The above-mentioned intelligent capture method of key information of traditional Chinese medicine text of the present invention includes three main steps, namely data preprocessing, feature extraction, and key information extraction.

[0025] Specifically, the received TCM text data are first normalized through data preprocessing to facilitate the subsequent extraction of key information.

[0026] Then, the word vector model is used to extract word features from each TCM text data to obtain the first word data set, which includes vocabulary corresponding to disease types, symptoms, drug names, drug dosages, pharmaceutical methods, etc.; semantic analysis is performed on each word data in the first word data set to obtain a streamlined second word data set, that is, other information not related to the prescription is filtered out, such as deleting the author of the prescription, the year of writing, the address of writing, etc.

[0027] Finally, the word data in the second word data set and the TCM knowledge graph are input into the pre-built deep learning model, and the deep learning model performs in-depth analysis on the data in the second word data set under the guidance of the TCM knowledge graph, thereby obtaining some prescription information. Thus, by repeating the above process, the prescription information of each TCM text data in the TCM text data set can be extracted, and all the extracted prescription information is integrated, thus completing the intelligent capture of the key information of the TCM text.

[0028] Therefore, the solution of the present invention can efficiently and accurately extract key information from traditional Chinese medicine texts, realize automated processing of traditional Chinese medicine texts, improve the efficiency and accuracy of information extraction, and is conducive to promoting the development of traditional Chinese medicine research and application.

[0029] It should be noted that there are many types of traditional Chinese medicine text data, such as ancient texts, modern documents, prescriptions written by well-known experts, etc., and the present invention is not limited to them.

[0030] Optionally, the preprocessing of each TCM text data in the TCM text data set includes: The text of each TCM text data is standardized, including word segmentation, removal of stop words, etc.; then, the redundant data in each TCM text data after the standardized processing is removed.

[0031] In the embodiment of the present invention, the preprocessing of the text includes word segmentation, stop word removal and removal of obviously irrelevant data. Among them, with the continuous development of traditional Chinese medicine technology, some traditional Chinese medicine vocabulary will be in a state of being stopped, which is more obvious than in other fields. Therefore, removing stop words is crucial to obtaining accurate prescription information.

[0032] Alternatively, if Figure 2 As shown, a word vector model is used to extract word features from each of the TCM text data to obtain a first word data set, and semantic analysis is performed on each word data in the first word data set to obtain a streamlined second word data set, including: Use the word vector model to extract word features from each of the TCM text data to obtain the corresponding first word data sets; Each word data in the first word data set is restored to the corresponding position of the original traditional Chinese medicine text data, and the word data is semantically analyzed based on the context sentence corresponding to the corresponding position using the first general large language model. If the semantic analysis result shows that it belongs to prescription information, the word data is retained, otherwise it is deleted, thereby obtaining a streamlined second word data set.

[0033] In the embodiment of the present invention, the present invention does not perform semantic analysis on each word data itself, but places each word data in the first word data set in the original text, and performs contextual semantic analysis based on the context sentence, which can improve the accuracy of the semantic analysis of the word data. After obtaining the semantic analysis result, it can be decided whether to retain it based on the semantic analysis result. Specifically, if the semantic analysis result shows that it belongs to prescription information, the word data is retained, otherwise it is deleted, thereby obtaining a streamlined second word data set.

[0034] The above semantic analysis is preferably implemented using a general large language model, such as the BERT large language model. BERT is a pre-trained language representation model proposed by Google in 2018. It has caused great changes in the field of natural language processing (NLP). The core feature of BERT is the use of a bidirectional Transformer encoder. Through large-scale unsupervised pre-training, it has learned deep language representations, which has enabled it to achieve significant performance improvements in various NLP tasks. The present invention uses the BERT large language model to perform an in-depth analysis of the true meaning of each word data in the corresponding sentence, thereby obtaining an analysis result of whether it belongs to prescription information.

[0035] Of course, the general large language model can also be an improved model of the BERT large language model, such as RoBERTa: the training method of BERT is optimized, including longer training time, larger data sets and dynamic masking strategies; XLNet: a better training method is introduced, using the permutation language model instead of the mask language model to solve the problems that may be caused by the NSP task in BERT; ERNIE: a model proposed by Baidu, which improves the model's ability to understand Chinese by introducing Chinese corpus such as knowledge graphs for pre-training.

[0036] Optionally, the second word dataset and the TCM knowledge graph are input into a pre-built deep learning model, and the deep learning model outputs a number of prescription information, including: Determine the text generation year information corresponding to each second word data in the second word data set, input the text generation year information and the traditional Chinese medicine knowledge graph into a second general large language model, and the second general large language model outputs a traditional Chinese medicine knowledge sub-graph matching the text generation year information; The second word dataset and the traditional Chinese medicine knowledge sub-graph are input into a pre-built deep learning model, and the deep learning model outputs a number of prescription information.

[0037] In the embodiment of the present invention, as mentioned above, the TCM text data is divided into ancient texts and modern documents, that is, the prescription contents in these received TCM texts actually have corresponding generation years. At the same time, medical theories in different periods have significant differences in drug use, treatment methods, compilation rules of medical literature, etc., which requires the use of the TCM knowledge graph of the corresponding period to conduct in-depth analysis of the second word data set, so as to obtain more accurate prescription information. However, the above-mentioned TCM knowledge sub-graph is constructed based on the latest and revised TCM knowledge. If the TCM knowledge sub-graph is used to analyze each second word data in the second word data set, the prescription information obtained is likely to be biased or even wrong, especially for ancient texts.

[0038] In this regard, the present invention also constructs a second universal large language model, and its construction method is the same as the aforementioned first universal large language model, which will not be repeated here. The aforementioned received Chinese medicine text data is configured with corresponding generation age information, based on which the text generation age information corresponding to each second word data in the second word data set can be determined, such as the Ming Dynasty, the current period, etc. The text generation age information and the Chinese medicine knowledge map are input into the second universal large language model, and the second universal large language model can obtain Chinese medicine related documents of the period corresponding to the generation age information on the Internet, such as documents introducing Chinese medicine theoretical knowledge or knowledge differences in several periods. Based on this information, it is possible to correct certain knowledge content in the Chinese medicine knowledge map, thereby obtaining a Chinese medicine knowledge sub-map that matches the text generation age information. Finally, the Chinese medicine knowledge sub-map and the aforementioned second word data set are input into the pre-constructed deep learning model, and the deep learning model performs a deep analysis on each second word data under the guidance of the Chinese medicine knowledge sub-map, thereby obtaining the corresponding prescription information.

[0039] It should be noted that the above-mentioned deep learning model can also be constructed based on the above-mentioned BERT large language model, that is, it constitutes a third general large language model. The advantage of such a setting is that: in reality, there is a general lack of annotated traditional Chinese medicine texts. If a vertical special model is constructed, it is easy to cause insufficient training intensity of the model due to insufficient training data, and ultimately fail to obtain accurate prediction results; and the general large language model itself has been trained in a relatively deep manner. When in use, only a small sample data is needed to fine-tune it, and a special model suitable for the above-mentioned several analysis and prediction tasks of the present invention can be obtained. Of course, the teacher-student network can also be used to slim down the special model so that it only retains the corresponding single analysis and prediction function, which can reduce the size of the model.

[0040] Optionally, the step of inputting the text generation time information and the TCM knowledge graph into a second general large language model, and the second general large language model outputting a TCM knowledge sub-graph matching the text generation time information, further includes: When the output prescription information reaches a preset number, the execution of the method is suspended, and the output corresponding prescription information is manually verified based on the traditional Chinese medicine text data, and the error rate is statistically obtained; A correction value is determined according to the error rate, and the text generation year information is downward expanded based on the correction value. The text generation year information and the traditional Chinese medicine knowledge graph after downward expansion are input into a second general large language model, and the second general large language model outputs the traditional Chinese medicine knowledge sub-graph that matches the text generation year information.

[0041] In an embodiment of the present invention, the above scheme of the present invention is to analyze each TCM text data in the TCM text data set one by one, and output the corresponding prescription information. Another improvement of the present invention is that when the output prescription information reaches a preset number (for example, 100), the execution of the method of the present invention is suspended, and the corresponding output prescription information is manually verified based on the TCM text data to determine whether there is an obvious recognition error, and the error rate is obtained based on the statistics. Then, the corresponding correction value is determined based on the statistical error rate, and the correction value is used to expand the aforementioned text generation age information downward, that is, to expand to the adjacent "newer" period, and the downward expanded text generation age information and TCM knowledge graph are input into the second general large language model, and the second general large language model outputs a TCM knowledge sub-graph based on the appropriate presence of "new" TCM knowledge content for in-depth analysis to obtain more accurate prescription information.

[0042] Optionally, the calculation formula of the correction value is as follows: ; In the formula, is the correction value, in years; is a constant; is the error rate.

[0043] In the embodiment of the present invention, the reason for the high error rate of prescription information recognition may be that the text generation year information determined above is not accurate enough, and more likely that the above-mentioned type of TCM related literature corresponding to the above-mentioned text generation year information is insufficient, resulting in the constructed TCM knowledge sub-graph being inaccurate. In this regard, the present invention determines the correction value based on the error rate and the corresponding positive correlation, and the correction value is at least Year, It can be 50 years, 100 years, etc., and can be determined according to the corresponding characteristics of different periods. For example, if there are sufficient TCM literature of the above type in a certain period, then the corresponding The smaller the year value, the larger the value.

[0044] Therefore, by appropriately expanding the text generation year information toward the current period, the supplement and "update" of TCM knowledge in the TCM knowledge sub-graph can be achieved, which is more conducive to adapting the obtained TCM knowledge sub-graph to the second word data of the above-mentioned text generation year information.

[0045] It should be noted that the prescription information extracted in the present invention is not a direct text extraction of the prescription content in the TCM text data, but based on the extracted prescription text information, the corresponding TCM knowledge sub-graph is integrated to correct some erroneous information (limited by the cognitive level at the time or caused by typos, etc.) to generate corresponding prescription information. The probability of the prescription information being verified to be valid in the future will also be greater. At the same time, the actual TCM knowledge graph is constantly developing and changing, but the development and changes in similar periods are related to each other. For example, a prescription for a certain type of disease in the early period was verified to be effective, but its medical theory was not understood, and when correcting the numerical value, the prescription information was verified to be effective. After the corresponding year, the prescription has updated the medical theory and medication method of this type of disease. Therefore, by appropriately extending the text generation year information to the current period, it is helpful to improve the above-mentioned TCM knowledge sub-graph, thereby facilitating the acquisition of more accurate prescription information.

[0046] like Figure 3 As shown, the embodiment of the present invention also discloses a system for intelligently capturing key information of traditional Chinese medicine texts, including a processing module and a storage module, wherein a computer program is stored in the storage module, and the processing module executes the computer program to implement the following method steps: Receiving a traditional Chinese medicine text data set, and preprocessing each traditional Chinese medicine text data in the traditional Chinese medicine text data set; Use a word vector model to extract word features from each of the TCM text data to obtain a first word data set, and perform semantic analysis on each word data in the first word data set to obtain a streamlined second word data set; Inputting the second word dataset and the TCM knowledge graph into a pre-built deep learning model, and the deep learning model outputs a number of prescription information; Repeat the above process to complete the processing of all the traditional Chinese medicine text data, and integrate the obtained prescription information.

[0047] An embodiment of the present invention further discloses an electronic device, comprising: a memory storing executable program code; a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the method described in the above embodiment.

[0048] The embodiment of the present invention further discloses a computer storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method described in the above embodiment is executed.

[0049] An embodiment of the present invention further discloses a computer program product, which includes computer code. When the computer code is executed by a processor of an electronic device, the method described in the above embodiment is implemented.

[0050] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0051] The method and related apparatus provided by the embodiment of the present application are described with reference to the method flow chart and / or structural diagram provided by the embodiment of the present application. Specifically, each process and / or box in the method flow chart and / or structural diagram, as well as the combination of the processes and / or boxes in the flow chart 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 processor 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 process in the process. Figure 1 Schematic diagram of one or more processes and / or structures Figure 1These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 Schematic diagram of one or more processes and / or structures Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A flow or multiple flows and / or structures illustrate the steps of the functions specified in one block or multiple blocks.

[0052] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A method for intelligently capturing key information of traditional Chinese medicine texts, characterized in that: The steps include: Receiving a traditional Chinese medicine text data set, and preprocessing each traditional Chinese medicine text data in the traditional Chinese medicine text data set; Use a word vector model to extract word features from each of the TCM text data to obtain a first word data set, and perform semantic analysis on each word data in the first word data set to obtain a streamlined second word data set; Inputting the second word dataset and the TCM knowledge graph into a pre-built deep learning model, and the deep learning model outputs a number of prescription information; Repeat the above process to complete the processing of all the traditional Chinese medicine text data, and integrate the obtained prescription information.

2. The method for intelligently capturing key information of traditional Chinese medicine text according to claim 1, characterized in that: Preprocessing each TCM text data in the TCM text data set includes: The text of each TCM text data is standardized, including word segmentation and removal of stop words; then, redundant data in each TCM text data after the standardized processing is removed.

3. The method for intelligently capturing key information of traditional Chinese medicine text according to claim 2, characterized in that: Using a word vector model to extract word features from each of the TCM text data to obtain a first word data set, performing semantic analysis on each word data in the first word data set to obtain a streamlined second word data set, including: Use the word vector model to extract word features from each of the TCM text data to obtain the corresponding first word data sets; Each word data in the first word data set is restored to the corresponding position of the original traditional Chinese medicine text data, and the word data is semantically analyzed based on the context sentence corresponding to the corresponding position using the first general large language model. If the semantic analysis result shows that it belongs to prescription information, the word data is retained, otherwise it is deleted, thereby obtaining a streamlined second word data set.

4. The method for intelligently capturing key information of traditional Chinese medicine text according to claim 3 is characterized by: The second word dataset and the TCM knowledge graph are input into a pre-built deep learning model, and the deep learning model outputs a number of prescription information, including: Determine the text generation year information corresponding to each second word data in the second word data set, input the text generation year information and the traditional Chinese medicine knowledge graph into a second general large language model, and the second general large language model outputs a traditional Chinese medicine knowledge sub-graph matching the text generation year information; The second word dataset and the traditional Chinese medicine knowledge sub-graph are input into a pre-built deep learning model, and the deep learning model outputs a number of prescription information.

5. The method for intelligently capturing key information of traditional Chinese medicine text according to claim 4 is characterized by: Inputting the text generation time information and the TCM knowledge graph into a second general large language model, and the second general large language model outputting a TCM knowledge sub-graph matching the text generation time information, further comprising: When the output prescription information reaches a preset number, the execution of the method is suspended, and the output corresponding prescription information is manually verified based on the traditional Chinese medicine text data, and the error rate is statistically obtained; A correction value is determined according to the error rate, and the text generation year information is downward expanded based on the correction value. The text generation year information and the traditional Chinese medicine knowledge graph after downward expansion are input into a second general large language model, and the second general large language model outputs the traditional Chinese medicine knowledge sub-graph that matches the text generation year information.

6. The method for intelligently capturing key information of traditional Chinese medicine text according to claim 5, characterized in that: The calculation formula of the correction value is as follows: ； In the formula, is the correction value, in years; is a constant; is the error rate.

7. A system for intelligently capturing key information of traditional Chinese medicine texts, comprising a processing module and a storage module, wherein a computer program is stored in the storage module, characterized in that: The processing module executes the computer program to implement the following method steps: Receiving a traditional Chinese medicine text data set, and preprocessing each traditional Chinese medicine text data in the traditional Chinese medicine text data set; Use a word vector model to extract word features from each of the TCM text data to obtain a first word data set, and perform semantic analysis on each word data in the first word data set to obtain a streamlined second word data set; Inputting the second word dataset and the TCM knowledge graph into a pre-built deep learning model, and the deep learning model outputs a number of prescription information; Repeat the above process to complete the processing of all the traditional Chinese medicine text data, and integrate the obtained prescription information.

8. An electronic device comprising: A memory storing executable program code; A processor coupled to the memory; characterized in that: the processor calls the executable program code stored in the memory to execute the method according to any one of claims 1-6.

9. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is executed.

10. A computer program product, characterized in that: The computer program product includes computer codes, and when the computer codes are executed by a processor of an electronic device, the method according to any one of claims 1 to 6 is implemented.

Images