A data analysis method and system for TCM constitution identification and conditioning

Abstract

The present invention relates to the field of data processing technology, and in particular to a data analysis method and system for TCM constitution identification and conditioning. When processing data for TCM constitution identification and conditioning, the present invention first divides the data into questionnaire data and doctor diagnosis data. For the questionnaire data, the data is divided into multiple types according to whether the data in the questionnaire data is easy to identify. Different data mapping methods are used for different types of data, thereby improving the accuracy of data mapping and further improving the accuracy of data processing. At the same time, in view of the characteristics that the data for TCM constitution identification and conditioning have different collection subjects, the method of setting accuracy weight assignment is used to set different accuracy weights for different types of data, thereby weakening the influence of inaccurate data on the results as much as possible, thereby improving the accuracy of TCM constitution identification data analysis.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a data analysis method and system for TCM constitution identification and conditioning. Background Art

[0002] As a component of the theoretical system of traditional Chinese medicine, constitution shows a relatively stable characteristic in terms of morphology, physiology and psychology. The identification of traditional Chinese medicine constitution is based on the theory of traditional Chinese medicine. According to the characteristics of human body shape, psychology and physiology, the constitution type is determined through comprehensive analysis. At present, the classification standard of traditional Chinese medicine constitution is Wang Qi's nine-point method, namely, nine basic types such as peaceful constitution, qi deficiency constitution, yang deficiency constitution, yin deficiency constitution, phlegm-damp constitution, damp-heat constitution, blood fatigue constitution, qi stagnation constitution and special constitution. Among them, peaceful constitution means that the physiological and psychological state of a person is normal, and the other eight constitutions (collectively referred to as non-peaceful constitution in this article) all indicate that a person may be in a sub-healthy state, which is more likely to evolve into certain diseases than peaceful constitution. Therefore, it is of great significance to identify peaceful constitution and non-peaceful constitution to judge the health status of the human body, seek medical treatment or make adjustments in time, and prevent diseases to achieve "preventive treatment".

[0003] Nowadays, we have entered the era of big data. Data mining and processing are mainly used to discover the hidden rules or relationships between data from a large amount of data. Usually, machine automatic recognition is adopted without too much human intervention. The use of data processing technology can provide intelligent and automated auxiliary means for decision analysis and formulation. With the promotion of medical informatization in medical and health service-related institutions, a large amount of health data has been accumulated in the field of medical and health care. Therefore, data mining and processing technology has great application potential in the field of medical care. However, there is still a lack of large-scale, standardized and unified databases in the field of Chinese medicine constitution, and big data methods have not been fully applied to this field, resulting in the inability to effectively utilize the information contained in a large amount of Chinese medicine constitution data generated by clinical, basic and health management practices, and it is difficult to guide the scientific research and industrialization practices in various fields of Chinese medicine. With the passage of time, health data will accumulate more and more, and the correlation between data will become more complex. Therefore, the application of effective data mining and processing technology in the field of medical care, especially in the field of Chinese medicine constitution, has great practical significance for the development of the medical field.

[0004] In the prior art, when processing data for TCM constitution identification and conditioning, only simple processing procedures such as data preprocessing and data mapping are performed. However, for medical data, different professional vocabulary may have different expressions, and some vocabulary has subtle differences but the meanings expressed are not exactly the same. Therefore, simple data mapping processing leads to low accuracy of data processing; and due to the unique characteristics of data in the medical field, data mapping processing methods in other fields do not have a good conversion application effect in the field of medical data; and, since TCM constitution identification and conditioning data come from different collection entities, there may be deviations in accuracy, which also affects the processing accuracy of TCM constitution identification and conditioning data. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a data analysis method and system for TCM constitution identification and conditioning, and proposes a data processing method that can improve data processing accuracy based on the data characteristics of TCM constitution identification and conditioning.

[0006] The present invention provides a data analysis method for TCM constitution identification and conditioning, comprising:

[0007] S1: Obtain data for TCM constitution identification and conditioning;

[0008] S2: preprocessing the data used for TCM constitution identification and conditioning;

[0009] S3: classifying the data used for TCM constitution identification and conditioning;

[0010] In S3, the data used for TCM constitution identification and conditioning are classified into a first data type, a second data type and a third data type, wherein the first data type is a non-confusing data type, the second data type is a field name easily confused data type, and the third data type is an option information easily confused data type;

[0011] S4: Perform differentiated data processing operations on different types of data;

[0012] The S4 is specifically:

[0013] S4.1: directly performing text data mapping on the data included in the first data type, mapping the text data included in the first data type to a digital data interval;

[0014] S4.2: Performing text data mapping on the data included in the second data type by table lookup, mapping the text data included in the second data type to a digital data interval;

[0015] S4.3: Performing text data mapping on the data included in the third data type through the first machine learning model, mapping the text data included in the third data type to a digital data interval;

[0016] S5: Assigning accuracy weights to the data used for TCM constitution identification and conditioning;

[0017] S6: Obtaining a data sample set for TCM constitution identification and conditioning according to the accuracy weight assignment;

[0018] S7: Input the data sample set into the second machine learning model to implement data analysis of TCM constitution identification and conditioning.

[0019] Preferably, the data includes questionnaire data and doctor's diagnosis data.

[0020] Preferably, the data preprocessing operation includes: outlier filtering and invalid sample deletion;

[0021] The outlier filtering is to remove information in the questionnaire data that is obviously inconsistent with reality;

[0022] The invalid sample deletion is: taking the survey object's questionnaire data as a sample, if a preset number of abnormal values ​​appear in the sample, the survey object's questionnaire data is considered invalid, and the survey object's questionnaire data is deleted;

[0023] Preferably, the preset number of times is 3;

[0024] Preferably, the S4.3 is specifically:

[0025] S4.3.1: Collecting a preset number of questionnaires filled in by users, and screening out information of a third data type from the questionnaires filled in by the preset number of users;

[0026] S4.3.2: Manually labeling the information of the third data type in the questionnaires filled in by the preset number of users, thereby forming a sample set;

[0027] S4.3.3: Using the sample set to train a first machine learning model;

[0028] S4.3.4: Input the data contained in the third data type to be mapped into the first machine learning model to obtain a mapping result.

[0029] Preferably, in S4.3.1, the preset quantity is 10,000 copies.

[0030] Preferably, in S4.3.3, the first machine learning model is one of an artificial neural network model, a recurrent neural network model, and a convolutional neural network model.

[0031] Preferably, in S5, the process of assigning the accuracy weight is specifically as follows: for the questionnaire data, hiring professionals to score the quality of the questionnaire according to the number of abnormal values ​​in S2, and hiring professionals to score the collection quality of the tongue diagnosis, facial diagnosis, sound diagnosis, and pulse wave signal data of the doctor's diagnosis data; then normalizing the data according to the quality score of the questionnaire and the collection quality score of the doctor's diagnosis data to obtain the accuracy weight assignments for the questionnaire data and the doctor's diagnosis data, respectively.

[0032] Preferably, in S6, the generation process of the sample set S is specifically as follows:

[0033]

[0034] Among them, a and b are the accuracy weight assignments of the questionnaire data and the accuracy weight assignments of the doctor's diagnosis data respectively, S1 is the questionnaire data set, and S2 is the doctor's diagnosis data set.

[0035] According to another aspect of the present invention, a data processing system for TCM constitution identification and conditioning is provided, wherein the system adopts any of the above-mentioned data analysis methods for TCM constitution identification and conditioning, and the system comprises:

[0036] A data acquisition module, used to acquire data for TCM constitution identification and conditioning;

[0037] A preprocessing module, used for preprocessing the data used for TCM constitution identification and conditioning;

[0038] A data type classification module, for classifying the data used for TCM constitution identification and conditioning;

[0039] A differentiated data processing module is used to perform differentiated data processing operations on different types of data;

[0040] An accuracy weight assignment module, used for assigning accuracy weights to the data used for TCM constitution identification and conditioning;

[0041] A sample set generation module, used to obtain a data sample set for TCM constitution identification and conditioning according to the accuracy weight assignment;

[0042] The data analysis module is used to input the data sample set into the second machine learning model to realize data analysis of TCM constitution identification and conditioning.

[0043] The embodiments of the present invention have the following technical effects:

[0044] When processing the data of TCM constitution identification and conditioning, the present invention first divides the data into questionnaire data and doctor diagnosis data. For the questionnaire data, the data is divided into multiple types according to whether the data in the questionnaire data is easy to identify. Different data mapping methods are used for different types of data, thereby improving the accuracy of data mapping and further improving the accuracy of data analysis.

[0045] At the same time, in view of the characteristics of different data collection subjects in TCM constitution identification and conditioning, the method of setting accuracy weight assignment is adopted to set different accuracy weights for different types of data, so as to weaken the impact of inaccurate data on the results as much as possible, thereby improving the accuracy of TCM constitution identification. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0047] Figure 1 is a flow chart of a data analysis method for TCM constitution identification and conditioning provided by Example 1 of the present invention;

[0048] Figure 2 It is a flowchart of performing differentiated data processing operations on different types of data provided by Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. 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 belong to the scope of protection of the present invention.

[0050] Embodiment 1, Figure 1 1 is a flow chart of a data analysis method for TCM constitution identification and conditioning provided by Example 1 of the present invention. Figure 1 , a data analysis method for TCM constitution identification and conditioning, specifically comprising:

[0051] S1: Obtain data for TCM constitution identification and conditioning;

[0052] Wherein, the data includes questionnaire data and doctor's diagnosis data;

[0053] The questionnaire data includes the basic information, lifestyle and risk factors of the survey subjects; the questionnaire data is sent to the survey subjects to invite them to fill in the questionnaire online; it is worth noting that in this embodiment, the meaning of filling in the questionnaire online includes but is not limited to filling in the questionnaire data through any method such as mobile phone app, Internet website, computer program, etc.

[0054] For example, the sample questionnaire is shown in Table 1, which is a sample questionnaire derived from the TCM constitution identification and conditioning system:

[0055]

[0056]

[0057]

[0058]

[0059] The doctor's diagnosis data includes data obtained by the doctor through tongue diagnosis, face diagnosis, voice diagnosis, and pulse wave signal diagnosis of the survey subject; the doctor's diagnosis data is obtained by the doctor performing actual diagnosis on the survey subject on the spot, and filling in the data acquisition module of TCM constitution identification and conditioning online; in this embodiment, the doctor's diagnosis data is image data and digital wave data;

[0060] S2: preprocessing the data used for TCM constitution identification and conditioning;

[0061] Now we have entered the era of big data. Data mining technology can be used to mine medical data to obtain the human health information that people want to obtain. Data mining is mainly used to discover the hidden rules or relationships between data from a large amount of data. It usually adopts the method of automatic machine recognition without too much human intervention. The use of data mining technology can provide intelligent and automated auxiliary means for decision analysis and formulation. With the promotion of medical informatization in medical and health service-related institutions, a large amount of health data has been accumulated in the medical and health field. Therefore, data mining technology has great application potential in the field of medical care.

[0062] The effectiveness of data mining is closely related to the quality of the data obtained. Generally, during the data acquisition stage, the data obtained may be affected by the mood, time, cultural level, feelings, etc. of the survey subjects, and the quality of the data obtained may vary. At this stage, the technical means that can be adopted are relatively limited. Therefore, it is generally necessary to perform data preprocessing operations on the obtained data.

[0063] The data preprocessing operations include: outlier filtering and invalid sample deletion;

[0064] Specifically, the outlier filtering is: removing information in the questionnaire data that is obviously inconsistent with reality;

[0065] For example, for the date of birth information, if January 1, 1900 appears, this information is obviously inconsistent with the actual situation. Therefore, during preprocessing, this information should be deleted. For example, the estimated daily water intake appears to be 10,000L. This information is also obviously inconsistent with the actual situation. Therefore, during preprocessing, this information should be deleted;

[0066] Specifically, the invalid sample deletion is: taking the survey object's questionnaire data as a sample, if a preset number of abnormal values ​​appear in the sample, the survey object's questionnaire data is considered invalid, and the survey object's questionnaire data is deleted;

[0067] Furthermore, the preset number of times is 3;

[0068] For example, for Wan Moumou (name ID number 0113), there are multiple abnormal values ​​in his questionnaire data (born on January 1, 1900, drinking 100 times / week, 100 liang / time, poor sleep 0 hours / day, regular exercise 200 times / week). It can be seen that the survey subject may have an improper attitude and be perfunctory when filling in the questionnaire information. The accuracy of other information of the survey subject may not be very high. Therefore, the spread questionnaire output of the spread subject is treated as invalid data and deleted;

[0069] S3: classifying the data used for TCM constitution identification and conditioning;

[0070] In the data mining and processing of TCM constitution identification and conditioning, due to the characteristics of complex data and different meanings of similar words, when mining the data of TCM constitution identification and conditioning, the data mining model may easily mistake similar but not completely identical information for the same information, thus affecting the accuracy of data mining. For example, hypertension and hyperlipidemia in personal medical history, and field names such as gynecological vulva, gynecological vagina, gynecological cervix, and gynecological uterine body, have different meanings. The mining model may think that they are the same information, thus affecting the accuracy of data mining. For example, high blood sugar and diabetes represent the same disease type, but the above information may not be identified as the same disease when the machine recognizes it, thus affecting the accuracy of data processing.

[0071] Therefore, in this embodiment, the data used for TCM constitution identification and conditioning are classified into a first data type, a second data type and a third data type, wherein the first data type is a non-confusing data type, the second data type is a field name easily confused data type, and the third data type is an option information easily confused data type;

[0072] Exemplarily, the first data type includes the basic information table in the questionnaire sample form, in which the field names and options of "Lifestyle and Risk Factors" such as smoking, dietary habits, drinking, exercise, sleep, life stress, etc. are not easily recognized by computers; the second data type includes information with easily confused field names such as gynecological vulva, gynecological vagina, gynecological cervix, and gynecological uterine body; the third data type includes information with easily confused option information in personal disease history and family disease history;

[0073] S4: Perform differentiated data processing operations on different types of data;

[0074] In step 3, the data of TCM identification and conditioning are classified according to whether they are easy to be confused. In this step, different types of data are processed differently according to the classification in step 3 to improve the accuracy of data processing and convert the questionnaire data into information that can be accurately recognized by the computer;

[0075] Among them, Figure 2 As shown, the S4 is specifically:

[0076] S4.1: directly performing text data mapping on the data included in the first data type, mapping the text data included in the first data type to a digital data interval;

[0077] For example, the names of the survey subjects may be numbered, for example, the name of the first survey subject is replaced by 1; for example, the gender of the survey subject, male is mapped to number 1, and female is mapped to number 2; for example, the text data of exercise status is mapped to number 1, and frequent exercise is mapped to number 1, occasional exercise is mapped to number 2, and no exercise is mapped to number 3;

[0078] S4.2: Performing text data mapping on the data included in the second data type by table lookup, mapping the text data included in the second data type to a digital data interval;

[0079] Since the second data type is a data type whose field name is easily confused, for a fixed sample questionnaire form, the field names in the sample form are generally relatively fixed. Therefore, a field name-number correspondence table can be set in advance, and the text data can be mapped to the digital data interval through the table to improve the accuracy of text data mapping, thereby improving the accuracy of data processing;

[0080] Specifically, S4.2 is specifically: establishing a field name and number correspondence table, considering the field names contained in the data of the second data type to be mapped as fixed numbers, and then mapping the field names contained in the data of the second data type in the questionnaire to specific numbers according to the field name and number correspondence table, thereby completing the digital mapping of this type of text data;

[0081] S4.3: Performing text data mapping on the data included in the third data type through the first machine learning model, mapping the text data included in the third data type to a digital data interval;

[0082] Since the third data type is an option information data type that is easily confused, and its information is filled in by the user according to actual conditions, a simple mapping table cannot meet the mapping requirements of this type of information. The machine learning model has a strong learning ability and therefore has a unique advantage in identifying unspecific information. Therefore, in this embodiment, the data contained in the third data type is mapped to text data through a machine learning model through a model to improve the mapping accuracy of this type of information.

[0083] Specifically, the S4.3 is as follows:

[0084] S4.3.1: Collecting a preset number of questionnaires filled in by users, and screening out information of a third data type from the questionnaires filled in by the preset number of users;

[0085] Wherein, the preset quantity is 10,000 copies;

[0086] S4.3.2: Manually labeling the information of the third data type in the questionnaires filled in by the preset number of users, thereby forming a sample set;

[0087] Specifically, hiring annotators with a medical background to manually annotate the information of the third data type in the questionnaires filled out by the preset number of users, and setting up an information review process to improve the accuracy of the annotations;

[0088] S4.3.3: Using the sample set to train a first machine learning model;

[0089] The first machine learning model is one of an artificial neural network model, a recurrent neural network model, and a convolutional neural network model;

[0090] S4.3.4: Input the data contained in the third data type to be mapped into the first machine learning model to obtain a mapping result.

[0091] In this step S4.3, through the above S4.3.1-S4.3.4, for the information in the questionnaire where the option information is easily confused, a good data mapping result is obtained;

[0092] It is worth emphasizing that the doctor's diagnosis data is image data and digital wave data, and therefore, no mapping processing is required.

[0093] S5: Assigning accuracy weights to the data used for TCM constitution identification and conditioning;

[0094] Through the above introduction, the data used for TCM constitution identification and conditioning includes questionnaire data and doctor's diagnosis data; wherein, the questionnaire data is filled in by the respondents themselves, and the doctor's diagnosis data is filled in by the doctor according to the respondents. The above two types of data play an important role in TCM constitution identification and conditioning, and their accuracy also affects the accuracy of the final identification result; however, since the above two types of data are acquired by different subjects, the accuracy of the data acquired by different subjects may deviate for the data of different respondents. Therefore, compared with the prior art, this embodiment sets different accuracy weights for different types of data by setting accuracy weight assignment, thereby weakening the impact of inaccurate data on the results as much as possible, thereby improving the accuracy of TCM constitution identification.

[0095] In this embodiment, the accuracy weight assignment process is specifically as follows: for the questionnaire data, hire professionals to score the quality of the questionnaire according to the number of abnormal values ​​in S2, and hire professionals to score the collection quality of the tongue diagnosis, facial diagnosis, sound diagnosis, and pulse wave signal data of the doctor's diagnosis data; then perform normalization processing based on the quality score of the questionnaire and the collection quality score of the doctor's diagnosis data, and obtain the accuracy weight assignment of the questionnaire data and the doctor's diagnosis data respectively;

[0096] For example, the questionnaire quality score of the survey subject No. 0006 is 6, and the collection quality score of the doctor's diagnosis data is 7. Then the normalized result a of the questionnaire quality score is: a=6 / (6+7)=0.46; the normalized result b of the collection quality score of the doctor's diagnosis data is: b=7 / (6+7)=0.54;

[0097] S6: Obtaining a data sample set for TCM constitution identification and conditioning according to the accuracy weight assignment;

[0098] Specifically, the generation process of the sample set S is as follows:

[0099]

[0100] Wherein, a and b are respectively the accuracy weight assignments of the questionnaire data and the accuracy weight assignments of the doctor's diagnosis data, S1 is the questionnaire data set, and S2 is the doctor's diagnosis data set;

[0101] S7: Inputting the data sample set into a second machine learning model to implement data analysis for TCM constitution identification and conditioning;

[0102] Specifically, the sample set S is input into the second machine learning model to achieve TCM constitution identification and conditioning results;

[0103] The second machine learning module is one of an artificial neural network model, a recurrent neural network model, and a convolutional neural network model;

[0104] S7 specifically includes: inputting the sample set obtained in S6 into the second machine learning model, and the output of the second machine learning is the constitution identification result;

[0105] Among them, the constitution identification result is specifically one of: balanced constitution, qi deficiency constitution, yang deficiency constitution, yin deficiency constitution, phlegm-damp constitution, damp-heat constitution, blood fatigue constitution, qi stagnation constitution, and special constitution.

[0106] Embodiment 2, this embodiment discloses a data processing system for TCM constitution identification and conditioning, the system adopts a data analysis method for TCM constitution identification and conditioning of embodiment 1, the system comprises:

[0107] A data acquisition module, used to acquire data for TCM constitution identification and conditioning;

[0108] A preprocessing module, used for preprocessing the data used for TCM constitution identification and conditioning;

[0109] A data type classification module, for classifying the data used for TCM constitution identification and conditioning;

[0110] A differentiated data processing module is used to perform differentiated data processing operations on different types of data;

[0111] An accuracy weight assignment module, used for assigning accuracy weights to the data used for TCM constitution identification and conditioning;

[0112] A sample set generation module, used to obtain a data sample set for TCM constitution identification and conditioning according to the accuracy weight assignment;

[0113] The data analysis module is used to input the data sample set into the second machine learning model to realize data analysis of TCM constitution identification and conditioning.

[0114] Embodiment 3: This embodiment discloses an electronic device, which includes one or more processors and a memory.

[0115] The processor may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0116] The memory may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor may run the program instructions to implement a data analysis method for TCM constitution identification and conditioning and / or other desired functions of any embodiment of the present application described above. Various contents such as initial external parameters, thresholds, etc. may also be stored in the computer-readable storage medium.

[0117] In one example, the electronic device may further include: an input device and an output device, and these components are interconnected through a bus system and / or other forms of connection mechanisms (not shown). The input device may include, for example, a keyboard, a mouse, etc. The output device may output various information to the outside, including early warning information, braking force, etc. The output device may include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, etc.

[0118] In addition, the electronic device may further include any other appropriate components according to specific application scenarios.

[0119] In addition to the above-mentioned methods and devices, the embodiments of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute each step or all steps of a data analysis method for TCM constitution identification and conditioning provided in any embodiment of the present application.

[0120] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0121] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps of a data analysis method for TCM constitution identification and conditioning provided in any embodiment of the present application.

[0122] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0123] It should be noted that the terms used in the present invention are only for describing specific embodiments, rather than limiting the scope of the present application. As shown in the present specification, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of more restrictions, the elements defined by the sentence "include one..." do not exclude the presence of other identical elements in the process, method or device including the elements.

[0124] It should also be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying 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 on the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", etc. should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A data analysis method for TCM constitution identification and conditioning, characterized in that: include: S1: Acquire data for TCM constitution identification and conditioning; the data includes questionnaire data and doctor's diagnosis data; S2: preprocessing the data used for TCM constitution identification and conditioning; The data preprocessing operations include: outlier filtering and invalid sample deletion; The outlier filtering is to remove information in the questionnaire data that is obviously inconsistent with reality; The invalid sample deletion is: taking the survey object's questionnaire data as a sample, if a preset number of abnormal values ​​appear in the sample, the survey object's questionnaire data is considered invalid, and the survey object's questionnaire data is deleted; S3: classifying the data used for TCM constitution identification and conditioning; In S3, the data used for TCM constitution identification and conditioning are classified into a first data type, a second data type and a third data type, wherein the first data type is a non-confusing data type, the second data type is a field name easily confused data type, and the third data type is an option information easily confused data type; S4: Perform differentiated data processing operations on different types of data; The S4 is specifically: S4.1: directly performing text data mapping on the data included in the first data type, mapping the text data included in the first data type to a digital data interval; S4.2: Performing text data mapping on the data included in the second data type by table lookup, mapping the text data included in the second data type to a digital data interval; S4.3: Performing text data mapping on the data contained in the third data type through the first machine learning model, mapping the text data contained in the third data type to a digital data interval; S4.3 is specifically: S4.3.1: Collecting a preset number of questionnaires filled in by users, and screening out information of a third data type from the questionnaires filled in by the preset number of users; S4.3.2: Manually labeling the information of the third data type in the questionnaires filled in by the preset number of users, thereby forming a sample set; S4.3.3: Using the sample set to train a first machine learning model; in S4.3.3, the first machine learning model is one of an artificial neural network model, a recurrent neural network model, and a convolutional neural network model; S4.3.4: inputting the data contained in the third data type to be mapped into the first machine learning model to obtain a mapping result; S5: Assigning accuracy weights to the data used for TCM constitution identification and conditioning; in S5, the process of assigning accuracy weights is specifically as follows: for the questionnaire data, hiring professionals to score the quality of the questionnaire according to the number of abnormal values ​​in S2, and hiring professionals to score the collection quality of the tongue diagnosis, facial diagnosis, voice diagnosis, and pulse wave signal data of the doctor's diagnosis data; then normalizing the questionnaire quality score and the doctor's diagnosis data collection quality score to obtain the accuracy weight assignment of the questionnaire data and the accuracy weight assignment of the doctor's diagnosis data respectively; S6: Obtaining a data sample set for TCM constitution identification and conditioning according to the accuracy weight assignment; In S6, the generation process of the sample set S is specifically as follows: Wherein, a and b are respectively the accuracy weight assignments of the questionnaire data and the accuracy weight assignments of the doctor's diagnosis data, S1 is the questionnaire data set, and S2 is the doctor's diagnosis data set; S7: Input the data sample set into the second machine learning model to implement data analysis of TCM constitution identification and conditioning.

2. A data analysis method for TCM constitution identification and conditioning according to claim 1, characterized in that: The preset number of times is 3.

3. A data analysis method for TCM constitution identification and conditioning according to claim 1, characterized in that: In S4.3.1, the preset quantity is 10,000 copies.

4. A data processing system for TCM constitution identification and conditioning, characterized in that: The system adopts the data processing method for TCM constitution identification and conditioning according to any one of claims 1 to 3, and the system comprises: A data acquisition module, used to acquire data for TCM constitution identification and conditioning; A preprocessing module, used for preprocessing the data used for TCM constitution identification and conditioning; A data type classification module, for classifying the data used for TCM constitution identification and conditioning; A differentiated data processing module is used to perform differentiated data processing operations on different types of data; An accuracy weight assignment module, used for assigning accuracy weights to the data used for TCM constitution identification and conditioning; A sample set generation module, used to obtain a data sample set for TCM constitution identification and conditioning according to the accuracy weight assignment; The data analysis module is used to input the data sample set into the second machine learning model to realize data analysis of TCM constitution identification and conditioning.

Images