Construction method and device of traditional Chinese medicine experience knowledge base and prescription determination method

Abstract

The invention discloses a traditional Chinese medicine experience knowledge base construction method and device and a prescription determination method, and the construction method comprises the steps: constructing a tongue pulse-symptom relation graph according to case data, calculating the mutual information degree of tongue pulse symptoms and prescriptions, generating a mutual reliability table, and carrying out the clustering of tongue pulse-symptom data to generate a clustering table; according to the method, a traditional Chinese medicine experience knowledge base is formed through integration, so that tongue vein data, symptom data and prescription data of case data can be scientifically and systematically analyzed, the relationship among the tongue vein data, the symptom data and the prescription data is visually displayed, and effective scientific analysis data support is provided for traditional Chinese medicine diagnosis and treatment; the accuracy of clinical diagnosis and treatment of the traditional Chinese medicine is improved, and the problem that the existing traditional Chinese medicine experience lacks effective support for formulating diagnosis and treatment schemes of the traditional Chinese medicine due to the lack of confluence analysis on the relationship among the tongue vein data, the symptom data and the prescription data is solved.

Description

Technical Field

[0001] The present invention relates to the field of traditional Chinese medicine experience knowledge bases, and particularly to a method for constructing a traditional Chinese medicine experience knowledge base, a device, and a method for determining a prescription. Background Art

[0002] In the field of traditional Chinese medicine, traditional experience inheritance mainly relies on the oral instruction and personal guidance between master and apprentice, as well as the research on ancient books. However, with the continuous accumulation of modern medical data, the traditional experience inheritance method faces many challenges.

[0003] A large amount of case data has been generated in traditional Chinese medicine clinical practice, including tongue and pulse data, symptom data, prescription data, etc. However, these data are often scattered and lack effective integration and analysis. For example, the internal relationship between tongue and pulse data, symptom data, and prescription data has not been systematically explored, and it is difficult to intuitively show the relationship between them, which lacks effective support for the diagnosis and treatment plan formulation of traditional Chinese medicine. Summary of the Invention

[0004] The present invention provides a method for constructing a traditional Chinese medicine experience knowledge base, a device, and a method for determining a prescription, which can solve the problem that the existing traditional Chinese medicine experience lacks effective support for the diagnosis and treatment plan formulation of traditional Chinese medicine due to the lack of integrated analysis of the relationship between tongue and pulse data, symptom data, and prescription data.

[0005] To solve the above technical problems, an embodiment of the present invention provides a method for constructing a traditional Chinese medicine experience knowledge base, including:

[0006] Obtaining a number of case data for each disease; wherein, the case data includes tongue and pulse data, symptom data, and prescription data;

[0007] For each disease, according to the tongue and pulse data and the symptom data, constructing a relationship graph for representing the relationship between the tongue and pulse and the symptoms;

[0008] For each disease, calculating the mutual trust degree between the tongue and pulse and the prescription according to the tongue and pulse data and the prescription data, and constructing a mutual trust degree table for representing the mutual trust degree between the tongue and pulse and the prescription according to the mutual trust degree between the tongue and pulse and the prescription;

[0009] For each disease, clustering the tongue and pulse data and the symptom data in a number of case data to obtain a number of typical categories composed of combinations of the tongue and pulse and the symptoms, determining the prescription data for each typical category according to each typical category, and constructing a clustering table for representing the clustering result between the tongue and pulse, the symptoms, and the prescription according to each typical category and the corresponding prescription data;

[0010] Constructing a traditional Chinese medicine experience knowledge base according to the relationship graph, the mutual trust degree table, and the clustering table of each disease.

[0011] Further, the tongue and pulse data include tongue symptom data and pulse condition data;

[0012] For each disease, according to the tongue and pulse data and the symptom data, constructing a relationship graph for characterizing the relationship between the tongue and pulse and the symptoms, including:

[0013] According to the tongue symptom data, the pulse condition data and the symptom data, taking the tongue symptoms, the pulse conditions and the symptoms as nodes, and taking the associations and the number of associations between the tongue symptoms and the pulse conditions, and between the pulse conditions and the symptoms as edges, respectively constructing a Sankey diagram and a chord diagram for characterizing the relationship between the tongue and pulse and the symptoms.

[0014] Further, the prescription data includes main medicine data and auxiliary medicine data;

[0015] Calculating the mutual trust degree between the tongue and pulse and the prescription according to the tongue and pulse data and the prescription data, including:

[0016] For each disease, under the condition of the same main medicine data, according to the tongue and pulse data and the auxiliary medicine data, calculating the mutual trust degree between each tongue symptom and each auxiliary medicine data and the mutual trust degree between each pulse condition and each auxiliary medicine data, and constructing a mutual trust degree table for characterizing the mutual trust degree between the tongue and pulse and the prescription according to the mutual trust degree between each tongue symptom and each auxiliary medicine data and the mutual trust degree between each pulse condition and each auxiliary medicine data.

[0017] Further, the calculation formulas for the mutual trust degree between each tongue symptom and each auxiliary medicine data and the mutual trust degree between each pulse condition and each auxiliary medicine data are:

[0018]

[0019] Wherein, PMI(a, b) is the mutual trust degree; a represents a tongue symptom or a pulse condition; b represents each auxiliary medicine; p(a, b) is the probability that each tongue symptom or each pulse condition and each auxiliary medicine appear simultaneously; p(a) is the probability that each tongue symptom or each pulse condition appears; p(b) is the probability that each auxiliary medicine appears.

[0020] Further, for each disease, clustering the tongue and pulse data and the symptom data in a number of case data to obtain a number of typical categories composed of combinations of the tongue and pulse and the symptoms, including:

[0021] For each disease, clustering the tongue and pulse data and the symptom data in a number of case data through the Kmeans clustering algorithm to obtain a number of typical categories composed of combinations of the tongue and pulse and the symptoms.

[0022] Further, according to each typical category, determining the prescription data of each typical category, and constructing a clustering table for characterizing the clustering results among the tongue and pulse, the symptoms and the prescription according to each typical category and the corresponding prescription data, including:

[0023] For each typical category, obtain the case data corresponding to each typical category, perform duplicate removal on the prescription data of the case data corresponding to each typical category, determine the main drug data and auxiliary drug data for each typical category, and construct a clustering table for characterizing the clustering results among tongue veins, symptoms, and prescriptions based on each typical category and the corresponding main drug data and auxiliary drug data.

[0024] Further, after obtaining several case data for each disease, it further includes:

[0025] Perform data preprocessing on the several case data for each disease obtained.

[0026] Based on the above method item embodiments, the present invention correspondingly provides device item embodiments;

[0027] An embodiment of the present invention provides a device for constructing a traditional Chinese medicine experience knowledge base, including: a data acquisition module, a relationship graph construction module, a mutual trust degree table construction module, a clustering table construction module, and a knowledge base construction module;

[0028] The data acquisition module is used to obtain several case data for each disease; wherein, the case data includes tongue vein data, symptom data, and prescription data;

[0029] The relationship graph construction module is used for each disease to construct a relationship graph for characterizing the relationship between tongue veins and symptoms based on the tongue vein data and symptom data;

[0030] The mutual trust degree table construction module is used for each disease to calculate the mutual trust degree between the tongue vein and the prescription according to the tongue vein data and the prescription data, and construct a mutual trust degree table for characterizing the mutual trust degree between the tongue vein and the prescription based on the mutual trust degree between the tongue vein and the prescription;

[0031] The clustering table construction module is used for each disease to cluster the tongue vein data and symptom data in several case data to obtain several typical categories composed of combinations of tongue veins and symptoms, determine the prescription data for each typical category according to each typical category, and construct a clustering table for characterizing the clustering results among tongue veins, symptoms, and prescriptions based on each typical category and the corresponding prescription data;

[0032] The knowledge base construction module is used to construct a traditional Chinese medicine experience knowledge base according to the relationship graph, mutual trust degree table, and clustering table of each disease.

[0033] Further, after the data acquisition module, it further includes: a data preprocessing module;

[0034] The data preprocessing module is used to perform data preprocessing on the several case data for each disease obtained.

[0035] Based on the above method embodiments, the present invention correspondingly provides method embodiments for determining prescriptions;

[0036] An embodiment of the present invention provides a method for determining a prescription based on a traditional Chinese medicine experience knowledge base, including:

[0037] Obtain the tongue and pulse data and symptom data of the patient to be diagnosed;

[0038] According to the tongue and pulse data and symptom data of the patient to be diagnosed, retrieve from the traditional Chinese medicine experience knowledge base and extract the corresponding prescription data; wherein, the traditional Chinese medicine experience knowledge base is constructed by a construction method of the traditional Chinese medicine experience knowledge base;

[0039] Take the extracted prescription data as the prescription for the patient to be diagnosed.

[0040] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0041] The present invention first constructs a relationship graph for characterizing the relationship between the tongue and pulse and symptoms through the tongue and pulse data and symptom data of each case; secondly, calculates the mutual information degree between the tongue and pulse symptoms and the prescription according to the tongue and pulse data, symptom data and prescription data of each case, and constructs a mutual trust degree table for characterizing the mutual trust degree among the tongue and pulse, symptoms and prescriptions according to the mutual trust degree between the tongue and pulse symptoms and the prescription; then clusters the tongue and pulse data and symptom data in the case data of several patients in each case to obtain several typical categories composed of combinations of the tongue and pulse and symptoms in each case, determines the prescriptions for each typical category according to each typical category, and constructs a clustering table for characterizing the clustering results among the tongue and pulse, symptoms and prescriptions according to each typical category and the corresponding prescription; finally, constructs a traditional Chinese medicine experience knowledge base according to the relationship graph, mutual trust degree table and clustering table of each case; that is, by generating a relationship graph for characterizing the relationship between the tongue and pulse and symptoms, a mutual trust degree table for characterizing the mutual trust degree among the tongue and pulse, symptoms and prescriptions, and a clustering table for characterizing the clustering results among the tongue and pulse, symptoms and prescriptions according to the case data, the tongue and pulse data, symptom data and prescription data of the case data can be scientifically and systematically analyzed, and the relationship between the tongue and pulse data, symptom data and prescription data can be intuitively displayed, providing effective scientific analysis data support for traditional Chinese medicine diagnosis and treatment, improving the accuracy of traditional Chinese medicine clinical diagnosis and treatment, and solving the problem that the existing traditional Chinese medicine experience lacks the integration and analysis of the relationship between the tongue and pulse data, symptom data and prescription data, resulting in the lack of effective support for the formulation of traditional Chinese medicine diagnosis and treatment plans. Description of the Drawings

[0042] Figure 1 : is a step flow chart of a method for constructing a traditional Chinese medicine experience knowledge base provided by an embodiment of the present invention;

[0043] Figure 2 : An example diagram of the Sankey diagram provided by the embodiment of the present invention;

[0044] Figure 3 : An example diagram of the chord diagram provided by the embodiment of the present invention;

[0045] Figure 4 : An example diagram of the clustering table provided by the embodiment of the present invention;

[0046] Figure 5 : A structural module diagram of a device for constructing a traditional Chinese medicine experience knowledge base provided by the embodiment of the present invention;

[0047] Figure 6 : A step flowchart of a method for determining a prescription based on a traditional Chinese medicine experience knowledge base provided by the embodiment of the present invention. Detailed implementation manners

[0048] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0049] Embodiment 1:

[0050] Refer to Figure 1 : A step flowchart of a method for constructing a traditional Chinese medicine experience knowledge base provided by the embodiment of the present invention; the construction method at least includes the following steps:

[0051] Step S1: Obtain a number of case data for each disease; wherein, the case data includes tongue and pulse data, symptom data, and prescription data;

[0052] Step S2: For each disease, construct a relationship diagram for characterizing the relationship between the tongue and pulse and symptoms according to the tongue and pulse data and the symptom data;

[0053] Step S3: For each disease, calculate the mutual trust degree between the tongue and pulse and the prescription according to the tongue and pulse data and the prescription data, and construct a mutual trust degree table for characterizing the mutual trust degree between the tongue and pulse and the prescription according to the mutual trust degree between the tongue and pulse and the prescription;

[0054] Step S4: For each disease, cluster the tongue and pulse data and the symptom data in the number of case data to obtain a number of typical categories composed of combinations of the tongue and pulse and symptoms, and determine the prescription data of each typical category according to each typical category, and construct a clustering table for characterizing the clustering results among the tongue and pulse, symptoms, and prescriptions according to each typical category and the corresponding prescription data;

[0055] Step S5: Construct a traditional Chinese medicine experience knowledge base according to the relationship diagram, mutual trust degree table, and clustering table of each disease syndrome.

[0056] In this embodiment, the tongue and pulse data includes tongue symptom data and pulse condition data.

[0057] For each disease syndrome, according to the tongue and pulse data and symptom data, construct a relationship diagram for characterizing the relationship between the tongue and pulse and symptoms, including:

[0058] According to the tongue symptom data, pulse condition data, and symptom data, use the tongue symptoms, pulse conditions, and symptoms as nodes, and use the associations and the number of associations between the tongue symptoms and pulse conditions, and between the pulse conditions and symptoms as edges to respectively construct a Sankey diagram and a chord diagram for characterizing the relationship between the tongue and pulse and symptoms.

[0059] Exemplarily, refer to Figure 2 , which is an example diagram of the Sankey diagram provided by the embodiment of the present invention; the Sankey diagram is mainly composed of edges, flows, and pivots, where the edges represent the flowing data, the flows represent the specific values of the flowing data, and the nodes represent different classifications. The width of the edge is displayed in proportion to the flow, and the wider the edge, the larger the value.

[0060] The specific construction method of the Sankey diagram is as follows:

[0061] 1. Count all types of "tongue texture", "tongue coating", "pulse condition", and "disease syndrome".

[0062] 2. Use the exhaustive method to count according to the types counted in step 1 to obtain the combinations of all data types in the first step (where the types of "tongue texture" and "tongue coating" are combined into one type).

[0063] 3. In the case data, count the number of cases that all meet the combination conditions of all data types obtained in step 2.

[0064] 4. Calculate according to the number of cases obtained in step 3:

[0065] 4.1. Node data nodes, the node data includes all "tongue texture", "tongue coating", "pulse condition", and "disease syndrome", as the nodes of the Sankey diagram.

[0066] 4.2. Link data links, which includes two sub-parts: ① The association and the number of associations from the tongue symptom or tongue coating to the pulse condition; ② The association and the number of associations from the pulse condition to the disease syndrome.

[0067] 5. Construct a Sankey diagram according to the node data and link number in step 4.

[0068] Exemplarily, refer to Figure 3, which is an example diagram of the chord diagram provided by the embodiments of the present invention; the chord diagram is mainly used to display the relationships between multiple objects. The line segment connecting any two points on the circle is called a chord, and the chord (the connection between two points) represents the association relationship between the two. The specific construction method of the chord diagram is as follows:

[0069] 1. Count the types of "symptoms", "tongue coating", "tongue body", and "pulse condition" in all case data;

[0070] 2. Generate a set of nodes data based on the four types of data counted in step 1, including important data items: a) Id: a unique identifier field; b) Name: the name of the node; c) Value: indicating the importance of the node, initially defaulted to 1; d) Category: indicating the type of the node;

[0071] 3. Traverse the case data to generate a set of links data. The method for each links data is as follows:

[0072] 3.1. Traverse the symptom data, query and return the id in nodes according to the name field, denoted as ids_symptom;

[0073] 3.2. Traverse the tongue body data, query and return the id in nodes according to the name field, denoted as ids_sheZhi;

[0074] 3.4. Traverse the tongue coating data, query and return the id in nodes according to the name field, denoted as ids_sheTai;

[0075] 3.5. Traverse the pulse condition data, query and return the id in nodes according to the name field, denoted as ids_xiang;

[0076] 3.6. Combine the above ids_symptom, ids_sheZhi, ids_sheTai, and ids_xiang data, denoted as ids_all;

[0077] 3.7. Calculate the link data according to ids_all;

[0078] 4. Traverse the node data nodes, count the number of times the id field appears in the link data links for each node data nodes, and then update each node data according to the count;

[0079] 5. Traverse the link data links, remove redundant and duplicate data, and obtain the ret_links data;

[0080] 7. Use the cates, nodes, and ret_links data obtained above to generate a chord diagram.

[0081] In this embodiment, referring to Figure 4 , which is an exemplary diagram of the mutual trust degree table provided by the embodiment of the present invention; the prescription data includes main medicine data and auxiliary medicine data; calculating the mutual trust degree between the tongue and pulse data and the prescription data includes:

[0082] For each disease, under the condition of the same main medicine data, according to the tongue and pulse data and the auxiliary medicine data, calculate the mutual trust degree between each tongue symptom and each auxiliary medicine data and the mutual trust degree between each pulse condition and each auxiliary medicine data, and construct a mutual trust degree table for characterizing the mutual trust degree between the tongue and pulse and the prescription according to the mutual trust degree between each tongue symptom and each auxiliary medicine data and the mutual trust degree between each pulse condition and each auxiliary medicine data.

[0083] In this embodiment, the calculation formulas for the mutual trust degree between each tongue symptom and each auxiliary medicine data and the mutual trust degree between each pulse condition and each auxiliary medicine data are:

[0084]

[0085] Where PMI(a, b) is the mutual trust degree; a represents a tongue symptom or a pulse condition; b represents each auxiliary medicine; p(a, b) is the probability of the simultaneous occurrence of each tongue symptom or each pulse condition and each auxiliary medicine; p(a) is the probability of the occurrence of each tongue symptom or each pulse condition; p(b) is the probability of the occurrence of each auxiliary medicine.

[0086] In this embodiment, for each disease, clustering the tongue and pulse data and the symptom data in a number of case data to obtain a number of typical categories composed of combinations of tongue and pulse and symptoms, including:

[0087] For each disease, clustering the tongue and pulse data and the symptom data in a number of case data through the Kmeans clustering algorithm to obtain a number of typical categories composed of combinations of tongue and pulse and symptoms.

[0088] Exemplarily, the clustering of the tongue and pulse data and the symptom data in a number of case data through the Kmeans clustering algorithm is specifically as follows:

[0089] 1. Traverse each case data, merge the "tongue quality", "tongue coating", "pulse condition", and "disease" fields of a single case to form a string;

[0090] 2. Merge the strings in the first step to form a string group data_str;

[0091] 3. Train data_str with the kmeans algorithm to obtain the clustering center;

[0092] 4. Select the string closest to the clustering center in data_str as the typical category of "tongue and pulse - symptom".

[0093] In this embodiment, with reference to Figure 4 , which is an example diagram of the clustering table provided by the embodiment of the present invention; determining the prescription data of each typical category according to each typical category, and constructing a clustering table for characterizing the clustering results among tongue veins, symptoms, and prescriptions according to each typical category and the corresponding prescription data, including:

[0094] According to each typical category, obtain the case data corresponding to each typical category, perform a duplicate removal operation on the prescription data of the case data corresponding to each typical category, determine the main drug data and auxiliary drug data of each typical category, and construct a clustering table for characterizing the clustering results among tongue veins, symptoms, and prescriptions according to each typical category and the corresponding main drug data and auxiliary drug data.

[0095] In this embodiment, after obtaining a number of case data for each disease, it further includes:

[0096] Perform data preprocessing on the obtained number of case data for each disease.

[0097] Embodiment 2:

[0098] With reference to Figure 5 , which is a structural module diagram of a device for constructing a traditional Chinese medicine experience knowledge base provided by the embodiment of the present invention; the device at least includes: a data acquisition module, a relationship graph construction module, a mutual trust degree table construction module, a clustering table construction module, and a knowledge base construction module;

[0099] The data acquisition module is used to obtain a number of case data for each disease; wherein, the case data includes tongue vein data, symptom data, and prescription data;

[0100] The relationship graph construction module is used to construct a relationship graph for characterizing the relationship between tongue veins and symptoms according to the tongue vein data and symptom data for each disease;

[0101] The mutual trust degree table construction module is used to calculate the mutual trust degree between the tongue veins and the prescriptions according to the tongue vein data and the prescription data for each disease, and construct a mutual trust degree table for characterizing the mutual trust degree between the tongue veins and the prescriptions according to the mutual trust degree between the tongue veins and the prescriptions;

[0102] The clustering table construction module is used to cluster the tongue vein data and symptom data in a number of case data for each disease to obtain a number of typical categories composed of combinations of tongue veins and symptoms, determine the prescription data of each typical category according to each typical category, and construct a clustering table for characterizing the clustering results among tongue veins, symptoms, and prescriptions according to each typical category and the corresponding prescription data;

[0103] The knowledge base construction module is used to construct a traditional Chinese medicine experience knowledge base according to the relationship graph, mutual trust degree table, and clustering table of each disease.

[0104] In this embodiment, after the data acquisition module, it further includes: a data preprocessing module;

[0105] The data preprocessing module is used to perform data preprocessing on the obtained several case data of each disease.

[0106] Embodiment 3:

[0107] Refer to Figure 6 : It is a step flowchart of a prescription determination method based on a traditional Chinese medicine experience knowledge base provided by an embodiment of the present invention; this method at least includes the following steps:

[0108] Step A1: Obtain the tongue and pulse data and symptom data of the patient to be diagnosed;

[0109] Step A2: According to the tongue and pulse data and symptom data of the patient to be diagnosed, retrieve from the traditional Chinese medicine experience knowledge base and extract the corresponding prescription data; wherein, the traditional Chinese medicine experience knowledge base is constructed by the construction method of a traditional Chinese medicine experience knowledge base in Embodiment 1;

[0110] Step A3: Use the extracted prescription data as the prescription for the patient to be diagnosed.

[0111] The above specific embodiments have further detailed the purpose, technical solution and beneficial effects of the present invention. It should be understood that the above is only the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. In particular, for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for constructing a traditional Chinese medicine experience knowledge base, characterized in that Including: Obtaining a number of case data for each disease; wherein, the case data includes tongue and pulse data, symptom data, and prescription data; For each disease, according to the tongue and pulse data and the symptom data, construct a relationship graph for characterizing the relationship between the tongue and pulse and the symptoms; For each disease, according to the tongue and pulse data and the prescription data, calculate the mutual trust degree between the tongue and pulse and the prescription, and construct a mutual trust degree table for characterizing the mutual trust degree between the tongue and pulse and the prescription according to the mutual trust degree between the tongue and pulse and the prescription; For each disease, cluster the tongue and pulse data and the symptom data in a number of case data to obtain a number of typical categories composed of combinations of the tongue and pulse and the symptoms, and determine the prescription data of each typical category according to each typical category, and construct a clustering table for characterizing the clustering results among the tongue and pulse, the symptoms, and the prescription according to each typical category and the corresponding prescription data; Construct a traditional Chinese medicine experience knowledge base according to the relationship graph, the mutual trust degree table, and the clustering table of each disease; 2. The construction method of a traditional Chinese medicine experience knowledge base according to claim 1, characterized in that The tongue and pulse data includes tongue symptom data and pulse condition data; For each disease, according to the tongue and pulse data and the symptom data, constructing a relationship graph for characterizing the relationship between the tongue and pulse and the symptoms includes: According to the tongue symptom data, the pulse condition data, and the symptom data, use the tongue symptoms, the pulse conditions, and the symptoms as nodes, and use the associations and the number of associations between the tongue symptoms and the pulse conditions, and between the pulse conditions and the symptoms as edges to respectively construct a Sankey diagram and a chord diagram for characterizing the relationship between the tongue and pulse and the symptoms; 3. The construction method of a traditional Chinese medicine experience knowledge base according to claim 2, characterized in that, The prescription data includes main medicine data and auxiliary medicine data; Calculating the mutual trust degree between the tongue and pulse and the prescription according to the tongue and pulse data and the prescription data includes: For each disease, in the case of the same main medicine data, according to the tongue and pulse data and the auxiliary medicine data, calculate the mutual trust degree between each tongue symptom and each auxiliary medicine data and the mutual trust degree between each pulse condition and each auxiliary medicine data, and construct a mutual trust degree table for characterizing the mutual trust degree between the tongue and pulse and the prescription according to the mutual trust degree between each tongue symptom and each auxiliary medicine data and the mutual trust degree between each pulse condition and each auxiliary medicine data; 4. The construction method of a traditional Chinese medicine experience knowledge base according to claim 3, characterized in that The calculation formulas for the mutual trust degree between each tongue symptom and each auxiliary medicine data and the mutual trust degree between each pulse condition and each auxiliary medicine data are: Wherein, PMI(a, b) is the mutual trust degree; a represents a tongue symptom or a pulse condition; b represents each auxiliary medicine material; p(a, b) is the probability that each tongue symptom or each pulse condition and each auxiliary medicine material appear simultaneously; p(a) is the probability that each tongue symptom or each pulse condition appears; p(b) is the probability that each auxiliary medicine material appears.

5. The construction method of a traditional Chinese medicine experience knowledge base according to claim 4, characterized in that For each disease, clustering the tongue and pulse data and the symptom data in a number of case data to obtain a number of typical categories composed of combinations of the tongue and pulse and the symptoms includes: For each disease, use the Kmeans clustering algorithm to cluster the tongue and pulse data and the symptom data in a number of case data to obtain a number of typical categories composed of combinations of the tongue and pulse and the symptoms; 6. The construction method of a traditional Chinese medicine experience knowledge base according to claim 5, characterized in that Determining the prescription data of each typical category according to each typical category, and constructing a clustering table for characterizing the clustering results among the tongue and pulse, the symptoms, and the prescription according to each typical category and the corresponding prescription data includes: For each typical category, obtain the case data corresponding to each typical category, perform duplicate removal on the prescription data of the case data corresponding to each typical category, determine the main medicine data and auxiliary medicine data for each typical category, and construct a clustering table for characterizing the clustering results among tongue and pulse manifestations, symptoms, and prescriptions based on each typical category and the corresponding main medicine data and auxiliary medicine data.

7. A method for constructing a traditional Chinese medicine experience knowledge base according to claim 6, characterized in that After obtaining a number of case data for each disease condition, it further includes: Performing data preprocessing on the obtained number of case data for each disease condition.

8. A device for constructing a traditional Chinese medicine experience knowledge base, characterized in that It includes: A data acquisition module, a relationship graph construction module, a mutual trust degree table construction module, a clustering table construction module, and a knowledge base construction module; The data acquisition module is used to obtain a number of case data for each disease condition; wherein, the case data includes tongue and pulse manifestation data, symptom data, and prescription data; The relationship graph construction module is used to construct a relationship graph for each disease condition to characterize the relationship between tongue and pulse manifestations and symptoms based on the tongue and pulse manifestation data and symptom data; The mutual trust degree table construction module is used to calculate the mutual trust degree between the tongue and pulse manifestations and prescriptions for each disease condition based on the tongue and pulse manifestation data and prescription data, and construct a mutual trust degree table for characterizing the mutual trust degree between the tongue and pulse manifestations and prescriptions based on the mutual trust degree between the tongue and pulse manifestations and prescriptions; The clustering table construction module is used to cluster the tongue and pulse manifestation data and symptom data in a number of case data for each disease condition to obtain a number of typical categories composed of combinations of tongue and pulse manifestations and symptoms, determine the prescription data for each typical category based on each typical category, and construct a clustering table for characterizing the clustering results among tongue and pulse manifestations, symptoms, and prescriptions based on each typical category and the corresponding prescription data; The knowledge base construction module is used to construct a traditional Chinese medicine experience knowledge base based on the relationship graph, mutual trust degree table, and clustering table for each disease condition.

9. The construction device of a traditional Chinese medicine experience knowledge base according to claim 8, characterized in that, After the data acquisition module, it further includes: a data preprocessing module; The data preprocessing module is used to perform data preprocessing on the obtained number of case data for each disease condition.

10. A method for determining a prescription based on the TCM experience knowledge base, characterized in that, It includes: Obtaining the tongue and pulse manifestation data and symptom data of the patient to be diagnosed; Retrieving from the traditional Chinese medicine experience knowledge base according to the tongue and pulse manifestation data and symptom data of the patient to be diagnosed, and extracting the corresponding prescription data; wherein, the traditional Chinese medicine experience knowledge base is constructed by the construction method of the traditional Chinese medicine experience knowledge base in claims 1-7; Taking the extracted prescription data as the prescription for the patient to be diagnosed.

Images