Weighted association rule method and system based on dosage of traditional Chinese medicinal materials in traditional Chinese medicine prescription

Abstract

The invention belongs to the field of traditional Chinese medicine data mining, and particularly relates to a weighted association rule method and system based on traditional Chinese medicine dosage of a traditional Chinese medicine prescription, comprising the following steps: collecting a target effective traditional Chinese medicine prescription, and recording all the applied medicine dosage of traditional Chinese medicines of the traditional Chinese medicine prescription; standardizing the names of the medicinal materials in the traditional Chinese medicine prescription, sending all data to a matrix construction module, and processing all the data by the matrix construction module to obtain a traditional Chinese medicine prescription dosage weight matrix; the rule association module obtains various parameters of a weighted association rule model in combination with the collected medicinal material set and the traditional Chinese medicine prescription dosage weight matrix; and the standard judgment module obtains various parameters of the weighted association rule model as judgment standards so as to screen the medicinal material with the strongest association among the medicinal materials. According to the method, the concept of the weighted association rule is provided, the importance of the medicinal material dosage on the influence of the association rule between the medicinal materials is evaluated more comprehensively, and related parameters of the weighted association rule are calculated.

Description

Technical Field

[0001] The invention belongs to the field of traditional Chinese medicine data mining, and in particular to a weighted association rule method and system based on the dosage of traditional Chinese medicine prescriptions and traditional Chinese medicines. Background Art

[0002] In the application of mining strongly associated medicinal material combinations in traditional Chinese medicine prescriptions, the traditional method is to use the Apriori association rule algorithm proposed by R. Agrawal et al. in 1993. This method is used to calculate all strong association rules between items in a large customer transaction database, and can also be applied to traditional Chinese medicine prescriptions to mine strongly associated medicinal material combinations. The basic steps of the Apriori algorithm are described as follows: 1. First, all frequent item sets are screened. 2. The association rules are screened according to relevant parameters such as support, support, and lift to find strong association rules that meet the conditions.

[0003] However, the existing Apriori algorithm has certain limitations. The algorithm assumes that all items have the same importance and only considers whether each item appears. That is, in the process of calculating association rules, only the frequency is used to filter out the strong association rules of interest. Traditional Chinese medicine prescriptions are complex combinations of medicinal materials, and also involve the dosage of each medicinal material in the Chinese medicine prescription. The traditional association rule algorithm is applied to mine the core medicinal material combination with strong association in the Chinese medicine prescription database, and only uses the frequency as the screening condition of the association rule. This method lacks comprehensiveness and also makes the rationality and accuracy of the screening results questionable. Therefore, it is particularly important to propose a new weighted association rule algorithm. Summary of the invention

[0004] The existing algorithm for screening association rule evaluation criteria in traditional Chinese medicine prescriptions cannot comprehensively screen out strongly associated core medicinal material combinations. The problem to be solved by the present invention is to provide a weighted association rule method and system based on the dosage of medicinal materials in traditional Chinese medicine prescriptions, so as to more comprehensively and accurately screen out the core medicinal material combination for treating target diseases.

[0005] The technical solution adopted by the present invention to achieve the above-mentioned purpose is: a weighted association rule method based on the dosage of Chinese medicinal materials in Chinese medicine prescriptions, comprising the following steps:

[0006] 1) The data collection module of the host computer collects the target effective Chinese medicine prescriptions in the database by keyword, and records all the dosages of the Chinese medicines used in the Chinese medicine prescriptions;

[0007] 2) The data standardization module normalizes the names of the medicinal materials in the collected traditional Chinese medicine prescriptions, and sends the traditional Chinese medicine prescriptions, medicinal material names and all the corresponding medicinal material dosages to the matrix construction module to construct the unweighted basic matrix M0. The matrix construction module normalizes the unweighted basic matrix to obtain the standardized unweighted basic matrix M 01 ;

[0008] 3) The matrix construction module constructs the dose weighted matrix M1 based on the collected traditional Chinese medicine prescriptions, medicinal materials names and medicinal materials doses, and standardizes the matrix M1 to obtain the standardized dose weighted matrix M. 11 ;

[0009] 4) According to the obtained standardized dose weighting matrix M 11 , obtain the medicinal material dosage weight matrix M2;

[0010] 5) The matrix building module will standardize the powerless base matrix M 01 and the dose weight matrix M2 of the medicinal materials to obtain the dose weight matrix M3 of the Chinese medicine prescriptions;

[0011] 6) The rule association module combines the collected medicinal materials and the dose weight matrix M3 of the traditional Chinese medicine prescriptions to obtain various parameters of the weighted association rule model;

[0012] 7) The standard discrimination module obtains the parameters of the weighted association rule model as judgment criteria based on the parameters of the weighted association rule algorithm to screen the medicinal materials with the strongest correlation between medicinal materials.

[0013] The construction of the unweighted basic matrix M0 is specifically as follows:

[0014] 2-1) After the collected Chinese medicine prescriptions and medicinal materials are treated with 0 and 1, the unrighteous basic matrix M0 is constructed as:

[0015]

[0016] Among them, i represents the i-th Chinese medicine prescription, j represents the j-th medicinal material; a ij Indicates whether the jth medicinal ingredient is in the i-th Chinese medicine prescription t i Appears in, if it appears then a ij =1, if it does not appear then a ij =0; Z represents a set of integers.

[0017] 2-2) Standardize M0 to obtain the standardized powerless basic matrix M 01 ,Right now:

[0018]

[0019] Among them, i represents the i-th Chinese medicine prescription, j represents the j-th medicinal material; a'ij Indicates a ij Divide by the total number of medicinal materials contained in each Chinese medicine prescription.

[0020] The step 3) is specifically:

[0021] 3-1) The collected Chinese medicine prescriptions, medicinal materials and the dosage of medicinal materials are used to construct the dose weighted matrix M1 as follows:

[0022]

[0023] Where i represents the i-th Chinese medicine prescription, j represents the j-th medicinal material; if the j-th medicinal material appears in the i-th Chinese medicine prescription, then b ij =s ij ,s ij Indicates the dosage of the jth medicinal material in the i-th prescription; if it does not appear, b ij =0;

[0024] 3-2) Standardize M1 to obtain the standardized dose weighting matrix M 11 as follows:

[0025]

[0026] Where i represents the i-th Chinese medicine prescription, j represents the j-th medicinal material; b ij Divide by the mean dose of all medicinal materials in the corresponding prescription (b' i. ) aver , and obtain the dosage weight of the medicinal material in all prescriptions.

[0027] The step 4) is specifically:

[0028] According to the standardized dose weighting matrix M 11 , then the medicinal material dosage weight matrix M2 is:

[0029]

[0030] Calculate the average of the dosage weights of each medicinal material in all prescriptions to obtain the medicinal material dosage weight w j .

[0031] The step 5) is specifically:

[0032] Normalized unweighted basis matrix M 01 The Chinese medicine dosage weight matrix M3 is calculated by combining the medicinal material dosage weight matrix M2;

[0033]

[0034] Where i represents the i-th Chinese medicine prescription, w' i Represents the dosage weight of the i-th traditional Chinese medicine prescription.

[0035] The step 6) is specifically:

[0036] 6-1) Let the set of medicinal materials H = {H1, H2, H3, ..., H n}, and set the medicinal material subset A = {H n1 ,H n2 ,H n3 ,...,H nk}, k represents k medicinal materials, then the weighted support wsup of the medicinal material subset A is:

[0037]

[0038] in, a represents the collection of all medicinal materials inj The sum of the prescription weights that are not 0; Represents the sum of all prescription weights;

[0039] 6-2) Let the medicinal material subset B = {H j}, then the weighted confidence wcon of the weighted association rule with the medicinal material subset A as the antecedent and the medicinal material subset B as the contingent is:

[0040]

[0041] 6-3) Obtain the weighted lift wlift with the medicinal material subset A as the antecedent and the medicinal material subset B as the posterior:

[0042]

[0043] According to the parameters of the weighted association rule algorithm, the parameters of the weighted association rule are obtained as the judgment criteria, specifically:

[0044] 7-1) Keep the rule that weighted lift wlift>1;

[0045] 7-2) Sort the retained rules according to the weighted support wsup, and analyze the results of the weighted association rule analysis according to wsup and the weighted lift wcon, namely:

[0046] The higher the scores of weighted support wsup and weighted lift wcon, the stronger the association rule is considered to be, that is, the stronger the association between the medicinal materials is. Finally, the medicinal materials with the strongest association between the medicinal materials are screened to obtain the core medicinal material combination for treating the target disease.

[0047] A weighted association rule system based on the dosage of Chinese medicinal materials in traditional Chinese medicine prescriptions, comprising: a data acquisition module, a data specification module, a matrix construction module, a rule association module and a standard discrimination module;

[0048] The data collection module is used to collect target effective Chinese medicine prescriptions in the database by keyword, and record all the dosages of medicinal materials used in the Chinese medicine prescriptions;

[0049] The data standardization module is used to call the names of medicinal materials in the traditional Chinese medicine prescriptions in the data standardization module for standardization processing, and send the standardized medicinal material names and all the corresponding medicinal material doses to the matrix construction module;

[0050] The matrix construction module is used to construct an unweighted basis matrix and a standardized unweighted basis matrix according to the standardized medicinal material names and all the corresponding medicinal material dosages; and process the Chinese medicine prescriptions, medicinal materials and the corresponding medicinal material dosages, the unweighted basis matrix and the standardized unweighted basis matrix to finally obtain the Chinese medicine prescription dosage weight matrix;

[0051] A rule association module is used to receive and obtain various parameters of a weighted association rule model based on the collected medicinal material set and the weight matrix of the dosage of the traditional Chinese medicine prescription;

[0052] The standard discrimination module is used to screen the medicinal materials with the strongest association among medicinal materials according to the various parameters of the weighted association rule model.

[0053] A weighted association rule system based on the dosage of Chinese medicinal materials in a traditional Chinese medicine prescription comprises a memory and a processor; the memory is used to store a computer program; the processor is used to implement a weighted association rule method based on the dosage of Chinese medicinal materials in a traditional Chinese medicine prescription when executing the computer program.

[0054] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, a weighted association rule method based on the dosage of Chinese medicinal materials in a traditional Chinese medicine prescription is implemented.

[0055] The present invention has the following beneficial effects and advantages:

[0056] 1. When screening the core medicinal material combination, the present invention comprehensively considers for the first time the influence and importance of the dosage of each medicinal material in the traditional Chinese medicine prescription on the association rule mining, and describes the association rule relationship between the medicinal materials more accurately, and can more reasonably and accurately screen out the core medicinal material combination for treating the target disease.

[0057] 2. The present invention proposes the concept of weighted association rules, more comprehensively evaluates the importance of the influence of medicinal material dosage on the association rules between medicinal materials, and calculates the relevant parameters of the weighted association rules. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 A flow chart of the method of the present invention based on the weighted association rule of the dosage of Chinese medicinal materials in Chinese medicine prescriptions;

[0059] Figure 2 The present invention is based on a system block diagram of a weighted association rule system for the dose of traditional Chinese medicine prescriptions and Chinese medicinal materials. DETAILED DESCRIPTION

[0060] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0061] like Figure 1 As shown, it is a flow chart of the method for weighted association rules based on the dosage of Chinese medicinal materials in Chinese medicine prescriptions of the present invention. The present invention provides a method for weighted association rules based on the dosage of Chinese medicinal materials in Chinese medicine prescriptions, comprising the following steps:

[0062] 1) Check the literature database, read the literature, collect effective Chinese medicine prescriptions for the treatment of target diseases, and record the dosages of all the medicinal materials used in the Chinese medicine prescriptions;

[0063] 2) According to the "Chinese Pharmacopoeia (2020 Edition)", the names of medicinal materials in the collected Chinese medicine prescriptions are standardized;

[0064] 3) After the collected Chinese medicine prescriptions and medicinal materials are treated with 0 and 1, the rightsless basic matrix M0 is constructed as follows:

[0065]

[0066] Among them, i represents the i-th Chinese medicine prescription, j represents the j-th medicinal material; a ij Indicates whether the jth medicinal ingredient is in the i-th Chinese medicine prescription t i Appears in, if it appears then a ij =1, if it does not appear then a ij =0; Z represents a set of integers.

[0067] Standardize M0 to obtain the standardized permissionless basic matrix M 01 ,Right now:

[0068]

[0069] Among them, i represents the i-th Chinese medicine prescription, j represents the j-th medicinal material; a' ij Indicates a ij Divide by the total number of medicinal materials contained in each Chinese medicine prescription.

[0070] 4) The dose weighted matrix M1 of the collected Chinese medicine prescriptions, medicinal materials and medicinal materials is constructed as follows:

[0071] The dose weighted matrix M1 of the collected Chinese medicine prescriptions, medicinal materials and medicinal materials is constructed as follows:

[0072]

[0073] Where i represents the i-th Chinese medicine prescription, j represents the j-th medicinal material; if the j-th medicinal material appears in the i-th Chinese medicine prescription, then b ij =s ij ,s ij Indicates the dosage of the jth medicinal material in the i-th prescription; if it does not appear, b ij =0;

[0074] Normalize M1 to obtain the standardized dose weighting matrix M 11 as follows:

[0075]

[0076] Where i represents the i-th Chinese medicine prescription, j represents the j-th medicinal material; b ij Divide by the mean dose of all medicinal materials in the corresponding prescription (b' i. ) aver , and obtain the dosage weight of the medicinal material in all prescriptions.

[0077] 5) According to the standardized dose weighting matrix M obtained in step 4) 11 Calculate and obtain the medicinal material dosage weight matrix M2;

[0078] According to the standardized dose weighting matrix M 11 , then the medicinal material dosage weight matrix M2 is:

[0079]

[0080] Calculate the average of the dosage weights of each medicinal material in all prescriptions to obtain the medicinal material dosage weight w j .

[0081] 6) According to the standardized unweighted basis matrix M obtained in step 3) 01 The Chinese medicine dosage weight matrix M3 is calculated by combining the medicinal material dosage weight matrix M2;

[0082]

[0083] Where i represents the i-th Chinese medicine prescription, w' i Represents the dosage weight of the i-th traditional Chinese medicine prescription.

[0084] 7) Combining the medicinal material data in step 1) and the matrix constructed in step 6), various parameters of the weighted association rule algorithm are calculated:

[0085] Suppose the medicinal material set H = {H1, H2, H3, ..., H n}, and set the medicinal material subset A = {H n1 ,H n2 ,H n3 ,...,Hnk}, k represents k medicinal materials, then the weighted support degree (wsup) calculation formula of subset A of medicinal materials is as follows:

[0086]

[0087] in a represents the collection of all medicinal materials inj The sum of the prescription weights that are not 0; Represents the sum of all prescription weights.

[0088] Assume that the medicinal material subset B = {H j}, the weighted confidence (wcon) calculation formula for weighted correlation rules with subset A of the medicinal material as the preceding term and subset B of the medicinal material as the latter term is as follows:

[0089]

[0090] The calculation formula of weighted lift is as follows:

[0091]

[0092] 8) The parameters of the weighted association rules calculated in step 7) are used as judgment criteria. First, the rules with weighted lift wlift>1 are retained; then the retained rules are sorted according to weighted support (wsup), and the results of the weighted association rule analysis are analyzed according to wsup and weighted lift (wcon). The higher the wsup and wcon scores, the stronger the association rule is considered to be, that is, the stronger the association between the medicinal materials is, and finally the core medicinal material combination for treating the target disease is screened.

[0093] like Figure 2 As shown, it is a system block diagram of the weighted association rule system based on the dosage of Chinese medicinal materials in Chinese medicine prescriptions of the present invention. The weighted association rule system based on the dosage of Chinese medicinal materials in Chinese medicine prescriptions of the present invention is characterized by comprising: a data acquisition module, a data specification module, a matrix construction module, a rule association module and a standard discrimination module;

[0094] The data collection module is used to collect target effective Chinese medicine prescriptions in the database through keyword methods, and simultaneously record all the dosages of medicinal materials applied in the Chinese medicine prescriptions;

[0095] The data standardization module is used to call the names of medicinal materials in the traditional Chinese medicine prescriptions in the data standardization module for standardization processing, and send the standardized medicinal material names and all the corresponding medicinal material doses to the matrix construction module;

[0096] A matrix construction module, which is used to construct an unweighted basis matrix and a standardized unweighted basis matrix according to the standardized names of Chinese medicinal materials and all the corresponding dosages of Chinese medicinal materials; and process the traditional Chinese medicine formula, Chinese medicinal materials, their corresponding dosages of Chinese medicinal materials, the unweighted basis matrix and the standardized unweighted basis matrix, and finally obtain the dosage weight matrix of the traditional Chinese medicine formula;

[0097] A rule association module, which is used to receive and obtain the parameters of the weighted association rule model according to the collected set of Chinese medicinal materials and the dosage weight matrix of the traditional Chinese medicine formula;

[0098] A standard discrimination module, which is used to screen the Chinese medicinal materials with the strongest association between Chinese medicinal materials according to the parameters of the weighted association rule model.

[0099] Example 1:

[0100] In this example, the weighted association rule analysis is carried out on the Chinese medicinal materials in the traditional Chinese medicine formula by the above method, and the core combination of Chinese medicinal materials for treating the target disease is screened out, as follows:

[0101] In order to test the effect of the weighted association rule evaluation parameters, the weighted association rule analysis is carried out on the traditional Chinese medicine formula for treating rheumatoid arthritis (RA) in this example. Five databases, namely China National Knowledge Infrastructure (CNKI), VIP Chinese Journal Service Platform, Wanfang Data Knowledge Service Platform, Web of Science and PubMed, are retrieved in this study. With "rheumatoid arthritis", "wangbi" and "rheumatoid arthritis" as keywords, the retrieval type is clinical research.

[0102] Inclusion criteria for formula data: (a) Initially clinically diagnosed with rheumatoid arthritis (RA); (b) The names of the Chinese medicinal materials included in the traditional Chinese medicine formula are complete and the dosages of the Chinese medicinal materials used are clear; (c) The traditional Chinese medicine formula has a definite curative effect on treating RA. Exclusion criteria for data: (a) Non-clinical research, such as animal research and review articles. (b) Non-oral traditional Chinese medicine formula, such as acupuncture and medicated bath. (c) The addition or subtraction of Chinese medicinal materials in the traditional Chinese medicine formula, incomplete composition or unclear dosage of Chinese medicinal materials.

[0103] According to the above inclusion and exclusion criteria, a total of 111 effective traditional Chinese medicine formulas for treating RA are collected. The information of the collected traditional Chinese medicine formulas includes the names of the traditional Chinese medicine formulas, the names of the Chinese medicinal materials in the formulas and the dosages of the Chinese medicinal materials used. The 111 formulas are sorted out, the names of the Chinese medicinal materials appearing in the formulas are standardized and duplicates are removed, and a total of 173 Chinese medicinal materials are collected.

[0104] The weighted association rule analysis is carried out on the above 111 traditional Chinese medicine formulas and 173 Chinese medicinal materials collected, and the results are shown in Table 1:

[0105] Table 1 Analysis results of one-item set weighted association rules

[0106]

[0107]

[0108] The wsup of single medicinal materials were arranged in descending order and then accumulated, and polynomial regression analysis was performed. The single medicinal materials corresponding to the instantaneous change rate of the polynomial curve greater than the average change rate were selected as core single medicinal materials. Finally, 54 core single medicinal materials were selected, as shown in Table 2.

[0109] Table 2 54 core single medicinal materials

[0110]

[0111]

[0112] The weighted association rules with the maximum number of preceding items of 1 were constructed and analyzed for the 54 core single herbs selected above, and the screening conditions were wcon>70% and wlift>1. The results were arranged in descending order according to wsup, and the results are shown in Table 3.

[0113] Table 3 Analysis results of weighted association rules with the maximum number of preceding items being 1

[0114]

[0115]

[0116] The weighted association rules with the maximum number of preceding items of 2 were constructed for the 54 core single herbs selected above, and the analysis was performed, with the screening conditions of wcon>70% and wlift>1. The results were sorted in descending order according to wsup, as shown in Table 4.

[0117] Table 4 Parameter analysis results of weighted association rules with a maximum number of preceding items of 2

[0118]

[0119]

[0120] According to the results of weighted association rule analysis, "white peony root-licorice-cinnamon twig" (wsup=0.23833; wcon=0.84293; wlift=1.39667) is the strongest weighted association rule combination, that is, white peony root, licorice and cinnamon twig are the core medicinal material combination for the treatment of RA.

[0121] By consulting the "Chinese Pharmacopoeia (2020 Edition)", the nature, taste, meridians, functions and main indications of white peony, licorice and cinnamon twig are collected as shown in Table 5.

[0122] Table 5 Nature, flavor, meridians, and functions of the three core medicinal materials

[0123]

[0124] White peony root, cinnamon twig and licorice are all included in the classic prescription "Guizhi Shaoyao Zhimu Decoction" recorded in "Jinkui Yaolue", and the main functions and main diseases treated by these three herbs are closely related to rheumatoid arthritis.

[0125] Ye Min et al. mentioned in the article "Research Progress on Chinese Medicine for the Treatment of Rheumatoid Arthritis by Regulating Intestinal Flora" that white peony has the function of regulating immunity, and the active ingredients of white peony can inhibit inflammatory response, thus having a therapeutic effect on RA. Li Xue et al. mentioned in the article "Research Progress on the Chemical Composition and Pharmacological Actions of Cinnamon Twig" that cinnamon twig has anti-inflammatory effects, and can promote vasodilation and improve blood circulation. Li Zhe et al. described in the article "Exploring the Function of Cinnamon Twig in Treating Rheumatoid Arthritis Based on the "Sweating Theory"" that cinnamon twig can play a therapeutic role in RA through sweating. Licorice has the effect of reducing toxicity and increasing efficacy.

[0126] The present invention comprehensively considers for the first time the influence and importance of the dosage of each medicinal material in the traditional Chinese medicine prescription on the association rule mining when screening the core medicinal material combination, describes the association rule relationship between the medicinal materials more accurately, can more reasonably and accurately screen out the core medicinal material combination for treating the target disease, and more comprehensively evaluates the importance of the influence of the medicinal material dosage on the association rules between the medicinal materials, and calculates the relevant parameters of the weighted association rules.

[0127] The above description is only an embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, expansion, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A weighted association rule method based on the dosage of Chinese medicinal materials in traditional Chinese medicine prescriptions, characterized in that: The following steps are involved: 1) The data collection module of the host computer collects the target effective Chinese medicine prescriptions in the database by keyword, and records all the dosages of the Chinese medicines used in the Chinese medicine prescriptions; 2) The data standardization module normalizes the names of the medicinal materials in the collected traditional Chinese medicine prescriptions, and sends the traditional Chinese medicine prescriptions, medicinal material names and all the corresponding medicinal material dosages to the matrix construction module to construct the unweighted basic matrix M0. The matrix construction module normalizes the unweighted basic matrix to obtain the standardized unweighted basic matrix M 01 ; 3) The matrix construction module constructs the dose weighted matrix M1 according to the collected Chinese medicine prescriptions, medicinal material names and medicinal material doses, and standardizes the matrix M1 to obtain the standardized dose weighted matrix M 11 ; 4) According to the obtained standardized dose weighting matrix M 11 , obtain the medicinal material dosage weight matrix M2; 5) The matrix construction module normalizes the unweighted basis matrix M 01 and the medicinal material dosage weight matrix M2, to obtain the Chinese medicine prescription dosage weight matrix M3; 6) The rule association module combines the collected medicinal material set and the Chinese medicine prescription dosage weight matrix M3 to obtain various parameters of the weighted association rule model; 7) The standard discrimination module obtains the parameters of the weighted association rule model as the judgment criteria according to the parameters of the weighted association rule algorithm to screen the medicinal materials with the strongest association between the medicinal materials.

2. A weighted association rule method based on the dosage of Chinese medicinal materials in Chinese medicine prescriptions according to claim 1, characterized in that: The construction of the unweighted basic matrix M0 is specifically as follows: 2-1) The collected Chinese medicine prescriptions and medicinal materials are processed into 0 and 1 to construct the unweighted basic matrix M0: Where i represents the i-th Chinese medicine prescription, j represents the j-th medicinal material; a ij Indicates whether the jth medicinal ingredient is in the i-th Chinese medicine prescription t i Appears in, if it appears then a ij =1, if it does not appear then a ij =0; Z represents a set of integers. 2-2) Standardize M0 to obtain the standardized unweighted basis matrix M 01 ,Right now: Where i represents the i-th Chinese medicine prescription, j represents the j-th medicinal material; a' ij Indicates a ij Divide by the total number of medicinal materials contained in each Chinese medicine prescription.

3. A weighted association rule method based on the dosage of Chinese medicinal materials in Chinese medicine prescriptions according to claim 1, characterized in that: The step 3) is specifically: 3-1) The collected Chinese medicine prescriptions, medicinal materials and the dosage of medicinal materials are used to construct the dose weighted matrix M1 as follows: Where i represents the i-th Chinese medicine prescription, j represents the j-th medicinal material; if the j-th medicinal material is in the i-th Chinese medicine prescription, i If b ij =s ij ,s ij Indicates the dosage of the jth medicinal material in the i-th prescription; if it does not appear, b ij =0; 3-2) Standardize M1 to obtain the standardized dose weighting matrix M 11 as follows: Where i represents the i-th Chinese medicine prescription, j represents the j-th medicinal material; b ij Divide by the mean dose of all medicinal materials in the corresponding prescription (b' i. ) aver , and obtain the dosage weight of the medicinal material in all prescriptions.

4. A weighted association rule method based on the dosage of Chinese medicinal materials in Chinese medicine prescriptions according to claim 1, characterized in that: The step 4) is specifically: According to the standardized dose weighting matrix M 11 , then the medicinal material dosage weight matrix M2 is: Calculate the average of the dosage weights of each medicinal material in all prescriptions to obtain the medicinal material dosage weight w j .

5. The weighted association rule method based on the dosage of Chinese medicinal materials in Chinese medicine prescriptions according to claim 1 is characterized in that: The step 5) is specifically: Normalized unweighted basis matrix M 01 The Chinese medicine dosage weight matrix M3 is calculated by combining the medicinal material dosage weight matrix M2; Where i represents the i-th Chinese medicine prescription, w' i Represents the dosage weight of the i-th traditional Chinese medicine prescription.

6. A weighted association rule method based on the dosage of Chinese medicinal materials in Chinese medicine prescriptions according to claim 1, characterized in that: The step 6) is specifically: 6-1) Let the set of medicinal materials H = {H1, H2, H3, ..., H n }, and set the medicinal material subset A = {H n1 ,H n2 ,H n3 ,...,H nk }, k represents k medicinal materials, then the weighted support wsup of the medicinal material subset A is: in, a represents the collection of all medicinal materials inj The sum of the prescription weights that are not 0; Represents the sum of all prescription weights; 6-2) Let the medicinal material subset B = {H j }, then the weighted confidence wcon of the weighted association rule with the medicinal material subset A as the antecedent and the medicinal material subset B as the contingent is: 6-3) Obtain the weighted lift wlift with the medicinal material subset A as the antecedent and the medicinal material subset B as the posterior:

7. A weighted association rule method based on the dosage of Chinese medicinal materials in Chinese medicine prescriptions according to claim 1, characterized in that: According to the parameters of the weighted association rule algorithm, the parameters of the weighted association rule are obtained as the judgment criteria, specifically: 7-1) Keep the rule that weighted lift wlift>1; 7-2) Sort the retained rules according to the weighted support wsup, and analyze the results of the weighted association rule analysis according to wsup and the weighted lift wcon, namely: The higher the scores of weighted support wsup and weighted lift wcon, the stronger the association rule is considered to be, that is, the stronger the association between the medicinal materials is. Finally, the medicinal materials with the strongest association between the medicinal materials are screened to obtain the core medicinal material combination for treating the target disease.

8. A weighted association rule system based on the dosage of Chinese medicinal materials in Chinese medicine prescriptions according to claim 1, characterized in that: include: Data collection module, data specification module, matrix construction module, rule association module and standard judgment module; The data collection module is used to collect target effective Chinese medicine prescriptions in the database by keyword, and record all the dosages of medicinal materials used in the Chinese medicine prescriptions; The data standardization module is used to call the names of medicinal materials in the traditional Chinese medicine prescriptions in the data standardization module for standardization processing, and send the standardized medicinal material names and all the corresponding medicinal material doses to the matrix construction module; The matrix construction module is used to construct an unweighted basis matrix and a standardized unweighted basis matrix according to the standardized medicinal material names and all the corresponding medicinal material dosages; and process the Chinese medicine prescriptions, medicinal materials and the corresponding medicinal material dosages, the unweighted basis matrix and the standardized unweighted basis matrix to finally obtain the Chinese medicine prescription dosage weight matrix; A rule association module is used to receive and obtain various parameters of a weighted association rule model based on the collected medicinal material set and the weight matrix of the dosage of the traditional Chinese medicine prescription; The standard discrimination module is used to screen the medicinal materials with the strongest association among medicinal materials according to the various parameters of the weighted association rule model.

9. A weighted association rule system based on the dosage of Chinese medicinal materials in traditional Chinese medicine prescriptions, characterized in that: It comprises a memory and a processor; the memory is used to store a computer program; the processor is used to implement a weighted association rule method based on the dosage of Chinese medicinal materials in a traditional Chinese medicine prescription as described in any one of claims 1-7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, a weighted association rule method based on the dosage of Chinese medicinal materials in a traditional Chinese medicine prescription as described in any one of claims 1 to 7 is implemented.

Images