CN109887604A - A kind of quantization decision-making system of names of disease of tcm similarity - Google Patents
A kind of quantization decision-making system of names of disease of tcm similarity Download PDFInfo
- Publication number
- CN109887604A CN109887604A CN201910142196.0A CN201910142196A CN109887604A CN 109887604 A CN109887604 A CN 109887604A CN 201910142196 A CN201910142196 A CN 201910142196A CN 109887604 A CN109887604 A CN 109887604A
- Authority
- CN
- China
- Prior art keywords
- disease
- module
- drug
- name
- study
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a kind of quantization decision-making systems of names of disease of tcm similarity, including name of disease input module A, are used to input basic name of disease and comparison name of disease;Prescription data processing module B is used to handle basic name of disease and comparison name of disease, obtains the application collection that drugs with a higher frequency composition data and low frequency drug categories data are constituted;Discrimination model generation module C obtains drug composition rule tree and drug categories rule tree, generates quantization decision model for handling drugs with a higher frequency composition data and low frequency drug categories data;Model is applied and result output module D, is used to drugs with a higher frequency composition data and low frequency drug categories data application obtaining the similarity of comparison name of disease and basic name of disease to decision model is quantified, and show output;Intelligent control module E is used for error outside the bag according to the entropy and rule set of drug composition rule tree and drug categories rule tree, adjusts the ration of division of drugs with a higher frequency and low frequency drug.
Description
Technical field
The present invention relates to data minings and Artificial smart field, specifically, being related specifically to cure the desease in one kind
The quantization decision-making system of name similarity.
Background technique
With the development of data technique and artificial intelligence, prescription information existing for script magnanimity is able to digitization and standard
Change, has laid solid foundation for further intelligence.Since traditional Chinese medicine is built upon with the mode of thinking that side surveys card to prescription number
According to unique mode of thinking in comprehension of information and association base, therefore data mining and artificial intelligence technology are used, especially most
The utilization of new machine learning method, so that computer simulation Chinese medicine is achieved with the process that side surveys card thinking.
Data mining generally refers to the process for being hidden in wherein information by algorithm search from a large amount of data, usually with
Computer science is related, and by statistics, online analysis and processing, information retrieval, machine learning, expert system (by past
The rule of thumb) and all multi-methods such as pattern-recognition realize above-mentioned target.Artificial intelligence is that research makes computer to simulate people
The subject of certain thought processes and intelligent behavior (such as study, reasoning, thinking, planning) mainly includes that computer realizes intelligence
Principle, be manufactured similarly to the computer of human brain intelligence, enable a computer to realize higher level application.Card is surveyed with side to refer mainly to
The interpretation of the cause, onset and process of an illness or symptom of its cured mainly object are speculated according to prescription flavour of a drug composition and its effectiveness.
Summary of the invention
It is an object of the invention to aiming at the shortcomings in the prior art, provide a kind of quantization judgement of names of disease of tcm similarity
System combines data mining, artificial intelligence with magnanimity tcm prescription data, demonstrate,proves thinking to simulate Chinese medicine and survey with side
Journey quantifies different names of disease of tcm similarities.
Technical problem solved by the invention can be realized using following technical scheme:
A kind of quantization decision-making system of names of disease of tcm similarity, including
Name of disease input module A is used to input basic name of disease and comparison name of disease;
Prescription data processing module B is used for the name of disease input module A basic name of disease inputted and comparison name of disease progress side
Agent data processing obtains the application collection that drugs with a higher frequency composition data and low frequency drug categories data are constituted;
Discrimination model generation module C is used to handle drugs with a higher frequency composition data and low frequency drug categories data,
Drug composition rule tree and drug categories rule tree are obtained, and with the ratio of both artificial intelligence regulatings, generates quantization and determines mould
Type;
Model is applied and result output module D, and the drugs with a higher frequency for being used to obtain prescription data processing module B forms number
According to the quantization decision model generated with low frequency drug categories data application to discrimination model generation module C, obtain comparison name of disease with
The similarity of basic name of disease, and show output;
Intelligent control module E is used for entropy and rule set according to drug composition rule tree and drug categories rule tree
The outer error of bag, adjust the ration of division of drugs with a higher frequency and low frequency drug.
Further, the prescription data processing module B includes basic name of disease prescription data aggregation module B1, study collection point
Demoulding block B2, drugs with a higher frequency study module B3, low frequency drug study module B4 and comparison name of disease set of applications B5.
Further, the basic name of disease prescription data aggregation module B1, based on the total library of basic data, intelligent screening
The correspondence study group and control group prescription data acquisition system of the basic name of disease and comparison name of disease that need to quantify, and pass through artificial intelligence skill
Art is standardized.
Further, the study collection partition module B2 is constructed based on basis name of disease prescription data aggregation module B1, study
Collection partition module B2 selects suitable low-and high-frequency to distinguish standard by artificial intelligence, by the basic prescription data set after standardization point
It is segmented into high and low frequency drug two datasets.
Further, the drugs with a higher frequency study module B3 is based on study collection partition module B2 building, drugs with a higher frequency study
Module B3 is based on drugs with a higher frequency data set, rejects low frequency drug composition, constitutes the study group and control group of drugs with a higher frequency;
The low frequency drug study module B4 is based on study collection partition module B2 building, and low frequency drug study module B4 is based on
Low frequency drug data collection rejects drugs with a higher frequency composition, with the categorical clusters low frequency drug of recipe drug, constitutes low frequency drug
Habit group and control group.
Further, by repeating basic name of disease prescription data aggregation module B1, study collection partition module B2, drugs with a higher frequency
Study module B3 and low frequency drug study module B4, building comparison name of disease set of applications B5, compares name of disease set of applications B5 and serves as reasons
The application collection that drugs with a higher frequency composition data and low frequency drug categories data are constituted.
Further, the discrimination model generation module C includes drug composition rule generation module C1, drug categories rule
Module C2 and in proportion combination quantization decision model C3.
Further, the drug composition rule generation module C1 is constructed based on drugs with a higher frequency study module B3, medicine group
Control group is closed based on the drug composition study group of prescription at rule generation module C1 to pass through on the basis of random forests algorithm
Artificial intelligence regulation generates a certain number of rule trees;
The drug categories rule module C2 is constructed based on low frequency drug study module B4, drug categories rule module C2 base
Control group is combined in the Category Learning of prescription, on the basis of random forests algorithm, regulates and controls to generate a fixed number by artificial intelligence
The rule tree of amount.
Further, the quantization of the combination in proportion decision model C3 is based on drug composition rule generation module C1 and medicine
The other rule module C2 building of species, combination quantization decision model C3 is according to the entropy of create-rule tree and the bag of rule set in proportion
Outer error, intelligence adjust the ration of division of study collection partition module B2.
Further, the model is applied and result output module D is based on comparison name of disease set of applications B5 and in proportion group
Resultant decision model C3 building, is used to compare drugs with a higher frequency composition data and low frequency drug in name of disease set of applications B5
Categorical data is applied to combination quantization decision model C3 in proportion, obtains the similarity of comparison name of disease and basic name of disease, and shows
Output.
Further, the intelligent control module E is based on combination quantization decision model C3 building in proportion, intelligent control mould
Block E is used to adjust study collection according to error outside the bag of the entropy and rule set of drug composition rule tree and drug categories rule tree
Decouple the ration of division of module B2.
Compared with prior art, the beneficial effects of the present invention are:
Data source based on prescription composition with names of disease of tcm pairing constructs various judgment rule tree set, and then combines structure
At names of disease of tcm judgment models, different name of disease similarities are quantified by the model, based on the system can quantify names of disease of tcm it
Between the degree of correlation.
It is (random gloomy with VS.net and R language tool packet using data mining technology and Artificial intellectual technology
Woods), simulation establishes Chinese medicine and differentiates that names of disease of tcm model and application model quantify the complete of different name of disease similarities with side's survey card thinking
Process.The problem of present invention is TCM Document research, especially same sick different name provides the new research method of one kind and quantization work
Tool.
Detailed description of the invention
Fig. 1 is the structural block diagram of the quantization decision-making system of names of disease of tcm similarity of the present invention.
Specific embodiment
To be easy to understand the technical means, the creative features, the aims and the efficiencies achieved by the present invention, below with reference to
Specific embodiment, the present invention is further explained.
Referring to a kind of Fig. 1 quantization decision-making system of names of disease of tcm similarity of the present invention, including
Name of disease input module A is used to input basic name of disease and comparison name of disease;
Prescription data processing module B, the prescription data processing module B include basic name of disease prescription data aggregation module
B1, study collection partition module B2, drugs with a higher frequency study module B3, low frequency drug study module B4 and comparison name of disease set of applications
B5 is used to carry out prescription data processing to the name of disease input module A basic name of disease inputted and comparison name of disease, obtains drugs with a higher frequency
The application collection that composition data and low frequency drug categories data are constituted;
Discrimination model generation module C, the discrimination model generation module C include drug composition rule generation module C1, medicine
The other rule module C2 of species and in proportion combination quantization decision model C3.It is used for drugs with a higher frequency composition data and low frequency drug
Categorical data is handled, and obtains drug composition rule tree and drug categories rule tree, and with the ratio of both artificial intelligence regulatings
Example generates quantization decision model;
Model is applied and result output module D, and the drugs with a higher frequency for being used to obtain prescription data processing module B forms number
According to the quantization decision model generated with low frequency drug categories data application to discrimination model generation module C, obtain comparison name of disease with
The similarity of basic name of disease, and show output;
Intelligent control module E is used for entropy and rule set according to drug composition rule tree and drug categories rule tree
The outer error of bag, adjust the ration of division of drugs with a higher frequency and low frequency drug.
Embodiment
It is basic name of disease with " quenching one's thirst ", " above disappearing " is to compare name of disease, for the similitude of quantization " above disappearing " and " quenching one's thirst ", answers
Card quantization names of disease of tcm Similarity Model is surveyed with side with Chinese medicine.
Name of disease input module A inputs basic name of disease " quenching one's thirst ", compares name of disease " above disappearing ".
Basic name of disease prescription data aggregation module B1 retrieves " quenching one's thirst " and " above disappearing " in the total library of basic data, excludes to disappear
Prescription extracts control group out at random after the thirsty and different name that may quench one's thirst.Data import basic database, normalized to form following format.
On the basis of basic name of disease prescription data aggregation module B1, study collection partition module B2, study collection partition are established
Module B2 selects suitable low-and high-frequency to distinguish standard by artificial intelligence, carries out to the basic prescription data set after having been standardized
Segmentation, is divided into high and low frequency drug two datasets, data save in the following format:
On the basis of study collects partition module B2, drugs with a higher frequency study module B3, drugs with a higher frequency study module B3 are established
Based on drugs with a higher frequency data set, low frequency drug composition is rejected, constitutes new study group and control group, data are protected in the following format
It deposits:
On the basis of study collects partition module B2, low frequency drug study module B4, low frequency drug study module B4 are established
Based on low frequency drug data collection, drugs with a higher frequency composition is rejected, with the classification (four natures and five flavors of drug, channel tropism, potency classes) of recipe drug
Low frequency drug is clustered, constitutes new study group and control group, data save in the following format:
Module B2, drugs with a higher frequency study module B3 and low frequency drug study module B4 are decoupled by repetitive learning collection, is established
Name of disease set of applications B5 is compared, comparison name of disease set of applications B5 is made of drugs with a higher frequency composition data and low frequency drug categories data
Application collection, data save in the following format:
On the basis of drugs with a higher frequency study module B3, drug composition rule generation module C1, drug composition rule are established
Generation module C1 closes control group based on the drug composition study group of prescription and passes through artificial intelligence on the basis of random forests algorithm
It can regulate and control to generate a certain number of rule trees, data save in the following format:
On the basis of low frequency drug study module B4, drug categories rule module C2, drug categories rule module are established
C2 combines control group based on the Category Learning of prescription, on the basis of random forests algorithm, generates one by artificial intelligence regulation
The rule tree of fixed number amount, data save in the following format:
On the basis of drug composition rule generation module C1 and drug categories rule module C2, combined amount in proportion is established
Change decision model C3, combination quantization decision model C3 is in proportion with artificial intelligence regulating collocation drug composition rule tree and drug class
The ratio of other rule tree, constitutes quantization decision model, and the outer error format of total bag is as follows
Compare name of disease set of applications B5 and in proportion combination quantization decision model C3 on the basis of, establish model apply with
As a result output module D generates the data application to comparison name of disease set of applications B5 to combination quantization decision model C3 in proportion
Quantization decision model, obtain the similarity 88% of comparison name of disease and basic name of disease, and export display.
On the basis of combination quantization decision model C3 in proportion, establish intelligent control module E, intelligent control module E according to
The entropy of create-rule tree and the outer error of the bag of rule set, the setting of the relevant rudimentaries such as ration of division of intelligent adjustment module B2.
The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention.The technology of the industry
Personnel are it should be appreciated that the present invention is not limited to the above embodiments, and the above embodiments and description only describe this
The principle of invention, without departing from the spirit and scope of the present invention, various changes and improvements may be made to the invention, these changes
Change and improvement all fall within the protetion scope of the claimed invention.The claimed scope of the invention by appended claims and its
Equivalent thereof.
Claims (11)
1. a kind of quantization decision-making system of names of disease of tcm similarity, it is characterised in that: including
Name of disease input module A is used to input basic name of disease and comparison name of disease;
Prescription data processing module B is used to carry out prescription number to the name of disease input module A basic name of disease inputted and comparison name of disease
According to processing, the application collection that drugs with a higher frequency composition data and low frequency drug categories data are constituted is obtained;
Discrimination model generation module C is used to handle drugs with a higher frequency composition data and low frequency drug categories data, obtain
Drug composition rule tree and drug categories rule tree, and with the ratio of both artificial intelligence regulatings, generate quantization decision model;
Model apply and result output module D, be used for by prescription data processing module B obtain drugs with a higher frequency composition data and
The quantization decision model that low frequency drug categories data application is generated to discrimination model generation module C obtains comparison name of disease and basis
The similarity of name of disease, and show output;
Intelligent control module E is used for the bag of the entropy and rule set according to drug composition rule tree and drug categories rule tree
Outer error adjusts the ration of division of drugs with a higher frequency and low frequency drug.
2. the quantization decision-making system of names of disease of tcm similarity according to claim 1, it is characterised in that: the prescription data
Processing module B include basic name of disease prescription data aggregation module B1, study collection partition module B2, drugs with a higher frequency study module B3,
Low frequency drug study module B4 and comparison name of disease set of applications B5.
3. the quantization decision-making system of names of disease of tcm similarity according to claim 2, it is characterised in that: the basis name of disease
Prescription data aggregation module B1, based on the total library of basic data, basic name of disease and comparison name of disease that intelligent screening needs to quantify
Correspondence study group and control group prescription data acquisition system, and be standardized by artificial intelligence technology.
4. the quantization decision-making system of names of disease of tcm similarity according to claim 3, it is characterised in that: the study collection point
Demoulding block B2 is constructed based on basis name of disease prescription data aggregation module B1, and study collection partition module B2 selects to close by artificial intelligence
Suitable low-and high-frequency distinguishes standard, is high and low frequency drug two datasets by the basic prescription Segmentation of Data Set after standardization.
5. the quantization decision-making system of names of disease of tcm similarity according to claim 4, it is characterised in that:
The drugs with a higher frequency study module B3 is based on study collection partition module B2 building, and drugs with a higher frequency study module B3 is based on high frequency
Drug data collection rejects low frequency drug composition, constitutes the study group and control group of drugs with a higher frequency;
The low frequency drug study module B4 is based on study collection partition module B2 building, and low frequency drug study module B4 is based on low frequency
Drug data collection rejects drugs with a higher frequency composition, with the categorical clusters low frequency drug of recipe drug, constitutes the study group of low frequency drug
And control group.
6. the quantization decision-making system of names of disease of tcm similarity according to claim 5, it is characterised in that: by repeating basis
Name of disease prescription data aggregation module B1, study collection partition module B2, drugs with a higher frequency study module B3 and low frequency drug study module
B4, building comparison name of disease set of applications B5, comparison name of disease set of applications B5 are by drugs with a higher frequency composition data and low frequency drug class
The application collection that other data are constituted.
7. the quantization decision-making system of names of disease of tcm similarity according to claim 6, it is characterised in that: the discrimination model
Generation module C includes drug composition rule generation module C1, drug categories rule module C2 and combines quantization judgement mould in proportion
Type C3.
8. the quantization decision-making system of names of disease of tcm similarity according to claim 7, it is characterised in that:
The drug composition rule generation module C1 is constructed based on drugs with a higher frequency study module B3, drug composition rule generation module
C1 closes control group based on the drug composition study group of prescription, on the basis of random forests algorithm, regulates and controls to give birth to by artificial intelligence
At a certain number of rule trees;
The drug categories rule module C2 is constructed based on low frequency drug study module B4, and drug categories rule module C2 is based on side
The Category Learning of agent combines control group, on the basis of random forests algorithm, is generated by artificial intelligence regulation a certain number of
Rule tree.
9. the quantization decision-making system of names of disease of tcm similarity according to claim 8, it is characterised in that:
The quantization of the combination in proportion decision model C3 is based on drug composition rule generation module C1 and drug categories rule module
C2 building, combination quantization decision model C3 is adjusted according to error outside the entropy of create-rule tree and the bag of rule set, intelligence in proportion
The ration of division of section study collection partition module B2.
10. the quantization decision-making system of names of disease of tcm similarity according to claim 9, it is characterised in that: the model is answered
It is based on comparison name of disease set of applications B5 and in proportion combination quantization decision model C3 building with result output module D, is used for
The drugs with a higher frequency composition data in name of disease set of applications B5 and low frequency drug categories data application will be compared to combined amount in proportion
Change decision model C3, obtain the similarity of comparison name of disease and basic name of disease, and shows output.
11. the quantization decision-making system of names of disease of tcm similarity according to claim 10, it is characterised in that: the intelligence is adjusted
Control module E be based in proportion combination quantization decision model C3 building, intelligent control module E be used for according to drug composition rule tree with
The entropy of drug categories rule tree and the outer error of the bag of rule set, adjust the ration of division of study collection partition module B2.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910142196.0A CN109887604B (en) | 2019-02-26 | 2019-02-26 | Quantitative determination system for similarity of traditional Chinese medicine disease names |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910142196.0A CN109887604B (en) | 2019-02-26 | 2019-02-26 | Quantitative determination system for similarity of traditional Chinese medicine disease names |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109887604A true CN109887604A (en) | 2019-06-14 |
CN109887604B CN109887604B (en) | 2023-01-17 |
Family
ID=66929410
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910142196.0A Active CN109887604B (en) | 2019-02-26 | 2019-02-26 | Quantitative determination system for similarity of traditional Chinese medicine disease names |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109887604B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110544538A (en) * | 2019-08-23 | 2019-12-06 | 上海中医药大学 | Five-organ attribution judging system based on concept of traditional Chinese medicine |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102985924A (en) * | 2011-02-14 | 2013-03-20 | 松下电器产业株式会社 | Similar case retrieval device and similar case retrieval method |
CN104794340A (en) * | 2015-04-17 | 2015-07-22 | 南京大学 | Intelligent processing system for traditional Chinese medicine information |
CN107103196A (en) * | 2017-04-26 | 2017-08-29 | 成都中医药大学 | A kind of tcm clinical practice data cleaning method |
CN109215777A (en) * | 2018-08-03 | 2019-01-15 | 电子科技大学 | TCM Document intelligent excavating and prescription aid decision-making method and system |
-
2019
- 2019-02-26 CN CN201910142196.0A patent/CN109887604B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102985924A (en) * | 2011-02-14 | 2013-03-20 | 松下电器产业株式会社 | Similar case retrieval device and similar case retrieval method |
CN104794340A (en) * | 2015-04-17 | 2015-07-22 | 南京大学 | Intelligent processing system for traditional Chinese medicine information |
CN107103196A (en) * | 2017-04-26 | 2017-08-29 | 成都中医药大学 | A kind of tcm clinical practice data cleaning method |
CN109215777A (en) * | 2018-08-03 | 2019-01-15 | 电子科技大学 | TCM Document intelligent excavating and prescription aid decision-making method and system |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110544538A (en) * | 2019-08-23 | 2019-12-06 | 上海中医药大学 | Five-organ attribution judging system based on concept of traditional Chinese medicine |
CN110544538B (en) * | 2019-08-23 | 2022-02-18 | 上海中医药大学 | Five-organ attribution judging system based on concept of traditional Chinese medicine |
Also Published As
Publication number | Publication date |
---|---|
CN109887604B (en) | 2023-01-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Khalil et al. | Energy efficiency prediction using artificial neural network | |
CN101447020B (en) | Pornographic image recognizing method based on intuitionistic fuzzy | |
CN107766787A (en) | Face character recognition methods, device, terminal and storage medium | |
CN106919951A (en) | A kind of Weakly supervised bilinearity deep learning method merged with vision based on click | |
CN106845528A (en) | A kind of image classification algorithms based on K means Yu deep learning | |
CN106934038B (en) | A kind of medical data duplicate checking and the method and system associated | |
Chen et al. | A cooperative cuckoo search–hierarchical adaptive neuro-fuzzy inference system approach for predicting student academic performance | |
Indra et al. | Application of C4. 5 Algorithm for Cattle Disease Classification | |
CN100416599C (en) | Not supervised classification process of artificial immunity in remote sensing images | |
CN105578472B (en) | A kind of wireless sensor network performance online Method for optimized planning based on immunity principle | |
CN109978074A (en) | Image aesthetic feeling and emotion joint classification method and system based on depth multi-task learning | |
Karthikeyan et al. | A hybrid clustering approach using artificial bee colony (ABC) and particle swarm optimization | |
Ayumi et al. | A study on medicinal plant leaf recognition using artificial intelligence | |
CN109887604A (en) | A kind of quantization decision-making system of names of disease of tcm similarity | |
Gavhale et al. | Identification of medicinal plant using Machine learning approach | |
Tang et al. | Intrusive tumor growth inspired optimization algorithm for data clustering | |
Horzyk | Associative graph data structures with an efficient access via AVB+ trees | |
Xue et al. | Tree-like branching network for multi-class classification | |
Wang et al. | Visual information computing and processing model based on artificial neural network | |
Chou et al. | Text mining technique for Chinese written judgment of criminal case | |
Gao et al. | Statistics and Analysis of Targeted Poverty Alleviation Information Integrated with Big Data Mining Algorithm | |
Canuto et al. | Population-based bio-inspired algorithms for cluster ensembles optimization | |
CN107563401A (en) | The Ensemble classifier recognition methods of integrated classification and cluster | |
CN110544538B (en) | Five-organ attribution judging system based on concept of traditional Chinese medicine | |
Chen | Applicability Analysis of Data-driven Methods for Adolescents Mental Health Surveillance |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |