Medical Flow Data Cleaning Method and System Based on Intelligent Processing

By collecting and analyzing the characteristic text information and keywords of the medical flow data, matching the database type and data attribute feature information, and scientifically matching the data cleaning tools, the accurate and efficient search and intelligent and precise cleaning of the medical flow data are achieved, and the problems of low efficiency and quality of the traditional Chinese medicine flow data cleaning in the existing technology are solved.

CN119400333BActive Publication Date: 2025-06-13BEIJING YAOYUN DATA TECH CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202510006546.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-01-03
Publication Date
2025-06-13
Estimated Expiration
2045-01-03

AI Technical Summary

Technical Problem

The existing medical flow data cleaning processing cannot achieve accurate and efficient search of the required medical flow data, nor can it intelligently and accurately screen medical flow data cleaning tools, resulting in a reduction in the quality and efficiency of medical flow data cleaning.

Method used

By collecting the search feature text data of the target medical flow, identifying the search keywords, matching the database type, collecting the target data, collecting data attribute feature information, scientifically matching the data cleaning tool type, and independently performing the data cleaning operation, and finally visualizing the output data cleaning results.

Benefits of technology

It realizes accurate and efficient search and intelligent and precise matching of medical flow data, improves the quality and efficiency of medical flow data cleaning, and ensures the accuracy and reliability of data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119400333B_ABST
    Figure CN119400333B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of pharmaceutical flow data processing, and discloses a pharmaceutical flow data cleaning method and system based on intelligent processing. The system includes a pharmaceutical flow information acquisition module, a pharmaceutical flow information cleaning module, and a pharmaceutical flow information cleaning output module. By combining big data storage of pharmaceutical flow data cleaning tool type information with intelligent recognition algorithms and target pharmaceutical flow data attribute feature text information for intelligent matching of target pharmaceutical flow data cleaning tool types, refined and intelligent analysis of target pharmaceutical flow data cleaning tool types is achieved, improving the quality of pharmaceutical flow data cleaning. According to the analysis parameters of the target pharmaceutical flow data cleaning tool type, the target pharmaceutical flow collected data is autonomously and efficiently cleaned, realizing efficient and accurate operation of the target pharmaceutical flow data cleaning operation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of pharmaceutical flow data processing, and specifically to a method and system for cleaning pharmaceutical flow data based on intelligent processing. Background Art

[0002] The analysis of pharmaceutical flow data can be completed by using data visualization tools, statistical analysis methods, machine learning models for data processing, etc. Among them, data cleaning and preprocessing are the basic steps of pharmaceutical flow data analysis. Cleaning and preprocessing include deleting duplicate data, handling missing values, standardizing data, etc. Through these steps, the quality and consistency of the data can be ensured, thereby improving the accuracy and reliability of pharmaceutical data analysis results. Users can achieve scientific and accurate analysis of pharmaceutical flow data by cleaning and preprocessing the pharmaceutical flow data, providing strong support for users to make pharmaceutical sales decisions; however, the existing cleaning and processing of pharmaceutical flow data cannot accurately and efficiently search for the required pharmaceutical flow data, nor can it intelligently and accurately screen pharmaceutical flow data cleaning tools, reducing the quality and efficiency of pharmaceutical flow data cleaning.

[0003] The Chinese invention patent with the publication number CN112148851B discloses a method for constructing a pharmaceutical knowledge Q&A system based on a knowledge graph. By collecting pharmaceutical knowledge data and through data cleaning and processing, a pharmaceutical knowledge graph is constructed; collecting natural language questions and converting them into semantic query graphs to search for question answers in the pharmaceutical knowledge graph and performing visual output, improving the retrieval accuracy of pharmaceutical knowledge answers; however, the above technical solutions cannot scientifically and accurately match the required data cleaning tools based on the collected characteristic information of pharmaceutical knowledge data, reducing the quality of the construction of the pharmaceutical knowledge graph and the reliability of the pharmaceutical knowledge Q&A system. Summary of the Invention

[0004] (I) Technical Problems to be Solved

[0005] To solve the above problems that the existing cleaning and processing of pharmaceutical flow data cannot accurately and efficiently search for the required pharmaceutical flow data, nor can it intelligently and accurately screen pharmaceutical flow data cleaning tools, reducing the quality and efficiency of pharmaceutical flow data cleaning, and to achieve the above purposes of accurately collecting the search feature text information of the target pharmaceutical flow, accurately identifying the search keywords of the pharmaceutical flow, accurately matching the database type of the pharmaceutical flow search, accurately collecting the target pharmaceutical flow data, collecting the attribute characteristic information of the target pharmaceutical flow data, scientifically matching the type of pharmaceutical flow data cleaning tools, autonomously and efficiently executing the pharmaceutical flow data cleaning operation, and visually outputting and displaying the pharmaceutical flow data cleaning results.

[0006] (II) Technical Solutions

[0007] The present invention is realized through the following technical solutions: A method for cleaning pharmaceutical flow data based on intelligent processing, the method comprising the following steps:

[0008] S1. Collect target pharmaceutical flow search feature text data;

[0009] S2. Perform search keyword recognition processing on the target pharmaceutical flow information search feature text information based on the target pharmaceutical flow search feature text data and the pharmaceutical flow search keyword data to generate target pharmaceutical flow search keyword recognition data;

[0010] S3. Perform database object matching processing required for target pharmaceutical flow information search based on the target pharmaceutical flow search keyword recognition data and the pharmaceutical flow search database to generate a target pharmaceutical flow search database;

[0011] S4. Perform target pharmaceutical flow information collection processing based on the target pharmaceutical flow search feature text data and the target pharmaceutical flow search database to generate target pharmaceutical flow collection data;

[0012] S5. Collect target pharmaceutical flow data attribute feature text data and perform data cleaning tool type analysis processing required for target pharmaceutical flow data cleaning with the pharmaceutical flow data cleaning tool type data to generate target pharmaceutical flow data cleaning tool type analysis data;

[0013] S6. Perform target pharmaceutical flow data cleaning processing based on the target pharmaceutical flow data cleaning tool type analysis data and the target pharmaceutical flow collection data to generate target pharmaceutical flow cleaning data;

[0014] S7. Construct target pharmaceutical flow cleaning output data and perform target pharmaceutical flow data cleaning result output feedback operation.

[0015] Preferably, the operation steps for collecting the target pharmaceutical flow search feature text data are as follows:

[0016] S11. Online collect the search feature text information of the target pharmaceutical product flow required by the user terminal through the data collection dialog box of the pharmaceutical flow management platform, and generate target pharmaceutical flow search feature text data , and the target pharmaceutical flow search feature text data includes name feature text data, specification feature text data, batch number feature text data, production date feature text data, expiration date feature text data, sales volume feature text data, selling price feature text data, and manufacturer feature text data for the target pharmaceutical product flow search.

[0017] The present invention accurately collects the target pharmaceutical flow search feature text information through a data collection dialog box, achieving the effect of providing reliable data support for accurately identifying the search keywords of target pharmaceutical flow data.

[0018] Preferably, the operation steps for identifying the search keywords of the target pharmaceutical flow search feature text information based on the target pharmaceutical flow search feature text data and the pharmaceutical flow search keyword data to generate the target pharmaceutical flow search keyword identification data are as follows:

[0019] S21. Establish a pharmaceutical flow search keyword data set , ; where represents the pharmaceutical flow search keyword data corresponding to the th type of pharmaceutical flow search feature text information, represents the maximum value of the number of types of pharmaceutical flow search feature text information. The pharmaceutical flow search feature text information represents the search feature text information corresponding to searching different types of pharmaceutical flow data, and the pharmaceutical flow search keyword data represents the pharmaceutical flow data search keywords corresponding to different types of pharmaceutical flow search feature text information; the pharmaceutical flow search keyword data includes name keywords, specification keywords, batch number keywords, production date keywords, expiration date keywords, sales volume keywords, selling price keywords, and manufacturer keywords for pharmaceutical product flow data search;

[0020] S22. Use the XGBoost algorithm to perform character matching of the target pharmaceutical flow search feature text data with the pharmaceutical flow search keyword data set and the pharmaceutical flow search keyword data to search for the pharmaceutical flow search keyword data corresponding to the target pharmaceutical flow search feature text data , and construct a target pharmaceutical flow search keyword identification data set , , where represents the th target pharmaceutical flow search keyword identification data, represents the th target pharmaceutical flow search keyword identification data.

[0021] The present invention intelligently identifies the target pharmaceutical flow search keywords through scientifically preset pharmaceutical flow search keywords combined with the XGBoost algorithm and the target pharmaceutical flow search feature text information, achieving the effect of improving the search accuracy of pharmaceutical flow data.

[0022] Preferably, the database object matching process for searching the target pharmaceutical flow information by using the target pharmaceutical flow search keyword recognition data and the pharmaceutical flow search database is as follows:

[0023] S31. Establish a set of pharmaceutical flow search databases , ; where represents the th pharmaceutical flow search database, represents the maximum value of the number of types of pharmaceutical flow search databases. The pharmaceutical flow search database refers to a pharmaceutical flow data search database composed of different types of pharmaceutical flow search keyword feature identifiers; the pharmaceutical flow search database includes the Chinese Essential Medicines Database, the Domestic Drug Database, the Chinese Medical Insurance Database, the Chinese Hospital Database, the Chinese Commercial Database, the Chinese Pharmacy Database, and the Chinese Medical Institution Database;

[0024] S32. Use the iterative deepening search algorithm to match the target pharmaceutical flow search keyword recognition data in the target pharmaceutical flow search keyword recognition data set with the pharmaceutical flow search databases in the pharmaceutical flow search database set to for pharmaceutical flow search keyword matching, search for the corresponding pharmaceutical flow search databases of the target pharmaceutical flow search keyword recognition data from to and , and construct a set of target pharmaceutical flow search databases , , where represents the th target pharmaceutical flow search database, and represents the th target pharmaceutical flow search database.

[0025] The present invention realizes the intelligent screening of the target pharmaceutical flow search database type by combining the standard setting of the pharmaceutical flow search database with the iterative deepening search algorithm and the target pharmaceutical flow search keyword, and achieves the effect of accurately searching the target pharmaceutical flow search database based on data analysis.

[0026] Preferably, the operation steps for collecting the target pharmaceutical flow information based on the target pharmaceutical flow search feature text data and the target pharmaceutical flow search database are as follows:

[0027] S41. Input the target pharmaceutical flow search feature text data into the target pharmaceutical flow search database set of the target pharmaceutical flow search databases to in the corresponding database search dialog boxes and perform the search and collection operation of the target pharmaceutical flow information;

[0028] S42. Generate the target pharmaceutical flow collection data by generating data identifiers for the target pharmaceutical flow information collected and searched in the databases corresponding to the target pharmaceutical flow search databases in step S41 to The target pharmaceutical flow collection data includes the target pharmaceutical flow data in EXCEL format, the target pharmaceutical flow data in web format, the target pharmaceutical flow data in picture format, the target pharmaceutical flow data in WORD format, and the target pharmaceutical flow data in PDF format.

[0029] The present invention efficiently and accurately collects and processes the target pharmaceutical flow information based on the target pharmaceutical flow search feature text information and the target pharmaceutical flow search database, achieving the effect of autonomously and efficiently searching the target pharmaceutical flow data.

[0030] Preferably, the operation steps for collecting the attribute feature text data of the target pharmaceutical flow data and performing the analysis and processing of the data cleaning tool type required for cleaning the pharmaceutical flow data with the data cleaning tool type data of the pharmaceutical flow are as follows:

[0031] S51. Use a two-way search algorithm to search and collect the attribute feature text information of the pharmaceutical flow data corresponding to the target pharmaceutical flow collection data and generate the target pharmaceutical flow data attribute feature text data The target pharmaceutical flow data attribute feature text data includes the file type attribute feature text data, the file size attribute feature text data, the occupied space attribute feature text data, and the file version attribute feature text data of the target pharmaceutical flow data;

[0032] S52. Establish a data set of the data cleaning tool types of the pharmaceutical flow , where represents the th data cleaning tool type data of the pharmaceutical flow, ​Represents the maximum value of the number of types of pharmaceutical flow data cleaning tools, and the types of pharmaceutical flow data cleaning tools include PowerQuery, OpenRefine, TableauPrep, and FineDataLink; the data of the types of pharmaceutical flow data cleaning tools represents the optimal information on the types of pharmaceutical flow data cleaning tools set for the text information of the attribute characteristics of different types of pharmaceutical flow data.

[0033] S53. Compare the text data of the attribute characteristics of the target pharmaceutical flow data with the data set of the types of pharmaceutical flow data cleaning tools and the data of the types of pharmaceutical flow data cleaning tools in the data set of the types of pharmaceutical flow data cleaning tools to perform character matching of the attribute characteristics of the pharmaceutical flow data, search for the data of the types of pharmaceutical flow data cleaning tools corresponding to the text data of the attribute characteristics of the target pharmaceutical flow data, and construct an analysis data set of the types of target pharmaceutical flow data cleaning tools . The specific operations for generating the analysis data set of the types of target pharmaceutical flow data cleaning tools are as follows:

[0034] S531. Parameter initialization: The population size of the bat for searching the data cleaning tool is , the maximum number of iterations is T, the objective function is F, and the position of the bat individual for searching the data cleaning tool in the search space of the data set of the types of pharmaceutical flow data cleaning tools , , ; and the velocity , the sound wave frequency , the sound wave loudness and the frequency ;

[0035] S532. Search for the position of the optimal bat individual for searching the data cleaning tool in the search space of the data set of the types of pharmaceutical flow data cleaning tools , that is, search for the position of the data of the types of pharmaceutical flow data cleaning tools in the search space of the data set of the types of pharmaceutical flow data cleaning tools that is most matched with the text data of the attribute characteristics of the target pharmaceutical flow data by the bat population for searching the data cleaning tool, and update the velocity and position. The velocity and position update formulas are as follows: , where , and where represents the bat individual for searching the data cleaning tool After iterations, in the data set of the medical flow data cleaning tool type the speed in the search space, representing the bat individual searching for the data cleaning tool After iterations, in the data set of the medical flow data cleaning tool type the speed in the search space; representing the bat individual searching for the data cleaning tool After iterations, in the data set of the medical flow data cleaning tool type the position in the search space, representing the bat individual searching for the data cleaning tool After iterations, in the data set of the medical flow data cleaning tool type the position in the search space;

[0036] S533. Generate random numbers rand1 and rand1 is a random number in the interval [0, 1]. When rand1 > , select an optimal bat individual searching for the data cleaning tool from the best bat individuals searching for the data cleaning tool, that is, in the data set of the medical flow data cleaning tool type search the search space for the medical flow data cleaning tool type data that best matches the target medical flow data attribute characteristic text data of the position. Near the selected optimal bat individual searching for the data cleaning tool, through the formula , , generate a local solution, that is, in the data set of the medical flow data cleaning tool type search the search space for the medical flow data cleaning tool type data that best matches the target medical flow data attribute characteristic text data , where represents the bat individual searching for the data cleaning tool After iterations, in the data set of the medical flow data cleaning tool type the sound intensity in the search space, representing the bat individual searching for the data cleaning tool After iterations, in the data set of the medical flow data cleaning tool type the frequency in the search space, representing the bat individual searching for the data cleaning tool After iterations, the initial frequency in the data set of the medical flow data cleaning tool type in the search space; otherwise, according to the formula , , update the search bat position of the data cleaning tool, that is, in the data set of the medical flow data cleaning tool type in the search space, search for and update the medical flow data cleaning tool type data matching the target medical flow data attribute feature text data where represents the search bat individual of the data cleaning tool After iterations, the sound intensity in the search space of the data set of the medical flow data cleaning tool type , represents the search bat individual of the data cleaning tool After iterations, the frequency in the search space of the data set of the medical flow data cleaning tool type , represents a random function with a value of , and are constants and 0 < < 1, > 0;

[0037] S534. Then generate another random number rand2, which is a random number on [0, 1]; when rand2 < , and at this time the fitness of the objective function F is better than the new solution in S533, that is, in the data set of the medical flow data cleaning tool type in the search space, search for and update the medical flow data cleaning tool type data matching the target medical flow data attribute feature text data whose fitness is better than that of the medical flow data cleaning tool type data matched in S533, then accept the medical flow data cleaning tool type data in the search space of the data set of the medical flow data cleaning tool type and update the position. The position update formula is as follows: , and according to the formula , , synchronously adjust and ; where represents the search bat individual of the data cleaning tool In the data set of the medical flow data cleaning tool type a new position in the search space, indicating the bat individual searched by the data cleaning tool In the data set of the medical flow data cleaning tool type an old position in the search space, indicating any number within the range of [-1, 1];

[0038] S535. Sort the fitness values of all individuals in the bat population searched by the data cleaning tool, and find the current best , that is, search in the data set of the medical flow data cleaning tool type for the position of the medical flow data cleaning tool type data that best matches the target medical flow data attribute characteristic text data in the search space; ;

[0039] S536. Repeat S532 to S535. When the maximum number of iterations T is satisfied, output the medical flow data cleaning tool type data that matches the target medical flow data attribute characteristic text data ;

[0040] S537. Take the medical flow data cleaning tool type data output in step S536 that matches the target medical flow data attribute characteristic text data , and construct an analysis data set of the target medical flow data cleaning tool type , , where represents the th analysis data of the target medical flow data cleaning tool type, represents the th analysis data of the target medical flow data cleaning tool type.

[0041] The present invention performs intelligent matching of the target medical flow data cleaning tool type by combining the information of the medical flow data cleaning tool type based on big data storage with the bat optimization algorithm and the text information of the target medical flow data attribute characteristics collected, achieving the effect of refined and intelligent analysis of the target medical flow data cleaning tool type.

[0042] Preferably, according to the analysis data of the target medical flow data cleaning tool type and the collected data of the target medical flow, the operation steps for performing target medical flow data cleaning processing to generate target medical flow cleaning data are as follows:

[0043] S61. When the analysis data set of the target pharmaceutical flow data cleaning tool type is generated, according to the analysis data set of the target pharmaceutical flow data cleaning tool type in the target pharmaceutical flow data cleaning tool type analysis data to the corresponding pharmaceutical flow data cleaning tool is used to clean the collected data of the target pharmaceutical flow perform data cleaning operations according to the attribute characteristics of the pharmaceutical flow data, and generate the cleaned data of the target pharmaceutical flow .

[0044] The present invention autonomously and efficiently cleans the collected data of the target pharmaceutical flow according to the analysis parameters of the target pharmaceutical flow data cleaning tool type, achieving the effect of efficient and accurate operation of the target pharmaceutical flow data cleaning operation.

[0045] Preferably, the operation steps of constructing the output data of the cleaned target pharmaceutical flow and performing the feedback operation of the output of the cleaned target pharmaceutical flow data are as follows:

[0046] S71. Obtain the cleaned data of the target pharmaceutical flow and construct the output data of the cleaned target pharmaceutical flow through data identification ;

[0047] S72. Push the output data of the cleaned target pharmaceutical flow to the pharmaceutical flow management platform through the Internet communication network and use a display screen to output the feedback of the cleaned pharmaceutical flow data.

[0048] The present invention constructs the output data of the cleaned target pharmaceutical flow based on data processing science, and cooperates with the display screen to visually and visually feedback and output the display operation of the cleaned target pharmaceutical flow data, achieving the effect of efficient collection and timely and accurate output feedback of the cleaned target pharmaceutical flow data.

[0049] A pharmaceutical flow data cleaning system based on intelligent processing is used to implement the pharmaceutical flow data cleaning method based on intelligent processing. The system includes a pharmaceutical flow information acquisition module, a pharmaceutical flow information cleaning module, and a pharmaceutical flow information cleaning output module;

[0050] The pharmaceutical flow information acquisition module includes a target pharmaceutical flow search feature information collection unit, a pharmaceutical flow search keyword storage unit, a target pharmaceutical flow search keyword recognition unit, a pharmaceutical flow search database storage unit, a target pharmaceutical flow search database matching unit, and a target pharmaceutical flow data collection unit;

[0051] The target medical drug flow search feature information collection unit collects target medical drug flow search feature text data through a data collection dialog box; the medical drug flow search keyword storage unit is used to store medical drug flow search keyword data; the target medical drug flow search keyword recognition unit performs search keyword recognition processing on the target medical drug flow information search feature text information based on the target medical drug flow search feature text data and the medical drug flow search keyword data, generating target medical drug flow search keyword recognition data; the medical drug flow search database storage unit is used to store the medical drug flow search database; the target medical drug flow search database matching unit performs database object matching processing required for target medical drug flow information search based on the target medical drug flow search keyword recognition data and the medical drug flow search database, generating the target medical drug flow search database; the target medical drug flow data collection unit performs target medical drug flow information collection processing based on the target medical drug flow search feature text data and the target medical drug flow search database, generating target medical drug flow collection data;

[0052] The medical drug flow information cleaning module includes a target medical drug flow data attribute feature information collection unit, a medical drug flow data cleaning tool storage unit, a target medical drug flow data cleaning tool analysis unit, and a target medical drug flow data cleaning operation execution unit;

[0053] The target medical drug flow data attribute feature information collection unit collects target medical drug flow data attribute feature text data; the medical drug flow data cleaning tool storage unit is used to store medical drug flow data cleaning tool type data; the target medical drug flow data cleaning tool analysis unit performs analysis processing on the data cleaning tool type required for target medical drug flow data cleaning based on the target medical drug flow data attribute feature text data and the medical drug flow data cleaning tool type data, generating target medical drug flow data cleaning tool type analysis data; the target medical drug flow data cleaning operation execution unit performs target medical drug flow data cleaning processing based on the target medical drug flow data cleaning tool type analysis data and the target medical drug flow collection data, generating target medical drug flow cleaning data;

[0054] The medical drug flow information cleaning output module includes a medical drug flow data cleaning result acquisition unit and a medical drug flow data cleaning result output feedback unit;

[0055] The medical drug flow data cleaning result acquisition unit is used to construct target medical drug flow cleaning output data; the medical drug flow data cleaning result output feedback unit pushes the target medical drug flow cleaning output data to the medical drug flow management platform through the Internet communication network and combines it with a display screen to execute the target medical drug flow data cleaning result output feedback operation.

[0056] (III) Beneficial effects

[0057] The present invention provides a method and system for cleaning pharmaceutical flow data based on intelligent processing, which has the following beneficial effects:

[0058] First, by using a data acquisition dialog box to accurately collect the target pharmaceutical flow search feature text information, it provides reliable data support for accurately identifying the target pharmaceutical flow data search keywords; scientifically presetting the pharmaceutical flow search keywords and combining intelligent search algorithms with the target pharmaceutical flow search feature text information to perform intelligent identification of the target pharmaceutical flow search keywords, which improves the accuracy of pharmaceutical flow data search; standardizing the setting of the pharmaceutical flow search database and combining intelligent search algorithms with the target pharmaceutical flow search keywords to perform intelligent screening of the target pharmaceutical flow search database type, realizing accurate search of the target pharmaceutical flow search database based on data analysis, and improving the search efficiency and quality of pharmaceutical flow data; efficiently and accurately collecting and processing the target pharmaceutical flow information based on the target pharmaceutical flow search feature text information and the target pharmaceutical flow search database, realizing autonomous and efficient search of the target pharmaceutical flow data, and improving the intelligence and reliability of pharmaceutical flow data search.

[0059] Second, by accurately collecting the target pharmaceutical flow data attribute feature text information based on the data attribute feature keywords, it provides real data support for scientifically matching the target pharmaceutical flow data cleaning tool type; based on the big data storage of the pharmaceutical flow data cleaning tool type information and combining intelligent recognition algorithms with the target pharmaceutical flow data attribute feature text information to perform intelligent matching of the target pharmaceutical flow data cleaning tool type, realizing refined and intelligent analysis of the target pharmaceutical flow data cleaning tool type, and improving the cleaning quality of pharmaceutical flow data; autonomously and efficiently cleaning the target pharmaceutical flow collected data according to the target pharmaceutical flow data cleaning tool type analysis parameters, realizing efficient and accurate operation of the target pharmaceutical flow data cleaning operation.

[0060] Third, by scientifically constructing the target pharmaceutical flow cleaning output data based on data processing and cooperating with the display screen to visually and intuitively feedback and output the target pharmaceutical flow data cleaning result display operation, it realizes efficient collection and timely and accurate output feedback of the target pharmaceutical flow cleaning data, and improves the efficiency and accuracy of pharmaceutical flow data cleaning feedback. Description of the Drawings

[0061] Figure 1 It is a schematic diagram of the modules of the pharmaceutical flow data cleaning system based on intelligent processing provided by the present invention;

[0062] Figure 2 It is a flowchart of the method for cleaning pharmaceutical flow data based on intelligent processing provided by the present invention. Detailed Embodiments

[0063] The following will clearly and completely describe the technical solutions in the embodiments of the present invention 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 of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0064] The embodiments of the medical flow data cleaning method and system based on intelligent processing are as follows:

[0065] Embodiment 1:

[0066] Please refer to Figure 1 - Figure 2 , for the medical flow data cleaning method based on intelligent processing, the method includes the following steps:

[0067] S1. Collect the target medical flow search feature text data;

[0068] S2. Perform search keyword recognition processing on the target medical flow search feature text data and the medical flow search keyword data to generate target medical flow search keyword recognition data;

[0069] S3. Perform database object matching processing required for target medical flow information search according to the target medical flow search keyword recognition data and the medical flow search database to generate a target medical flow search database;

[0070] S4. Perform target medical flow information collection processing based on the target medical flow search feature text data and the target medical flow search database to generate target medical flow collection data;

[0071] S5. Collect the target medical flow data attribute feature text data and perform data cleaning tool type analysis processing required for target medical flow data cleaning with the medical flow data cleaning tool type data to generate target medical flow data cleaning tool type analysis data;

[0072] S6. Perform target medical flow data cleaning processing according to the target medical flow data cleaning tool type analysis data and the target medical flow collection data to generate target medical flow cleaning data;

[0073] S7. Construct the target medical flow cleaning output data and perform the target medical flow data cleaning result output feedback operation.

[0074] Furthermore, please refer to Figure 1 - Figure 2 , the operation steps for collecting the target medical flow search feature text data are as follows:

[0075] S11. Online collect the search feature text information of the target pharmaceutical product flow required by the user terminal through the data collection dialog box of the pharmaceutical flow management platform, and generate the target pharmaceutical flow search feature text data The target pharmaceutical flow search feature text data includes the name feature text data, specification feature text data, batch number feature text data, production date feature text data, expiration date feature text data, sales volume feature text data, selling price feature text data, and manufacturer feature text data for the search of the target pharmaceutical product flow.

[0076] The operation steps for identifying the search keywords of the target pharmaceutical flow information search feature text information based on the target pharmaceutical flow search feature text data and the pharmaceutical flow search keyword data are as follows:

[0077] S21. Establish a pharmaceutical flow search keyword data set , ; where represents the pharmaceutical flow search keyword data corresponding to the th type of pharmaceutical flow search feature text information type, represents the maximum value of the number of pharmaceutical flow search feature text information types. The pharmaceutical flow search feature text information type represents the search feature text information corresponding to the search of different types of pharmaceutical flow data, and the pharmaceutical flow search keyword data represents the pharmaceutical flow data search keywords corresponding to different types of pharmaceutical flow search feature text information; the pharmaceutical flow search keyword data includes the name keyword, specification keyword, batch number keyword, production date keyword, expiration date keyword, sales volume keyword, selling price keyword, and manufacturer keyword for the search of pharmaceutical product flow data;

[0078] S22. Use the XGBoost algorithm to match the characters of the target pharmaceutical flow search feature text data with the pharmaceutical flow search keyword data set and the pharmaceutical flow search keyword data to search for the pharmaceutical flow search keyword data corresponding to the target pharmaceutical flow search feature text data , and construct a target pharmaceutical flow search keyword identification data set , , where represents the th target pharmaceutical flow search keyword identification data, represents the th target pharmaceutical flow search keyword identification data.

[0079] Match the data identified by the search keywords for the target pharmaceutical flow with the database objects required for searching the target pharmaceutical flow information in the pharmaceutical flow search database. The operation steps for generating the target pharmaceutical flow search database are as follows:

[0080] S31. Establish a set of pharmaceutical flow search databases , ; where represents the th pharmaceutical flow search database, represents the maximum value of the number of types of pharmaceutical flow search databases. The pharmaceutical flow search database represents a pharmaceutical flow data search database composed of different types of pharmaceutical flow search keyword features; the pharmaceutical flow search database includes the Chinese Essential Medicine Database, the Domestic Drug Database, the Chinese Medical Insurance Database, the Chinese Hospital Database, the Chinese Commercial Database, the Chinese Pharmacy Database, and the Chinese Medical Institution Database;

[0081] S32. Use the iterative deepening search algorithm to match the target pharmaceutical flow search keyword identification data in the set of the target pharmaceutical flow search keyword identification data to with the pharmaceutical flow search databases in the set of pharmaceutical flow search databases to search for the pharmaceutical flow search databases corresponding to the target pharmaceutical flow search keyword identification data to and construct a set of target pharmaceutical flow search databases , , where represents the th target pharmaceutical flow search database, represents the th target pharmaceutical flow search database, represents the th target pharmaceutical flow search database.

[0082] The operation steps for collecting the target pharmaceutical flow information based on the target pharmaceutical flow search feature text data and the target pharmaceutical flow search database to generate the target pharmaceutical flow collection data are as follows:

[0083] S41. Input the target pharmaceutical flow search feature text data into the corresponding database search dialog boxes of the target pharmaceutical flow search databases in the set of target pharmaceutical flow search databases from to and execute the search and collection operation of the target pharmaceutical flow information;

[0084] S42. Search the target pharmaceutical flow search database in step S41 to The target pharmaceutical flow information collected by searching the corresponding database is used to generate target pharmaceutical flow collection data through data identification , and the target pharmaceutical flow collection data includes target pharmaceutical flow data in EXCEL format, target pharmaceutical flow data in web version, target pharmaceutical flow data in picture version, target pharmaceutical flow data in WORD version, and target pharmaceutical flow data in PDF version.

[0085] Through the target pharmaceutical flow search feature information collection unit, the target pharmaceutical flow search feature text information is accurately collected using a data collection dialog box, providing reliable data support for accurately identifying the search keywords of target pharmaceutical flow data; the pharmaceutical flow search keyword storage unit and the target pharmaceutical flow search keyword recognition unit cooperate with each other to scientifically preset the pharmaceutical flow search keywords and use intelligent search algorithms and the target pharmaceutical flow search feature text information to intelligently identify the target pharmaceutical flow search keywords, improving the accuracy of pharmaceutical flow data search; the pharmaceutical flow search database storage unit and the target pharmaceutical flow search database matching unit cooperate with each other to standardize the setting of the pharmaceutical flow search database and use intelligent search algorithms and the target pharmaceutical flow search keywords to intelligently screen the type of target pharmaceutical flow search database, realizing accurate search of the target pharmaceutical flow search database based on data analysis and improving the search efficiency and quality of pharmaceutical flow data; the target pharmaceutical flow data collection unit efficiently and accurately collects and processes the target pharmaceutical flow information based on the target pharmaceutical flow search feature text information and the target pharmaceutical flow search database, realizing autonomous and efficient search of target pharmaceutical flow data and improving the intelligence and reliability of pharmaceutical flow data search.

[0086] Further, please refer to Figure 1 - Figure 2 , the operation steps for collecting the target pharmaceutical flow data attribute feature text data and analyzing and processing the data cleaning tool type required for target pharmaceutical flow data cleaning with the pharmaceutical flow data cleaning tool type data are as follows:

[0087] S51. Use a two-way search algorithm to search and collect the pharmaceutical flow data attribute feature text information corresponding to the target pharmaceutical flow collection data and generate target pharmaceutical flow data attribute feature text data , and the target pharmaceutical flow data attribute feature text data includes file type attribute feature text data, file size attribute feature text data, occupied space attribute feature text data, and file version attribute feature text data of the target pharmaceutical flow data;

[0088] S52. Establish a data set of data cleaning tool types for the flow of medical drugs , ; where represents the th type of data cleaning tool type data for the flow of medical drugs, represents the maximum value of the number of data cleaning tool types for the flow of medical drugs. The data cleaning tool types for the flow of medical drugs include PowerQuery, OpenRefine, TableauPrep, and FineDataLink; the data cleaning tool type data for the flow of medical drugs represents the optimal data cleaning tool type information set for the text information of the attribute characteristics of different types of medical drug flow data;

[0089] S53. Match the text data of the attribute characteristics of the target medical drug flow data with the data cleaning tool type data in the data set of data cleaning tool types for the flow of medical drugs to perform character matching of the attribute characteristics of the medical drug flow data, search for the data cleaning tool type data corresponding to the text data of the attribute characteristics of the target medical drug flow data , and construct an analysis data set of the target data cleaning tool type for the flow of medical drugs . The specific operations for generating the analysis data set of the target data cleaning tool type for the flow of medical drugs are as follows: S531. Parameter initialization: The population size of the data cleaning tool search bats is

[0090] , the maximum number of iterations is T, the objective function is F, and the position of the data cleaning tool search bat individual in the search space of the data set of data cleaning tool types for the flow of medical drugs is , ; and the velocity , the sound wave frequency , the sound wave loudness , and the frequency are ;

[0091] S532. Search for the position of the optimal data cleaning tool search bat individual in the data cleaning tool search bat population in the search space of the data set of data cleaning tool types for the flow of medical drugs , that is, search for the data cleaning tool type data in the search space of the data set of data cleaning tool types for the flow of medical drugs that is the most matched with the text data of the attribute characteristics of the target medical drug flow data by the data cleaning tool search bat population , which is the most matched data cleaning tool type data Position, and update the speed and position. The speed and position update formulas are as follows: , , where represents the speed of the data cleaning tool searching for bat individuals at iterations in the data set of the data cleaning tool type for the flow of medicine in the search space, represents the speed of the data cleaning tool searching for bat individuals at iterations in the data set of the data cleaning tool type for the flow of medicine in the search space; represents the position of the data cleaning tool searching for bat individuals at iterations in the data set of the data cleaning tool type for the flow of medicine in the search space, represents the position of the data cleaning tool searching for bat individuals at iterations in the data set of the data cleaning tool type for the flow of medicine in the search space;

[0092] S533. Generate random numbers rand1 and rand1 is a random number in the interval [0, 1]. When rand1 > , select an optimal data cleaning tool searching for bat individuals from the best data cleaning tool searching for bat individuals, that is, search in the data set of the data cleaning tool type for the flow of medicine in the search space for the data cleaning tool type data of the flow of medicine that best matches the target text data of the attribute characteristics of the flow of medicine . At the position near the selected optimal data cleaning tool searching for bat individuals, through the formula , , , generate a local solution, that is, search in the data set of the data cleaning tool type for the flow of medicine in the search space for the data cleaning tool type data of the flow of medicine that best matches the target text data of the attribute characteristics of the flow of medicine . Among them , where represents the sound wave loudness of the data cleaning tool searching for bat individuals at iterations in the data set of the data cleaning tool type for the flow of medicine in the search space, represents the sound wave loudness of the data cleaning tool searching for bat individuals at After the th iteration, the frequency in the search space, represents the search bat individual of the data cleaning tool In After the th iteration, the initial frequency in the search space; otherwise, according to the formula , , update the search bat position of the data cleaning tool, that is, in the data set of the data cleaning tool type for the pharmaceutical flow direction Search space to update the data cleaning tool type data of the pharmaceutical flow direction that matches the target pharmaceutical flow direction data attribute feature text data , where represents the search bat individual of the data cleaning tool In After the th iteration, the sound wave loudness in the search space of the data set of the data cleaning tool type for the pharmaceutical flow direction , represents the search bat individual of the data cleaning tool In After the th iteration, the frequency in the search space of the data set of the data cleaning tool type for the pharmaceutical flow direction represents a random function with a value of , And are constants and 0 < < 1, > 0;

[0093] S534. Then generate another random number rand2, and rand2 is a random number on [0, 1]; when rand2 < , and at this time the fitness of the objective function F is better than the new solution in S533, that is, in the data set of the data cleaning tool type for the pharmaceutical flow direction Search space to search for the data cleaning tool type data of the pharmaceutical flow direction that matches the target pharmaceutical flow direction data attribute feature text data , The fitness of is better than the data cleaning tool type data of the pharmaceutical flow direction matched in S533 , then accept the data cleaning tool type data of the pharmaceutical flow direction In the search space of the data set of the data cleaning tool type for the pharmaceutical flow direction, and update the position. The position update formula is as follows: , and according to the formula , , , synchronously adjust And ; among which represents that the data cleaning tool searches for bat individuals in the data set of the data cleaning tool type for the flow of medicine at a new position in the search space, represents that the data cleaning tool searches for bat individuals in the data set of the data cleaning tool type for the flow of medicine at an old position in the search space, represents any number within the range of [-1, 1];

[0094] S535. Sort the fitness values of all individuals in the bat population searched by the data cleaning tool, and find the current best , that is, search in the data set of the data cleaning tool type for the flow of medicine to find the position of the data of the data cleaning tool type for the flow of medicine that best matches the text data of the attribute characteristics of the target medicine flow data ;

[0095] S536. Repeat S532 to S535. When the maximum number of iterations T is satisfied, output the data of the data cleaning tool type for the flow of medicine that matches the text data of the attribute characteristics of the target medicine flow data ;

[0096] S537. Take the data of the data cleaning tool type for the flow of medicine output in step S536 that matches the text data of the attribute characteristics of the target medicine flow data , and construct the analysis data set of the data cleaning tool type for the target medicine flow data , , among which represents the th analysis data of the data cleaning tool type for the target medicine flow data, represents the th analysis data of the data cleaning tool type for the target medicine flow data.

[0097] The operation steps for performing the target medicine flow data cleaning process based on the analysis data of the data cleaning tool type for the target medicine flow data and the target medicine flow collection data to generate the target medicine flow cleaning data are as follows:

[0098] S61. When the generation of the analysis data set of the data cleaning tool type for the target medicine flow data is completed, according to the analysis data of the data cleaning tool type for the target medicine flow data in the analysis data set of the data cleaning tool type for the target medicine flow data from the th The corresponding pharmaceutical flow data cleaning tool collects data on the target pharmaceutical flow Perform data cleaning operations according to the attribute characteristics of the pharmaceutical flow data, and generate the cleaned data of the target pharmaceutical flow .

[0099] Through the target pharmaceutical flow data attribute feature information collection unit, accurately collect the text information of the target pharmaceutical flow data attribute features based on the data attribute feature keywords, providing real data support for scientifically matching the types of target pharmaceutical flow data cleaning tools; the pharmaceutical flow data cleaning tool storage unit and the target pharmaceutical flow data cleaning tool analysis unit cooperate with each other, and based on the big data storage of the pharmaceutical flow data cleaning tool type information, combine the intelligent recognition algorithm with the text information of the target pharmaceutical flow data attribute features to perform intelligent matching of the target pharmaceutical flow data cleaning tool type, realizing refined and intelligent analysis of the target pharmaceutical flow data cleaning tool type, and improving the quality of pharmaceutical flow data cleaning; the target pharmaceutical flow data cleaning operation execution unit autonomously and efficiently cleans the collected data of the target pharmaceutical flow according to the analysis parameters of the target pharmaceutical flow data cleaning tool type, realizing efficient and accurate operation of the target pharmaceutical flow data cleaning operation.

[0100] Further, please refer to Figure 1 - Figure 2 , and the operating steps for constructing the cleaned output data of the target pharmaceutical flow and performing the output feedback operation of the target pharmaceutical flow data cleaning result are as follows:

[0101] S71. Obtain the cleaned data of the target pharmaceutical flow And construct the cleaned output data of the target pharmaceutical flow through data identification ;

[0102] S72. Push the cleaned output data of the target pharmaceutical flow To the pharmaceutical flow management platform through the Internet communication network and use a display screen to perform feedback output of the pharmaceutical flow cleaning data.

[0103] Through the cooperation of the pharmaceutical flow data cleaning result acquisition unit and the pharmaceutical flow data cleaning result output feedback unit, scientifically construct the cleaned output data of the target pharmaceutical flow based on data processing, and cooperate with the display screen to visually and intuitively perform the feedback output of the target pharmaceutical flow data cleaning result display operation, realizing efficient collection and timely and accurate output feedback of the target pharmaceutical flow cleaning data, and improving the efficiency and accuracy of the pharmaceutical flow data cleaning feedback.

[0104] Example 2:

[0105] Please refer to Figure 1 - Figure 2, a pharmaceutical flow data cleaning system based on intelligent processing, is used to implement a pharmaceutical flow data cleaning method based on intelligent processing. The system includes a pharmaceutical flow information acquisition module, a pharmaceutical flow information cleaning module, and a pharmaceutical flow information cleaning output module;

[0106] The pharmaceutical flow information acquisition module includes a target pharmaceutical flow search feature information collection unit, a pharmaceutical flow search keyword storage unit, a target pharmaceutical flow search keyword recognition unit, a pharmaceutical flow search database storage unit, a target pharmaceutical flow search database matching unit, and a target pharmaceutical flow data collection unit;

[0107] The target pharmaceutical flow search feature information collection unit collects target pharmaceutical flow search feature text data through a data collection dialog box; the pharmaceutical flow search keyword storage unit is used to store pharmaceutical flow search keyword data; the target pharmaceutical flow search keyword recognition unit performs search keyword recognition processing on the target pharmaceutical flow information search feature text information based on the target pharmaceutical flow search feature text data and the pharmaceutical flow search keyword data, and generates target pharmaceutical flow search keyword recognition data; the pharmaceutical flow search database storage unit is used to store the pharmaceutical flow search database; the target pharmaceutical flow search database matching unit performs database object matching processing required for target pharmaceutical flow information search based on the target pharmaceutical flow search keyword recognition data and the pharmaceutical flow search database, and generates a target pharmaceutical flow search database; the target pharmaceutical flow data collection unit performs target pharmaceutical flow information collection processing based on the target pharmaceutical flow search feature text data and the target pharmaceutical flow search database, and generates target pharmaceutical flow collection data;

[0108] The pharmaceutical flow information cleaning module includes a target pharmaceutical flow data attribute feature information collection unit, a pharmaceutical flow data cleaning tool storage unit, a target pharmaceutical flow data cleaning tool analysis unit, and a target pharmaceutical flow data cleaning operation execution unit;

[0109] The target pharmaceutical flow data attribute feature information collection unit collects target pharmaceutical flow data attribute feature text data; the pharmaceutical flow data cleaning tool storage unit is used to store pharmaceutical flow data cleaning tool type data; the target pharmaceutical flow data cleaning tool analysis unit performs data cleaning tool type analysis processing required for target pharmaceutical flow data cleaning based on the target pharmaceutical flow data attribute feature text data and the pharmaceutical flow data cleaning tool type data, and generates target pharmaceutical flow data cleaning tool type analysis data; the target pharmaceutical flow data cleaning operation execution unit performs target pharmaceutical flow data cleaning processing based on the target pharmaceutical flow data cleaning tool type analysis data and the target pharmaceutical flow collection data, and generates target pharmaceutical flow cleaning data;

[0110] The pharmaceutical flow information cleaning and output module includes a pharmaceutical flow data cleaning result acquisition unit and a pharmaceutical flow data cleaning result output and feedback unit;

[0111] The pharmaceutical flow data cleaning result acquisition unit is used to construct the target pharmaceutical flow cleaning and output data; the pharmaceutical flow data cleaning result output and feedback unit pushes the target pharmaceutical flow cleaning and output data to the pharmaceutical flow management platform through the Internet communication network and combines it with the display screen to execute the target pharmaceutical flow data cleaning result output and feedback operation.

[0112] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for cleaning medical flow data based on intelligent processing, characterized in that: The method comprises the following steps: S1. Collect target medicine flow search feature text data O; S2, performing search keyword recognition processing on the target medicine flow information search feature text information according to the target medicine flow search feature text data and the medicine flow search keyword data, and generating target medicine flow search keyword recognition data; S3, according to the target medicine flow search keyword identification data and the medicine flow search database, the database object matching processing required for the target medicine flow information search is performed to generate the target medicine flow search database, and the target medicine flow search database set B'=(b' v1 ,…,b' v2 ), 1≤v1≤v≤v2≤τ, where b' v1 represents the v1th target medicine flow search database, b' v2 represents the v2th target medicine flow search database, τ represents the maximum number of medicine flow search database types; S4, performing target medicine flow information collection and processing based on the target medicine flow search feature text data and the target medicine flow search database to generate target medicine flow collection data; The S4 comprises the following steps: S41, inputting the O into the b' in the B' v1 to b' v2 In the corresponding database search dialog box, the search and collection operation of the target medicine flow information is performed; S42, b' in S41 v1 to b' v2 The target medicine flow information collected by the corresponding database search is used to generate target medicine flow collection data P through data identification, and the target medicine flow collection data includes EXCEL version target medicine flow data, web page version target medicine flow data, picture version target medicine flow data, WORD version target medicine flow data and PDF version target medicine flow data; S5. Collecting target medicine flow data attribute feature text data and performing data cleaning tool type analysis processing on medicine flow data cleaning tool type data required for target medicine flow data cleaning, and generating target medicine flow data cleaning tool type analysis data; The S5 comprises the following steps: S51, using a bidirectional search algorithm to search and collect the attribute feature text information of the medicine flow data corresponding to P, and generate target medicine flow data attribute feature text data L; S52. Establish a data set of medicine flow data cleaning tool type C = (c1,…,c k ,…,c υ ), k = 1, 2, 3, ..., υ; where c k represents the kth type of medicine flow data cleaning tool type data, υ represents the maximum number of medicine flow data cleaning tool types, the medicine flow data cleaning tool types include PowerQuery, OpenRefine, TableauPrep and FineDataLink, and the medicine flow data cleaning tool type data represents the optimal medicine flow data cleaning tool type information set for different types of medicine flow data attribute feature text information; S53, the L and the c in the C k Perform a character match on the attribute feature of the medicine flow data to search for the c corresponding to the L k , and construct the target medicine flow data cleaning tool type analysis data set C′; S6, performing target medicine flow data cleaning processing according to the target medicine flow data cleaning tool type analysis data and the target medicine flow collection data to generate target medicine flow cleaning data; S7. Construct the target medicine flow cleaning output data and execute the target medicine flow data cleaning result output feedback operation.

2. The method for cleaning medicine flow data based on intelligent processing according to claim 1, characterized in that: The S1 comprises the following steps: S11. Collect the search feature text information of the target pharmaceutical product flow required by the user end online through the data collection dialog box of the pharmaceutical flow management platform, and generate the target pharmaceutical flow search feature text data O.

3. The method for cleaning medicine flow data based on intelligent processing according to claim 2 is characterized in that: The S2 comprises the following steps: S21. Establishing a data set of keywords for searching medicine flow where a u represents the medicine flow search keyword data corresponding to the u-th medicine flow search feature text information type, Indicates the maximum number of types of feature text information for drug flow search; S22, using XGBoost algorithm to add O, A and a u Perform character matching of the medical flow search feature text information to search for the a corresponding to the O u , and construct a target medical flow search keyword recognition data set where a′ u1 represents the u1th target medicine flow search keyword recognition data, a′ u2 Represents the u2th target medical flow search keyword recognition data.

4. The method for cleaning medicine flow data based on intelligent processing according to claim 3 is characterized by: The S3 comprises the following steps: S31, establish a medicine flow search database set B = (b1, ..., b v ,…,b τ ), v = 1, 2, 3, ..., τ; where b v represents a vth medicine flow search database, wherein the medicine flow search database represents a medicine flow data search database composed of different types of medicine flow search keyword feature identifiers; S32, using an iterative deepening search algorithm to find the a' in A' u1 to a' u2 with the b in the b v Perform drug flow search keyword matching to search for the target drug flow search keyword identification data a' u1 to a' u2 The corresponding medicine flow search database b v , and construct a target medicine flow search database set B′.

5. The method for cleaning medicine flow data based on intelligent processing according to claim 4 is characterized in that: The specific operation of generating the target medicine flow data cleaning tool type analysis data set C′ in S53 is as follows: S531, parameter initialization: the data cleaning tool searches for the bat population size υ, the maximum number of iterations T, the objective function F, the data cleaning tool searches for the position H of the bat individual i in the C search space i ,i=1,2,…,υ;and speed◇ i , sound wave frequency Loudness of sound waves Sum frequency Λ i ; S532, searching the C search space for the optimal data cleaning tool to search for the bat individual position H* in the bat population, that is, the data cleaning tool searches the bat population in the C search space for the c that best matches the L. k location; S533, generate a random number rand1, rand1 is a random number in the interval [0, 1], when rand1>Λ i , select an optimal data cleaning tool to search for bat individuals in the optimal data cleaning tool search for bat individuals, that is, search for the C that best matches the L in the C search space k The optimal data cleaning tool is used to search for the location of the bat individual and generate a local solution, that is, to search for the c that best matches the L in the C search space. k ; Update the data cleaning tool to search for bat positions, that is, update the search for the c that matches the L in the C search space k location; S534, generate another random number rand2, rand2 is a random number on [0, 1]; when At this time, the fitness of the objective function F is better than the new solution in S533, that is, the c matching the L is searched in the C search space. k The fitness of is better than the c matched in S533 k , then accept the c k position in the C search space and update the position; S535, sorting the fitness values ​​of all individuals in the bat population searched by the data cleaning tool, and finding the current optimal H*, that is, searching the C that best matches the L in the C search space. k location; S536, repeat S532 to S535, and when the maximum number of iterations T is met, output the c that matches the L k ; S537, outputting the c that matches the L in step S536 k , and construct the target medicine flow data cleaning tool type analysis data set C'=(c' k1 ,…,c' k2 ), 1≤k1≤k≤k2≤υ, where c' k1 represents the type of analysis data of the k1th target medical flow data cleaning tool, c' k2 Represents the k2th target medical flow data cleaning tool type analysis data.

6. The method for cleaning medicine flow data based on intelligent processing according to claim 5, characterized in that: The S6 comprises the following steps: S61, when the generation of C' is completed, according to the c' in C' k1 to c′ k2 The corresponding medicine flow data cleaning tool performs data cleaning operations on the P according to the attribute characteristics of the medicine flow data, and generates target medicine flow cleaning data P′.

7. The method for cleaning medicine flow data based on intelligent processing according to claim 6, characterized in that: The S7 comprises the following steps: S71, obtaining the P' and constructing the target medicine flow cleaning output data g through data identification; S72, pushing the g to the medicine flow management platform through the Internet communication network and using a display screen to output the medicine flow cleaning data feedback.

8. A medicine flow data cleaning system based on intelligent processing, used to implement the medicine flow data cleaning method based on intelligent processing according to any one of claims 1 to 7, characterized in that: The system includes a medicine flow information acquisition module, a medicine flow information cleaning module, and a medicine flow information cleaning output module.

Citation Information

Patent Citations

  • A method for constructing a medical knowledge question-answering system based on knowledge graph

    CN112148851B

  • Intelligent data matching method in medicine flow direction data cleaning

    CN112685392A

  • Target data determination method and device, computer equipment, a storage medium

    CN114020733A