A Health Knowledge Mining Method, System and Terminal Based on a Dynamic Feedback Mechanism

Abstract

The present invention relates to a health knowledge mining method, system and terminal based on a dynamic feedback mechanism, belonging to the technical field of knowledge mining, and includes: collecting the health knowledge mining requirements input by the user and the current time point; determining the requirement type based on the health knowledge mining requirements; selecting a mining database based on the requirement type; retrieving the database update time point and the update reference time value based on the mining database; calculating an update interval time value based on the current time point and the database update time point; updating and adjusting the mining database based on the update interval time value and the update reference time value to obtain an adjusted database; performing knowledge mining from the adjusted database based on the health knowledge mining requirements to generate a health knowledge mining result, and outputting the health knowledge mining result. The present invention has the effect of improving the accuracy of the mining result.

Description

Technical Field

[0001] The present invention relates to the technical field of knowledge mining, and in particular, to a health knowledge mining method, system and terminal based on a dynamic feedback mechanism. Background Art

[0002] Knowledge mining technology is a method that combines various technologies such as data mining, machine learning, natural language processing, statistical analysis, etc. to extract useful knowledge such as patterns, relationships, trends and rules from a large amount of data, so as to provide support for decision-making.

[0003] When performing knowledge mining on health knowledge, generally after clarifying the mining objectives and problems, it is necessary to collect data related to the problems from various sources, preprocess the original data, then perform feature selection on the preprocessed data, and establish a model to discover knowledge in the data. Finally, the mined knowledge and information are transformed into actual decisions and actions to support medical decision-making and health management.

[0004] When performing knowledge mining, it is necessary to collect data related to the problems from various sources, and the distribution and characteristics of the data collected from various sources are likely to change over time, so errors or biases are likely to occur during the collection, storage and processing processes, which in turn affects the accuracy of the mining results. Summary of the Invention

[0005] In order to improve the accuracy of the mining results, the present invention provides a health knowledge mining method, system and terminal based on a dynamic feedback mechanism.

[0006] In a first aspect, the present invention provides a health knowledge mining method based on a dynamic feedback mechanism, adopting the following technical solutions:

[0007] A health knowledge mining method based on a dynamic feedback mechanism includes:

[0008] S1: Collect the health knowledge mining requirements input by the user and the current time point;

[0009] S2: Determine the requirement type based on the health knowledge mining requirements;

[0010] S3: Select a mining database based on the requirement type;

[0011] S4: Retrieve the database update time point and the update reference time value based on the mining database;

[0012] S5: Calculate the update interval time value based on the current time point and the database update time point;

[0013] S6: Update and adjust the mining database based on the update interval time value and the update reference time value to obtain an adjusted database;

[0014] S7: Conduct knowledge mining from the adjusted database based on the health knowledge mining requirement to generate a health knowledge mining result, and output the health knowledge mining result.

[0015] Optionally, step S3 includes the following steps:

[0016] S31: Determine single-type source information based on the requirement type;

[0017] S32: Retrieve source storage information and source content information based on the single-type source information;

[0018] S33: Determine a source retrieval reference value based on the source storage information;

[0019] S34: Determine a content overlap value based on the source content information;

[0020] S35: Determine an overlap reference value based on the content overlap value;

[0021] S36: Select the single-type source information based on the source retrieval reference value and the overlap reference value to form selected source information;

[0022] S37: Determine a selected database based on the selected source information, and use the selected database as the mining database.

[0023] Optionally, step S33 includes the following steps:

[0024] S331: Retrieve a source storage value and storage type information based on the source storage information;

[0025] S332: Sort the source storage values from largest to smallest, and use the sorting result as a storage sorting value;

[0026] S333: Determine a type retrieval speed value based on the storage type information;

[0027] S334: Sort the type retrieval speed values from largest to smallest, and use the sorting result as a retrieval speed sorting value;

[0028] S335: Calculate the difference between the storage sorting value and the retrieval speed sorting value as a sorting deviation value;

[0029] S336: Generate a sorting deviation reference value based on the sorting deviation value and the storage sorting value, and use the sorting deviation reference value as the source retrieval reference value.

[0030] Optionally, step S336 includes the following steps:

[0031] S3361: Determine whether the sorting deviation value is greater than a preset sorting deviation reference value;

[0032] S3362: If so, calculate the difference between the sorting deviation value and the sorting deviation reference value and use it as the sorting anomaly deviation value;

[0033] S3363: Calculate the quotient of the source storage value and the type retrieval speed value and use it as the retrieval time value;

[0034] S3364: Sort in ascending order based on the retrieval time value, and use the sorting result as the retrieval time sorting value;

[0035] S3365: Generate an initial time sorting reference value based on the retrieval time sorting value;

[0036] S3366: Adjust the initial time sorting reference value based on the sorting anomaly deviation value to form a time sorting adjustment reference value, and use the time sorting adjustment reference value as the sorting deviation reference value;

[0037] S3367: If not, generate a storage sorting reference value based on the storage sorting value, and use the storage sorting reference value as the sorting deviation reference value.

[0038] Optionally, step S34 includes the following steps:

[0039] S341: Retrieve content keyword information and content repeated word information based on the source content information;

[0040] S342: Determine the keyword coincidence value based on the content keyword information;

[0041] S343: Determine the repeated word coincidence value based on the content repeated word information;

[0042] S344: Determine the comprehensive coincidence value based on the keyword coincidence value and the repeated word coincidence value;

[0043] S345: Determine whether there is a coincidence between the content keyword information and the content repeated word information;

[0044] S346: If so, determine the content overlapping word information based on the content keyword information and the content repeated word information;

[0045] S347: Determine the adjusted coincidence value based on the content overlapping word information and the comprehensive coincidence value, and use the adjusted coincidence value as the content coincidence value;

[0046] S348: If the answer is no, directly use the comprehensive coincidence value as the content coincidence value.

[0047] Optionally, step S342 includes the following steps:

[0048] S3421: Determine keyword meaning information and content synonym information based on the content keyword information;

[0049] S3422: Determine meaning-related word information based on the keyword meaning information;

[0050] S3423: Combine the content synonym information and the meaning-related word information to form key comprehensive word information;

[0051] S3424: Conduct a coincidence query based on the key comprehensive word information to obtain comprehensive coincidence word information and the corresponding comprehensive coincidence count value;

[0052] S3425: Determine a coincidence reference benchmark value based on the comprehensive coincidence word information;

[0053] S3426: Calculate the product value between the coincidence reference benchmark value and the comprehensive coincidence count value and use it as the single coincidence word coincidence value;

[0054] S3427: Calculate the sum value of all the single coincidence word coincidence values and use it as the keyword coincidence value.

[0055] Optionally, step S343 includes the following steps:

[0056] S3431: Retrieve single repeated word information and the corresponding repeated count value based on the content repeated word information;

[0057] S3432: Sort based on the repeated count value from largest to smallest, and use the sorting result as the repeated count sorting value;

[0058] S3433: Determine the repeated count sorting influence value based on the repeated count sorting value;

[0059] S3434: Conduct a coincidence query based on the single repeated word information to obtain repeated coincidence word information and the corresponding repeated coincidence count value;

[0060] S3435: Determine the repeated coincidence word coincidence value based on the repeated coincidence word information and the repeated coincidence count value;

[0061] S3436: Adjust the repeated coincidence word coincidence value based on the repeated count sorting influence value to obtain the repeated word coincidence value.

[0062] Optionally, the step S347 includes the following steps:

[0063] S3471: Retrieving the keyword count value based on the content keyword information;

[0064] S3472: Retrieving the repeated word count value based on the content repeated word information;

[0065] S3473: Calculating the sum value between the keyword count value and the repeated word count value and using it as the reference word count value;

[0066] S3474: Retrieving the overlapping word count value based on the content overlapping word information;

[0067] S3475: Calculating the quotient value between the overlapping word count value and the reference word count value and using it as the overlapping word occupancy ratio;

[0068] S3476: Determining the overlapping word occupancy ratio influence value based on the overlapping word occupancy ratio;

[0069] S3477: Adjusting the comprehensive overlap value based on the overlapping word occupancy ratio influence value to form the adjusted overlap value.

[0070] In a second aspect, the present invention provides a health knowledge mining system based on a dynamic feedback mechanism, adopting the following technical solution:

[0071] A health knowledge mining system based on a dynamic feedback mechanism, comprising:

[0072] An acquisition module, configured to acquire health knowledge mining requirements and the current time point;

[0073] A memory, configured to store a health knowledge mining method according to any one of the first aspect;

[0074] A processor, configured to load and execute the program in the memory.

[0075] In a third aspect, the present invention provides a terminal, adopting the following technical solution:

[0076] A terminal, comprising a memory and a processor, wherein a computer program is stored on the memory and can be loaded and executed by the processor to perform a health knowledge mining method according to any one of the first aspect.

[0077] In summary, the present invention includes at least one of the following beneficial technical effects:

[0078] 1. By collecting the health knowledge mining requirements and the current time point and determining the requirement type, then selecting the mining database according to the requirement type to retrieve the database update time point and the update reference time value, calculating the update interval time value, and then updating and adjusting the mining database to obtain the adjusted database, and then performing knowledge mining from the adjusted database according to the health knowledge mining requirements to generate and output the health knowledge mining result, so as to perform knowledge mining after updating the database to be used, thereby improving the accuracy of the mining result;

[0079] 2. By determining the single-type source information according to the requirement type and retrieving the source storage information and the source content information, thereby determining the source retrieval reference value and the coincidence reference value and selecting the single-type source information to form the selected source information, and determining the selected database as the mining database according to the selected source information, thereby improving the accuracy of the obtained mining database;

[0080] 3. By retrieving the source storage value and the storage type information from the source storage information to obtain the storage sorting value and the retrieval speed sorting value respectively, calculating the sorting deviation value and generating the sorting deviation reference value with the storage sorting value, and using the sorting deviation reference value as the source retrieval reference value, thereby improving the accuracy of the obtained source retrieval reference value. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 is the flowchart of the method for mining health knowledge based on a dynamic feedback mechanism according to an embodiment of the present application;

[0082] Figure 2 is the flowchart of the method for selecting a mining database according to an embodiment of the present application;

[0083] Figure 3 is the flowchart of the method for determining the source retrieval reference value according to an embodiment of the present application;

[0084] Figure 4 is the flowchart of the method for generating the sorting deviation reference value according to an embodiment of the present application;

[0085] Figure 5 is the flowchart of the method for determining the content coincidence value according to an embodiment of the present application;

[0086] Figure 6 is the flowchart of the method for determining the keyword coincidence value according to an embodiment of the present application;

[0087] Figure 7 is the flowchart of the method for determining the repeated word coincidence value according to an embodiment of the present application;

[0088] Figure 8 is the flowchart of the method for determining the adjusted coincidence value according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0089] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

[0090] A health knowledge mining method based on a dynamic feedback mechanism, which collects the health knowledge mining requirements and the current time point, selects the mining database and adjusts it according to the current time point, and then conducts knowledge mining from the adjusted database through the health knowledge mining requirements to generate and output the health knowledge mining results, so as to conduct knowledge mining after updating the database to be used, thereby improving the accuracy of the mining results.

[0091] Refer to Figure 1 , embodiments of the present invention disclose a health knowledge mining method based on a dynamic feedback mechanism, which includes:

[0092] S1: Collect the health knowledge mining requirements input by the user and the current time point.

[0093] Among them, the health knowledge mining requirement refers to the requirement corresponding to the user's desire to conduct knowledge mining on health-related knowledge at the current time, and the health knowledge mining requirement is obtained through user input. The current time point refers to the time point at which the current time is located, and the current time point is obtained by querying the database that records time in real time at the current time.

[0094] S2: Determine the requirement type based on the health knowledge mining requirements.

[0095] Among them, the requirement type refers to the knowledge type corresponding to the user's need to conduct knowledge mining on health-related knowledge. The requirement type includes types such as disease-related knowledge, healthy lifestyle, healthcare information, emerging health technologies and concepts, etc. By retrieving the theme corresponding to the health knowledge mining requirement and querying the preset type database according to the requirement theme to obtain the requirement type, the type database pre-stores different requirement themes and the corresponding requirement types, and the type database is obtained through pre-input.

[0096] S3: Select the mining database based on the requirement type.

[0097] Among them, the mining database refers to the database that needs to be used for knowledge mining. By retrieving the database related to the requirement type involved in knowledge mining through the requirement type and then selecting the retrieved database, the mining database is obtained for convenient subsequent use. The specific selection steps of the mining database refer to S31 to S37.

[0098] S4: Retrieve the database update time point and the update reference time value based on the mining database.

[0099] Among them, the database update time point refers to the time point when the database last updated the content, and the update reference time value refers to the maximum interval time value that the database can tolerate when updating the content. The mining database stores the database update time point and the update reference time value, and the database update time point and the update reference time value are retrieved through the mining database for convenient subsequent use.

[0100] S5: Calculate the update interval time value based on the current time point and the database update time point.

[0101] Among them, the update interval time value refers to the time value between the time when the mining database last updated the content and the current time. By calculating the interval time between the current time point and the database update time point, the update interval time value is obtained for convenient subsequent use.

[0102] S6: Based on the update interval time value and the update reference time value, adjust the mining database to obtain an adjusted database.

[0103] Among them, the adjusted database refers to the database after adjusting the mining database. By judging whether the update interval time value is greater than the update reference time value, when the update interval time value is greater than the update reference time value, it indicates that the mining database needs to be updated and adjusted at this time. Therefore, the source corresponding to the mining database is reread and updated to obtain the adjusted database. When the update interval time value is not greater than the update reference time value, it indicates that the mining database does not need to be updated and adjusted at this time. Therefore, the mining database is directly used as the adjusted database for convenient subsequent use.

[0104] S7: Conduct knowledge mining from the adjusted database based on the health knowledge mining requirements to generate a health knowledge mining result and output the health knowledge mining result.

[0105] Among them, the health knowledge mining result refers to the result obtained after knowledge mining according to the user's requirements. Knowledge mining is carried out from the adjusted database through the health knowledge mining requirements, thereby generating and outputting the health knowledge mining result. Thus, knowledge mining is carried out after updating the database to be used, thereby improving the accuracy of the mining result. The specific steps of conducting knowledge mining are prior art and will not be elaborated further.

[0106] In step S3, in order to further ensure the rationality of the mining database, it is necessary to conduct a further separate analysis and calculation on the mining database, which is specifically described in detail through the following steps.

[0107] Refer to Figure 2 , the selection method of the mining database includes the following steps:

[0108] S31: Determine the single-type source information based on the requirement type.

[0109] Among them, the single-type source information refers to the source information of the database involved in a single requirement type. Different requirement types correspond to different single-type source information. By inputting the requirement type into a preset source query database, the single-type source information can be obtained, which is convenient for subsequent use. The source query database pre-stores different requirement types and their corresponding single-type source information, and the source query database is obtained through pre-input.

[0110] S32: Retrieve the source storage information and source content information based on the single-type source information.

[0111] Among them, the source storage information refers to the data situation information stored in the database corresponding to the single-type source information, and the source content information refers to the data content information stored in the database corresponding to the single-type source information. By using the single-type source information to retrieve the corresponding database as the source database, the source storage information and source content information are stored in the source database. Retrieving the source storage information and source content information through the source database is convenient for subsequent use.

[0112] S33: Determine the source retrieval reference value based on the source storage information.

[0113] Among them, the source retrieval reference value refers to the reference value corresponding to the selection reference based on the retrieval situation of the source database. By analyzing the source storage information, the source retrieval reference value can be determined, which is convenient for subsequent use. The specific steps for determining the source retrieval reference value refer to S331 to S336.

[0114] S34: Determine the content coincidence value based on the source content information.

[0115] Among them, the content coincidence value refers to the coincidence situation when the stored content of the source database corresponding to a single requirement type coincides with that of other databases. By analyzing the source content information, the content coincidence value can be determined, which is convenient for subsequent use. The specific steps for determining the content coincidence value refer to S341 to S348.

[0116] S35: Determine the coincidence reference value based on the content coincidence value.

[0117] Among them, the coincidence reference value refers to the reference value corresponding to the selection reference based on the coincidence situation of the stored content of the source database. Different content coincidence values correspond to different coincidence reference values. By inputting the content coincidence value into a preset coincidence reference database, the coincidence reference value can be obtained, which is convenient for subsequent use. The coincidence reference database pre-stores different content coincidence values and their corresponding coincidence reference values, and the coincidence reference database is obtained through pre-input.

[0118] S36: Retrieve the reference value and the coincidence reference value based on the source to select the single-type source information to form the selected source information.

[0119] Among them, the selected source information refers to the source information corresponding to the selected single-type source information. By calculating the sum value of the source retrieval reference value and the coincidence reference value, sorting the calculated sum value, and selecting the single-type source information corresponding to the preset benchmark selection number according to the sorting result as the selected source information for convenient subsequent use. The benchmark selection number refers to the number of sources to be selected, which is obtained through pre-input.

[0120] S37: Determine the selected database based on the selected source information and use the selected database as the mining database.

[0121] Among them, the selected database refers to the database corresponding to the selected source database. By using the source database corresponding to the selected source information as the selected database and using the selected database as the mining database, the accuracy of the obtained mining database can be improved.

[0122] In step S33, in order to further ensure the rationality of the source retrieval reference value, it is necessary to perform a further separate analysis and calculation on the source retrieval reference value, which is specifically described in detail through the following steps.

[0123] Refer to Figure 3 , the method for determining the source retrieval reference value includes the following steps:

[0124] S331: Retrieve the source storage value and the storage type information based on the source storage information.

[0125] Among them, the source storage value refers to the storage value of the data stored in the source database, and the storage type information refers to the type information of the data stored in the source database. Retrieving the source database based on the source storage information, where the source storage value and the storage type information are stored in the source database, facilitating subsequent use by retrieving the source storage value and the storage type information from the source database.

[0126] S332: Sort the source storage values from largest to smallest and use the sorting result as the storage sorting value.

[0127] Among them, the storage sorting value refers to the result value corresponding to the sorting based on the storage amount of the data. By sorting the source storage values from largest to smallest and using the sorting result as the storage sorting value, it is convenient for subsequent use.

[0128] S333: Determine the type retrieval speed value based on the storage type information.

[0129] Among them, the type retrieval speed value refers to the speed value corresponding to the retrieval of data types. Different storage type information corresponds to different type retrieval speed values. By inputting the storage type information into a preset retrieval speed database to obtain the type retrieval speed value, it is convenient for subsequent use. The retrieval speed database pre-stores different storage type information and the corresponding type retrieval speed values, and the retrieval speed database is obtained through pre-input.

[0130] S334: Sort in descending order based on the type retrieval speed value, and use the sorting result as the retrieval speed sorting value.

[0131] Among them, the retrieval speed sorting value refers to the result value corresponding to the sorting based on the retrieval speed of the data. By sorting the type retrieval speed value in descending order and using the sorting result as the retrieval speed sorting value, it is convenient for subsequent use.

[0132] S335: Calculate the difference between the storage sorting value and the retrieval speed sorting value and use it as the sorting deviation value.

[0133] Among them, the sorting deviation value refers to the deviation value corresponding to the situation where there is a deviation in the sorting results of the data storage amount and the retrieval speed. By calculating the difference between the storage sorting value and the retrieval speed sorting value and using it as the sorting deviation value, it is convenient for subsequent use.

[0134] S336: Generate a sorting deviation reference value based on the sorting deviation value and the storage sorting value, and use the sorting deviation reference value as the source retrieval reference value.

[0135] Among them, the sorting deviation reference value refers to the reference value corresponding to the reference for source selection based on the deviation situation of the sorting. By analyzing the sorting deviation value and the storage sorting value, a sorting deviation reference value is generated, and the sorting deviation reference value is used as the source retrieval reference value to improve the accuracy of the obtained source retrieval reference value. The specific generation steps of the sorting deviation reference value refer to S3361 to S3367.

[0136] In step S336, in order to further ensure the rationality of the sorting deviation reference value, it is necessary to perform a further separate analysis and calculation on the sorting deviation reference value, which is specifically described in detail through the following steps.

[0137] Refer to Figure 4 , the generation method of the sorting deviation reference value includes the following steps:

[0138] S3361: Determine whether the sorting deviation value is greater than a preset sorting deviation reference value. If it is, execute S3362; if not, execute S3367.

[0139] Among them, the sorting deviation reference value refers to the maximum deviation value that can be tolerated when directly sorting by storage capacity, and the sorting deviation reference value is obtained through pre-input. By judging whether the sorting deviation value is greater than the preset sorting deviation reference value, it is determined whether the storage sorting value can be directly used for reference.

[0140] S3362: Calculate the difference between the sorting deviation value and the sorting deviation reference value and use it as the sorting anomaly deviation value.

[0141] Among them, the sorting anomaly deviation value refers to the deviation value corresponding to the existence of an abnormal deviation in sorting. When the sorting deviation value is greater than the preset sorting deviation reference value, it indicates that the storage sorting value cannot be directly used for reference at this time. Therefore, the difference between the sorting deviation value and the sorting deviation reference value is calculated and used as the sorting anomaly deviation value for subsequent use.

[0142] S3363: Calculate the quotient of the source storage value and the type retrieval speed value and use it as the retrieval time value.

[0143] Among them, the retrieval time value refers to the time value required for retrieving data. By calculating the quotient of the source storage value and the type retrieval speed value and using it as the retrieval time value, it is convenient for subsequent use.

[0144] S3364: Sort in ascending order based on the retrieval time value and use the sorting result as the retrieval time sorting value.

[0145] Among them, the retrieval time sorting value refers to the result value corresponding to the sorting based on the retrieval time. By sorting the retrieval time value in ascending order and using the sorting result as the retrieval time sorting value, it is convenient for subsequent use.

[0146] S3365: Generate an initial time sorting reference value based on the retrieval time sorting value.

[0147] Among them, the initial time sorting reference value refers to the initial reference value corresponding to the reference based on the sorting result of the retrieval time. Different retrieval time sorting values correspond to different initial time sorting reference values. By inputting the initial time sorting reference value into the preset initial time sorting reference database, the initial time sorting reference value is obtained for subsequent use. The initial time sorting reference database stores different retrieval time sorting values and the corresponding initial time sorting reference values, and the initial time sorting reference database is obtained through pre-input.

[0148] S3366: Adjust the initial time sorting reference value based on the sorting anomaly deviation value to form an adjusted time sorting reference value, and use the adjusted time sorting reference value as the sorting deviation reference value.

[0149] Among them, the time sorting adjustment reference value refers to the reference value corresponding to the adjustment of the sorting reference of the retrieval time for the abnormal deviation of sorting. By inputting the sorting abnormal deviation value into a preset sorting abnormal deviation database to obtain the sorting abnormal deviation influence value, and calculating the product value between the sorting abnormal deviation influence value and the initial time sorting reference value to obtain the time sorting adjustment reference value, and using the time sorting adjustment reference value as the sorting deviation reference value, thereby improving the accuracy of the obtained sorting deviation reference value. The sorting abnormal deviation database pre-stores different sorting abnormal deviation values and the corresponding sorting abnormal deviation influence values, and the sorting abnormal deviation database is obtained through pre-input.

[0150] S3367: Generate a storage sorting reference value based on the stored sorting value, and use the storage sorting reference value as the sorting deviation reference value.

[0151] Among them, the storage sorting reference value refers to the reference value corresponding to the reference based on the sorting result of the storage volume. When the sorting deviation value is not greater than the preset sorting deviation reference value, it indicates that the stored sorting value can be directly used for reference at this time. Therefore, by inputting the stored sorting value into a preset storage sorting reference database to obtain the storage sorting reference value, and using the storage sorting reference value as the sorting deviation reference value, thereby improving the accuracy of the obtained sorting deviation reference value. The storage sorting reference database pre-stores different stored sorting values and the corresponding storage sorting reference values, and the storage sorting reference database is obtained through pre-input.

[0152] In step S34, in order to further ensure the rationality of the content coincidence value, it is necessary to perform a more detailed separate analysis and calculation on the content coincidence value, which is specifically described in detail through the following steps.

[0153] Refer to Figure 5 , the method for determining the content coincidence value includes the following steps:

[0154] S341: Retrieve the content keyword information and content repeated word information based on the source content information.

[0155] Among them, the content keyword information refers to the information corresponding to the keywords in the data content stored in the source database, and the content repeated word information refers to the information corresponding to the words that appear repeatedly in the data content stored in the source database. The source content information includes the content keyword information and the content repeated word information. Retrieving the content keyword information and the content repeated word information through the source content information facilitates subsequent use.

[0156] S342: Determine the keyword coincidence value based on the content keyword information.

[0157] Among them, the keyword coincidence value refers to the coincidence degree value corresponding to the coincidence of keywords in different source databases. By analyzing the keyword information of the content, the keyword coincidence value is determined to facilitate subsequent use. The specific steps for determining the keyword coincidence value refer to S3421 to S3427.

[0158] S343: Determine the repetition word coincidence value based on the content repetition word information.

[0159] Among them, the repetition word coincidence value refers to the coincidence degree value corresponding to the coincidence of repetition words in different source databases. By analyzing the content repetition word information, the repetition word coincidence value is determined to facilitate subsequent use. The specific steps for determining the repetition word coincidence value refer to S3431 to S3436.

[0160] S344: Determine the comprehensive coincidence value based on the keyword coincidence value and the repetition word coincidence value.

[0161] Among them, the comprehensive coincidence value refers to the comprehensive coincidence degree value corresponding to the coincidence of keywords and repetition words. By calculating the sum value between the keyword coincidence value and the repetition word coincidence value, and taking the sum value as the comprehensive coincidence value to facilitate subsequent use.

[0162] S345: Determine whether there is a coincidence between the content keyword information and the content repetition word information. If yes, execute S346; if no, execute S348.

[0163] Among them, by judging whether there is a coincidence between the content keyword information and the content repetition word information, it is judged whether the comprehensive coincidence value needs to be adjusted.

[0164] S346: Determine the content overlapping word information based on the content keyword information and the content repetition word information.

[0165] Among them, the content overlapping word information refers to the word information when the keywords and repetition words in the source database overlap. When there is a coincidence between the content keyword information and the content repetition word information, it means that the comprehensive coincidence value needs to be adjusted at this time. Therefore, the content keyword information and the content repetition word information are matched, and the words that match are used as the content overlapping word information to facilitate subsequent use.

[0166] S347: Determine the adjusted coincidence value based on the content overlapping word information and the comprehensive coincidence value, and use the adjusted coincidence value as the content coincidence value.

[0167] Among them, the adjusted coincidence value refers to the coincidence value corresponding to the comprehensive coincidence value after adjustment. By analyzing the content overlapping word information and the comprehensive coincidence value, the adjusted coincidence value is determined, and the adjusted coincidence value is used as the content coincidence value, thereby improving the accuracy of the obtained content coincidence value. The specific steps for determining the adjusted coincidence value refer to S3471 to S3477.

[0168] S348: Directly use the comprehensive coincidence value as the content coincidence value.

[0169] Among them, when there is no coincidence between the content keyword information and the content repeated word information, it means that the comprehensive coincidence value does not need to be adjusted at this time. Therefore, the comprehensive coincidence value is directly used as the content coincidence value, thereby improving the accuracy of the obtained content coincidence value.

[0170] In step S342, in order to further ensure the rationality of the keyword coincidence value, it is necessary to perform a further separate analysis and calculation on the keyword coincidence value, which is specifically described in detail through the following steps.

[0171] Refer to Figure 6 , the method for determining the keyword coincidence value includes the following steps:

[0172] S3421: Determine the keyword meaning information and the content synonym information based on the content keyword information.

[0173] Among them, the keyword meaning information refers to the meaning information represented by the keyword, and the content synonym information refers to the word information with the same meaning as the keyword. Different content keyword information corresponds to different keyword meaning information and content synonym information. By inputting the content keyword information into the preset keyword database, the keyword meaning information and the content synonym information can be obtained for convenient subsequent use. The keyword database pre-stores different content keyword information and the corresponding keyword meaning information and content synonym information, and the keyword database is obtained through pre-input.

[0174] S3422: Determine the meaning-related word information based on the keyword meaning information.

[0175] Among them, the meaning-related word information refers to the word information that is related to the meaning represented by the keyword. Different keyword meaning information corresponds to different meaning-related word information. By inputting the keyword meaning information into the preset meaning database, the meaning-related word information can be obtained for convenient subsequent use. The meaning database pre-stores different keyword meaning information and the corresponding meaning-related word information, and the meaning database is obtained through pre-input.

[0176] S3423: Combine the content synonym information and the meaning-related word information to form the key comprehensive word information.

[0177] Among them, the key comprehensive word information refers to the comprehensive information corresponding to the words that are associated with or have the same meaning as the keyword, and the words that are associated with or have the same meaning as the keyword are called comprehensive words. By combining the content synonym information and the meaning-related word information, and using the corresponding words after combination as the key comprehensive word information, it is convenient for subsequent use.

[0178] S3424: Perform a coincidence query based on the key comprehensive word information to obtain the comprehensive coincidence word information and the corresponding comprehensive coincidence count value.

[0179] Among them, the comprehensive coincidence word information refers to the word information corresponding to the coincidence with the comprehensive word, and the comprehensive coincidence count value refers to the count value corresponding to the coincidence with the comprehensive word. By performing a coincidence query on the key comprehensive word information, the comprehensive words with coincidences are used as the comprehensive coincidence word information, and the coincidence count corresponding to the comprehensive coincidence word information is counted, and the count result is used as the comprehensive coincidence count value, which is convenient for subsequent use.

[0180] S3425: Determine the coincidence reference benchmark value based on the comprehensive coincidence word information.

[0181] Among them, the coincidence reference benchmark value refers to the benchmark value corresponding to the reference of the comprehensive words with coincidences, and different comprehensive coincidence word information corresponds to different coincidence reference benchmark values. By inputting the comprehensive coincidence word information into the preset coincidence reference benchmark database to obtain the coincidence reference benchmark value, it is convenient for subsequent use. The coincidence reference benchmark database stores different comprehensive coincidence word information and the corresponding coincidence reference benchmark values, and the coincidence reference benchmark database is obtained through pre-input.

[0182] S3426: Calculate the product value between the coincidence reference benchmark value and the comprehensive coincidence count value and use it as the single coincidence word coincidence value.

[0183] Among them, the single coincidence word coincidence value refers to the reference value corresponding to the coincidence of a single coincidence word. By calculating the product value between the coincidence reference benchmark value and the comprehensive coincidence count value and using it as the single coincidence word coincidence value, it is convenient for subsequent use.

[0184] S3427: Calculate the sum value of all the single coincidence word coincidence values and use it as the keyword coincidence value.

[0185] Among them, by calculating the sum value of all the single coincidence word coincidence values and using it as the keyword coincidence value, the accuracy of the obtained keyword coincidence value is improved.

[0186] In step S343, in order to further ensure the rationality of the repeated word coincidence value, it is necessary to perform a more detailed separate analysis and calculation on the repeated word coincidence value, which is specifically described in detail through the following steps.

[0187] Reference Figure 7 , the method for determining the repetition coincidence value includes the following steps:

[0188] S3431: Retrieve single repetition word information and the corresponding repetition count value based on the content repetition word information.

[0189] Among them, the single repetition word information refers to the information of a single repetition word, and the repetition count value refers to the number of times when a single repetition word exists. The single repetition word information is retrieved through the content repetition word information, and the number of times corresponding to the single repetition word information is counted, and the counting result is used as the repetition count value for subsequent use.

[0190] S3432: Sort from large to small based on the repetition count value, and use the sorting result as the repetition count sorting value.

[0191] Among them, the repetition count sorting value refers to the result value obtained by sorting according to the number of repetitions of a single repetition word. By sorting the repetition count value from large to small and using the sorting result as the repetition count sorting value, it is convenient for subsequent use.

[0192] S3433: Determine the repetition count sorting influence value based on the repetition count sorting value.

[0193] Among them, the repetition count sorting influence value refers to the influence degree value generated by the repetition count sorting value on the selection. Different repetition count sorting values correspond to different repetition count sorting influence values. By inputting the repetition count sorting value into the preset repetition count sorting influence database to obtain the repetition count sorting influence value, it is convenient for subsequent use. The repetition count sorting influence database pre-stores different repetition count sorting values and the corresponding repetition count sorting influence values, and the repetition count sorting influence database is obtained through pre-input.

[0194] S3434: Conduct a coincidence query based on the single repetition word information to obtain the repeated coincidence word information and the corresponding repeated coincidence count value.

[0195] Among them, the repeated coincidence word information refers to the word information corresponding to the coincidence with a single repetition word, and this word is used as the repeated coincidence word. The repeated coincidence count value refers to the number of times corresponding to the coincidence of the repeated coincidence word. By conducting a coincidence query on the single repetition word information, the word corresponding to the coincidence with a single repetition word is used as the repeated coincidence word information, and the number of times corresponding to the repeated coincidence word information is counted, and the counting result is used as the repeated coincidence count value for subsequent use.

[0196] S3435: Determine the repeated coincidence value based on the repeated coincidence word information and the repeated coincidence count value.

[0197] Among them, the overlapping value of repeated overlapping words refers to the overlapping degree value corresponding to the existence of repeated overlapping words. By querying the preset repeated overlapping reference database with the repeated overlapping word information, the repeated overlapping reference value is obtained. Then, the product value between the repeated overlapping times value and the repeated overlapping reference value is calculated and used as the overlapping value of repeated overlapping words for subsequent use. The repeated overlapping reference database pre-stores different repeated overlapping word information and the corresponding repeated overlapping reference values, and the repeated overlapping reference database is obtained through pre-input.

[0198] S3436: Adjust the overlapping value of repeated overlapping words based on the influence value of sorting by the number of repetitions to obtain the overlapping value of repeated words.

[0199] Among them, by calculating the product value between the influence value of sorting by the number of repetitions and the overlapping value of repeated overlapping words, and updating and adjusting the overlapping value of repeated overlapping words with the product value, and then using the updated overlapping value of repeated overlapping words as the overlapping value of repeated words, the accuracy of the obtained overlapping value of repeated words is improved.

[0200] In step S347, in order to further ensure the rationality of the adjusted overlapping value, it is necessary to perform a further separate analysis and calculation on the adjusted overlapping value, which is specifically described in detail through the following steps.

[0201] Refer to Figure 8 , the determination method of the adjusted overlapping value includes the following steps:

[0202] S3471: Retrieve the keyword count value based on the content keyword information.

[0203] Among them, the keyword count value refers to the count value corresponding to the keyword. By retrieving the keywords in the content keyword information and counting the number, and using the counting result as the keyword count value for subsequent use.

[0204] S3472: Retrieve the repeated word count value based on the content repeated word information.

[0205] Among them, the repeated word count value refers to the count value corresponding to the repeated word. By retrieving the repeated words in the content repeated word information and counting the number, and using the counting result as the repeated word count value for subsequent use.

[0206] S3473: Calculate the sum value between the keyword count value and the repeated word count value and use it as the reference word count value.

[0207] Among them, the reference word count value refers to the count value corresponding to the word that needs to be referenced. By calculating the sum value between the keyword count value and the repeated word count value and using it as the reference word count value for subsequent use.

[0208] S3474: Retrieve the number value of overlapping words based on the content overlapping word information.

[0209] Among them, the number value of overlapping words refers to the corresponding number value of overlapping words. By retrieving the overlapping words in the content overlapping word information and counting the number, and taking the counting result as the number value of overlapping words, it is convenient for subsequent use.

[0210] S3475: Calculate the quotient between the number value of overlapping words and the number value of reference words and use it as the overlapping word occupancy ratio.

[0211] Among them, the overlapping word occupancy ratio refers to the proportional value of the number of overlapping words to the number of reference words. By calculating the quotient between the number value of overlapping words and the number value of reference words and using it as the overlapping word occupancy ratio, it is convenient for subsequent use.

[0212] S3476: Determine the overlapping word occupancy ratio influence value based on the overlapping word occupancy ratio.

[0213] Among them, the overlapping word occupancy ratio influence value refers to the influence value of the overlapping word occupancy ratio on the comprehensive coincidence value. Different overlapping word occupancy ratios correspond to different overlapping word occupancy ratio influence values. By inputting different overlapping word occupancy ratios into the preset overlapping word occupancy ratio influence database to obtain the overlapping word occupancy ratio influence value, it is convenient for subsequent use. The overlapping word occupancy ratio influence database pre-stores different overlapping word occupancy ratios and the corresponding overlapping word occupancy ratio influence values, and the overlapping word occupancy ratio influence database is obtained through pre-input.

[0214] S3477: Adjust the comprehensive coincidence value based on the overlapping word occupancy ratio influence value to form an adjusted coincidence value.

[0215] Among them, by calculating the product value between the overlapping word occupancy ratio influence value and the comprehensive coincidence value, and updating and adjusting the comprehensive coincidence value with the product value, and taking the updated comprehensive coincidence value as the adjusted coincidence value, the accuracy of the obtained adjusted coincidence value is improved.

[0216] Based on the same inventive concept, an embodiment of the present invention provides a health knowledge mining system based on a dynamic feedback mechanism, including:

[0217] An acquisition module, configured to acquire health knowledge mining requirements and the current time point;

[0218] A memory, configured to store a health knowledge mining method as described above;

[0219] A processor, configured to load and execute the program in the memory.

[0220] Based on the same inventive concept, an embodiment of the present invention provides a terminal, including a memory and a processor. A computer program is stored on the memory and can be loaded and executed by the processor to perform a health knowledge mining method based on a dynamic feedback mechanism as described above.

[0221] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the above division of each functional module is used as an example. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. For the specific working processes of the above-described system, device, and unit, reference can be made to the corresponding processes in the foregoing method embodiments, which will not be elaborated herein.

[0222] The above are only the preferred embodiments of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the concept of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art, several improvements and refinements made without departing from the principle of the present invention should also be regarded as within the protection scope of the present invention.

Claims

1. A health knowledge mining method based on a dynamic feedback mechanism, characterized in that, Including: S1: Collect the health knowledge mining requirements input by the user and the current time point; S2: Determine the requirement type based on the health knowledge mining requirements; S3: Select a mining database based on the requirement type; S4: Retrieve the database update time point and the update reference time value based on the mining database; S5: Calculate the update interval time value based on the current time point and the database update time point; S6: Update and adjust the mining database based on the update interval time value and the update reference time value to obtain an adjusted database; S7: Mine knowledge from the adjusted database based on the health knowledge mining requirements to generate a health knowledge mining result, and output the health knowledge mining result; The step S3 includes the following steps: S31: Determine single-type source information based on the requirement type; S32: Retrieve source storage information and source content information based on the single-type source information; S33: Determine a source retrieval reference value based on the source storage information; S34: Determine a content overlap value based on the source content information; S35: Determine an overlap reference value based on the content overlap value; S36: Select the single-type source information based on the source retrieval reference value and the overlap reference value to form selected source information; S37: Determine a selected database based on the selected source information, and use the selected database as the mining database; The step S33 includes the following steps: S331: Retrieve a source storage value and storage type information based on the source storage information; S332: Sort the source storage value from largest to smallest, and use the sorting result as a storage sorting value; S333: Determine a type retrieval speed value based on the storage type information; S334: Sort the type retrieval speed value from largest to smallest, and use the sorting result as a retrieval speed sorting value; S335: Calculate the difference between the storage sorting value and the retrieval speed sorting value and use it as a sorting deviation value; S336: Generate a sorting deviation reference value based on the sorting deviation value and the storage sorting value, and use the sorting deviation reference value as the source retrieval reference value.

2. The health knowledge mining method based on a dynamic feedback mechanism according to claim 1, wherein The step S336 includes the following steps: S3361: Determine whether the sorting deviation value is greater than a preset sorting deviation reference value; S3362: If so, calculate the difference between the sorting deviation value and the sorting deviation reference value and use it as a sorting abnormal deviation value; S3363: Calculate the quotient of the source storage value and the type retrieval speed value and use it as a retrieval time value; S3364: Sort the retrieval time value from smallest to largest, and use the sorting result as a retrieval time sorting value; S3365: Generate an initial retrieval time sorting reference value based on the retrieval time sorting value; S3366: Adjust the initial retrieval time sorting reference value based on the sorting abnormal deviation value to form an adjusted retrieval time sorting reference value, and use the adjusted retrieval time sorting reference value as the sorting deviation reference value; S3367: If the answer is no, generate a storage sorting reference value based on the stored sorting value, and use the storage sorting reference value as the sorting deviation reference value.

3. A method for mining health knowledge based on a dynamic feedback mechanism according to claim 1, characterized in that The step S34 includes the following steps: S341: Retrieve content keyword information and content repeated word information based on the source content information; S342: Determine the keyword coincidence value based on the content keyword information; S343: Determine the repeated word coincidence value based on the content repeated word information; S344: Determine the comprehensive coincidence value based on the keyword coincidence value and the repeated word coincidence value; S345: Determine whether there is a coincidence between the content keyword information and the content repeated word information; S346: If the answer is yes, determine the content overlapping word information based on the content keyword information and the content repeated word information; S347: Determine the adjusted coincidence value based on the content overlapping word information and the comprehensive coincidence value, and use the adjusted coincidence value as the content coincidence value; S348: If the answer is no, directly use the comprehensive coincidence value as the content coincidence value.

4. A health knowledge mining method based on a dynamic feedback mechanism according to claim 3, characterized in that, The step S342 includes the following steps: S3421: Determine the keyword meaning information and content synonym information based on the content keyword information; S3422: Determine the meaning related word information based on the keyword meaning information; S3423: Combine the content synonym information and the meaning related word information to form key comprehensive word information; S3424: Conduct a coincidence query based on the key comprehensive word information to obtain the comprehensive coincidence word information and the corresponding comprehensive coincidence times value; S3425: Determine the coincidence reference benchmark value based on the comprehensive coincidence word information; S3426: Calculate the product value between the coincidence reference benchmark value and the comprehensive coincidence times value and use it as the single coincidence word coincidence value; S3427: Calculate the sum value of all the single coincidence word coincidence values and use it as the keyword coincidence value.

5. A health knowledge mining method based on a dynamic feedback mechanism according to claim 3, characterized in that The step S343 includes the following steps: S3431: Retrieve the single repeated word information and the corresponding repeated times value based on the content repeated word information; S3432: Sort the repeated times values from largest to smallest, and use the sorting result as the repeated times sorting value; S3433: Determine the repeated times sorting influence value based on the repeated times sorting value; S3434: Conduct a coincidence query based on the single repeated word information to obtain the repeated coincidence word information and the corresponding repeated coincidence times value; S3435: Determine the repeated coincidence word coincidence value based on the repeated coincidence word information and the repeated coincidence times value; S3436: Adjust the repeated coincidence word coincidence value based on the repeated times sorting influence value to use it as the repeated word coincidence value.

6. A method for mining health knowledge based on a dynamic feedback mechanism according to claim 3, characterized in that, The step S347 includes the following steps: S3471: Retrieve the keyword number value based on the content keyword information; S3472: Retrieve the repeated word number value based on the content repeated word information; S3473: Calculate the sum value between the keyword number value and the repeated word number value and use it as the reference word number value; S3474: Retrieve the overlapping word number value based on the content overlapping word information; S3475: Calculate the quotient of the number value of the overlapping words and the number value of the reference words as the overlapping word occupancy ratio; S3476: Determine the overlapping word occupancy ratio influence value based on the overlapping word occupancy ratio; S3477: Adjust the comprehensive coincidence value based on the overlapping word occupancy ratio influence value to form the adjusted coincidence value.

7. A health knowledge mining system based on a dynamic feedback mechanism, characterized in that, Comprising: An acquisition module, configured to acquire the health knowledge mining requirement and the current time point; A memory, configured to store a health knowledge mining method based on a dynamic feedback mechanism according to any one of claims 1 to 6; A processor, configured to load and execute the program in the memory.

8. A terminal, characterized in that, Comprising a memory and a processor, wherein a computer program is stored on the memory and can be loaded and executed by the processor to implement a health knowledge mining method based on a dynamic feedback mechanism according to any one of claims 1 to 6.

Images