Diet health management method and equipment based on big data analysis and medium

Abstract

The invention discloses a diet health management method and equipment based on big data analysis and a medium, and relates to the technical field of big data health management, and the method comprises the steps: searching similar user group data from a historical data warehouse based on diet behavior characteristics and diet health management indexes, and constructing a diet behavior reference portrait, performing difference calculation on the dietary behavior characteristics and the dietary behavior reference portrait to output a dietary behavior deviation measure, and mapping the dietary behavior deviation measure into a corresponding management event type; according to the management event type, candidate diet health intervention paths are retrieved from an intervention strategy library, constraint screening and performability verification are performed on the candidate diet health intervention paths one by one, and a target diet health management intervention path is determined; and collecting the execution effect of the target diet health management intervention path in real time, recording the diet behavior change condition of the user, and forming health feedback data. According to the invention, the accuracy, the performability and the continuous optimization capability of diet health management are improved.

Description

Technical Field

[0001] This invention relates to the field of big data health management technology, and in particular to a dietary health management method, device and medium based on big data analysis. Background Technology

[0002] With the rapid development of information technology, mobile internet, and wearable devices, dietary health management technologies are gradually evolving from traditional experience-based guidance to data-driven approaches. Current dietary health management technologies typically rely on user-reported dietary records, physiological indicator collection, and basic nutritional models to assess and provide recommendations regarding an individual's dietary structure, energy intake, and health risks. In recent years, with the widespread application of big data analytics in the healthcare field, some solutions have begun to incorporate multi-source data fusion, historical sample comparison, and statistical analysis methods to conduct a more systematic analysis of user dietary behavior. These technologies have, to some extent, improved the scientific rigor and quantifiability of dietary health management, promoting the development of personalized health management models.

[0003] From the overall application of existing technologies, there is still room for improvement in the dynamic analysis and continuous management of dietary behavior in relevant dietary health management programs. Existing programs mostly focus on static data or analysis results within a single time period, failing to fully explore the patterns of dietary behavior changes over time, particularly in time window segmentation, behavioral feature extraction, and multi-dimensional comparative analysis, where a systematic data processing mechanism has not yet been established. Furthermore, some technologies rely heavily on fixed thresholds or general models when assessing user dietary behavior, lacking methods for comparing similar user groups based on big data analysis. This makes it difficult to objectively reflect the relative deviation of an individual's dietary behavior within a group, thus affecting the targeting and feasibility of subsequent intervention strategies. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a dietary health management method based on big data analysis to solve the problem of difficulty in performing dynamic deviation analysis of users' dietary behavior based on big data.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a dietary health management method based on big data analysis, which includes collecting basic health data of users and forming a dietary health dataset through aggregation and processing; The diet and health dataset is divided according to the preset time window rules, and the diet behavior management data in each time window is statistically analyzed to extract diet behavior characteristics. Through the preset diet behavior management rules, multi-dimensional comparative analysis is carried out to generate diet and health management indicators. Based on dietary behavior characteristics and dietary health management indicators, similar user group data are retrieved from historical data warehouse to construct a dietary behavior benchmark profile. The difference between dietary behavior characteristics and dietary behavior benchmark profile is calculated to output a dietary behavior deviation measure, and the dietary behavior deviation measure is mapped to the corresponding management event type. Candidate dietary health intervention paths are retrieved from the intervention strategy library based on the type of management event. Constraint screening and feasibility verification are performed on each candidate dietary health intervention path to determine the target dietary health management intervention path. The system collects real-time data on the effectiveness of the target dietary health management intervention pathway and records changes in users' dietary behavior to generate health feedback data.

[0007] As a preferred embodiment of the dietary health management method based on big data analysis described in this invention, the specific steps for forming a dietary health dataset through aggregation processing are as follows: The system deduplicates user basic health data according to user identifier and time information, and identifies and corrects missing and abnormal data to obtain valid health data. Valid health data from different sources are standardized by unifying fields, converting formats, and standardizing units of measurement to form standardized health data. Based on user identification and time dimension, the standardized health data are correlated and fused to generate a dietary health dataset.

[0008] As a preferred embodiment of the dietary health management method based on big data analysis described in this invention, the steps of dividing the dietary health dataset according to a preset time window rule and statistically analyzing the dietary behavior management data within each time window to extract dietary behavior features are as follows. The diet and health dataset is divided into time periods according to the preset time window rules, and the diet and health data belonging to the same time window are aggregated to form diet behavior management data. Using dietary behavior management data as the statistical object, dietary intake data and eating behavior data within a time window are summarized to generate dietary status statistics. Based on the dietary status statistics, dietary behavior is structured and summarized to form dietary behavior characteristics.

[0009] As a preferred embodiment of the dietary health management method based on big data analysis described in this invention, the generation of dietary health management indicators refers to inputting dietary behavior characteristics into preset dietary behavior management rules, performing comparative judgments item by item according to different dietary behavior dimensions, obtaining multi-dimensional comparative analysis results for each time window, and converting each multi-dimensional comparative analysis result into indicator details according to a preset coding table, using the user identifier and the start and end time of the window as the primary keys, and writing the indicator details as dietary health management indicators.

[0010] As a preferred embodiment of the dietary health management method based on big data analysis described in this invention, the specific steps for constructing a baseline profile of dietary behavior by retrieving similar user group data from a historical data warehouse based on dietary behavior characteristics and dietary health management indicators are as follows: For dietary behavior characteristics and dietary health management indicators, the corresponding field values ​​are read one by one according to the preset field mapping table, and then structured and encapsulated to form search keys; Using the search key as a condition, perform similar user searches in the historical data warehouse to obtain a set of candidate users, and calculate the similarity score between the dietary behavior characteristics and dietary health management indicators of each candidate user in the set and the corresponding fields of the current user. The candidate users are sorted from high to low according to their similarity scores. The top N candidate users are selected as the similar user group. At the same time, the dietary behavior characteristics and dietary health management indicators of the similar user group under the same time window are extracted to form similar user group data. The dietary behavior characteristics of similar user groups are statistically aggregated to obtain the baseline level of dietary behavior, and the dietary health management indicators are statistically aggregated to obtain the baseline status of dietary health. The baseline levels of dietary behavior and the baseline status of dietary health are uniformly packaged into a structured record to form a baseline profile of dietary behavior.

[0011] As a preferred embodiment of the dietary health management method based on big data analysis described in this invention, the specific steps of calculating the difference between dietary behavior characteristics and dietary behavior benchmark profiles to output a dietary behavior deviation metric, and mapping the dietary behavior deviation metric to corresponding management event types, are as follows. Calculate the field difference between the current user's field values ​​and the field values ​​of the dietary behavior baseline profile, determine the direction of the difference according to the field type, and summarize and aggregate the field difference and direction of the difference to form a dietary behavior deviation metric. Using dietary behavior deviation metrics as input, a preset management event triggering rule table is loaded and deviation features are used as triggering conditions. By traversing the preset management event triggering rule table, the deviation fields and deviation levels in the dietary behavior deviation metrics are matched with the triggering conditions. If the triggering conditions are met, the corresponding management event type will be output. If multiple triggering conditions are met at the same time, the management event type with the highest priority will be selected as the target output according to the priority in the preset management event triggering rule table.

[0012] As a preferred embodiment of the dietary health management method based on big data analysis described in this invention, the steps of retrieving candidate dietary health intervention paths from the intervention strategy library according to the type of management event, performing constraint screening and feasibility verification on each candidate dietary health intervention path, and determining the target dietary health management intervention path are as follows: Using the type of management event as the search entry, perform a matching search in the intervention strategy library and output candidate dietary health intervention paths; Read the constraints corresponding to each candidate dietary health intervention path one by one, and compare the constraints with the current user's dietary health dataset, dietary behavior characteristics and dietary health management indicators item by item. Eliminate candidate dietary health intervention paths that do not meet the constraints to form the filtered intervention path. The feasibility of each candidate dietary health intervention path in the screened intervention pathways was checked one by one, and the unfeasible candidate dietary health intervention paths were eliminated to form an executable intervention path. The executable intervention paths are sorted according to the preset priority rules, and the highest priority executable intervention path is selected as the target dietary health management intervention path.

[0013] As a preferred embodiment of the dietary health management method based on big data analysis described in this invention, the formation of health feedback data refers to initiating execution monitoring with the target dietary health management intervention path as the execution object, collecting the execution effect data of the target dietary health management intervention path in real time, and simultaneously recording the changes in the user's dietary behavior during the intervention execution period, and integrating the execution effect data with the changes in dietary behavior.

[0014] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the dietary health management method based on big data analysis as described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the dietary health management method based on big data analysis as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By standardizing and aggregating multi-source user health data, and combining time window segmentation and statistical analysis to extract dietary behavior characteristics, and based on the retrieval of similar user groups and the construction of benchmark profiles from historical data warehouses, this invention enables the use of big data analysis to conduct group benchmark comparisons and deviation measurement expressions of individual dietary behaviors, thereby mapping the deviation results into management event types that can drive decision-making. At the same time, through event-driven intervention path constraint screening, feasibility verification, and execution effect feedback mechanisms, a closed-loop dietary health management process covering assessment, decision-making, and feedback is constructed, improving the accuracy, feasibility, and continuous optimization capabilities of dietary health management. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a dietary health management method based on big data analysis.

[0019] Figure 2 A flowchart for the aggregation and processing of dietary health datasets.

[0020] Figure 3 A flowchart for constructing a baseline profile of dietary behavior.

[0021] Figure 4 A flowchart for generating health feedback data. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4 This is one embodiment of the present invention, which provides a dietary health management method based on big data analysis, including the following steps: S1. Collect basic health data from users and aggregate and process it to form a dietary health dataset.

[0026] It should be noted that the user's basic health data includes demographic information, physiological indicators, past health status data, and dietary records.

[0027] S1.1 Deduplicating user basic health data according to user identifier and time information, and identifying and correcting missing and abnormal data to obtain valid health data.

[0028] Furthermore, a unified data alignment rule is established for basic user health data based on user identification and time information. By forming a comparable data positioning basis through a unified time granularity, duplicate health data records of the same user at the same time dimension are identified on this comparable data positioning basis. Duplicate health data records are merged according to the data source credibility, field completeness, and record time priority rules. Subsequently, missing data identification is performed on each field of the deduplicated health data to identify the missing type and correct the missing fields according to the preset completion rules. After the missing data correction is completed, anomaly detection is performed on the health data to identify abnormal value ranges, inconsistent units, and field logic anomalies. According to the anomaly type, the corresponding data correction processing is completed, and valid health data is output.

[0029] It should be noted that the data completion rules are pre-configured missing data processing strategies based on data field type, business risk level, and data source credibility. They are used to complete data completion and retain completion marks when missing data occurs, using historical valid values, adjacent time window interpolation, or rule inference methods in a predetermined priority order.

[0030] S1.2. Standardize the fields, format, and units of measurement of valid health data from different sources to form standardized health data. Then, based on user identification and time dimension, link and merge the standardized health data to generate a dietary health dataset.

[0031] Furthermore, valid health data from different sources undergoes unified field processing, standardizing data from each source into uniform field names and semantics; standardized conversions are performed on the time format, data type, and enumeration values ​​of the unified field health data to form standardized health data; unit identification, verification, and conversion are performed on indicator fields involving measurement units in the standardized health data to form data results with consistent units; finally, using user identifier and time dimension as the association primary key, the data results with consistent units are aligned and merged to generate a dietary health dataset.

[0032] S2. Divide the diet and health dataset according to the preset time window rules, perform statistical analysis on the diet behavior management data in each time window, extract diet behavior characteristics, and perform multi-dimensional comparative analysis through preset diet behavior management rules to generate diet and health management indicators.

[0033] S2.1 Divide the diet and health dataset into time periods according to the preset time window rules, and aggregate the diet and health data belonging to the same time window to form diet behavior management data.

[0034] Furthermore, each record in the diet and health dataset is divided according to its timestamp based on a preset time window rule (such as a fixed period or sliding window, such as daily, weekly, or monthly). All diet and health data records whose timestamps fall within the same time window are aggregated into the specific time interval corresponding to the preset time window, and structured diet behavior management data is formed with the user identifier and the time window as the joint primary key.

[0035] It should be noted that the time window rule is set by configuring three parameters: start time, window length (e.g., 1 day, 7 days), and sliding step (e.g., 1 day).

[0036] S2.2 Using dietary behavior management data as the statistical object, summarize the dietary intake data and eating behavior data within the time window to generate dietary status statistical results, and summarize the dietary behavior in a structured manner based on the dietary status statistical results to form dietary behavior characteristics.

[0037] Furthermore, dietary intake data and eating behavior data are extracted for each time window. Dietary intake data is summed or averaged according to dimensions such as nutrient composition, food category, and intake amount to generate intake-related statistical indicators. Eating behavior data is statistically analyzed or pattern identified according to dimensions such as meal frequency, meal time distribution, and meal regularity to generate behavioral statistical indicators. Intake-related statistical indicators and behavioral statistical indicators are combined to form dietary status statistical results. Based on dietary status statistical results, eating behavior is structured and summarized to form dietary behavior characteristics.

[0038] S2.3 Input dietary behavior characteristics into the preset dietary behavior management rules, perform comparative judgments item by item according to different dietary behavior dimensions, obtain multi-dimensional comparative analysis results for each time window, and convert each multi-dimensional comparative analysis result into indicator details according to the preset coding table. Using user identifier and window start and end time as the primary key, write the indicator details as dietary health management indicators.

[0039] Furthermore, dietary behavior characteristics are input into preset dietary behavior management rules. For each preset time window, a comparison is performed item by item against the corresponding pattern in the preset dietary behavior management rules, according to different dietary behavior dimensions such as nutritional balance, meal regularity, food diversity, and reasonable energy intake. Each characteristic value in the dietary behavior characteristics is numerically compared or logically matched with the judgment conditions of the same dimension in the preset dietary behavior management rules. For example, it is judged whether the average daily vegetable intake is lower than the recommended lower limit or whether the number of meals per day is less than 3. Based on the comparison results of dietary behavior characteristics and preset dietary behavior management rules in each dietary behavior dimension, it is determined whether each dietary behavior dimension meets the corresponding judgment conditions and marked as met or not met, forming a multi-dimensional comparative analysis result. Each multi-dimensional comparative analysis result is converted into a structured indicator detail item according to the coding mapping relationship in the preset coding table. For example, "insufficient vegetable intake" is mapped to the code "NUT-001". Using the user identifier and the start and end time of the preset time window as the joint primary key, all indicator details items are written as dietary health management indicators.

[0040] It should be noted that the dietary behavior management rules are based on the dietary behavior characteristics and dietary health management indicators of similar user groups in the same time window in the historical data warehouse. The baseline distribution of each dietary behavior dimension is obtained through statistical aggregation, and the quantiles or cluster centers of the baseline distribution are used as the judgment conditions. The judgment criteria are the rules and standards used to determine whether dietary behavior characteristics meet the standards. They are determined by statistical analysis of the dietary behavior characteristics of similar user groups in the historical data warehouse under the corresponding time window. The coding table is a correspondence table used to map the comparative analysis results of various dietary behaviors to standardized indicator codes. It is uniformly defined and configured according to dietary behavior dimensions, deviation types and management needs to ensure the structured expression and consistency of dietary health management indicators.

[0041] S3. Based on dietary behavior characteristics and dietary health management indicators, retrieve similar user group data from the historical data warehouse to construct a dietary behavior benchmark profile, calculate the difference between dietary behavior characteristics and dietary behavior benchmark profile, output dietary behavior deviation measure, and map the dietary behavior deviation measure to the corresponding management event type.

[0042] S3.1 For dietary behavior characteristics and dietary health management indicators, read the corresponding field values ​​one by one according to the preset field mapping table, and encapsulate them in a structured way to form a search key.

[0043] Furthermore, for each field in the dietary behavior characteristics and dietary health management indicators, the matching field values ​​in the dietary behavior characteristics and dietary health management indicators are read one by one according to the correspondence between source fields and target fields defined in the preset field mapping table; the read field values ​​are arranged according to the order and data type specified in the preset field mapping table; the arranged field values ​​are encapsulated in a structured manner to generate a key-value pair set containing all mapped field values, and the key-value pair set is used as the retrieval key.

[0044] It should be noted that the field mapping table is a structured configuration table used to define the correspondence, order, and data type between each source field in dietary behavior characteristics and dietary health management indicators and the target field in the search key; the field mapping table is configured based on the set of feature fields required for similar user retrieval and similarity calculation, after the participation scope, correspondence, arrangement order, and data type of each field are uniformly defined in the feature design stage.

[0045] S3.2. Using the search key as a condition, perform a similar user search in the historical data warehouse to obtain a set of candidate users, and calculate the similarity score between the dietary behavior characteristics and dietary health management indicators of each candidate user in the set and the corresponding fields of the current user.

[0046] Furthermore, all field values ​​in the search key are matched against the dietary behavior characteristics and dietary health management indicators of each user stored in the historical data warehouse, and user records where all fields are non-empty and cover the dimensions contained in the search key are retained to form a candidate user set. For each candidate user in the candidate user set, the field values ​​corresponding to the search key in the candidate user's dietary behavior characteristics and dietary health management indicators are extracted. The similarity between the corresponding field values ​​in the current user's dietary behavior characteristics and dietary health management indicators and the corresponding field values ​​in the candidate user's indicators is calculated item by item, as expressed by: ; In the formula, Is the current user With candidate users In the Similarity across fields; It is the current user; They are candidate users; It is the first of the current user's dietary behavior characteristics or dietary health management indicators. Numeric values ​​for each field; It is the first of the candidate users' dietary behavior characteristics or dietary health management indicators. Numeric values ​​for each field; It is the first The standard deviation of each field's values ​​across all user field values ​​in the historical data warehouse; It is a very small positive number (e.g.) ), used to avoid the denominator being zero; It is the sequence number of the field in the search key; among them, dietary behavior characteristics include numerical characteristics such as the intake of various nutrients, total energy intake, number of food categories, number of meals, and distribution regularity of meal time. The dietary behavior characteristics are processed to unify the dimensions, converting features with different units of measurement into dimensionless or standardized values ​​with a unified scale.

[0047] S3.3 Sort each candidate user according to the similarity score from high to low, select the top N candidate users as the similar user group, and extract the dietary behavior characteristics and dietary health management indicators of the similar user group under the same time window to form similar user group data.

[0048] Furthermore, the similarity scores of each candidate user in the candidate user set are sorted in descending order of value from high to low. The top N candidate users (N is 30 in this example) are selected from the sorting results to form a similar user group. For each candidate user in the similar user group, their dietary behavior characteristics under the same time window as the current user are extracted from the historical data warehouse. At the same time, dietary health management indicators for each candidate user in the similar user group under the same time window are extracted. The extracted dietary behavior characteristics and dietary health management indicators are summarized to form similar user group data.

[0049] S3.4 Statistically aggregate the various dietary behavior characteristics of similar user groups to obtain the baseline level of dietary behavior, and statistically aggregate the various dietary health management indicators to obtain the baseline status of dietary health.

[0050] Furthermore, for each dietary behavior characteristic in the data of similar user groups, a statistical aggregate value is calculated for each similar user group. The aggregation method includes the mean, median, or mode, forming a baseline level for dietary behavior. The above statistical aggregation operation is repeated for all dietary behavior characteristics to obtain a complete set of baseline levels for dietary behavior. For each dietary health management indicator in the data of similar user groups, a statistical aggregate value is calculated for each similar user group. The aggregation method includes the compliance rate, average deviation level, or the most frequently occurring indicator status, forming a baseline status for dietary health.

[0051] S3.5. Unify the baseline level of dietary behavior and the baseline state of dietary health into a structured record to form a baseline profile of dietary behavior.

[0052] Furthermore, each feature value in the dietary behavior baseline level is aligned with each indicator value in the dietary health baseline state according to the field name corresponding to it in the search key; all aligned field values ​​are organized into a key-value pair set with a unified structure, including field name, baseline value and its data type metadata, and the key-value pair set is used as a complete structured record to form a dietary behavior baseline profile.

[0053] It should be noted that the feature values ​​in the baseline level of dietary behavior are obtained by statistically aggregating the corresponding dietary behavior feature fields in similar user group data (such as truncated mean or mode). The indicator values ​​in the baseline state of dietary health are obtained by statistically aggregating the corresponding dietary health management indicator fields (such as compliance rate, average deviation level, or most frequently occurring state) from data of similar user groups.

[0054] S3.6 Calculate the field difference between the current user's field value and the field value of the dietary behavior baseline profile, determine the direction of difference according to field type, and summarize and aggregate the field difference and direction of difference to form a dietary behavior deviation metric.

[0055] Furthermore, the algorithm iterates through each field in the current user's dietary behavior characteristics and dietary health management indicators to obtain field values. It extracts the baseline values ​​for the corresponding fields from the dietary behavior baseline profile, calculates the field difference between the current user's field value and the corresponding baseline value in the dietary behavior baseline profile, uses the difference (current value minus baseline value) for continuous fields, uses the grade difference for ordered categorical fields, and marks whether unordered categorical fields are equal. It determines the direction of the difference based on the field type and business meaning; for example, energy intake higher than the baseline value is considered a "positive deviation," and vegetable intake lower than the baseline value is considered a "negative deviation." Finally, it organizes and summarizes the field differences and their corresponding directions by field name to form a dietary behavior deviation metric.

[0056] It should be noted that the current user field value is the original numerical value or coded value directly extracted from the current user's dietary behavior characteristics and dietary health management indicators, and corresponds one-to-one with each field in the dietary behavior benchmark profile.

[0057] S3.7. Using dietary behavior deviation measurement as input, load the preset management event trigger rule table and use deviation characteristics as trigger conditions. By traversing the preset management event trigger rule table, match the deviation field and deviation level in dietary behavior deviation measurement with the trigger conditions.

[0058] Furthermore, all deviation fields and deviation level information are extracted from the dietary behavior deviation metric. The system then iterates through the pre-defined management event triggering rule table. For each rule, the deviation fields and deviation levels in the dietary behavior deviation metric are compared and matched against the triggering conditions. If a deviation field or level matches a triggering condition, the corresponding management event is recorded for later processing. During the matching process, for each deviation field, the value in the dietary behavior deviation metric is compared one by one with the corresponding standard value or range in the pre-defined management event triggering rule table to determine if the triggering condition is met. For example, if a deviation field in the dietary behavior deviation metric has a value of 10%, and the corresponding deviation level threshold in the pre-defined management event triggering rule table is 5%-15%, then the deviation field is considered to meet the triggering condition. After traversing all rules, all successfully matched management events are summarized.

[0059] It should be noted that the management event trigger rule table is constructed based on the correlation analysis results between the dietary behavior benchmark profiles of similar user groups in the historical data warehouse and the corresponding health outcomes (such as weight changes, blood sugar fluctuations, etc.). Deviation fields with clear health impacts are solidified as trigger conditions and stored in the form of structured tables.

[0060] S3.8 If the triggering conditions are met, the corresponding management event type will be output. If multiple triggering conditions are met at the same time, the management event type with the highest priority in the preset management event triggering rule table will be selected as the target output.

[0061] Furthermore, it determines whether the dietary behavior deviation metric meets any of the triggering conditions in the preset management event triggering rule table. If it does, it records the management event type associated with the triggering condition. If the dietary behavior deviation metric meets multiple triggering conditions, it obtains the priority values ​​defined for each of these triggering conditions in the preset management event triggering rule table. All management event types that meet the conditions are sorted in descending order according to their corresponding priority values. The management event type with the highest priority value in the sorting results is selected and used as the target output.

[0062] S4. Based on the type of management event, retrieve candidate dietary health intervention paths from the intervention strategy library, perform constraint screening and feasibility verification on each candidate dietary health intervention path, and determine the target dietary health management intervention path.

[0063] S4.1. Using the management event type as the search entry point, perform a matching search in the intervention strategy library and output candidate dietary health intervention paths.

[0064] Furthermore, using the management event type as the search entry point, all intervention strategy records that perfectly match the management event type are found in the intervention strategy library. The complete intervention steps, execution conditions, and expected goals corresponding to each record are extracted to form candidate dietary health intervention paths.

[0065] It should be noted that the intervention strategy library is a collection of strategies that have been validated and proven effective in historical dietary health intervention cases, such as dietary adjustment programs to control energy intake, dietary structure optimization programs to increase vegetable and whole grain intake, and behavioral intervention programs to regulate meal times; these strategies are stored internally in a structured manner.

[0066] S4.2 Read the constraints corresponding to each candidate dietary health intervention path, and compare the constraints with the current user's dietary health dataset, dietary behavior characteristics and dietary health management indicators one by one. Eliminate candidate dietary health intervention paths that do not meet the constraints to form the filtered intervention paths.

[0067] Furthermore, the constraints associated with each candidate dietary health intervention path are read one by one. These constraints include structured rules such as the applicable population scope, dietary behavior characteristics restrictions, and dietary health management indicator status. The judgment fields in each constraint are compared item by item with the actual values ​​of the corresponding fields in the current user's dietary health dataset, dietary behavior characteristics, and dietary health management indicators. The comparison operation is implemented through exact matching, interval inclusion judgment, or logical relationship verification. For example, it is determined whether the current user's age is within the age range specified by the constraint, or whether the current user's vegetable intake is higher than the minimum threshold set by the constraint (which is the pre-set lower limit of vegetable intake for a specific intervention path, used to limit the minimum dietary baseline level applicable to the intervention path, and the minimum threshold is determined based on the statistics of previous intervention effects and nutritional recommendation standards). If any constraint of a candidate dietary health intervention path is not met, the candidate dietary health intervention path is removed from the candidate set. After completing the constraint verification of all candidate dietary health intervention paths, the paths that meet all constraints are retained to form the filtered intervention paths.

[0068] S4.3. Check the feasibility of each candidate dietary health intervention path in the screened intervention path one by one, and eliminate the unfeasible candidate dietary health intervention paths to form an executable intervention path.

[0069] Furthermore, each candidate dietary health intervention path in the screened intervention path is examined one by one for the execution resources it depends on, the user's current physiological state support, and the feasibility of the intervention cycle. If a candidate dietary health intervention path has problems such as unavailable resources, conflict with the user's current health state, or the required execution cycle exceeds the allowable range, the candidate dietary health intervention path is determined to be unexecutable and is eliminated. All candidate dietary health intervention paths that pass the executability verification are retained to form executable intervention paths.

[0070] S4.4 Sort the executable intervention paths according to the preset priority rules, and select the executable intervention path with the highest priority as the target dietary health management intervention path.

[0071] Furthermore, the priority value associated with each executable intervention path in the intervention strategy library is read, and all executable intervention paths are sorted from high to low according to the priority value. If multiple executable intervention paths have the same highest priority value, any one of them is retained or further sorted in ascending order according to the intervention path identifier to ensure uniqueness. The executable intervention path ranked first in the sorting results is selected as the target dietary health management intervention path.

[0072] S5. Real-time collection of the implementation effect of the target dietary health management intervention path, and recording of changes in user dietary behavior to form health feedback data.

[0073] S5.1. Start execution monitoring with the target dietary health management intervention path as the execution object, collect the execution effect data of the target dietary health management intervention path in real time, and simultaneously record the changes in the user's dietary behavior during the intervention execution period. Integrate the execution effect data and the changes in dietary behavior to form health feedback data.

[0074] Furthermore, execution monitoring is initiated using the target dietary health management intervention path as the execution target. Real-time data on the execution effectiveness of the target dietary health management intervention path is collected, including intervention completion rate, user compliance score, and changes in physiological indicators. Simultaneously, changes in user dietary behavior during the intervention period are recorded, covering fields such as daily food intake types, meal time distribution, energy intake, and nutrient composition. The execution effectiveness data and dietary behavior changes are time-aligned and horizontally stitched together using timestamps within the intervention period as the association key. User identifiers and intervention path identifiers are added to the stitched records to form structured records. All structured records are then aggregated to form health feedback data.

[0075] This embodiment also provides a computer device applicable to the dietary health management method based on big data analysis, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the dietary health management method based on big data analysis as proposed in the above embodiment.

[0076] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0077] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the dietary health management method based on big data analysis as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0078] In summary, this invention achieves standardized aggregation of multi-source user health data, extraction of dietary behavior characteristics through time window segmentation and statistical analysis, and construction of benchmark profiles based on similar user groups from historical data warehouses. This enables the use of big data analysis to perform group-based benchmark comparisons and deviation measurement of individual dietary behaviors, and then maps the deviation results into management event types that can drive decision-making. Simultaneously, through event-driven intervention path constraint screening, feasibility verification, and execution effect feedback mechanisms, a closed-loop dietary health management process covering assessment, decision-making, and feedback is constructed, improving the accuracy, feasibility, and continuous optimization capabilities of dietary health management.

[0079] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A dietary health management method based on big data analysis, characterized in that: include, Collect users' basic health data and aggregate and process it to form a dietary health dataset; The diet and health dataset is divided according to the preset time window rules, and the diet behavior management data in each time window is statistically analyzed to extract diet behavior characteristics. Through the preset diet behavior management rules, multi-dimensional comparative analysis is carried out to generate diet and health management indicators. Based on dietary behavior characteristics and dietary health management indicators, similar user group data are retrieved from historical data warehouse to construct a dietary behavior benchmark profile. The difference between dietary behavior characteristics and dietary behavior benchmark profile is calculated to output a dietary behavior deviation measure, and the dietary behavior deviation measure is mapped to the corresponding management event type. Candidate dietary health intervention paths are retrieved from the intervention strategy library based on the type of management event. Constraint screening and feasibility verification are performed on each candidate dietary health intervention path to determine the target dietary health management intervention path. The system collects real-time data on the effectiveness of the target dietary health management intervention pathway and records changes in users' dietary behavior to generate health feedback data.

2. The dietary health management method based on big data analysis as described in claim 1, characterized in that: The specific steps for forming a dietary health dataset through aggregation processing are as follows. The system deduplicates user basic health data by user identifier and time information, and identifies and corrects missing and abnormal data to obtain valid health data. Valid health data from different sources are standardized by unifying fields, converting formats, and standardizing units of measurement to form standardized health data. Based on user identification and time dimension, the standardized health data are correlated and fused to generate a dietary health dataset.

3. The dietary health management method based on big data analysis as described in claim 2, characterized in that: The dietary health dataset is divided according to a preset time window rule, and statistical analysis is performed on the dietary behavior management data within each time window to extract dietary behavior features. The specific steps are as follows: The diet and health dataset is divided into time periods according to the preset time window rules, and the diet and health data belonging to the same time window are aggregated to form diet behavior management data. Using dietary behavior management data as the statistical object, dietary intake data and eating behavior data within a time window are summarized to generate dietary status statistics. Based on the dietary status statistics, dietary behavior is structured and summarized to form dietary behavior characteristics.

4. The dietary health management method based on big data analysis as described in claim 1, characterized in that: The generation of dietary health management indicators refers to inputting dietary behavior characteristics into preset dietary behavior management rules, performing comparative judgments item by item according to different dietary behavior dimensions, obtaining multi-dimensional comparative analysis results for each time window, and converting each multi-dimensional comparative analysis result into indicator details according to a preset coding table. The indicator details are written as dietary health management indicators with user identifier and window start and end time as the primary keys.

5. The dietary health management method based on big data analysis as described in claim 3 or 4, characterized in that: The method involves retrieving data from a historical data warehouse to construct a baseline profile of dietary behavior based on dietary behavior characteristics and dietary health management indicators. The specific steps are as follows: For dietary behavior characteristics and dietary health management indicators, the corresponding field values ​​are read one by one according to the preset field mapping table, and then structured and encapsulated to form search keys; Using the search key as a condition, perform similar user searches in the historical data warehouse to obtain a set of candidate users, and calculate the similarity score between the dietary behavior characteristics and dietary health management indicators of each candidate user in the set and the corresponding fields of the current user. The candidate users are sorted from high to low according to their similarity scores. The top N candidate users are selected as the similar user group. At the same time, the dietary behavior characteristics and dietary health management indicators of the similar user group under the same time window are extracted to form similar user group data. The dietary behavior characteristics of similar user groups are statistically aggregated to obtain the baseline level of dietary behavior, and the dietary health management indicators are statistically aggregated to obtain the baseline status of dietary health. The baseline levels of dietary behavior and the baseline status of dietary health are uniformly packaged into a structured record to form a baseline profile of dietary behavior.

6. The dietary health management method based on big data analysis as described in claim 5, characterized in that: The specific steps for calculating the difference between dietary behavior characteristics and the baseline dietary behavior profile, outputting a dietary behavior deviation metric, and mapping the dietary behavior deviation metric to the corresponding management event type are as follows. Calculate the field difference between the current user's field values ​​and the field values ​​of the dietary behavior baseline profile, determine the direction of the difference according to the field type, and summarize and aggregate the field difference and direction of the difference to form a dietary behavior deviation metric. Using dietary behavior deviation metrics as input, a preset management event triggering rule table is loaded and deviation features are used as triggering conditions. By traversing the preset management event triggering rule table, the deviation fields and deviation levels in the dietary behavior deviation metrics are matched with the triggering conditions. If the triggering conditions are met, the corresponding management event type will be output. If multiple triggering conditions are met at the same time, the management event type with the highest priority will be selected as the target output according to the priority in the preset management event triggering rule table.

7. The dietary health management method based on big data analysis as described in claim 6, characterized in that: The steps are as follows: Candidate dietary health intervention paths are retrieved from the intervention strategy database based on the type of management event; constraint screening and feasibility verification are performed on each candidate dietary health intervention path; and the target dietary health management intervention path is determined. Using the type of management event as the search entry, perform a matching search in the intervention strategy library and output candidate dietary health intervention paths; Read the constraints corresponding to each candidate dietary health intervention path one by one, and compare the constraints with the current user's dietary health dataset, dietary behavior characteristics and dietary health management indicators item by item. Eliminate candidate dietary health intervention paths that do not meet the constraints to form the filtered intervention path. The feasibility of each candidate dietary health intervention path in the screened intervention pathways was checked one by one, and the unfeasible candidate dietary health intervention paths were eliminated to form an executable intervention path. The executable intervention paths are sorted according to the preset priority rules, and the highest priority executable intervention path is selected as the target dietary health management intervention path.

8. The dietary health management method based on big data analysis as described in claim 7, characterized in that: The formation of health feedback data refers to initiating execution monitoring with the target dietary health management intervention path as the execution object, collecting execution effect data of the target dietary health management intervention path in real time, and simultaneously recording changes in users' dietary behavior during the intervention execution period, and integrating the execution effect data with the changes in dietary behavior.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the dietary health management method based on big data analysis as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the dietary health management method based on big data analysis as described in any one of claims 1 to 8.

Images