Standardized food fast inspection information pushing method and system

Abstract

The invention provides a food fast inspection information standardized pushing method and system, and the method comprises the steps: carrying out the screening and evaluation of to-be-inspected items of each food in different regions, determining the to-be-inspected items of each food in different regions, and building a first mapping relation between the food information of each food in different regions and the corresponding to-be-inspected items, dynamically adjusting corresponding inspection items through a mixed algorithm model fusing risk weight calculation and machine learning; when food detection is carried out, food information is obtained according to the unique label code of the food to be detected, the item to be detected is determined according to the food information and the first mapping relation, and the item to be detected is pushed, specifically, it can be ensured that the item to be detected closely follows the risk change through an intelligent algorithm dynamic evaluation method, and the risk change can be accurately detected. And regional differences and special requirements can be considered, so that the food safety is guaranteed to the maximum extent while the efficiency is improved.

Description

Technical Field

[0001] This invention belongs to the field of standardized information push technology for rapid food testing, and specifically relates to a standardized method and system for pushing rapid food testing information. Background Technology

[0002] To ensure food safety, it is necessary to conduct rapid testing on food in various regions regularly or irregularly. However, due to the wide variety of food types and the numerous and ever-changing testing items required for each type of food, the technical requirements for rapid testing personnel in various regions are very high and difficult to standardize. Summary of the Invention

[0003] Based on this, the present invention provides a standardized method and system for pushing rapid food testing information, which aims to intelligently push rapid testing information of food in various regions.

[0004] A first aspect of this invention provides a standardized method for pushing rapid food testing information, the method comprising: By screening and evaluating the test items for various foods in different regions, the test items for various foods in different regions are determined, and a first mapping relationship between food information and corresponding test items for various foods in different regions is established. When conducting food testing, the food information is obtained based on the unique label code of the food to be tested, and the items to be tested are determined based on the food information and the first mapping relationship, and the items to be tested are pushed out. In the step of screening and evaluating the test items of various foods in different regions to determine the test items, a hybrid algorithm model that integrates risk weight calculation and machine learning is used to dynamically adjust the test items.

[0005] Furthermore, after the step of obtaining food information based on the unique label code of the food to be inspected, determining the items to be inspected based on the food information and the first mapping relationship, and pushing the items to be inspected during food testing, the following steps are included: Establish a second mapping relationship between each test item and its corresponding rapid test reagent; When conducting food testing, the target rapid test reagent is determined based on the test items and the second mapping relationship; Obtain the current rapid test reagent, compare it with the target rapid test reagent, and determine whether the rapid test reagent information is consistent; If so, proceed with the subsequent rapid food testing procedure.

[0006] Furthermore, the step of determining the required testing items for various foods in different regions by screening and evaluating the testing items includes: A multi-dimensional data collection and integration network is constructed, which includes a basic data layer and an external risk data layer. The basic data layer includes the attributes of the food itself and historical testing data, while the external risk data layer includes safety incident data, supply chain data, and consumer feedback data. The data in the multi-dimensional data acquisition and integration network is preprocessed, and the preprocessed data is subjected to feature extraction and feature selection to determine the target features; For each potential testing item, values ​​are assigned based on two dimensions: the probability of risk occurrence and the degree of risk impact. A comprehensive risk score is calculated for each item. The potential testing items include existing testing items and newly added candidate items. Determine whether the overall risk score is greater than the first threshold; If the overall risk score is greater than the first threshold, the corresponding item will be identified as an item to be inspected. The random forest algorithm is used to train a random forest model to identify high-risk items based on the labels of whether historical security events were caused by undetected items and the target features. At the same time, a reinforcement learning algorithm is introduced so that the random forest model can dynamically adjust the weights and composition of the items to be detected based on the detection results. Obtain the comprehensive risk score corresponding to the high-risk project, and determine whether the comprehensive risk score corresponding to the high-risk project is greater than the second threshold, wherein the second threshold is less than the first threshold; If the overall risk score of a high-risk item is greater than the second threshold, then the corresponding high-risk item will be identified as an item to be inspected.

[0007] Furthermore, the step of determining the required testing items for various foods in different regions by screening and evaluating the testing items for each food item also includes: The inspection items are dynamically adapted according to the distribution links, wherein the distribution links include at least the production link, the warehousing link and the retail link; The detection items are adjusted according to special scenarios, where the special scenarios include at least major events, seasonal changes, and emergencies.

[0008] Furthermore, in the step of preprocessing the data in the multi-dimensional data acquisition and integration network, and extracting and selecting features from the preprocessed data to determine the target features, the extracted features include at least food type, raw material composition, production process, frequency of historical exceedances, number of safety incidents, frequency of negative consumer feedback, and toxicity of risky substances. In feature selection, the variance selection method is used to retain features with discriminative power. At the same time, the mutual information method is used to calculate the mutual information value between the features and the adjustment results of the inspection items. Features whose mutual information values ​​meet the preset values ​​are selected and included in the training of the random forest model.

[0009] Furthermore, the step of assigning values ​​to each potential detection item from two dimensions—risk occurrence probability and risk impact degree—and calculating the comprehensive risk score for each item includes: Calculate the probability of risk occurrence based on the frequency of exceeding standards in historical tests and the incidence of safety incidents involving similar foods. Values ​​are assigned based on the scope of the population consuming the product, the harmful consequences, and the social impact, and the degree of risk impact is calculated. The risk score is obtained by weighting and summing the probability of occurrence and the degree of impact of the risk.

[0010] A second aspect of this invention provides a standardized food rapid testing information push system for implementing the standardized food rapid testing information push method described in the first aspect, the system comprising: The evaluation module is used to screen and evaluate the test items of various foods in different regions, determine the test items of various foods in different regions, and establish the first mapping relationship between food information of various foods in different regions and the corresponding test items. The first determining module is used to obtain food information based on the unique label code of the food to be inspected when food testing is carried out, and to determine the items to be inspected based on the food information and the first mapping relationship, and to push the items to be inspected. In the step of screening and evaluating the test items of various foods in different regions to determine the test items, a hybrid algorithm model that integrates risk weight calculation and machine learning is used to dynamically adjust the test items.

[0011] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the standardized push method for rapid food testing information provided in the first aspect.

[0012] A fourth aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the standardized push method for rapid food testing information provided in the first aspect.

[0013] This invention provides a standardized method and system for pushing rapid food testing information. By screening and evaluating the test items for various foods in different regions, the system determines the required test items for each food in different regions and establishes a first mapping relationship between food information and corresponding test items. Specifically, a hybrid algorithm model integrating risk weight calculation and machine learning is used to dynamically adjust the test items. When food testing is conducted, food information is obtained based on the unique label code of the food to be tested. Based on the food information and the first mapping relationship, the test items are determined and pushed to the system. The method of dynamic evaluation using intelligent algorithms ensures that the test items keep pace with risk changes while also taking into account regional differences and special needs, maximizing food safety while improving efficiency. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating the implementation of a standardized method for pushing rapid food testing information according to Embodiment 1 of the present invention. Figure 2 This is a structural block diagram of a standardized food rapid testing information push system provided in Embodiment 2 of the present invention; Figure 3 This is a structural block diagram of an electronic device provided in Embodiment 3 of the present invention. Detailed Implementation

[0015] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0016] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0018] Example 1 According to an embodiment of the present invention, a method for standardized push of rapid food testing information is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0019] This first embodiment provides a standardized method for pushing rapid food testing information, which can be used in electronic devices, such as computers. Please refer to... Figure 1 , Figure 1 The flowchart of a standardized method for pushing rapid food testing information provided in Embodiment 1 of the present invention is shown, specifically including steps S01 to S02.

[0020] Step S01: By screening and evaluating the test items of various foods in different regions, the test items of various foods in different regions are determined, and a first mapping relationship between the food information of various foods in different regions and the corresponding test items is established.

[0021] In this embodiment of the invention, a hybrid algorithm model integrating risk weight calculation and machine learning is used to dynamically adjust the testing items. Specifically, a multi-dimensional data collection and integration network is constructed, which includes a basic data layer and an external risk data layer. The basic data layer includes the food's inherent attributes and historical testing data, while the external risk data layer includes safety incident data, supply chain data, and consumer feedback data. It should be noted that the food's inherent attributes include type, raw material composition, production process, distribution channels, and shelf life. Historical testing data includes real-time connections between testing equipment at testing sites in various regions and the cloud data center. After each test, the system automatically uploads the test results (such as the specific value of a certain indicator and whether it is qualified), testing methods (such as rapid testing methods and laboratory testing methods), reagent brands and batches used, and other information. The system performs structured processing on the uploaded data, classifies and stores it according to dimensions such as food type, testing items, and testing time, and establishes an index for quick retrieval.

[0022] In order to collect safety incident data, web crawling technology is used to set specific keywords (such as "food poisoning", "food recall", "illegal additives", etc.) to regularly crawl information released by official platforms. The crawled information is filtered and analyzed to extract key information such as the types of food involved, risky substances, and the location and time of the incident, and then automatically entered into the risk database. Supply chain data includes environmental data of raw material origins, changes in supplier qualifications, and pollution risks during transportation. For example, to collect environmental data of raw material origins, the system connects with relevant environmental data platforms to obtain information such as soil heavy metal content and water quality indicators, and links them to the corresponding raw material origins. When food raw materials come from this origin, the relevant environmental data is automatically matched. To collect data on changes in supplier qualifications, the system connects with the enterprise credit information system to obtain information such as supplier qualification certifications and administrative penalty records in real time. When supplier qualifications change, the system automatically alerts and updates the supply chain risk level of the relevant food. To obtain pollution risks during transportation, transport vehicles are equipped with GPS positioning and temperature and humidity sensors, and upload location and environmental data at regular intervals. When a cold chain break occurs (temperature exceeds the set range and continues for a period of time), the system automatically records and marks the transported batch of food as high-risk. Data from the multi-dimensional data acquisition and integration network is preprocessed, and feature extraction and selection are performed on the preprocessed data to determine target features. It should be noted that preprocessing includes data cleaning and data standardization. In data cleaning, missing values ​​in the basic data layer and external risk data layer are handled using mean imputation (suitable for numerical data such as detection index values) and mode imputation (suitable for categorical data such as food types). Outliers are then identified and removed. For data in the detection results that significantly deviate from the normal range, professional knowledge is used to determine whether they are reasonable anomalies; if they are unreasonable anomalies, they are deleted. Finally, duplicate data is deduplicated to ensure the uniqueness of each data point. In data standardization, numerical data of different magnitudes (such as risk occurrence probability, detection index values, etc.) are converted to the [0,1] interval and standardized using min-max. Categorical data (such as food types, detection methods, etc.) are one-hot encoded and converted into binary vector form for easier model processing. Furthermore, during the feature extraction and selection process, features related to food inspection items are extracted from the preprocessed data, including food type, raw material composition, production process, frequency of historical exceedances, number of safety incidents, frequency of negative consumer feedback, and toxicity level of risk substances. Further, the variance selection method is used to retain discriminative features. At the same time, the mutual information method is used to calculate the mutual information value between the features and the adjustment results of the inspection items, and features whose mutual information values ​​meet the preset values ​​are included in the training of the random forest model. For each potential testing item, a risk score is calculated based on two dimensions: the probability of risk occurrence and the degree of risk impact. The potential testing items include existing testing items and newly added candidate items. Specifically, the probability of risk occurrence is calculated based on the frequency of exceedances in historical tests and the incidence of safety incidents involving similar foods. In this embodiment of the invention, for existing testing items, the number of times the same food was tested in various regions over the past 12 months is statistically analyzed, and this number is divided by the total number of tests to obtain the exceedance frequency of that item. This frequency serves as the main reference indicator for the probability of risk occurrence. The calculation formula is as follows: ; Where, N 超标 N represents the number of times the project exceeded the standard in the past 12 months. 总检测 This refers to the total number of tests conducted in the project over the past 12 months. For newly added candidate items, an initial probability value is assigned based on the frequency of occurrence of the item in similar food safety incidents in other regions or historically, combined with expert assessments of the potential risks of the item in current food products. The calculation formula is as follows: ; Wherein, P0 is the initial probability value, P1 is the frequency of safety incidents involving similar foods in other regions, and P2 is the expert assessment probability. As detection data accumulates, the probability value is continuously adjusted. The calculation formula is as follows: ; The risk impact is assessed based on the target consumer group, the harmful consequences, and the social impact. Specific weighting methods include: assigning weights based on the target consumer group (e.g., infants, pregnant women, general population), with infant food having the highest weight (1.0), food for pregnant women 0.9, food for the general population 0.6, and other special population foods weighted between 0.6 and 0.9 depending on the specific circumstances; assigning weights based on the toxicity level of the hazardous substances (e.g., carcinogens 1.0, teratogens 0.9, short-term poisoning substances 0.7, other harmful substances weighted between 0.5 and 0.9 depending on their toxicity level); and referencing the public opinion intensity of similar historical safety incidents (e.g., number of media reports, social media discussions), with large-scale public opinion events (more than 1000 media reports) weighted at 1.0, medium-scale public opinion events (300-1000 media reports) weighted at 0.7, and localized small-scale events (less than 300 media reports) weighted at 0.3. Based on the probability of risk occurrence and the degree of risk impact, a weighted sum is obtained to obtain a comprehensive risk score. For example, the comprehensive risk score = probability of risk occurrence × 0.6 + degree of risk impact × 0.4 (the weights can be adjusted by experts according to the actual situation). When the score is higher than the set first threshold, the item is included in the scope of inspection. Determine whether the overall risk score is greater than the first threshold; If the overall risk score is greater than the first threshold, the corresponding item will be identified as an item to be inspected. A random forest algorithm is used to train a random forest model to identify high-risk items based on labels indicating whether historical safety incidents were caused by undetected items and the target features. Simultaneously, a reinforcement learning algorithm is introduced, enabling the random forest model to dynamically adjust the weights and composition of items to be inspected based on detection results. In this embodiment, the dataset used to train the random forest model consists of collected food safety incident cases from previous years. Each case includes extracted features and a label indicating whether an undetected item caused a safety incident (1 for yes, 0 for no). These cases are divided into training and testing sets in a 7:3 ratio. The training set is then used to train the random forest model, setting the number of decision trees to 100 and the maximum depth to 10. Other parameters are optimized using a grid search method, achieving an accuracy of over 85% on the testing set. For newly added potential detection items, their features are input into the trained model to obtain a high-risk probability. When the high-risk probability exceeds a third threshold, the item is designated as a key evaluation item, i.e., a high-risk item. Furthermore, the reinforcement learning algorithm's learning process involves initializing a Q-table, selecting an action based on the current state in each detection cycle, obtaining a reward after executing the action, and updating the Q-value. ; in, The current Q value (an assessment of how "good" or "bad" the action is) when taking action a in state s. Let r be the learning rate, and r be the immediate reward the agent receives upon transitioning to the new state s' after performing action a in state s. As a discount factor, Let be the value of the "optimal action" in state s' as perceived by the agent, where s' is the new state the agent transitions to after executing action a, and a' is the action that can be chosen in the new state s'. The state space is the current set of items to be checked and the weights of each item. The action space includes adding an item, deleting an item, increasing the weight of an item, and decreasing the weight of an item. The reward function is as follows: if no safety event occurs after detection, a positive reward of R1=10 is given; if a safety event occurs, a negative reward of R2=-20 is given. Corresponding rewards or penalties are given based on the improvement or decrease in detection efficiency (e.g., a positive reward for shortening detection time and a negative reward for extending it). Furthermore, the model's performance is evaluated using metrics such as accuracy, recall, and F1 score, while also monitoring changes in the safety event incidence rate and consumer complaint volume after adjustments to the checked items. Further, the model is evaluated at preset intervals, and model parameters (such as coefficients in risk weight calculation and the number of decision trees in the random forest) are adjusted based on the evaluation results to continuously optimize model performance. Obtain the comprehensive risk score corresponding to the high-risk project, and determine whether the comprehensive risk score corresponding to the high-risk project is greater than the second threshold, wherein the second threshold is less than the first threshold; If the overall risk score of a high-risk item is greater than the second threshold, then the corresponding high-risk item will be identified as an item to be inspected.

[0023] In other embodiments of the present invention, the testing items are dynamically adapted according to the distribution process. The distribution process includes at least production, warehousing, and retail stages. For example, in the production stage, the focus is on detecting risks introduced by raw materials, such as pesticide and veterinary drug residues in agricultural products, and heavy metal content in raw materials for processed foods, with testing frequency once per batch. In the warehousing stage, storage-related testing items such as "microbial contamination" and "oxidative spoilage" are added. For refrigerated foods, "total bacterial count" and "coliform bacteria" are tested daily; for foods stored at room temperature, testing is conducted every 3 days. In the retail stage, the focus is on items such as "label compliance" (e.g., altered shelf life, incorrect ingredient labeling) and "risks of near-expiry foods," with random sampling of on-sale foods conducted daily, with a sampling rate of no less than 5%. The system adjusts testing items based on specific scenarios, including at least major events, seasonal changes, and emergencies. For example, during major events, one month prior to the event, the testing frequency of high-frequency items (such as microorganisms and heavy metals) is doubled based on the event's scale and number of participants, and the testing of food additive usage is increased. During seasonal changes, the system automatically adjusts the priority of testing items according to the season. In summer (June-August), the priority of items such as "total bacterial count" and "coliform bacteria" is increased by 30%; in winter (December-February), the priority of "nitrite content" (such as in pickled products) is increased by 20%. In emergencies, when a food safety emergency (such as avian influenza) occurs in a certain area, the system includes specific testing items (such as virus testing) for relevant foods (such as poultry) in the required testing items within one hour and simultaneously transmits this information to testing sites nationwide, while increasing the sampling rate of such foods to over 50%.

[0024] Step S02: When conducting food testing, obtain food information based on the unique label code of the food to be tested, determine the items to be tested based on the food information and the first mapping relationship, and push the items to be tested.

[0025] Understandably, during testing, rapid testing personnel first scan the unique label code on the food to be tested, thus identifying the type of food. The backend system then automatically pushes the previously associated testing items based on the food type, enabling standardized and unified testing of the same food across different regions at the same time, ensuring the efficiency and reliability of food inspection. Furthermore, the rapid testing reagents used for each testing item for various foods have been standardized. During testing, personnel need to scan the QR code on the rapid testing reagent to verify its information. The backend system then determines whether the information of the rapid testing reagent meets the testing requirements for the current food item; only if it does can the subsequent testing be completed.

[0026] Specifically, a second mapping relationship is established between each test item and its corresponding rapid test reagent; When conducting food testing, the target rapid test reagent is determined based on the test items and the second mapping relationship; Obtain the current rapid test reagent, compare it with the target rapid test reagent, and determine whether the rapid test reagent information is consistent; If the rapid test reagent information is consistent, then proceed with the subsequent rapid food testing procedure.

[0027] In summary, the standardized food rapid testing information push method in the above embodiments of the present invention screens and evaluates the test items of various foods in different regions to determine the test items for each food in different regions, and establishes a first mapping relationship between food information of various foods in different regions and the corresponding test items. Specifically, a hybrid algorithm model integrating risk weight calculation and machine learning is used to dynamically adjust the test items. When food testing is conducted, food information is obtained based on the unique label code of the food to be tested, and the test items are determined based on the food information and the first mapping relationship, and then pushed to the relevant authorities. Specifically, the method of dynamic evaluation through intelligent algorithms ensures that the test items keep pace with risk changes while also taking into account regional differences and special needs, maximizing food safety while improving efficiency.

[0028] Example 2 Please see Figure 2 , Figure 2 This is a structural block diagram of a standardized food rapid testing information push system 200 provided in Embodiment 2 of the present invention. This standardized food rapid testing information push system 200 is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0029] Specifically, the standardized food rapid testing information push system 200 includes: an evaluation module 21 and a first determination module 22, wherein: The evaluation module 21 is used to screen and evaluate the test items of various foods in different regions, determine the test items of various foods in different regions, and establish a first mapping relationship between the food information of various foods in different regions and the corresponding test items. The first determining module 22 is used to obtain food information based on the unique label code of the food to be inspected when food testing is carried out, and to determine the items to be inspected based on the food information and the first mapping relationship, and to push the items to be inspected. In the step of screening and evaluating the test items of various foods in different regions to determine the test items, a hybrid algorithm model that integrates risk weight calculation and machine learning is used to dynamically adjust the test items.

[0030] Furthermore, in some optional embodiments of the present invention, the food rapid testing information standardization push system 200 further includes: A module is established to create a second mapping relationship between each test item and its corresponding rapid test reagent. The second determination module is used to determine the target rapid test reagent based on the test items and the second mapping relationship when conducting food testing. The judgment module is used to obtain the current rapid test reagent, compare it with the target rapid test reagent, and determine whether the rapid test reagent information is consistent. The execution module is used to perform subsequent food rapid testing operations when the rapid test reagent information is consistent.

[0031] Furthermore, in some optional embodiments of the present invention, the evaluation module 21 includes: The building unit is used to construct a multi-dimensional data collection and integration network. The multi-dimensional data collection and integration network includes a basic data layer and an external risk data layer. The basic data layer includes the attributes of the food itself and historical testing data. The external risk data layer includes safety incident data, supply chain data, and consumer feedback data. The preprocessing unit is used to preprocess the data in the multi-dimensional data acquisition and integration network, and to extract and select features from the preprocessed data to determine target features. The extracted features include at least food type, raw material composition, production process, frequency of past detection exceeding standards, number of safety incidents, frequency of negative consumer feedback, and toxicity of risky substances. In feature selection, the variance selection method is used to retain discriminative features. At the same time, the mutual information method is used to calculate the mutual information value between the features and the adjustment results of the inspection items. Features with mutual information values ​​that meet the preset value are selected and included in the random forest model training. The calculation unit is used to assign values ​​to each potential detection item from two dimensions: the probability of risk occurrence and the degree of risk impact, and to calculate the comprehensive risk score of each item. The potential detection items include existing items to be detected and newly added candidate items. The first judgment unit is used to determine whether the overall risk score is greater than the first threshold. The first determining unit is used to determine the corresponding item as an item to be inspected if the comprehensive risk score is greater than the first threshold. The training unit is used to train a random forest model to identify high-risk items based on the label of whether historical security events were caused by undetected items and the target features, using the random forest algorithm. At the same time, a reinforcement learning algorithm is introduced so that the random forest model can dynamically adjust the weights and composition of the items to be detected based on the detection results. The second judgment unit is used to obtain the comprehensive risk score corresponding to the high-risk project and determine whether the comprehensive risk score corresponding to the high-risk project is greater than the second threshold, wherein the second threshold is less than the first threshold. The second determining unit is used to determine the corresponding high-risk item as an item to be inspected if the comprehensive risk score corresponding to the high-risk item is greater than the second threshold.

[0032] Furthermore, in some optional embodiments of the present invention, the evaluation module 21 further includes: An adaptation unit is used to dynamically adapt the inspection items according to the circulation links, wherein the circulation links include at least the production link, the warehousing link and the retail link; The triggering unit is used to trigger adjustments to the detection items based on special scenarios, wherein the special scenarios include at least major events, seasonal changes, and emergencies.

[0033] Furthermore, in some optional embodiments of the present invention, the computing unit includes: The first calculation subunit is used to calculate the probability of risk occurrence based on the frequency of exceeding standards in historical tests and the incidence rate of safety incidents involving similar foods; The second calculation subunit is used to assign values ​​based on the scope of the population that consumes the product, the harmful consequences, and the social impact, and to calculate the degree of risk impact. The third calculation subunit is used to perform a weighted summation based on the probability of risk occurrence and the degree of risk impact to obtain a comprehensive risk score. Example 3 In another aspect, the present invention also proposes an electronic device, please refer to [link to relevant documentation]. Figure 3 The electronic device shown is an embodiment of the present invention, including a memory 20, a processor 10, and a computer program 30 stored in the memory and executable on the processor. When the processor 10 executes the computer program 30, it implements the food rapid testing information standardization push method as described above.

[0034] In some embodiments, the processor 10 may be a central processing unit (CPU), controller, microcontroller, microprocessor or other data processing chip, used to run program code stored in memory 20 or process data, such as executing access restriction programs.

[0035] The memory 20 includes at least one type of readable storage medium, such as flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 20 can be an internal storage unit of an electronic device, such as the hard disk of the electronic device. In other embodiments, the memory 20 can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. Furthermore, the memory 20 can include both internal and external storage units of the electronic device. The memory 20 can be used not only to store application software and various types of data of the electronic device, but also to temporarily store data that has been output or will be output.

[0036] It should be pointed out that, Figure 3 The structure shown does not constitute a limitation on the electronic device. In other embodiments, the electronic device may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0037] This invention also proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the standardized push method for rapid food testing information as described above.

[0038] Those skilled in the art will understand that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0039] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0040] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0041] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0042] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A food fast inspection information standardization pushing method, characterized in that, The method comprises: By screening and evaluating the to-be-inspected items of each food in different regions, determining the to-be-inspected items of each food in different regions, and establishing a first mapping relationship between the food information of each food in different regions and the corresponding to-be-inspected items; When food detection is performed, the food information is obtained according to the unique label code of the to-be-inspected food, the to-be-inspected items are determined according to the food information and the first mapping relationship, and the to-be-inspected items are pushed. In the step of determining the to-be-inspected items of each food in different regions by screening and evaluating the to-be-inspected items of each food in different regions, a hybrid algorithm model of risk weight calculation and machine learning is used to dynamically adjust the to-be-inspected items.

2. The food fast inspection information standardization pushing method according to claim 1, characterized in that, The step of determining the to-be-inspected items of each food in different regions by screening and evaluating the to-be-inspected items of each food in different regions further comprises: A second mapping relationship between each to-be-inspected item and the corresponding rapid detection reagent is established. When food detection is performed, the target rapid detection reagent is determined according to the to-be-inspected items and the second mapping relationship. The current rapid detection reagent is obtained and compared with the target rapid detection reagent, and it is determined whether the rapid detection reagent information is consistent. If yes, the subsequent food rapid detection operation is performed. 3.The food fast inspection information standardization pushing method according to claim 2, characterized in that, The step of determining the to-be-inspected items of each food in different regions by screening and evaluating the to-be-inspected items of each food in different regions comprises: A multi-dimensional data collection and integration network is constructed, which comprises a basic data layer and an external risk data layer, the basic data layer comprises food properties and historical detection data, and the external risk data layer comprises safety event data, supply chain data and consumer feedback data; The data in the multi-dimensional data collection and integration network is preprocessed, and the preprocessed data is subjected to feature extraction and feature selection to determine target features; Each potential detection item is valued from two dimensions of risk occurrence probability and risk impact degree, and the risk comprehensive score of each item is calculated, the potential detection item includes existing to-be-inspected items and new candidate items; It is judged whether the risk comprehensive score is greater than a first threshold value; If it is judged that the risk comprehensive score is greater than the first threshold value, the corresponding item is determined as a to-be-inspected item; A random forest algorithm is used to train a random forest model to identify high-risk items according to the label of whether a historical safety event is caused by an undetected item and the target features, and a reinforcement learning algorithm is introduced to enable the random forest model to dynamically adjust the weight and composition of the to-be-inspected items according to the detection results; The risk comprehensive score corresponding to the high-risk item is obtained, and it is judged whether the risk comprehensive score corresponding to the high-risk item is greater than a second threshold value, wherein the second threshold value is less than the first threshold value; If it is judged that the risk comprehensive score corresponding to the high-risk item is greater than the second threshold value, the corresponding high-risk item is determined as a to-be-inspected item.

4. The food fast inspection information standardization pushing method according to claim 3, characterized in that, The step of determining the to-be-inspected items of each food in different regions by screening and evaluating the to-be-inspected items of each food in different regions further comprises: According to the dynamic adaptation of the circulation link, the circulation link at least includes a production link, a storage link and a retail link; According to the special scene triggering the adjustment of the detection item, the special scene at least includes a major event, seasonal change and an emergency.

5. The food fast inspection information standardization pushing method according to claim 4, characterized in that, In the step of preprocessing the data in the multi-dimensional data collection and integration network, and performing feature extraction and feature selection on the preprocessed data to determine the target features, the extracted features at least include food category, raw material composition, production process, historical detection over-standard frequency, safety event occurrence frequency, consumer negative feedback frequency and risk substance toxicity; In the feature selection, the variance selection method is used to retain the features with distinguishing characteristics, and the mutual information method is used to calculate the mutual information value between the features and the adjustment results of the detection items, and the features with the mutual information value meeting the preset value are selected into the random forest model training.

6. The food fast inspection information standardization pushing method according to claim 5, characterized in that, The step of assigning values to each potential detection item from the risk occurrence probability and the risk impact degree, and calculating the risk comprehensive score of each item includes: According to the over-standard frequency in the historical detection and the safety event occurrence rate of similar foods, the risk occurrence probability is calculated; According to the scope of the consumer population, the harm consequences and the social impact, the risk impact degree is calculated; According to the risk occurrence probability and the risk impact degree, the weighted sum is calculated to obtain the risk comprehensive score.

7. A food fast inspection information standardization pushing system, characterized in that, The system for implementing the food fast detection information standardization pushing method according to any one of claims 1-6 comprises: An evaluation module is configured to determine the detection items of different foods in different regions by screening and evaluating the detection items of different foods in different regions, and establish a first mapping relationship between the food information of different foods in different regions and the corresponding detection items; A first determination module is configured to, when food detection is performed, acquire food information according to a unique tag code of the food to be detected, determine the detection items according to the food information and the first mapping relationship, and push the detection items. In the step of determining the detection items of different foods in different regions by screening and evaluating the detection items of different foods in different regions, a hybrid algorithm model of risk weight calculation and machine learning is used to dynamically adjust the detection items.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the food fast detection information standardization pushing method according to any one of claims 1-6.

9. An electronic device, comprising: The computer program stored in the memory and executable on the processor, when the processor executes the program, implements the food fast detection information standardization pushing method according to any one of claims 1-6.

Images