Data association analysis method and device based on food safety event emergency disposal

Abstract

The invention discloses a data association analysis method and device based on food safety event emergency disposal. The method comprises the following steps: S1, acquiring food data; s2, constructing a database linked list according to the food data; s3, obtaining a food safety traceability flow chart according to the association of the food safety links in the database linked list; s4, according to the food safety traceability flow chart, obtaining a food safety traceability information model through Bayesian network learning; and S5, according to the origin of the scene where the food emergency occurs, through the food safety traceability information model, obtaining traceability information of the food safety emergency. By adopting the technical scheme of the invention, the direct influence and potential threats of the food safety emergencies are effectively mastered, and the problem that the emergency disposal efficiency of the food safety emergencies is influenced is avoided.

Description

Technical Field

[0001] The present invention belongs to the technical field of big data correlation analysis, and particularly relates to a data correlation analysis method and device for emergency disposal of food safety incidents. Background Art

[0002] Food safety has received increasing attention. Food safety supervision is a complex task because there are problems such as few supervision resources, a long chain of links, many types of operations, and heavy safety responsibilities in food safety supervision itself. Therefore, to effectively carry out food safety supervision and improve the emergency response ability to food safety emergencies, it is necessary to change the traditional supervision method, use the concept of intelligent supervision to comprehensively analyze and judge food safety emergencies, and provide a scientific basis for the supervision department to formulate effective emergency response measures. The key to intelligent supervision is the mining and utilization of data. To conduct a more scientific and comprehensive analysis and judgment of food safety emergencies, it is necessary to conduct correlation and fusion analysis on various business data related to food safety. The business data involved includes four aspects: First, food safety data itself: mainly refers to planting and breeding data, production and sales data, and catering circulation data; second, food safety supervision data, mainly refers to administrative license data, registration and filing data, supervision and inspection data, sampling and monitoring data, and risk communication data; third, food safety entity data, mainly refers to registration and registration data, credit supervision data, administrative penalty data, law enforcement inspection data, serious violation data, accident handling data, and responsibility interview data; fourth, social supervision information, mainly refers to online public opinion data, complaint and reporting data, social supervision data, and media exposure data. Moreover, these data generally have problems such as inconsistent standards, diverse types, multiple storage methods, and poor comprehensive utilization. At present, there is no correlation analysis technology method that can integrate various links and types of data of food safety risks for emergency disposal of food safety emergencies. Especially when a food safety emergency occurs, it is impossible to scientifically analyze the impact degree, affected range, and action area of the food safety emergency, and it is impossible to effectively support the supervision department to make decisions on emergency disposal of food safety incidents and control and eliminate the impact of food safety emergencies.

[0003] The current application requirements for food safety data are no longer limited to the query requirements of relevant information of a single enterprise, but put forward higher requirements for the correlation between enterprise-related data. Food data includes enterprise data and supervision data. Due to different perspectives on operations, the key information items for data leadership and linking are not the same, and the lack of an effective data correlation analysis method results in the breakage of the key information chain, inability to form a closed loop, and insufficient traceability and impact analysis. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a data correlation analysis method and device based on the emergency disposal of food safety incidents, effectively controlling the direct impact and potential threats of food safety emergencies, and avoiding problems that affect the emergency disposal efficiency of food safety emergencies.

[0005] To achieve the above object, the present invention adopts the following technical solutions: A data correlation analysis method based on the emergency disposal of food safety incidents, comprising: Step S1, obtaining food data; Step S2, constructing a database linked list according to the food data; Step S3, obtaining a food safety traceability flowchart according to the correlation of food safety links in the database linked list; Step S4, obtaining a food safety traceability information model through Bayesian network learning according to the food safety traceability flowchart; Step S5, obtaining the traceability information of food safety emergencies through the food safety traceability information model according to the scene origin of food emergencies.

[0006] Preferably, in step S2, the correlation information of food safety is extracted through semantic recognition technology in the NLP field according to the food data to obtain a database linked list.

[0007] Preferably, in step S3, a food safety traceability flowchart is obtained through an expert system according to the correlation of food safety links in the database linked list.

[0008] Preferably, step S4 includes: According to the food safety traceability flowchart, obtaining the relationship between variables in the food circulation process; Performing Bayesian network learning according to the relationship between variables in the food circulation process to obtain a food safety traceability information model.

[0009] Preferably, step S5 includes: Based on the scene origin of food emergencies, identifying the range of data indicators that may be affected by emergencies through the food safety traceability information model; According to the range of data indicators that may be affected by emergencies, obtaining the traceability information of food safety emergencies through the food safety traceability information model.

[0010] The present invention also provides a data correlation analysis device based on the emergency disposal of food safety incidents, comprising: An acquisition module for acquiring food data; A construction module for constructing a database linked list according to the food data; An extraction module for obtaining a food safety traceability flowchart according to the database linked list; A training module, configured to obtain a food safety traceability information model through Bayesian network learning according to a food safety traceability flowchart. An analysis module, configured to obtain traceability information of a food safety emergency according to the origin of the scenario where the food emergency occurs and through the food safety traceability information model.

[0011] Preferably, a construction module extracts associated information on food safety from food data through semantic recognition technology in the NLP field to obtain a database linked list.

[0012] Preferably, an extraction module obtains a food safety traceability flowchart through an expert system based on the association of food safety links in the database linked list.

[0013] The embodiments of the present invention have the following beneficial effects: The present invention constructs a database linked list according to food data; obtains a food safety traceability flowchart according to the association of food safety links in the database linked list; obtains a food safety traceability information model through Bayesian network learning according to the food safety traceability flowchart; and obtains traceability information of a food safety emergency according to the origin of the scenario where the food emergency occurs and through the food safety traceability information model. By adopting the technical solution of the present invention, the direct impact and potential threats of food safety emergencies can be effectively controlled, and the problem of affecting the emergency response efficiency of food safety emergencies can be avoided. Description of the Drawings

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0015] Figure 1 It is a flowchart of a data association analysis method for food safety event emergency response based on the embodiments of the present invention. Detailed Embodiments

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0017] To make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the drawings and specific embodiments.

[0018] Example 1: As Figure 1 shown, an embodiment of the present invention provides a data association analysis method based on emergency disposal of food safety incidents, including: Step S1, obtaining food data; Step S2, constructing a database linked list according to the food data; Step S3, obtaining a food safety traceability flowchart through an expert system based on the association of food safety links in the database linked list; Step S4, obtaining a food safety traceability information model through Bayesian network learning according to the food safety traceability flowchart; Step S5, obtaining the traceability information of food safety emergencies through the food safety traceability information model according to the scene origin of food emergencies.

[0019] As an implementation manner of the embodiment of the present invention, in step S2, according to the food data, the associated information of food safety is extracted through semantic recognition technology in the NLP field to obtain a database linked list.

[0020] Furthermore, the database linked list is expressed in the form of a data dictionary, and the data dictionary forms a database linked list through the assembly between data tables. Among them, the data tables are generally associated in different dimensions by fields such as [Unified Social Credit Code], [Production Batch Number], [Product Code], [License ID], [Production License Number], and [Unit Name], and can be associated and assembled according to actual needs. The data in the data table is based on the [Production Product Information Table] as the main line, and the upstream and downstream are associated to form a data chain. Among them, in the product production and processing stage, the [Product Batch] is used as the main line of association. If there is data, it can be traced through the product batch code on the single product. When entering the product sales link, the [Commodity Barcode] can be used for upstream and downstream tracing. The alternative associated field is the unified social credit code of each link institution. Among them, the [Unified Social Credit Code of Raw and Auxiliary Material Enterprises] of the raw material provider can be found in the relevant table of the production enterprise; the [Unified Social Credit Code of the Production Enterprise] of the production enterprise can be found in the relevant table of the operating enterprise, so as to ensure the full-process data association. The database linked list formed according to the data dictionary is shown in Table 1.

[0021] Table 1

[0022] As an implementation manner of the embodiment of the present invention, in step S3, an expert system is constructed through the existing food traceability information, and at the same time, a food safety traceability flowchart is obtained through the expert system based on the association of food safety links in the database linked list.

[0023] Firstly, use the relationships of the food safety-related chains in the actual application scenario as the input of the expert system; then, according to the relationships of the food safety-related chains in the actual application scenario, correspond the food safety links to the data tables in the database linked list and extract the key fields of the data tables to generate a food safety flow chart; secondly, find the nodes with the closest correlation in every two unconnected sub-graphs of the flow chart and add connections to make the whole flow chart fully connected; finally, mark the key nodes in the flow chart by means of keyword detection; thus, a food safety traceability flow chart is generated.

[0024] As an implementation manner of an embodiment of the present invention, step S4 includes: According to the food safety traceability flow chart, obtain the relationships between the variables in the food circulation process; According to the relationships between the variables in the food circulation process, perform Bayesian network learning to obtain a food safety traceability information model.

[0025] Furthermore, the relationships between the variables in the food circulation process are as follows: in the food circulation process, there are six types of primary influencing factors and 18 secondary influencing factors, namely Personnel factors: work attitude, knowledge level Environmental factors: operation environment, warehousing environment risk, sales environment risk Technical factors: monitoring technology, detection technology, information technology Management factors: insufficient food safety publicity, inaccurate food supply and demand prediction, accident emergency plan handling, accident compensation ability Equipment factors: transportation tools, inventory isolation measures, warehousing equipment risk Consumer factors: improper eating methods, eating habits, fluke mentality Among them, the 18 secondary influencing factors have two states of strong correlation and weak correlation, and the primary influencing factors have three states of high, medium, and low. Different states represent the characteristics of the influencing factors. The target variable "food safety" represents the quality of food and has three states of high, medium, and low. Environmental and equipment factors have a direct impact on the target variable. Among them, weak correlation is defined as the influence of parameter values on the quality of the product, that is, when the parameter is within the qualified range, it is a weak correlation.

[0026] For example, a set of data on personnel factors is shown in Table 2: Table 2

[0027] For example, a set of data on food quality is shown in Table 3: Table 3

[0028] As an implementation manner of an embodiment of the present invention, step S5 includes: Step S51: Based on the scenario origin of the food emergency, through the food safety traceability information model based on Bayesian network reasoning, identify the range of data indicators that the emergency may affect, specifically including: S511: Determine the scenario origin (event occurrence point) Determine the scenario origin of the food emergency, and use the information of the scenario origin as the input data of the food safety traceability information model. The input data includes first-level factors such as human factors, environmental factors, technical factors, management factors, equipment factors, consumer factors, etc., or second-level factors such as poor working environment, risk in storage environment, degree of food safety publicity, inaccurate food supply and demand prediction, etc.

[0029] S512: Reasoning of the Bayesian network After the scenario origin is determined, the Bayesian network makes inferences based on known conditions. The reasoning process is as follows: 1). Take the specific information of the event occurrence (such as the type, location, time, etc. of the emergency) as the input, and update the relevant nodes in the Bayesian network. For example, assume that a pollution event occurs in a certain production plant area, and the event involves a certain batch of food. At this time, the status of the node "environmental factor" can be set to "low".

[0030] 2). The Bayesian network will make inferences according to the conditional probabilities between the nodes. For example, if the "environmental factor" is set to "low", the possible changes in factors such as "production equipment" or "storage environment" will be calculated, and then the possibility of affecting food quality will be determined.

[0031] 3). Since there are known probability relationships between each node (such as the influence probability of "environmental factor" on "equipment risk"), the Bayesian network calculates the updated probabilities of other relevant nodes (such as "equipment risk", "product quality", etc.) according to the known input.

[0032] S5113: Identify the data indicators that the emergency may affect The final result of the Bayesian network reasoning is to identify all the data indicators that the emergency may affect. According to the size of the conditional probability values, determine the value of each factor. For example, under the condition that the "environmental factor" is given as "low", calculate the probabilities of the food quality being "high", "medium", and "low" respectively, and the one with the largest probability is the value of the target element. Through these reasoning results, the range of food safety risks that the emergency may affect can be identified, so as to provide guidance for emergency handling.

[0033] Step S52. Based on the range of data indicators that may be affected by the emergency, use the food safety traceability information model for Bayesian network reasoning to obtain the traceability information of the food safety emergency, specifically including: S521. Determine the reasoning target and initial data The origin data of the emergency scenario: namely the location, time, production batch, etc. of the event (such as food contamination, quality problems).

[0034] The identified range of data indicators: from the previous Bayesian reasoning, obtain the key indicators that may be affected (such as environmental factors, equipment status, personnel factors, product quality, etc.).

[0035] The origin data of the emergency scenario and the identified range of data indicators are used as inputs for further reasoning about the source, impact path, and subsequent traceability information of the entire event.

[0036] S522. Trace specific information from the database Based on the dependency relationships between variables, deduce all relevant links from the event source (such as the pollution source) to the impact chain. First, determine which link has problems according to the range of data indicators determined by the Bayesian network, and then query the relevant data through fields such as [Unified Social Credit Code], [Production Batch Number], [Product Code], [License ID], [Production License Number], [Unit Name]; if there are problems in the production link, queries can be made through the [Production Product Information Table]; if there are problems in the product sales link, the [Commodity Barcode] can be used for up and down traceability.

[0037] In the embodiment of the present invention, through the combination of an expert system, a database linked list, and a Bayesian network, the traceability information of the food safety emergency is obtained, realizing data correlation analysis based on the emergency disposal of food safety events, providing data support for a scientific and reasonable method for potential impacts of food safety emergencies, and improving the emergency disposal ability of food safety emergencies.

[0038] Embodiment 2: The embodiment of the present invention provides a data correlation analysis device based on the emergency disposal of food safety events, including: An acquisition module, used to acquire food data; A construction module, used to construct a database linked list according to the food data; An extraction module, used to obtain a food safety traceability flow chart according to the database linked list; A training module, used to obtain a food safety traceability information model through Bayesian network learning according to the food safety traceability flow chart; An analysis module, configured to obtain traceability information of a food safety emergency through a food safety traceability information model according to the origin of the scenario where the food emergency occurs.

[0039] As an implementation manner of an embodiment of the present invention, a construction module extracts associated information on food safety from food data through semantic recognition technology in the NLP field to obtain a database linked list.

[0040] As an implementation manner of an embodiment of the present invention, an extraction module obtains a food safety traceability flowchart through an expert system based on the associations of food safety links in the database linked list.

[0041] The above-described embodiments are only descriptions of the preferred embodiments of the present invention and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A data correlation analysis method based on the emergency response to food safety incidents, characterized in that, Including: Step S1, obtaining food data; Step S2, constructing a database linked list according to the food data; Step S3, obtaining a food safety traceability flowchart based on the associations of food safety links in the database linked list; Step S4, obtaining a food safety traceability information model through Bayesian network learning according to the food safety traceability flowchart; Step S5, obtaining the traceability information of food safety emergencies through the food safety traceability information model based on the scene origin of food emergencies.

2. The data correlation analysis method based on the emergency response to food safety incidents according to claim 1, wherein In step S2, the associated information of food safety is extracted through semantic recognition technology in the NLP field according to the food data to obtain a database linked list.

3. The data correlation analysis method based on emergency response to food safety incidents according to claim 2, wherein In step S3, a food safety traceability flowchart is obtained through an expert system based on the associations of food safety links in the database linked list.

4. The data association analysis method based on the emergency response to food safety incidents according to claim 3, wherein Step S4 includes: Obtaining the relationships between variables in the food circulation process according to the food safety traceability flowchart; Performing Bayesian network learning according to the relationships between variables in the food circulation process to obtain a food safety traceability information model.

5. The data correlation analysis method based on the emergency disposal of food safety incidents according to claim 4, wherein, Step S5 includes: Based on the scene origin of food emergencies, identifying the range of data indicators that may be affected by emergencies through the food safety traceability information model; Obtaining the traceability information of food safety emergencies through the food safety traceability information model according to the range of data indicators that may be affected by emergencies.

6. A data correlation analysis device based on emergency response to food safety incidents, characterized in that, Including: An acquisition module for obtaining food data; A construction module for constructing a database linked list according to the food data; An extraction module for obtaining a food safety traceability flowchart according to the database linked list; A training module for obtaining a food safety traceability information model through Bayesian network learning according to the food safety traceability flowchart; An analysis module for obtaining the traceability information of food safety emergencies through the food safety traceability information model based on the scene origin of food emergencies.

7. The data association analysis device based on the emergency response to food safety incidents according to claim 6, wherein, The construction module extracts the associated information of food safety through semantic recognition technology in the NLP field according to the food data to obtain a database linked list.

8. The data association analysis device based on food safety incident emergency handling according to claim 7, wherein The extraction module obtains a food safety traceability flowchart through an expert system based on the associations of food safety links in the database linked list.

Images