A method for establishing a food safety knowledge base

Abstract

The present invention relates to a method for establishing a food safety knowledge base, the method comprising: acquiring knowledge data related to food safety; establishing a food safety knowledge base according to the knowledge data; and determining output recommended knowledge based on the food safety knowledge base and input demand information by performing knowledge reasoning and knowledge synthesis. The present invention acquires knowledge data of a variety of feedback food information, establishes a corresponding food safety knowledge base, uses the query results of the demand information in the food safety knowledge base, performs knowledge reasoning and knowledge synthesis, forms corresponding recommended knowledge, pushes it to users efficiently and quickly, feeds back the food knowledge required by users, and performs dynamic early warning to ensure the efficiency of food management.

Description

Technical Field

[0001] The present invention relates to the field of food information technology, and in particular to a method for establishing a food safety knowledge base. Background Art

[0002] Supervising and regulating food through a variety of intelligent methods and creating the best supervision model for enterprises and governments are the main development directions of food supervision and regulation in the future. The technical means of big data technology are to analyze and process a large amount of data. The formalized and scientific mining and collection of food safety information, the management of enterprises, and the supervision of the government are effective ways to continuously optimize the resource allocation of food enterprises and regulatory departments to improve the safety, standardization and operability of food processing. Establishing a food safety knowledge base and entering the actual products of enterprises into the system in real time to realize data sharing can greatly reduce the work pressure of regulatory departments, alleviate the problem of insufficient number of personnel in food regulatory agencies, enhance the work innovation of regulatory departments under the background of big data, and strengthen the actual work efficiency of food supervision.

[0003] In the Chinese patent CN107680026A “A popular science esophagus system and its implementation method”, the main purpose is to establish an interactive food safety knowledge base for the public, providing the public with food safety knowledge query, food safety problem consultation, food industry chain and food safety, food safety incidents and other content. It mainly popularizes science to the public through mobile phones, and does not involve the establishment, management and application of food safety knowledge base. In the Chinese patent CN111666269A “Food Safety Big Data Automatic Coding and Full Chain Traceability System”, the food data recorded at the industrial chain nodes in the material flow sequence is automatically encoded after identifying the characteristics, and food safety risk assessment and early warning are carried out. Information can be traced according to the coding information, food safety hazard points can be determined, and the sharing of multi-category full chain big data can be realized, but it lacks content such as product information storage, call, and joint use, that is, it lacks a knowledge base for food. Therefore, how to establish a comprehensive food safety knowledge base to effectively supplement the existing food supervision technology is an urgent problem to be solved. Summary of the invention

[0004] In view of this, it is necessary to provide a method for establishing a food safety knowledge base in order to achieve the purpose of establishing and effectively utilizing the food safety knowledge base.

[0005] The present invention provides a method for establishing a food safety knowledge base, comprising:

[0006] Acquire knowledge data related to food safety;

[0007] Establishing a food safety knowledge base based on the knowledge data;

[0008] Based on the food safety knowledge base, knowledge reasoning and knowledge synthesis are performed according to input demand information to determine output recommended knowledge.

[0009] Furthermore, the knowledge data includes first knowledge and second knowledge, the first knowledge and the second knowledge are in the form of definitions, concepts, terms, graphics, videos, tables, and documents, and the acquisition of knowledge data related to food safety includes:

[0010] The first knowledge recorded in books, documents or computer systems is read, wherein the first knowledge includes basic food content information, legal and regulatory standards information, and additive information related to food safety; the second knowledge on the knowledge carrier is read using a crawler method, wherein the second knowledge includes record information and experience information of knowledge personnel during their study, work and research.

[0011] Furthermore, establishing a food safety knowledge base based on the knowledge data includes:

[0012] Extract key knowledge information from the knowledge data, classify the knowledge, and determine the corresponding classification information; establish at least one basic database based on the knowledge data and the corresponding classification information; integrate the at least one basic database into an Excel spreadsheet and upload it to the big data cloud for storage; associate the at least one basic database with knowledge, establish a knowledge network structure, and determine the food safety knowledge base. Extract key knowledge information from the knowledge data, classify the knowledge, and determine the corresponding classification information, including:

[0013] Using clustering method, regression method, correlation method, principal component analysis method, and structured processing method to extract the key information corresponding to the knowledge data;

[0014] Determine, according to the key information, classification information corresponding to the knowledge data, wherein the classification information includes at least one of people, machines, materials, methods, and environment;

[0015] According to the number of occurrences of terms in the knowledge data, the corresponding knowledge feature value is determined and classified into the corresponding knowledge group, wherein the knowledge group includes at least one of food standards, attributes, ingredients, products, and inputs; the knowledge feature value is expressed by the following formula:

[0016] y=a0+a1×x1+…an×xn

[0017] In the above formula, y represents the knowledge feature value, xn represents the nth food attribute term in the knowledge data, and an represents the number of occurrences of the term corresponding to the nth food attribute term.

[0018] Furthermore, the establishing of at least one basic database based on the knowledge data and the corresponding classification information includes:

[0019] Based on the knowledge data, the classification information and the knowledge group, a corresponding two-dimensional table is determined, and the two-dimensional table forms the corresponding basic database; wherein the basic database includes at least one of a food material database, a food database, a catering database, a food production database, a dietary nutrition database, a standard specification database, a nutritional health effect database, a production entity database, an input database, an additive database, and a foodborne disease and health monitoring database.

[0020] Furthermore, the step of establishing at least one basic database based on the knowledge data and the corresponding classification information further comprises:

[0021] Using information mining, data mining is performed on the database information in the basic database to form mining information; using data fusion, data fusion is performed on the database information in the basic database to form fusion information;

[0022] A professional database is established based on the mined information and the fused information, wherein the professional database includes at least one of a food safety risk management database, a hazard traceability database, a supervision management database and a health management database.

[0023] Furthermore, the step of associating knowledge of at least one of the basic databases, establishing a knowledge network structure, and determining the food safety knowledge base includes:

[0024] Functionally associating the database information in the basic database and the professional database through internal keywords, external keywords, and classification information to form corresponding function association information, wherein the function association information includes an equality relationship, a containment relationship, a contained relationship, a linear relationship, or a nonlinear relationship between different database information; determining the association level words corresponding to the different database information according to the function association information, wherein the association level words include hypernyms, hyponyms, synonyms, and related words;

[0025] According to the associated hierarchical words, a knowledge network structure of multiple levels is formed, wherein the knowledge network structure includes a knowledge core, a knowledge main line, knowledge branches and extended knowledge.

[0026] Furthermore, establishing a food safety knowledge base based on the knowledge data further includes:

[0027] Collect various production information from various food production chains, and / or obtain product information circulating in the survey market, and / or obtain food event information related to the food safety knowledge base through crawlers; add the various production information to the Excel table according to the production information and / or the product information and / or the food event information, and update the corresponding basic database; based on the updated basic database, update the food safety knowledge base.

[0028] Furthermore, based on the food safety knowledge base, knowledge reasoning and knowledge synthesis are performed according to the input demand information to determine the output recommended knowledge including:

[0029] According to the input demand information, the food safety knowledge base is queried, starting from the core knowledge points in the constructed knowledge network structure, the corresponding upper-level knowledge, lower-level knowledge, and horizontal-level knowledge are searched. The associated data information associated with the demand information is determined; the associated data information is fused to determine the associated fusion information; and the recommended knowledge is output according to the associated data information and the associated fusion information.

[0030] Further, the associated fusion information includes risk level information, and the performing knowledge fusion on the associated data information to determine the associated fusion information includes:

[0031] Calculate according to the dynamic indicator data in the associated data information to determine the corresponding risk level information; construct a corresponding secondary database according to the risk level information;

[0032] In the secondary database, the text in the risk level information is identified, and different colors are displayed according to different risk level information to provide system warning reminders to reflect different risks.

[0033] Compared with the prior art, the beneficial effects of the present invention include: first, collecting and feeding back a variety of knowledge data on food safety as a multi-faceted source of information; then, establishing the knowledge data as a food safety knowledge base, based on large-scale knowledge, to facilitate subsequent retrieval of information; finally, calling the database knowledge in the food safety big data, performing intelligent association, knowledge fusion and food safety-related risk calculation on the knowledge, so as to determine the final recommended knowledge and feed it back to the user. In summary, the present invention obtains a variety of knowledge data on feedback food information, establishes a corresponding food safety knowledge base, uses the query results of demand information in the food safety knowledge base, performs knowledge reasoning and knowledge synthesis, forms corresponding recommended knowledge, and pushes it to users efficiently and quickly, feeds back the food knowledge required by users, and performs dynamic early warning to ensure the efficiency of food management. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A schematic diagram of a process of an embodiment of a method for establishing a food safety knowledge base provided by the present invention;

[0035] Figure 2 The present invention provides Figure 1 A schematic diagram of a flow chart of an embodiment of step S1 in FIG.

[0036] Figure 3 The present invention provides Figure 1 A schematic diagram of a flow chart of an embodiment of step S2 in FIG.

[0037] Figure 4 The present invention provides Figure 3 A schematic diagram of a flow chart of an embodiment of step S21;

[0038] Figure 5 A schematic diagram of a table header of an embodiment of a standard library provided by the present invention;

[0039] Figure 6 A schematic diagram of a table header of an embodiment of a food attribute library provided by the present invention;

[0040] Figure 7 A schematic diagram of a table header of an embodiment of a food material library provided by the present invention;

[0041] Figure 8 A schematic diagram of a table header of an embodiment of a product library provided by the present invention;

[0042] Fig. 9 A schematic diagram of a table header of an embodiment of a business directory database provided by the present invention;

[0043] Fig.10 A schematic diagram of a table header of an embodiment of an input library provided by the present invention;

[0044] Fig.11 A schematic diagram of a header of an embodiment of an additive library provided by the present invention;

[0045] Fig.12 A schematic diagram of a table header of an embodiment of an equipment library provided by the present invention;

[0046] Fig.13 A schematic diagram of a table header of an embodiment of a food safety incident database provided by the present invention;

[0047] Fig.14 The present invention provides Figure 3 A schematic diagram of a flow chart of an embodiment of step S24;

[0048] Fig.15 A schematic diagram of an association of an embodiment of the knowledge association provided by the present invention;

[0049] Fig.16 The present invention provides Figure 1 A schematic diagram of a process of updating a food safety knowledge base in step S2 in an embodiment;

[0050] Fig.17 The present invention provides Figure 1 A schematic diagram of a flow chart of an embodiment of step S3;

[0051] Fig.18 The present invention provides Figure 7 A schematic diagram of a flow chart of an embodiment of step S32;

[0052] Fig.19 This is a structural schematic diagram of an embodiment of a system for establishing a food safety knowledge base provided by the present invention. DETAILED DESCRIPTION

[0053] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0054] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0055] In the description of the present invention, reference to an "embodiment" means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive of other embodiments. It is explicitly and implicitly understood by those skilled in the art that the described embodiments may be combined with other embodiments.

[0056] The present invention provides a method for establishing a food safety knowledge base, which is applied to the food safety monitoring and management process. Based on the collection of multiple knowledge data, a food safety knowledge base is established, which provides a new idea for further improving food safety supervision. The following are detailed descriptions:

[0057] The embodiment of the present invention provides a method for establishing a food safety knowledge base, combining Figure 1FIG1 is a flow chart of an embodiment of a method for establishing a food safety knowledge base provided by the present invention, comprising steps S1 to S3, wherein: in step S1, knowledge data related to food safety is obtained; wherein the knowledge data includes definitions, concepts, terms, graphics, videos, tables, documents related to food safety; or food quality and safety fragment (island) knowledge (or data) is obtained through dictionaries, lexicons, literature, classrooms, websites, and network searches; food safety knowledge-related data is collected according to people, machines, materials, methods, and environment, mainly including laws and regulations, basic information on food, product information, processing information, and equipment information during production and transportation; in step S2, a food safety knowledge base is established based on the knowledge data; in step S3, based on the food safety knowledge base, knowledge reasoning and knowledge synthesis are performed according to the input demand information to determine the output recommended knowledge.

[0058] In an embodiment of the present invention, first, various knowledge data on food safety are collected and fed back as multi-faceted information sources; then, the knowledge data is established as a food safety knowledge base based on large-scale knowledge to facilitate subsequent retrieval of information; finally, the database knowledge in the food safety big data is called, and the knowledge is intelligently associated, knowledge is integrated, and risk calculations related to food safety are performed to determine the final recommended knowledge and feed it back to the user.

[0059] It should be noted that the technical solution of the present invention mainly includes two aspects, one is the establishment of a food safety knowledge base, and the other is the application of the food safety knowledge base under the conditions of new industrial management and government supervision. Among them, the establishment of a food safety knowledge base includes the collection of knowledge fragments, the classification of fragments, the construction of professional knowledge structure, the combination of knowledge associations, and the improvement of knowledge learning; and the application of the food safety knowledge base is the application of the food knowledge base in food quality and safety risk analysis, risk control, hazard traceability, enterprise production management, market supervision and management, dietary health management, etc. The steps are knowledge demand information input, knowledge reasoning based on the knowledge base, knowledge synthesis, and knowledge output.

[0060] As a preferred embodiment, the knowledge data includes first knowledge and second knowledge, the first knowledge and the second knowledge are in the form of definitions, concepts, terms, graphics, videos, tables, documents, etc. Figure 2 Come and see, Figure 2 The present invention provides Figure 1The flowchart of an embodiment of step S1 in the figure includes steps S11 to S12, wherein: in step S11, the first knowledge recorded in books, documents or stored in a computer system is read, wherein the first knowledge includes basic food content information, legal and regulatory standards information, and additive information related to food safety; in step S12, the second knowledge on the knowledge carrier is read in a crawler manner, wherein the second knowledge includes record information and experience information of knowledge personnel during their study, work and research.

[0061] In the embodiment of the present invention, various knowledge data are obtained through various channels to ensure the richness and diversity of the data, which facilitates the establishment of a subsequent food safety knowledge base.

[0062] It should be noted that the food safety knowledge base requires a large amount of data support. Knowledge fragments include the collection of definitions, concepts, terms, graphics, videos, tables and documents related to food safety. The collection of knowledge is divided into two parts. One part is the knowledge that has been recorded in books, documents or stored in computer systems, including basic food content information, laws, regulations and standards, additives, etc.; the other part is to obtain relevant knowledge about food information from the knowledge carriers of knowledge-based personnel, mainly the records and experiences of knowledge-based personnel during their study, work and research.

[0063] As a preferred embodiment, Figure 3 Come and see, Figure 3 The present invention provides Figure 1 Step S2 is a flow chart of an embodiment, comprising steps S21 to S24, wherein: in step S21, key knowledge information is extracted from the knowledge data, and knowledge classification is performed to determine corresponding classification information; in step S22, at least one basic database is established based on the knowledge data and the corresponding classification information; in step S23, the at least one basic database is integrated into a two-dimensional data table and uploaded to the big data cloud for storage; in step S24, knowledge association is performed on at least one basic database, a knowledge network structure is established, and the food safety knowledge base is determined.

[0064] In the embodiment of the present invention, knowledge classification is performed based on the collected knowledge data, and a corresponding basic database is established, and then the correlation between the knowledge data is determined, and the knowledge network structure is established using the correlation. The two-dimensional data table includes an Excel table.

[0065] It should be noted that the establishment of a knowledge base includes knowledge collection, classification, establishment of professional knowledge, knowledge structure construction, knowledge association combination and knowledge learning improvement. The knowledge in the knowledge base includes facts, concepts and data. Facts are short-term information stored in the database; concepts are knowledge extracted from the common experience of researchers, which can be divided into standards and rules. The former is the condensation of knowledge, and the latter is to guide other personnel on how to derive hypotheses from unknown or known facts; data is the improvement of knowledge, and the knowledge base needs a lot of data support.

[0066] As a preferred embodiment, Figure 4 Come and see, Figure 4 The present invention provides Figure 3 Step S21 in the flowchart of an embodiment includes steps S211 to S213, wherein: in step S211, a clustering method, a regression method, a correlation method, a principal component analysis method, and a structured processing method are used to extract the key information corresponding to the knowledge data; in step S212, based on the key information, the classification information corresponding to the knowledge data is determined, wherein the classification information includes at least one of people, machines, materials, methods, and environments; in step S213, based on the number of occurrences of terms in the knowledge data, the corresponding knowledge feature value is determined and classified into the corresponding knowledge group;

[0067] The knowledge characteristic value is expressed by the following formula:

[0068] y=a0+a1×x1+…an×xn

[0069] In the above formula, y represents the knowledge feature value, xn represents the nth food attribute term in the knowledge data, and an represents the number of occurrences of the term corresponding to the nth food attribute term.

[0070] In the embodiment of the present invention, classification information is used to effectively classify knowledge data, and the number of term occurrences is used to determine the corresponding knowledge group, so as to subsequently form the correlation between data. According to the closeness of the knowledge relationship, it is divided into several knowledge groups, and the existing clustering, regression, correlation, principal component analysis, structured processing and other methods are used to extract key knowledge information, and classified according to the five aspects of people, machines, materials, methods, and environment; at the same time, the knowledge group is determined based on the knowledge characteristic value, and various knowledge about food is divided into food standards, attributes, ingredients, products, inputs, etc.; food is divided into cereals, potatoes, starch, livestock meat, etc. In the knowledge characteristic value formula, y is the characteristic value of food quality and safety terms, xn is the food quality and safety attribute term in the document, an is the coefficient, and is the number of term occurrences. The knowledge characteristic value can also be used as a key control point in the production and processing of food, and there is a detailed classification of knowledge in the process of knowledge organization and structuring.

[0071] As a preferred embodiment, the above-mentioned step S22 specifically includes: based on the knowledge data, the classification information and the knowledge group, determining a corresponding two-dimensional table, and the two-dimensional table forms the corresponding basic database; wherein the basic database includes at least one of a food material database, a food database, a catering database, a food production database, a dietary nutrition database, a standard specification database, a nutritional health effect database, a production entity database, an input database, an additive database, and a foodborne disease and health monitoring database.

[0072] In the embodiment of the present invention, the classified knowledge data is effectively sorted using a basic database to form a variety of basic databases that feed back different information.

[0073] In a specific embodiment of the present invention, a high-dimensional basic database is established based on the collected knowledge fragments and classification information, and the basic database includes several tables, each of which (two-dimensional tables) includes several keywords and several grouping classifications. Professional knowledge includes using the basic database and the production process database (including: inspection and testing database, full-chain circulation database), using information mining and data fusion technology to establish a high-dimensional professional database (food safety risk management database, hazard traceability database, supervision management database, health management database, etc.).

[0074] In a specific embodiment of the present invention, Figure 5 Come and see, Figure 5 The schematic diagram of the header of the embodiment of the standard library provided by the present invention, the basic database includes multiple standard libraries, the standard library is mainly a database established based on the laws and regulations and national standards of various foods, among which the national standards for the detection limits of some substances in food are mainly included, the information of the standard library includes the name of the standard, classification, index name, index type, upper and lower limits and detection method, classification 1, 2, 3 are standard type, standard function, standard range, standard type is national standard, local standard, industry standard; standard function includes detection standard, inspection specification, laws and regulations, planting specification, management specification, coding specification, quality standard, transaction specification, processing specification, hygiene specification, technical specification, sampling specification, control specification, operation specification, service specification; standard range covers all foods. Among them, the import method of the standard library is to collect various laws and regulations on food and national, local and industry standards, condense and mine their important information, and convert unstructured information into structured information and fill it into the standard table database.

[0075] In a specific embodiment of the present invention, Figure 6 Come and see, Figure 6This is a schematic diagram of the header of an embodiment of the food attribute library provided by the present invention. The basic database includes multiple food attribute libraries. The purpose of the food attribute library is to collect the properties of food and form a database after obtaining a large amount of information. Once the information of the enterprise's products is input by the enterprise end or the regulatory end, it can be evaluated whether the food properties are within the normal range and whether some basic properties are reasonable, and quality evaluation can be performed. The food attributes of this system are classified by type: basic properties, physical properties, chemical properties, microbial properties, etc. The basic properties mainly include density and bulk density; physical properties mainly include specific heat, thermal conductivity, boiling point, freezing point, smoke point, viscosity coefficient, dielectric constant, pH, oxidation, reducibility, and thermal stability; microbial properties include the content of Escherichia coli, mold, yeast, and pathogenic bacteria. The pathogenic bacteria mainly include Salmonella, Shigella, and Staphylococcus aureus.

[0076] In a specific embodiment of the present invention, Figure 7 Come and see, Figure 7 The schematic diagram of the header of an embodiment of the food material library provided by the present invention, the basic database includes a variety of food material libraries, the food material library is to collect and summarize some nutritional information of food materials, the basic energy and nutritional components of food processed from different food materials can be obtained by calculation, and provide a reference for the daily nutritional components ingested by the human body. The information collected by the food material library includes the name of the food material, classification, edible part, proportion of edible part, main place of production, main processed products, and main nutritional components; classification 1, 2, and 3 are respectively the type, food type, and eating method. The main nutrients are: protein, fat, carbohydrates, dietary fiber, minerals (sodium, potassium, calcium, iron and zinc) and vitamins (VA, VC and B vitamins).

[0077] In a specific embodiment of the present invention, Figure 8 Come and see, Figure 8 This is a schematic diagram of the header of an embodiment of the product library provided by the present invention. The basic database includes multiple product libraries. Each enterprise processes food materials to obtain pre-packaged or catering products. A database is established based on the information of various products. The main purpose is to establish a data foundation for the traceability system, and to build a complete food traceability system by taking food and food origin as traceable attributes. The information of the enterprise directory library includes enterprise name, classification, social credit code, business license code, registration number, business scope, address, and legal person. Classifications 1, 2, and 3 are type, category, and product type, respectively.

[0078] In a specific embodiment of the present invention, Fig. 9 Come and see, Fig. 9The schematic diagram of the header of an embodiment of the enterprise directory library provided by the present invention, the basic database includes a variety of enterprise directory libraries, the enterprise directory library is a database established with various food and catering companies as the main body, the purpose is to collect the basic information and main business product types of many domestic enterprises, and the product library can play a complementary role. The two databases make the entire food safety knowledge base more complete. The enterprise directory library is a database established with various food and catering companies as the main body. The information of the enterprise directory library includes enterprise name, classification, social credit code, business license code, registration number, business scope, address, legal person, and classifications 1, 2, and 3 are type, category, and product type respectively.

[0079] In a specific embodiment of the present invention, Fig.10 Come and see, Fig.10 This is a schematic diagram of the header of an embodiment of the input library provided by the present invention. The basic database includes multiple input libraries. Inputs are substances used or added in the processing of food, among which agricultural inputs are substances used or added in the production process of agricultural products, including fertilizers, pesticides, veterinary drugs, feed, feed additives and other agricultural production materials. The input library is a database established based on various inputs. The information of each input includes input type, classification, residue, reference standard, harmful effect, maximum usage, main manufacturer, main application unit, and detection method. Classification 1, 2, and 3 are name, classification, and status respectively. The input types include pesticides, feed, fertilizers, veterinary drugs, and newly approved veterinary drugs; the classifications include insecticides, fungicides, and herbicides; and the states include powder and liquid.

[0080] In a specific embodiment of the present invention, Fig.11 Come and see, Fig.11 The schematic diagram of the header of an embodiment of the additive library provided by the present invention, the basic database includes a variety of additive libraries, currently there are 23 categories of food additives in my country, more than 2000 varieties, through the standardized use of food additives, can improve the stability and quality of food, accelerate the pace of development of my country's food industry, but some enterprises are driven by interests, do not use food additives in a standardized manner, thus affecting the safety and quality of food, therefore, additives play a certain role in the relevant departments to play a supervisory function and ensure food quality and safety. The additive library established by this method is a database based on various additives, and the additive library is a database based on various additives. The information of each additive includes the name, classification, production standard, harmful effect, maximum usage, main manufacturer, and detection method of the additive. Classification 1, 2, and 3 are respectively type, classification, and state. The types of additives are expanded: additives, processing aids, natural flavors for food, synthetic flavors for food, nutritional supplements, preservatives, and probiotics; the state of additives is mainly gas, liquid, solid, and powder.

[0081] In a specific embodiment of the present invention, Fig.12 Come and see, Fig.12 This is a schematic diagram of the header of an embodiment of the equipment library provided by the present invention. The basic database includes multiple equipment libraries. The equipment library is composed of equipment information used for food processing and transportation. The information in the equipment library mainly includes equipment name, classification, model, production capacity, specifications, installed power, and technical parameters, among which category, type, and processing node are the three classification bases of the equipment. Classifications 1, 2, and 3 are category, type, and processing node, respectively. The equipment library can search for equipment name, equipment type, and manufacturer, present equipment information, and can download information with one click. This information can be input by each equipment company into the information of the equipment it sells, and then the food processing company can query and download information according to its own needs, and contact the seller to buy and sell equipment.

[0082] In a specific embodiment of the present invention, Fig.13 Come and see, Fig.13 This is a schematic diagram of a table header of an embodiment of a food safety incident database provided by the present invention. The basic database includes multiple food safety incident databases. The food safety incident database is obtained by querying network information and performing structured processing on the information.

[0083] As a preferred embodiment, the above step S22 specifically further includes: using information mining to perform data mining on the database information in the basic database to form mining information; using data fusion to perform data fusion on the database information in the basic database to form fusion information; and establishing a professional database based on the mining information and the fusion information, wherein the professional database includes at least one of a food safety risk management database, a hazard traceability database, a supervision management database, and a health management database. In an embodiment of the present invention, information mining and data fusion technology are used to establish a high-dimensional professional database.

[0084] As a preferred embodiment, the above step S23 specifically includes: the data obtained by various means are integrated into an Excel table, uploaded to the big data cloud for storage, and some information is displayed in the form of pie charts and bar charts to achieve a large-screen visualization, so that the information display is more obvious and the method is richer, and the cloud data can be supplemented and modified later. In the embodiment of the present invention, the knowledge data is effectively stored.

[0085] As a preferred embodiment, Fig.14 Come and see, Fig.14 The present invention provides Figure 3Step S24 in the flowchart of an embodiment includes steps S241 to S243, wherein: in step S241, the database information in the basic database and the professional database is functionally associated through internal keywords, external keywords, and classification information to form corresponding functional association information, wherein the functional association information includes an equal relationship, a containment relationship, a contained relationship, a linear relationship, or a nonlinear relationship between different database information; in step S242, the associated hierarchical words corresponding to different database information are determined according to the functional association information, wherein the associated hierarchical words include hypernyms, hyponyms, synonyms, and related words; in step S243, a knowledge network structure of multiple levels is formed according to the associated hierarchical words, wherein the knowledge network structure includes a knowledge core, a knowledge main line, knowledge branches, and extended knowledge. In the embodiment of the present invention, functional association is used to effectively construct a multi-level knowledge network structure.

[0086] In a specific embodiment of the present invention, Fig.15The schematic diagram of the association of the embodiment of the knowledge association provided by the present invention is that the knowledge base and other food safety databases are interconnected and complementary, and the information of the products can be shared, especially the food safety big data and the food safety knowledge base realize the knowledge association, the logical structure is first established between the knowledge, and then the knowledge is associated according to the logic, the knowledge information is associated together, and the input data is automatically associated and combined according to the logical structure, and presented in the relationship of hyponyms, the same knowledge only appears once, and then connected with other knowledge in the form of association. The knowledge base first needs to construct the knowledge structure, and the collected basic database and the standard professional database are functionally associated through internal or external keywords, classifications, data, etc., to form a knowledge network structure of several levels. The knowledge network level includes database layer, keyword layer, classification layer, and data layer, which mainly completes data transmission services. Functional association includes: equality, inclusion, inclusion, linear or nonlinear weighted calculation (sum, product), equality can be the knowledge content of the same level, inclusion and inclusion can be described by the definition of hyponyms. In addition, the knowledge structure also consists of knowledge core, knowledge main line, knowledge branch and extended knowledge. The knowledge core is the primary knowledge information diffracted from the knowledge center, that is, the knowledge main line logic is linked downward, and the further downward link is the knowledge branch, and the extended knowledge is the supplementary explanation of the knowledge. The correlation coefficient r is used to represent the closeness of the relationship between keywords and knowledge in the knowledge base, so that the correlation coefficients r of the knowledge core, knowledge main line, knowledge branch and extended knowledge are 0.8~1.0, 0.6~0.8, 0.4~0.6, and 0.2~0.4 respectively, so as to calculate the correlation coefficient. The overall logic unfolds the association of data knowledge from three aspects: enterprise management, supervision management and food safety. There is a certain functional relationship between the associations of various knowledge, for example, f1 represents hyponyms and f2 represents hypernyms.

[0087] As a preferred embodiment, Fig.16 Come and see, Fig.16 The present invention provides Figure 1 The flowchart of an embodiment of updating the food safety knowledge base in step S2 is shown in the figure. The above step S2 also includes steps S25 to S27, wherein: in step S25, a variety of production information in a variety of food production chains is collected, and / or product information flowing in the survey market is obtained, and / or food event information related to the food safety knowledge base is obtained by crawling; in step S26, according to the production information and / or the product information and / or the food event information, the various production information is added to the Excel table, and the corresponding basic database is updated; in step S27, based on the updated basic database, the food safety knowledge base is updated. In the embodiment of the present invention, the food safety knowledge base is effectively maintained and updated.

[0088] In a specific embodiment of the present invention, the food safety knowledge base is a huge system data, and the information of food is not static. The information of products produced by various enterprises every day is not exactly the same. It is necessary to update the knowledge base in time and continuously improve the content of the existing database, including supplementing, correcting and appending knowledge. By judging the degree of attention to previous knowledge points, appending and arranging the knowledge base, and dynamically identifying and learning food quality and safety knowledge fragments through the definition or description of previous knowledge content, knowledge learning will continuously generate new knowledge levels, the construction of new knowledge relationships, and new knowledge fragments. The update of knowledge can be done by the enterprise through the data generated by production, the product information is input into Excel with a standard header, and uploaded to the database to complete an update of the enterprise system; it can also be done by the supervisory and management department by investigating the product information flowing in the market, and integrating its information into the system; it can also be done by the crawler language to automatically collect information related to the knowledge base every day, and extract text-related knowledge through information structuring processing and enter it into the knowledge base in time.

[0089] As a preferred embodiment, Fig.17 Come and see, Fig.17 The present invention provides Figure 1 Step S3 in the flowchart of an embodiment, the above step S3 includes steps S31 to S33, wherein: in step S31, according to the input demand information, a query is performed in the food safety knowledge base, starting from the core knowledge points in the constructed knowledge network structure, the corresponding upper-level knowledge, lower-level knowledge, and horizontal-level knowledge are searched. Determine the associated data information associated with the demand information; in step S32, perform knowledge fusion on the associated data information to determine the associated fusion information; in step S33, output the recommended knowledge based on the associated data information and the associated fusion information. In an embodiment of the present invention, based on the food safety knowledge base established above, the recommended knowledge required by the user is obtained quickly and effectively, which facilitates the user's management of food information.

[0090] As a preferred embodiment, Fig.18 Come and see, Fig.18 The present invention provides Figure 7Step S32 in the flowchart of an embodiment, the above step S32 includes steps S321 to S323, wherein: in step S321, the corresponding risk level information is determined according to the dynamic indicator data in the associated data information; in step S322, the corresponding secondary database is constructed according to the risk level information; in step S323, in the secondary database, the text in the risk level information is identified, and different colors are displayed according to different risk level information to provide system warning reminders to reflect different risks. In the embodiment of the present invention, the associated data information is used for calculation to achieve effective warning.

[0091] In a specific embodiment of the present invention, the application of the knowledge base mainly includes food quality and safety risk analysis, risk control, hazard traceability, enterprise production management, market supervision and management, and dietary health management. Through information query and risk assessment, potential food safety incidents can be mined, including foodborne diseases (food poisoning), illegal additives (poisoning), raw materials and product authenticity traceability (intentional or unintentional adulteration) identification and hazard (over-limited substances) identification, or play a role in risk analysis, traceability and supervision. Dietary management can call on the knowledge information of the nutritional characteristics of food in the knowledge base, calculate the nutritional components of the food according to the proportion by inputting the weight of the food, and complete the evaluation of the diet. The steps of knowledge base application are knowledge demand information input, knowledge reasoning based on the knowledge base, knowledge synthesis, and knowledge output;

[0092] The application of the knowledge base can search for keywords such as food attributes, relevant standards, safety management, risk analysis, etc. in the food safety knowledge system (webpage or APP), input structured and unstructured information, including query words, pictures, videos, and short sentences, to obtain relevant knowledge content and obtain various information related to the product.

[0093] Among them, the acquisition of knowledge data information needs to go through a systematic reasoning process, which starts from the core knowledge points, and the logical progression and transmission of the upper, lower and horizontal knowledge of each knowledge point, until the result is output;

[0094] Among them, after inputting information, the system needs to output complete knowledge information, and the output form includes dynamic / static graphs, tables, text descriptions or their combinations of knowledge. Among them, knowledge fusion plays an important role. Knowledge fusion integrates and displays the relevant contents of various databases in the knowledge base, knowledge associations in knowledge base applications and risk calculations. According to the needs of query or knowledge matching, all relevant information is extracted from the knowledge base according to text recognition, and different interfaces are displayed. Different links are established to associate them with the associated positions. The content is selected and merged according to your own requirements. Some of the information needs to be formed into several indicators through certain mathematical calculations to show the level of risk assessment. Then the results are evaluated and output to form a secondary database, and finally the relevant knowledge is displayed. According to the results of the risk level, the system will promptly issue a warning reminder. The levels are black, green, blue, orange, and red, which will be more different according to the risk level. In the secondary database, the system recognizes the text in the risk level and then displays different colors for risk warning.

[0095] The technical solution of the present invention is described in more detail below in conjunction with pork-related food information:

[0096] First, establish an application system for food safety knowledge base. The establishment of the food safety knowledge base system includes two sub-platforms, one is the food safety big data sharing and intelligent supervision cloud platform and the intelligent knowledge center IKC platform. The former is a collection of food safety big data, which includes all the contents of food safety big data. The knowledge base, as one of the contents, collects network information manually, fills in an independent unified header, and finally uploads it to the cloud platform. Nine knowledge databases including laws, regulations and standards, food safety incidents, food attributes, food materials, products, enterprise directories, inputs, additives and processing logistics equipment have been established. As an information platform, the knowledge base includes a large amount of information about food, which can be queried and called and analyzed by other databases. At the same time, it combines data with the intelligent knowledge center IKC platform, and can use the content in the knowledge base to associate information. It is mainly through the association between the hypernyms and hyponyms of each information, with food as the highest level, and then starts to associate food information through hyponyms such as product type, implementation standard, and production producer. When all products are associated downward to a certain extent, they can be associated upward through hypernyms, so that most of the knowledge is associated;

[0097] Second, pork big data collection. Among them, data acquisition is the collection of data mainly adopts the methods of manual network query and crawler query download. When collecting data on fresh pork, crawlers and manual query methods can be used to capture the basic information of fresh pork on multiple community food delivery platforms such as JD.com, Taobao, Meituan, Ele.me, Duoduodaojia, etc., including basic information such as pork parts, pig breeds, breeding methods, certificates of conformity, brands, manufacturers, production dates and sales volumes, as well as information on processing node acquisition and transportation, slaughtering and cutting, packaging and pre-cooling, and warehousing. Procurement and transportation include transportation methods, transportation tools, train / batch numbers, and transportation routes, and receive information including qualifications, quality inspection and testing; slaughter and cutting include equipment models and main technical parameters; packaging and pre-cooling include packaging equipment, packaging and pre-cooling equipment models and main technical parameters; warehousing information includes enterprise information, warehouse information, warehousing quality grade, storage management, quality inspection reports, etc.; data cleaning and completion: the collected data will appear in different types, and it is necessary to remove abnormal and inconsistent data, and complete necessary data, such as adding dates, addresses, tasks and events. For pigs transported by land, the train number should be clearly stated, and the vehicles should be confirmed to be different. The system automatically obtains the time and address to complete the current node information. Data integration: organize different data sources, and uniformly process the information uploaded by sub-nodes such as procurement and transportation, slaughter and cutting, packaging and pre-cooling, and warehousing;

[0098] Third, the update and improvement of pork knowledge starts with the improvement of known lacking knowledge. In the process of constructing the knowledge base, only individual knowledge content of multiple categories may be entered. For example, in the product type of meat products, the knowledge base only has knowledge content related to fresh pork and Yurun ham, lacking knowledge of other meat products. This requires additional knowledge. According to the level of attention to knowledge, that is, the number of searches for entry names and knowledge names by users in the knowledge intelligence center, these will be promptly reflected to the background of the knowledge base, and managers will supplement and add to the knowledge content in a timely manner. As the knowledge base is continuously improved, new entries and knowledge, that is, knowledge fragments, will be generated, which will generate new knowledge levels and logical relationships, and the knowledge base's large screen or knowledge graph will update the knowledge content;

[0099] Fourth, the food data assessment module establishes a quality and safety risk assessment and early warning model to give a risk assessment and early warning result judgment. The specific implementation steps are as follows: the quality and safety risk assessment and early warning model is established according to industry standards and regulations, and is modified according to the model in the patent of "Food Safety Big Data Automatic Coding and Full Chain Traceability System", and the relevant tables are improved, and then the risk assessment is calculated. The relative risk value can be used a is the critical limit of the indicator; b is the actual value of the indicator; c is the daily safe intake; d is the estimated daily intake; f is the correction factor of the safe intake; Fj is the risk index of the jth control point, Pjk is the relative risk of the kth indicator at the jth node; F is the comprehensive risk index, which is obtained by the arithmetic square root of the maximum and minimum values ​​of Fj. Arithmetic square algorithm: relative risk value of a single indicator Node risk index, comprehensive risk index of the entire chain For example, when Pjk≥1, the single indicator is detected as exceeding the standard and unqualified, and when Pjk﹤1, the single indicator is detected as qualified. For example, the total mercury content in fresh pork measured at the storage node is 0.1mg / kg, the key limit of the indicator is 0.2mg / kg, the daily safe intake is 0.04mg / d, the daily intake is 0.05mg / d, and the correction factor f is 1.2, then the relative risk value Pjk is 0.75; if multiple indicators are measured in the storage node, the node risk index needs to be calculated; the maximum and minimum node risk index arithmetic square values ​​obtained from the nodes of acquisition and transportation, slaughter and cutting, packaging and pre-cooling, and storage are used as the comprehensive risk index.

[0100] Fifth, after entering Yurun ham in the intelligent knowledge center, a series of related information about this food will appear, including the attributes, hypernyms and hyponyms of Yurun ham. The attributes of Yurun ham contain many knowledge points, including the nutritional ingredients, product types, processing methods, manufacturers, storage conditions and shelf life of Yurun ham; the hypernym of Yurun ham is meat products, including attributes and product types, and the product types include meat products, beverages, and convenience foods. These product types will continue to lead to a series of products; the hyponyms of Yurun ham are ingredients, including fresh pork and salt. Fresh pork has hyponyms of live pigs, transportation companies, fine processing plants, and testing, which contain a lot of knowledge information about fresh pork. The related information of knowledge can be seen in the knowledge graph, among which fresh pork is the knowledge core, and fine processing plants, transportation companies, testing, and Yurun ham ingredients are the main knowledge lines. The monosodium glutamate in Yurun ham ingredients is all branch knowledge. According to the number of ingredients, the r value of white sugar is 0.4+0.2 / 5=0.44.

[0101] Sixth, knowledge fusion. If a food safety incident occurs, more detailed information needs to be known. The knowledge base can match the relevant information. For example, if the cadmium metal in fresh pork exceeds the standard, the relevant knowledge base will be searched by searching the keywords of fresh pork and cadmium metal, and the information in the standard library, input library, and product library will be searched, and the relevant knowledge will be displayed in separate interfaces. The inspection specifications, management specifications, processing specifications, hygiene specifications, and operation specifications related to fresh pork appear in the interface of the standard library, and the detection methods and requirements of cadmium metal in fresh pork are clearly displayed; the interface of the input library mainly displays the requirements, minimum limits and harmful effects of cadmium metal in various pork foods; the product interface mainly displays the relevant product information, including the manufacturer, and the same batch of products are warned at the same time. According to the risk assessment calculation, the various risk assessment results of fresh pork are classified into the secondary database for example to complete the fusion and merging of knowledge.

[0102] The embodiment of the present invention also provides a system for establishing a food safety knowledge base, combining Fig.19 Come and see, Fig.19 A schematic diagram of a structure of an embodiment of a food safety knowledge base establishment system provided by the present invention, comprising:

[0103] An acquisition unit 1901 is used to acquire knowledge data related to food safety;

[0104] An establishing unit 1902 is used to establish a food safety knowledge base according to the knowledge data;

[0105] The output unit 1903 is used to perform knowledge reasoning and knowledge synthesis based on the food safety knowledge base and the input demand information to determine the output recommended knowledge.

[0106] The more specific implementation methods of each unit of the food safety knowledge base establishment system can refer to the description of the food safety knowledge base establishment method of the present invention, and have similar beneficial effects, which will not be repeated here.

[0107] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the method for establishing a food safety knowledge base as described above is implemented.

[0108] Generally speaking, the computer instructions for implementing the method of the present invention can be carried by any combination of one or more computer-readable storage media. Non-transitory computer-readable storage media can include any computer-readable media except for the signal itself that is temporarily propagating.

[0109] Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or devices, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection with one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or device.

[0110] Computer program code for performing the operation of the present invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages, in particular, Python language suitable for neural network computing and platform frameworks based on TensorFlow, PyTorch, etc. can be used. The program code can be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0111] An embodiment of the present invention further provides a computing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for establishing a food safety knowledge base as described above is implemented.

[0112] The computer-readable storage medium and computing device provided according to the above embodiments of the present invention can be implemented with reference to the contents specifically described in the method for establishing a food safety knowledge base according to the present invention, and have similar beneficial effects as the method for establishing a food safety knowledge base as described above, which will not be repeated here.

[0113] The present invention discloses a method for establishing a food safety knowledge base. First, a variety of knowledge data on food safety are collected and fed back as multi-faceted information sources; then, the knowledge data is established as a food safety knowledge base, based on large-scale knowledge, to facilitate subsequent retrieval of information; finally, the database knowledge in the food safety big data is called, and the knowledge is intelligently associated, knowledge fused, and risk calculations related to food safety are performed to determine the final recommended knowledge and feed it back to the user.

[0114] The technical solution of the present invention establishes a corresponding food safety knowledge base by acquiring knowledge data of various feedback food information, and uses the query results of demand information in the food safety knowledge base to perform knowledge reasoning and knowledge synthesis to form corresponding recommended knowledge, which is pushed to users efficiently and quickly, and the food knowledge required by users is fed back, and dynamic early warning is performed to ensure the efficiency of food management.

[0115] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for establishing a food safety knowledge base, characterized in that: include: Acquire knowledge data related to food safety; Extracting key knowledge information from the knowledge data, classifying the knowledge, and determining corresponding classification information; Establishing at least one basic database based on the knowledge data and the corresponding classification information; Using information mining, data mining is performed on the database information in the basic database to form mining information; Using data fusion, data fusion is performed on the database information in the basic database to form fusion information; Establishing a professional database based on the mined information and the fused information, wherein the professional database includes at least one of a food safety risk management database, a hazard traceability database, a supervision management database, and a health management database; Functionally associating the database information in the basic database and the professional database through internal keywords, external keywords, and classification information to form corresponding function association information; determining the association level words corresponding to different database information according to the function association information; and forming a knowledge network structure of multiple levels according to the association level words to establish a food safety knowledge base; According to the input demand information, query the food safety knowledge base, start from the core knowledge points in the constructed knowledge network structure, search for corresponding upper-level knowledge, lower-level knowledge, and horizontal-level knowledge, and determine the associated data information associated with the demand information; Calculate the corresponding risk level information based on the dynamic indicator data in the associated data information; According to the risk level information, a corresponding secondary database is constructed; In the secondary database, the text in the risk level information is identified, and different colors are displayed according to different risk level information to provide system warning reminders, reflect different risks, and generate associated fusion information; The recommended knowledge is output according to the associated data information and the associated fusion information.

2. The method for establishing a food safety knowledge base according to claim 1, characterized in that: The knowledge data includes first knowledge and second knowledge, the first knowledge and the second knowledge are in the form of definitions, concepts, terms, graphics, videos, tables, and documents, and the acquisition of knowledge data related to food safety includes: Reading the first knowledge recorded in a book, a document or a computer system, wherein the first knowledge includes basic food content information, legal and regulatory standard information, and additive information related to food safety; The second knowledge on the knowledge carrier is read by a crawler method, wherein the second knowledge includes record information and experience information of the knowledge personnel during study, work and research.

3. The method for establishing a food safety knowledge base according to claim 1, characterized in that: The establishment of a food safety knowledge base includes: Integrate the at least one basic database into a two-dimensional data table and upload it to a big data cloud for storage; At least one of the basic databases is knowledge-associated to establish a knowledge network structure and determine the food safety knowledge base.

4. The method for establishing a food safety knowledge base according to claim 3, characterized in that: The extracting key knowledge information from the knowledge data, classifying the knowledge, and determining corresponding classification information includes: Using clustering method, regression method, correlation method, principal component analysis method, and structured processing method to extract the key knowledge information corresponding to the knowledge data; Determine, according to the key knowledge information, classification information corresponding to the knowledge data, wherein the classification information includes at least one of people, machines, materials, methods, and environment; According to the number of occurrences of terms in the knowledge data, corresponding knowledge feature values ​​are determined and classified into corresponding knowledge groups, wherein the knowledge group includes at least one of food standards, attributes, ingredients, products, inputs, etc.

5. The method for establishing a food safety knowledge base according to claim 4, characterized in that: The step of establishing at least one basic database based on the knowledge data and the corresponding classification information includes: Based on the knowledge data, the classification information and the knowledge group, a corresponding two-dimensional table is determined, and the two-dimensional table forms the corresponding basic database; Among them, the basic database includes at least one of a food material database, a food database, a catering database, a food production database, a dietary nutrition database, a standard specification database, a nutritional health effect database, a production entity database, an input database, an additive database, and a foodborne disease and health monitoring database.

6. The method for establishing a food safety knowledge base according to claim 5, characterized in that: The function association information includes an equality relationship, a containment relationship, a contained relationship, a linear relationship or a nonlinear relationship between different database information; The associated hierarchical words include hypernyms, hyponyms, synonyms and related words; The knowledge network structure includes knowledge core, knowledge main line, knowledge branches and extended knowledge.

7. The method for establishing a food safety knowledge base according to claim 3, characterized in that: The step of establishing a food safety knowledge base based on the knowledge data further comprises: Collect various production information in various food production chains, and / or obtain product information flowing in the investigated market, and / or obtain food event information related to the food safety knowledge base through crawlers; According to the production information and / or the product information and / or the food event information, the multiple production information is added to the two-dimensional data table, and the corresponding basic database is updated; Based on the updated basic database, the food safety knowledge base is updated.

Images