CN110688557A - Food safety event-oriented early warning method - Google Patents
Food safety event-oriented early warning method Download PDFInfo
- Publication number
- CN110688557A CN110688557A CN201910898339.0A CN201910898339A CN110688557A CN 110688557 A CN110688557 A CN 110688557A CN 201910898339 A CN201910898339 A CN 201910898339A CN 110688557 A CN110688557 A CN 110688557A
- Authority
- CN
- China
- Prior art keywords
- food safety
- event
- events
- food
- early warning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 235000013305 food Nutrition 0.000 title claims abstract description 246
- 238000000034 method Methods 0.000 title claims abstract description 48
- 239000013598 vector Substances 0.000 claims abstract description 27
- 230000011218 segmentation Effects 0.000 claims abstract description 22
- 238000011161 development Methods 0.000 claims abstract description 11
- 230000018109 developmental process Effects 0.000 claims abstract description 11
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 230000009193 crawling Effects 0.000 claims abstract description 9
- 239000000654 additive Substances 0.000 claims description 15
- 239000000273 veterinary drug Substances 0.000 claims description 13
- 238000001514 detection method Methods 0.000 claims description 11
- 235000013336 milk Nutrition 0.000 claims description 11
- 239000008267 milk Substances 0.000 claims description 11
- 210000004080 milk Anatomy 0.000 claims description 11
- 230000000996 additive effect Effects 0.000 claims description 10
- 229920000877 Melamine resin Polymers 0.000 claims description 9
- 230000036541 health Effects 0.000 claims description 9
- JDSHMPZPIAZGSV-UHFFFAOYSA-N melamine Chemical compound NC1=NC(N)=NC(N)=N1 JDSHMPZPIAZGSV-UHFFFAOYSA-N 0.000 claims description 9
- 239000000843 powder Substances 0.000 claims description 9
- 244000005700 microbiome Species 0.000 claims description 8
- 239000000447 pesticide residue Substances 0.000 claims description 6
- 239000003814 drug Substances 0.000 claims description 5
- 229940079593 drug Drugs 0.000 claims description 5
- 238000007689 inspection Methods 0.000 claims description 5
- 239000003344 environmental pollutant Substances 0.000 claims description 4
- 238000002474 experimental method Methods 0.000 claims description 4
- 231100000719 pollutant Toxicity 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 3
- 239000002778 food additive Substances 0.000 claims description 2
- 235000013373 food additive Nutrition 0.000 claims description 2
- 238000012797 qualification Methods 0.000 claims description 2
- 238000009877 rendering Methods 0.000 abstract description 9
- 230000007246 mechanism Effects 0.000 abstract description 7
- 238000012545 processing Methods 0.000 abstract description 4
- 238000009792 diffusion process Methods 0.000 abstract 1
- 239000000575 pesticide Substances 0.000 description 8
- 235000015219 food category Nutrition 0.000 description 7
- 235000005911 diet Nutrition 0.000 description 5
- 239000004599 antimicrobial Substances 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000000378 dietary effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 239000004278 EU approved seasoning Substances 0.000 description 2
- BUGBHKTXTAQXES-UHFFFAOYSA-N Selenium Chemical compound [Se] BUGBHKTXTAQXES-UHFFFAOYSA-N 0.000 description 2
- 230000000895 acaricidal effect Effects 0.000 description 2
- 239000000642 acaricide Substances 0.000 description 2
- 230000000844 anti-bacterial effect Effects 0.000 description 2
- 230000002141 anti-parasite Effects 0.000 description 2
- 230000002421 anti-septic effect Effects 0.000 description 2
- 239000003963 antioxidant agent Substances 0.000 description 2
- 239000003096 antiparasitic agent Substances 0.000 description 2
- 229940125687 antiparasitic agent Drugs 0.000 description 2
- 229940064004 antiseptic throat preparations Drugs 0.000 description 2
- 239000003899 bactericide agent Substances 0.000 description 2
- 235000013361 beverage Nutrition 0.000 description 2
- 235000013339 cereals Nutrition 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 235000013365 dairy product Nutrition 0.000 description 2
- 239000000645 desinfectant Substances 0.000 description 2
- 235000011194 food seasoning agent Nutrition 0.000 description 2
- 235000003599 food sweetener Nutrition 0.000 description 2
- 230000002363 herbicidal effect Effects 0.000 description 2
- 239000004009 herbicide Substances 0.000 description 2
- 239000002917 insecticide Substances 0.000 description 2
- 235000013372 meat Nutrition 0.000 description 2
- 235000013622 meat product Nutrition 0.000 description 2
- 210000000653 nervous system Anatomy 0.000 description 2
- 239000003755 preservative agent Substances 0.000 description 2
- 239000003128 rodenticide Substances 0.000 description 2
- 229910052711 selenium Inorganic materials 0.000 description 2
- 239000011669 selenium Substances 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 235000013555 soy sauce Nutrition 0.000 description 2
- 239000003765 sweetening agent Substances 0.000 description 2
- 239000002562 thickening agent Substances 0.000 description 2
- 235000021419 vinegar Nutrition 0.000 description 2
- 239000000052 vinegar Substances 0.000 description 2
- 230000009278 visceral effect Effects 0.000 description 2
- 206010016952 Food poisoning Diseases 0.000 description 1
- 208000019331 Foodborne disease Diseases 0.000 description 1
- 238000012369 In process control Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000009395 breeding Methods 0.000 description 1
- 230000001488 breeding effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000037213 diet Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010965 in-process control Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 230000036186 satiety Effects 0.000 description 1
- 235000019627 satiety Nutrition 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Educational Administration (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Computer Security & Cryptography (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention relates to a food safety event oriented early warning method, which comprises the following steps: crawling data of food safety events of 2000 years to the present from the internet using a crawler mechanism of dynamic rendering and incremental updating; constructing and loading a dictionary special for the food safety field; preprocessing by removing repetition, removing noise, segmenting words by utilizing a word segmentation tool, removing stop words and performing part-of-speech tagging; generating word vectors by the preprocessed text features, and judging whether the words are food safety events or not; determining a time type of the food safety event according to the development characteristics; analyzing and evaluating the event, checking the level of the event, and early warning; analyzing the cause of the food safety event, judging the harmfulness to other foods and issuing early warning information. The method comprises the steps of analyzing and processing food safety events issued on a network in time, constructing an early warning model according to results, predicting the damage degree of the events through early warning of the food safety events, taking corresponding measures in time to prevent large-scale diffusion and preventing events with damage trends from occurring.
Description
Technical Field
The invention relates to the fields of artificial intelligence, computer technology and food safety, in particular to a food safety event-oriented early warning method.
Background
With the development of society and economy, people have more and more abundant lives, and for people who have solved the problem of satiety, the eating health is more important, so that the food safety becomes one of the most concerned problems. But in order to achieve commercial benefits, merchants do not choose means to make illegal things, food safety events occur sometimes, and how to enable consumers to know detailed conditions in time and prevent and control the occurring food safety events provides basis for food safety problem management, control and decision making.
The existing food safety early warning mainly comprises two types: pre-preventive and in-process control. Due to the lack of open link data and open knowledge bases, detection data of each step of planting (breeding), processing, packaging, storing, transporting, selling and consuming after food generation cannot be directly acquired, so that the prior prevention is very difficult. In order to solve these problems, researchers have adopted a method combining physical and virtual methods to search existing data from various aspects so as to accurately warn.
Due to the rapid development of the internet, the data of food safety events can be acquired easily and timely. The preprocessing stage after data acquisition is also researched by predecessors, and all that needs to be done is to change the preprocessing stage according to the self condition. The difficulty is how to design and analyze the cause of the occurred food safety event to early warn the possible food safety event.
Therefore, the current food safety early warning method is lack of an early warning method which can classify the occurred food safety events, early warn according to categories, analyze causes of the food safety events and influence other foods.
Disclosure of Invention
In order to solve the problem of early warning of food safety, the invention provides an early warning method for analyzing and judging the properties (potential safety hazard type events, single type events, expandable type events and frequent type events) and causes (additives exceed standard, pesticide residues, veterinary drug residues, food-borne microorganisms exceed standard, pollutants exceed standard and the like) of a food safety event and possibly influencing other foods, the early warning method comprises the following steps:
the food safety event is crawled from each large portal website on the Internet in time by using a crawler mechanism with dynamic rendering and incremental updating, the occurrence time of the event is 2000 years to the present (namely 2019), and the data format comprises structured data (detection data) and unstructured data (news text data). The dynamic rendering refers to the simulation click of a webpage needing to click the next page and other operations by using a Selenium tool; the incremental update means that only the updated event of the current day is crawled, and the crawl is not repeated for the crawled event.
The food safety event constructs a dictionary special for the food safety field, and comprises proper nouns in the fields of food safety events, food enterprise information, pesticides, veterinary drugs and food additives. The dictionary of the food safety events is constructed based on a large amount of knowledge about the food safety field, and is a tool for correctly identifying words in the field when the words are segmented, so that errors in segmentation are prevented.
Preprocessing according to the food safety event and the special dictionary in the food safety field, wherein the preprocessing comprises the operations of removing duplication, removing noise, utilizing an open source word segmentation tool jieba to perform word segmentation, removing stop words, labeling parts of speech and the like;
performing mathematical representation on the words in the food safety event text according to the words, generating word vectors by using a one-hot model, wherein each element in the vectors is associated with one word in a word stock, the corresponding element in the vectors is set to be 1, and other elements are set to be 0;
judging whether the food safety event is a food safety event according to the judgment, wherein if the food safety event is structured data (detection data), judging whether the food safety event is the food safety event according to the characteristics of qualified and unqualified detection results and the like; and if the data is unstructured data (news text data), matching keywords in the title and the text by adopting a regular matching method, and judging whether the data is a food safety event or not according to the characteristics of keyword-to-word vectors, word frequency and the like. The regular matching method is to match the title and text and check whether words such as food names, additive names, microorganism names, etc. are included and the occurrence frequency (i.e., word frequency) is included.
And dividing related food safety events into four types of events such as potential safety hazard type events, single type events, expandable type events and frequency type events according to the development characteristics of the related food safety events.
And analyzing and evaluating the event according to the law of the health administration department, verifying the level of the event, and issuing notice, I-level, II-level, III-level and IV-level early warning information to the judgment result of the food safety event.
And judging the harmfulness to other foods and issuing early warning information according to the cause analysis of the food safety events. The cause of the food safety event refers to the cause (veterinary drug residue, pesticide residue, overproof additive, overproof food-borne microorganism, overproof pollutant and the like) of the food safety event.
Preferably, the acquisition of the food safety event is derived from a food safety news event crawled from the internet, and the time of occurrence of the acquisition event ranges from 2000 years to the present (i.e., 2019 years);
preferably, the dictionary in the food safety field includes dictionaries of food safety incidents, food enterprise information, food categories, pesticides, veterinary drugs, additives and the like. The food safety event dictionary describes food safety events based on a word-level and corpus combination method, and comprises original characteristics (time, place, cause, victim, scale, harm degree, food name, food category and the like) of each food safety event; the food enterprise information is divided according to administrative regions, and each enterprise information is contained under each administrative region; the food category dictionary is used for special dietary food, milk and dairy products, beverages, grains and products thereof, meat and meat products and the like; the pesticide dictionary is divided into insecticide, acaricide, herbicide, rodenticide, bactericide and the like; veterinary drugs are classified into disinfectant antiseptics, antimicrobials, synthetic antimicrobials, antiparasitics, drugs acting on the visceral system, drugs acting on the nervous system, and the like; the additives are further classified into bleaches, antioxidants, colorants, sweeteners, thickeners, preservatives, and the like. The specific names of the dictionaries such as food enterprise information, pesticide, veterinary drug and the like belong to the seasonings under the food category dictionary, such as soy sauce, vinegar and the like.
Preferably, the preprocessing method of the food safety event comprises the operations of removing duplication and noise of the document, performing word segmentation by using an open source word segmentation tool jieba, removing stop words, labeling parts of speech and the like. The stop words refer to certain words or characters that are filtered before or after processing the text data for efficiency. The part-of-speech tagging refers to performing part-of-speech tagging on a divided word, such as nouns, verbs, adjectives and the like.
Preferably, the word vector representation of the food safety event is that a word vector is generated by using a one-hot model, each element in the vector is associated with a word in a word stock, the corresponding element in the vector is set to be 1, and other elements are set to be 0;
preferably, the determination of whether the food safety event is a food safety event means that if the food safety event is structured data (detection data), the food safety event is determined according to the characteristics of qualified and unqualified detection results and the like; and if the data is unstructured data (news text data), matching keywords in the title and the text by adopting a regular matching method, and judging whether the data is a food safety event or not according to the characteristics of keyword-to-word vectors, word frequency and the like. The regular matching method is to match the title and text and check whether words such as food names, additive names, microorganism names, etc. are included and the occurrence frequency (i.e., word frequency) is included.
Preferably, the related food safety events are classified into safety hazard type events, singleton type events, expandable type events and frequency type events according to their development characteristics. The potential safety hazard events refer to sampling inspection results of related departments of the country and the like, and are unqualified but do not cause certain consequences. The single event is an event of primary school department of Sichuan experiment school, food in school canteens is going to go moldy and go bad, and other school canteens can not do the same according to moral and legal requirements. The extensible event refers to the event of milk powder of three-deer, and aims at the problem of adding melamine, the melamine is not only added by a factory of one three-deer, but also added by milk powder produced by other factories. The frequent event refers to an event which does not occur for the first time and has occurred before.
Preferably, the event is analyzed and evaluated according to laws of the health administration department, the level of the event is determined, and the judgment result of the food safety event is issued with notice, early warning information of level I (mild), level II (moderate), level III (severe) and level IV (warning).
Preferably, the cause analysis of the food safety event judges the damage type of other foods and issues early warning information. The cause of the food safety event refers to the cause (veterinary drug residue, pesticide residue, overproof additives and the like) causing the food safety event. And then judging whether the food safety events can cause the occurrence possibility of other food safety events in other types of foods, and then issuing early warning information to the food safety events which are possible to occur. The early warning information is different from the five types of early warning information and is used for corpus early warning. I.e. to issue a text message that an additive may have a certain effect in a food product.
The invention provides a food safety event-oriented early warning method, which is characterized in that a crawler mechanism with dynamic rendering and incremental updating is used for crawling data about food safety events from 2000 to 2019 on the Internet; constructing a dictionary special for the food safety field; preprocessing the acquired data by removing the weight, removing the noise, performing word segmentation by using an open source word segmentation tool jieba, removing stop words, performing part-of-speech tagging and the like; vectorizing the characteristics of the preprocessed text to generate a word vector, and judging whether the word vector is a food safety event; dividing related food safety events into potential safety hazard type events, single type events, expandable type events and frequent type events according to development characteristics of the food safety events; analyzing and evaluating the event according to the law of the health administration department, verifying the level of the event, and issuing notice, I-level, II-level, III-level and IV-level early warning information to the judgment result of the food safety event so that the government-related department can quickly manage, control and process the event; and (3) aiming at the cause analysis of food safety events, judging the harmfulness to other foods and issuing early warning information.
By means of the method, classified early warning and range early warning can be effectively carried out on food safety through the steps of obtaining data, judging the nature of a food safety event and analyzing the cause of the food safety event, and the effect of the cause of the food safety event on other foods can be prevented, so that a good method is provided for classified early warning and range early warning on food safety and preventing the factors of the occurred food safety event from further causing other food safety events.
Drawings
Fig. 1 is a schematic flow chart of a food safety event-oriented early warning method according to a preferred embodiment of the present invention;
fig. 2 is a schematic diagram illustrating food safety event analysis of a food safety event-oriented early warning method according to a preferred embodiment of the present invention;
fig. 3 is a schematic diagram illustrating classification of food safety events according to a method for early warning of food safety events according to a preferred embodiment of the present invention;
fig. 4 is a schematic diagram illustrating classification and early warning of food safety events according to an early warning method for food safety events in accordance with a preferred embodiment of the present invention;
fig. 5 is a schematic diagram of cause warning analysis of a food safety event-oriented warning method according to a preferred embodiment of the present invention;
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
Fig. 1 is a schematic flow chart of a food safety event-oriented early warning method according to a preferred embodiment of the present invention, and as shown in fig. 1, the present invention provides a food safety event-oriented early warning method, including:
s101, crawling food safety events from the Internet.
In this embodiment, the food safety event crawled from the internet in time by using a crawler mechanism with dynamic rendering and incremental updating is a food safety event crawled from a website of each large portal, and the authenticity of the event is guaranteed because many readers have large influence. The dynamic rendering refers to the simulation click of a webpage needing to click the next page and other operations by using a Selenium tool; the incremental update refers to only crawling updated events, and crawling is not repeated for the crawled events.
For example, taking Xinhua network as an example, a 'dietary word theory' link can be opened, a red blacklist is arranged in the link, a plurality of food safety events are published in the list in real time, and a crawler mechanism can be designed to crawl the food safety events to be stored in txt. The crawler mechanism is used for crawling data by using a dynamic rendering and incremental updating method, wherein the dynamic rendering is to simulate a mouse to click a display more button to load more news, the incremental updating is based on a time sequence, the deadline of a crawling event is recorded, and a newly issued food safety event is crawled from the last time of interception.
S102, constructing a dictionary in the field of food safety.
In this embodiment, the dictionary in the food safety field refers to a set of words and related information related to the food safety field. Such as dictionaries of food safety incidents, food enterprise information, food categories, pesticide names, additive names, etc. The food safety event dictionary describes food safety events based on a word-level and corpus combined method, and comprises original characteristics (time, place, cause, victim, scale, prohibited item involved, degree of harm, food name, food belonging category and the like) of each food safety event; the food enterprise information is divided according to administrative regions, and each enterprise information is contained under each administrative region; the food category dictionary is also special dietary food, milk and dairy products, beverages, grains and products thereof, meat and meat products and the like; the pesticide dictionary is divided into insecticide, acaricide, herbicide, rodenticide, bactericide and the like; veterinary drugs are classified into disinfectant antiseptics, antimicrobials, synthetic antimicrobials, antiparasitics, drugs acting on the visceral system, drugs acting on the nervous system, and the like; the additives are further classified into bleaches, antioxidants, colorants, sweeteners, thickeners, preservatives, and the like. The specific names of the dictionaries such as food enterprise information, pesticide, veterinary drug and the like belong to the seasonings under the food category dictionary, such as soy sauce, vinegar and the like. The food safety event dictionary is constructed based on a large number of food safety events, the existing dictionaries can be found in pesticide dictionaries, veterinary drug dictionaries and the like, and the dictionaries are an effective tool for correctly identifying words in the field when used for word segmentation, so that word segmentation errors are prevented. Figure 2 is an analytical schematic of a food safety event.
S103, preprocessing the data through operations of removing duplication, removing noise, performing word segmentation by utilizing an open source word segmentation tool jieba and the like
In this embodiment, the data is preprocessed by removing the repetition, removing the noise, performing word segmentation by using the open source word segmentation tool jieba, removing stop words, performing part-of-speech tagging, and the like, so that the data is normalized. The stop words refer to certain words or characters that are filtered before or after processing the text data for efficiency. The part-of-speech tagging refers to part-of-speech tagging of nouns, verbs, adjectives and the like after word segmentation.
And S104, vectorizing the processed text features to generate word vectors, and judging whether the text is a food safety event or not.
In this embodiment, the words in the food safety event text are represented mathematically, a one-hot model is used to generate a word vector, each element in the vector is associated with a word in the thesaurus, the corresponding element in the vector is set to 1, and the other elements are set to 0. The judgment of whether the event is a food safety event refers to that whether the event is the food safety event is judged according to the characteristics of qualification, unqualified detection results and the like if the event is structured data (detection data); and if the data is unstructured data (news text data), matching keywords in the title and the text by adopting a regular matching method, and judging whether the data is a food safety event or not according to the characteristics of keyword-to-word vectors, word frequency and the like. The regular matching method is to match the title and text and check whether words such as food names, additive names, microorganism names, etc. are included and the occurrence frequency (i.e., word frequency) is included.
And S105, dividing related food safety events into potential safety hazard type events, single type events, extensible type events and frequent events according to development characteristics of the related food safety events.
In this embodiment, the related food safety events are classified into a hidden danger type event, a single type event, an expandable type event, and a frequent event according to their development characteristics. The potential safety hazard events refer to sampling inspection results of related departments of the country and the like, and are unqualified but do not cause certain consequences. The single event is an event of primary school department of Sichuan experiment school, food in school canteens is going to go moldy and go bad, and other school canteens can not do the same according to moral and legal requirements. The extensible event refers to the event of milk powder of three-deer, and aims at the problem of adding melamine, the melamine is not only added by a factory of one three-deer, but also added by milk powder produced by other factories. The frequent event refers to an event which does not occur for the first time and has occurred before. As shown in fig. 3. The method specifically comprises the following steps:
acquiring structured and unstructured food safety events through a crawler, determining the food safety events which are unqualified to be acquired and are not multiple-occurrence-period-determined to be potential safety hazard type events in the acquired event library, and issuing a notice; for the obtained unstructured food safety events, food safety accidents are caused, but the individual improper behaviors (mostly non-conforming to the sanitary inspection standards) are determined to be single events due to judgment; and the obtained food safety event is determined to be an expandable event due to expandable type judgment (such as additive standard exceeding, veterinary drug, pesticide residue standard exceeding, food-borne microorganism standard exceeding and the like); and if the obtained food safety events are more than 3 times in the food safety event library and the interval time is less than 5 years, determining the events as frequent events. For example, the melamine overproof milk powder event with three degrees of weight like 'borrow soul' is not only a frequent event but also an expandable event.
And S106, analyzing and evaluating the event according to laws of the health administration department, verifying the level of the event, and issuing notice, I-level, II-level, III-level and IV-level early warning information to the judgment result of the food safety event.
In this embodiment, an early warning level is set and an early warning is performed according to the determination result of the food safety event. The early warning level is divided into five early warning levels of notification, I level (mild), II level (moderate), III level (severe) and IV level (warning). The early warning information of the I level, the II level, the III level and the IV level is determined according to the division of related departments of the state on food safety accidents, and the early warning information is notified based on the method. The notification early warning information is used for warning national notifications of events which have potential safety hazards but do not cause accidents; the I-level early warning information indicates that the food safety event meets one of the following conditions:
1. the accident influence range relates to more than 2 villages and towns in a county-level administrative region, and serious harm is brought to public diet safety;
2. the number of the injured people is more than 30 and less than 100, and no death cases occur;
3. other general food safety incidents recognized by county level governments.
The level II early warning information indicates that the food safety event meets one of the following conditions:
1. the accident influence range relates to more than 2 county-level administrative regions in the city-level administrative region, and serious harm is brought to the dietary safety of people;
2. causing more than 100 people to be injured or death cases to occur;
3. other major food safety incidents recognized by the municipality level government;
the grade III early warning information indicates that the food safety event meets one of the following conditions:
1. the accident harm is serious, and the influence range relates to more than two city-level administrative areas in province;
2. more than 100 people are injured and death cases occur;
3. cause more than 10 cases of death;
4. food poisoning accidents happen in schools, and more than 50 people are injured;
5. more than 50 people are injured in national or regional major activities and important meetings;
6. other major food safety accidents recognized by provincial government
The IV-level early warning information indicates that the food safety event meets one of the following conditions:
1. the accident hazard range spans provincial administrative districts and tends to be further expanded;
2. out of provincial disposal range;
3. other major food safety accidents identified by the state department or the state department authority;
if the potential safety hazard event is a single event, an expandable event or a frequent event, the potential safety hazard event is announced to the whole country, the level of the event is determined according to the analysis and evaluation of the accident by the law of the health administration department, the early warning is sent out corresponding to the early warning level, the potential safety hazard event is announced to the whole country, the people in the whole country supervise the event and prevent the people from being injured, and the relevant government departments quickly manage, control and process the event. As shown in fig. 4.
And S107, analyzing the cause of the food safety event, judging the harmfulness to other foods and issuing early warning information.
In this embodiment, the cause analysis of the food safety event determines the damage type to other foods and issues warning information. The cause of the food safety event refers to the cause (veterinary drug residue, pesticide residue, overproof additive, overproof food-borne microorganism, overproof pollutant and the like) of the food safety event. And then judging whether the food safety events can cause the occurrence possibility of other food safety events in other types of foods, and then issuing early warning information to the food safety events which are possible to occur. The early warning information is different from the five types of early warning information and is used for corpus early warning. I.e. to issue a text message that an additive may have a certain effect in a food product. As shown in fig. 5.
The invention provides a food safety event-oriented early warning method, which is characterized in that a crawler mechanism with dynamic rendering and incremental updating is used for crawling data related to food safety events from 2000 years to the present, namely 2019 years, from the Internet; constructing and loading a dictionary special for the food safety field; preprocessing the acquired data by removing the weight, removing the noise, performing word segmentation by using an open source word segmentation tool jieba, removing stop words, performing part-of-speech tagging and the like; vectorizing the preprocessed text features to generate word vectors; dividing related food safety events into potential safety hazard type events, single type events, expandable type events and frequent type events according to development characteristics of the food safety events; analyzing and evaluating the event according to law of health administration departments, verifying the level of the event, and issuing notice, I-level, II-level, III-level and IV-level early warning information to the judgment result of the food safety event; analyzing the cause of the food safety event, judging the harmfulness to other foods and issuing early warning information. Through reclassification of food safety events, the early warning information is expanded to the early warning information release, namely, the food safety event contains the warning to the food enterprises, and the food enterprises are urged to seriously fulfill the responsibility of the enterprises and serve as food safety employees; and secondly, the method plays a role of supervision and leading of government functional departments, and enables the government related departments to rapidly manage, control and process the events after the events occur. The double pipes are arranged simultaneously, the food quality safety is ensured,
finally, the method of the present invention is only a preferred embodiment and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. A food safety event-oriented early warning method is characterized by comprising the following steps:
crawling food safety events from various main portal websites on the Internet, wherein the occurrence time range of the events is 2000 years to the present (namely 2019 years);
and constructing a special dictionary in the food safety field according to the food safety event, wherein the special dictionary comprises proper nouns in the food safety event, food enterprise information, proper nouns in the fields of agriculture, veterinary medicines and food additives and the like. The dictionary of the food safety events is constructed based on a large amount of knowledge about the food safety field, and is a tool for correctly identifying words in the field when the words are segmented, so that errors in segmentation are prevented.
The food safety event is loaded with a dictionary special for the food safety field for preprocessing, and the steps comprise the operations of removing duplication, removing noise, utilizing an open-source word segmentation tool jieba to segment words, removing stop words, labeling parts of speech and the like;
carrying out numerical representation on the words in the food safety event text according to the words, generating word vectors by using a one-hot model, wherein each element in the vectors is related to one word in a word stock, the corresponding element in the vectors is set to be 1, and other elements are set to be 0;
judging whether the food safety event is a food safety event according to the judgment, wherein if the food safety event is structured data (detection data), judging whether the food safety event is the food safety event according to the characteristics of qualified and unqualified detection results and the like; and if the data is unstructured data (news text data), matching keywords in the title and the text by adopting a regular matching method, and judging whether the data is a food safety event or not according to the characteristics of keyword-to-word vectors, word frequency and the like.
And dividing related food safety events into potential safety hazard type events, single type events, expandable type events and frequency type events according to the development characteristics of the related food safety events. The potential safety hazard events refer to sampling inspection results of related departments of the country and the like, and are unqualified but do not cause certain consequences. The single event is an event of primary school department of Sichuan experiment school, food in school canteens is going to go moldy and go bad, and other school canteens can not do the same according to moral and legal requirements. The extensible event refers to the event of milk powder of three-deer, and aims at the problem of adding melamine, the melamine is not only added by a factory of one three-deer, but also added by milk powder produced by other factories. The frequent event refers to an event which does not occur for the first time and has occurred before.
And analyzing and evaluating the event according to the law of the health administration department, verifying the level of the event, and issuing notice, I-level, II-level, III-level and IV-level early warning information to the judgment result of the food safety event.
And judging the hazard type of other foods and issuing early warning information according to the cause analysis of the food safety events. The cause of the food safety event refers to the cause (veterinary drug residue, pesticide residue, overproof additive, overproof food-borne microorganism, overproof pollutant and the like) of the food safety event. And analyzing the cause to judge whether other food safety events can be caused in other types of food, and then issuing early warning information to the food safety events which are possible to occur.
2. The method as claimed in claim 1, wherein the food safety event data is derived from announcements issued by government agencies, news information issued by each provincial and municipal food safety website, news events issued by food safety columns in each news website, and the like.
3. The food safety event-oriented early warning method according to claim 1, wherein a food safety domain-specific dictionary is constructed according to the food safety event, the dictionary comprises food safety domain professional knowledge, and keywords which are most frequently used for describing the food safety event are extracted from a large number of food safety events, and the food safety domain professional dictionary is constructed according to an existing food safety domain dictionary and the downloaded food safety event.
4. The food safety event-oriented early warning method as claimed in claim 1, wherein the data is preprocessed through de-duplication, de-noising, word segmentation and other operations, de-duplication is performed according to a built special dictionary for the food safety field, word segmentation is performed by using an open source word segmentation tool, and then word de-stop and part-of-speech tagging are performed.
5. The food safety event-oriented early warning method according to claim 1, wherein the vectorization of the processed text features to generate a word vector means that words in the text of the food safety event are represented numerically, a one-hot model is used to generate the word vector, each element in the vector is associated with a word in a word bank, the corresponding element in the vector is set to 1, and the other elements are set to 0; judging whether the event is a food safety event or not, namely judging whether the event is the food safety event or not according to the characteristics of qualification, unqualified detection results and the like if the event is structured data; and if the data is unstructured data, matching keywords in the title and the text by adopting a regular matching method, and judging whether the data is a food safety event or not according to the characteristics of word-to-word vectors, word frequency and the like of the keywords.
6. The food safety event-oriented early warning method according to claim 1, wherein the related food safety events are classified into potential safety hazard type events, single type events, expandable type events and frequent type events according to development characteristics, the potential safety hazard events refer to sampling inspection results of related departments of the country and the like, and the events which are unqualified but do not cause certain consequences, the single events refer to events of primary schools and schools of experiment in Sichuan, food in canteens is going to go moldy and go bad and is used, that is, according to moral and legal requirements, other schools and canteens do not do the same type of events, the expandable events refer to milk powder events of sanlu, aiming at the problem of melamine addition, the expandable events are not only added by one manufacturer of sanlu, and melamine is added to milk powder produced by other manufacturers, the frequent events refer to events which do not occur for the first time, but also events that have previously occurred.
7. The food safety event-oriented early warning method as claimed in claim 1, wherein the event is analyzed and evaluated according to laws of health administration departments, the level of the event is determined, and the judgment result of the food safety event is notified, i-level, ii-level, iii-level and iv-level early warning information is issued.
8. The food safety event-oriented early warning method according to claim 1, wherein the cause of the food safety event is a cause of occurrence of the food safety event, and then whether the cause of the food safety event causes occurrence possibility of other food safety events in other types of food is judged, and then early warning information is issued for the food safety event which is likely to occur.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910898339.0A CN110688557A (en) | 2019-09-23 | 2019-09-23 | Food safety event-oriented early warning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910898339.0A CN110688557A (en) | 2019-09-23 | 2019-09-23 | Food safety event-oriented early warning method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110688557A true CN110688557A (en) | 2020-01-14 |
Family
ID=69109857
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910898339.0A Pending CN110688557A (en) | 2019-09-23 | 2019-09-23 | Food safety event-oriented early warning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110688557A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112395482A (en) * | 2020-11-06 | 2021-02-23 | 北京工商大学 | Food safety incident early warning platform |
| CN113360782A (en) * | 2021-06-07 | 2021-09-07 | 武汉理工大学 | Internet data-oriented food safety risk identification method and system |
| CN113592338A (en) * | 2021-08-09 | 2021-11-02 | 新疆大学 | Food quality management safety risk pre-screening model |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103853700A (en) * | 2012-11-29 | 2014-06-11 | 中国科学院计算机网络信息中心 | Event forewarning method based on regions and object information discovery |
| CN107491499A (en) * | 2017-07-27 | 2017-12-19 | 杭州中奥科技有限公司 | A kind of public sentiment method for early warning based on unstructured data |
| CN109241429A (en) * | 2018-09-05 | 2019-01-18 | 食品安全与营养（贵州）信息科技有限公司 | A kind of food safety public sentiment monitoring method and system |
| CN109347801A (en) * | 2018-09-17 | 2019-02-15 | 武汉大学 | A kind of vulnerability exploit methods of risk assessment based on multi-source word insertion and knowledge mapping |
| JP2019109860A (en) * | 2017-12-20 | 2019-07-04 | 株式会社富士通エフサス | Support device, display control method and display control program |
| CN110008311A (en) * | 2019-04-04 | 2019-07-12 | 北京邮电大学 | A kind of product information security risk monitoring method based on semantic analysis |
-
2019
- 2019-09-23 CN CN201910898339.0A patent/CN110688557A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103853700A (en) * | 2012-11-29 | 2014-06-11 | 中国科学院计算机网络信息中心 | Event forewarning method based on regions and object information discovery |
| CN107491499A (en) * | 2017-07-27 | 2017-12-19 | 杭州中奥科技有限公司 | A kind of public sentiment method for early warning based on unstructured data |
| JP2019109860A (en) * | 2017-12-20 | 2019-07-04 | 株式会社富士通エフサス | Support device, display control method and display control program |
| CN109241429A (en) * | 2018-09-05 | 2019-01-18 | 食品安全与营养（贵州）信息科技有限公司 | A kind of food safety public sentiment monitoring method and system |
| CN109347801A (en) * | 2018-09-17 | 2019-02-15 | 武汉大学 | A kind of vulnerability exploit methods of risk assessment based on multi-source word insertion and knowledge mapping |
| CN110008311A (en) * | 2019-04-04 | 2019-07-12 | 北京邮电大学 | A kind of product information security risk monitoring method based on semantic analysis |
Non-Patent Citations (2)
| Title |
|---|
| 凌文婧 等: "基于Hadoop的食品安全预警系统架构", 《食品安全导刊》 * |
| 陈敏: "《认知计算导论》", 31 May 2017, 华中科技大学出版社 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112395482A (en) * | 2020-11-06 | 2021-02-23 | 北京工商大学 | Food safety incident early warning platform |
| CN112395482B (en) * | 2020-11-06 | 2021-11-19 | 北京工商大学 | Food safety incident early warning platform |
| CN113360782A (en) * | 2021-06-07 | 2021-09-07 | 武汉理工大学 | Internet data-oriented food safety risk identification method and system |
| CN113592338A (en) * | 2021-08-09 | 2021-11-02 | 新疆大学 | Food quality management safety risk pre-screening model |
| CN113592338B (en) * | 2021-08-09 | 2023-09-12 | 新疆大学 | Food quality management safety risk pre-screening model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Robson et al. | A comprehensive review of food fraud terminologies and food fraud mitigation guides | |
| Zhu et al. | The role of media reporting in food safety governance in China: A dairy case study | |
| CN110688557A (en) | Food safety event-oriented early warning method | |
| Falessi et al. | A comprehensive characterization of NLP techniques for identifying equivalent requirements | |
| Mawudeku et al. | Global public health intelligence network (GPHIN) | |
| US20130297519A1 (en) | System and method for identifying potential legal liability and providing early warning in an enterprise | |
| WO2012174219A1 (en) | Apparatus and method for the automatic discovery of control events from the publication of documents | |
| Hatchard et al. | Tweeting about public health policy: Social media response to the UK Government’s announcement of a Parliamentary vote on draft standardised packaging regulations | |
| Johnson-Hall | Ensuring food safety by preventing food recalls: The impact of locus of failure, regulatory agency discovery, breadth, and firm size on corrective action | |
| Hirata et al. | Uncovering the impact of COVID-19 on shipping and logistics | |
| Suhaimi et al. | Consumers’ purchase decision based on intrinsic and extrinsic factors related to food safety issues: a review | |
| Gopal et al. | Machine learning based classification of online news data for disaster management | |
| Almanza et al. | The impact of publishing foodservice inspection scores | |
| Soon | Food fraud countermeasures and consumers: A future agenda | |
| Blench | Global public health intelligence network (GPHIN) | |
| Ma et al. | On the causation correlation of maritime accidents based on data mining techniques | |
| WO2006114328A1 (en) | Test mining systems and methods for early detection and warning | |
| Jones et al. | Estimating thresholds in occupancy when species detection is imperfect | |
| Zuo et al. | A food safety prescreening method with domain-specific information using online reviews | |
| EP2377054A1 (en) | Apparatus and method for the automatic discovery of control events from the publication of documents | |
| Gilbert | To blame is human: A quantitative systematic review of the relationship between outcome severity of large‐scale crises and attributions of blame | |
| Pilankar et al. | Detecting violation of human rights via social media | |
| Jishag et al. | Automated review analyzing system using sentiment analysis | |
| Gottipati et al. | Exploring the impact of COVID-19 on aviation industry: A text mining approach | |
| As-Salafiyah | COVID-19 and Halal Food: An Application of Text Analytics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200114 |
|
| WD01 | Invention patent application deemed withdrawn after publication |