CN105976125A - Dynamic layered early warning modeling method for food safety risk - Google Patents
Dynamic layered early warning modeling method for food safety risk Download PDFInfo
- Publication number
- CN105976125A CN105976125A CN201610333405.6A CN201610333405A CN105976125A CN 105976125 A CN105976125 A CN 105976125A CN 201610333405 A CN201610333405 A CN 201610333405A CN 105976125 A CN105976125 A CN 105976125A
- Authority
- CN
- China
- Prior art keywords
- node
- food safety
- early warning
- safety risk
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
- G06N5/041—Abduction
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Economics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Strategic Management (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Computational Linguistics (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a dynamic layered early warning modeling method for food safety risk. Based on a tree structure and the hierarchy analysis, the relation, the prediction conditions and the predication indexes of food safety key factors are taken as the attributes of the factor nodes to perform tree association modeling, and a layered early warning model for food safety risk is established. The dynamic layered early warning modeling method includes the steps of determining an index system for food safety risk early warning; establishing the tree structure of an early warning model; setting up the early warning indexes, calculation formula, warning conditions and early warning mode of food safety risk contained in the layer node, and establishing the layered early warning model of food safety risk; and obtaining the early warning prediction state value of nodes. The method is used for dynamic layered early warning for food safety risk, so that the resources can be effectively utilized, the working cost is reduced, the expenditure is saved, and the food safety risk predication ability, and decision-making ability and defense capability of major public security incidents are improved.
Description
Technical field
The invention belongs to knowledge engineering and field of artificial intelligence, relate to occurred events of public safety cooperation decision method, particularly relate to
A kind of dynamic layered early warning modeling method of food safety risk.
Background technology
At present, China's Safety of Food Quality general form allows of no optimist, and alimentary toxicosis and food origin disease remain the food of China
The matter of utmost importance of safety, great food safety affair happens occasionally.The serious stern form faced in view of current China Safety of Food Quality,
The processing agricultural product quality and safety management of China must be from the pipe afterwards of current " market sampling observation, media exposure, afterwards strike "
Reason pattern, is changed into " whole-process control, product back-tracing, sincere guarantee, risk assessment, harm early warning and emergency response " as early as possible
Pre-event management pattern.
The arrival of data age, the decision support behavior for food safety assessment brings new impact and challenge.To food safety
Carry out the existing method of early warning mainly include method for early warning based on step analysis, method for early warning based on fuzzy mathematics, based on
The method for early warning of Delphi, method for early warning based on neutral net and based on seasonal effect in time series trend prediction.Pre-based on above-mentioned early warning
Survey method can carry out corresponding Food Safety Analysis, but, these method for early warning are the most simply theorizing and Formal
Face gives computing formula and the method for early warning, and lacks concrete implementation, and does not support that user is according to different applied field
Scape is dynamically set up Early-warning Model and then is predicted early warning for food safety risk.At present in field of food safety, still lack pin
Modeling method to early warning, is both for greatly single incident in actual application and carries out early warning, it is impossible to directly apply to decision support system
System.
Summary of the invention
In order to overcome above-mentioned the deficiencies in the prior art, the present invention provides the dynamic layered early warning modeling method of a kind of food safety risk,
Based on step analysis and tree structure, use Method of Knowledge Reasoning, all regard all key factors as factor nodes, by relevant for institute
Connection relation, predicted condition, prediction index, as the attribute of node, carry out tree-like association modeling, create dividing of food safety risk
Layer Early-warning Model, for the dynamic layered early warning of food safety risk.
The principle of the present invention is: the present invention is directed to the dynamic layered early warning application of food safety risk, according to layering, by each index
Carry out tree-shaped arrangement by low-level early warning toward high-level early warning, set up dynamic multilayer Early-warning Model, model node object is retouched
State definition.Use Multiway Tree Structure, give the concrete structure of each node and the definition of tree construction, by each internodal connect into
Row hierarchical structure defines, and each node can freely define, and the concrete form of multiway tree can be set modeling according to concrete scene,
Use prediction engine that whole tree structure model is driven, it is achieved layering early warning.
Present invention provide the technical scheme that
The dynamic layered early warning modeling method of a kind of food safety risk, based on tree structure and step analysis, by food safety wind
Danger key factor is considered as the factor nodes of tree structure, using key factor incidence relation, predicted condition, prediction index as the factor
The attribute of node, carries out tree-like association modeling, creates the layering Early-warning Model of food safety risk, for food safety risk
Dynamic layered early warning；Comprise the steps:
1) index system of food safety risk early warning is determined；
2) tree structure of Early-warning Model is set up, by node and the child node of region setting at different levels tree structure, region layering and tree
Shape structural stratification is corresponding；
3) according to step 1) index system of described food safety risk early warning, for each layer of tree structure, this layer is set
Food safety risk warning index, computing formula, early-warning conditions and the modes of warning that node comprises, sets up food safety risk
Layering Early-warning Model；
4) according to step 3) the layering Early-warning Model of described food safety risk, obtain the early warning and alert state value of node；Root again
Demonstrate the early warning and alert state of each node according to early warning and alert state value with distinguishing, thus complete dynamically dividing of food safety risk
Layer early warning.
For the dynamic layered early warning modeling method of above-mentioned food safety risk, further, step 1) described food safety wind
The index system of danger early warning breaks out the data of system with specific reference to food origin disease Surveillance system and food origin disease, refers to
Mark system primary election, selects the vibrio cholerae examination case load in pathogeny detection, shigella dysenteriae inspection case load, rotavirus inspection
Case load and the total case load of intestinal tract disease, as the index system of described food safety risk early warning.
For the dynamic layered early warning modeling method of above-mentioned food safety risk, further, step 2) described tree structure is
Three-decker, the ground floor of tree structure is county, and each county is a node；The second layer is city, and each city is a node；
The third layer of tree structure is for saving, and only one of which node, for the root node of tree structure；The second layer is the child node of third layer；
Ground floor is leaf node, for the child node of the second layer.
For the dynamic layered early warning modeling method of above-mentioned food safety risk, further, step 2) described every node layer
Structure is defined as: node=(index, formula, early-warning conditions, modes of warning).
For the dynamic layered early warning modeling method of above-mentioned food safety risk, further, step 3) described set up food peace
The layering Early-warning Model of full blast danger, arranges every node layer and specifically includes following steps:
31) according to step 1) index system of described food safety risk early warning sets described index；Described index represents this joint
The food safety factor of point；Described index is a four-tuple (pointer type, initial value, unit, value of calculation)；
32) setup parameter variable, the value of described parametric variable is described finger target value, or according to the formula pair of described setting
The further calculated result of index；
33) described early-warning conditions is set according to described parametric variable；Described early-warning conditions be one hexa-atomic group (priority, state,
State description, state trend, sub-condition, subpattern)；Wherein said sub-condition is that a tlv triple (affiliated condition, close by constraint
System, state description)；
34) modes of warning of this layer is set according to described early-warning conditions；
35) model established is stored.
For the dynamic layered early warning modeling method of above-mentioned food safety risk, further, set described index and include: suddenly
Random vibrio inspection case load, shigella dysenteriae inspection case load, rotavirus inspection case load and the total case load of intestinal tract disease；
Formula by setting is calculated each bacterial disease number of cases respectively divided by the ratio of the total case load of disease and all antibacterial cases again
Number is divided by the ratio of the total case load of disease；
Setting two parametric variables afterwards as count1 and count2, described count1 value accounts for disease according to each bacterial disease number of cases
The ratio of total case load obtains；The ratio that described count2 value accounts for the total case load of disease according to all bacterial disease number of cases obtains；
Finally according to described count1 value and count2 value, set early-warning conditions, obtain modes of warning.
For the dynamic layered early warning modeling method of above-mentioned food safety risk, further, step 4) described in carry out food peace
The dynamic layered early warning of full blast danger specifically includes following steps:
41) the layering Early-warning Model object of the food safety risk that described establishment obtains, internal memory stress model index object are initialized；
42) input food safety risk data stream format data, mate index corresponding with node for formatted data, give joint
The Criterion Attribute assignment of point；
43) data to described input, calculate according to formula corresponding in node, obtain result of calculation；
44) according to step 43) described result of calculation, each node is carried out conditional judgment, obtains the early warning and alert state of node
Value, is divided into two classes according to the early warning and alert state value of node: be respectively risky node or devoid of risk node by node；Pass through again
Arrange different colours to carry out being layered early warning.
For the dynamic layered early warning modeling method of above-mentioned food safety risk, further, step 44) represent wind by redness
Danger node, green expression devoid of risk node.
Compared with prior art, the invention has the beneficial effects as follows:
The present invention provides the dynamic layered early warning modeling method of a kind of food safety risk, based on step analysis and tree structure, adopts
Use knowledge reasoning technology, all regard all key factors as factor nodes, using relevant, prediction index as node
Attribute, carries out tree-like association modeling, creates and obtain being layered Early-warning Model.
The invention provides for food safety based on chromatographic analysis, the method for the dynamic early-warning model modeling of knowledge reasoning, energy
Enough improve food safety risk predictive ability, the decision-making capability of great occurred events of public safety and defence capability, reduce job costs,
Cut down expenses, effectively utilize various resource, make safe all departments information and knowledge obtain effective integration and analysis, main efficacy results bag
Containing following some:
(1) by introducing tree and step analysis, it is achieved that food safety risk Early-warning Model dynamic modelling method, user is met
According to the requirement of different being customized of sight modelings, it is adapted to different sights, and forms sight collection, meet actual concrete
Needs；
(2) carry out flexibly according to factor nodes being layered early warning modeling, preferably embody different Food Safety State predictions,
Food Safety State real time data is carried out mode input according to time series, it is achieved the real-time risk profile to food safety risk；
(3) introduce Method of Knowledge Reasoning, utilize knowledge reasoning model, carry out the intelligent decision of automatization, comment for food safety
Estimate the decision support service that staff provides more intelligent, fine, various, for promoting food safety risk early warning level tool
Significant, for promote food safety risk reply intellectually and automatically level provide one feasible, advanced and effective
New way.
(4) use knowledge engineering technology that system is designed, can in knowledge and the collection of information, process, identify storage
The problem solving Semantic Heterogeneous during Deng, can make staff have more deep, correct understanding to related notion, thoroughly
Change fuzzy semantics present in conventional system.
In a word, the data at risk assessment center can effectively be utilized by food safety multilamellar method for early warning based on step analysis,
Contribute to forming good knowledge precipitation, rise simultaneously for the research of Risk-warning Inference Forecast and occurred events of public safety defence decision-making etc.
To important effect.Along with the lasting use of method system, technical solution of the present invention can make in system each staff in work
Make and in the middle of study, increase knowledge and skills, making information and the knowledge can be in the free flow of system all departments, by the wound of knowledge
New and restructuring, makes knowledge rise in value further.
Accompanying drawing explanation
Fig. 1 is the FB(flow block) of the dynamic layered early warning modeling method that the present invention provides.
Fig. 2 is the attribute definition figure of the Early-warning Model tree structure interior joint object that the present invention sets up.
Detailed description of the invention
Below in conjunction with the accompanying drawings, further describe the present invention by embodiment, but limit the scope of the present invention never in any form.
The present invention provides the dynamic layered early warning modeling method of a kind of food safety risk, based on step analysis and tree structure, adopts
By Method of Knowledge Reasoning, all regard all key factors as factor nodes, using all Correlation Criterias, prediction index as node
Attribute, carries out tree-like association modeling, creates the layering Early-warning Model of food safety risk, dynamically building for food safety risk
Mode division layer early warning.
Hierarchy Analysis Method is the simple and easy method that a kind of problem complex mainly for some, more fuzzy makes a policy, and is
In decision making process, non-quantitation event is done quantitative analysis, subjective judgment is done the effective ways of objective analysis.It is particularly well-suited to
Some are difficult to the problem of complete quantitative analysis, and hierarchical structure is the key decomposed and simplify comprehensive challenge clearly.
In the systematic analysis carrying out field of food safety problem, usually face one by answering that the many factors that is mutually related is constituted
Miscellaneous changeable data system.The decision-making of such issues that step analysis is and sequence provide a kind of simple and practical modeling method.Should
During by step analysis decision problem, it is necessary first to by problem methodization, stratification, construct a stratified structural model and (pass
Rank property hierarchical structure).Under this model, challenge is broken down into the ingredient of element or factor, and these elements are again by it
Attribute and relation form some levels, and the element of last layer time plays dominating role as criterion element relevant to next level.Pass rank
Property hierarchical structure is middle-level can be divided three classes: (1) is top: only one of which element in this level, and typically it is to analyze to ask
The predeterminated target of topic or desired result, the most top also referred to as destination layer.(2) intermediate layer: contain as reality in this level
Existing intermediate link involved by target, it can be made up of several levels, including criterion, the sub-criterion of required consideration, therefore
Intermediate layer is also referred to as rule layer.(3) lowermost layer: this level includes as realizing the alternative all kinds of basic elements of target,
Therefore lowermost layer is also referred to as basic element layer.
The detailed degree that hierachy number in recursive hierarchy structure is analyzed with the complexity of problem and needs is relevant, and general hierachy number is not
Restricted.Using step analysis, in conjunction with tree shape model form, carry out food safety risk Early-warning Model modeling, user can also
Model is developed, by the difference arranged, it is achieved risk profile and reverse reason are reviewed.
Knowledge reasoning refers to the mode of thinking of simulating human, according to domain knowledge and rule, particular problem makes inferences solved
Plant mechanism.In field of food safety, make corresponding with risk mechanism according to the knowledge in Food Safety Knowledge storehouse with rule
During decision-making, knowledge reasoning will necessarily be used.According to the real time data of input, the risk profile hierarchy model that coupling is set up, coupling
One or more rule being suitable in model rule base, and according to certain Inferential Control, infer the conclusion of problem.
For food safety data, carrying out being layered early warning mainly has two class models, a class to be based on single food hazard detection index
Food safety Early-warning Model, a class is food safety Early-warning Model based on multiple food hazardous material Testing index.Based on single
The food harm of food hazard detection index quickly identify and response early warning it is crucial that about the definition of food harm abnormal conditions.
On the basis of determining this definition, then determine the threshold value of normal condition.When the Testing index of hazardous material does not meets the threshold of normal condition
During value, produce early warning.Food Safety State early warning based on multiple food hazardous material Testing index, refer to danger multiple in food
Pest is monitored, and by the extent of injury indexation of each hazardous material, is i.e. calculated its corresponding hazardous material by concrete detection data
Pollution index, then the risk index of all kinds of hazardous material of COMPREHENSIVE CALCULATING, finally obtain Food Safety State evaluation.Its core is to utilize
Multiple hazardous material monitor values draw Food Safety State rational evaluation.
Fig. 1 is the FB(flow block) of the dynamic layered early warning modeling method that the present invention provides.Enteropathogenic series with food origin disease
As a example by detection sample, save, according at certain, the data that the Sentinel point hospital of counties and cities having under its command gathers, use the present invention to provide method to carry out food
Cause intestinal tract disease outburst risk multilamellar early warning tree-model modeling, specifically comprise the following steps that
Step 1: according to food origin disease Surveillance system, the data of food origin disease outburst system, carry out at the beginning of index system
Choosing, selects the vibrio cholerae examination case load in pathogeny detection, shigella dysenteriae inspection case load, and rotavirus checks case load,
The total case load of intestinal tract disease is index system；
Step 2: set up model tree diagram, by region setting model node, ground floor is county, and each county is a node, the
Two Ceng Wei cities, each city is a node, and third layer is province, only one of which node.That is, each node of ground floor is leaf
Node, the second layer is intermediate node, and third layer is root node.The same child node that county's node is this city's node being under the jurisdiction of a city,
All of city node is the child node of province's node.
Step 3: for model tree diagram, the respective attributes of each node object in successively arranging every layer, obtains in each layer each
The layering modes of warning of node；The respective attributes of node object specifically can set according to concrete problem, in the embodiment of the present invention,
The respective attributes of node object includes: index, formula, early-warning conditions and the modes of warning comprised；Wherein, the phase of node object
Attribute is answered specifically to set also dependent on the food-safety problem being specific to.Modes of warning can include two kinds: one, node exists
Security risk, carries out the layering early warning that this layering is corresponding；Two, there is not security risk in node, without early warning.
In the present embodiment, ground floor is layering at county level.In each node of ground floor, the index comprised is set, including suddenly
Random vibrio inspection case load, shigella dysenteriae inspection case load, rotavirus inspection case load and the total case load of intestinal tract disease；Formula is
Each bacterial disease number of cases (i.e. vibrio cholerae examination case load, shigella dysenteriae inspection case load, rotavirus inspection case load and intestinal disease
Sick total case load) divided by the ratio of the total case load of disease, and all bacterial disease number of cases are divided by the ratio of the total case load of disease；?
In the arranging of early-warning conditions, define two variablees count1, count2；Count1 is used for controlling the early warning of this layer of this node and shows；
Count2 is used for judging whether that needs carry out high-rise early warning.Wherein, count1 represents that bacterial disease number of cases accounts for the ratio of the total case load of disease
The example bacterial species number more than 10%, such as, without bacterial disease number of cases ratio more than 10%, then count1 is 0, if
Vibrio cholerae examination case load, shigella dysenteriae inspection case load, rotavirus inspection case load ratio any one more than 10%, then
Count1 is 1, and any two value is more than 10%, then count1 is 2, all exceedes, and count1 is 3；Count2 is according to institute
The ratio having bacterial disease number of cases to account for the total case load of disease carrys out value, the value of count2 specifically: when all bacterial disease number of cases divided by
When the ratio of the total case load of disease is more than or equal to 40%, then count2 is 2；When all bacterial disease number of cases are divided by the total case load of disease
Ratio less than 40% and more than or equal to 20% time, then count2 is 1；It is otherwise 0；Early-warning conditions is set as: if count1
More than or equal to 1, then there is intestinal tract disease at county level outburst risk, carry out early warning at county level；If count2 is equal to 2, early warning is set
Pattern, for moving towards next node layer (i.e. second layer node, city-level node), carries out next layer of early warning and calculates；
Fig. 2 is the definition figure of the tree structure interior joint object of Early-warning Model.As it can be seen, define it for each model node
Structure is: node=(index, formula, early-warning conditions, modes of warning)；Wherein, index=(pointer type, initial value, single
Position, value of calculation), index refers to the food safety factor that this model node represents；Pointer type is the description to this factor, just
Initial value is the initial assignment of this factor, can be empty；Unit is the numerical value unit of this factor；Value of calculation is used for being stored in actual mould
The end value of this factor in type operation；Formula refers to when this factor is to pass through the calculated value of formula according to actual monitoring value
Corresponding computing formula, if this factor is the factor of actual monitoring, then this is empty；Early-warning conditions=(priority, state,
State description, state trend, sub-condition, subpattern)；Sub-condition=(affiliated condition, restriction relation, state description)；Each
Individual node can bind multiple condition, it is possible to arranges condition check logic ("AND" or "or") on node, if selecting "AND",
Same first order conditions below this node all take with and using result predicting the outcome as node；If selecting "or", below this node
All take with first order conditions or and each condition under result predicting the outcome as node, node can be comprised many sub-conditions,
Its sub-condition check logic ("AND" or "or") is equally set in the first order condition under node, if selecting "AND",
The sub-condition of same one-level below this condition all take with and using result predicting the outcome as father's condition；If selecting "or", under the conditions of being somebody's turn to do
The sub-condition of same one-level in face all takes or and using result predicting the outcome as father's condition.The priority of condition refers to this condition
Be first judge or after judge；State refers to the result predicting this node, meets the requirements, and predicts the outcome as true, is not inconsistent
Closing is then false；State description, refers to the explanation to state outcome；State is moved towards, and refers to according to predicting the outcome, this
The next node that node points to；Affiliated condition, refers to the father's condition described belonging to sub-condition；Restriction relation, refers to antithetical phrase condition
Condition formula describe；State description, refers to the explanation to condition operation result.Modes of warning, refers to this node and is in prediction
In calculating, or it is not predicted calculating.
The second layer, in like manner, arranges index, formula, rule and the mode of rule as ground floor node, but, in index
Vibrio cholerae examination case load, shigella dysenteriae inspection case load, rotavirus inspection case load and the total case load of intestinal tract disease, refer to
Be citywide case load, i.e. the child node (ground floor node, the case load in Ji Ge county) of second layer node is added,
Value to the comprised parameter of second layer node.
In the present embodiment, third layer is the root of tree structure, and step 3, step 4 are same, set index, formula, early warning
Condition and modes of warning, but, this layer (third layer) is the root of tree structure, so, modes of warning no longer arranges next
Layer moves towards node.
When being predicted early warning, obtain the early warning result of node, then district according to the layering Early-warning Model of the food safety risk set up
Demonstrate the early warning result of each node with dividing；Specifically include:
41) the early warning tree-model object that establishment obtains, internal memory stress model index object are initialized；
42) formatted data (such as data stream format data) conduct input data are obtained from food origin disease Surveillance system, will
The formatted data index corresponding with node is mated, to the Criterion Attribute assignment of node；
Input data match different nodes according to different index names, the most in the following order by formatted data and tree structure
The index that node is corresponding is mated: the first left node of the ground floor of first tree structure is last node, again to the right
To the right, last node, the node of last third layer such put in order, by data the first left node of the second layer
The targets match that stream format data is corresponding with node, completes the achievement data coupling of node；
43): input data calculate according to formula corresponding in node, obtain result of calculation；
44): according to above-mentioned result of calculation, each node is carried out conditional judgment, the early warning and alert state value obtaining node (has two
Individual value, the most risky node or devoid of risk node), different values shows by different colors, according to the difference of color, enters
Row layering early warning；Risk node, green expression devoid of risk node is represented by redness；When certain hierarchy node reddens, then it represents that this layer
Level node has food safety risk.
It should be noted that publicizing and implementing the purpose of example is that help is further appreciated by the present invention, but those skilled in the art
It is understood that various substitutions and modifications are all possible without departing from the present invention and spirit and scope of the appended claims.
Therefore, the present invention should not be limited to embodiment disclosure of that, and the scope of protection of present invention defines with claims
Scope is as the criterion.
Claims (9)
1. a dynamic layered early warning modeling method for food safety risk, based on tree structure and step analysis, by food safety
Risk key factor is considered as the factor nodes of tree structure, using key factor incidence relation, predicted condition, prediction index as because of
The attribute of child node, carries out tree-like association modeling, creates the layering Early-warning Model of food safety risk, for food safety risk
Dynamic layered early warning；Comprise the steps:
1) index system of food safety risk early warning is determined；
2) tree structure of Early-warning Model is set up, by node and the child node of region setting at different levels tree structure, region layering and tree
Shape structural stratification is corresponding；
3) according to step 1) index system of described food safety risk early warning, for each layer of tree structure, this layer is set
Food safety risk warning index, computing formula, early-warning conditions and the modes of warning that node comprises, sets up food safety risk
Layering Early-warning Model；
4) according to step 3) the layering Early-warning Model of described food safety risk, obtain the early warning and alert state value of node；Root again
Demonstrate the early warning and alert state of each node according to early warning and alert state value with distinguishing, thus complete dynamically dividing of food safety risk
Layer early warning.
2. the dynamic layered early warning modeling method of food safety risk as claimed in claim 1, is characterized in that, step 1) described
The index system of food safety risk early warning breaks out the number of system with specific reference to food origin disease Surveillance system and food origin disease
According to, carry out index system primary election, select the vibrio cholerae examination case load in pathogeny detection, shigella dysenteriae inspection case load, wheel
Shape virus examination case load and the total case load of intestinal tract disease, as the index system of described food safety risk early warning.
3. the dynamic layered early warning modeling method of food safety risk as claimed in claim 1, is characterized in that, step 2) described
Tree structure is three-decker, and the ground floor of tree structure is county, and each county is a node；The second layer is city, and each city is
One node；The third layer of tree structure is for saving, and only one of which node, for the root node of tree structure；The second layer is third layer
Child node；Ground floor is leaf node, for the child node of the second layer.
4. the dynamic layered early warning modeling method of food safety risk as claimed in claim 1, is characterized in that, step 2) described
The structure of every node layer is defined as: node=(index, formula, early-warning conditions, modes of warning).
5. the dynamic layered early warning modeling method of food safety risk as claimed in claim 1, is characterized in that, step 3) described
Set up the layering Early-warning Model of food safety risk, every node layer be set and specifically include following steps:
31) according to step 1) index system of described food safety risk early warning sets described index；Described index represents this joint
The food safety factor of point；Described index is a four-tuple (pointer type, initial value, unit, value of calculation)；
32) setup parameter variable, the value of described parametric variable is described finger target value, or according to the formula pair of described setting
The further calculated result of index；
33) described early-warning conditions is set according to described parametric variable；Described early-warning conditions be one hexa-atomic group (priority, state,
State description, state trend, sub-condition, subpattern)；Wherein said sub-condition is that a tlv triple (affiliated condition, close by constraint
System, state description)；
34) modes of warning of this layer is set according to described early-warning conditions；
35) model established is stored.
6. the dynamic layered early warning modeling method of food safety risk as claimed in claim 5, is characterized in that, set described index
Including: vibrio cholerae examination case load, shigella dysenteriae inspection case load, rotavirus inspection case load and the total case load of intestinal tract disease；
First, the formula by setting is calculated each bacterial disease number of cases respectively divided by the ratio of the total case load of disease and all antibacterials
Case load is divided by the ratio of the total case load of disease；
Afterwards, two parametric variables are set as count1 and count2；Count1 is for representing the alert status of this layer of this node；
Count2 is used for representing whether and carries out high-rise early warning；Described count1 value accounts for the total case of disease according to each bacterial disease number of cases
The ratio of number obtains；The ratio that described count2 value accounts for the total case load of disease according to all bacterial disease number of cases obtains；
Finally, according to described count1 value and count2 value, set early-warning conditions, obtain modes of warning.
7. the dynamic layered early warning modeling method of food safety risk as claimed in claim 6, is characterized in that, described parametric variable
Count1 accounts for the ratio of the total case load of the disease bacterial species number more than 10% equal to bacterial disease number of cases；Described parametric variable count2
The ratio accounting for the total case load of disease according to all bacterial disease number of cases carrys out value, when all bacterial disease number of cases are divided by the total case load of disease
When ratio is more than or equal to 40%, count2 is equal to 2；When described count1 is more than or equal to 1, the alert status of this layer of this node
For risky；When described count2 is equal to 2, modes of warning calculates for carrying out next layer of early warning.
8. the dynamic layered early warning modeling method of food safety risk as claimed in claim 1, is characterized in that, step 4) described
Carry out the dynamic layered early warning of food safety risk and specifically include following steps:
41) the layering Early-warning Model object of the food safety risk that described establishment obtains, internal memory stress model index object are initialized；
42) input food safety risk data stream format data, mate index corresponding with node for formatted data, give joint
The Criterion Attribute assignment of point；
43) data to described input, calculate according to formula corresponding in node, obtain result of calculation；
44) according to step 43) described result of calculation, each node is carried out conditional judgment, obtains the early warning and alert state of node
Value, is divided into two classes according to the early warning and alert state value of node: be respectively risky node or devoid of risk node by node；Pass through again
Arrange different colours carry out be layered early warning show.
9. the dynamic layered early warning modeling method of food safety risk as claimed in claim 7, is characterized in that, step 44) use
Red expression risk node, represents devoid of risk node by green, is achieved in being layered early warning and shows.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610333405.6A CN105976125A (en) | 2016-05-19 | 2016-05-19 | Dynamic layered early warning modeling method for food safety risk |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610333405.6A CN105976125A (en) | 2016-05-19 | 2016-05-19 | Dynamic layered early warning modeling method for food safety risk |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105976125A true CN105976125A (en) | 2016-09-28 |
Family
ID=56955273
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610333405.6A Pending CN105976125A (en) | 2016-05-19 | 2016-05-19 | Dynamic layered early warning modeling method for food safety risk |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105976125A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107862479A (en) * | 2017-12-07 | 2018-03-30 | 钟永松 | A kind of food security early warning system based on cloud computing |
| CN110059854A (en) * | 2019-03-13 | 2019-07-26 | 阿里巴巴集团控股有限公司 | Method and device for risk identification |
| CN110889635A (en) * | 2019-11-29 | 2020-03-17 | 北京金和网络股份有限公司 | Method for performing emergency drilling on food safety event processing |
| CN112148749A (en) * | 2020-11-24 | 2020-12-29 | 车智互联（北京）科技有限公司 | Data analysis method, computing device and storage medium |
-
2016
- 2016-05-19 CN CN201610333405.6A patent/CN105976125A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107862479A (en) * | 2017-12-07 | 2018-03-30 | 钟永松 | A kind of food security early warning system based on cloud computing |
| CN110059854A (en) * | 2019-03-13 | 2019-07-26 | 阿里巴巴集团控股有限公司 | Method and device for risk identification |
| CN110889635A (en) * | 2019-11-29 | 2020-03-17 | 北京金和网络股份有限公司 | Method for performing emergency drilling on food safety event processing |
| CN110889635B (en) * | 2019-11-29 | 2022-10-04 | 北京金和网络股份有限公司 | Method for performing emergency drilling on food safety event processing |
| CN112148749A (en) * | 2020-11-24 | 2020-12-29 | 车智互联（北京）科技有限公司 | Data analysis method, computing device and storage medium |
| CN112148749B (en) * | 2020-11-24 | 2021-04-20 | 车智互联（北京）科技有限公司 | Data analysis method, computing device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chowdhury et al. | Advantages and limitations of artificial intelligence | |
| CN105976125A (en) | Dynamic layered early warning modeling method for food safety risk | |
| Luo et al. | A BIM-based code compliance checking process of deep foundation construction plans | |
| Lee et al. | An efficient design support system based on automatic rule checking and case-based reasoning | |
| Lin et al. | Using a network framework to quantitatively select ecological indicators | |
| CN107832913A (en) | The Forecasting Methodology and system to monitoring data trend based on deep learning | |
| Chen et al. | Safety supervision of tower crane operation on construction sites: An evolutionary game analysis | |
| CN107545375A (en) | A kind of high-risk equipment fault quantifying risk evaluation method of petroleum and petrochemical industry | |
| Yang et al. | Development of a risk monitoring system for nuclear power plants based on GO-FLOW methodology | |
| Liu et al. | Using fuzzy logic to generate conditional probabilities in Bayesian belief networks: a case study of ecological assessment | |
| CN109857080A (en) | A kind of intelligent ship lock operational system based on BIM | |
| CN103425884B (en) | A kind of fuzzy prediction method of subway transfer channel traffic state | |
| Su et al. | Inclusion of task dependence in human reliability analysis | |
| Balali et al. | Selecting appropriate structural system: Application of PROMETHEE decision making method | |
| Rossignol et al. | Rethinking vulnerability analysis and governance with emphasis on a participatory approach | |
| CN110782147A (en) | Fuzzy Petri network-based oil spill disaster chain risk analysis method | |
| Sarip et al. | Fuzzy logic application for house price prediction | |
| Hong et al. | Performance balance oriented product structure optimization involving heterogeneous uncertainties in intelligent manufacturing with an industrial network | |
| CN113610397A (en) | Petrochemical enterprise safety evaluation method based on PSO-BP neural network | |
| Comes et al. | Enhancing robustness in multi-criteria decision-making: A scenario-based approach | |
| Meng et al. | CBR-RBR fusion based parametric rapid construction method of bridge BIM model | |
| Zhao et al. | Study of fault diagnosis method based on fuzzy Bayesian network and application in CTCS-3 train control system | |
| Xu et al. | Fire Safety Assessment of High-Rise Buildings Based on Fuzzy Theory and Radial Basis Function Neural Network. | |
| Song et al. | Quality Risk Management Algorithm for Cold Storage Construction Based on Bayesian Networks | |
| Xu | Risk Assessment of Green Intelligent Building Based on Artificial Intelligence |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160928 |