CN114117032A - Method and device for generating plan based on real-time emergency data and electronic equipment - Google Patents
Method and device for generating plan based on real-time emergency data and electronic equipment Download PDFInfo
- Publication number
- CN114117032A CN114117032A CN202111490281.XA CN202111490281A CN114117032A CN 114117032 A CN114117032 A CN 114117032A CN 202111490281 A CN202111490281 A CN 202111490281A CN 114117032 A CN114117032 A CN 114117032A
- Authority
- CN
- China
- Prior art keywords
- plan
- prediction
- emergency data
- time emergency
- real
- 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
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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Animal Behavior & Ethology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of computer science, in particular to a method, a device and electronic equipment for generating a plan based on real-time emergency data, wherein the method comprises the steps of obtaining current real-time emergency data, respectively carrying out local optimal prediction and global optimal prediction on the current real-time emergency data based on a plan knowledge graph to obtain a first prediction plan and a second prediction plan, the plan knowledge graph is constructed based on historical real-time emergency data information and corresponding historical plan information, carrying out semantic analysis on the first prediction plan and the second prediction plan, respectively comparing semantic analysis results with target semantics, determining a target plan in the first prediction plan and the second prediction plan, generating the plans by introducing different modes, and further selecting the optimal plan by screening different plans, the correctness and the effectiveness of the generated plan are fully ensured.
Description
Technical Field
The invention relates to the technical field of computer science, in particular to a method and a device for generating a plan based on real-time emergency data and electronic equipment.
Background
In real life, a large number of emergencies are filled, particularly real-time emergency data similar to an airport, in order to better deal with the emergencies, some plans need to be made in advance, in the prior art, the airport is taken as an example, the main form of the current airport plans is a paper plan, when the emergencies occur, emergency treatment personnel need to browse corresponding treatment processes, because the airport emergencies have the suddenness, the urgency and the high hazard, the plans made in advance can not deal with the emergencies successfully, meanwhile, the forms of the emergencies are various, sometimes the plans can not correspond to the plans one to one, and therefore, the existing plans have a great problem in form.
Therefore, a method, an apparatus and an electronic device for generating a plan based on real-time emergency data are needed to overcome the above-mentioned drawbacks.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method and an apparatus for generating a plan based on real-time emergency data, and an electronic device, so as to solve the problem that an effective plan cannot be generated based on current real-time emergency data in the prior art.
According to a first aspect, an embodiment of the present invention provides a method for generating a plan based on real-time emergency data, including:
acquiring current real-time emergency data;
respectively performing local optimal prediction and global optimal prediction on the current real-time emergency data based on a plan knowledge graph to obtain a first prediction plan and a second prediction plan, wherein the plan knowledge graph is constructed based on historical real-time emergency data information and corresponding historical plan information;
and performing semantic analysis on the first prediction plan and the second prediction plan, comparing semantic analysis results with target semantics respectively, and determining the target plan in the first prediction plan and the second prediction plan.
According to the method for generating the plan based on the real-time emergency data, provided by the embodiment of the invention, the current real-time emergency data is combined with the knowledge map, so that the correct and effective plan can be made under any real-time emergency data, meanwhile, the plan is generated by introducing different modes, and then the optimal plan is selected by screening different plans, so that the correctness and the effectiveness of the generated plan are fully ensured.
With reference to the first implementation manner of the first aspect, in the first implementation manner of the first aspect, the performing, based on a plan knowledge graph, local optimal prediction and global optimal prediction on the current real-time emergency data to obtain a first prediction plan and a second prediction plan respectively includes:
performing local optimal prediction on the current real-time emergency data based on a plan knowledge graph, and determining the first prediction plan;
and inputting the current real-time emergency data into a plan prediction model, and determining the second prediction plan, wherein the plan prediction model is obtained based on the plan knowledge graph training.
According to the method for generating the plan based on the real-time emergency data, provided by the embodiment of the invention, the plan is generated by introducing two different modes of local optimal prediction and global optimal prediction, and then the optimal plan is selected by screening different plans, so that the accuracy and the effectiveness of the generated plan are fully ensured.
With reference to the first embodiment of the first aspect, in a second embodiment of the first aspect, the method further includes:
updating the plan knowledge graph based on the corresponding relation between the target plan and the current real-time emergency data;
and training the plan prediction model based on the updated plan knowledge graph, and determining the updated plan prediction model.
According to the method for generating the plan based on the real-time emergency data, provided by the embodiment of the invention, the real-time emergency data acquired each time and the plan generated correspondingly are used for updating the knowledge graph and the prediction model, so that the timeliness of the prediction model and the knowledge graph is ensured, and the accuracy and the effectiveness of the generated plan in the subsequent plan generation process are further ensured.
With reference to the first aspect, in a third implementation manner of the first aspect, the performing semantic analysis on the first prediction plan and the second prediction plan, comparing semantic analysis results with target semantics respectively, and determining a target plan in the first prediction plan and the second prediction plan includes:
if the semantic analysis result of the first prediction plan is the same as the target semantic, taking the first prediction plan as a target plan;
otherwise, the second prediction plan is used as a target plan.
According to the method for generating the plan based on the real-time emergency data, provided by the embodiment of the invention, the generated plan is ensured to always accord with the preset target through the comparison between the semantic analysis result of the generated plan and the target semantic, and the accuracy and the effectiveness of the generated plan are further ensured.
With reference to the first implementation manner of the first aspect, in a fourth implementation manner of the first aspect, the performing a local optimal prediction on the current real-time emergency data based on a plan knowledge graph, and determining the first prediction plan includes:
respectively matching the current real-time emergency data based on the feature words in the predetermined case knowledge graph to obtain the feature words corresponding to the current real-time emergency data;
and selecting the entity and the entity behavior corresponding to the characteristic words as the first prediction plan based on the entity and the entity behavior in the characteristic words and the plan knowledge graph.
According to the method for generating the plan based on the real-time emergency data, provided by the embodiment of the invention, the current real-time emergency data information is described by using the characteristic words, so that the real-time emergency data information can be normally identified to generate the corresponding plan when the real-time emergency data information which does not exist in the knowledge map is met, and the accuracy and the effectiveness of the generated plan in the subsequent plan generating process are further ensured.
With reference to the first aspect or the third embodiment of the first aspect, in a fifth embodiment of the first aspect, the method further includes:
if the semantic analysis result of the first prediction plan and the semantic analysis result of the first prediction plan are the same as the target semantic, comparing the semantic analysis results with standby target semantics respectively, and determining a target plan in the first prediction plan and the second prediction plan.
According to the method for generating the plan based on the real-time emergency data, provided by the embodiment of the invention, through setting the plurality of target semantics, even if the semantics of the generated two plans are the same in a certain aspect, the most appropriate target plan can be screened out through further comparison, and the accuracy and the effectiveness of the generated plan are further ensured.
With reference to the first embodiment of the first aspect, in a sixth embodiment of the first aspect, the method further includes:
and taking the predetermined plan knowledge graph as a training set, and sending the predetermined plan knowledge graph into a deep neural network for training to obtain the predetermined plan prediction model.
According to the method for generating the plan based on the real-time emergency data, provided by the embodiment of the invention, the corresponding plan generation model is generated by introducing the deep neural network, so that the accuracy and the effectiveness of the generated plan in the subsequent plan generation process are further ensured.
According to a second aspect, an embodiment of the present invention provides an apparatus for generating a plan based on real-time emergency data, including:
the acquisition module is used for acquiring current real-time emergency data;
the first processing module is used for respectively carrying out local optimal prediction and global optimal prediction on the current real-time emergency data based on a plan knowledge graph to obtain a first prediction plan and a second prediction plan, wherein the plan knowledge graph is constructed based on historical real-time emergency data information and corresponding historical plan information;
and the second processing module is used for performing semantic analysis on the first prediction plan and the second prediction plan, comparing semantic analysis results with target semantics respectively, and determining a target plan from the first prediction plan and the second prediction plan.
According to the method for generating the plan based on the real-time emergency data, provided by the embodiment of the invention, the current real-time emergency data is combined with the knowledge map, so that the correct and effective plan can be made under any real-time emergency data, meanwhile, the plan is generated by introducing different modes, and then the optimal plan is selected by screening different plans, so that the correctness and the effectiveness of the generated plan are fully ensured.
According to a third aspect, an embodiment of the present invention provides an electronic device, including: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing therein computer instructions, and the processor executing the computer instructions to perform the method for generating a plan based on real-time emergency data according to the first aspect or any one of the embodiments of the first aspect.
According to a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, which stores computer instructions for causing a computer to execute the method for generating a plan based on real-time emergency data described in the first aspect or any one of the implementation manners of the first aspect.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow diagram of a method for generating a protocol based on real-time emergency data according to an embodiment of the invention;
FIG. 2 is a flow diagram of a method for generating a protocol based on real-time emergency data, according to an embodiment of the invention;
FIG. 3A is a flow diagram of relationship extraction in generating a knowledge graph according to an embodiment of the invention;
FIG. 3B is a schematic diagram of relationship extraction in the process of generating a knowledge graph according to an embodiment of the invention;
FIG. 3C is a flow diagram of relationship completion in generating a knowledge-graph according to an embodiment of the invention;
FIG. 3D is a flow diagram of matching text instances in a process of generating a knowledge-graph according to an embodiment of the invention;
FIG. 4 is a flow diagram of a method for generating a protocol based on real-time emergency data, according to an embodiment of the invention;
fig. 5 is a block diagram of a device for generating a plan based on real-time emergency data according to an embodiment of the present invention;
fig. 6 is a schematic diagram of a hardware structure of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
According to the method for generating the plan based on the real-time emergency data, provided by the embodiment of the invention, the real-time emergency data acquired each time and the plan generated correspondingly are used for updating the knowledge graph and the prediction model, so that the timeliness of the prediction model and the knowledge graph is ensured, and the accuracy and the effectiveness of the generated plan in the subsequent plan generation process are further ensured.
In accordance with an embodiment of the present invention, there is provided an embodiment of a method for real-time emergency data generation protocol, it should be noted that the steps illustrated in the flowchart of the accompanying drawings may be performed in a computer system such as a set of computer executable instructions, and that while a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in an order different than here.
In this embodiment, a method for generating a plan based on real-time emergency data is provided, and may be used for electronic devices, such as a computer, a server, a tablet computer, and the like, fig. 1 is a flowchart of a method for generating a plan based on real-time emergency data according to an embodiment of the present invention, and as shown in fig. 1, the flowchart includes the following steps:
and S11, acquiring the current real-time emergency data.
Specifically, the current real-time emergency data may include, but is not limited to: current person, time, event, location, etc. of various real-time emergency data.
For example, in practical applications, by taking real-time emergency data of an airport a as an example, it is assumed that a fire occurs in an airport a. Then in the above example, terminal a and the fire are both real-time emergency data.
And S12, respectively carrying out local optimal prediction and global optimal prediction on the current real-time emergency data based on a plan knowledge graph to obtain a first prediction plan and a second prediction plan, wherein the plan knowledge graph is constructed based on historical real-time emergency data information and corresponding historical plan information.
Specifically, according to the corresponding relation in the knowledge graph and the obtained current real-time emergency data information, local optimal prediction and global optimal prediction are carried out, and a first prediction plan and a second prediction plan are obtained.
For example, there is a corresponding relationship in the knowledge graph K that the entity B needs to perform processing on the entity a, but the success rate of processing the entity a by the entity B is not high through global optimal prediction, so the global optimal prediction concludes that the generated second prediction plan is: the entity B does not process the entity A, but in local consideration, historical factors are not considered, local optimal prediction is obtained, and the first prediction plan is generated as follows: entity B processes entity a.
Details about this step will be described later.
And S13, performing semantic analysis on the first prediction plan and the second prediction plan, comparing semantic analysis results with target semantics respectively, and determining the target plan in the first prediction plan and the second prediction plan.
Specifically, the target plan is determined from the first prediction plan and the second prediction plan according to the target semantics.
For example, also taking the knowledge graph K and the entities a and B as examples, the generated first predicted solution and second predicted solution are also the same as the examples described above, and assuming that the target semantics is that the entity a needs to be processed, the first predicted solution and the target semantics are the same, and the second predicted solution and the target semantics are different, so the first predicted solution is selected as the target solution.
Details about this step will be described later.
According to the method for generating the plan based on the real-time emergency data, provided by the embodiment of the invention, the current real-time emergency data is combined with the knowledge map, so that the correct and effective plan can be made under any real-time emergency data, meanwhile, the plan is generated by introducing different modes, and then the optimal plan is selected by screening different plans, so that the correctness and the effectiveness of the generated plan are fully ensured.
In this embodiment, a method for generating a plan based on real-time emergency data is provided, which may be used in electronic devices, such as a computer, a server, a tablet computer, and the like, fig. 2 is a flowchart of generating a plan based on real-time emergency data according to an embodiment of the present invention, and as shown in fig. 2, the flowchart includes the following steps:
and S21, acquiring the current real-time emergency data.
Please refer to S11 in fig. 1, which is not described herein again.
And S22, respectively carrying out local optimal prediction and global optimal prediction on the current real-time emergency data based on a plan knowledge graph to obtain a first prediction plan and a second prediction plan, wherein the plan knowledge graph is constructed based on historical real-time emergency data information and corresponding historical plan information.
Specifically, S22 includes:
s221, local optimal prediction is carried out on the current real-time emergency data based on a plan knowledge graph, and the first prediction plan is determined.
(1) And respectively matching the current real-time emergency data based on the characteristic words in the predetermined case knowledge graph to obtain the characteristic words corresponding to the current real-time emergency data.
Specifically, the entities existing in the knowledge graph are directly selected corresponding feature words, and the entities not existing in the knowledge graph are searched for similar entities to select corresponding feature words.
For example, suppose that there are feature words nurse and police in the knowledge graph, wherein the nurse corresponds to the feature words: rescue, medical knowledge, feature words corresponding to police: rescue knowledge, criminal investigation knowledge. Supposing that the obtained current real-time emergency data information comprises two entities, namely a police PL and a doctor DR, wherein the police PL can directly obtain feature word rescue knowledge and criminal investigation knowledge according to a knowledge graph, but because the doctor DR entity cannot find a corresponding entity in the knowledge graph, a feature word of a similar entity nurse is used as the feature word of the doctor DR, so that the feature word of the doctor DR is as follows: rescue, medical knowledge.
It should be noted that the entity classification and the feature word relationship list are only examples, and in practical applications, the practical applications are not limited to the correspondence described in the examples.
(2) And selecting the entity and the entity behavior corresponding to the characteristic words as the first prediction plan based on the entity and the entity behavior in the characteristic words and the plan knowledge graph.
For example, an emergency event X exists, the emergency event X is a fire X, a fire fighter FM exists at the emergency event X, the current real-time emergency data information is acquired and includes the fire X and the fire fighter FM, and the characteristic words of the fire X are judged through the knowledge graph: high temperature, harm, the characteristic word of obtaining firemen FM is: fire and rescue knowledge can be suppressed. Obviously, the firefighter FM can suppress the fire x, and thus, the first prediction case is that the firefighter FM suppresses the fire x.
S222, inputting the current real-time emergency data into a plan prediction model, and determining the second prediction plan, wherein the plan prediction model is obtained based on the plan knowledge graph training.
Specifically, the plan prediction model performs global prediction according to global information.
For example, an emergency event SX exists, the emergency event SX is a special fire SX, a fireman FM _1 exists at the place where the emergency event SX occurs, the current real-time emergency data information is acquired, the current real-time emergency data information includes the special fire SX, the fireman FM _1 determines that the characteristic words of the special fire SX are as follows through a knowledge graph: high temperature, harm, difficult putting out, the characteristic word who obtains fire fighter FM _1 is: fire and rescue knowledge can be suppressed. Meanwhile, according to the investigation of the historical plan, the prediction model finds that the probability of extinguishing the special fire sx is extremely low, and the difficulty of extinguishing the special fire sx by a fireman is extremely high, so that the second prediction plan is that the fireman FM _1 withdraws, and the special fire sx does not need to be extinguished.
And S23, performing semantic analysis on the first prediction plan and the second prediction plan, comparing semantic analysis results with target semantics respectively, and determining the target plan in the first prediction plan and the second prediction plan.
Specifically, S23 includes:
s231, if the semantic analysis result of the first prediction plan is the same as the target semantic, taking the first prediction plan as a target plan.
For example, assume that there are target semantics: entity A1 executes entity B1, and if entity A1 executes entity B1 as a result of semantic analysis performed on the first predicted solution, the first predicted solution is taken as a target solution.
S232, if not, taking the second prediction plan as a target plan.
For example, in practical applications, it is assumed that there are target semantics: the entity a1 does not execute the entity B1, and performs semantic analysis on the first prediction plan to obtain a semantic analysis result that the entity a1 executes the entity B1, which is obviously different from the target semantic, and at the same time, performs semantic analysis on the first prediction plan to obtain a semantic analysis result that the entity a1 does not execute the entity B1, and the semantic analysis result of the second prediction plan is the same as the target semantic, and selects the second prediction plan as the target plan.
According to the method for generating the plan based on the real-time emergency data, provided by the embodiment of the invention, the current real-time emergency data is combined with the knowledge map, so that the correct and effective plan can be made under any real-time emergency data, meanwhile, the plan is generated by introducing different modes, and then the optimal plan is selected by screening different plans, so that the correctness and the effectiveness of the generated plan are fully ensured.
In this embodiment, a concept of a knowledge graph is introduced, and for building the knowledge graph, firstly, relationship labeling between entities is realized in an unsupervised manner based on depth-based inclusion clustering, as shown in fig. 3A, specifically including:
s301, inputting an emergency plan sample library;
specifically, the step is to input historical plans and historical real-time emergency data into the value model.
S302, feature extraction and vectorization are realized;
specifically, the dependency parsing algorithm is used in combination, and the characteristic mode of the context window is used for extracting the characteristic words.
S303, deep embedding clustering is realized;
specifically, as shown in fig. 3B, after data is input into the self-encoder, the self-encoder performs self-encoding operation on the data through its encoding layer, and then inputs the self-encoded data into the decoding layer to reconstruct the data, but in this embodiment, only the self-encoded data in the encoding layer is needed, and the data in the encoding layer in the self-encoder is input into the relative entropy optimization clustering model to be trained, and the clustering result is output.
Further, when the method is used, the following formula is involved:
wherein q isijIs the probability of assigning data i to cluster j, ziRepresenting data i, j representing cluster j, uiDenotes the cluster center and α denotes the degree of freedom of the t-distribution.
Wherein p isijTarget distribution, fjAs follows:
fj＝∑iqij
finally, the target distribution is optimized by the following formula KL (P | | Q):
s304, labeling the relation label;
and according to the clustering result obtained in the step S303, finding out the feature word which can represent the features most according to the comparison result of the weights.
In particular, the following formula is involved:
wherein, Wi，kDegree of representation of the representative feature i for the class k, CiIs a collection of clusters containing feature i.
Referring to fig. 3C, the process for further refining the knowledge-graph is as follows:
s311, inputting the training set data into a model;
specifically, the obtained clustering result is input into a value relation completion model.
S312, matching the text example;
referring to fig. 3D, the instance matching may be performed in the form of such a triplet as < entity T, correspondence Y, entity P >. The method comprises the following specific steps:
s3121, randomly combining to form a triple group;
for example, find three words, firefighter, execute, command.
S3122, distinguishing entities and corresponding relations in the triples;
for example, firefighters and commands are both entities and executions are correspondences.
S3123, forming a text and searching a corresponding example;
for example, there is a "firefighter execution command" in the example, the match was successful.
S3124, outputting the formed triplet
Finally, still taking the above-mentioned triples as an example, the text is output.
S313, extracting the relation template;
specifically, firstly, a text instance list is input, all text instances in the list are traversed, the similarity between a representation vector of the text instance and each cluster is calculated, and the text instance is divided into a first class cluster with the confidence coefficient larger than a threshold value. If the similarity of a certain text instance and the existing clusters is lower than the threshold value, a new class cluster is created by taking the vector as the center. And finally, taking the mass center of each cluster as the extracted relation template.
S314A1, expanding the text example;
after the template generation is finished, the template with high confidence coefficient is sent into a training set, wherein the confidence coefficient Conf (p)i) The calculation can be made by the following formula:
pos is the number of similar generated templates and examples, neg is the number of dissimilar generated templates and examples, unk is the number of similarity between generated templates and examples, and W is the number of similarity between generated templates and examples which cannot be judgednegGenerating the number w of the template entity characteristic words which are not similar to the example characteristic wordsunkTo generate the number of template entity feature words whose similarity to the instance feature words can not be judged.
S314A2, suppressing semantic drift;
the drift of the semantics is suppressed by using the markov random field, and specifically, the markov random field uses the following formula:
wherein E is the value of the energy function, xt，ytA hidden variable is represented by a number of hidden variables,
representing the transposed matrix of the corresponding parameter.
Further, after the corresponding text example with the semantic deviation suppressed is obtained, the confidence degree is calculated, and the text example with the confidence degree is deleted.
Further, the confidence may be obtained by the following formula:
wherein, sim (P)i，Ci) Selected triplet PiAnd example CiSimilarly, P is the total number of instances.
S314B, the relationship information is output.
Specifically, the data is the result of multiple calculations.
Further, by training the historical real-time emergency data and the historical plans, a plurality of corresponding relations may finally appear, and in order to ensure the correctness of the corresponding relations, the embodiment introduces a convolutional neural network to calculate the relations, wherein the propagation rule of each convolutional layer is as follows:
wherein, therein
Is the adjacency matrix of undirected graph G plus self-join (i.e., each vertex and itself plus an edge), INIs an identity matrix;
is that
A degree matrix of (c); hlIs the active cell matrix of layer I; wlIs a parameter matrix for each layer.
Further, after the training is completed, the confidence of the relationship between the two entities can be judged by the following formula:
g(e1，R，e2)＝URf(e1WRe2+br)
where f is a standard non-linear function, e1、e2Is a knowledge representation g (e) of two different nodes in the knowledge-graph1，R，e2) Higher function score indicates entity e1And e2The higher the likelihood of being in the relationship R.
As a specific application example of this embodiment, as shown in fig. 4, the method for generating a plan based on real-time emergency data includes:
s1, acquiring current real-time emergency data;
s2, performing local optimal prediction on the current real-time emergency data based on a plan knowledge graph, and determining the first prediction plan;
s3, inputting the current real-time emergency data into a plan prediction model, and determining the second prediction plan, wherein the plan prediction model is obtained based on the plan knowledge graph training;
s4, judging whether the semantic analysis result of the first prediction solution is the same as the target semantic, if so, executing step S5; otherwise, go to step S6;
s5, taking the first prediction plan as a target plan;
s6, taking the second prediction plan as a target plan;
s7, updating the knowledge graph and the plan prediction model;
and S8, outputting the target plan.
In a specific scenario, for example in the field of civil aviation, the method may be used, specifically as follows:
suppose a fire disaster occurs in a terminal building, two rooms exist in the terminal building and need to be rescued, but only one room can be rescued, wherein the room A contains a large number of valuable instruments, the rescue difficulty is simple, the historical rescue success rate is high, and one passenger exists in the room B, but the rescue difficulty is difficult, and the historical rescue success rate is low. Although the plan a output by the global prediction is the rescue room a, the plan B generated by the local prediction is the rescue room B, but there is a target semantic "first rescuers" and the plan B is the same as the target semantic by comparison, so the plan B is set as the target plan.
The embodiment further provides a device for generating a plan based on real-time emergency data, which is used to implement the foregoing embodiments and preferred embodiments, and the description of the device is omitted. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.
The present embodiment provides a device for generating a plan based on real-time emergency data, as shown in fig. 5, including:
an obtaining module 51, configured to obtain current real-time emergency data;
the first processing module 52 is configured to perform local optimal prediction and global optimal prediction on the current real-time emergency data based on a plan knowledge graph, so as to obtain a first prediction plan and a second prediction plan, where the plan knowledge graph is constructed based on historical real-time emergency data information and corresponding historical plan information;
a second processing module 53, configured to perform semantic analysis on the first prediction plan and the second prediction plan, compare semantic analysis results with target semantics respectively, and determine a target plan from the first prediction plan and the second prediction plan
The real-time emergency data-based protocol generation apparatus in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC circuit, a processor and memory executing one or more software or fixed programs, and/or other devices that may provide the above-described functionality.
Further functional descriptions of the modules are the same as those of the corresponding embodiments, and are not repeated herein.
An embodiment of the present invention further provides an electronic device, which includes the device for generating a plan based on real-time emergency data shown in fig. 6.
Referring to fig. 6, fig. 6 is a schematic structural diagram of an electronic device according to an alternative embodiment of the present invention, and as shown in fig. 6, the electronic device may include: at least one processor 61, such as a CPU (Central Processing Unit), at least one communication interface 63, memory 64, at least one communication bus 62. Wherein a communication bus 62 is used to enable the connection communication between these components. The communication interface 63 may include a Display (Display) and a Keyboard (Keyboard), and the optional communication interface 63 may also include a standard wired interface and a standard wireless interface. The Memory 64 may be a high-speed RAM Memory (volatile Random Access Memory) or a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. The memory 64 may optionally be at least one memory device located remotely from the processor 61. Wherein the processor 61 may be in connection with the apparatus described in fig. 6, an application program is stored in the memory 64, and the processor 61 calls the program code stored in the memory 64 for performing any of the above-mentioned method steps.
The communication bus 62 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus 62 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 6, but this is not intended to represent only one bus or type of bus.
The memory 64 may include a volatile memory (RAM), such as a random-access memory (RAM); the memory may also include a non-volatile memory (english: non-volatile memory), such as a flash memory (english: flash memory), a hard disk (english: hard disk drive, abbreviated: HDD) or a solid-state drive (english: SSD); the memory 64 may also comprise a combination of the above types of memory.
The processor 61 may be a Central Processing Unit (CPU), a Network Processor (NP), or a combination of CPU and NP.
The processor 61 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a Programmable Logic Device (PLD), or a combination thereof. The PLD may be a Complex Programmable Logic Device (CPLD), a field-programmable gate array (FPGA), a General Array Logic (GAL), or any combination thereof.
Optionally, the memory 64 is also used to store program instructions. Processor 61 may invoke program instructions to implement a method for generating a protocol based on real-time emergency data as shown in any of the embodiments of the present application.
The embodiment of the invention also provides a non-transitory computer storage medium, wherein the computer storage medium stores computer executable instructions which can execute the method for generating the plan based on the real-time emergency data in any method embodiment. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk (Hard Disk Drive, abbreviated as HDD), a Solid State Drive (SSD), or the like; the storage medium may also comprise a combination of memories of the kind described above.
Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope defined by the appended claims.
Claims (10)
1. A method for generating a plan based on real-time emergency data is characterized by comprising the following steps:
acquiring current real-time emergency data;
respectively performing local optimal prediction and global optimal prediction on the current real-time emergency data based on a plan knowledge graph to obtain a first prediction plan and a second prediction plan, wherein the plan knowledge graph is constructed based on historical real-time emergency data information and corresponding historical plan information;
and performing semantic analysis on the first prediction plan and the second prediction plan, comparing semantic analysis results with target semantics respectively, and determining the target plan in the first prediction plan and the second prediction plan.
2. The method of claim 1, wherein the performing local optimal prediction and global optimal prediction on the current real-time emergency data based on a plan knowledge graph to obtain a first prediction plan and a second prediction plan comprises:
performing local optimal prediction on the current real-time emergency data based on a plan knowledge graph, and determining the first prediction plan;
and inputting the current real-time emergency data into a plan prediction model, and determining the second prediction plan, wherein the plan prediction model is obtained based on the plan knowledge graph training.
3. The method of claim 2, further comprising:
updating the plan knowledge graph based on the corresponding relation between the target plan and the current real-time emergency data;
and training the plan prediction model based on the updated plan knowledge graph, and determining the updated plan prediction model.
4. The method of claim 1, wherein performing semantic analysis on the first prediction plan and the second prediction plan and comparing the semantic analysis results with target semantics respectively, and determining a target plan in the first prediction plan and the second prediction plan comprises:
if the semantic analysis result of the first prediction plan is the same as the target semantic, taking the first prediction plan as a target plan;
otherwise, the second prediction plan is used as a target plan.
5. The method of claim 2, wherein the locally optimal prediction of the current real-time emergency data based on a plan knowledge-graph is performed, and wherein determining the first predicted plan comprises:
respectively matching the current real-time emergency data based on the feature words in the predetermined case knowledge graph to obtain the feature words corresponding to the current real-time emergency data;
and selecting the entity and the entity behavior corresponding to the characteristic words as the first prediction plan based on the entity and the entity behavior in the characteristic words and the plan knowledge graph.
6. The method of claim 1 or 4, further comprising:
if the semantic analysis result of the first prediction plan and the semantic analysis result of the first prediction plan are the same as the target semantic, comparing the semantic analysis results with standby target semantics respectively, and determining a target plan in the first prediction plan and the second prediction plan.
7. The method of claim 2, further comprising:
and taking the predetermined plan knowledge graph as a training set, and sending the predetermined plan knowledge graph into a deep neural network for training to obtain the predetermined plan prediction model.
8. A device for generating a plan based on real-time emergency data is characterized by comprising:
the acquisition module is used for acquiring current real-time emergency data;
the first processing module is used for respectively carrying out local optimal prediction and global optimal prediction on the current real-time emergency data based on a plan knowledge graph to obtain a first prediction plan and a second prediction plan, wherein the plan knowledge graph is constructed based on historical real-time emergency data information and corresponding historical plan information;
and the second processing module is used for performing semantic analysis on the first prediction plan and the second prediction plan, comparing semantic analysis results with target semantics respectively, and determining a target plan from the first prediction plan and the second prediction plan.
9. An electronic device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform the steps of the method of any one of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111490281.XA CN114117032A (en) | 2021-12-08 | 2021-12-08 | Method and device for generating plan based on real-time emergency data and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111490281.XA CN114117032A (en) | 2021-12-08 | 2021-12-08 | Method and device for generating plan based on real-time emergency data and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114117032A true CN114117032A (en) | 2022-03-01 |
Family
ID=80367380
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111490281.XA Pending CN114117032A (en) | 2021-12-08 | 2021-12-08 | Method and device for generating plan based on real-time emergency data and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114117032A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116561385A (en) * | 2023-07-10 | 2023-08-08 | 中国人民解放军军事科学院系统工程研究院 | Knowledge representation-based plan quick matching recommendation method |
| CN116681305A (en) * | 2023-06-05 | 2023-09-01 | 中国标准化研究院 | Emergency decision method based on knowledge graph |
| CN116934127A (en) * | 2023-09-19 | 2023-10-24 | 中国铁塔股份有限公司吉林省分公司 | Emergency plan generation method and system based on intelligent platform |
-
2021
- 2021-12-08 CN CN202111490281.XA patent/CN114117032A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116681305A (en) * | 2023-06-05 | 2023-09-01 | 中国标准化研究院 | Emergency decision method based on knowledge graph |
| CN116681305B (en) * | 2023-06-05 | 2024-04-26 | 中国标准化研究院 | Emergency decision method based on knowledge graph |
| CN116561385A (en) * | 2023-07-10 | 2023-08-08 | 中国人民解放军军事科学院系统工程研究院 | Knowledge representation-based plan quick matching recommendation method |
| CN116561385B (en) * | 2023-07-10 | 2023-10-13 | 中国人民解放军军事科学院系统工程研究院 | Knowledge representation-based plan quick matching recommendation method |
| CN116934127A (en) * | 2023-09-19 | 2023-10-24 | 中国铁塔股份有限公司吉林省分公司 | Emergency plan generation method and system based on intelligent platform |
| CN116934127B (en) * | 2023-09-19 | 2023-11-24 | 中国铁塔股份有限公司吉林省分公司 | Emergency plan generation method and system based on intelligent platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114117032A (en) | Method and device for generating plan based on real-time emergency data and electronic equipment | |
| CN111523640B (en) | Training method and device for neural network model | |
| JP6355683B2 (en) | Risk early warning method, apparatus, storage medium, and computer program | |
| WO2022105118A1 (en) | Image-based health status identification method and apparatus, device and storage medium | |
| US20230080230A1 (en) | Method for generating federated learning model | |
| CN108111399B (en) | Message processing method, device, terminal and storage medium | |
| CN114004210A (en) | Emergency plan generating method, system, equipment and medium based on neural network | |
| CN114942984A (en) | Visual scene text fusion model pre-training and image-text retrieval method and device | |
| CN113656587B (en) | Text classification method, device, electronic equipment and storage medium | |
| CN111566646A (en) | Electronic device for obfuscating and decoding data and method for controlling the same | |
| CN111143178A (en) | User behavior analysis method, device and equipment | |
| EP4105895A2 (en) | Human-object interaction detection method, neural network and training method therefor, device, and medium | |
| CN112926308A (en) | Method, apparatus, device, storage medium and program product for matching text | |
| CN113434722B (en) | Image classification method, device, equipment and computer readable storage medium | |
| CN114707002A (en) | Disaster emergency decision generation method, device, equipment and storage medium | |
| CN112528040B (en) | Detection method for guiding drive corpus based on knowledge graph and related equipment thereof | |
| CN112445914A (en) | Text classification method, device, computer equipment and medium | |
| CN110704614B (en) | Information processing method and device for predicting user group type in application | |
| CN116777646A (en) | Artificial intelligence-based risk identification method, apparatus, device and storage medium | |
| US20220207427A1 (en) | Method for training data processing model, electronic device and storage medium | |
| CN108961079A (en) | Insure the method, apparatus storage medium and electronic equipment of family's identification | |
| CN114091909A (en) | Collaborative development method, system, device and electronic equipment | |
| CN113312552A (en) | Data processing method, device, electronic equipment and medium | |
| CN113704256A (en) | Data identification method and device, electronic equipment and storage medium | |
| CN112906652A (en) | Face image recognition method and device, electronic equipment and storage medium |
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 |