CN112488896A - Emergency plan generation method and device, computer equipment and storage medium - Google Patents
Emergency plan generation method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN112488896A CN112488896A CN202011442500.2A CN202011442500A CN112488896A CN 112488896 A CN112488896 A CN 112488896A CN 202011442500 A CN202011442500 A CN 202011442500A CN 112488896 A CN112488896 A CN 112488896A
- Authority
- CN
- China
- Prior art keywords
- emergency
- plan
- emergency plan
- segment
- historical
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 230000001788 irregular Effects 0.000 claims abstract description 71
- 230000003068 static effect Effects 0.000 claims abstract description 28
- 238000002372 labelling Methods 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 13
- 239000012634 fragment Substances 0.000 claims description 41
- 239000013598 vector Substances 0.000 claims description 35
- 238000004590 computer program Methods 0.000 claims description 13
- 238000000605 extraction Methods 0.000 claims description 13
- 238000001914 filtration Methods 0.000 claims description 9
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 230000004044 response Effects 0.000 claims description 5
- 238000000638 solvent extraction Methods 0.000 claims 2
- 238000012545 processing Methods 0.000 abstract description 5
- 230000011218 segmentation Effects 0.000 description 14
- 230000008520 organization Effects 0.000 description 12
- 230000008569 process Effects 0.000 description 9
- 230000006870 function Effects 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 6
- 230000008878 coupling Effects 0.000 description 6
- 238000010168 coupling process Methods 0.000 description 6
- 238000005859 coupling reaction Methods 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000013507 mapping Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 108010069898 fibrinogen fragment X Proteins 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides an emergency plan generation method and device, computer equipment and a storage medium, relates to the technical field of data processing, and is used for improving the accuracy of emergency plan generation. The method mainly comprises the following steps: dividing the historical emergency plan into a plurality of emergency plan segments; acquiring a label set of each emergency plan segment through an emergency plan label labeling model, wherein the label set comprises a structural label set and a semantic label set, and the emergency plan label labeling model is obtained by training according to a sample emergency plan segment and a label set corresponding to the sample emergency plan segment; determining an emergency plan frame according to a Cartesian product of a structural tag set and a semantic tag set, wherein the emergency plan frame comprises a plurality of plan entries; determining the emergency plan section of each plan entry according to the calculated static correlation and dynamic correlation between each emergency plan section and the target irregular emergency; and generating an emergency plan corresponding to the target irregular emergency through the emergency plan segment of each plan entry.
Description
Technical Field
The present application relates to the field of data processing technologies, and in particular, to an emergency plan generation method and apparatus, a computer device, and a storage medium.
Background
The emergency plan is a full-period activity and system which is made for quickly and effectively preventing emergencies, responding to emergencies, disposing in the middle of the emergencies, recovering and rebuilding after the emergencies, and is in a core position in an emergency management system of China. In recent years, due to the fact that coupling between human activities is continuously enhanced, particularly, influence of the human activities on external environment is continuously enhanced, occurrence rules, internal mechanisms and external influence of emergencies are increasingly complex, a large number of unconventional emergencies are formed, and an existing emergency plan is difficult to effectively deal with.
The irregular emergency event often involves dynamic evolution of a regular emergency event and close coupling of multiple regular emergency events, which presents many challenges for generating a plan for the irregular emergency event. At present, the most similar cases are searched from a historical case base to serve as the basis for generating the emergency plan, and the non-conventional emergency events generally do not repeatedly occur in the history, so that the very similar cases are difficult to find in the historical case base, and the emergency plan is generated aiming at the non-conventional emergency events which have the characteristics of unobvious occurrence rule, multi-disaster dynamic coupling and the like.
Disclosure of Invention
The embodiment of the application provides an emergency plan generating method and device, computer equipment and a storage medium, which are used for generating an emergency plan for an unconventional emergency with the characteristics of unobvious occurrence rule, multi-disaster dynamic coupling and the like, and improving the accuracy of generating the emergency plan.
The embodiment of the invention provides an emergency plan generating method, which comprises the following steps:
dividing the historical emergency plan into a plurality of emergency plan segments;
acquiring a label set of each emergency plan segment through an emergency plan label labeling model, wherein the label set comprises a structural label set and a semantic label set, and the emergency plan label labeling model is obtained by training according to a sample emergency plan segment and a label set corresponding to the sample emergency plan segment;
determining an emergency plan frame according to the Cartesian product of the structural tag set and the semantic tag set, wherein the emergency plan frame comprises a plurality of plan entries;
determining the emergency plan section of each plan entry according to the calculated static correlation and dynamic correlation between each emergency plan section and the target irregular emergency;
and generating an emergency plan corresponding to the target irregular emergency through the emergency plan segment of each plan entry.
The embodiment of the invention provides an emergency plan generating device, which comprises:
the dividing module is used for dividing the historical emergency plan into a plurality of emergency plan segments;
the acquisition module is used for acquiring a label set of each emergency plan segment through an emergency plan label labeling model, wherein the label set comprises a structural label set and a semantic label set, and the emergency plan label labeling model is obtained by training according to a sample emergency plan segment and a label set corresponding to the sample emergency plan segment;
the determining module is used for determining an emergency plan frame according to the Cartesian product of the structural label set and the semantic label set, wherein the emergency plan frame comprises a plurality of plan entries;
the determining module is further used for determining the emergency plan section of each plan entry according to the calculated static correlation and dynamic correlation between each emergency plan section and the target irregular emergency;
and the generating module is used for generating an emergency plan corresponding to the target irregular emergency through the emergency plan segment of each plan entry.
A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the emergency plan generation method when executing the computer program.
A computer-readable storage medium storing a computer program which, when executed by a processor, implements the emergency plan generating method described above.
The invention provides an emergency plan generating method, an emergency plan generating device, computer equipment and a storage medium, wherein a historical emergency plan is divided into a plurality of emergency plan segments; acquiring a label set of each emergency plan segment through an emergency plan label labeling model, wherein the label set comprises a structural label set and a semantic label set, and then determining an emergency plan frame according to the Cartesian product of the structural label set and the semantic label set, wherein the emergency plan frame comprises a plurality of plan entries; determining the emergency plan section of each plan entry according to the calculated static correlation and dynamic correlation between each emergency plan section and the target irregular emergency; and generating an emergency plan corresponding to the target irregular emergency through the emergency plan segment of each plan entry. Compared with the method that the most similar cases are searched from the historical case library to serve as the basis for generating the emergency plans at present, the method divides the historical emergency plans into a plurality of emergency plan segments, then generates an emergency plan frame and the emergency plan segments of all plan entries in the emergency plan frame according to the emergency plan segments, and finally generates the emergency plans corresponding to the target abnormal emergency events through the emergency plan segments of all plan entries. The emergency plan generated by the invention is determined according to the emergency plan segments, and the emergency plan segments are formed by separating a plurality of historical emergency plans, so the emergency plan can be generated for unconventional emergency events with the characteristics of unobvious occurrence rule, multi-disaster dynamic coupling and the like, and the accuracy of generating the emergency plan is improved.
Drawings
Fig. 1 is a flowchart of an emergency plan generation method provided in the present application;
FIG. 2 is a flow chart of an emergency plan generation method provided by the present application;
FIG. 3 is a network architecture diagram of a scene information extraction model provided herein;
FIG. 4 is a flow chart of the matching degree calculation provided by the present application;
FIG. 5 is a flow chart of determining an emergency protocol section for each protocol entry provided herein;
FIG. 6 is a flow chart of determining an emergency protocol section for each protocol entry provided herein;
fig. 7 is a block diagram illustrating an emergency plan generating apparatus according to an embodiment of the present application;
FIG. 8 is a schematic diagram of a computer device provided in one embodiment of the present application.
Detailed Description
In order to better understand the technical solutions described above, the technical solutions of the embodiments of the present application are described in detail below with reference to the drawings and the specific embodiments, and it should be understood that the specific features of the embodiments and the embodiments of the present application are detailed descriptions of the technical solutions of the embodiments of the present application, and are not limitations of the technical solutions of the present application, and the technical features of the embodiments and the embodiments of the present application may be combined with each other without conflict.
First embodiment
Referring to fig. 1, a method for generating an emergency plan according to a first embodiment of the present invention is shown, and the method specifically includes steps S10-S50:
and step S10, dividing the historical emergency plan into a plurality of emergency plan segments.
In this embodiment, the goal of this step is to segment a complete historical contingency plan into segments. To achieve this goal, three segmentation schemes are specifically provided: coarse-grained segmentation, fine-grained segmentation, and mixed segmentation. Specifically, the historical emergency plan is divided according to natural sections to obtain a plurality of emergency plan segments, and each natural section forms one emergency plan segment; or the historical emergency plan is segmented according to punctuations to obtain a plurality of emergency plan segments, and each sentence forms one emergency plan segment; or the historical emergency plan is segmented according to the natural segments and the punctuations to obtain a plurality of emergency plan segments.
The coarse-grained division is to divide the historical emergency plan according to natural sections, and each natural section forms an emergency plan segment; and fine-grained division is to divide the historical emergency plan according to punctuations, and each sentence forms an emergency plan segment. Because the natural segment may be too long, the semantic information is not single, and the sentence may be over-broken, resulting in incomplete speech information, therefore further providing mixed segmentation, the specific scheme is: firstly, performing coarse-grained segmentation on an emergency plan, and if the length of an obtained natural segment is lower than a specified threshold value, directly taking the natural segment as a final segmentation result, otherwise, performing fine-grained segmentation on the natural segment; when fine-grained segmentation is carried out on the natural segment, firstly, a middle punctuation point is searched, compromise segmentation is carried out according to the punctuation point, if the length of a segmentation result is lower than a threshold value, segmentation is stopped, otherwise, the segmentation result is continuously segmented according to the fine-grained segmentation mode.
And step S20, acquiring a label set of each emergency plan segment through the emergency plan label labeling model.
The tag set comprises a structural tag set and a semantic tag set, the tag set reflects structural information and semantic information of the emergency plan fragment at the same time, and the tag set is formed in the following manner:
Γ＝Λ×Z
and Λ is a structural tag set, and Z is a semantic tag set, so as to encode semantic information of the emergency plan fragment. The cartesian products of the two constitute the fragment tag set. Lambda is mainly used for coding structural information of the emergency plan fragments, and Z is mainly used for coding semantic information of the emergency plan fragments.
For example: and the 'organization architecture primary title' in a Cartesian product set of the 'organization architecture primary title' and the 'organization architecture primary content' indicates that a certain emergency plan section is a part of the primary title of the organization architecture section.
In this embodiment, the emergency plan label labeling model is obtained by training according to the sample emergency plan segment and the label set corresponding to the sample emergency plan segment. The sample emergency plan fragments are obtained by separating the sample emergency plan fragments, and the specific separation mode is the same as the content described in the step S10; the label set corresponding to the sample emergency plan fragment also comprises a structural label set and a semantic label set, specifically labeled labels can be labeled manually, and then the sample emergency plan fragment and the label set corresponding to the sample emergency plan fragment are trained through a conditional random field model to obtain an emergency plan label labeling model.
Specifically, the emergency Plan tag labeling model may be abstractly represented as a mapping function f (Plan; Parameter) of the emergency Plan to the semantic tag sequence, where Plan (seg) is (Labels)1,…,segn) For a historical emergency plan consisting of n segments, each segment segi(1. ltoreq. i. ltoreq.n) is obtained by the step of dividing the emergency response plan, and Labels ═ l1,…,ln) For sequences consisting of n tags, each tag liE is lambada multiplied by Z (1 ≦ i ≦ n). The Parameter is a model Parameter and is estimated by a conditional random field model.
And step S30, determining an emergency plan frame according to the Cartesian product of the structural label set and the semantic label set.
For the embodiment of the present invention, the emergency plan frame includes a plurality of plan entries, and after the emergency plan fragment set is obtained, the emergency plan fragment set is reassembled to obtain a final new emergency plan, and the reassembling step includes: and selecting an emergency plan frame and selecting emergency plan contents.
The emergency plan framework is selected with the goal of constructing an overall structure for the new emergency plan. The tag set of the emergency plan fragment already contains the structural and semantic information of the emergency plan, and can be used as a basis for determining an emergency plan frame, and the specific method is to select part or all plan entries from a cartesian product set of the structural tag set and the semantic tag set as the emergency plan frame, for example, all plan entries are sorted according to the use frequency, and N plan entries ranked at the top are used as the emergency plan frame.
For example: Λ ═ primary title, primary content, secondary title, secondary content }, Z ═ organization architecture, leadership, participation organization, response hierarchy, response startup condition }, the cartesian product set of both has 4 × 5 ═ 20 plan entries, 5 of them can be selected as the emergency plan framework:
{ organization architecture primary title, leader organization secondary content, participation organization secondary title and participation machine, wherein the emergency plan generated needs to contain organization architecture primary chapters, the sections contain leader organization and participation organization secondary chapters, and each secondary chapter needs specific content.
And step S40, determining the emergency plan section of each plan entry according to the calculated static correlation and dynamic correlation of each emergency plan section and the target irregular emergency.
And step S50, generating an emergency plan corresponding to the target irregular emergency through the emergency plan segment of each plan entry.
Compared with the basis of searching the most similar case from a historical case library as the emergency plan generation at present, the emergency plan generation method divides the historical emergency plan into a plurality of emergency plan segments, then generates an emergency plan frame and the emergency plan segments of all plan entries in the emergency plan frame according to the emergency plan segments, and finally generates the emergency plan corresponding to the target irregular emergency event through the emergency plan segments of all plan entries. The emergency plan generated by the invention is determined according to the emergency plan segments, and the emergency plan segments are formed by separating a plurality of historical emergency plans, so the emergency plan can be generated for unconventional emergency events with the characteristics of unobvious occurrence rule, multi-disaster dynamic coupling and the like, and the accuracy of generating the emergency plan is improved.
Second embodiment
As shown in fig. 2, before dividing the historical emergency plan into a plurality of emergency plan segments, the emergency plan generating method according to the second embodiment of the present invention further includes:
step S201, a scene information quadruple set of the target irregular emergency is obtained through a scene information extraction model.
The input and output of the embodiment are presented in text form, namely, the aim is to generate a text-type emergency plan, the utilized resource is a large amount of text-type emergency plans which are already established at present, and the input information is also text description of the given unconventional emergency. The first premise that the existing emergency plan is applicable to the given irregular emergency is that the scene of the emergency event processed by the existing emergency plan is matched with the scene of the irregular emergency event, the first premise that the historical emergency plan is applicable to the target irregular emergency is that the scene of the emergency event processed by the historical emergency plan is matched with the scene of the target irregular emergency event, and for this reason, the first step of the embodiment is to extract the normalized scene information of the emergency event, that is, to obtain the scene information quadruple set of the target irregular emergency event.
In order to extract the emergency normalized scene information, firstly, a normalized description mode of the emergency scene information is designed. Specifically, the incident scene information is represented using a set of tuples:
S＝{T1,…,Tn}
Ti＝{Namei,Typei,Upperi,Loweri}(i＝1,…,n)
wherein n represents the number of attributes contained in the scene information quadruplet set, NameiIndicating attribute name, TypeiRepresenting the property type, UpperiRepresenting the upper bound of the attribute watch value, LoweriThe Lower limit (i is more than or equal to 1 and less than or equal to n) of the attribute monitoring value is expressed, and the Lower limit and the n both need to satisfy Loweri≤Upperi. For convenience of understanding, the embodiment provides a specific example of the normalized scene information of the emergency event: for the sentence "a non-passenger ship with 3000 tons or more collides with a ship and endangers the life safety of more than 10 and less than 30 persons" in an emergency plan, the corresponding quaternary set is S ═ a great face<Infinity, 3000 tons for non-passenger ship and disaster-bearing body>，<Life safety, disaster-bearing body, 30 persons, 10 persons>}。
Then, the scene information extraction model is trained according to the embodiment, and the scene information extraction model is obtained by training according to the historical emergency plan and the scene information quadruple set corresponding to the historical emergency plan, namely, the scene information quadruple labeling is performed on the historical emergency plan according to the above method, and then the model training is performed according to the labeled data to obtain the scene information extraction model. The model can be abstractly represented as a mapping function f (Text; Parameter) of a Text sequence to a tetrad set, wherein Text is a Text description of a sentence in an emergency plan or a target irregular emergency, the quadrate is a tetrad set of scene information contained in the Text description, and Parameter is a Parameter of a scene information extraction model.
Specifically, the scene information extraction model is estimated based on a deep neural network of an encoder-decoder structure, and the network architecture is shown in fig. 3. The input of the deep neural network for extracting the normalized scene information is an emergency scene description text formed by word sequences, then the emergency scene description text enters an encoder neural network to be encoded to obtain intermediate representation in a high-dimensional vector form, and then the intermediate representation information enters a decoder neural network to be decoded to obtain a final scene information quadruple set. In order to facilitate the control of the decoder on the Start and the End of the decoding process, two flags are added, four-tuple < Start, Null > and < End, Null >, indicating the Start of the decoding process when the decoder outputs the former and the End of the decoding process when the decoder outputs the latter.
The emergency normalized scene information extraction stage can be further subdivided into two sub-stages: extracting the normalized scene information of the emergency in the historical emergency plan and extracting the normalized scene information in the target non-conventional emergency description text. In the two sub-stages, the historical emergency plan and the description text of the target irregular emergency are both regarded as texts with the same format, and the scene information extraction model is used for realizing the two sub-stages in a unified manner.
Step S202, invalid plans in the historical emergency plans are filtered by calculating the matching degree of the scene information quadruple set of the historical emergency plans and the scene information quadruple set of the target irregular emergency.
Specifically, the filtering the invalid plans in the historical emergency plans by calculating the matching degree of the scene information quadruple set of the historical emergency plans and the scene information quadruple set of the target irregular emergency, includes:
1. by the formula Match (S, S') ═ ΣT′∈S′maxT∈S(Match (T, T')) calculating the matching degree of the scene information quadruple set of the historical emergency plan and the scene information quadruple set of the target irregular emergency;
2. determining the historical emergency plans with the matching degrees lower than a specified threshold value as invalid plans, and filtering the invalid plans;
s and S' respectively represent scene information quadruple sets of the historical emergency plan and the target abnormal emergency; match (T, T ') is used to calculate the matching degree between any two scene information quadruples T and T ' in S and S '.
The specific calculation flow is shown in fig. 4. Firstly, comparing Type fields of four-tuple of scene information, entering subsequent matching only when the types are the same, and otherwise, determining that the scene information is not matched; then, comparing the Name fields of the four-tuple of the two scene information, entering subsequent matching only when the names are the same, and otherwise, determining that the fields are not matched; finally, comparing the Upper field and the Lower field of the four-tuple of the two scene information, wherein the comparison result comprises four situations of the scene information, which are respectively: completely matching, namely the Upper field and the Lower field of the four-tuple of the two scene information are respectively and completely equal; matching is contained, namely an interval formed by a Lower field and an Upper field of one scene information quadruple is in an interval of another scene information quadruple; cross matching, namely forming an intersection relation between an interval formed by a Lower field and an Upper field of one scene information quadruple and an interval of another scene information quadruple; and the mismatch is that the interval formed by the Lower field and the Upper field of one scene information quadruple is completely irrelevant in the interval of another scene information quadruple. For the four matching results, A, B, C, D four values are returned to reflect the scores of the four matching conditions, and A, B, C, D four values are required to satisfy A > B or C ≧ 0> D.
After the matching score between any two scene information quadruplets T and T ' in S and S ' is obtained through calculation, according to a formula Match (S, S ') -sigmaT′∈S′maxT∈S(Match (T, T')) calculating the matching sum of S and SAnd scoring, wherein historical emergency plans below a specified threshold are identified as invalid emergency plans and other historical emergency plans are identified as candidate emergency plans for the target irregular emergency.
Third embodiment
As shown in fig. 5 and 6, the emergency plan generating method according to the third embodiment of the present invention, wherein the determining the emergency plan section of each plan entry according to the calculated static and dynamic correlations of each emergency plan section with the target irregular emergency event includes:
step 501, determining the emergency plan segment with the highest static relevance to the target irregular emergency as the emergency plan segment of the plan entry.
In this embodiment, after obtaining the valid historical emergency plan segment set for the target irregular emergency according to the second embodiment, the next step is to extract a subset that is closely related to the target irregular emergency. And specifically, the correlation between each emergency plan segment and the target irregular emergency is calculated, and the emergency plan segments with the correlation lower than a specified threshold are excluded from subsequent processing. To distinguish the correlation calculated in the subsequent stage, the correlation in this step is called static correlation.
The specific calculation method of the static correlation of the emergency plan fragment is as follows: let a certain emergency plan fragment or target unconventional emergency text be described as X ═ w1,…,wnIn which w1,…,wnFor each word therein, n being the number of words, an initial vector representation of X is first calculated as follows:
wherein, Pr (w)i) Is the word wiFrequency of occurrence, Pr (w) in the set of all contingency plan fragmentsi(ii) a l) is the word wiThe frequency of occurrence in all the emergency plan fragment sets marked as l labels is obtained by counting the fragment sets; vec (w)i) Is thatWord wiThe vector representation of (1) can be obtained through any word vector representation model such as word2vec, Glove and the like; the label is a label in the label set, and a is a model hyper-parameter specified in advance.
Further, after calculating the static correlation between each emergency plan segment and the target irregular emergency, the common part of all the vectors is subtracted from the initial vector to enhance the discrimination of the segment vector. The method comprises the following specific steps:
arranging initial vectors (organized by column vectors) of all the fragments in the effective emergency plan fragment set into a matrix M ═ vec0(X1),…,vec0(Xn)]Wherein n is the number of all fragments;
arranging the initial vectors (organized as column vectors) of all segments labeled l as a matrix Ml＝[vec0(X1),…,vec0(Xm)]Wherein m is the number of all fragments tagged with l;
calculate the first two principal component vectors u of M0And u1，MlFirst two principal component vectors v0And v1；
The final vector representation of a certain contingency plan segment X with label l is calculated as follows:
where λ is a hyper-parameter specified in advance.
And finally, calculating the correlation between the vector representation of the emergency plan fragment and the vector representation of the text description of the given abnormal emergency, wherein the specific calculation mode can be obtained by any vector distance function (such as vector cosine distance, vector Euclidean distance and the like). And taking the emergency plan sections with the correlation exceeding a specified threshold as candidate resources to enter subsequent steps for processing.
And finally, calculating the static correlation between the vector representation of the emergency plan fragment and the vector representation of the text description of the target irregular emergency, wherein the specific calculation mode can be obtained by any vector distance function (such as vector cosine distance, vector Euclidean distance and the like). And taking the emergency plan sections with the correlation exceeding a specified threshold as candidate resources to enter subsequent steps for processing.
And 502, excluding the emergency plan segment with the highest static correlation, and calculating the dynamic correlation between the remaining emergency plan segments and the target irregular emergency.
Specifically, the dynamic correlation of each emergency plan segment with the target irregular emergency event is calculated by the following formula:
wherein, the set of the emergency plan fragments is C ═ { C ═ C1,…,CnEach emergency plan fragment is Ci(1≤i≤n)，rel0(CiN) is an emergency plan section CiStatic correlation with newly selected Emergency plan fragment N, rel0(CiTarget) is an emergency plan section CiStatic relevance to Target non-conventional emergency text description Target, alpha and beta are specified hyper-parameters.
The basic idea of the dynamic correlation formula is that when a new emergency plan segment is selected as the content of the new emergency plan, if the similarity of the remaining emergency plan segments is high, the correlation of the remaining emergency plan segments should be properly reduced to prevent a large number of semantically repeated emergency plan segments from being continuously selected as the content of the new emergency plan; meanwhile, if the remaining emergency plan segments are from the same historical emergency plans, the relevance of the remaining emergency plan segments should be properly adjusted to be high so as to be beneficial to selecting the emergency plan segments with continuous contents as the contents of the new emergency plans. The importance of these two factors is regulated by α and β.
And step 503, determining the emergency plan segment with the highest dynamic relevance to the target irregular emergency as the emergency plan segment of the plan entry.
And 504, eliminating the emergency plan sections with the highest dynamic relevance, and calculating the dynamic relevance of the remaining emergency plan sections until the number of the emergency plan sections of the plan entry meets a specified value.
The new contingent plan segment selection process described above is repeated until a segment meeting the specified number is selected for the plan entry. The selection process is adopted for all the plan items determined in the emergency plan frame selection stage, and after all the plan items are selected, the emergency plan segments in the new content set form the content of the new emergency plan.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
In an embodiment, an emergency plan generating device is provided, and the emergency plan generating device corresponds to the emergency plan generating methods in the above embodiments one to one. As shown in fig. 7, the functional modules of the emergency plan generating device are described in detail as follows:
the division module 10 is used for dividing the historical emergency plan into a plurality of emergency plan segments;
an obtaining module 20, configured to obtain a tag set of each emergency plan fragment through an emergency plan tag labeling model, where the tag set includes a structural tag set and a semantic tag set, and the emergency plan tag labeling model fragment is obtained by training according to a sample emergency plan and a tag set corresponding to the sample emergency plan;
a determining module 30, configured to determine an emergency plan frame according to a cartesian product of the structural tag set and the semantic tag set, where the emergency plan frame includes a plurality of plan entries;
the determining module 30 is further configured to determine an emergency plan section of each plan entry according to the calculated static correlation and dynamic correlation between each emergency plan section and the target irregular emergency event;
and the generating module 40 is configured to generate an emergency plan corresponding to the target irregular emergency event through the emergency plan segment of each plan entry.
Further, the apparatus further comprises:
the obtaining module 20 is further configured to obtain a scene information quadruple set of the target irregular emergency through a scene information extraction model, where the scene information extraction model is obtained by training according to the historical emergency plan and the scene information quadruple set corresponding to the historical emergency plan;
and the filtering module is used for filtering invalid plans in the historical emergency plans by calculating the matching degree of the scene information quadruple set of the historical emergency plans and the scene information quadruple set of the target irregular emergency.
Specifically, the scene information quadruplet set includes an attribute name, an attribute type, an upper limit value and a lower limit value, and the filtering module is configured to:
by the formula Match (S, S') ═ ΣT′∈S′maxT∈S(Match (T, T')) calculating the matching degree of the scene information quadruple set of the historical emergency plan and the scene information quadruple set of the target irregular emergency;
determining the historical emergency plans with the matching degrees lower than a specified threshold value as invalid plans, and filtering the invalid plans;
s and S' respectively represent scene information quadruple sets of the historical emergency plan and the target abnormal emergency; match (T, T ') is used to calculate the matching degree between any two scene information quadruples T and T ' in S and S '.
Further, the dividing module 10 is specifically configured to:
dividing the historical emergency plan according to natural segments to obtain a plurality of emergency plan segments, wherein each natural segment forms one emergency plan segment; or
Dividing the historical emergency plan according to punctuations to obtain a plurality of emergency plan segments, wherein each sentence forms one emergency plan segment; or
And segmenting the historical emergency plan according to the natural sections and the punctuations to obtain a plurality of emergency plan segments.
Specifically, the determining module 30 is configured to:
determining an emergency plan segment with the highest static relevance to a target irregular emergency as an emergency plan segment of the plan entry;
excluding the emergency plan segment with the highest static correlation, and calculating the dynamic correlation between the remaining emergency plan segments and the target irregular emergency;
determining an emergency plan segment with the highest dynamic relevance to a target irregular emergency as an emergency plan segment of the plan entry;
and excluding the emergency plan section with the highest dynamic correlation, and calculating the dynamic correlation of the remaining emergency plan section emergency plan sections until the number of the emergency plan sections of the plan entry meets a specified numerical value.
By the formula
Calculating an initial vector of each emergency plan segment and the target irregular emergency;
calculating the static correlation of the emergency plan segment and the target irregular emergency according to the initial vector of the emergency plan segment and the initial vector of the target irregular emergency;
wherein the text description of the emergency response protocol fragment or the target irregular emergency is X ═ w1,…,wn}，w1,…,wnFor each word therein, n is the number of words, Pr (w)i) Is the word wiFrequency of occurrence, Pr (w) in the set of all contingency plan fragmentsi(ii) a l) is the word wiFrequency of occurrence in all the set of contingency plan segments labeled as l-tags, vec (w)i) Is the word wiThe label is a label in the label set, and a is a model hyper-parameter.
Arranging the initial vectors of all the segments in the emergency plan segmentsColumn is a matrix M ═ vec0(X1),…,vec0(Xn)]And n is the number of all fragments;
arranging the initial vectors of all the emergency plan segments with the labels of l into a matrix Ml＝[vec0(X1),…,vec0(Xm)]And m is the number of all fragments labeled with l;
calculate the first two principal component vectors u of M0And u1，MlFirst two principal component vectors v0And v1；
By the formula
And calculating a final vector of the emergency plan fragment X with the label of l, wherein the lambda is a specified hyper-parameter.
Calculating the dynamic correlation of each emergency plan segment and the target irregular emergency through a formula:
wherein, the set of the emergency plan fragments is C ═ { C ═ C1,…,CnEach emergency plan fragment is Ci(1≤i≤n)，rel0(CiN) is an emergency plan section CiStatic correlation with newly selected Emergency plan fragment N, rel0(CiTarget) is an emergency plan section CiStatic relevance to Target non-conventional emergency text description Target, alpha and beta are specified hyper-parameters.
For specific limitations of the emergency plan generating device, reference may be made to the above limitations of the emergency plan generating method, which are not described herein again. The various modules in the above-described apparatus may be implemented in whole or in part by software, hardware, and combinations thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 8. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement an emergency protocol generation method.
In one embodiment, a computer device is provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
dividing the historical emergency plan into a plurality of emergency plan segments;
acquiring a label set of each emergency plan segment through an emergency plan label labeling model, wherein the label set comprises a structural label set and a semantic label set, and the emergency plan label labeling model is obtained by training according to a sample emergency plan segment and a label set corresponding to the sample emergency plan segment;
determining an emergency plan frame according to the Cartesian product of the structural tag set and the semantic tag set, wherein the emergency plan frame comprises a plurality of plan entries;
determining the emergency plan section of each plan entry according to the calculated static correlation and dynamic correlation between each emergency plan section and the target irregular emergency;
and generating an emergency plan corresponding to the target irregular emergency through the emergency plan segment of each plan entry.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
dividing the historical emergency plan into a plurality of emergency plan segments;
acquiring a label set of each emergency plan segment through an emergency plan label labeling model, wherein the label set comprises a structural label set and a semantic label set, and the emergency plan label labeling model is obtained by training according to a sample emergency plan segment and a label set corresponding to the sample emergency plan segment;
determining an emergency plan frame according to the Cartesian product of the structural tag set and the semantic tag set, wherein the emergency plan frame comprises a plurality of plan entries;
determining the emergency plan section of each plan entry according to the calculated static correlation and dynamic correlation between each emergency plan section and the target irregular emergency;
and generating an emergency plan corresponding to the target irregular emergency through the emergency plan segment of each plan entry.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (10)
1. An emergency plan generating method, comprising:
dividing the historical emergency plan into a plurality of emergency plan segments;
acquiring a label set of each emergency plan segment through an emergency plan label labeling model, wherein the label set comprises a structural label set and a semantic label set, and the emergency plan label labeling model is obtained by training according to a sample emergency plan segment and a label set corresponding to the sample emergency plan segment;
determining an emergency plan frame according to the Cartesian product of the structural tag set and the semantic tag set, wherein the emergency plan frame comprises a plurality of plan entries;
determining the emergency plan section of each plan entry according to the calculated static correlation and dynamic correlation between each emergency plan section and the target irregular emergency;
and generating an emergency plan corresponding to the target irregular emergency through the emergency plan segment of each plan entry.
2. The emergency protocol generation method of claim 1, wherein prior to the partitioning of the historical emergency protocol into a plurality of emergency protocol segments, the method further comprises:
acquiring a scene information quadruple set of a target irregular emergency through a scene information extraction model, wherein the scene information extraction model is obtained by training according to a historical emergency plan and a scene information quadruple set corresponding to the historical emergency plan;
and filtering invalid plans in the historical emergency plans by calculating the matching degree of the scene information quadruple set of the historical emergency plans and the scene information quadruple set of the target irregular emergency.
3. The emergency plan generating method of claim 2, wherein the scene information quadruplet set comprises an attribute name, an attribute type, an upper limit value and a lower limit value, and the filtering the invalid plan in the historical emergency plan by calculating a matching degree between the scene information quadruplet set of the historical emergency plan and the scene information quadruplet set of the target irregular emergency event comprises:
by the formula Match (S, S') ═ ΣT′∈S′maxT∈S(Match (T, T')) calculating the matching degree of the scene information quadruple set of the historical emergency plan and the scene information quadruple set of the target irregular emergency;
determining the historical emergency plans with the matching degrees lower than a specified threshold value as invalid plans, and filtering the invalid plans;
s and S' respectively represent scene information quadruple sets of the historical emergency plan and the target abnormal emergency; match (T, T ') is used to calculate the matching degree between any two scene information quadruples T and T ' in S and S '.
4. The emergency protocol generation method of claim 1, wherein the partitioning of the historical emergency protocol into a plurality of emergency protocol segments comprises:
dividing the historical emergency plan according to natural segments to obtain a plurality of emergency plan segments, wherein each natural segment forms one emergency plan segment; or
Dividing the historical emergency plan according to punctuations to obtain a plurality of emergency plan segments, wherein each sentence forms one emergency plan segment; or
And segmenting the historical emergency plan according to the natural sections and the punctuations to obtain a plurality of emergency plan segments.
5. The emergency protocol generation method of claim 2, wherein determining the emergency protocol segment for each protocol entry according to the calculated static and dynamic correlations of each emergency protocol segment with the target irregular emergency event comprises:
determining an emergency plan segment with the highest static relevance to a target irregular emergency as an emergency plan segment of the plan entry;
excluding the emergency plan segment with the highest static correlation, and calculating the dynamic correlation between the remaining emergency plan segments and the target irregular emergency;
determining an emergency plan segment with the highest dynamic relevance to a target irregular emergency as an emergency plan segment of the plan entry;
and excluding the emergency plan section with the highest dynamic correlation, and calculating the dynamic correlation of the remaining emergency plan section emergency plan sections until the number of the emergency plan sections of the plan entry meets a specified numerical value.
6. The emergency protocol generation method of claim 5, wherein before determining the emergency protocol segment for each protocol entry based on the calculated static and dynamic correlations of each emergency protocol segment with the target irregular emergency event, the method further comprises:
by the formula
Calculating an initial vector of each emergency plan segment and the target irregular emergency;
calculating the static correlation of the emergency plan segment and the target irregular emergency according to the initial vector of the emergency plan segment and the initial vector of the target irregular emergency;
wherein the text description of the emergency response protocol fragment or the target irregular emergency is X ═ w1,…,wn}，w1,…,wnFor each word therein, n is the number of words, Pr (w)i) Is the word wiFrequency of occurrence, Pr (w) in the set of all contingency plan fragmentsi(ii) a l) is the word wiFrequency of occurrence in all the set of contingency plan segments labeled as l-tags, vec (w)i) Is the word wiThe label is a label in the label set, and a is a model hyper-parameter.
7. The emergency protocol generation method of claim 5, wherein before determining the emergency protocol segment for each protocol entry based on the calculated static and dynamic correlations of each emergency protocol segment with the target irregular emergency event, the method further comprises:
calculating the dynamic correlation of each emergency plan segment and the target irregular emergency through a formula:
wherein, the set of the emergency plan fragments is C ═ { C ═ C1,…,CnEach emergency plan fragment is Ci(1≤i≤n)，rel0(CiN) is an emergency plan section CiStatic correlation with newly selected Emergency plan fragment N, rel0(CiTarget) is an emergency plan section CiIs not related to the targetThe conventional emergency text describes the static relevance of Target, and alpha and beta are designated hyper-parameters.
8. An emergency plan generating apparatus, comprising:
the dividing module is used for dividing the historical emergency plan into a plurality of emergency plan segments;
the acquisition module is used for acquiring a label set of each emergency plan segment through an emergency plan label labeling model, wherein the label set comprises a structural label set and a semantic label set, and the emergency plan label labeling model segment is obtained by training according to a sample emergency plan and a label set corresponding to the sample emergency plan;
the determining module is used for determining an emergency plan frame according to the Cartesian product of the structural label set and the semantic label set, wherein the emergency plan frame comprises a plurality of plan entries;
the determining module is further used for determining the emergency plan section of each plan entry according to the calculated static correlation and dynamic correlation between each emergency plan section and the target irregular emergency;
and the generating module is used for generating an emergency plan corresponding to the target irregular emergency through the emergency plan segment of each plan entry.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the emergency plan generation method of any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, carries out the emergency plan generating method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011442500.2A CN112488896B (en) | 2020-12-08 | 2020-12-08 | Emergency plan generation method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011442500.2A CN112488896B (en) | 2020-12-08 | 2020-12-08 | Emergency plan generation method and device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112488896A true CN112488896A (en) | 2021-03-12 |
| CN112488896B CN112488896B (en) | 2022-11-04 |
Family
ID=74941143
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011442500.2A Active CN112488896B (en) | 2020-12-08 | 2020-12-08 | Emergency plan generation method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112488896B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113222361A (en) * | 2021-04-25 | 2021-08-06 | 北京中交华安科技有限公司 | Emergency rescue scheme generation method for emergency |
| CN113780785A (en) * | 2021-09-01 | 2021-12-10 | 上海智眭科技有限公司 | Plan management system and management method thereof |
| CN113822546A (en) * | 2021-09-01 | 2021-12-21 | 上海智眭科技有限公司 | Automatic plan management system and management method thereof |
| CN115204752A (en) * | 2022-09-13 | 2022-10-18 | 深圳市城市公共安全技术研究院有限公司 | Emergency handling scheme generation method, system, device and storage medium |
| CN116739871A (en) * | 2023-08-14 | 2023-09-12 | 深圳星网信通科技股份有限公司 | Emergency plan updating method, emergency plan updating device and storage medium |
| 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 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184217A (en) * | 2011-05-04 | 2011-09-14 | 西安电子科技大学 | Emergency plan generating system and method |
| CN107945082A (en) * | 2017-10-09 | 2018-04-20 | 中国电子科技集团公司第二十八研究所 | A kind of emergency preplan generation method and system |
| CN108665141A (en) * | 2018-04-03 | 2018-10-16 | 山东科技大学 | A method of extracting emergency response procedural model automatically from accident prediction scheme |
| CN108874078A (en) * | 2018-09-04 | 2018-11-23 | 江苏警官学院 | A kind of intelligent emergency program management system |
| CN111177284A (en) * | 2019-12-31 | 2020-05-19 | 清华大学 | Emergency plan model generation method, device and equipment |
-
2020
- 2020-12-08 CN CN202011442500.2A patent/CN112488896B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184217A (en) * | 2011-05-04 | 2011-09-14 | 西安电子科技大学 | Emergency plan generating system and method |
| CN107945082A (en) * | 2017-10-09 | 2018-04-20 | 中国电子科技集团公司第二十八研究所 | A kind of emergency preplan generation method and system |
| CN108665141A (en) * | 2018-04-03 | 2018-10-16 | 山东科技大学 | A method of extracting emergency response procedural model automatically from accident prediction scheme |
| CN108874078A (en) * | 2018-09-04 | 2018-11-23 | 江苏警官学院 | A kind of intelligent emergency program management system |
| CN111177284A (en) * | 2019-12-31 | 2020-05-19 | 清华大学 | Emergency plan model generation method, device and equipment |
Non-Patent Citations (1)
| Title |
|---|
| 刘 彤 等: ""基于SIF-Fasttext算法的应急预案快速生成"", 《中国科技论文》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113222361A (en) * | 2021-04-25 | 2021-08-06 | 北京中交华安科技有限公司 | Emergency rescue scheme generation method for emergency |
| CN113780785A (en) * | 2021-09-01 | 2021-12-10 | 上海智眭科技有限公司 | Plan management system and management method thereof |
| CN113822546A (en) * | 2021-09-01 | 2021-12-21 | 上海智眭科技有限公司 | Automatic plan management system and management method thereof |
| CN115204752A (en) * | 2022-09-13 | 2022-10-18 | 深圳市城市公共安全技术研究院有限公司 | Emergency handling scheme generation method, system, device and storage medium |
| CN116739871A (en) * | 2023-08-14 | 2023-09-12 | 深圳星网信通科技股份有限公司 | Emergency plan updating method, emergency plan updating device and storage medium |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112488896B (en) | 2022-11-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112488896B (en) | Emergency plan generation method and device, computer equipment and storage medium | |
| CN111222305B (en) | Information structuring method and device | |
| CN111444723B (en) | Information extraction method, computer device, and storage medium | |
| CN112015900B (en) | Medical attribute knowledge graph construction method, device, equipment and medium | |
| CN112084381A (en) | Event extraction method, system, storage medium and equipment | |
| EP3312736A1 (en) | Apparatus program & method for data property recognition | |
| CN112580346B (en) | Event extraction method and device, computer equipment and storage medium | |
| CN110309504B (en) | Text processing method, device, equipment and storage medium based on word segmentation | |
| CN110543637A (en) | Chinese word segmentation method and device | |
| CN113282729B (en) | Knowledge graph-based question and answer method and device | |
| CN113127626B (en) | Recommendation method, device, equipment and readable storage medium based on knowledge graph | |
| CN112766319A (en) | Dialogue intention recognition model training method and device, computer equipment and medium | |
| CN115827819A (en) | Intelligent question and answer processing method and device, electronic equipment and storage medium | |
| CN113806493A (en) | Entity relationship joint extraction method and device for Internet text data | |
| CN111967253A (en) | Entity disambiguation method and device, computer equipment and storage medium | |
| CN115203372A (en) | Text intention classification method and device, computer equipment and storage medium | |
| CN117291192B (en) | Government affair text semantic understanding analysis method and system | |
| CN117131868A (en) | Document-level entity relationship-oriented joint extraction method and device based on two stages of table-graph | |
| CN114548325B (en) | Zero sample relation extraction method and system based on dual contrast learning | |
| CN116186223A (en) | Financial text processing method, device, equipment and storage medium | |
| US12050557B2 (en) | Computerized systems and methods of data compression | |
| US20190188270A1 (en) | Generating an executable code based on a document | |
| CN115879460A (en) | Method, device, equipment and medium for identifying new label entity facing text content | |
| CN114372157A (en) | Insurance semantic understanding method, device, equipment and medium based on knowledge graph | |
| Silva et al. | A new approach to the pos tagging problem using evolutionary computation |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |