CN115952297A - Subway fire scene-oriented emergency plan knowledge graph construction method and device - Google Patents
Subway fire scene-oriented emergency plan knowledge graph construction method and device Download PDFInfo
- Publication number
- CN115952297A CN115952297A CN202211614429.0A CN202211614429A CN115952297A CN 115952297 A CN115952297 A CN 115952297A CN 202211614429 A CN202211614429 A CN 202211614429A CN 115952297 A CN115952297 A CN 115952297A
- Authority
- CN
- China
- Prior art keywords
- entity
- fire
- data set
- subway
- emergency plan
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a subway fire scene oriented emergency plan knowledge graph construction method and device. According to the invention, the corpus of the subway fire emergency plan is constructed by labeling the unstructured text data set, and the purposes of extracting and sorting related text semantic data of the subway fire by using a named entity identification method are achieved. Based on the method, a BilSTM-CRF method is used for conducting named entity identification on a material library to form an entity data set, the mutual relation between entity data attributes and entities in the entity data set is defined, a triple is established based on the entity data set and the mutual relation between the entity data attributes and the entities in the entity data set to establish an emergency plan knowledge map facing a subway fire scene, namely the emergency plan knowledge map facing the subway fire scene is established according to the extracted entity set, the potential relation and the organization mode of emergency information can be deeply mined, and the requirements of emergency treatment and accuracy of subway emergency events are met.
Description
Technical Field
The invention relates to the field of knowledge graph construction, in particular to a subway fire scene oriented emergency plan knowledge graph construction method and device.
Background
The subway emergency usually has the characteristics of emergency, urgency, unpredictability, uncertainty in evolution and the like, so that the emergency plan of the subway emergency is particularly important. However, currently, most emergency cases and emergency plans of subway emergencies are stored in the form of common documents, information integration and information combing are lacked, related workers are difficult to comprehensively master information resources, and the emergency cases and the emergency plans cannot quickly respond and give appropriate emergency treatment decisions when the emergencies occur. At present, a knowledge graph stores and expresses knowledge by using a graph structure, case texts are processed in a more efficient mode to visually present the interrelation between entities, and the knowledge graph is widely applied to the fields of medical treatment, historical data, finance and the like, but related applications are not developed in subway emergency plans.
Disclosure of Invention
The invention aims to provide a subway fire scene-oriented emergency plan knowledge map construction method and device, which can deeply mine the potential relation and organization mode of emergency information and meet the requirements of urgency and accuracy of emergency treatment of subway emergency.
In order to achieve the purpose, the invention provides the following scheme:
an emergency plan knowledge graph construction method for a subway fire scene comprises the following steps:
acquiring an unstructured text data set of a subway fire emergency plan; the unstructured text data set comprises: text information resources of the subway fire emergency plan and text data of the subway fire emergency plan; the text data includes: emergency treatment principle and treatment regulation of subway fire;
carrying out text annotation on the unstructured text data set to construct a corpus of a subway fire emergency plan;
carrying out named entity identification on the corpus by using a BilSTM-CRF method to form an entity data set, and defining the correlation between entity data attributes and entities in the entity data set;
establishing triples based on the entity data sets and interrelationships between entity data attributes and entities in the entity data sets;
and constructing an emergency plan knowledge graph facing the subway fire scene based on the triples.
Preferably, the text labeling is performed on the unstructured text data set to construct a corpus of subway fire emergency plans, and the method specifically includes:
determining an entity concept related to an emergency plan of a subway fire scene; the entity concepts comprise: fire, fire department, fire process evaluation and fire process personnel;
and performing text annotation on the unstructured text data set according to the entity concept to construct a corpus of the subway fire emergency plan.
Preferably, a BMEO labeling method is adopted to perform text labeling on the unstructured text data set according to the entity concept to construct a corpus of subway fire emergency plans; wherein, B represents the beginning of the entity concept, M represents the middle part of the entity concept, E represents the end of the entity concept, and O represents the non-entity concept.
Preferably, conducting named entity recognition on the corpus by using a BilSTM-CRF method to form an entity data set, specifically comprising:
performing word vector mapping on an entity concept obtained by performing text labeling on the unstructured text data set to obtain a vector matrix;
taking the vector matrix as the input of a BilSTM-CRF model to obtain an output result; the BilSTM-CRF model comprises the following components: two-way long-time and short-time memory network model and CRF layer; the bidirectional long-short time memory network model comprises a forward long-short time memory network model and a backward long-short time memory network model;
and finishing named entity recognition of the corpus based on the output result to form an entity data set.
Preferably, the vector matrix is used as an input of the BiLSTM-CRF model to obtain an output result, and specifically includes:
inputting each vector matrix into a forward long-short time memory network model and a backward long-short time memory network model in sequence to obtain a first output result and a second output result;
splicing the first output result and the second output result into a long input sequence from left to right and from right to left respectively, transmitting the obtained output probability serving as input to the CRF layer, and converting the output probability into a numerical value for output; this value is output as the output result.
Preferably, defining the interrelationship between the entity data attributes and the entities in the entity data set comprises:
defining a relationship between a fire and a fire department as an inclusion relationship;
defining the relation between the fire and the fire disposal process as a concurrency relation;
defining the relation between the fire disposal flow and the fire disposal evaluation as a sequence relation;
the relationship between the fire management personnel and the fire management department is defined as a subordinate relationship.
Preferably, the entity data set and the correlation between the entity data attribute and the entity in the entity data set are subjected to data extraction and data fusion to establish the triple; the triplet includes: entity-entity relationship-entity and entity-attribute relationship-entity.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
according to the subway fire scene oriented emergency plan knowledge graph construction method, text labeling is carried out on the unstructured text data set, a corpus of subway fire emergency plans is constructed, and the purpose of extracting and sorting related text semantic data of subway fires by using a named entity recognition method can be achieved. Based on the method, a BilSTM-CRF method is used for conducting named entity identification on a material library to form an entity data set, the mutual relation between entity data attributes and entities in the entity data set is defined, a triple is established based on the entity data set and the mutual relation between the entity data attributes and the entities in the entity data set, then an emergency plan knowledge graph facing a subway fire scene is established based on the triple, the purpose of establishing the emergency plan knowledge graph facing the subway fire scene according to the extracted entity set is achieved, further the potential relation and the organization mode of emergency information can be deeply mined, the requirements of urgency and accuracy of emergency treatment of subway emergency events are met, and the method has important practical and theoretical values for safety emergency management.
Corresponding to the method for constructing the subway fire scene oriented emergency plan knowledge graph, the invention also provides a subway fire scene oriented emergency plan knowledge graph constructing device, which comprises the following steps:
the data set acquisition module is used for acquiring an unstructured text data set of the subway fire emergency plan; the unstructured text data set comprises: text information resources of the subway fire emergency plan and text data of the subway fire emergency plan; the text data includes: emergency treatment principle and treatment regulation of subway fire;
the corpus construction module is used for carrying out text annotation on the unstructured text data set and constructing a corpus of the subway fire emergency plan;
the entity data determining module is used for carrying out named entity identification on the corpus by using a BilSTM-CRF method to form an entity data set and defining the correlation between entity data attributes and entities in the entity data set;
the triple construction module is used for establishing a triple based on the entity data set and the correlation between the entity data attribute and the entity in the entity data set;
and the knowledge map construction module is used for constructing an emergency plan knowledge map facing the subway fire scene based on the triples.
The technical effect achieved by the device provided by the invention is the same as that achieved by the method for constructing the knowledge map of the emergency plan for the subway fire scene provided by the invention, so that the method is not repeated herein.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a flow chart of an emergency plan knowledge graph construction method for a subway fire scene, provided by the invention;
fig. 2 is a schematic diagram of data transmission of a bidirectional long and short term memory network according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of the emergency plan knowledge graph construction device for the subway fire scene provided by the invention.
Detailed Description
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 only a part of the embodiments of the present invention, and not all of the embodiments. 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.
Based on the related cases of the existing subway fire scene and the adopted emergency plan, the related text semantic data of the subway fire is extracted and arranged by using a named entity identification method, and an emergency plan knowledge graph facing the subway fire scene is set up according to the extracted entity set, so that the potential relation and organization mode of emergency information can be deeply mined, the requirements on urgency and accuracy of emergency treatment of subway emergency events are met, and the method has very important practical and theoretical values for safe emergency management. Therefore, the invention provides an emergency plan knowledge graph construction method and device for a subway fire scene.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
As shown in fig. 1, the method for constructing a knowledge graph of an emergency plan for a fire scene of a subway, provided by the invention, comprises the following steps:
s1: unstructured text data sets relating to subway fire emergency plans are collected.
S1.1: the method for collecting the related text information resources of the subway fire emergency plan mainly comprises the following steps: fire emergency disposal flow, fire emergency disposal evaluation, basic setting data and the like.
S1.2: collect text data such as subway conflagration emergency treatment principle, processing regulation, mainly include: a fire emergency operation specification manual, a fire emergency operation regulation and system and the like.
S2: and carrying out text annotation on the unstructured text data set and constructing a corpus of the subway fire emergency plan.
S2.1: related entity concepts related to the subway fire scene emergency plan are determined, and the concepts mainly comprise fire, fire disposal departments, fire disposal processes and the like.
S2.2: and according to the entity concept determined in the step S2.1, carrying out text annotation on the text data collected in the step S1 and constructing a corpus of the subway fire emergency plan. The specific labeling method is as follows:
and (3) carrying out entity concept labeling by adopting a BMEO labeling method, wherein B represents the beginning of the entity concept, M represents the middle part of the entity concept, E represents the end of the entity concept, and O represents a non-entity concept. For example, event subjects, event handling personnel, event occurrence locations, event handling organizations, event handling devices, and the like are labeled in the text dataset using labels, such as OBJE, NAME, PLACE, ORG, EQUI, and the like, respectively.
S3: and (3) naming entity identification is carried out on the corpus by using a BilSTM-CRF method to form a corresponding entity data set, and the attribute and the interrelation between the entities are defined.
S3.1: and carrying out named entity identification on the corpus by using a BilSTM-CRF method to form a corresponding entity data set. The specific operation steps are as follows:
firstly, performing word vector mapping on the entity concept marked in the step S2, and initializing a random matrix of n x d, wherein n is the length of the matrix, namely the length of the semantic data to be identified, d is the dimension of the attribute vector used for representing each element in the semantic data, and the formed vector matrix of n x d is used as the input of the model.
Secondly, a bidirectional Long-Short term memory network (LSTM) formed by adding a forward Long-Short term memory network (LSTM) and a backward LSTM, i.e. a BiLSTM, is specifically shown in fig. 2. And (3) sequentially and respectively inputting each (n, d) in the previous step into a forward LSTM network and a backward LSTM network, splicing output results into a long input sequence from left to right and from right to left, and transmitting output probability calculated by a BilSTM layer as input to a CRF (Conditional Random Fields) layer.
The neural elements of the LSTM mainly comprise an input threshold, a forgetting threshold, an output threshold and historical time transfer permission characteristics, wherein the operation flow of the forward LSTM or the backward LSTM is as follows:
(1) Inputting the sequence characteristic x of the current time interval t t And the predicted sequence characteristic h of the previous time interval t-1 And after weighted summation, calculating by a sigmoid function to obtain forgetting weight output, namely forgetting gate.
(2) And calculating the input sequence characteristic information by using a sigmoid function, and then updating the current sequence characteristic information to obtain input weight output, namely an input gate.
(3) Characterizing the sequence by x t And the predicted sequence characteristic h of the previous time interval t-1 After weighting, the real-time sequence characteristics of the current time interval are obtained through tanh function processing
(4) Characterizing real-time sequences
And historical sequence feature C t-1 Are respectively related to the input weight i t And a forgetting weight f t Weighted summation is carried out to obtain the historical sequence characteristic C t 。
(5) Inputting tanh function to predict sequence characteristic predicted value tanh (C) of next time interval t ) And by inputting the weight o t Correcting to obtain a predicted time sequence characteristic h t 。
h t ＝o t tanh(C t )
In the formula: f. of t 、i t And o t Respectively the output of the forgetting threshold, the input threshold and the output threshold,
and &>
Is x t Output weights on three thresholds, <' > based on the sum of the weights>
And &>
Is h t-1 Output weights at three thresholds, <' >>
And &>
Respectively real-time sequence characteristic>
Weight of (a), b f 、b i 、b o And b C To calculate the deviation between three thresholds and the real-time sequence feature.
Then, the probability value of the BilSTM output is converted into a numerical value to be output after being calculated by a CRF layer, and the specific calculation formula is as follows:
in the formula: for a given sequence x, the sequence notation result is y, Z (x) is a normalization factor, emit (x) i ,y i ) Is the output probability of LSTM, trans (y) i-1 ,y i ) And P (y | x) is the numerical value output after the calculation and conversion of the CRF layer.
And finally, finishing named entity recognition to form a corresponding entity data set.
S3.2: on the basis of step S3.1, entity attributes are defined. The entity attributes in the data set of step S2 are defined as follows:
(1) And (3) fire hazard: fire type, fire class, fire location, impact area, etc.
(2) And a fire disaster treatment department: department names, department functions, department constituents, personnel names, etc.
(3) Fire hazard disposal flow: process name, process content, process progress, etc.
S3.3: on the basis of step S3.1 and step S3.2, the interrelationship between the entity attributes is determined. The specific relationship is determined as follows: the fire and fire treatment departments are in an inclusion relationship, the fire and fire treatment processes are in a concurrent relationship, the fire treatment processes and the fire treatment evaluations are in a sequential relationship, and the fire treatment personnel and the fire treatment departments are in a subordinate relationship, and the like.
S4: and performing data extraction and data fusion on the data source of the emergency plan of the subway fire scene so as to complete the establishment of the triples.
S4.1: and (4) performing data extraction and data fusion on the entity, attribute and relationship data set involved in the step (S3).
S4.2: and establishing triples of entity-entity relation-entity, entity-attribute relation-entity and the like, so as to facilitate the query and storage of data.
S5: and constructing an emergency plan knowledge graph facing the subway fire scene.
And based on the triples in the step S4, mapping the correlated knowledge structures into nodes and edges of the knowledge graph, and establishing the subway fire scene oriented emergency plan knowledge graph.
Based on the description, the subway fire emergency plan knowledge graph is constructed, decision support is provided for development and implementation of the subway fire emergency plan, a corpus for the subway fire emergency plan is also established, ordered organization and storage of emergency plan data in the subway fire scene are achieved, and a data foundation and a construction framework are provided for construction of the same type of subway emergency plan knowledge graph.
In addition, corresponding to the method for constructing the emergency plan knowledge map for the fire scene of the subway provided by the invention, the invention also provides a device for constructing the emergency plan knowledge map for the fire scene of the subway, as shown in fig. 3, the device comprises: the system comprises a data set acquisition module 1, a corpus construction module 2, an entity data determination module 3, a triple construction module 4 and a knowledge graph construction module 5.
The data set acquisition module 1 is used for acquiring an unstructured text data set of a subway fire emergency plan. The unstructured text data set includes: text information resources of the subway fire emergency plan and text data of the subway fire emergency plan. The text data includes: emergency treatment principle and treatment regulation for subway fire.
The corpus building module 2 is used for carrying out text annotation on the unstructured text data set and building a corpus of the subway fire emergency plan.
The entity data determination module 3 is used for carrying out named entity identification on the material base by using a BilSTM-CRF method to form an entity data set, and defining the entity data attribute and the interrelation between entities in the entity data set.
The triple construction module 4 is configured to establish a triple based on the entity data set and the entity data attribute in the entity data set and the correlation between the entities.
The knowledge map building module 5 is used for building an emergency plan knowledge map facing the subway fire scene based on the triples.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the description of the method part.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (8)
1. A subway fire scene oriented emergency plan knowledge graph construction method is characterized by comprising the following steps:
acquiring an unstructured text data set of a subway fire emergency plan; the unstructured text data set comprises: text information resources of the subway fire emergency plan and text data of the subway fire emergency plan are acquired; the text data includes: emergency treatment principle and treatment regulation of subway fire;
carrying out text annotation on the unstructured text data set to construct a corpus of subway fire emergency plans;
carrying out named entity identification on the corpus by using a BilSTM-CRF method to form an entity data set, and defining the correlation between entity data attributes and entities in the entity data set;
establishing triples based on the entity data sets and interrelationships between entity data attributes and entities in the entity data sets;
and constructing an emergency plan knowledge graph facing the subway fire scene based on the triples.
2. The method for constructing an emergency plan knowledge graph for a subway fire scene as claimed in claim 1, wherein said text labeling is performed on said unstructured text data set to construct a corpus of subway fire emergency plans, specifically comprising:
determining an entity concept related to an emergency plan of a subway fire scene; the entity concept comprises: fire, fire department, fire process evaluation and fire process personnel;
and performing text annotation on the unstructured text data set according to the entity concept to construct a corpus of the subway fire emergency plan.
3. The subway fire scene-oriented emergency plan knowledge map construction method as claimed in claim 2, wherein a BMEO labeling method is adopted to perform text labeling on said unstructured text data set according to said entity concept to construct a corpus of subway fire emergency plans; wherein, B represents the beginning of the entity concept, M represents the middle part of the entity concept, E represents the end of the entity concept, and O represents the non-entity concept.
4. The method for constructing the fire scene of the subway oriented emergency plan knowledge map as claimed in claim 1, wherein a BilSTM-CRF method is used for conducting named entity recognition on said corpus to form an entity data set, and specifically comprises:
performing word vector mapping on an entity concept obtained by performing text labeling on the unstructured text data set to obtain a vector matrix;
taking the vector matrix as the input of a BilSTM-CRF model to obtain an output result; the BilSTM-CRF model comprises the following components: two-way long-time and short-time memory network model and CRF layer; the bidirectional long-short term memory network model comprises a forward long-short term memory network model and a backward long-short term memory network model;
and finishing named entity recognition of the corpus based on the output result to form an entity data set.
5. The method for constructing an emergency plan knowledge graph for a subway fire scene as claimed in claim 4, wherein said vector matrix is used as an input of a BilSTM-CRF model to obtain an output result, specifically comprising:
inputting each vector matrix into a forward long-short time memory network model and a backward long-short time memory network model in sequence to obtain a first output result and a second output result;
splicing the first output result and the second output result into a long input sequence from left to right and from right to left respectively, and converting the obtained output probability into a numerical value for output after the obtained output probability is used as input and transmitted to the CRF layer; this value is output as the output result.
6. The method for constructing an emergency plan knowledge graph for a subway fire scene as claimed in claim 2, wherein defining the correlation between entity data attributes and entities in said entity data set comprises:
defining a relationship between a fire and a fire department as an inclusion relationship;
defining the relation between the fire and the fire disposal process as a concurrency relation;
defining the relation between the fire disposal flow and the fire disposal evaluation as a sequence relation;
the relationship between the fire management personnel and the fire management department is defined as a subordinate relationship.
7. The method for constructing the subway fire scene-oriented emergency plan knowledge graph according to claim 1, wherein the triples are established by performing data extraction and data fusion on the entity data set and the interrelationship between entity data attributes and entities in the entity data set; the triplet includes: entity-entity relationship-entity and entity-attribute relationship-entity.
8. The utility model provides an emergent plan knowledge map construction equipment towards subway fire scene which characterized in that includes:
the data set acquisition module is used for acquiring an unstructured text data set of the subway fire emergency plan; the unstructured text data set comprises: text information resources of the subway fire emergency plan and text data of the subway fire emergency plan; the text data includes: emergency treatment principle and treatment regulation of subway fire;
the corpus construction module is used for carrying out text annotation on the unstructured text data set and constructing a corpus of the subway fire emergency plan;
the entity data determining module is used for carrying out named entity identification on the corpus by using a BilSTM-CRF method to form an entity data set and defining the correlation between entity data attributes and entities in the entity data set;
the triple construction module is used for establishing a triple based on the entity data set and the correlation between the entity data attribute and the entity in the entity data set;
and the knowledge map construction module is used for constructing an emergency plan knowledge map facing the subway fire scene based on the triples.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211614429.0A CN115952297A (en) | 2022-12-15 | 2022-12-15 | Subway fire scene-oriented emergency plan knowledge graph construction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211614429.0A CN115952297A (en) | 2022-12-15 | 2022-12-15 | Subway fire scene-oriented emergency plan knowledge graph construction method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115952297A true CN115952297A (en) | 2023-04-11 |
Family
ID=87281701
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211614429.0A Pending CN115952297A (en) | 2022-12-15 | 2022-12-15 | Subway fire scene-oriented emergency plan knowledge graph construction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115952297A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116521303A (en) * | 2023-07-04 | 2023-08-01 | 四川易诚智讯科技有限公司 | Dynamic display method and system of emergency plan based on multi-source data fusion |
| CN117077631A (en) * | 2023-10-16 | 2023-11-17 | 中国电建集团西北勘测设计研究院有限公司 | Knowledge graph-based engineering emergency plan generation method |
-
2022
- 2022-12-15 CN CN202211614429.0A patent/CN115952297A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116521303A (en) * | 2023-07-04 | 2023-08-01 | 四川易诚智讯科技有限公司 | Dynamic display method and system of emergency plan based on multi-source data fusion |
| CN116521303B (en) * | 2023-07-04 | 2023-09-12 | 四川易诚智讯科技有限公司 | Dynamic display method and system of emergency plan based on multi-source data fusion |
| CN117077631A (en) * | 2023-10-16 | 2023-11-17 | 中国电建集团西北勘测设计研究院有限公司 | Knowledge graph-based engineering emergency plan generation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115952297A (en) | Subway fire scene-oriented emergency plan knowledge graph construction method and device | |
| CN107368468B (en) | Operation and maintenance knowledge map generation method and system | |
| Asteris et al. | A novel heuristic algorithm for the modeling and risk assessment of the COVID-19 pandemic phenomenon | |
| Mangalathu et al. | Deep learning-based classification of earthquake-impacted buildings using textual damage descriptions | |
| CN111160005B (en) | Event prediction method and device based on event evolution knowledge ontology and terminal equipment | |
| Zhang et al. | Semiautomated social media analytics for sensing societal impacts due to community disruptions during disasters | |
| Bhardwaj et al. | Implementation of ID3 algorithm | |
| CN113392986B (en) | Highway bridge information extraction method based on big data and management maintenance system | |
| CN108984775B (en) | Public opinion monitoring method and system based on commodity comments | |
| CN111898852A (en) | Method for constructing knowledge graph of hidden troubles and accidents of enterprise, electronic equipment and storage medium | |
| CN109614612A (en) | A kind of Chinese text error correction method based on seq2seq+attention | |
| Miranker et al. | Text mining and semantic triples: Spatial analyses of text in applied humanitarian forensic research | |
| CN115455202A (en) | Emergency event affair map construction method | |
| CN113761192B (en) | Text processing method, text processing device and text processing equipment | |
| CN117236676A (en) | RPA process mining method and device based on multi-mode event extraction | |
| CN115099344A (en) | Model training method and device, user portrait generation method and device, and equipment | |
| Mironela | Analysis of data volumes circulating in SNs after the occurrence of an earthquake | |
| Nabian et al. | Accelerating stochastic assessment of post-earthquake transportation network connectivity via machine-learning-based surrogates | |
| CN112508603A (en) | Method and device for mining potential customer information of endowment community | |
| Alexeyev et al. | Decision-making support system for experts of penal law | |
| Masmas et al. | Knowledge-based decision support system for emergency management: The pandemic framework | |
| Hassan et al. | A communication and tracking ontology for mobile systems in the event of a large scale disaster | |
| Zahera et al. | Semantic-based end-to-end learning for typhoon intensity prediction | |
| Zuhairoh et al. | Continuous-time Hybrid Markov/semi-Markov Model with Sojourn Time Approach in the Spread of Infectious Diseases | |
| Phengsuwan et al. | Context-based knowledge discovery and querying for social media data |
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 |