CN114707004B - Method and system for extracting and processing case-affair relation based on image model and language model - Google Patents

Method and system for extracting and processing case-affair relation based on image model and language model Download PDF

Info

Publication number
CN114707004B
CN114707004B CN202210569919.7A CN202210569919A CN114707004B CN 114707004 B CN114707004 B CN 114707004B CN 202210569919 A CN202210569919 A CN 202210569919A CN 114707004 B CN114707004 B CN 114707004B
Authority
CN
China
Prior art keywords
event
main body
dimensional
information
type information
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.)
Active
Application number
CN202210569919.7A
Other languages
Chinese (zh)
Other versions
CN114707004A (en
Inventor
龚小刚
王艺丹
赵帅
陈建
张辰
章九鼎
陈祖歌
王红凯
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
State Grid Zhejiang Electric Power Co Ltd
Information and Telecommunication Branch of State Grid Zhejiang Electric Power Co Ltd
Original Assignee
State Grid Zhejiang Electric Power Co Ltd
Information and Telecommunication Branch of State Grid Zhejiang Electric Power Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by State Grid Zhejiang Electric Power Co Ltd, Information and Telecommunication Branch of State Grid Zhejiang Electric Power Co Ltd filed Critical State Grid Zhejiang Electric Power Co Ltd
Priority to CN202210569919.7A priority Critical patent/CN114707004B/en
Publication of CN114707004A publication Critical patent/CN114707004A/en
Application granted granted Critical
Publication of CN114707004B publication Critical patent/CN114707004B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/367—Ontology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Algebra (AREA)
  • Animal Behavior & Ethology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Computational Linguistics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Software Systems (AREA)
  • Image Analysis (AREA)
  • Processing Or Creating Images (AREA)

Abstract

The invention provides a method and a system for extracting and processing a matter relation based on an image model and a language model, wherein the method comprises the following steps: extracting event type information and a second event main body corresponding to each event type information based on the language model; selecting a second event main body corresponding to the event type information, determining a logic image between the first event main body and the selected second event main body, and identifying the logic image based on a neural network model to obtain event logic information; generating a case relation map according to the event type information and the event logic information between the first event main body and the second event main body; determining corresponding one-dimensional event type information and one-dimensional event logic information in the physical relationship map, and determining a corresponding second event main body as a primary action event main body according to the one-dimensional event logic information; and the primary action event main body and/or the secondary action event main body act according to the corresponding one-dimensional event type information and/or two-dimensional event type information.

Description

Method and system for extracting and processing case-affair relation based on image model and language model
Technical Field
The invention relates to the technical field of natural language processing, in particular to a method and a system for extracting and processing a matter relation based on an image model and a language model.
Background
Events are one of the core concepts of human society, and social activities of people tend to be event-driven. The evolution laws and patterns that occur sequentially in time and space between events are a valuable piece of knowledge. The event relationship is obtained based on the subject of the event, in the power grid system, there are many subjects of the event, and after a first subject has a certain first event, the subject relationship based on the first event may be accompanied by a plurality of second events of other subjects.
Chinese patent publication No. CN110110870A discloses an intelligent monitoring method for equipment failure based on event map technology, which utilizes maps to detect and monitor failure of power equipment, perform failure traceability analysis and maintenance analysis, but cannot analyze the physical relationship among multiple subjects, and cannot perform automatic and efficient coordination processing after a certain subject fails.
For example, after a certain power plant as a first main body fails, corresponding events may be generated along with other multiple main bodies, for example, a first second main body may repair the power plant, a second main body performs scheduling power supply, and performs power supply processing on a power supply area of the first main body.
Disclosure of Invention
The embodiment of the invention provides a method and a system for extracting and processing a matter relation based on an image model and a language model, which can automatically cooperate and schedule a plurality of other subjects to process after an emergency occurs in a certain subject, thereby improving the event processing efficiency.
In a first aspect of the embodiments of the present invention, a method for extracting and processing a case relationship based on an image model and a language model is provided, including:
acquiring the fact relation data uploaded by a first event main body based on a fact relation template, and extracting event type information in the fact relation data and a second event main body corresponding to each event type information based on a language model;
when one event type information is identified, selecting a second event main body corresponding to the event type information, determining a logic image between the first event main body and the selected second event main body, and identifying the logic image based on a neural network model to obtain event logic information;
generating a case relation graph according to the event type information and the event logic information between all the first event main bodies and all the second event main bodies;
receiving emergency information uploaded by any first event main body, determining corresponding one-dimensional event type information and one-dimensional event logic information in a matter relation graph according to the first event main body and the emergency information, and determining a corresponding second event main body as a primary action event main body according to the one-dimensional event logic information;
extracting all primary action event main bodies, determining two-dimensional event type information and two-dimensional event logic information corresponding to the primary action event main bodies and the one-dimensional event type information, extracting corresponding secondary action event main bodies according to the two-dimensional event type information and the two-dimensional event logic information, and performing actions by the primary action event main bodies and/or the secondary action event main bodies according to the corresponding one-dimensional event type information and/or the two-dimensional event type information;
and if the actions of the primary action event main body and the secondary action event main body do not meet the preset requirement, making a maneuver plan.
Optionally, in a possible implementation manner of the first aspect, the obtaining of the case relationship data uploaded by the first event body based on the case relationship template includes:
initializing an original relation template at a first event body, wherein the original relation template comprises at least one first body vacancy and one second body vacancy, an event type vacancy is arranged between the first body vacancy and the second body vacancy, a plurality of logic images which can be selected are arranged between the first body vacancy and the second body vacancy, and each logic image has event logic information corresponding to the logic image;
and receiving the configuration information of the matter relation of the first event main body, and filling the original relation template according to the configuration information of the matter relation to obtain the matter relation data.
Optionally, in a possible implementation manner of the first aspect, the receiving event relationship configuration information of the first event body, and filling the original relationship template according to the event relationship configuration information to obtain event relationship data includes:
filling a first event body into a first body vacancy of the original relationship template;
extracting all second event bodies in the event relation configuration information, establishing second body vacancies corresponding to all the second event bodies in the original relation template to obtain an event relation template, and filling the second event bodies into the second body vacancies;
extracting event type information respectively corresponding to all second event bodies in the event relationship configuration information, and filling the event type information into an event type vacancy;
extracting selection information in the configuration information of the matter relationship, and selecting a logic image between the first main body vacancy and the second main body vacancy according to the selection information;
obtaining event relation data based on all of the first body slots, the second body slots, the event type slots, and the logical image.
Optionally, in a possible implementation manner of the first aspect, the obtaining event relation data based on all of the first body slots, the second body slots, the event type slots, and the logical image includes:
if the repeated second main body vacancy exists, all repeated logic images corresponding to the repeated second main body vacancy are obtained;
if the repeated logic images are judged to be completely the same, one of the second main body vacancies is reserved, and the rest of the second main body vacancies, the event type vacancies corresponding to the second main body vacancies and the logic images are deleted;
selecting the reserved event type vacancy at the second main body vacancy, and establishing a corresponding number of new event type vacancies at the lower part of the selected event type vacancy according to the number of the deleted second main body vacancies;
filling the event type information in the deleted event type vacancy into new event type vacancies respectively;
and obtaining the event relation data according to the first main body vacancy, the second main body vacancy, the previous event type vacancy, the new event type vacancy and the logic image.
Optionally, in a possible implementation manner of the first aspect, if it is determined that there is no duplicate second body vacancy, the event type vacancy and the logical image are used to obtain the event relationship data.
Optionally, in a possible implementation manner of the first aspect, the generating a case relationship graph according to the event type information and the event logic information between all the first event subjects and all the second event subjects includes:
establishing an initialization map according to all first event main bodies, wherein the initialization map is provided with nodes corresponding to each first event main body;
determining a second event main body corresponding to each first event main body, and connecting the node of the first event main body with the node of the corresponding second event main body according to the event type information and the event logic information;
and after judging that all the nodes of the first event main body are respectively connected with the nodes of the corresponding second event main body, generating a affair relation graph.
Optionally, in a possible implementation manner of the first aspect, the receiving emergency information uploaded by any one first event main body, determining, according to the first event main body and the emergency information, corresponding one-dimensional event type information and one-dimensional event logic information in a case relationship graph, and determining, according to the one-dimensional event logic information, that a corresponding second event main body is a primary action event main body further includes:
acquiring an emergency node corresponding to a first event main body uploading emergency information, and determining corresponding one-dimensional event type information according to the attribute of the emergency information;
determining one-dimensional event logic information corresponding to the one-dimensional event type information, and determining a second event main body corresponding to the first event main body according to the one-dimensional event logic information;
and taking the second event main body as a primary action event main body, and taking a node corresponding to the second event main body as a primary action node.
Optionally, in a possible implementation manner of the first aspect, the extracting all primary action event bodies, determining two-dimensional event type information and two-dimensional event logic information of the primary action event body corresponding to the one-dimensional event type information, and extracting corresponding secondary action event bodies according to the two-dimensional event type information and the two-dimensional event logic information, where the acting of the primary action event body and/or the secondary action event body according to the corresponding one-dimensional event type information and/or the two-dimensional event type information includes:
obtaining corresponding two-dimensional event type information in a primary action event main body according to the attribute of the one-dimensional event type information;
determining two-dimensional event logic information corresponding to the two-dimensional event type information, and determining a secondary action node corresponding to the primary action node according to the two-dimensional event logic information;
and extracting a second event body corresponding to the secondary action node as a secondary action event body.
Optionally, in a possible implementation manner of the first aspect, if the actions of the primary action event body and the secondary action event body do not meet preset requirements, the making a maneuver plan includes:
counting all the first-stage action nodes to obtain the number of the first-stage nodes, and counting all the second-stage action nodes to obtain the number of the second-stage nodes;
acquiring one-dimensional response time corresponding to the completion of the one-dimensional event type information by all the first-level action nodes, and acquiring two-dimensional response time corresponding to the completion of the two-dimensional event type information by all the second-level action nodes;
calculating according to the number of the primary nodes, the number of the secondary nodes, the one-dimensional response time and the two-dimensional response time to obtain a processing coefficient corresponding to the emergency information;
Figure DEST_PATH_IMAGE001
wherein the content of the first and second substances,
Figure 28247DEST_PATH_IMAGE002
in order to calculate the processing coefficients,
Figure 353794DEST_PATH_IMAGE003
is a first normalized value of the first normalized value,
Figure 733872DEST_PATH_IMAGE004
is a value of a standard quantity of the substance,
Figure 683986DEST_PATH_IMAGE005
in order to be a first-order weight,
Figure 807277DEST_PATH_IMAGE006
is as follows
Figure 777812DEST_PATH_IMAGE007
The one-dimensional response time of the individual level one action nodes,
Figure 305354DEST_PATH_IMAGE008
is as follows
Figure 255947DEST_PATH_IMAGE007
The standard response time of an individual level one action node,
Figure 308830DEST_PATH_IMAGE009
is an upper limit value of the number of the primary action nodes,
Figure 843665DEST_PATH_IMAGE010
the number of the nodes at the first level,
Figure 100002_DEST_PATH_IMAGE011
in order to be the secondary weight, the weight is,
Figure 770470DEST_PATH_IMAGE012
is as follows
Figure 327135DEST_PATH_IMAGE013
The two-dimensional response time of each secondary action node,
Figure 793495DEST_PATH_IMAGE014
is as follows
Figure 983694DEST_PATH_IMAGE013
The standard response time of each secondary action node,
Figure 391322DEST_PATH_IMAGE015
is the upper limit value of the number of the secondary action nodes,
Figure 803325DEST_PATH_IMAGE016
the number of the secondary nodes;
and if the processing coefficient is judged to be larger than the preset coefficient, judging that the actions of the primary action event main body and the secondary action event main body do not meet the preset requirement, and making a exercise plan for the extracted primary action event main body and the extracted secondary action event main body.
In a second aspect of the embodiments of the present invention, there is provided a system for extracting and processing a case relation based on an image model and a language model, including:
the system comprises an extraction module, a language model acquisition module and a display module, wherein the extraction module is used for acquiring the event relation data uploaded by a first event main body based on a event relation template, and extracting event type information in the event relation data and a second event main body corresponding to each event type information based on the language model;
the identification module is used for selecting a second event main body corresponding to the event type information when one piece of event type information is identified, determining a logic image between the first event main body and the selected second event main body, and identifying the logic image based on a neural network model to obtain event logic information;
the generation module is used for generating a case relation map according to the event type information and the event logic information between all the first event main bodies and all the second event main bodies;
the determining module is used for receiving the emergency information uploaded by any one first event main body, determining corresponding one-dimensional event type information and one-dimensional event logic information in a matter relation graph according to the first event main body and the emergency information, and determining a corresponding second event main body as a primary action event main body according to the one-dimensional event logic information;
the extraction module is used for extracting all the primary action event main bodies, determining two-dimensional event type information and two-dimensional event logic information of the primary action event main bodies corresponding to the one-dimensional event type information, extracting corresponding secondary action event main bodies according to the two-dimensional event type information and the two-dimensional event logic information, and enabling the primary action event main bodies and/or the secondary action event main bodies to act according to the corresponding one-dimensional event type information and/or the two-dimensional event type information;
and the formulating module is used for formulating a maneuver plan if the actions of the primary action event main body and the secondary action event main body do not meet the preset requirement.
In a third aspect of the embodiments of the present invention, a storage medium is provided, in which a computer program is stored, which, when being executed by a processor, is adapted to implement the method according to the first aspect of the present invention and various possible designs of the first aspect of the present invention.
The invention provides a method and a system for extracting and processing a matter relation based on an image model and a language model. The event relation data uploaded by each first event main body can be obtained according to the event relation template, corresponding information is identified according to the language model and the neural network model, and the event relation map is constructed, so that the event main bodies can establish corresponding relations among the event main bodies according to corresponding event type information and event logic information. According to the invention, the emergency information is processed according to the affair relation map, and the corresponding primary action event main body and secondary action event main body are rapidly determined, so that the corresponding primary action event main body and secondary action event main body rapidly respond to participate in the processing of the corresponding emergency, the processing efficiency when a plurality of main bodies cooperate is improved, and the main bodies which do not need to cooperate do not need to prepare, thereby avoiding resource waste.
According to the technical scheme provided by the invention, the original relation template is processed according to the event relation configuration information of different first event main bodies, and the corresponding vacancy is established at the original relation template to obtain the event relation template uniquely corresponding to each first event main body.
The invention can process the emergency information through the affair relation map, and determine the corresponding primary action event main body and secondary action event main body by combining the relation of each node in the affair relation map, so that the invention can automatically determine the primary action event main body which is directly processed according to the relation of the nodes after the emergency occurs, and also can determine the secondary action event main body which is indirectly processed, thereby improving the determination efficiency of the action event main body and the solution efficiency of the emergency. The invention can comprehensively calculate according to the number of the primary nodes, the number of the secondary nodes, the one-dimensional response time and the two-dimensional response time to obtain the processing coefficient, judge the response speed of the plurality of action nodes when processing the emergency through the processing coefficient, judge whether to make a corresponding exercise plan according to the processing coefficient, avoid the lower efficiency when similar events occur in the follow-up process, and improve the linkage processing capacity and efficiency among the plurality of event nodes, the event main bodies and the action main bodies.
Drawings
FIG. 1 is a flowchart of a first embodiment of a method for extracting and processing a case relationship based on an image model and a language model according to the present invention;
FIG. 2 is a flowchart illustrating a second embodiment of a method for extracting and processing a case relationship based on an image model and a language model according to the present invention;
fig. 3 is a configuration diagram of a first embodiment of a case relation extraction processing system based on an image model and a language model according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are 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.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein.
It should be understood that, in various embodiments of the present invention, the sequence numbers of the processes do not mean the execution sequence, and the execution sequence of the processes should be determined by the functions and the internal logic of the processes, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
It should be understood that in the present application, "comprising" and "having" and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
It should be understood that, in the present invention, "a plurality" means two or more. "and/or" is merely an association describing an associated object, meaning that three relationships may exist, e.g., a and/or B, may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "comprises A, B and C" and "comprises A, B, C" means that all three of A, B, C comprise, "comprises A, B or C" means that one of A, B, C comprises, "comprises A, B and/or C" means that any 1 or any 2 or 3 of A, B, C comprises.
It should be understood that in the present invention, "B corresponding to a", "a corresponds to B", or "B corresponds to a" means that B is associated with a, and B can be determined from a. Determining B from a does not mean determining B from a alone, but may be determined from a and/or other information. And the matching of A and B means that the similarity of A and B is greater than or equal to a preset threshold value.
As used herein, "if" may be interpreted as "at … …" or "when … …" or "in response to a determination" or "in response to a detection", depending on the context.
The technical solution of the present invention will be described in detail below with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments.
The invention provides a method for extracting and processing a matter relation based on an image model and a language model, which comprises the following steps of:
step S110, obtaining the event relation data uploaded by the first event body based on the event relation template, and extracting event type information in the event relation data and a second event body corresponding to each event type information based on the language model. The method comprises the steps of firstly obtaining the fact relation data uploaded by a first event main body based on a fact relation template, and extracting corresponding information according to a language model. The first event agent may be a power station, the event type information may be that a thermal generator of the power station has a fault, the event type information and the second event agent corresponding to the event type information may be plural, the 1 st second event agent may be a maintenance team, and the 2 nd second event agent may be another power station. The power generation station with the fault can be maintained through the maintenance team, the power generation station with the fault can be replaced by other power generation stations to generate power, and the power is supplied to the corresponding power utilization area.
Each first event main body can correspond to an input terminal, the input terminal can upload corresponding matter relation data to a server through the input terminal, and the server obtains the matter relation data uploaded by the first event main bodies based on the matter relation template and carries out corresponding processing.
In a possible implementation manner of the technical solution provided by the present invention, as shown in fig. 2, step S110 includes:
step S1101, initializing an original relationship template at a first event body, where the original relationship template includes at least one first body vacancy and one second body vacancy, an event type vacancy is provided between the first body vacancy and the second body vacancy, a plurality of selectable logical images are provided between the first body vacancy and the second body vacancy, and each logical image has event logical information corresponding thereto. The invention firstly carries out initialization processing of the original relation template at each first event body, the original relation template only comprises one second body vacancy, and the invention only sets one second body vacancy in the original relation template because the number of the second event bodies corresponding to different types of the first event bodies may be different. An event type slot is disposed between the first body slot and the second body slot.
The first body slot, the second body slot, and the event type slot may be the input of corresponding text, e.g., the text that the first body slot may input is power station a, etc. The logical image is an image preset in the original relationship template, and may be in the form of an arrow, for example, → ", etc., such as the first body space → the second body space, that is, it can be understood that when the event type information corresponding to the first event body corresponding to the first body space occurs, the processing of responding through the second event body corresponding to the second body space is required. The logical image can be regarded as a correspondence between two event subjects.
Step S1102, receiving the configuration information of the matter relationship of the first event body, and filling the original relationship template according to the configuration information of the matter relationship to obtain the matter relationship data. The invention processes and fills the original relation template according to the matter relation configuration information to obtain corresponding matter relation data. It is to be understood that the case relationship configuration information may be actively configured by the staff at the first event subject.
In a possible implementation manner of the technical solution provided by the present invention, step S1102 includes:
filling a first event body into a first body vacancy of the original relationship template. Each original relationship template has only one first event body, which as mentioned above may be the power plant a, which will now be filled into the corresponding first body vacancies.
And extracting all second event bodies in the event relation configuration information, establishing second body vacancies corresponding to all the second event bodies in the original relation template to obtain the event relation template, and filling the second event bodies into the second body vacancies. The invention establishes new second main body vacancies at the original relation template according to the quantity of the second event main bodies, so that all the second event main bodies corresponding to the first event main body have the second main body vacancies corresponding to the first event main body, the second event main body can be a power station B, and the power station B is filled in the corresponding second main body vacancies at this time.
And extracting the event type information corresponding to all the second event bodies in the event relation configuration information respectively, and filling the event type information into an event type vacancy. The event type information includes, for example, a power generation failure, and the power generation failure is filled in the event type empty space at this time.
And extracting selection information in the configuration information of the matter relationship, and selecting a logic image between the first main body vacancy and the second main body vacancy according to the selection information. Since the logical image is preset, the user can select the logical image based on the selection information, for example, by pointing the first body space to the second body space with an arrow pointing towards the second body space between the first body space and the second body space.
Obtaining event relation data based on all of the first body slots, the second body slots, the event type slots, and the logical image. The case relation data at this time may be the case relation data corresponding to one first event body.
In a possible embodiment, the obtaining of the event relationship data based on all the first body slots, the second body slots, the event type slots, and the logic images includes:
and if the repeated second body vacancy exists, acquiring all repeated logic images corresponding to the repeated second body vacancy. When the user configures the second event body, the first event body may have the same second event body when multiple types of event type information occur, so that repeated second body vacancies exist at this time, and the invention determines all repeated logical images corresponding to the repeated second body vacancies.
And if the repeated logic images are judged to be completely the same, one of the second body vacancies is reserved, and the rest of the second body vacancies, the event type vacancies corresponding to the second body vacancies and the logic images are deleted. It is understood that, at this time, a plurality of same event logic information and different event type information exist between one second event body and the corresponding first event body. Therefore, only one second body vacancy can be reserved at this time, and the rest of second body vacancies, the event type vacancies corresponding to the second body vacancies and the logic images can be deleted, so that the vacancy number and the data volume of the physical relation template are reduced.
And selecting the reserved event type vacancy at the second body vacancy, and establishing a corresponding number of new event type vacancies at the lower part of the selected event type vacancy according to the number of the deleted second body vacancies. The invention establishes a corresponding number of new event type vacancies at the lower part of the selected event type vacancy, so that a plurality of different event type information can be filled into different event type vacancies, and only one identical second body vacancy and identical logic image are reserved.
And filling the event type information in the deleted event type vacancy into the new event type vacancy respectively. The invention respectively fills the corresponding event type information into the new event type vacancy, so that all the event type information can be respectively stored in the corresponding event type vacancy.
And obtaining the event relation data according to the first main body vacancy, the second main body vacancy, the previous event type vacancy, the new event type vacancy and the logic image. The invention can obtain the data of the physical relationship according to the vacancy of the second event type, thus effectively reducing the data quantity value and the vacancy quantity of the data of the physical relationship.
In one possible embodiment, if it is determined that there is no duplicate second body vacancy, the event type vacancy and the logical image are used to obtain the event relationship data. In this case, there is no need to combine a plurality of second body slots, but only one event type slot. Therefore, the matter relation data can be directly obtained.
After the fact relation data are obtained, the event type information in the fact relation data and a second event main body corresponding to each event type information are extracted based on the language model. Because the event type information and the second event main body are characters, the identification can be carried out based on the language model to obtain corresponding information.
Step S120, when one event type information is identified, selecting a second event main body corresponding to the event type information, determining a logic image between the first event main body and the selected second event main body, and identifying the logic image based on a neural network model to obtain event logic information. When the language model determines that an event type information is obtained, the invention determines a second event main body corresponding to the event type information, determines a corresponding logic image, and identifies the logic image through the neural network model to obtain the event logic information, wherein the logic image can be pointed by an arrow, and as mentioned above, if the arrow points to the second event main body for the first event main body, the second event main body responds when the event type information of the corresponding type occurs. And if the arrow points to the first event body for the second event body, the first event body responds when the second event body generates event type information of a corresponding type. The neural network model may be a convolutional neural network model having an image recognition function, and the event logic information may be that the second event body responds when the corresponding type of event type information occurs, or the first event body responds when the corresponding type of event type information occurs, as described above.
And step S130, generating a case relation graph according to the event type information and the event logic information between all the first event main bodies and all the second event main bodies. The incident relationship graph may have a plurality of first event subjects and a plurality of relationships between the first event subjects and/or the second event subjects. The first event body and the second event body are in a relative relationship, for example, the input terminal at the power station a is taken as the first event body, and the input terminal at the power station B is taken as the second event body. If the incoming terminal at station B is the subject of the first event, then the incoming terminal at station a may be the subject of the first event.
In one possible implementation manner, the technical solution provided by the present invention, in step S130, includes:
and establishing an initialization map according to all the first event main bodies, wherein the initialization map has nodes corresponding to each first event main body. The invention will firstly establish an initialization map based on all the first event subjects, and it can be understood that each input terminal can be regarded as a first event subject, i.e. each second event subject can also be converted into a corresponding first event subject, and as mentioned above, the first event subject only has a different viewing angle relative to the second event subject input terminal. Each initialization map has nodes corresponding to each first event body, and the number of the nodes is the same as that of the first event bodies.
And determining a second event main body corresponding to each first event main body, and connecting the node of the first event main body with the node of the corresponding second event main body according to the event type information and the event logic information. The invention starts from the perspective of each first event body, determines each corresponding second event body, and connects the nodes of the first event body and the nodes of the second event body by combining the event type information and the event logic information, and at the moment, the invention has the connection relation of a plurality of first event bodies.
And after judging that all the nodes of the first event main body are respectively connected with the nodes of the corresponding second event main body, generating a affair relation graph. After all the first event main bodies are respectively connected with the corresponding second event main bodies, the physical relationship map at the moment reaches a completely connected state.
Step S140, receiving the emergency information uploaded by any first event main body, determining corresponding one-dimensional event type information and one-dimensional event logic information in the event relationship graph according to the first event main body and the emergency information, and determining a corresponding second event main body as a primary action event main body according to the one-dimensional event logic information. After an emergency occurs in a certain main body, the input terminal of the corresponding main body can be used as a first event main body to upload emergency information, at this time, the server can determine corresponding one-dimensional event type information and one-dimensional event logic information in the event relationship map according to the first event main body and the emergency information, for example, the first event main body is a power generation station a, the emergency information is a power generation motor fault, the corresponding one-dimensional event type information can be the power generation motor fault, the one-dimensional event logic information can be maintenance processing required for a team C, the main body corresponding to the team C at this time is a second event main body, the team C is a first-level action event main body, and the first-level action event main body can be regarded as a main body for directly processing the event of the emergency information.
In one possible implementation manner, the technical solution provided by the present invention, in step S140, includes:
acquiring an emergency node corresponding to a first event main body uploading emergency information, and determining corresponding one-dimensional event type information according to the attribute of the emergency information. According to the method, the node of the first event main body in the matter relation graph is taken as an emergency node, corresponding one-dimensional event type information is determined according to the attribute of the emergency information, and the attribute of the emergency information at the moment can be understood as the fault attribute of the generator motor. At this time, corresponding one-dimensional event type information, namely, a generator motor fault, is obtained, and the generator motor needs to be maintained.
And determining one-dimensional event logic information corresponding to the one-dimensional event type information, and determining a second event main body corresponding to the first event main body according to the one-dimensional event logic information. The invention determines a second event body corresponding to the first event body according to the one-dimensional event logic information, namely, determines a corresponding team group C.
And taking the second event main body as a primary action event main body, and taking a node corresponding to the second event main body as a primary action node. Since the team C is a main body that directly processes the event corresponding to the emergency information, the second event main body is used as the primary action event main body, and the node corresponding to the second event main body is used as the primary action node.
Step S150, extracting all primary action event main bodies, determining two-dimensional event type information and two-dimensional event logic information of the primary action event main bodies corresponding to the one-dimensional event type information, extracting corresponding secondary action event main bodies according to the two-dimensional event type information and the two-dimensional event logic information, and acting by the primary action event main bodies and/or the secondary action event main bodies according to the corresponding one-dimensional event type information and/or the two-dimensional event type information. After all the primary action event main bodies are obtained, the two-dimensional event type information corresponding to the primary action event main bodies and the one-dimensional event type information is obtained, for example, the one-dimensional event type information is a generator motor fault and needs to be maintained, the two-dimensional event type information at this time can be used for preparing a new generator motor component, the two-dimensional event logic information corresponding to the prepared new generator motor component is determined at this time, the corresponding secondary action event main bodies can be determined according to the two-dimensional event type information and the two-dimensional event logic information at this time, and the secondary action event main bodies can be a back office D for preparing the new generator motor component. It can be understood that the secondary action event body does not directly process the event corresponding to the emergency event information, but processes the event in an indirect manner. The invention can rapidly determine the main body of the event corresponding to the direct and indirect processing emergency information through the event relation map, thereby improving the determination efficiency of the main body.
In one possible implementation manner of the technical solution provided by the present invention, step S150 includes:
and obtaining corresponding two-dimensional event type information in the primary action event main body according to the attribute of the one-dimensional event type information. The attribute of the one-dimensional event type information may be that the generator motor is out of order and needs to be maintained, and at this time, the two-dimensional event type information may be that a new generator motor component needs to be prepared.
And determining two-dimensional event logic information corresponding to the two-dimensional event type information, and determining a secondary action node corresponding to the primary action node according to the two-dimensional event logic information. The invention can be based on the two-dimensional event logic information corresponding to the two-dimensional event type information, namely the two-dimensional event logic information can be an arrow at the moment, and the arrow points to the secondary action event main body from the primary action event main body. The first-level action event main body corresponds to the first-level action node in the case relation graph, and the second-level action event main body corresponds to the second-level action node in the case relation graph.
And extracting a second event body corresponding to the secondary action node as a secondary action event body. The invention can extract the primary action event main body and the secondary action event main body, so that the primary action event main body and the secondary action event main body carry out corresponding cooperative processing operation according to corresponding event type information, thereby solving the emergency rapidly and efficiently.
And step S160, if the actions of the primary action event main body and the secondary action event main body do not meet the preset requirement, making a maneuver plan.
In one possible implementation manner, the technical solution provided by the present invention, in step S160, includes:
and counting all the first-stage action nodes to obtain the number of the first-stage nodes, and counting all the second-stage action nodes to obtain the number of the second-stage nodes. The invention can obtain the number of the primary nodes and the number of the secondary nodes, and the more the number of the primary nodes and the number of the secondary nodes are, the more the nodes which need to cooperate and participate in event processing are proved. Therefore, the invention can carry out comprehensive calculation according to the number of the primary nodes and the number of the secondary nodes to obtain the deviant of the processing coefficient
Figure 61657DEST_PATH_IMAGE017
Obtaining the offset value of the processing coefficient, if the number of the primary nodes and the number of the secondary nodes are more, the more the subjects involved are proved to be, so that the offset value of the processing coefficient is larger at the moment, and the processing coefficient is further caused to have a tendency of becoming larger.
And acquiring one-dimensional response time corresponding to the completion of the one-dimensional event type information by all the first-level action nodes, and acquiring two-dimensional response time corresponding to the completion of the two-dimensional event type information by all the second-level action nodes. The one-dimensional response time may be regarded as the time for the first-level action node to complete the one-dimensional event type information, for example, the time for repairing the generator motor of the power station a, and the two-dimensional response time corresponding to the two-dimensional event type information may be the time for preparing a new generator motor component, and the like.
And calculating according to the number of the primary nodes, the number of the secondary nodes, the one-dimensional response time and the two-dimensional response time to obtain a processing coefficient corresponding to the emergency information. The invention can combine the number of the first-level nodes, the number of the second-level nodes, the one-dimensional response time and the two-dimensional response time to carry out multi-dimensional fusion calculation, and obtain a processing coefficient which is used for processing the emergency information and fusing multiple dimensions.
The processing coefficient is calculated by the following formula,
Figure 118081DEST_PATH_IMAGE018
wherein the content of the first and second substances,
Figure 821245DEST_PATH_IMAGE019
in order to calculate the processing coefficients,
Figure 349495DEST_PATH_IMAGE020
is a first normalized value of the first normalized value,
Figure 564182DEST_PATH_IMAGE021
is a standard quantity value and is used as a standard quantity value,
Figure 611466DEST_PATH_IMAGE022
in order to be a first-order weight,
Figure 586727DEST_PATH_IMAGE023
is as follows
Figure 848687DEST_PATH_IMAGE024
The one-dimensional response time of the individual level one action nodes,
Figure 222992DEST_PATH_IMAGE025
is as follows
Figure 412532DEST_PATH_IMAGE024
The standard response time of an individual level one action node,
Figure 841852DEST_PATH_IMAGE026
is an upper limit value of the number of the primary action nodes,
Figure 87981DEST_PATH_IMAGE027
the number of the nodes at one level is,
Figure 898255DEST_PATH_IMAGE028
in order to be the secondary weight, the weight is,
Figure 274790DEST_PATH_IMAGE029
is as follows
Figure 124279DEST_PATH_IMAGE013
The two-dimensional response time of each secondary action node,
Figure 208168DEST_PATH_IMAGE030
is as follows
Figure 161824DEST_PATH_IMAGE013
The standard response time of each secondary action node,
Figure 533812DEST_PATH_IMAGE015
is the upper limit value of the number of the secondary action nodes,
Figure 168581DEST_PATH_IMAGE031
the number of secondary nodes.
By passing
Figure 859807DEST_PATH_IMAGE032
The difference between the one-dimensional response time and the standard response time of all the primary action nodes can be obtained if
Figure 249112DEST_PATH_IMAGE033
The larger the response time of all primary action nodes is proved to be, the longer the response time is, the more the response time is proved to be
Figure 315892DEST_PATH_IMAGE034
The average processing difference time of all the primary action nodes can be obtained. By passing
Figure 8649DEST_PATH_IMAGE035
The difference between the two-dimensional response time and the standard response time of all secondary action nodes can be obtained if
Figure 90567DEST_PATH_IMAGE036
The larger the response time of all secondary action nodes is proved to be, the longer the response time is, the more the response time is proved to be
Figure 368448DEST_PATH_IMAGE037
The average processing difference time of all secondary action nodes can be obtained. By offset value
Figure 207703DEST_PATH_IMAGE038
To pair
Figure 568232DEST_PATH_IMAGE039
And performing offset processing on the calculated value to obtain a final processing coefficient, wherein the larger the processing coefficient is, the lower the processing efficiency of the corresponding primary action nodes and secondary action nodes on the emergency is proved to be.
And if the processing coefficient is judged to be larger than the preset coefficient, judging that the actions of the primary action event main body and the secondary action event main body do not meet the preset requirement, and making a exercise plan for the extracted primary action event main body and the extracted secondary action event main body. When the processing coefficient is greater than the preset coefficient, the fact that the processing of the emergency by the primary action node and the secondary action node is too low at the moment is proved, and corresponding emergency needs to be trained, so that the exercise plan can be made for the extracted primary action event main body and the extracted secondary action event main body, and corresponding training is conducted. The exercise program may be to reprocess the corresponding fault once again according to the incident information.
In order to implement the method for extracting and processing the case-of-affairs relationship based on the image model and the language model, the invention also provides a system for extracting and processing the case-of-affairs relationship based on the image model and the language model, as shown in fig. 3, comprising:
the system comprises an extraction module, a language model acquisition module and a display module, wherein the extraction module is used for acquiring the event relation data uploaded by a first event main body based on a event relation template, and extracting event type information in the event relation data and a second event main body corresponding to each event type information based on the language model;
the identification module is used for selecting a second event main body corresponding to the event type information when one piece of event type information is identified, determining a logic image between the first event main body and the selected second event main body, and identifying the logic image based on a neural network model to obtain event logic information;
the generation module is used for generating a case relation map according to the event type information and the event logic information between all the first event main bodies and all the second event main bodies;
the determining module is used for receiving the emergency information uploaded by any one first event main body, determining corresponding one-dimensional event type information and one-dimensional event logic information in a matter relation graph according to the first event main body and the emergency information, and determining a corresponding second event main body as a primary action event main body according to the one-dimensional event logic information;
the extraction module is used for extracting all the primary action event main bodies, determining two-dimensional event type information and two-dimensional event logic information of the primary action event main bodies corresponding to the one-dimensional event type information, extracting corresponding secondary action event main bodies according to the two-dimensional event type information and the two-dimensional event logic information, and enabling the primary action event main bodies and/or the secondary action event main bodies to act according to the corresponding one-dimensional event type information and/or the two-dimensional event type information.
The present invention also provides a storage medium having a computer program stored therein, the computer program being executable by a processor to implement the methods provided by the various embodiments described above.
The storage medium may be a computer storage medium or a communication medium. Communication media includes any medium that facilitates transfer of a computer program from one place to another. Computer storage media may be any available media that can be accessed by a general purpose or special purpose computer. For example, a storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium. Of course, the storage medium may also be integral to the processor. The processor and the storage medium may reside in an Application Specific Integrated Circuits (ASIC). Additionally, the ASIC may reside in user equipment. Of course, the processor and the storage medium may reside as discrete components in a communication device. The storage medium may be read-only memory (ROM), random-access memory (RAM), CD-ROMs, magnetic tapes, floppy disks, optical data storage devices, and the like.
The present invention also provides a program product comprising execution instructions stored in a storage medium. The at least one processor of the device may read the execution instructions from the storage medium, and the execution of the execution instructions by the at least one processor causes the device to implement the methods provided by the various embodiments described above.
In the above embodiments of the terminal or the server, it should be understood that the Processor may be a Central Processing Unit (CPU), other general-purpose processors, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the present invention may be embodied directly in a hardware processor, or in a combination of the hardware and software modules within the processor.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and these modifications or substitutions do not depart from the spirit of the corresponding technical solutions of the embodiments of the present invention.

Claims (6)

1. The method for extracting and processing the affair relation based on the image model and the language model is characterized by comprising the following steps:
acquiring the fact relation data uploaded by a first event main body based on a fact relation template, and extracting event type information in the fact relation data and a second event main body corresponding to each event type information based on a language model;
when one event type information is identified, selecting a second event main body corresponding to the event type information, determining a logic image between the first event main body and the selected second event main body, and identifying the logic image based on a neural network model to obtain event logic information;
generating a case relation graph according to the event type information and the event logic information between all the first event main bodies and all the second event main bodies;
receiving emergency information uploaded by any first event main body, determining corresponding one-dimensional event type information and one-dimensional event logic information in a matter relation graph according to the first event main body and the emergency information, and determining a corresponding second event main body as a primary action event main body according to the one-dimensional event logic information;
extracting all primary action event main bodies, determining two-dimensional event type information and two-dimensional event logic information of the primary action event main bodies corresponding to the one-dimensional event type information, extracting corresponding secondary action event main bodies according to the two-dimensional event type information and the two-dimensional event logic information, and enabling the primary action event main bodies and the secondary action event main bodies to act according to the corresponding one-dimensional event type information and the two-dimensional event type information;
if the actions of the primary action event main body and the secondary action event main body do not meet the preset requirement, a maneuver plan is made;
the generating of the event relation graph according to the event type information and the event logic information between all the first event main bodies and all the second event main bodies comprises the following steps:
establishing an initialization map according to all first event main bodies, wherein the initialization map is provided with nodes corresponding to each first event main body;
determining a second event main body corresponding to each first event main body, and connecting the node of the first event main body with the node of the corresponding second event main body according to the event type information and the event logic information;
after judging that all the nodes of the first event main body are respectively connected with the nodes of the corresponding second event main body, generating a matter relation graph;
the receiving of the emergency information uploaded by any one of the first event bodies, determining corresponding one-dimensional event type information and one-dimensional event logic information in a case relationship graph according to the first event body and the emergency information, and determining a corresponding second event body as a primary action event body according to the one-dimensional event logic information, further includes:
acquiring an emergency node corresponding to a first event main body uploading emergency information, and determining corresponding one-dimensional event type information according to the attribute of the emergency information;
determining one-dimensional event logic information corresponding to the one-dimensional event type information, and determining a second event main body corresponding to the first event main body according to the one-dimensional event logic information;
taking the second event main body as a primary action event main body, and taking a node corresponding to the second event main body as a primary action node;
the extracting all the primary action event main bodies, determining two-dimensional event type information and two-dimensional event logic information of the primary action event main bodies corresponding to the one-dimensional event type information, extracting corresponding secondary action event main bodies according to the two-dimensional event type information and the two-dimensional event logic information, and acting by the primary action event main bodies and the secondary action event main bodies according to the corresponding one-dimensional event type information and the two-dimensional event type information comprises the following steps:
obtaining corresponding two-dimensional event type information in a primary action event main body according to the attribute of the one-dimensional event type information;
determining two-dimensional event logic information corresponding to the two-dimensional event type information, and determining a secondary action node corresponding to the primary action node according to the two-dimensional event logic information;
extracting a second event main body corresponding to the secondary action node as a secondary action event main body;
if the actions of the primary action event main body and the secondary action event main body do not meet the preset requirements, a maneuver plan is formulated, which comprises the following steps:
counting all the first-stage action nodes to obtain the number of the first-stage nodes, and counting all the second-stage action nodes to obtain the number of the second-stage nodes;
acquiring one-dimensional response time corresponding to the completion of the one-dimensional event type information by all the first-level action nodes, and acquiring two-dimensional response time corresponding to the completion of the two-dimensional event type information by all the second-level action nodes;
calculating according to the number of the primary nodes, the number of the secondary nodes, the one-dimensional response time and the two-dimensional response time to obtain a processing coefficient corresponding to the emergency information;
Figure 60079DEST_PATH_IMAGE002
wherein the content of the first and second substances,
Figure DEST_PATH_IMAGE003
in order to calculate the processing coefficients,
Figure 43209DEST_PATH_IMAGE004
is a first normalized value of the first normalized value,
Figure DEST_PATH_IMAGE005
is a standard quantity value and is used as a standard quantity value,
Figure 159064DEST_PATH_IMAGE006
in order to be a first-order weight,
Figure DEST_PATH_IMAGE007
is as follows
Figure DEST_PATH_IMAGE009
The one-dimensional response time of the individual level one action nodes,
Figure 47386DEST_PATH_IMAGE010
is as follows
Figure 742940DEST_PATH_IMAGE009
The standard response time of an individual level one action node,
Figure DEST_PATH_IMAGE011
is an upper limit value of the number of the primary action nodes,
Figure 829845DEST_PATH_IMAGE012
the number of the nodes at one level is,
Figure DEST_PATH_IMAGE013
in order to be the secondary weight, the weight is,
Figure 647760DEST_PATH_IMAGE014
is as follows
Figure DEST_PATH_IMAGE015
The two-dimensional response time of each secondary action node,
Figure 429902DEST_PATH_IMAGE016
is as follows
Figure 991464DEST_PATH_IMAGE015
The standard response time of each secondary action node,
Figure DEST_PATH_IMAGE017
is the upper limit value of the number of the secondary action nodes,
Figure 339400DEST_PATH_IMAGE018
the number of secondary nodes;
and if the processing coefficient is judged to be larger than the preset coefficient, judging that the actions of the primary action event main body and the secondary action event main body do not meet the preset requirement, and making a exercise plan for the extracted primary action event main body and the extracted secondary action event main body.
2. The method for extracting and processing a case relation based on an image model and a language model according to claim 1,
the acquiring of the event relation data uploaded by the first event main body based on the event relation template comprises the following steps:
initializing an original relation template at a first event body, wherein the original relation template comprises at least one first body vacancy and one second body vacancy, an event type vacancy is arranged between the first body vacancy and the second body vacancy, a plurality of logic images which can be selected are arranged between the first body vacancy and the second body vacancy, and each logic image has event logic information corresponding to the logic image;
and receiving the configuration information of the matter relation of the first event main body, and filling the original relation template according to the configuration information of the matter relation to obtain the matter relation data.
3. The method for extracting and processing a case relation based on an image model and a language model according to claim 2,
the receiving of the configuration information of the matter relationship of the first event main body and the filling of the original relationship template according to the configuration information of the matter relationship to obtain the data of the matter relationship comprises the following steps:
filling a first event body into a first body vacancy of the original relation template;
extracting all second event bodies in the event relation configuration information, establishing second body vacancies corresponding to all the second event bodies in the original relation template to obtain an event relation template, and filling the second event bodies into the second body vacancies;
extracting event type information respectively corresponding to all second event bodies in the event relationship configuration information, and filling the event type information into an event type vacancy;
extracting selection information in the configuration information of the matter relationship, and selecting a logic image between the first main body vacancy and the second main body vacancy according to the selection information;
obtaining event relation data based on all of the first body slots, the second body slots, the event type slots, and the logical image.
4. The method for extracting and processing case relation based on image model and language model according to claim 3,
obtaining event relation data based on all of the first body slots, the second body slots, the event type slots and the logic images, including:
if the repeated second main body vacancy exists, all repeated logic images corresponding to the repeated second main body vacancy are obtained;
if the repeated logic images are judged to be completely the same, one of the second main body vacancies is reserved, and the rest of the second main body vacancies, the event type vacancies corresponding to the second main body vacancies and the logic images are deleted;
selecting the reserved event type vacancy at the second main body vacancy, and establishing a corresponding number of new event type vacancies at the lower part of the selected event type vacancy according to the number of the deleted second main body vacancies;
filling the event type information in the deleted event type vacancy into new event type vacancies respectively;
and obtaining the event relation data according to the first main body vacancy, the second main body vacancy, the previous event type vacancy, the new event type vacancy and the logic image.
5. The method for extracting and processing event relation based on image model and language model according to claim 4,
and if the second main body vacancy is judged not to be repeated, obtaining the event relation data according to the current first main body vacancy, the second main body vacancy, the event type vacancy and the logic image.
6. A system for extracting and processing a case-based relationship between an image model and a language model, which is applied to the method for extracting and processing a case-based relationship between an image model and a language model according to any one of claims 1 to 5, the system comprising:
the system comprises an extraction module, a language model acquisition module and a display module, wherein the extraction module is used for acquiring the event relation data uploaded by a first event main body based on a event relation template, and extracting event type information in the event relation data and a second event main body corresponding to each event type information based on the language model;
the identification module is used for selecting a second event main body corresponding to the event type information when one piece of event type information is identified, determining a logic image between the first event main body and the selected second event main body, and identifying the logic image based on a neural network model to obtain event logic information;
the generation module is used for generating a case relation map according to the event type information and the event logic information between all the first event main bodies and all the second event main bodies;
the determining module is used for receiving the emergency information uploaded by any one first event main body, determining corresponding one-dimensional event type information and one-dimensional event logic information in a matter relation graph according to the first event main body and the emergency information, and determining a corresponding second event main body as a primary action event main body according to the one-dimensional event logic information;
the extraction module is used for extracting all the primary action event main bodies, determining two-dimensional event type information and two-dimensional event logic information of the primary action event main bodies corresponding to the one-dimensional event type information, extracting corresponding secondary action event main bodies according to the two-dimensional event type information and the two-dimensional event logic information, and enabling the primary action event main bodies and the secondary action event main bodies to act according to the corresponding one-dimensional event type information and the two-dimensional event type information;
the formulating module is used for formulating a maneuver plan if the actions of the primary action event main body and the secondary action event main body do not meet the preset requirement;
the generating of the event relation graph according to the event type information and the event logic information between all the first event main bodies and all the second event main bodies comprises the following steps:
establishing an initialization map according to all first event main bodies, wherein the initialization map is provided with nodes corresponding to each first event main body;
determining a second event main body corresponding to each first event main body, and connecting the node of the first event main body with the node of the corresponding second event main body according to the event type information and the event logic information;
after judging that all the nodes of the first event main body are respectively connected with the nodes of the corresponding second event main body, generating a matter relation graph;
the receiving of the emergency information uploaded by any one of the first event bodies, determining corresponding one-dimensional event type information and one-dimensional event logic information in a case relationship graph according to the first event body and the emergency information, and determining a corresponding second event body as a primary action event body according to the one-dimensional event logic information, further includes:
acquiring an emergency node corresponding to a first event main body uploading emergency information, and determining corresponding one-dimensional event type information according to the attribute of the emergency information;
determining one-dimensional event logic information corresponding to the one-dimensional event type information, and determining a second event main body corresponding to the first event main body according to the one-dimensional event logic information;
taking the second event main body as a primary action event main body, and taking a node corresponding to the second event main body as a primary action node;
the extracting all the primary action event main bodies, determining two-dimensional event type information and two-dimensional event logic information of the primary action event main bodies corresponding to the one-dimensional event type information, extracting corresponding secondary action event main bodies according to the two-dimensional event type information and the two-dimensional event logic information, and acting by the primary action event main bodies and the secondary action event main bodies according to the corresponding one-dimensional event type information and the two-dimensional event type information comprises the following steps:
obtaining corresponding two-dimensional event type information in a primary action event main body according to the attribute of the one-dimensional event type information;
determining two-dimensional event logic information corresponding to the two-dimensional event type information, and determining a secondary action node corresponding to the primary action node according to the two-dimensional event logic information;
extracting a second event main body corresponding to the secondary action node as a secondary action event main body;
if the actions of the primary action event main body and the secondary action event main body do not meet the preset requirements, a maneuver plan is formulated, and the exercise plan comprises the following steps:
counting all the first-stage action nodes to obtain the number of the first-stage nodes, and counting all the second-stage action nodes to obtain the number of the second-stage nodes;
acquiring one-dimensional response time corresponding to the completion of the one-dimensional event type information by all the first-level action nodes, and acquiring two-dimensional response time corresponding to the completion of the two-dimensional event type information by all the second-level action nodes;
calculating according to the number of the primary nodes, the number of the secondary nodes, the one-dimensional response time and the two-dimensional response time to obtain a processing coefficient corresponding to the emergency information;
Figure DEST_PATH_IMAGE019
wherein the content of the first and second substances,
Figure 593795DEST_PATH_IMAGE003
in order to calculate the processing coefficients,
Figure 320356DEST_PATH_IMAGE004
is a first normalized value of the first normalized value,
Figure 606981DEST_PATH_IMAGE005
is a standard quantity value and is used as a standard quantity value,
Figure 950369DEST_PATH_IMAGE006
in order to be a first-order weight,
Figure 375665DEST_PATH_IMAGE007
is a first
Figure 116088DEST_PATH_IMAGE009
The one-dimensional response time of the individual level one action nodes,
Figure 957136DEST_PATH_IMAGE010
is as follows
Figure 201036DEST_PATH_IMAGE009
The standard response time of an individual primary action node,
Figure 407020DEST_PATH_IMAGE011
is an upper limit value of the number of the primary action nodes,
Figure 634739DEST_PATH_IMAGE012
the number of the nodes at the first level,
Figure 13899DEST_PATH_IMAGE013
is the weight of the second level weight,
Figure 456513DEST_PATH_IMAGE014
is as follows
Figure 879404DEST_PATH_IMAGE015
The two-dimensional response time of each secondary action node,
Figure 79572DEST_PATH_IMAGE016
is as follows
Figure 777270DEST_PATH_IMAGE015
The standard response time of each secondary action node,
Figure 949757DEST_PATH_IMAGE017
is the upper limit value of the number of the secondary action nodes,
Figure 809128DEST_PATH_IMAGE018
the number of the secondary nodes;
and if the processing coefficient is judged to be larger than the preset coefficient, judging that the actions of the primary action event main body and the secondary action event main body do not meet the preset requirement, and making a exercise plan for the extracted primary action event main body and the extracted secondary action event main body.
CN202210569919.7A 2022-05-24 2022-05-24 Method and system for extracting and processing case-affair relation based on image model and language model Active CN114707004B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210569919.7A CN114707004B (en) 2022-05-24 2022-05-24 Method and system for extracting and processing case-affair relation based on image model and language model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210569919.7A CN114707004B (en) 2022-05-24 2022-05-24 Method and system for extracting and processing case-affair relation based on image model and language model

Publications (2)

Publication Number Publication Date
CN114707004A CN114707004A (en) 2022-07-05
CN114707004B true CN114707004B (en) 2022-08-16

Family

ID=82176617

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210569919.7A Active CN114707004B (en) 2022-05-24 2022-05-24 Method and system for extracting and processing case-affair relation based on image model and language model

Country Status (1)

Country Link
CN (1) CN114707004B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114969382B (en) * 2022-07-19 2022-10-21 国网浙江省电力有限公司信息通信分公司 Entity generation method based on event chain inference of event graph

Citations (32)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5920828A (en) * 1997-06-02 1999-07-06 Baker Hughes Incorporated Quality control seismic data processing system
CN107018095A (en) * 2017-03-29 2017-08-04 西安电子科技大学 Crosspoint analogue system and method based on discrete event
CN107908671A (en) * 2017-10-25 2018-04-13 南京擎盾信息科技有限公司 Knowledge mapping construction method and system based on law data
CN108052576A (en) * 2017-12-08 2018-05-18 国家计算机网络与信息安全管理中心 A kind of reason knowledge mapping construction method and system
CN110110870A (en) * 2019-06-05 2019-08-09 厦门邑通软件科技有限公司 A kind of equipment fault intelligent control method based on event graphical spectrum technology
CN110781317A (en) * 2019-10-29 2020-02-11 北京明略软件系统有限公司 Method and device for constructing event map and electronic equipment
CN110955575A (en) * 2019-11-14 2020-04-03 国网浙江省电力有限公司信息通信分公司 Business system fault positioning method based on correlation analysis model
CN110968699A (en) * 2019-11-01 2020-04-07 数地科技(北京)有限公司 Logic map construction and early warning method and device based on event recommendation
CN110968700A (en) * 2019-11-01 2020-04-07 数地科技(北京)有限公司 Domain event map construction method and device fusing multi-class affairs and entity knowledge
CN111639494A (en) * 2020-04-20 2020-09-08 北京大学 Case affair relation determining method and system
CN111768077A (en) * 2020-05-28 2020-10-13 国网浙江省电力有限公司绍兴供电公司 Intelligent power grid trip event identification method based on knowledge graph
CN112487207A (en) * 2020-12-09 2021-03-12 Oppo广东移动通信有限公司 Image multi-label classification method and device, computer equipment and storage medium
CN112905801A (en) * 2021-02-08 2021-06-04 携程旅游信息技术（上海）有限公司 Event map-based travel prediction method, system, device and storage medium
CN112966079A (en) * 2021-03-02 2021-06-15 中国电子科技集团公司第二十八研究所 Event portrait oriented text analysis method for dialog system
CN113268591A (en) * 2021-04-17 2021-08-17 中国人民解放军战略支援部队信息工程大学 Air target intention evidence judging method and system based on affair atlas
WO2021175009A1 (en) * 2020-03-02 2021-09-10 深圳壹账通智能科技有限公司 Early warning event graph construction method and apparatus, device, and storage medium
CN113379214A (en) * 2021-06-02 2021-09-10 国网福建省电力有限公司 Method for automatically filling and assisting decision of power grid accident information based on affair map
CN113407726A (en) * 2020-11-02 2021-09-17 中科嘉速（北京）信息技术有限公司 Emergency disposal plan method and system
WO2021189971A1 (en) * 2020-10-26 2021-09-30 平安科技（深圳）有限公司 Medical plan recommendation system and method based on knowledge graph representation learning
CN113535739A (en) * 2021-09-16 2021-10-22 国网浙江省电力有限公司信息通信分公司 Data market layer table establishing method based on power grid energy data
CN113590824A (en) * 2021-07-30 2021-11-02 平安科技（深圳）有限公司 Method and device for constructing causal graph and related equipment
WO2021254457A1 (en) * 2020-06-17 2021-12-23 第四范式（北京）技术有限公司 Method and device for constructing knowledge graph, computer device, and storage medium
CN113961678A (en) * 2021-09-15 2022-01-21 北京三快在线科技有限公司 Intelligent dialogue method and device based on affair map
CN114020936A (en) * 2022-01-06 2022-02-08 北京融信数联科技有限公司 Construction method and system of multi-modal affair map and readable storage medium
CN114064937A (en) * 2022-01-14 2022-02-18 云孚科技(北京)有限公司 Method and system for automatically constructing case map
CN114090794A (en) * 2021-11-29 2022-02-25 中国平安人寿保险股份有限公司 Event map construction method based on artificial intelligence and related equipment
CN114153988A (en) * 2021-12-02 2022-03-08 青岛科技大学 Chemical abnormal event detection method and system based on matter map
CN114281972A (en) * 2021-12-28 2022-04-05 亿迅信息技术有限公司 Dialog control method, system storage medium and server based on subject object tracking and cognitive inference
CN114398500A (en) * 2022-01-29 2022-04-26 哈尔滨工业大学 Event prediction method based on graph-enhanced pre-training model
CN114417004A (en) * 2021-11-10 2022-04-29 南京邮电大学 Method, device and system for fusing knowledge graph and case graph
CN114429126A (en) * 2021-12-17 2022-05-03 中国石油大学(华东) True and false message identification method based on reinforcement learning and affair knowledge graph
WO2022088526A1 (en) * 2020-10-30 2022-05-05 南京师范大学 Geographic knowledge acquisition method

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10803133B2 (en) * 2013-04-29 2020-10-13 Moogsoft Inc. System for decomposing events from managed infrastructures that includes a reference tool signalizer
US10013653B2 (en) * 2016-01-26 2018-07-03 Università della Svizzera italiana System and a method for learning features on geometric domains
CN111428049B (en) * 2020-03-20 2023-07-21 北京百度网讯科技有限公司 Event thematic generation method, device, equipment and storage medium

Patent Citations (33)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5920828A (en) * 1997-06-02 1999-07-06 Baker Hughes Incorporated Quality control seismic data processing system
CN107018095A (en) * 2017-03-29 2017-08-04 西安电子科技大学 Crosspoint analogue system and method based on discrete event
CN107908671A (en) * 2017-10-25 2018-04-13 南京擎盾信息科技有限公司 Knowledge mapping construction method and system based on law data
CN108052576A (en) * 2017-12-08 2018-05-18 国家计算机网络与信息安全管理中心 A kind of reason knowledge mapping construction method and system
WO2020244262A1 (en) * 2019-06-05 2020-12-10 厦门邑通软件科技有限公司 Device fault intelligent monitoring method based on event graph technology
CN110110870A (en) * 2019-06-05 2019-08-09 厦门邑通软件科技有限公司 A kind of equipment fault intelligent control method based on event graphical spectrum technology
CN110781317A (en) * 2019-10-29 2020-02-11 北京明略软件系统有限公司 Method and device for constructing event map and electronic equipment
CN110968699A (en) * 2019-11-01 2020-04-07 数地科技(北京)有限公司 Logic map construction and early warning method and device based on event recommendation
CN110968700A (en) * 2019-11-01 2020-04-07 数地科技(北京)有限公司 Domain event map construction method and device fusing multi-class affairs and entity knowledge
CN110955575A (en) * 2019-11-14 2020-04-03 国网浙江省电力有限公司信息通信分公司 Business system fault positioning method based on correlation analysis model
WO2021175009A1 (en) * 2020-03-02 2021-09-10 深圳壹账通智能科技有限公司 Early warning event graph construction method and apparatus, device, and storage medium
CN111639494A (en) * 2020-04-20 2020-09-08 北京大学 Case affair relation determining method and system
CN111768077A (en) * 2020-05-28 2020-10-13 国网浙江省电力有限公司绍兴供电公司 Intelligent power grid trip event identification method based on knowledge graph
WO2021254457A1 (en) * 2020-06-17 2021-12-23 第四范式（北京）技术有限公司 Method and device for constructing knowledge graph, computer device, and storage medium
WO2021189971A1 (en) * 2020-10-26 2021-09-30 平安科技（深圳）有限公司 Medical plan recommendation system and method based on knowledge graph representation learning
WO2022088526A1 (en) * 2020-10-30 2022-05-05 南京师范大学 Geographic knowledge acquisition method
CN113407726A (en) * 2020-11-02 2021-09-17 中科嘉速（北京）信息技术有限公司 Emergency disposal plan method and system
CN112487207A (en) * 2020-12-09 2021-03-12 Oppo广东移动通信有限公司 Image multi-label classification method and device, computer equipment and storage medium
CN112905801A (en) * 2021-02-08 2021-06-04 携程旅游信息技术（上海）有限公司 Event map-based travel prediction method, system, device and storage medium
CN112966079A (en) * 2021-03-02 2021-06-15 中国电子科技集团公司第二十八研究所 Event portrait oriented text analysis method for dialog system
CN113268591A (en) * 2021-04-17 2021-08-17 中国人民解放军战略支援部队信息工程大学 Air target intention evidence judging method and system based on affair atlas
CN113379214A (en) * 2021-06-02 2021-09-10 国网福建省电力有限公司 Method for automatically filling and assisting decision of power grid accident information based on affair map
CN113590824A (en) * 2021-07-30 2021-11-02 平安科技（深圳）有限公司 Method and device for constructing causal graph and related equipment
CN113961678A (en) * 2021-09-15 2022-01-21 北京三快在线科技有限公司 Intelligent dialogue method and device based on affair map
CN113535739A (en) * 2021-09-16 2021-10-22 国网浙江省电力有限公司信息通信分公司 Data market layer table establishing method based on power grid energy data
CN114417004A (en) * 2021-11-10 2022-04-29 南京邮电大学 Method, device and system for fusing knowledge graph and case graph
CN114090794A (en) * 2021-11-29 2022-02-25 中国平安人寿保险股份有限公司 Event map construction method based on artificial intelligence and related equipment
CN114153988A (en) * 2021-12-02 2022-03-08 青岛科技大学 Chemical abnormal event detection method and system based on matter map
CN114429126A (en) * 2021-12-17 2022-05-03 中国石油大学(华东) True and false message identification method based on reinforcement learning and affair knowledge graph
CN114281972A (en) * 2021-12-28 2022-04-05 亿迅信息技术有限公司 Dialog control method, system storage medium and server based on subject object tracking and cognitive inference
CN114020936A (en) * 2022-01-06 2022-02-08 北京融信数联科技有限公司 Construction method and system of multi-modal affair map and readable storage medium
CN114064937A (en) * 2022-01-14 2022-02-18 云孚科技(北京)有限公司 Method and system for automatically constructing case map
CN114398500A (en) * 2022-01-29 2022-04-26 哈尔滨工业大学 Event prediction method based on graph-enhanced pre-training model

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Anatomical object recognition and labeling by atlas-based focused non-rigid registration and region-growing;Leiguang Gong 等;《2008 International Conference on Audio, Language and Image Processing》;20081231;全文 *
基于数据科学思维的情报事理逻辑揭示与科学解读;刘如等;《情报理论与实践》;20180314(第08期);全文 *

Also Published As

Publication number Publication date
CN114707004A (en) 2022-07-05

Similar Documents

Publication Publication Date Title
Zhang et al. A gated peripheral-foveal convolutional neural network for unified image aesthetic prediction
CN110472090B (en) Image retrieval method based on semantic tags, related device and storage medium
CN107798027B (en) Information popularity prediction method, information recommendation method and device
CN111242019B (en) Video content detection method and device, electronic equipment and storage medium
CN114707004B (en) Method and system for extracting and processing case-affair relation based on image model and language model
CN110490237B (en) Data processing method and device, storage medium and electronic equipment
CN111522735B (en) Shunting method and device for test experiment
CN115423353A (en) Power distribution network resource consumption scheduling method and device, electronic equipment and storage medium
CN116862454B (en) Indoor building design management method and system
CN111967620A (en) Photovoltaic module diagnosis method, device, equipment and readable storage medium
CN112132208A (en) Image conversion model generation method and device, electronic equipment and storage medium
CN110502670A (en) Network social intercourse relationship knowledge mapping generation method and system based on artificial intelligence
CN109359675A (en) Image processing method and equipment
CN117112902A (en) Off-line reinforcement learning commodity recommendation system based on multi-mode contrast learning
CN114663261B (en) Data processing method suitable for training and examination system
CN116522131A (en) Object representation method, device, electronic equipment and computer readable storage medium
CN116166910A (en) Social media account vermicelli water army detection method, system, equipment and medium
CN114693435A (en) Intelligent return visit method and device for collection list, electronic equipment and storage medium
CN112565797B (en) Page request processing method and device, electronic equipment and medium
CN113673229A (en) Electric power marketing data interaction method, system and storage medium
CN112199290A (en) UI element positioning method and system based on machine learning
CN113190589A (en) Content distribution method and device suitable for distribution system and storage medium
CN112270296A (en) Cloud platform based smart city visual management system and method
CN111125536A (en) Information pushing method and device, computer equipment and storage medium
CN116562653B (en) Distributed energy station area line loss monitoring method and system

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