CN115964503A - Safety risk prediction method and system based on community equipment facilities - Google Patents
Safety risk prediction method and system based on community equipment facilities Download PDFInfo
- Publication number
- CN115964503A CN115964503A CN202111630708.1A CN202111630708A CN115964503A CN 115964503 A CN115964503 A CN 115964503A CN 202111630708 A CN202111630708 A CN 202111630708A CN 115964503 A CN115964503 A CN 115964503A
- Authority
- CN
- China
- Prior art keywords
- time
- equipment
- community
- historical
- facility
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention provides a security risk prediction method and a security risk prediction system based on community equipment facilities, wherein the method comprises the following steps: respectively constructing a historical moment dynamic knowledge graph set and a current moment dynamic knowledge graph of the target community based on equipment safety data of each equipment facility in the target community at different moments; sequentially inputting historical time dynamic knowledge maps in the historical time dynamic knowledge map set into an equipment facility historical knowledge map prediction model according to a time sequence relation to obtain a first equipment facility safety risk prediction result; inputting the current-time dynamic knowledge graph into a current-time knowledge graph prediction model of the equipment facility to obtain a second equipment facility safety risk prediction result; and fusing the first equipment facility safety risk prediction result and the second equipment facility safety risk prediction result to obtain a final equipment facility safety risk prediction result. The method improves the accuracy of community security risk prediction.
Description
Technical Field
The invention relates to the technical field of community risk assessment, in particular to a security risk prediction method and system based on community equipment facilities.
Background
The conventional community equipment facility safety risk studying and judging method is mainly an expert evaluating method, requires mutual discussion of multiple related professionals, draws collected community equipment facility data into a table, and performs manual analysis based on the table data, so that a large amount of manpower and material resources are consumed, the method is easily influenced by subjective assumption of related personnel, and the prediction accuracy is insufficient.
In recent years, with rapid development of technologies such as machine learning, big data analysis, and artificial intelligence, community equipment and facility security risk research and judgment methods based on machine learning have attracted extensive attention and have been primarily attempted. A related safety detection model is established through a large amount of community equipment facility safety data in the prior art, and new community data are predicted and risk study and judgment based on the model.
However, although the conventional machine learning method effectively reduces the consumption of manpower and material resources and the time for prediction, the requirement for data is extremely strict, and structured data needs to be used. Due to the fact that the situation of community equipment and facility data is complex, effective structured data is difficult to extract and form, the data has isomerism, effective information cannot be updated and intercepted in real time through a traditional method, influences existing among the data are ignored, certain limitation exists, and the accuracy of community risk prediction is reduced. Therefore, there is a need for a method and system for predicting security risk based on equipment facilities to solve the above problems.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a security risk prediction method and system based on community equipment facilities.
The invention provides a security risk prediction method based on community equipment facilities, which comprises the following steps:
respectively constructing a historical moment dynamic knowledge graph set and a current moment dynamic knowledge graph of a target community based on equipment safety data of each equipment facility in the target community at different moments;
sequentially inputting the historical moment dynamic knowledge maps in the historical moment dynamic knowledge map set into an equipment facility historical knowledge map prediction model according to a time sequence relation to obtain a first equipment facility safety risk prediction result of the target community at the next moment; the device facility historical knowledge map prediction model is obtained by training a neural network through a sample historical time dynamic knowledge map corresponding to each time before a sample time;
inputting the current-time dynamic knowledge graph into a device facility current-time knowledge graph prediction model to obtain a second device facility security risk prediction result of the target community at the next time; the device facility current moment knowledge graph prediction model is obtained by training a relation graph convolution network through a sample moment safety dynamic knowledge graph corresponding to a sample moment;
and fusing the first equipment facility security risk prediction result and the second equipment facility security risk prediction result to obtain a final equipment facility security risk prediction result of the target community at the next moment.
According to the security risk prediction method based on the community equipment facilities, provided by the invention, based on the equipment security data of each equipment facility in the target community at different time, a historical time dynamic knowledge graph set and a current time dynamic knowledge graph of the target community are respectively constructed, and the method comprises the following steps:
acquiring entity types in the target community and entity relationships among each entity according to the equipment safety data;
according to the entity type and the entity relation, historical time information is used as a timestamp to form a quadruple corresponding to each historical time information, and a dynamic knowledge map set of the historical time is constructed based on the quadruple corresponding to each historical time information according to a time sequence relation;
according to the entity type and the entity relation, the current time information is used as a time stamp to form a quadruple corresponding to the current time information, and according to the quadruple corresponding to the current time information, a dynamic knowledge map at the current time is obtained;
and the tail entity in the quadruple corresponding to each historical moment information and the quadruple corresponding to the current moment information is a safety risk result of the equipment and the facility to be predicted.
According to the safety risk prediction method based on the community equipment facility, the historical knowledge map prediction model of the equipment facility is obtained by constructing a multilayer perceptron and a gating circulation unit.
According to the safety risk prediction method based on the community equipment facilities, provided by the invention, the equipment facility historical knowledge map prediction model is obtained through the following steps:
taking any sample time as a reference time, acquiring a sample historical time dynamic knowledge graph corresponding to each historical time of a target community before the reference time, wherein a quadruple corresponding to the sample historical time dynamic knowledge graph is composed of a head entity, an entity relation, a first sample tail entity and timestamp information;
inputting head entities, entity relations and timestamp information in a plurality of sample historical moment dynamic knowledge maps into the multilayer perceptron for training to obtain an aggregation vector output by the multilayer perceptron;
sequentially inputting a plurality of aggregation vectors into the gated circulation unit according to a time sequence relation for training, and outputting a predicted first sample tail entity;
and performing loss calculation according to the first sample tail entity and the predicted first sample tail entity, and if the obtained loss value meets a first preset training threshold value, obtaining an equipment facility historical knowledge map prediction model.
According to the safety risk prediction method based on the community equipment facilities, provided by the invention, the current-time knowledge graph prediction model of the equipment facilities is obtained through the following steps:
acquiring a sample time safety dynamic knowledge graph based on sample equipment safety data of a target community at a sample time; the security dynamic knowledge graph at the sample time is generated by a quadruple including a head entity, an entity relation, a second tail entity and timestamp information;
inputting head entities, entity relationships and timestamp information in the multiple sample time safety dynamic knowledge graphs into a relationship graph convolution network for training, and outputting predicted second sample tail entities;
and performing loss calculation according to the second sample tail entity and the predicted second sample tail entity, and if the obtained loss value meets a second preset training threshold value, obtaining a knowledge graph prediction model of the equipment at the current moment.
According to the security risk prediction method based on the community equipment facilities, provided by the invention, the first equipment facility security risk prediction result and the second equipment facility security risk prediction result are fused to obtain the final equipment facility security risk prediction result of the target community at the next moment, and the method comprises the following steps:
adjusting the weights corresponding to the first equipment facility safety risk prediction result and the second equipment facility safety risk prediction result respectively through a particle swarm algorithm;
and summing the first equipment facility security risk prediction result and the second equipment facility security risk prediction result after the weight adjustment, and determining a final equipment facility security risk prediction result of the target community at the next moment.
The invention also provides a security risk prediction system based on community equipment facilities, which comprises:
the time sequence knowledge graph building module is used for respectively building a historical moment dynamic knowledge graph set and a current moment dynamic knowledge graph of the target community based on the equipment safety data of each equipment facility in the target community at different moments;
the historical knowledge map risk prediction module is used for sequentially inputting the historical moment dynamic knowledge maps in the historical moment dynamic knowledge map set into an equipment facility historical knowledge map prediction model according to a time sequence relation to obtain a first equipment facility safety risk prediction result of the target community at the next moment; the device facility historical knowledge map prediction model is obtained by training a neural network through a sample historical time dynamic knowledge map corresponding to each time before a sample time;
the current-time knowledge graph prediction module is used for inputting the current-time dynamic knowledge graph into an equipment facility current-time knowledge graph prediction model to obtain a second equipment facility safety risk prediction result of the target community at the next time; the device facility current moment knowledge graph prediction model is obtained by training a relation graph convolution network through a sample moment safety dynamic knowledge graph corresponding to a sample moment;
and the fusion reasoning module is used for fusing the first equipment facility safety risk prediction result and the second equipment facility safety risk prediction result to obtain a final equipment facility safety risk prediction result of the target community at the next moment.
The invention also provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the steps of the security risk prediction method based on the community equipment facility.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the community equipment facility-based security risk prediction method as described in any of the above.
The invention also provides a computer program product comprising a computer program which, when executed by a processor, performs the steps of the method for predicting security risk based on a community equipment facility as described in any one of the above.
According to the safety risk prediction method and system based on the community equipment facilities, the triples of the community knowledge maps are expanded into the quadruples with the time sequence information, so that the safety dynamic knowledge maps of the community equipment facilities are constructed based on the quadruples, the safety risk level of the community equipment facilities at a future moment is predicted according to the constructed historical moment dynamic knowledge maps and the current moment dynamic knowledge maps, and the accuracy of the community safety risk prediction is improved.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic flow chart of a security risk prediction method based on community equipment facilities according to the present invention;
FIG. 2 is a schematic structural diagram of a security risk prediction system based on community equipment facilities provided by the present invention;
fig. 3 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
As machine learning is widely applied to risk study and judgment and risk inference, a machine learning model for community risk prediction and evaluation is also receiving wide attention. Due to the rise of the knowledge graph, the storage mode of the data is greatly changed, the real world data can be displayed and stored in the form of the knowledge graph, and the comprehensive community data information can be more abundantly displayed through the knowledge graph, so that the requirement on the data of the algorithm for predicting based on the knowledge graph is not as strict as that of other algorithms.
The conventional community knowledge graph is a static knowledge graph constructed based on static data, and the characteristics of a target entity and the characteristics of connected neighbor entities are extracted through the static knowledge graph, so that data are processed, and a community risk prediction result is obtained. However, the above static knowledge graph-based community risk inference method fails to consider the influence of historical data on the present and future, because the present static knowledge graph is formed in a triple form, the established static knowledge graph cannot reflect the characteristic that the comprehensive risk of the community changes along with time, the characteristic that the comprehensive community data has time sequence is not considered, the risk can be researched and judged only according to the present data flat-laying type expansion, the risk data information changing along with time cannot be processed, the expansion of knowledge graph information and the autonomous learning of a model cannot be completed, and the total comprehensive risk of the community to occur at the future time cannot be predicted.
The invention provides a community security risk prediction method based on a time sequence dynamic knowledge graph, which constructs the time sequence dynamic knowledge graph according to the historical data of community equipment facilities in a quadruple (head entity, relation, tail entity and time) mode, excavates the mutual influence among data according to the constructed time sequence dynamic knowledge graph, effectively utilizes the security historical data information of the community equipment facilities, and predicts the security risk of the future community equipment facilities based on the entity and the relation at the next moment.
Fig. 1 is a schematic flow chart of a security risk prediction method based on a community facility according to the present invention, and as shown in fig. 1, the present invention provides a security risk prediction method based on a community facility, including:
In the invention, firstly, equipment safety data of a target community are screened and extracted, and in order to utilize the time sequence characteristics of the equipment safety data, a time stamp is introduced to construct a quadruple corresponding to each community equipment facility, wherein the quadruple comprises { a head entity, a relation between entities, a tail entity and a time stamp }. Specifically, based on timestamp information, a dynamic knowledge graph of historical time is constructed according to quadruples of historical time before the current time by taking the current time as a reference time, and the dynamic knowledge graph is sorted according to a time sequence relation, so that a set of dynamic knowledge graphs related to the historical time is obtained; and further, constructing the dynamic knowledge graph at the current time according to the quadruple at the current time.
Since the degree of the security risk of the equipment facilities in the community can change along with the time, the change process of the security risk of the equipment facilities in the community can be comprehensively reflected by representing the characteristics related to the risk elements at different times. The method comprises the steps of inputting each target entity and the connected relation characteristics thereof at each moment into an equipment facility historical knowledge map prediction model constructed by a Multi-Layer perceptron (MLP) and a Gated Recurrent Neural (GRU) by using the existing entity data, relation data and time data, namely a historical moment dynamic knowledge map set, aggregating the characteristics of each target entity and the connected relation thereof at the historical moment through the MLP, and inputting the information obtained by aggregation into the GRU according to the time sequence, so that the probability of the equipment facility security event of the target community at the next moment, namely a first equipment facility security risk prediction result, is predicted based on the historical moment dynamic knowledge map set.
103, inputting the dynamic knowledge graph at the current moment into a knowledge graph prediction model of the equipment at the current moment to obtain a second equipment safety risk prediction result of the target community at the next moment; the equipment facility current-time knowledge graph prediction model is obtained by training a relation graph convolution network through a sample-time safety dynamic knowledge graph corresponding to a sample time.
In the invention, global information of the dynamic knowledge Graph at the current time is aggregated through a knowledge Graph prediction model of the equipment facility at the current time, which is obtained by training a Relational Graph Convolutional Network (RGCN for short). Specifically, the device facility current-time knowledge graph prediction model aggregates information in the current-time knowledge graph, performs weighted summation on information of a target entity and neighbor entities of the target entity at the current time, and obtains the probability of a device facility security event occurring in a target community at the current time through a Softmax activation function, namely, a second device facility security risk prediction result.
And 104, fusing the first equipment facility security risk prediction result and the second equipment facility security risk prediction result to obtain a final equipment facility security risk prediction result of the target community at the next moment.
According to the method, the prediction result of the historical knowledge map prediction model of the equipment facility and the prediction result of the knowledge map prediction model of the equipment facility at the current moment are subjected to weighted fusion to obtain the final prediction result of the target community, so that an event which will occur in the target community at a future moment, such as community risk level rise, can be obtained, the comprehensive risk trend of the safety of the equipment facility of the community can be effectively predicted, a reasonable theoretical basis is provided for decision-makers, early warning prevention and control are facilitated in each link of the safety of the equipment facility of the community, and the safety of activities is guaranteed.
According to the safety risk prediction method based on the community equipment facilities, the triples of the community knowledge maps are expanded into the quadruples with the time sequence information, so that the safety dynamic knowledge maps of the community equipment facilities are constructed based on the quadruples, the safety risk level of the community equipment facilities at a future moment is predicted according to the constructed historical moment dynamic knowledge maps and the constructed current moment dynamic knowledge maps, and the accuracy of the community safety risk prediction is improved.
On the basis of the above embodiment, the respectively constructing a historical time dynamic knowledge graph set and a current time dynamic knowledge graph of the target community based on the device security data of each device facility in the target community at different times includes:
In the invention, firstly, security risk events possibly occurring in community equipment facilities are analyzed, different types of entities are extracted from various data information according to acquired heterogeneous data (namely equipment security data) of the community equipment, the connection relation among the entities and corresponding timestamps of the entities are defined to construct a quadruple, and the security dynamic knowledge maps corresponding to the community equipment facilities at historical time and current time are established according to the quadruple and the timestamp information.
and the four-tuple corresponding to each historical moment information and the tail entity in the four-tuple corresponding to the current moment information are the safety risk results of the equipment and the facilities to be predicted.
In the invention, the community equipment and facility safety data mainly aim at the equipment and facility data of six systems of community gas, fire fighting, power supply and distribution, heating ventilation and air conditioning, water supply and drainage and elevators, including gas, temperature of a gas well chamber, environmental data of a gas pressure regulating station, combustible gas, indoor gas, fire hydrant water pressure, fire water tank liquid level, spraying tail end water pressure, electrical parameters, heating ventilation and air conditioning water tanks, smoke fire detectors, water quality of water supply and drainage, elevator running speed, door opening and closing speed and the like. Further, according to the characteristics that the interaction between the community equipment facilities may have risks, the safety risks of the community equipment facilities at the future time, the connection, succession and continuation characteristics between the past historical examples and the data at the current time, the community equipment facility safety dynamic knowledge graph with the time sequence relationship is constructed.
In order to construct a complex scene of a community system, the invention uses the knowledge graph for the scene construction of the community system. A knowledge graph is essentially a semantic network, a graph-based data structure, which is composed of entities and edges, wherein the entities represent things existing in the real world, and the edges represent connection relationships between the entities. In order to determine the entity type and the entity characteristics in the knowledge graph, according to the composition characteristics and the risk characteristics of the community system, the reliability of various facilities, various community personnel and related units in the community system, the artificially generated unsafe behavior in the community system, the unsafe state of the facilities, the defects of management of the related units and an event chain caused by the combined action of environmental factors are comprehensively considered, and the dynamic knowledge graph for evaluating the risk of the community system is constructed.
In the invention, the entity names in the knowledge graph finally determined based on the community equipment facility comprise four main classes which are respectively dividedComprises the following steps: various kinds of people related to the community, various facilities within the community, various units related to the community system, and various types of accidents caused thereby. Specifically, in the present invention, the security dynamic knowledge graph based on the community facility is a set composed of four-tuple, each of which represents the information of objective fact in the form of g =<s,r,o,t k >Wherein s represents a head entity, o represents a tail entity, and the head entity s and the tail entity o belong to an entity E; r represents the relationship between the head entity s and the tail entity o, and R belongs to R; t is t k Time stamp information is represented.
Different entity types are provided under different entities, and the method specifically comprises the following steps:
people of all types = { community residents, staff };
facility = { gas well, gas pressure regulating station, gas terminal, fire hydrant, fire water tank, elevator, electric parameter collector, heating and ventilation air conditioner, smoke alarm, electrical equipment, well lid, pressure regulator };
each type unit = { community property unit, community living committee };
risk class = { low risk, medium risk, high risk }.
And simultaneously, defining the connection relation between the entities in the knowledge graph:
relationship set = { having, attribute, normal, mild anomaly, error, failure };
the attribute types are respectively characterized by = { temperature, operating speed, acceleration, door opening and closing speed, combustible gas concentration, smoke concentration, current, load, pressure and water level };
and the timestamp information is added, so that a safety dynamic knowledge graph based on the community equipment facilities is constructed, and the conversion from a community system risk assessment index system to a community system knowledge graph is realized. It is noted that the knowledge-graph can be<Entity, relationship, entity>Or<Entity, attribute value>Two expression modes of (1). In particular, when employed<Entity, attribute value>In such a way that at different times, the relevant characteristics of the gas or tank, such as the gas concentration or the water pressure in the tank, change over time, for example at t 1 At time, the tank level and water pressure are both at low load, at t 2 At all times, the tank water level and water pressure are at high load. According to the method, the time information is added, the obtained security dynamic knowledge map is constructed, the influence of the attribute characteristics of the community equipment facilities changing along with time on the community security risk is considered, and the community security risk prediction is carried out by combining the characteristic change trend of the community equipment facilities at the historical time and the real-time characteristics of the community equipment facilities at the current time. It should be noted that, in the present invention, the tail entity in the dynamic knowledge graph is obtained through model prediction, and besides the risk level of the community equipment and facilities is obtained through prediction, the attribute values of different entities can also be predicted, for example, the load condition of the water level of the water tank is predicted in the above embodiment, and when the load is high, the risk of community safety can also be evaluated.
On the basis of the embodiment, the historical knowledge map prediction model of the equipment facility is constructed by a multilayer perceptron and a gating circulation unit.
In the invention, the model of the multilayer perceptron is simpler, the speed of information aggregation can be accelerated by using the multilayer perceptron, and the aggregation effect meets the design requirement; and the gating circulation unit can better capture the dependence relationship with larger time step distance in the time sequence and has better presentation effect on the time sequence data.
On the basis of the above embodiment, the equipment facility historical knowledge map prediction model is obtained by the following steps:
taking any sample time as a reference time, acquiring a sample historical time dynamic knowledge graph corresponding to each historical time of a target community before the reference time, wherein a quadruple corresponding to the sample historical time dynamic knowledge graph is composed of a head entity, an entity relation, a first sample tail entity and timestamp information;
inputting head entities, entity relations and timestamp information in a plurality of sample historical moment dynamic knowledge maps into the multilayer perceptron for training to obtain an aggregation vector output by the multilayer perceptron;
sequentially inputting a plurality of aggregation vectors into the gated circulation unit according to a time sequence relation for training, and outputting a predicted first sample tail entity;
and performing loss calculation according to the first sample tail entity and the predicted first sample tail entity, and if the obtained loss value meets a first preset training threshold value, obtaining an equipment facility historical knowledge map prediction model.
In the invention, by training the multi-layer perceptron and the gating cycle unit, the historical knowledge map prediction model of the equipment facility obtained by training can identify historical repeated events and predict future events by copying historical known facts. Since the community equipment security data can change along with the change of time, in the equipment and facility historical knowledge map prediction model, historical information needs to be inherited, screened and forgotten.
Further, the invention constructs a sample historical time dynamic knowledge graph of historical time before the reference time based on a certain sample time as the reference time, and for the constructed sample historical time dynamic knowledge graph, in the training process of the model, the characteristics of the target entity (namely, the head entity) under each sample historical time need to be aggregated, and the characteristics of the target entity comprise the self-attribute of the target entity, the relationship between adjacent entities and the timestamp information.
Specifically, if the query (s, r, k ) With target-specific entities s and time steps t k Historical vocabulary of relationships r of time
Then the model will increase the estimated probability of the selected subject entity (tail entity) in the historical vocabulary when trained. In the training process, after the dynamic knowledge graph of the sample at the historical moment is input into the multilayer perceptron, firstly, the multilayer perceptron is used for generating a containing head realityBank, relation, timestamp based aggregate vector @>
t k ＝t k-1 +t u ；
Wherein, W c ∈R 3dxN And b c ∈R N As a learnable parameter, t u Is one unit time step. In the invention, the index vector is an N-dimensional vector, and N represents the cardinality of the whole community equipment facility security history entity dictionary E.
The method for predicting the security risk of the community equipment facility to be generated at a future time through the community data existing in past historical time needs a neural network which can store and memorize the past security information of the community equipment facility and can predict the security risk of the community equipment facility at a future time. The recurrent neural network is a neural network that processes time-series data, which refers to data collected at different time points, such data reflecting the state or degree of change of a certain object, phenomenon, etc. over time. The invention selects the gate control circulation unit as a time sequence Network to be used for establishing a facility historical knowledge map prediction model with a multilayer sensing mechanism from the viewpoints of information screening and Network lightweight in the aspect of better representing effect in the face of time sequence data.
Specifically, the gated loop unit includes two gates: reset gate and refresh gate, through which the flow of information can be controlled. GRUs have the following characteristics: 1. the safety information of community equipment facilities at the current moment can be captured, and historical information is screened and combined with new information; 2. the number of GRU model parameters does not increase with time. Therefore, all past recorded community data can be trained by the GRU to predict the safety risk of community equipment and facilities at a future time.
Further, for each sample, vectors aggregated at historical time
Inputting the information into the GRU in the time series order, and using a Softmax function to output the information so as to obtain the probability that the equipment facility historical knowledge base prediction model predicts the object entity:
wherein the content of the first and second substances,
represents t k The contents memorized by the GRU at the moment; p (c) is a vector equal to the vocabulary of the security entity of the entire community facility and represents the prediction probability of the historical knowledge-graph prediction model of the facility. By using GRU, the invention can discard part of history information, filter unimportant information and prevent the size of inherited vector from being excessively accumulated along with time.
Specifically, the multilayer perceptron aggregates information at each historical moment, the aggregated information is screened through a gating cycle unit (GRU), and finally the obtained information is subjected to Softmax normalization to obtain the prediction probability output by the historical knowledge graph prediction model of the facility, namely the prediction probability of each tail entity o. And when the error between the tail entity and the real tail entity obtained by the prediction of the equipment facility historical knowledge map prediction model is smaller than a preset training threshold, judging that the model completes training. It should be noted that the historical knowledge map prediction model of the equipment and facility constructed by the invention not only can process historical data, but also can process updated information at the current moment, so that the model can be continuously optimized and adjusted.
On the basis of the above embodiment, the knowledge-graph prediction model of the current time of the equipment facility is obtained by the following steps:
acquiring a sample time safety dynamic knowledge graph based on sample equipment safety data of a target community at a sample time; the sample time safety dynamic knowledge graph is generated by a quadruple including a head entity, an entity relation, a second tail entity and timestamp information;
inputting head entities, entity relations and timestamp information in the plurality of sample time safety dynamic knowledge maps into a relation map convolution network for training, and outputting a second sample tail entity obtained through prediction;
and performing loss calculation according to the second sample tail entity and the predicted second sample tail entity, and if the obtained loss value meets a second preset training threshold value, obtaining a knowledge graph prediction model of the equipment at the current moment.
In the invention, the security risk of the community equipment facility is influenced by historical time data, and the security data of the community equipment facility can be directly influenced because the information at the current time is changed, such as the change of conditions such as temperature and the like. Therefore, it is necessary to integrate data such as a community environment at the current time based on community history data to comprehensively predict a community risk.
In order to better discover the correlation in the knowledge graph of the security risk of the community equipment facility and complete the information aggregation of the knowledge graph at the current time, the invention trains the relation graph convolution network, obtains the prediction model of the knowledge graph at the current time of the equipment facility when the error between the predicted value and the true value meets the preset threshold value, calculates the information and the deviation of the adjacent entity of the given entity through the model, and the adjacent entity can be expanded to a two-hop distance, thereby constructing high-order adjacent information and capturing the potential risk.
The relationship graph convolution network can realize heterogeneous graph modeling, can combine multiple relationship information of entities in a community equipment facility security dynamic knowledge graph, classifies the neighbor entities by considering different relationships when processing neighbor entity data, introduces different weight parameters for the neighbor entities of each relationship, and performs total aggregation after aggregating the neighbor entities belonging to the same relationship type. The aggregator formula for the graph convolution network is as follows:
wherein p is g R represents all relation sets in the community equipment facility security dynamic knowledge graph for the probability of the target entity at the current moment,
representing the number of neighbor entities having r relation with the head entity s; c. C s Is a normalized parameter, the invention selects c s Is the number of neighbor entities having an r relationship with the head entity s, i.e. < >>
Represents a weight parameter corresponding to a neighbor entity having an r relationship, based on a value of a variable>
Represents a characteristic of the neighbor entity, is asserted>
Represents the weight parameter corresponding to the head entity s itself, is/are>
Representing the characteristics of the target entity s. Local graph structures for connections between each relationship and entity by aggregating information of neighboring entities of a target entity, i.e. </or >>
A piece of information under the same kind of relationship is gathered on each entity. By aggregating all the messages of the relationships in the dynamic knowledge-graph at the current time, i.e.
The overall information for each entity is further calculated. Finally, by adding the information of the own entity, i.e. w s q s And the information obtained after the summation is subjected to a Softmax activation function, so that the target entity probability p at the current moment output by the relational graph convolution network is obtained g 。
Further, the model is operated by a two-layer aggregation: firstly, independently aggregating neighbor entities with the same relationship, wherein for each type of relationship, the directionality of the relationship is also considered, and meanwhile, the self-connection relationship is added for the self-connection; further, after aggregating all the neighbors of different relations, the total aggregation is performed again.
In the present invention, the same query quadruple (s, r, are, t) as before is given k ) The current-time knowledge-graph prediction model predicts a fact by selecting an object entity from the entire entity vocabulary epsilon, by treating the predicted fact as a completely new fact without reference to history.
On the basis of the foregoing embodiment, the merging the first device facility security risk prediction result and the second device facility security risk prediction result to obtain a final device facility security risk prediction result of the target community at the next time includes:
adjusting the weights corresponding to the first equipment facility safety risk prediction result and the second equipment facility safety risk prediction result respectively through a particle swarm algorithm;
and summing the first equipment facility security risk prediction result and the second equipment facility security risk prediction result after the weight adjustment, and determining a final equipment facility security risk prediction result of the target community at the next moment.
In the present invention, (s, r, k ) And performing prediction, wherein the historical knowledge graph prediction model of the equipment and the knowledge graph prediction model of the current moment of the equipment give the prediction object entity with the highest probability in the candidate space. In order to ensure that the sum of the probabilities of all the entities in the entity vocabulary epsilon appears as 1, the weight between the respective predicted results of the equipment facility historical knowledge graph prediction model and the equipment facility current time knowledge graph prediction model is adjusted by adding a correlation coefficient w (namely, dynamic weight), wherein the correlation coefficient w is a learnable parameter.
The method adopts a Particle Swarm Optimization (PSO) to find the optimal solution through cooperation and information sharing among individuals in a group, and the optimal correlation coefficient w is obtained.
Specifically, the prediction results of the two models are initialized to a group of random particles (i.e. a random solution) by a particle swarm algorithm; the optimal solution is then found by iteration. In each iteration, the particle passes through two "extrema" (p) tracks id ，p gd ) To update itself. After finding these two optimal values, the particle updates its velocity and position by the following formula:
v i ＝w·v id +c 1 ·r 1 ·(p id -x id )+c 2 ·r 2 ·(p gd -x id )；
x id ＝x id +v id ；
wherein v is id Denotes the particle velocity, x id Representing the current position of the particle, w is an inertia factor; c. C 1 And c 2 Are learning factors, all typically set to 2; r is 1 And r 2 Is a random number with a value ranging from 0 to 1.
The dynamic weight w can be changed linearly in the particle swarm algorithm searching process, and can also be changed dynamically according to a certain measure function of the particle swarm algorithm performance. The invention adopts the following steps that:
w＝(w ini -w end )(G k -g)/G k +w end ；
wherein G is k Is the maximum number of iterations, w ini Is the initial inertial weight, w end Is the inertia weight when iterating to the maximum evolution algebra.
Because the particle swarm algorithm needs a membership function to adjust the weight w, the position (marked as f) of a prediction result obtained by fusion in the matrix is recorded in the training process 1 ) The position of the result in the matrix (denoted as f) with the real data 2 ) In comparison, L is the objective function, denoted as f 1 And f 2 The difference, as L is smaller, the weight w is more consistent with the desired design:
L＝|f 1 -f 2 |；
finally, according to the weight calculated by the particle swarm algorithm, probability predictions obtained by the two models can be combined, and the final prediction is the tail entity o t Entity with highest combined probability:
p(o|s,p,t)＝w*p(c)+(1-w)*p(g)；
o t ＝argmax o∈ε p(o|s,r,t)；
and p (o | s, p, t) is the prediction of the model on each entity in the entity set E, and is determined by the probability of the equipment facility historical knowledge graph prediction model and the probability of the equipment facility current knowledge graph prediction model on the community equipment facility safety risk prediction, and the prediction probabilities of all tail entities are added to be 1. In the invention, the prediction results output by the two models are fused to obtain the final prediction result. And further, screening the predicted target tail entities, and selecting the tail entity with the highest probability as a prediction result of the model for the safety risk of the community equipment facilities. Specifically, the prediction result comprises the risk level of the community to be predicted, which is divided into three levels, namely low risk, medium risk and high risk, so that the management personnel can be assisted in making decisions.
In one embodiment, a security risk prediction process based on community equipment facilities is generally described. Firstly, by analyzing a security risk event of community equipment facilities, extracting different types of entities from various data information according to acquired heterogeneous data of the community equipment, defining the connection relation among the entities and corresponding timestamps of the entities, thereby constructing a quadruple, and establishing a security dynamic knowledge graph of the community equipment facilities, wherein the security dynamic knowledge graph comprises a historical moment dynamic knowledge graph set and a current moment dynamic knowledge graph which are used as data sets to be input of corresponding models;
then, using a multilayer perceptron in the historical knowledge map prediction model of the equipment facility, aggregating historical information in the dynamic knowledge map set at the historical moment, sending the aggregated information into a gating circulation unit, and obtaining a first equipment facility safety risk prediction result of the target community at the next moment through a Softmax activation function;
meanwhile, a multi-relation aggregator in the knowledge graph prediction model of the equipment facility at the current moment is used for aggregating the target entities, the relation among the entities and the time information at the current moment, and a second equipment facility safety risk prediction result of the target community at the next moment is obtained through linear regression and a Softmax activation function;
and finally, weighting the prediction results obtained by the equipment facility historical knowledge graph prediction model and the equipment facility current-time knowledge graph prediction model respectively, updating the weight through a particle swarm algorithm to obtain the final prediction result of the model, realizing reasoning of the security incident risk of the community equipment facility, and predicting the risk level of the community.
The method screens the community equipment data, selects six system data in the community equipment as a data set for constructing the dynamic knowledge map, adds the traditional knowledge map triple into the time dimension by introducing the concept of the timestamp to become the knowledge map quadruple, increases the precedence information of the entity and the relation in the knowledge map on the time dimension, and can dynamically display the safety data of the community equipment facilities. And then according to the prediction object entity with the highest probability generated by the equipment facility historical knowledge graph prediction model and the equipment facility current time knowledge graph prediction model, the automatic updating of the weight is realized through a particle swarm algorithm, the weighted summation processing is carried out on the prediction results of the two models, and finally the community safety risk prediction result with higher prediction accuracy is obtained.
The security risk prediction system based on the community equipment facilities provided by the invention is described below, and the security risk prediction system based on the community equipment facilities described below and the security risk prediction method based on the community equipment facilities described above can be correspondingly referred to each other.
Fig. 2 is a schematic structural diagram of a security risk prediction system based on a community equipment facility, as shown in fig. 2, the security risk prediction system based on a community equipment facility includes a time sequence knowledge graph construction module 201, a historical knowledge graph risk prediction module 202, a current time knowledge graph prediction module 203, and a fusion inference module 204, where the time sequence knowledge graph construction module 201 is configured to respectively construct a historical time dynamic knowledge graph set and a current time dynamic knowledge graph of a target community based on device security data of each equipment facility in the target community at different times; the historical knowledge graph risk prediction module 202 is configured to sequentially input historical time dynamic knowledge graphs in the historical time dynamic knowledge graph set into an equipment facility historical knowledge graph prediction model according to a time sequence relationship, so as to obtain a first equipment facility security risk prediction result of the target community at the next time; the device facility historical knowledge map prediction model is obtained by training a neural network through a sample historical time dynamic knowledge map corresponding to each time before a sample time; the current-time knowledge graph prediction module 203 is configured to input the current-time dynamic knowledge graph into an equipment facility current-time knowledge graph prediction model, so as to obtain a second equipment facility security risk prediction result of the target community at the next time; the device facility current time knowledge graph prediction model is obtained by training a relation graph convolution network through a sample time safety dynamic knowledge graph corresponding to a sample time; the fusion inference module 204 is configured to fuse the first equipment facility security risk prediction result and the second equipment facility security risk prediction result to obtain a final equipment facility security risk prediction result of the target community at the next moment.
According to the safety risk prediction system based on the community equipment facilities, the triples of the community knowledge maps are expanded into the quadruples with the time sequence information, so that the safety dynamic knowledge maps of the community equipment facilities are constructed based on the quadruples, the safety risk level of the community equipment facilities at a future moment is predicted according to the constructed historical moment dynamic knowledge maps and the constructed current moment dynamic knowledge maps, and the accuracy of the community safety risk prediction is improved.
The system provided by the present invention is used for executing the above method embodiments, and for the specific processes and details, reference is made to the above embodiments, which are not described herein again.
Fig. 3 is a schematic structural diagram of an electronic device provided in the present invention, and as shown in fig. 3, the electronic device may include: a Processor (Processor) 301, a communication Interface (Communications Interface) 302, a Memory (Memory) 303 and a communication bus 304, wherein the Processor 301, the communication Interface 302 and the Memory 303 complete communication with each other through the communication bus 304. The processor 301 may invoke logic instructions in the memory 303 to perform a community equipment facility-based security risk prediction method comprising: respectively constructing a historical moment dynamic knowledge graph set and a current moment dynamic knowledge graph of a target community based on equipment safety data of each equipment facility in the target community at different moments; sequentially inputting the historical moment dynamic knowledge maps in the historical moment dynamic knowledge map set into an equipment facility historical knowledge map prediction model according to a time sequence relation to obtain a first equipment facility safety risk prediction result of the target community at the next moment; the device facility historical knowledge map prediction model is obtained by training a neural network through a sample historical time dynamic knowledge map corresponding to each time before a sample time; inputting the current-time dynamic knowledge graph into a device facility current-time knowledge graph prediction model to obtain a second device facility security risk prediction result of the target community at the next time; the device facility current moment knowledge graph prediction model is obtained by training a relation graph convolution network through a sample moment safety dynamic knowledge graph corresponding to a sample moment; and fusing the first equipment facility security risk prediction result and the second equipment facility security risk prediction result to obtain a final equipment facility security risk prediction result of the target community at the next moment.
In addition, the logic instructions in the memory 303 may be implemented in the form of software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer readable storage medium, the computer program including program instructions, when the program instructions are executed by a computer, the computer being capable of executing the security risk prediction method based on community equipment facilities provided by the above methods, the method including: respectively constructing a historical time dynamic knowledge graph set and a current time dynamic knowledge graph of a target community based on equipment safety data of each equipment facility in the target community at different times; sequentially inputting historical time dynamic knowledge maps in the historical time dynamic knowledge map set into an equipment facility historical knowledge map prediction model according to a time sequence relation to obtain a first equipment facility safety risk prediction result of the target community at the next time; the device facility historical knowledge map prediction model is obtained by training a neural network through a sample historical time dynamic knowledge map corresponding to each time before a sample time; inputting the current-time dynamic knowledge graph into a device facility current-time knowledge graph prediction model to obtain a second device facility security risk prediction result of the target community at the next time; the device facility current moment knowledge graph prediction model is obtained by training a relation graph convolution network through a sample moment safety dynamic knowledge graph corresponding to a sample moment; and fusing the first equipment facility security risk prediction result and the second equipment facility security risk prediction result to obtain a final equipment facility security risk prediction result of the target community at the next moment.
In yet another aspect, the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, the computer program being implemented by a processor to perform the community equipment facility-based security risk prediction method provided in the foregoing embodiments, the method including: respectively constructing a historical moment dynamic knowledge graph set and a current moment dynamic knowledge graph of a target community based on equipment safety data of each equipment facility in the target community at different moments; sequentially inputting historical time dynamic knowledge maps in the historical time dynamic knowledge map set into an equipment facility historical knowledge map prediction model according to a time sequence relation to obtain a first equipment facility safety risk prediction result of the target community at the next time; the device facility historical knowledge map prediction model is obtained by training a neural network through a sample historical time dynamic knowledge map corresponding to each time before a sample time; inputting the current-time dynamic knowledge graph into a device facility current-time knowledge graph prediction model to obtain a second device facility security risk prediction result of the target community at the next time; the device facility current moment knowledge graph prediction model is obtained by training a relation graph convolution network through a sample moment safety dynamic knowledge graph corresponding to a sample moment; and fusing the first equipment facility security risk prediction result and the second equipment facility security risk prediction result to obtain a final equipment facility security risk prediction result of the target community at the next moment.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A security risk prediction method based on community equipment facilities is characterized by comprising the following steps:
respectively constructing a historical moment dynamic knowledge graph set and a current moment dynamic knowledge graph of a target community based on equipment safety data of each equipment facility in the target community at different moments;
sequentially inputting the historical moment dynamic knowledge maps in the historical moment dynamic knowledge map set into an equipment facility historical knowledge map prediction model according to a time sequence relation to obtain a first equipment facility safety risk prediction result of the target community at the next moment; the device facility historical knowledge map prediction model is obtained by training a neural network through a sample historical time dynamic knowledge map corresponding to each time before a sample time;
inputting the current-time dynamic knowledge graph into a device facility current-time knowledge graph prediction model to obtain a second device facility security risk prediction result of the target community at the next time; the device facility current moment knowledge graph prediction model is obtained by training a relation graph convolution network through a sample moment safety dynamic knowledge graph corresponding to a sample moment;
and fusing the first equipment facility security risk prediction result and the second equipment facility security risk prediction result to obtain a final equipment facility security risk prediction result of the target community at the next moment.
2. The community equipment facility-based security risk prediction method according to claim 1, wherein the respectively constructing the historical time dynamic knowledge graph set and the current time dynamic knowledge graph of the target community based on the equipment security data of each equipment facility in the target community at different times comprises:
acquiring entity types in the target community and entity relationships among all entities according to the equipment safety data;
according to the entity type and the entity relation, historical time information is used as a timestamp to form a quadruple corresponding to each historical time information, and a historical time dynamic knowledge graph set is constructed based on the quadruple corresponding to each historical time information according to a time sequence relation;
according to the entity type and the entity relation, the current time information is used as a time stamp to form a quadruple corresponding to the current time information, and according to the quadruple corresponding to the current time information, a dynamic knowledge map at the current time is obtained;
and the tail entity in the quadruple corresponding to each historical moment information and the quadruple corresponding to the current moment information is a safety risk result of the equipment and the facility to be predicted.
3. The community equipment facility-based security risk prediction method according to claim 1, wherein the equipment facility historical knowledge map prediction model is constructed by a multi-layer perceptron and a gated loop unit.
4. The community equipment facility-based security risk prediction method of claim 3, wherein the equipment facility historical knowledge graph prediction model is obtained by:
taking any sample time as a reference time, acquiring a sample historical time dynamic knowledge graph corresponding to each historical time of a target community before the reference time, wherein a quadruple corresponding to the sample historical time dynamic knowledge graph is composed of a head entity, an entity relation, a first sample tail entity and timestamp information;
inputting head entities, entity relations and timestamp information in a plurality of sample historical moment dynamic knowledge maps into the multilayer perceptron for training to obtain an aggregation vector output by the multilayer perceptron;
sequentially inputting a plurality of aggregation vectors into the gate control circulation unit for training according to a time sequence relation, and outputting a first sample tail entity obtained through prediction;
and performing loss calculation according to the first sample tail entity and the predicted first sample tail entity, and if the obtained loss value meets a first preset training threshold value, obtaining an equipment facility historical knowledge map prediction model.
5. The community equipment facility-based security risk prediction method of claim 1, wherein the equipment facility current time knowledge graph prediction model is obtained by the following steps:
acquiring a sample time safety dynamic knowledge graph based on sample equipment safety data of a target community at a sample time; the sample time safety dynamic knowledge graph is generated by a quadruple including a head entity, an entity relation, a second tail entity and timestamp information;
inputting head entities, entity relations and timestamp information in the plurality of sample time safety dynamic knowledge maps into a relation map convolution network for training, and outputting a second sample tail entity obtained through prediction;
and performing loss calculation according to the second sample tail entity and the predicted second sample tail entity, and if the obtained loss value meets a second preset training threshold value, obtaining a knowledge graph prediction model of the equipment at the current moment.
6. The community-equipment-facility-based security risk prediction method according to claim 1, wherein the fusing the first equipment facility security risk prediction result and the second equipment facility security risk prediction result to obtain a final equipment facility security risk prediction result of the target community at the next time includes:
adjusting the weights corresponding to the first equipment facility safety risk prediction result and the second equipment facility safety risk prediction result respectively through a particle swarm algorithm;
and summing the first equipment facility security risk prediction result and the second equipment facility security risk prediction result after the weight adjustment, and determining a final equipment facility security risk prediction result of the target community at the next moment.
7. A security risk prediction system based on community equipment facilities, comprising:
the time sequence knowledge graph building module is used for respectively building a historical moment dynamic knowledge graph set and a current moment dynamic knowledge graph of the target community based on the equipment safety data of each equipment facility in the target community at different moments;
the historical knowledge map risk prediction module is used for sequentially inputting historical time dynamic knowledge maps in the historical time dynamic knowledge map set into an equipment facility historical knowledge map prediction model according to a time sequence relation to obtain a first equipment facility safety risk prediction result of the target community at the next time; the device facility historical knowledge map prediction model is obtained by training a neural network through a sample historical time dynamic knowledge map corresponding to each time before a sample time;
the current-time knowledge graph prediction module is used for inputting the current-time dynamic knowledge graph into an equipment facility current-time knowledge graph prediction model to obtain a second equipment facility safety risk prediction result of the target community at the next time; the device facility current moment knowledge graph prediction model is obtained by training a relation graph convolution network through a sample moment safety dynamic knowledge graph corresponding to a sample moment;
and the fusion reasoning module is used for fusing the first equipment facility safety risk prediction result and the second equipment facility safety risk prediction result to obtain a final equipment facility safety risk prediction result of the target community at the next moment.
8. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the computer program implements the steps of the community equipment facility-based security risk prediction method according to any one of claims 1 to 6.
9. A non-transitory computer readable storage medium having a computer program stored thereon, wherein the computer program when executed by a processor implements the steps of the community equipment facility-based security risk prediction method according to any one of claims 1 to 6.
10. A computer program product comprising a computer program, wherein the computer program when executed by a processor implements the steps of the community equipment facility-based security risk prediction method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111630708.1A CN115964503B (en) | 2021-12-28 | 2021-12-28 | Safety risk prediction method and system based on community equipment facilities |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111630708.1A CN115964503B (en) | 2021-12-28 | 2021-12-28 | Safety risk prediction method and system based on community equipment facilities |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115964503A true CN115964503A (en) | 2023-04-14 |
| CN115964503B CN115964503B (en) | 2023-07-07 |
Family
ID=85888409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111630708.1A Active CN115964503B (en) | 2021-12-28 | 2021-12-28 | Safety risk prediction method and system based on community equipment facilities |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115964503B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117371623A (en) * | 2023-12-06 | 2024-01-09 | 佳源科技股份有限公司 | Electric energy meter running state early warning method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112749285A (en) * | 2021-01-21 | 2021-05-04 | 北京明略昭辉科技有限公司 | Resource early warning method, system, equipment and medium based on knowledge graph |
| CN112800237A (en) * | 2021-01-19 | 2021-05-14 | 中国再保险(集团)股份有限公司 | Prediction method and device based on knowledge graph embedded representation and computer equipment |
| WO2021103492A1 (en) * | 2019-11-28 | 2021-06-03 | 福建亿榕信息技术有限公司 | Risk prediction method and system for business operations |
| CN113642826A (en) * | 2021-06-02 | 2021-11-12 | 中国海洋大学 | Supplier default risk prediction method |
| CN113822494A (en) * | 2021-10-19 | 2021-12-21 | 平安科技（深圳）有限公司 | Risk prediction method, device, equipment and storage medium |
-
2021
- 2021-12-28 CN CN202111630708.1A patent/CN115964503B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021103492A1 (en) * | 2019-11-28 | 2021-06-03 | 福建亿榕信息技术有限公司 | Risk prediction method and system for business operations |
| CN112800237A (en) * | 2021-01-19 | 2021-05-14 | 中国再保险(集团)股份有限公司 | Prediction method and device based on knowledge graph embedded representation and computer equipment |
| CN112749285A (en) * | 2021-01-21 | 2021-05-04 | 北京明略昭辉科技有限公司 | Resource early warning method, system, equipment and medium based on knowledge graph |
| CN113642826A (en) * | 2021-06-02 | 2021-11-12 | 中国海洋大学 | Supplier default risk prediction method |
| CN113822494A (en) * | 2021-10-19 | 2021-12-21 | 平安科技（深圳）有限公司 | Risk prediction method, device, equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 姚春晓: "基于长短时记忆神经网络的脑血管疾病预测系统研究", 中国优秀硕士学位论文全文数据库 (医药卫生科技辑), vol. 2020, no. 1, pages 070 - 159 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117371623A (en) * | 2023-12-06 | 2024-01-09 | 佳源科技股份有限公司 | Electric energy meter running state early warning method and system |
| CN117371623B (en) * | 2023-12-06 | 2024-03-01 | 佳源科技股份有限公司 | Electric energy meter running state early warning method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115964503B (en) | 2023-07-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yang et al. | A network traffic forecasting method based on SA optimized ARIMA–BP neural network | |
| CN106453293B (en) | A kind of network security situation prediction method based on improved BPNN | |
| CN112966714B (en) | Edge time sequence data anomaly detection and network programmable control method | |
| Pranolo et al. | Robust LSTM With tuned-PSO and bifold-attention mechanism for analyzing multivariate time-series | |
| CN111079860B (en) | Method for constructing multi-granularity response model of emergency based on scene similarity | |
| Biswas et al. | Weather forecast prediction: an integrated approach for analyzing and measuring weather data | |
| CN115629160A (en) | Air pollutant concentration prediction method and system based on space-time diagram | |
| CN114023399A (en) | Air particulate matter analysis early warning method and device based on artificial intelligence | |
| CN115964503B (en) | Safety risk prediction method and system based on community equipment facilities | |
| Zhang et al. | Automatic traffic anomaly detection on the road network with spatial-temporal graph neural network representation learning | |
| Li et al. | A LSTM-based method for comprehension and evaluation of network security situation | |
| Cao et al. | Fast and explainable warm-start point learning for AC Optimal Power Flow using decision tree | |
| CN115964504B (en) | Food safety risk prediction method and system | |
| Xu et al. | Fire Safety Assessment of High-Rise Buildings Based on Fuzzy Theory and Radial Basis Function Neural Network. | |
| CN115796017A (en) | Interpretable traffic cognition method based on fuzzy theory | |
| Bhatt et al. | Automated forest fire prediction systems: a comprehensive review | |
| Li et al. | Information granularity with the self-emergence mechanism for event detection in WSN-based tunnel health monitoring | |
| CN113884807A (en) | Power distribution network fault prediction method based on random forest and multi-layer architecture clustering | |
| Elsarrar et al. | Analysis of forest fire data using neural network rule extraction with human understandable rules | |
| Gorrostieta-Hurtado et al. | Modeling key parameters for greenhouse using fuzzy clustering techniques | |
| Cao et al. | Field detection of indoor fire threat situation based on LSTM-Kriging network | |
| Aftabi et al. | A variational autoencoder framework for robust, physics-informed cyberattack recognition in industrial cyber-physical systems | |
| Hurtado et al. | Modeling Key Parameters for Greenhouse Using Fuzzy Clustering Techniques. | |
| Sheng et al. | Network traffic anomaly detection method based on chaotic neural network | |
| CN115455408B (en) | Network space deduction and security assessment method and device |
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 |