CN114155390B - Clustering device for classifying metadata of classified objects - Google Patents
Clustering device for classifying metadata of classified objects Download PDFInfo
- Publication number
- CN114155390B CN114155390B CN202111463988.1A CN202111463988A CN114155390B CN 114155390 B CN114155390 B CN 114155390B CN 202111463988 A CN202111463988 A CN 202111463988A CN 114155390 B CN114155390 B CN 114155390B
- Authority
- CN
- China
- Prior art keywords
- attack
- neural network
- basis function
- radial basis
- function neural
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Abstract
The invention relates to clustering equipment for classifying metadata of classified objects, and belongs to the technical field of artificial intelligence. The apparatus comprises: the network construction component is used for establishing a radial basis function neural network with a multilayer structure, and the number of hidden layers of the network is in direct proportion to the total number of the types of the attack objects to be analyzed; the intelligent classification component is used for classifying the aggressivity of the to-be-analyzed type of attack object according to the output data of the output layer of the radial basis function neural network after multiple times of training; and the target clustering component is used for clustering various kinds of attacks in the attack group based on the classification result. The invention also relates to a clustering method for classifying the metadata of the classified objects. According to the invention, the metadata related to the aggressivity of each kind of the attack object in the attack object group can be analyzed by adopting the radial basis function neural network with the customized structure, and the clustering operation is carried out on various attack objects in the attack object group based on the analysis result, so that the aggressivity-based classification of the attack objects is completed.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to clustering equipment for classifying metadata of classified objects.
Background
Metadata (Metadata), also called intermediary data and relay data, is data (data about data) describing data, and is mainly information describing data attribute (property) for supporting functions such as indicating storage location, history data, resource search, file record, and the like. Metadata is an electronic catalog, and in order to achieve the purpose of creating a catalog, the contents or features of data must be described and collected, so as to achieve the purpose of assisting data retrieval. Dublin Core Metadata Initiative (DCMI) is an application of Metadata, and is a workshop sponsored by International library computer Center (OCLC) and National Center for Supercomputing Applications (NCSA) in 2 months 1995, and 52 requests from librarians and computer experts are created to jointly specify a set of features describing electronic files on the network.
Metadata is information about the organization, data domain and its relationship of data, i.e., metadata is data about data, and clustering based on metadata is to divide objects having the same metadata into the same category for subsequent execution of object management. The method is particularly applied to the subdivided application field, for example, various types of attacks in the future attack group are used as various classified objects, whether the attacks exist or not is used as first metadata of the classified objects, the time when the attacks occur most frequently is used as second metadata of the classified objects, and the classified objects, namely various types of attacks, of the attack group are classified, so that subsequent defense and treatment are performed by an attack defender conveniently.
However, although the habit of a certain type of action is basically fixed, there is a certain difference in the habit particularly for the attack objects allocated to the attack object group, and even if it can be determined that a certain kind of attack object has an offensive nature, it is impossible to determine the time when the attack of the kind of attack object is habitually generated every day, and the absence of these metadata causes the attack object defender to be difficult to actually perform the clustering operation of the attack objects.
Disclosure of Invention
In order to solve the above problems, the present invention provides a clustering device for classifying metadata of classified objects, which can perform intelligent analysis on metadata related to aggressivity of each kind of an incoming matter group by using a radial basis function neural network with a customized structure, perform clustering operation based on the metadata on various types of the incoming matters of the incoming matter group based on an analysis result, and complete aggressivity-based classification on the incoming matters, thereby facilitating a defense party of the incoming matters to subsequently formulate a safe and effective campus management strategy.
Therefore, the present invention needs to have at least the following three important points:
(1) establishing a radial basis function neural network with a multilayer structure, wherein the multilayer structure comprises a single input layer, a single output layer and a set number of hidden layers, input data of the input layer of the radial basis function neural network are a plurality of digital images captured in a time-sharing manner in an activity area of a certain kind of incoming object in an incoming object group, and output data of the output layer of the radial basis function neural network are the time when whether the kind of incoming object has aggressivity and the probability of attack action occurring every day is the highest when the kind of incoming object has aggressivity;
(2) in the built radial basis function neural network with the multilayer structure, the number of hidden layers is in direct proportion to the total number of types of the attack objects to be analyzed, so that the radial basis function neural network can be matched with various attack object monitoring areas with different types and complexity;
(3) and classifying the aggressivity of the to-be-analyzed type of the incoming objects according to the output data of the output layer of the radial basis function neural network after multiple times of training, so that whether the to-be-analyzed type of the incoming objects has the aggressivity or not and the moment with the highest probability of attack action occurring every day when the to-be-analyzed type of the incoming objects has the aggressivity are used as two items of metadata to realize intelligent clustering operation on the various types of the incoming objects of the incoming object group.
According to a first aspect of the present invention, there is provided a clustering apparatus that classifies metadata of a classification object, the apparatus including:
the network construction component is used for establishing a radial basis function neural network with a multilayer structure, the multilayer structure is respectively a single input layer, a single output layer and a set number of hidden layers, input data of the input layer of the radial basis function neural network are a plurality of digital images captured in a time-sharing mode in an activity area of a certain kind of incoming material of an incoming material group, output data of the output layer of the radial basis function neural network are the time when whether the kind of incoming material has aggressivity and the probability of attack action occurring every day when the kind of incoming material has the aggressivity is the highest, and the numerical value of the set number is in direct proportion to the total number of kinds of the incoming materials to be analyzed;
the network training component is connected with the network construction component and used for adopting a moment when a certain kind of attack object is known to have aggressivity or not and is known to have the highest probability of attack action occurring every day when the certain kind of attack object has the aggressivity as output data of an output layer of the radial basis function neural network established by the network construction component, and adopting a plurality of digital images captured by the kind of attack object in a time division mode in an activity area of the certain kind of attack object as input data of an input layer of the radial basis function neural network established by the network construction component to train the radial basis function neural network established by the network construction component;
the intelligent classification component is used for adopting a plurality of digital images captured in a time-sharing manner in an active region of a certain kind of incoming material to be analyzed of an incoming material group as input data of an input layer of the radial basis function neural network after multiple times of training is executed by the network training component so as to operate the radial basis function neural network after multiple times of training, and classifying the aggressivity of the kind of incoming material to be analyzed according to the output data of the output layer of the radial basis function neural network after multiple times of training;
the target clustering component is connected with the intelligent analysis component and is used for clustering various types of the incoming objects of the incoming object group based on the classification result of the aggressivity of the intelligent classification component to the various types of the incoming objects;
the input data of the input layer of the radial basis function neural network is a plurality of digital images captured in a time-sharing mode in an active region of a certain kind of incoming object in an incoming object group, and the digital images comprise: the capturing visual fields of the plurality of digital images are the same and have the same resolution, and a plurality of capturing moments corresponding to the plurality of digital images are uniformly distributed in the time length of one day;
wherein, the output data of the output layer of the radial basis function neural network is whether the kind of attack is aggressive or not and the time when the probability of attack action is the highest every day when the kind of attack is aggressive comprises: the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments;
wherein, the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments, and the method comprises the following steps: when the attack identification data is equal to 1, the type of attack has aggressivity, and the frequent attack moment is a binary code corresponding to the moment with the highest probability of attack action occurring every day;
the clustering operation of various kinds of incoming materials of the incoming material group based on the classification result of the aggressivity of the various kinds of incoming materials of the incoming material group by the intelligent classification component comprises the following steps: and classifying a plurality of kinds of attack objects with the same attack identification data in the classification result into the same attack object class set.
According to a second aspect of the present invention, there is provided a clustering device for classifying metadata of a classification object, the device comprising a memory and one or more processors, the memory storing a computer program configured to be executed by the one or more processors to perform the steps of:
establishing a radial basis function neural network of a multilayer structure, wherein the multilayer structure comprises a single input layer, a single output layer and a set number of hidden layers respectively, input data of the input layer of the radial basis function neural network are a plurality of digital images captured in a time-sharing manner in an activity area of a certain kind of incoming objects in an incoming object group, output data of the output layer of the radial basis function neural network are moments of whether the kind of incoming objects is aggressive and have the highest probability of attack action occurring every day when the kind of incoming objects is aggressive, and the number of the set number is in direct proportion to the total number of the kinds of incoming objects to be analyzed;
adopting a moment when a certain kind of attack object is known to have aggressivity or not and is known to have the highest probability of attack action occurring every day when the certain kind of attack object has aggressivity as output data of an output layer of the established radial basis function neural network, and adopting a plurality of digital images captured by the certain kind of attack object in a time division mode in an activity area of the certain kind of attack object as input data of an input layer of the established radial basis function neural network to train the established radial basis function neural network;
the method comprises the steps that a plurality of digital images captured in a time-sharing mode in a certain kind of attack object activity area to be analyzed of an attack object group are used as input data of an input layer of a radial basis function neural network after multiple times of training to operate the radial basis function neural network after multiple times of training, and the aggressivity of the kind of attack object to be analyzed is classified according to output data of the output layer of the radial basis function neural network after multiple times of training;
clustering each kind of attack object of the attack object group based on the classification result of the aggressivity of each kind of attack object of the attack object group;
the input data of the input layer of the radial basis function neural network is a plurality of digital images captured in a time-sharing mode in an active region of a certain kind of incoming object in an incoming object group, and the digital images comprise: the capturing visual fields of the digital images are the same and the resolution is equal, and a plurality of capturing moments respectively corresponding to the digital images are uniformly distributed in the time length of one day;
wherein, the output data of the output layer of the radial basis function neural network is whether the kind of attack is aggressive or not and the time when the probability of attack action is the highest every day when the kind of attack is aggressive comprises: the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments;
wherein, the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments, and the method comprises the following steps: when the attack identification data is equal to 1, the type of attack object has aggressivity, and the frequent attack moment is a binary code corresponding to the moment with the highest probability of attack action occurring every day;
the clustering operation of various kinds of incoming materials of the incoming material group based on the classification result of the aggressivity of the various kinds of incoming materials of the incoming material group by the intelligent classification component comprises the following steps: and classifying a plurality of kinds of attack objects with the same attack identification data in the classification result into the same attack object class set.
According to a third aspect of the present invention, there is provided a clustering method of classifying metadata of a classification object, the method comprising:
establishing a radial basis function neural network of a multilayer structure, wherein the multilayer structure comprises a single input layer, a single output layer and a set number of hidden layers respectively, input data of the input layer of the radial basis function neural network are a plurality of digital images captured in a time-sharing manner in an activity area of a certain kind of incoming objects in an incoming object group, output data of the output layer of the radial basis function neural network are moments of whether the kind of incoming objects is aggressive and have the highest probability of attack action occurring every day when the kind of incoming objects is aggressive, and the number of the set number is in direct proportion to the total number of the kinds of incoming objects to be analyzed;
adopting a moment when a certain kind of attack object is known to have aggressivity or not and is known to have the highest probability of attack action occurring every day when the certain kind of attack object has aggressivity as output data of an output layer of the established radial basis function neural network, and adopting a plurality of digital images captured by the certain kind of attack object in a time division mode in an activity area of the certain kind of attack object as input data of an input layer of the established radial basis function neural network to train the established radial basis function neural network;
the method comprises the steps that a plurality of digital images captured in a time-sharing mode in a certain kind of attack object activity area to be analyzed of an attack object group are used as input data of an input layer of a radial basis function neural network after multiple times of training to operate the radial basis function neural network after multiple times of training, and the aggressivity of the kind of attack object to be analyzed is classified according to output data of the output layer of the radial basis function neural network after multiple times of training;
clustering each kind of attack object of the attack object group based on the classification result of the aggressivity of each kind of attack object of the attack object group;
the input data of the input layer of the radial basis function neural network is a plurality of digital images captured in a time-sharing mode in an active region of a certain kind of incoming object in an incoming object group, and the digital images comprise: the capturing visual fields of the digital images are the same and the resolution is equal, and a plurality of capturing moments respectively corresponding to the digital images are uniformly distributed in the time length of one day;
wherein, the output data of the output layer of the radial basis function neural network is whether the kind of attack is aggressive or not and the time when the probability of attack action is the highest every day when the kind of attack is aggressive comprises: the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments;
wherein, the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments, and the method comprises the following steps: when the attack identification data is equal to 1, the type of attack object has aggressivity, and the frequent attack moment is a binary code corresponding to the moment with the highest probability of attack action occurring every day;
the clustering operation of various kinds of incoming materials of the incoming material group based on the classification result of the aggressivity of the various kinds of incoming materials of the incoming material group by the intelligent classification component comprises the following steps: and classifying a plurality of types of attack matters with the same attack identification data in the classification result into the same attack species set.
Drawings
Embodiments of the invention will now be described with reference to the accompanying drawings, in which:
fig. 1 is a technical flowchart of a clustering apparatus that classifies metadata of classified objects according to the present invention.
Fig. 2 is a schematic structural diagram of a clustering apparatus that classifies metadata of classification objects according to embodiment 1 of the present invention.
Fig. 3 is a schematic structural diagram of a clustering device that classifies metadata of classification objects according to embodiment 2 of the present invention.
Fig. 4 is a schematic structural diagram of a clustering device that classifies metadata of classification objects according to embodiment 3 of the present invention.
Fig. 5 is a schematic structural diagram of a clustering device that classifies metadata of classification objects according to embodiment 4 of the present invention.
Fig. 6 is a schematic structural diagram of a clustering device that classifies metadata of classification objects according to embodiment 5 of the present invention.
Fig. 7 is a flowchart illustrating steps of a clustering method for classifying metadata of classification objects according to embodiment 6 of the present invention.
Detailed Description
The process of dividing a collection of physical or abstract objects into classes composed of similar objects is called clustering. The cluster generated by clustering is a collection of a set of data objects that are similar to objects in the same cluster and distinct from objects in other clusters. There are a number of classification problems in the natural and social sciences. Clustering analysis, also known as cluster analysis, is a statistical analysis method for studying (sample or index) classification problems. The clustering analysis originates from taxonomy, but clustering is not equal to classification. Clustering differs from classification in that the class into which the clustering is required to be divided is unknown. The clustering analysis content is very rich, and a system clustering method, an ordered sample clustering method, a dynamic clustering method, a fuzzy clustering method, a graph theory clustering method, a clustering forecasting method and the like are adopted.
Currently, the clustering operation of the classified objects by using the metadata of the classified objects is a clustering operation mode which uses a lot of metadata, however, in practical application, the accurate and reliable acquisition of the metadata is a difficult problem. For example, for monitoring areas of incoming things (e.g., various weapons), especially for monitoring areas of heaven, sea, and side incoming things, the defending party needs to clarify the aggressivity of different kinds of incoming things defended by the defending party, including two items of metadata, i.e., whether the two items are offensive and the time when the two items are used to perform an attack action, but even under the same kind, the aggressivity of different incoming things is very different, so that each monitoring area of incoming things cannot complete clustering of the aggressivity of each kind defended by the defending party, and the subsequent defense strategy is lack of pertinence and effectiveness.
In order to overcome the defects, the invention builds a clustering device for classifying the metadata of the classified objects, establishes a radial basis function neural network of a customized structure suitable for the complexity of the belongings in the jurisdiction of each belongings monitoring area, and realizes the intelligent analysis of the offensive metadata of different kinds of belongings defended by the clustering device on the basis of a targeted training mechanism, thereby ensuring the pertinence and the effectiveness of a subsequently formulated defense strategy.
As shown in fig. 1, a technical flowchart of a clustering device for classifying metadata of a classification object according to the present invention is shown.
As shown in fig. 1, the specific technical process of the present invention is as follows:
firstly, aiming at the garden of a certain monitoring area of the incoming objects, respectively collecting different digital images of the active area of each kind of the incoming objects at different times of the day, these digital images are used for the subsequent execution of the basic data of the radial basis function neural network for the offensive identification of an attack, when the aggressiveness of a respective kind of attack is known, the associated digital image can be used as training data for the radial basis function neural network, when the aggressiveness of the corresponding kind of an attack is unknown, the related digital image can be used as usage data of the radial basis function neural network to perform intelligent identification of the aggressiveness of the corresponding kind of an attack, the metadata related to the aggressivity of the attack object has two types, one is whether the kind of attack object has the aggressivity, and the other is the time when the probability of the attack action is the highest every day when the attack object has the aggressivity;
secondly, aiming at the monitoring area of the incoming object, a radial basis function neural network of a multilayer structure is established, the multilayer structure is respectively a single input layer, a single output layer and a set number of hidden layers, the input data of the input layer of the radial basis function neural network are a plurality of digital images captured in a time-sharing mode in the activity area of a certain kind of incoming object in the incoming object group, the output data of the output layer of the radial basis function neural network is the moment whether the kind of incoming object has aggressivity or not and the highest probability of attack action occurring every day when the kind of incoming object has the aggressivity, wherein the number of the hidden layers is in direct proportion to the total number of the kinds of incoming objects to be analyzed, and therefore the established radial basis function neural network can be matched with the monitoring areas of the incoming objects with different kinds and complexity;
thirdly, performing multiple training of known aggressive attack species on the established radial basis function neural network to ensure the accuracy of the radial basis function neural network for subsequent metadata identification;
and finally, performing analysis of offensive metadata on the unknown offensive type of the attack monitoring area by using the radial basis function neural network after multiple times of training, and performing clustering operation based on the metadata on various types of the offensive of the attack monitoring area defense based on analysis results.
The key point of the invention is that whether the attack is aggressive or not and the moment when the attack action probability is highest every day when the attack is aggressive are selected as the reference data of the attack clustering operation required by developing the defense strategy for the attack monitoring area, particularly, the invention selects the radial basis function neural network with a multilayer structure as the artificial identification model, and in order to meet the requirements of various attack monitoring areas with different types and complexity, the number of the built hidden layers of the radial basis function neural network is in direct proportion to the total number of the types of the attacks to be analyzed, and simultaneously, a targeted training mechanism is adopted, thereby realizing the accurate analysis of two different metadata related to the aggressivity.
Hereinafter, a clustering apparatus that classifies metadata of a classification object of the present invention will be specifically described by way of example.
Example 1
Fig. 2 is a schematic structural diagram of a clustering apparatus that classifies metadata of classification objects according to embodiment 1 of the present invention.
As shown in fig. 2, the clustering apparatus that classifies metadata of a classification object includes the following components:
the network construction component is used for establishing a radial basis function neural network with a multilayer structure, the multilayer structure is respectively a single input layer, a single output layer and a set number of hidden layers, input data of the input layer of the radial basis function neural network are a plurality of digital images captured in a time-sharing mode in an activity area of a certain kind of incoming material of an incoming material group, output data of the output layer of the radial basis function neural network are the time when whether the kind of incoming material has aggressivity and the probability of attack action occurring every day when the kind of incoming material has the aggressivity is the highest, and the numerical value of the set number is in direct proportion to the total number of kinds of the incoming materials to be analyzed;
the network training component is connected with the network construction component and used for adopting a moment when a certain type of attack object is known to have aggressivity or not and the highest probability of attack action occurring every day when the certain type of attack object has the aggressivity as output data of an output layer of the radial basis function neural network established by the network construction component, and adopting a plurality of digital images captured by the type of attack object in a time division manner in an activity area of the certain type of attack object as input data of an input layer of the radial basis function neural network established by the network construction component to train the radial basis function neural network established by the network construction component;
the intelligent classification component is used for adopting a plurality of digital images captured in a time-sharing manner in an active region of a certain kind of incoming material to be analyzed of an incoming material group as input data of an input layer of the radial basis function neural network after multiple times of training is executed by the network training component so as to operate the radial basis function neural network after multiple times of training, and classifying the aggressivity of the kind of incoming material to be analyzed according to the output data of the output layer of the radial basis function neural network after multiple times of training;
the target clustering component is connected with the intelligent analysis component and is used for clustering various types of the incoming objects of the incoming object group based on the classification result of the aggressivity of the intelligent classification component to the various types of the incoming objects;
the input data of the input layer of the radial basis function neural network is a plurality of digital images captured in a time-sharing mode in an active region of a certain kind of incoming object in an incoming object group, and the digital images comprise: the capturing visual fields of the digital images are the same and the resolution is equal, and a plurality of capturing moments respectively corresponding to the digital images are uniformly distributed in the time length of one day;
wherein, the output data of the output layer of the radial basis function neural network is whether the kind of attack is aggressive or not and the time when the probability of attack action is the highest every day when the kind of attack is aggressive comprises: the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments;
wherein, the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments, and the method comprises the following steps: when the attack identification data is equal to 1, the type of attack object has aggressivity, and the frequent attack moment is a binary code corresponding to the moment with the highest probability of attack action occurring every day;
the clustering operation of various kinds of incoming materials of the incoming material group based on the classification result of the aggressivity of the various kinds of incoming materials of the incoming material group by the intelligent classification component comprises the following steps: and classifying a plurality of kinds of attack objects with the same attack identification data in the classification result into the same attack object class set.
Example 2
Fig. 3 is a schematic structural diagram of a clustering device that classifies metadata of classification objects according to embodiment 2 of the present invention.
As shown in fig. 3, unlike embodiment 1 of the present invention, the clustering device that classifies metadata of a classification object further includes:
and the big data storage component is connected with the target clustering component through a wireless network and is used for storing clustering operation results of the target clustering component on various kinds of attacks in the attack group.
Example 3
Fig. 4 is a schematic structural diagram of a clustering device that classifies metadata of classification objects according to embodiment 3 of the present invention.
As shown in fig. 4, unlike embodiment 1 of the present invention, the clustering device that classifies metadata of a classification object further includes:
and the parameter configuration component is respectively connected with the network construction component, the network training component and the intelligent classification component and is used for respectively realizing the field configuration of the working parameters of the network construction component, the network training component and the intelligent classification component.
Example 4
Fig. 5 is a schematic structural diagram of a clustering device that classifies metadata of classification objects according to embodiment 4 of the present invention.
As shown in fig. 5, unlike embodiment 1 of the present invention, the clustering device that classifies metadata of a classification object further includes:
and the synchronous coordination component is respectively connected with the network construction component, the network training component and the intelligent classification component and is used for realizing the cooperative control of the respective actions of the network construction component, the network training component and the intelligent classification component.
In any of the foregoing embodiments, optionally, in the clustering device that classifies metadata of classified objects:
the clustering operation of various kinds of attacks of the attack object group based on the classification result of the aggressivity of various kinds of attacks of the attack object group by the intelligent classification component comprises the following steps: classifying a plurality of types of attack objects with the same frequent attack time in the classification result into the same attack species set;
wherein, the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments, and the method comprises the following steps: when the attack identification data is equal to 0, the attack-like object has no aggressivity, and the frequent attack moment is binary codes corresponding to null symbols;
the classification of multiple kinds of attack objects with the same attack identification data in the classification result into the same attack species set comprises the following steps: classifying a plurality of kinds of attack objects with attack identification data equal to 1 in the classification result into an attack type attack species set, and classifying a plurality of kinds of attack objects with attack identification data equal to 0 in the classification result into a mild type attack species set;
the classification of the multiple kinds of attack objects with the same frequent attack time in the classification result into the same attack object class set comprises the following steps: classifying the multiple kinds of attacks with the same frequent attack time in the classification result into a kind set of the same kind of attack species set, and classifying the multiple kinds of attacks with attack identification data equal to 1 in the classification result into a subset of the kind set of the same kind of attack species set.
In any of the foregoing embodiments, optionally, in the clustering device that classifies metadata of classified objects:
the radial basis function neural network uses a radial basis function as an activation function and is composed of a multilayer structure;
wherein, in the multilayer structure of the radial basis function neural network, the numerical transformation from the input layer to the hidden layer is nonlinear, and the numerical transformation from the hidden layer to the output layer is linear.
Example 5
Fig. 6 is a block diagram showing a configuration of a clustering device that classifies metadata of classification objects according to embodiment 5 of the present invention.
As shown in fig. 6, the clustering device for classifying metadata of a classification object includes a memory and N processors, N being a positive integer greater than or equal to 1, the memory storing a computer program configured to be executed by the N processors to perform the steps of:
establishing a radial basis function neural network of a multilayer structure, wherein the multilayer structure comprises a single input layer, a single output layer and a set number of hidden layers respectively, input data of the input layer of the radial basis function neural network are a plurality of digital images captured in a time-sharing manner in an activity area of a certain kind of incoming objects in an incoming object group, output data of the output layer of the radial basis function neural network are moments of whether the kind of incoming objects is aggressive and have the highest probability of attack action occurring every day when the kind of incoming objects is aggressive, and the number of the set number is in direct proportion to the total number of the kinds of incoming objects to be analyzed;
adopting a moment when a certain kind of attack object is known to have aggressivity or not and the highest probability of attack action occurring every day when the certain kind of attack object has the aggressivity as output data of an output layer of the established radial basis function neural network, and adopting a plurality of digital images captured by the certain kind of attack object in a time-sharing manner in an active region of the certain kind of attack object as input data of an input layer of the established radial basis function neural network to train the established radial basis function neural network;
the method comprises the steps that a plurality of digital images captured in a time-sharing mode in a certain type of attack object moving area to be analyzed of an attack object group are used as input data of an input layer of a radial basis function neural network after multiple times of training is carried out to operate the radial basis function neural network after the multiple times of training, and the aggressivity of the type of attack object to be analyzed is classified according to output data of an output layer of the radial basis function neural network after the multiple times of training;
clustering each kind of attack object of the attack object group based on the classification result of the aggressivity of each kind of attack object of the attack object group;
the input data of the input layer of the radial basis function neural network is a plurality of digital images captured in a time-sharing mode in an active region of a certain kind of incoming object in an incoming object group, and the digital images comprise: the capturing visual fields of the digital images are the same and the resolution is equal, and a plurality of capturing moments respectively corresponding to the digital images are uniformly distributed in the time length of one day;
wherein, the output data of the output layer of the radial basis function neural network is whether the kind of attack is aggressive or not and the time when the probability of attack action is the highest every day when the kind of attack is aggressive comprises: the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments;
wherein, the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments, and the method comprises the following steps: when the attack identification data is equal to 1, the type of attack object has aggressivity, and the frequent attack moment is a binary code corresponding to the moment with the highest probability of attack action occurring every day;
the clustering operation of various kinds of incoming materials of the incoming material group based on the classification result of the aggressivity of the various kinds of incoming materials of the incoming material group by the intelligent classification component comprises the following steps: and classifying a plurality of kinds of attack objects with the same attack identification data in the classification result into the same attack object class set.
Example 6
Fig. 7 is a flowchart illustrating steps of a clustering method for classifying metadata of classification objects according to embodiment 6 of the present invention.
As shown in fig. 7, the clustering method for classifying the metadata of the classification object includes:
establishing a radial basis function neural network of a multilayer structure, wherein the multilayer structure comprises a single input layer, a single output layer and a set number of hidden layers respectively, input data of the input layer of the radial basis function neural network are a plurality of digital images captured in a time-sharing manner in an activity area of a certain kind of incoming objects in an incoming object group, output data of the output layer of the radial basis function neural network are moments of whether the kind of incoming objects is aggressive and have the highest probability of attack action occurring every day when the kind of incoming objects is aggressive, and the number of the set number is in direct proportion to the total number of the kinds of incoming objects to be analyzed;
adopting a moment when a certain kind of attack object is known to have aggressivity or not and is known to have the highest probability of attack action occurring every day when the certain kind of attack object has aggressivity as output data of an output layer of the established radial basis function neural network, and adopting a plurality of digital images captured by the certain kind of attack object in a time division mode in an activity area of the certain kind of attack object as input data of an input layer of the established radial basis function neural network to train the established radial basis function neural network;
the method comprises the steps that a plurality of digital images captured in a time-sharing mode in a certain kind of attack object activity area to be analyzed of an attack object group are used as input data of an input layer of a radial basis function neural network after multiple times of training to operate the radial basis function neural network after multiple times of training, and the aggressivity of the kind of attack object to be analyzed is classified according to output data of the output layer of the radial basis function neural network after multiple times of training;
clustering operation is carried out on various types of the incoming materials of the incoming material group based on the classification result of the aggressiveness of the various types of the incoming materials of the incoming material group;
the input data of the input layer of the radial basis function neural network is a plurality of digital images captured in a time-sharing mode in an active region of a certain kind of incoming object in an incoming object group, and the digital images comprise: the capturing visual fields of the digital images are the same and the resolution is equal, and a plurality of capturing moments respectively corresponding to the digital images are uniformly distributed in the time length of one day;
wherein, the output data of the output layer of the radial basis function neural network is whether the kind of attack is aggressive or not and the time when the probability of attack action is the highest every day when the kind of attack is aggressive comprises: the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments;
wherein, the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments, and the method comprises the following steps: when the attack identification data is equal to 1, the type of attack object has aggressivity, and the frequent attack moment is a binary code corresponding to the moment with the highest probability of attack action occurring every day;
the clustering operation of various kinds of incoming materials of the incoming material group based on the classification result of the aggressivity of the various kinds of incoming materials of the incoming material group by the intelligent classification component comprises the following steps: and classifying a plurality of kinds of attack objects with the same attack identification data in the classification result into the same attack object class set.
In addition, Metadata (Metadata) is data (data about other data) describing other data, or structural data (structured data) for providing information on a certain resource. Metadata is data that describes an object such as an information resource or data, and is used for the purpose of: identifying a resource; evaluating the resources; tracking changes in the use of the resource; the method realizes simple and efficient management of a large amount of networked data; the information resources are effectively discovered, searched and integrally organized, and the used resources are effectively managed.
Generally, the basic features of metadata are:
a) once the metadata is established, it can be shared. The structure and integrity of the metadata depend on the value and use environment of the information resources; the development and utilization environment of metadata is often a varied distributed environment; either format cannot fully meet the different needs of different groups;
b) metadata is first a coding scheme. Metadata is an encoding system used to describe digital information resources, especially network information resources, which results in fundamental differences between metadata and conventional data encoding systems; the most important feature and function of metadata is to build a machine understandable framework for digitized information resources.
The metadata system constructs a logical framework and a basic model of the e-government affairs, so that the functional characteristics, the operation mode and the overall performance of the system operation of the e-government affairs are determined. The operation of the e-government affairs is implemented based on the metadata. It has the main functions as follows: description functions, integration functions, control functions, and agent functions.
Since metadata is also data, it can be stored and retrieved in a database in a data-like manner. The use of data elements will be made accurate and efficient if the organization of the data elements is provided while the metadata describing the data elements is provided. The user may first view his metadata when using the data in order to be able to obtain the information he wants.
In the foregoing specification, the invention has been described with reference to specific embodiments. However, one of ordinary skill in the art appreciates that various modifications and changes can be made without departing from the scope of the present invention as set forth in the claims below. Accordingly, the specification and figures are to be regarded in an illustrative rather than a restrictive sense, and all such modifications are intended to be included within the scope of present invention.
Benefits, other advantages, and solutions to problems have been described above with regard to specific embodiments. However, the benefits, advantages, solutions to problems, and any element(s) that may cause any benefit, advantage, or solution to occur or become more pronounced are not to be construed as a critical, required, or essential feature or element of any or all the claims. As used herein, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Claims (10)
1. A clustering device that classifies metadata of a classification object, the device comprising:
the network construction component is used for establishing a radial basis function neural network with a multilayer structure, the multilayer structure is respectively a single input layer, a single output layer and a set number of hidden layers, input data of the input layer of the radial basis function neural network are a plurality of digital images captured in a time-sharing mode in an activity area of a certain kind of incoming material of an incoming material group, output data of the output layer of the radial basis function neural network are the time when whether the kind of incoming material has aggressivity and the probability of attack action occurring every day when the kind of incoming material has the aggressivity is the highest, and the numerical value of the set number is in direct proportion to the total number of kinds of the incoming materials to be analyzed;
the network training component is connected with the network construction component and used for adopting a moment when a certain kind of attack object is known to have aggressivity or not and is known to have the highest probability of attack action occurring every day when the certain kind of attack object has the aggressivity as output data of an output layer of the radial basis function neural network established by the network construction component, and adopting a plurality of digital images captured by the kind of attack object in a time division mode in an activity area of the certain kind of attack object as input data of an input layer of the radial basis function neural network established by the network construction component to train the radial basis function neural network established by the network construction component;
the intelligent classification component is used for adopting a plurality of digital images captured in a time-sharing manner in an active region of a certain kind of incoming material to be analyzed of an incoming material group as input data of an input layer of the radial basis function neural network after multiple times of training is executed by the network training component so as to operate the radial basis function neural network after multiple times of training, and classifying the aggressivity of the kind of incoming material to be analyzed according to the output data of the output layer of the radial basis function neural network after multiple times of training;
the target clustering component is connected with the intelligent analysis component and is used for clustering various types of the incoming objects of the incoming object group based on the classification result of the aggressivity of the intelligent classification component to the various types of the incoming objects;
the input data of the input layer of the radial basis function neural network is a plurality of digital images captured in a time-sharing mode in an active region of a certain kind of incoming object in an incoming object group, and the digital images comprise: the capturing visual fields of the digital images are the same and the resolution is equal, and a plurality of capturing moments respectively corresponding to the digital images are uniformly distributed in the time length of one day;
wherein, the output data of the output layer of the radial basis function neural network is whether the kind of attack is aggressive or not and the time when the probability of attack action is the highest every day when the kind of attack is aggressive comprises: the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments;
wherein, the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments, and the method comprises the following steps: when the attack identification data is equal to 1, the type of attack object has aggressivity, and the frequent attack moment is a binary code corresponding to the moment with the highest probability of attack action occurring every day;
the clustering operation of various kinds of incoming materials of the incoming material group based on the classification result of the aggressivity of the various kinds of incoming materials of the incoming material group by the intelligent classification component comprises the following steps: and classifying a plurality of kinds of attack objects with the same attack identification data in the classification result into the same attack object class set.
2. The clustering device that classifies metadata of a classification object according to claim 1, wherein the device further comprises:
and the big data storage component is connected with the target clustering component through a wireless network and is used for storing clustering operation results of the target clustering component on various kinds of attacks in the attack group.
3. The clustering device for classifying metadata of classification objects according to claim 1, characterized in that the device further comprises:
and the parameter configuration component is respectively connected with the network construction component, the network training component and the intelligent classification component and is used for respectively realizing the field configuration of the working parameters of the network construction component, the network training component and the intelligent classification component.
4. The clustering device for classifying metadata of classification objects according to claim 1, characterized in that the device further comprises:
and the synchronous coordination component is respectively connected with the network construction component, the network training component and the intelligent classification component and is used for realizing the cooperative control of the respective actions of the network construction component, the network training component and the intelligent classification component.
5. The clustering device for classifying metadata of classification objects according to any one of claims 1 to 4, characterized by:
the clustering operation of various kinds of attacks of the attack object group based on the classification result of the aggressivity of various kinds of attacks of the attack object group by the intelligent classification component comprises the following steps: and classifying a plurality of kinds of attack objects with the same frequent attack time in the classification result into the same attack object class set.
6. The clustering device for classifying metadata of classification objects according to claim 5, wherein:
the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments, and the method comprises the following steps: when the attack identification data is equal to 0, the type of attack object has no aggressivity, and the frequent attack moment is binary coding corresponding to a null symbol;
the classification of multiple kinds of attack objects with the same attack identification data in the classification result into the same attack species set comprises the following steps: classifying the various kinds of attack objects with attack identification data equal to 1 in the classification result into an attack type attack species set, and classifying the various kinds of attack objects with attack identification data equal to 0 in the classification result into a mild type attack species set.
7. The clustering device that classifies metadata of a classification object according to claim 6, wherein:
classifying a plurality of kinds of attack objects with the same frequent attack time in the classification result into the same attack object class set comprises the following steps: classifying the multiple kinds of attacks with the same frequent attack time in the classification result into a kind set of the same kind of attack species set, and classifying the multiple kinds of attacks with attack identification data equal to 1 in the classification result into a subset of the kind set of the same kind of attack species set.
8. The clustering device for classifying metadata of classification objects according to any one of claims 1 to 4, wherein:
the radial basis function neural network uses a radial basis function as an activation function and is composed of a multilayer structure;
wherein, in the multilayer structure of the radial basis function neural network, the numerical transformation from the input layer to the hidden layer is nonlinear, and the numerical transformation from the hidden layer to the output layer is linear.
9. A clustering device that classifies metadata of a classified object, the device comprising a memory and one or more processors, the memory storing a computer program configured to be executed by the one or more processors to perform the steps of:
establishing a radial basis function neural network of a multilayer structure, wherein the multilayer structure comprises a single input layer, a single output layer and a set number of hidden layers respectively, input data of the input layer of the radial basis function neural network are a plurality of digital images captured in a time-sharing manner in an activity area of a certain kind of incoming objects in an incoming object group, output data of the output layer of the radial basis function neural network are moments of whether the kind of incoming objects is aggressive and have the highest probability of attack action occurring every day when the kind of incoming objects is aggressive, and the number of the set number is in direct proportion to the total number of the kinds of incoming objects to be analyzed;
adopting a moment when a certain kind of attack object is known to have aggressivity or not and is known to have the highest probability of attack action occurring every day when the certain kind of attack object has aggressivity as output data of an output layer of the established radial basis function neural network, and adopting a plurality of digital images captured by the certain kind of attack object in a time division mode in an activity area of the certain kind of attack object as input data of an input layer of the established radial basis function neural network to train the established radial basis function neural network;
the method comprises the steps that a plurality of digital images captured in a time-sharing mode in a certain kind of attack object activity area to be analyzed of an attack object group are used as input data of an input layer of a radial basis function neural network after multiple times of training to operate the radial basis function neural network after multiple times of training, and the aggressivity of the kind of attack object to be analyzed is classified according to output data of the output layer of the radial basis function neural network after multiple times of training;
clustering each kind of attack object of the attack object group based on the classification result of the aggressivity of each kind of attack object of the attack object group;
the input data of the input layer of the radial basis function neural network are a plurality of digital images captured in a time-sharing mode in the active region of a certain kind of incoming material of an incoming material group, and the digital images comprise: the capturing visual fields of the digital images are the same and the resolution is equal, and a plurality of capturing moments respectively corresponding to the digital images are uniformly distributed in the time length of one day;
wherein, the output data of the output layer of the radial basis function neural network is whether the kind of attack is aggressive or not and the time when the probability of attack action is the highest every day when the kind of attack is aggressive comprises: the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments;
wherein, the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments, and the method comprises the following steps: when the attack identification data is equal to 1, the type of attack object has aggressivity, and the frequent attack moment is a binary code corresponding to the moment with the highest probability of attack action occurring every day;
the clustering operation of various types of incoming materials of the incoming material group based on the offensive classification result of the various types of incoming materials of the intelligent classification component comprises the following steps: and classifying a plurality of kinds of attack objects with the same attack identification data in the classification result into the same attack object class set.
10. A clustering method for classifying metadata of a classification object, the method comprising:
establishing a radial basis function neural network of a multilayer structure, wherein the multilayer structure comprises a single input layer, a single output layer and a set number of hidden layers respectively, input data of the input layer of the radial basis function neural network are a plurality of digital images captured in a time-sharing manner in an activity area of a certain kind of incoming objects in an incoming object group, output data of the output layer of the radial basis function neural network are moments of whether the kind of incoming objects is aggressive and have the highest probability of attack action occurring every day when the kind of incoming objects is aggressive, and the number of the set number is in direct proportion to the total number of the kinds of incoming objects to be analyzed;
adopting a moment when a certain kind of attack object is known to have aggressivity or not and is known to have the highest probability of attack action occurring every day when the certain kind of attack object has aggressivity as output data of an output layer of the established radial basis function neural network, and adopting a plurality of digital images captured by the certain kind of attack object in a time division mode in an activity area of the certain kind of attack object as input data of an input layer of the established radial basis function neural network to train the established radial basis function neural network;
the method comprises the steps that a plurality of digital images captured in a time-sharing mode in a certain kind of attack object activity area to be analyzed of an attack object group are used as input data of an input layer of a radial basis function neural network after multiple times of training to operate the radial basis function neural network after multiple times of training, and the aggressivity of the kind of attack object to be analyzed is classified according to output data of the output layer of the radial basis function neural network after multiple times of training;
clustering each kind of attack object of the attack object group based on the classification result of the aggressivity of each kind of attack object of the attack object group;
the input data of the input layer of the radial basis function neural network is a plurality of digital images captured in a time-sharing mode in an active region of a certain kind of incoming object in an incoming object group, and the digital images comprise: the capturing visual fields of the digital images are the same and the resolution is equal, and a plurality of capturing moments respectively corresponding to the digital images are uniformly distributed in the time length of one day;
wherein, the output data of the output layer of the radial basis function neural network is whether the kind of attack is aggressive or not and the time when the probability of attack action is the highest every day when the kind of attack is aggressive comprises: the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments;
wherein, the output data of the output layer of the radial basis function neural network comprises attack identification data and frequent attack moments, and the method comprises the following steps: when the attack identification data is equal to 1, the type of attack object has aggressivity, and the frequent attack moment is a binary code corresponding to the moment with the highest probability of attack action occurring every day;
the clustering operation of various types of incoming materials of the incoming material group based on the classification result of the aggressivity of the various types of incoming materials of the incoming material group by the intelligent classification component comprises the following steps: and classifying a plurality of kinds of attack objects with the same attack identification data in the classification result into the same attack object class set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111463988.1A CN114155390B (en) | 2021-12-02 | 2021-12-02 | Clustering device for classifying metadata of classified objects |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111463988.1A CN114155390B (en) | 2021-12-02 | 2021-12-02 | Clustering device for classifying metadata of classified objects |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114155390A CN114155390A (en) | 2022-03-08 |
| CN114155390B true CN114155390B (en) | 2022-05-06 |
Family
ID=80456269
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111463988.1A Active CN114155390B (en) | 2021-12-02 | 2021-12-02 | Clustering device for classifying metadata of classified objects |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114155390B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106411896A (en) * | 2016-09-30 | 2017-02-15 | 重庆邮电大学 | APDE-RBF neural network based network security situation prediction method |
| CN110782481A (en) * | 2019-10-18 | 2020-02-11 | 华中光电技术研究所（中国船舶重工集团有限公司第七一七研究所） | Unmanned ship intelligent decision method and system |
| CN110826877A (en) * | 2019-10-24 | 2020-02-21 | 四川航天系统工程研究所 | Fire power distribution method for multiple point targets |
| CN113408699A (en) * | 2021-06-16 | 2021-09-17 | 中国地质科学院 | Lithology identification method and system based on improved radial basis function neural network |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7362223B2 (en) * | 2004-06-30 | 2008-04-22 | Lockhead Martin Corporation | Chemical, biological, radiological, and nuclear weapon detection system with environmental acuity |
| US10198667B2 (en) * | 2015-09-02 | 2019-02-05 | Pocketguardian, Llc | System and method of detecting offensive content sent or received on a portable electronic device |
| US11195067B2 (en) * | 2018-12-21 | 2021-12-07 | Ambient AI, Inc. | Systems and methods for machine learning-based site-specific threat modeling and threat detection |
-
2021
- 2021-12-02 CN CN202111463988.1A patent/CN114155390B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106411896A (en) * | 2016-09-30 | 2017-02-15 | 重庆邮电大学 | APDE-RBF neural network based network security situation prediction method |
| CN110782481A (en) * | 2019-10-18 | 2020-02-11 | 华中光电技术研究所（中国船舶重工集团有限公司第七一七研究所） | Unmanned ship intelligent decision method and system |
| CN110826877A (en) * | 2019-10-24 | 2020-02-21 | 四川航天系统工程研究所 | Fire power distribution method for multiple point targets |
| CN113408699A (en) * | 2021-06-16 | 2021-09-17 | 中国地质科学院 | Lithology identification method and system based on improved radial basis function neural network |
Non-Patent Citations (2)
| Title |
|---|
| Entity resolution for media metadata based;Qi Gu等;《Multimedia Tools & Applications》;20190813;第219-242页 * |
| 基于SVC的云服务化作战资源聚类重组研究;赵仁君 等;《第四届中国指挥控制大会论文集》;20160704;第383-389页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114155390A (en) | 2022-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hossain et al. | Crime prediction using spatio-temporal data | |
| Wares et al. | Data stream mining: methods and challenges for handling concept drift | |
| McElwee et al. | Deep learning for prioritizing and responding to intrusion detection alerts | |
| Popp et al. | Countering terrorism through information technology | |
| Ahmadi et al. | Modeling recurring concepts in data streams: a graph-based framework | |
| Watts et al. | R&D cluster quality measures and technology maturity | |
| Li et al. | Locally weighted fusion of structural and attribute information in graph clustering | |
| Alghanmi et al. | Machine learning approaches for anomaly detection in IoT: an overview and future research directions | |
| Cho et al. | Predicting future technological convergence patterns based on machine learning using link prediction | |
| Zhang et al. | Foundations of intelligent knowledge management | |
| Win et al. | PCPD: A parallel crime pattern discovery system for large-scale spatiotemporal data based on fuzzy clustering | |
| CN110737821B (en) | Similar event query method, device, storage medium and terminal equipment | |
| Zhou et al. | A statistics based approach for extracting priority rules from trained neural networks | |
| CN114155390B (en) | Clustering device for classifying metadata of classified objects | |
| Capdevila et al. | Mining urban events from the tweet stream through a probabilistic mixture model | |
| CN110740111B (en) | Data leakage prevention method and device and computer readable storage medium | |
| Pohl et al. | Active online learning for social media analysis to support crisis management | |
| Sujatha et al. | A new efficient SIF-based FCIL (SIF–FCIL) mining algorithm in predicting the crime locations | |
| Jung et al. | DAC: Descendant-aware clustering algorithm for network-based topic emergence prediction | |
| Panigrahi et al. | An efficient approach to detect sudden changes in vegetation index time series for land change detection | |
| Rajesh et al. | Fuzzy based privacy preserving classification of data streams | |
| Alfred et al. | Data summarization approach to relational domain learning based on frequent pattern to support the development of decision making | |
| Johnson et al. | GT-CHES: Graph transformation for classification in human evolutionary systems | |
| Siddique et al. | Enhancing comprehensibility of software clustering results | |
| Jousselme et al. | Uncertainty representations for information retrieval with missing data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP02 | Change in the address of a patent holder |
Address after: Room 327, third floor, No. 2, Stadium South Road, Shijingshan District, Beijing 100043 Patentee after: Beijing Zhongke Zhiyi Technology Co.,Ltd. Address before: Room 331, third floor, No. 2, Stadium South Road, Shijingshan District, Beijing 100043 Patentee before: Beijing Zhongke Zhiyi Technology Co.,Ltd. |
|
| CP02 | Change in the address of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: Room 327, third floor, No. 2, Stadium South Road, Shijingshan District, Beijing 100043 Patentee after: Beijing Zhongke Zhiyi Technology Co.,Ltd. Address before: Room 327, third floor, No. 2, Stadium South Road, Shijingshan District, Beijing 100043 Patentee before: Beijing Zhongke Zhiyi Technology Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |