Disclosure of Invention
In view of the above, an object of the embodiments of the present invention is to provide a method and a system for diagnosing object anomalies based on model identification, which acquire candidate object behavior data corresponding to a target online session user; inputting the candidate object behavior data into a preset abnormal knowledge point prediction network, and generating information for representing whether each object behavior data block comprises an abnormal behavior knowledge point; the candidate object behavior data includes a plurality of object behavior data chunks; based on the behavior knowledge points of the object behavior data blocks, generating related abnormal node information corresponding to related object behavior data block sequences; determining an abnormal mining result of the object behavior data block based on the related abnormal node information; and generating an abnormal positioning root cause corresponding to the target online session user based on the abnormal mining result, and performing abnormal interception diagnosis on the target online session user based on the abnormal positioning root cause, so that the accuracy of the subsequent abnormal interception diagnosis on the target online session user can be improved.
According to an aspect of the embodiment of the present invention, there is provided a method and a system for diagnosing an abnormality of an object based on model identification, the method including:
acquiring candidate object behavior data corresponding to a target online session user;
inputting the candidate object behavior data into a preset abnormal knowledge point prediction network, and generating information for representing whether each object behavior data block comprises an abnormal behavior knowledge point or not; the candidate object behavior data comprises a plurality of object behavior data blocks;
based on the behavior knowledge points of the object behavior data blocks, generating related abnormal node information corresponding to related object behavior data block sequences; the related object behavior data block sequence comprises a plurality of mutually connected object behavior data blocks;
determining an abnormal mining result of the object behavior data block based on the related abnormal node information;
and generating an abnormal positioning root cause corresponding to the target online session user based on the abnormal mining result, and carrying out abnormal interception diagnosis on the target online session user based on the abnormal positioning root cause.
In an alternative embodiment, the generating, based on the behavior knowledge points of the object behavior data blocks, related abnormal node information corresponding to the related object behavior data block sequences includes;
determining a first behavior knowledge point of a target object behavior data block in all the object behavior data blocks;
determining associated object behavior data blocks which accord with behavior association conditions with the target object behavior data blocks in all the object behavior data blocks; the related object behavior data block sequence comprises:
the target object behavior data blocks and the associated object behavior data blocks;
determining second behavior knowledge points corresponding to the behavior data blocks of the associated objects;
generating associated abnormal node information corresponding to the associated object behavior data blocks based on the second behavior knowledge points;
and generating the related abnormal node information based on the first behavior knowledge point and the related abnormal node information.
In an alternative embodiment, the generating the related abnormal node information based on the first behavior knowledge point and the related abnormal node information includes:
generating overall abnormal node information corresponding to the candidate object behavior data based on the behavior knowledge points of each object behavior data block;
calculating the first behavior knowledge points, the associated abnormal node information and the overall abnormal node information based on a set fusion model to generate fusion characteristic information;
and generating the related abnormal node information based on the fusion characteristic information.
In an alternative embodiment, the method further comprises:
acquiring template object behavior data for updating network weight parameters of an initialized neural network;
determining a dismantling rule for dismantling the template object behavior data;
based on the dismantling rule, dismantling the template object behavior data to generate a plurality of dismantling member object behavior data;
generating an abnormal mining array corresponding to the template object behavior data based on the first behavior trigger information and the abnormal training content of each disassembled member object behavior data;
the first behavior trigger information and the second behavior trigger information corresponding to the same behavior data of the disassembled member object are correspondingly associated and configured;
the first behavior trigger information is used for representing the position of the disassembled member object behavior data in the candidate object behavior data; the second behavior trigger information is used for representing the position of a preset behavior knowledge point generated based on abnormal training content corresponding to the behavior data of the disassembled member object in the abnormal mining array;
generating sample learning data and sample test data based on the template object behavior data and an anomaly mining array corresponding to each of the template object behavior data;
updating the network weight parameters of the initialized neural network through the sample learning data to generate an updated neural network;
and after the updated neural network is tested through the sample test data and meets the set test validity conditions, the updated neural network is used as the abnormal knowledge point prediction network.
In an alternative embodiment, the generating the anomaly mining array corresponding to the template object behavior data based on the first behavior trigger information of the respective disassembled member object behavior data and the anomaly training content includes:
acquiring abnormal prior label configuration information generated by carrying out abnormal prior label configuration on the template object behavior data;
generating the abnormal training content corresponding to each disassembled member object behavior data based on the abnormal coding mapping information of the abnormal priori label configuration information included in each disassembled member object behavior data;
determining whether abnormal coding mapping information of the abnormal priori label configuration information included in the behavior data of each disassembled member object is in a preset mapping interval or not based on the abnormal training content, and generating preset behavior knowledge points corresponding to the behavior data of each disassembled member object;
generating the first behavior trigger information corresponding to each piece of disassembled member object behavior data based on the position of each piece of disassembled member object behavior data in the template object behavior data;
and generating an abnormal mining array corresponding to the template object behavior data based on the first behavior trigger information and the preset behavior knowledge points corresponding to the disassembled member object behavior data.
In an alternative embodiment, when the attribute type of the target online session user is a first preset attribute type; after the generating of the abnormal positioning root cause corresponding to the target online session user based on the abnormal mining result, the method further comprises:
determining abnormal thermodynamic diagram information of the target online session user based on the abnormal positioning root cause;
when the abnormal thermodynamic diagram information has an abnormal session link larger than a set thermodynamic value, judging that the target online session user is an abnormal online session user;
acquiring target online session user behavior trigger information of the target online session user;
and generating interception configuration information corresponding to the target online session user based on the target online session user behavior trigger information.
In an alternative embodiment, when the attribute type of the target online session user is a second preset attribute type; after the generating of the abnormal positioning root cause corresponding to the target online session user based on the abnormal mining result, the method further comprises:
determining abnormal thermodynamic diagram information of the target online session user based on the abnormal positioning root cause;
when the abnormal thermodynamic diagram information has an abnormal session link larger than a set thermodynamic value, acquiring cooperative node information of the target online session user;
and generating interception configuration information corresponding to the target online session user based on the cooperative node information.
According to another aspect of the embodiment of the present invention, there is provided a method and a system for diagnosing an abnormality of an object based on model identification, the system including:
the acquisition unit is used for acquiring candidate object behavior data corresponding to the target online session user;
the first generation unit is used for inputting the candidate object behavior data into a preset abnormal knowledge point prediction network and generating information for representing whether each object behavior data block comprises an abnormal behavior knowledge point or not; the candidate object behavior data comprises a plurality of object behavior data blocks;
the second generation unit is used for generating related abnormal node information corresponding to the related object behavior data block sequence based on the behavior knowledge points of the object behavior data blocks; the related object behavior data block sequence comprises a plurality of mutually connected object behavior data blocks;
the determining unit is used for determining an abnormal mining result of the object behavior data block based on the related abnormal node information;
and the diagnosis unit is used for generating an abnormality locating root cause corresponding to the target online session user based on the abnormality mining result and carrying out abnormality interception diagnosis on the target online session user based on the abnormality locating root cause.
According to another aspect of an embodiment of the present invention, there is provided an electronic apparatus including: the device comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus; the memory is used for storing a computer program; the processor is configured to implement the object abnormality diagnosis method step based on model identification described in any one of the above when executing the computer program.
According to another aspect of the embodiments of the present invention, there is provided a readable storage medium having stored thereon a computer program which, when executed by a processor, can perform the steps of the above-described object abnormality diagnosis method based on model identification.
The foregoing objects, features and advantages of embodiments of the invention will be more readily apparent from the following detailed description of the embodiments taken in conjunction with the accompanying drawings.
Detailed Description
In order to enable those skilled in the art to better understand the present invention, a technical solution of the present embodiment of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiment of the present invention, and it is apparent that the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. All other embodiments, which can be made by those skilled in the art without the benefit of the teachings of this invention, are intended to fall within the scope of the invention.
The terms first, second, third and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented, for example, in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Fig. 1 shows an exemplary component diagram of a server 100. The server 100 may include one or more processors 104, such as one or more Central Processing Units (CPUs), each of which may implement one or more hardware threads. The server 100 may also include any storage medium 106 for storing any kind of information such as code, settings, data, etc. For example, and without limitation, storage medium 106 may include any one or more of the following combinations: any type of RAM, any type of ROM, flash memory devices, hard disks, optical disks, etc. More generally, any storage medium may store information using any technique. Further, any storage medium may provide volatile or non-volatile retention of information. Further, any storage medium may represent fixed or removable components of server 100. In one case, the server 100 may perform any of the operations of the associated instructions when the processor 104 executes the associated instructions stored in any storage medium or combination of storage media. The server 100 also includes one or more drive units 108, such as a hard disk drive unit, an optical disk drive unit, etc., for interacting with any storage media.
The server 100 also includes input/output 110 (I/O) for receiving various inputs (via input unit 112) and for providing various outputs (via output unit 114). One particular output mechanism may include a presentation device 116 and an associated Graphical User Interface (GUI) 118. The server 100 may also include one or more network interfaces 120 for exchanging data with other devices via one or more communication units 122. One or more communication buses 124 couple the components described above together.
The communication unit 122 may be implemented in any manner, for example, via a local area network, a wide area network (e.g., the internet), a point-to-point connection, etc., or any combination thereof. The communication unit 122 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers 100, etc., governed by any protocol or combination of protocols.
Fig. 2 is a schematic flow chart of a method and a system for diagnosing object abnormality based on model recognition according to an embodiment of the present invention, which may be executed by the server 100 shown in fig. 1, and detailed steps of the method and the system for diagnosing object abnormality based on model recognition are described below.
Step S110, candidate object behavior data corresponding to a target online session user is obtained;
step S120, inputting the candidate object behavior data into a preset abnormal knowledge point prediction network, and generating information for representing whether each object behavior data block comprises abnormal behavior knowledge points or not; the candidate object behavior data comprises a plurality of object behavior data blocks;
step S130, based on the behavior knowledge points of the object behavior data blocks, generating related abnormal node information corresponding to related object behavior data block sequences; the related object behavior data block sequence comprises a plurality of mutually connected object behavior data blocks;
step S140, determining an abnormal mining result of the object behavior data block based on the related abnormal node information;
and step S150, generating an abnormality locating root cause corresponding to the target online session user based on the abnormality mining result, and carrying out abnormality interception diagnosis on the target online session user based on the abnormality locating root cause.
Based on the above steps, the embodiment obtains candidate object behavior data corresponding to the target online session user; inputting the candidate object behavior data into a preset abnormal knowledge point prediction network, and generating information for representing whether each object behavior data block comprises an abnormal behavior knowledge point; the candidate object behavior data includes a plurality of object behavior data chunks; based on the behavior knowledge points of the object behavior data blocks, generating related abnormal node information corresponding to related object behavior data block sequences; determining an abnormal mining result of the object behavior data block based on the related abnormal node information; and generating an abnormal positioning root cause corresponding to the target online session user based on the abnormal mining result, and performing abnormal interception diagnosis on the target online session user based on the abnormal positioning root cause, so that the accuracy of the subsequent abnormal interception diagnosis on the target online session user can be improved.
In an alternative embodiment, the generating, based on the behavior knowledge points of the object behavior data blocks, related abnormal node information corresponding to the related object behavior data block sequences includes;
determining a first behavior knowledge point of a target object behavior data block in all the object behavior data blocks;
determining associated object behavior data blocks which accord with behavior association conditions with the target object behavior data blocks in all the object behavior data blocks; the related object behavior data block sequence comprises:
the target object behavior data blocks and the associated object behavior data blocks;
determining second behavior knowledge points corresponding to the behavior data blocks of the associated objects;
generating associated abnormal node information corresponding to the associated object behavior data blocks based on the second behavior knowledge points;
and generating the related abnormal node information based on the first behavior knowledge point and the related abnormal node information.
In an alternative embodiment, the generating the related abnormal node information based on the first behavior knowledge point and the related abnormal node information includes:
generating overall abnormal node information corresponding to the candidate object behavior data based on the behavior knowledge points of each object behavior data block;
calculating the first behavior knowledge points, the associated abnormal node information and the overall abnormal node information based on a set fusion model to generate fusion characteristic information;
and generating the related abnormal node information based on the fusion characteristic information.
In an alternative embodiment, the method further comprises:
acquiring template object behavior data for updating network weight parameters of an initialized neural network;
determining a dismantling rule for dismantling the template object behavior data;
based on the dismantling rule, dismantling the template object behavior data to generate a plurality of dismantling member object behavior data;
generating an abnormal mining array corresponding to the template object behavior data based on the first behavior trigger information and the abnormal training content of each disassembled member object behavior data;
the first behavior trigger information and the second behavior trigger information corresponding to the same behavior data of the disassembled member object are correspondingly associated and configured;
the first behavior trigger information is used for representing the position of the disassembled member object behavior data in the candidate object behavior data; the second behavior trigger information is used for representing the position of a preset behavior knowledge point generated based on abnormal training content corresponding to the behavior data of the disassembled member object in the abnormal mining array;
generating sample learning data and sample test data based on the template object behavior data and an anomaly mining array corresponding to each of the template object behavior data;
updating the network weight parameters of the initialized neural network through the sample learning data to generate an updated neural network;
and after the updated neural network is tested through the sample test data and meets the set test validity conditions, the updated neural network is used as the abnormal knowledge point prediction network.
In an alternative embodiment, the generating the anomaly mining array corresponding to the template object behavior data based on the first behavior trigger information of the respective disassembled member object behavior data and the anomaly training content includes:
acquiring abnormal prior label configuration information generated by carrying out abnormal prior label configuration on the template object behavior data;
generating the abnormal training content corresponding to each disassembled member object behavior data based on the abnormal coding mapping information of the abnormal priori label configuration information included in each disassembled member object behavior data;
determining whether abnormal coding mapping information of the abnormal priori label configuration information included in the behavior data of each disassembled member object is in a preset mapping interval or not based on the abnormal training content, and generating preset behavior knowledge points corresponding to the behavior data of each disassembled member object;
generating the first behavior trigger information corresponding to each piece of disassembled member object behavior data based on the position of each piece of disassembled member object behavior data in the template object behavior data;
and generating an abnormal mining array corresponding to the template object behavior data based on the first behavior trigger information and the preset behavior knowledge points corresponding to the disassembled member object behavior data.
In an alternative embodiment, when the attribute type of the target online session user is a first preset attribute type; after the generating of the abnormal positioning root cause corresponding to the target online session user based on the abnormal mining result, the method further comprises:
determining abnormal thermodynamic diagram information of the target online session user based on the abnormal positioning root cause;
when the abnormal thermodynamic diagram information has an abnormal session link larger than a set thermodynamic value, judging that the target online session user is an abnormal online session user;
acquiring target online session user behavior trigger information of the target online session user;
and generating interception configuration information corresponding to the target online session user based on the target online session user behavior trigger information.
In an alternative embodiment, when the attribute type of the target online session user is a second preset attribute type; after the generating of the abnormal positioning root cause corresponding to the target online session user based on the abnormal mining result, the method further comprises:
determining abnormal thermodynamic diagram information of the target online session user based on the abnormal positioning root cause;
when the abnormal thermodynamic diagram information has an abnormal session link larger than a set thermodynamic value, acquiring cooperative node information of the target online session user;
and generating interception configuration information corresponding to the target online session user based on the cooperative node information.
Fig. 3 shows a functional block diagram of an object abnormality diagnosis system 200 according to model-based recognition, which may correspond to the steps performed by the above-described method according to the function implemented by the object abnormality diagnosis system 200 according to model-based recognition, according to an embodiment of the present invention. The object abnormality diagnosis system 200 according to model recognition may be understood as the above-mentioned server 100 or a processor of the server 100, or may be understood as a component which is independent of the above-mentioned server 100 or processor and implements the functions of the present invention under the control of the server 100, as shown in fig. 3, and the functions of the respective functional modules of the object abnormality diagnosis system 200 according to model recognition will be described in detail below.
An obtaining unit 210, configured to obtain candidate object behavior data corresponding to a target online session user;
a first generating unit 220, configured to input the candidate object behavior data into a preset abnormal knowledge point prediction network, and generate a rule for characterizing whether each object behavior data block includes an abnormal behavior knowledge point; the candidate object behavior data comprises a plurality of object behavior data blocks;
a second generating unit 230, configured to generate relevant abnormal node information corresponding to the relevant object behavior data block sequence based on the behavior knowledge points of the object behavior data blocks; the related object behavior data block sequence comprises a plurality of mutually connected object behavior data blocks;
a determining unit 240, configured to determine an abnormal mining result of the object behavior data block based on the related abnormal node information;
and the diagnosis unit 250 is used for generating an abnormality locating root cause corresponding to the target online session user based on the abnormality mining result and carrying out abnormality interception diagnosis on the target online session user based on the abnormality locating root cause.
In an alternative embodiment, the second generating unit 230 is specifically further configured to;
determining a first behavior knowledge point of a target object behavior data block in all the object behavior data blocks;
determining associated object behavior data blocks which accord with behavior association conditions with the target object behavior data blocks in all the object behavior data blocks; the related object behavior data block sequence comprises:
the target object behavior data blocks and the associated object behavior data blocks;
determining second behavior knowledge points corresponding to the behavior data blocks of the associated objects;
generating associated abnormal node information corresponding to the associated object behavior data blocks based on the second behavior knowledge points;
and generating the related abnormal node information based on the first behavior knowledge point and the related abnormal node information.
In an alternative embodiment, the second generating unit 230 is specifically further configured to;
generating overall abnormal node information corresponding to the candidate object behavior data based on the behavior knowledge points of each object behavior data block;
calculating the first behavior knowledge points, the associated abnormal node information and the overall abnormal node information based on a set fusion model to generate fusion characteristic information;
and generating the related abnormal node information based on the fusion characteristic information.
In an alternative embodiment, the first generating unit 220 is specifically further configured to;
acquiring template object behavior data for updating network weight parameters of an initialized neural network;
determining a dismantling rule for dismantling the template object behavior data;
based on the dismantling rule, dismantling the template object behavior data to generate a plurality of dismantling member object behavior data;
generating an abnormal mining array corresponding to the template object behavior data based on the first behavior trigger information and the abnormal training content of each disassembled member object behavior data;
the first behavior trigger information and the second behavior trigger information corresponding to the same behavior data of the disassembled member object are correspondingly associated and configured;
the first behavior trigger information is used for representing the position of the disassembled member object behavior data in the candidate object behavior data; the second behavior trigger information is used for representing the position of a preset behavior knowledge point generated based on abnormal training content corresponding to the behavior data of the disassembled member object in the abnormal mining array;
generating sample learning data and sample test data based on the template object behavior data and an anomaly mining array corresponding to each of the template object behavior data;
updating the network weight parameters of the initialized neural network through the sample learning data to generate an updated neural network;
and after the updated neural network is tested through the sample test data and meets the set test validity conditions, the updated neural network is used as the abnormal knowledge point prediction network.
In an alternative embodiment, the determining unit 240 is specifically further configured to;
acquiring abnormal prior label configuration information generated by carrying out abnormal prior label configuration on the template object behavior data;
generating the abnormal training content corresponding to each disassembled member object behavior data based on the abnormal coding mapping information of the abnormal priori label configuration information included in each disassembled member object behavior data;
determining whether abnormal coding mapping information of the abnormal priori label configuration information included in the behavior data of each disassembled member object is in a preset mapping interval or not based on the abnormal training content, and generating preset behavior knowledge points corresponding to the behavior data of each disassembled member object;
generating the first behavior trigger information corresponding to each piece of disassembled member object behavior data based on the position of each piece of disassembled member object behavior data in the template object behavior data;
and generating an abnormal mining array corresponding to the template object behavior data based on the first behavior trigger information and the preset behavior knowledge points corresponding to the disassembled member object behavior data.
In an alternative embodiment, the diagnostic unit 250 is specifically further configured to;
determining abnormal thermodynamic diagram information of the target online session user based on the abnormal positioning root cause;
when the abnormal thermodynamic diagram information has an abnormal session link larger than a set thermodynamic value, judging that the target online session user is an abnormal online session user;
acquiring target online session user behavior trigger information of the target online session user;
and generating interception configuration information corresponding to the target online session user based on the target online session user behavior trigger information.
In an alternative embodiment, the diagnostic unit 250 is specifically further configured to;
determining abnormal thermodynamic diagram information of the target online session user based on the abnormal positioning root cause;
when the abnormal thermodynamic diagram information has an abnormal session link larger than a set thermodynamic value, acquiring cooperative node information of the target online session user;
and generating interception configuration information corresponding to the target online session user based on the cooperative node information.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.