CN118227741B - Crown block system alarm log analysis method and device, electronic equipment and storage medium - Google Patents
Crown block system alarm log analysis method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN118227741B CN118227741B CN202410643052.4A CN202410643052A CN118227741B CN 118227741 B CN118227741 B CN 118227741B CN 202410643052 A CN202410643052 A CN 202410643052A CN 118227741 B CN118227741 B CN 118227741B
- Authority
- CN
- China
- Prior art keywords
- alarm
- type
- information
- alarm information
- word
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Alarm Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides an overhead travelling crane system alarm log analysis method, an apparatus, an electronic device and a storage medium, wherein after data preparation is carried out by traversing an alarm information set, a time sequence vector of each alarm type is determined, for any alarm type, a semantic vector of the alarm type is extracted, and for any two alarm types, the association strength between any two alarm types is determined, so that whether any two alarm types are associated alarm types or not can be determined according to the time sequence vector of any two alarm types, the semantic vector and the association strength between any two alarm types, the whole mining process only needs to completely traverse alarm information in a log library, the mining efficiency is improved, in addition, the time sequence characteristics and the semantic information of the alarm information of a single alarm type are considered in the whole mining mode, and meanwhile, the time sequence association degree between the alarm information of the two alarm types is combined, and the mining accuracy of the alarm association is improved.
Description
Technical Field
The present invention relates to the field of log processing technologies, and in particular, to a method and apparatus for analyzing an alarm log of an overhead travelling crane system, an electronic device, and a storage medium.
Background
In order to monitor the states of all key components, potential faults or abnormal conditions in the crown block system are detected, and warning is sent out in advance, the warning system can monitor the crown block system in real time, so that the system is kept to operate in an optimal state, and the operation safety and the operation efficiency are improved. However, a large number of devices in the overhead travelling crane system can generate a large number of alarm logs, in addition, for the overhead travelling crane system, especially for a large-scale and complex overhead travelling crane system, single alarm information may not fully reflect the overall condition of the system, if operation and maintenance monitoring personnel analyze each alarm information one by one, the problem of low efficiency exists, and meanwhile, it is difficult to effectively and fully monitor and analyze the alarm logs, so that the problems existing in the system cannot be located in time. Therefore, it is necessary to perform relevance analysis of the alarm information, and the relevance analysis can help to find potential relations among multiple alarms, and screen the alarm information with relevance from massive alarm logs, so that the running state of the system can be understood more comprehensively and more efficiently.
The current way of performing relevance analysis on alarm information generally uses an association rule mining algorithm, such as Apriori algorithm, to discover frequent association rules between alarms by identifying co-occurrence patterns between alarms. However, the rule mining algorithm requires multiple traversals Shi Gaojing of the logs, which results in less efficient mining when the number of historical alert logs is large. In addition, the mining algorithm has certain requirements on the distribution of the alarm data, and only frequently occurring items can be mined, if the support degree is set to be low, the problems that the number of the determined frequent item sets is too large, the number of the association rules established based on the frequent item sets is too large, and the association rule mining effect is poor are caused.
Disclosure of Invention
The invention provides an analysis method, an analysis device, electronic equipment and a storage medium for alarm logs of an overhead travelling crane system, which are used for solving the defects of low log relevance mining efficiency and poor mining effect in the prior art.
The invention provides an analysis method of an alarm log of an overhead travelling crane system, which comprises the following steps:
Traversing each alarm information in the alarm information set according to the time stamp sequence, constructing a word list of each alarm type and a word vector list corresponding to each alarm information, and dividing a plurality of alarm subsets based on the time stamp of each alarm information; the alarm subset consists of triplets formed by alarm type fields, id and timestamp fields of each alarm message;
determining a timing vector for each alarm type based on the plurality of alarm subsets;
Extracting semantic vectors of any alarm type based on word lists of the alarm type and word vector lists corresponding to alarm information belonging to the alarm type aiming at the alarm type;
Aiming at any two alarm types, acquiring an analysis alarm sequence, and determining the association strength between the any two alarm types based on a word vector list corresponding to alarm information in the analysis alarm sequence and the word list of the any two alarm types;
For any two alarm types, determining whether the any two alarm types are associated alarm types based on the time sequence vectors, the semantic vectors and the association strength between the any two alarm types.
According to the method for analyzing the alarm log of the crown block system provided by the invention, each alarm information sequenced according to the time stamp in the alarm information set is traversed, a word list of each alarm type and a word vector list corresponding to each alarm information are constructed, and a plurality of alarm subsets are divided based on the time stamp of each alarm information, and the method concretely comprises the following steps:
creating an empty vocabulary for each alarm type;
Traversing each alarm information ordered by time stamp in the alarm information set, determining an alarm type field, a time stamp field and an alarm description field of the current alarm information aiming at the current alarm information, updating a word list of an alarm type indicated by the alarm type field of the current alarm information based on each word of the alarm description field of the current alarm information, creating a word vector list corresponding to the current alarm information, and distributing the current alarm information to an established alarm subset or a newly-built alarm subset based on the time stamp field of the current alarm information and distributing the current alarm information to the newly-built alarm subset;
The word list of any alarm type comprises word vectors of word segmentation contained in alarm information belonging to the any alarm type and the occurrence frequency of corresponding word segmentation in the alarm information belonging to the any alarm type; the word vector list corresponding to any alarm information comprises the serial numbers of each word of the alarm description field of the any alarm information in the word list of the alarm type indicated by the alarm type field of the any alarm information.
According to the method for analyzing the alarm log of the crown block system provided by the invention, the method for analyzing the alarm log of the crown block system distributes the current alarm information to the created alarm subset or the newly-built alarm subset based on the timestamp field of the current alarm information and distributes the current alarm information to the newly-built alarm subset, and specifically comprises the following steps:
If the timestamp field of the current alarm information exceeds the maximum time of the current created latest alarm sub-set, creating a new alarm sub-set, adding a triplet formed by the alarm type field, the id and the timestamp field of the current alarm information into the new alarm sub-set, and determining the maximum time of the new alarm sub-set based on the timestamp field of the current alarm information and a preset time window length;
and if the time stamp field of the current alarm information does not exceed the maximum time of the current created latest alarm sub-set, adding a triplet formed by the alarm type field, the id and the time stamp field of the current alarm information into the current created latest alarm sub-set.
According to the method for analyzing the alarm log of the crown block system provided by the invention, the time sequence vector of each alarm type is determined based on the plurality of alarm subsets, and the method specifically comprises the following steps:
Sequentially determining the quantity of alarm information belonging to any alarm type contained in the plurality of alarm subsets aiming at any alarm type;
And determining a time sequence vector of any alarm type based on the quantity of alarm information belonging to the alarm type contained in each alarm subset.
According to the analysis method of the crown block system alarm log provided by the invention, for any alarm type, the semantic vector of any alarm type is extracted based on the word list of any alarm type and the word vector list corresponding to the alarm information belonging to the alarm type, and the analysis method specifically comprises the following steps:
Determining a template text of any alarm type based on the word list of any alarm type and a word vector list corresponding to alarm information belonging to the any alarm type;
And extracting the semantic vector of any alarm type based on the template text of any alarm type.
According to the analysis method of the alarm log of the crown block system provided by the invention, the template text of any alarm type is determined based on the word list of any alarm type and the word vector list corresponding to the alarm information belonging to the any alarm type, and the analysis method specifically comprises the following steps:
determining key word segmentation in the word list of any alarm type based on the frequency of each word segmentation in the word list of any alarm type;
Determining the sequence among key segmentation words in the word list of any alarm type based on a word vector list corresponding to the alarm information of the any alarm type;
generating template text of any alarm type based on the sequence among the keyword segmentation words in the word list of the alarm type.
According to the method for analyzing the alarm log of the crown block system, which is provided by the invention, aiming at any two alarm types, an analysis alarm sequence is obtained, and the association strength between any two alarm types is determined based on a word vector list corresponding to alarm information in the analysis alarm sequence and the word list of the any two alarm types, and the method specifically comprises the following steps:
Based on the alarm subsets, acquiring a preset number of alarm subsets to be combined into a subset to be analyzed; wherein the subset to be analyzed contains the latest alarm information belonging to one of the two alarm types;
After deleting the triples of which the alarm type field is not the any two alarm types in the subset to be analyzed, aiming at any triplet in the subset to be analyzed, splicing word vector sequences corresponding to corresponding alarm information into any triplet based on the id of the alarm information in the any triplet to obtain the analysis alarm sequence; the word vector sequence corresponding to the corresponding alarm information is constructed based on the word vector list corresponding to the corresponding alarm information and the word list of the alarm type to which the corresponding alarm information belongs;
Inputting the analysis alarm sequence into a correlation analysis model to obtain the correlation strength between any two alarm types output by the correlation analysis model.
The invention also provides an alarm log analysis device of the crown block system, which comprises:
the data traversing unit is used for traversing each alarm information in the alarm information set according to the time stamp sequence, constructing a word list of each alarm type and a word vector list corresponding to each alarm information, and dividing a plurality of alarm subsets based on the time stamp of each alarm information; the alarm subset consists of triplets formed by alarm type fields, id and timestamp fields of each alarm message;
a time sequence feature extraction unit, configured to determine a time sequence vector of each alarm type based on the plurality of alarm subsets;
The semantic information extraction unit is used for extracting semantic vectors of any alarm type based on word lists of the any alarm type and word vector lists corresponding to alarm information belonging to the any alarm type aiming at the any alarm type;
the association strength analysis unit is used for acquiring an analysis alarm sequence aiming at any two alarm types, and determining association strength between any two alarm types based on a word vector list corresponding to alarm information in the analysis alarm sequence and word lists of the any two alarm types;
the confirming unit is used for determining whether any two alarm types are associated alarm types or not according to the time sequence vectors, the semantic vectors and the association strength between the any two alarm types for any two alarm types.
The invention also provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes any one of the above crown block system alarm log analysis methods when executing the program.
The invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of analysis of an alarm log of an overhead travelling crane system as described in any of the above.
The invention also provides a computer program product comprising a computer program which, when executed by a processor, implements the method for analyzing the alarm log of the crown block system according to any one of the above.
According to the crown block system alarm log analysis method, the crown block system alarm log analysis device, the electronic equipment and the storage medium, each alarm information in the alarm information set is traversed, a word list of each alarm type and a word vector list corresponding to each alarm information are constructed, a plurality of alarm subsets are divided based on the time stamp of each alarm information, then the time sequence vector of each alarm type is determined based on the plurality of alarm subsets, for any alarm type, the semantic vector of the alarm type is extracted based on the word list of the alarm type and the word vector list corresponding to the alarm information belonging to the alarm type, for any two alarm types, the analysis alarm sequence is acquired from the alarm information set, and the association strength between any two alarm types is determined based on the word vector list corresponding to the alarm information in the analysis alarm sequence and the word list of any two alarm types, so that whether any two alarm types are associated or not can be determined based on the time sequence vector of any two alarm types, the semantic vector and the association strength between any two alarm types, the whole mining process is more accurate, and the association between the two alarm types can be mined by combining the time sequence information, and the time sequence information can be more accurately mined, and the association characteristics can be more accurately judged.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of an alarm log analysis method of an overhead travelling crane system;
FIG. 2 is a flow chart of the method for traversing the alarm information set provided by the invention;
FIG. 3 is a schematic diagram of the structure of the alarm log analysis device of the crown block system provided by the invention;
Fig. 4 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Fig. 1 is a flow chart of an analysis method of an alarm log of an overhead travelling crane system, as shown in fig. 1, the method includes:
step 110, traversing each alarm information in the alarm information set according to the time stamp sequence, constructing a word list of each alarm type and a word vector list corresponding to each alarm information, and dividing a plurality of alarm subsets based on the time stamp of each alarm information; the alarm subset consists of triplets formed by alarm type fields, id and timestamp fields of each alarm message;
Step 120, determining a timing vector of each alarm type based on the plurality of alarm subsets;
Step 130, for any alarm type, extracting a semantic vector of the any alarm type based on a word list of the any alarm type and a word vector list corresponding to alarm information belonging to the any alarm type;
Step 140, for any two alarm types, acquiring an analysis alarm sequence from the alarm information set, and determining the association strength between any two alarm types based on a word vector list corresponding to alarm information in the analysis alarm sequence and the word list of the any two alarm types;
step 150, for any two alarm types, determining whether the any two alarm types are associated alarm types based on the timing vectors, the semantic vectors and the association strength between the any two alarm types.
Specifically, by completely traversing the alarm information ordered by time stamps in the alarm information set acquired from the alarm information once, a word list of each alarm type and a word vector list corresponding to each alarm information can be constructed. The alarm type can be preset based on the operation and maintenance requirements of the actual application scene. The word list of any alarm type contains word vectors of the words contained in the alarm information belonging to the alarm type (namely, the alarm type field is the alarm type) and the frequency of occurrence of the corresponding words in the alarm information belonging to the alarm type; the word vector list corresponding to any alarm information contains the sequence number of each word in the word list of the alarm type indicated by the alarm type field of the alarm information, and it is required to be noted that the sequence of the corresponding word in the word vector list corresponding to any alarm information is consistent with the sequence of the corresponding word in the alarm description field of the alarm information. Meanwhile, in the process of traversing the alarm information sets, a plurality of alarm subsets are also divided from the alarm information sets based on the time stamp of each alarm information. Each alarm subset consists of triples formed by alarm type fields, id and time stamp fields of each alarm information, namely, the alarm subset comprises a plurality of triples, and each triplet corresponds to one alarm information.
In some embodiments, as described in FIG. 2, each alert information in the alert information set may be traversed based on the following steps:
Step 210, creating an empty vocabulary for each alarm type;
step 220, traversing each alarm information in the alarm information set according to the time stamp sequence, determining an alarm type field, a time stamp field and an alarm description field of the current alarm information aiming at the current alarm information, updating a word list of the alarm type indicated by the alarm type field of the current alarm information based on each word of the alarm description field of the current alarm information, creating a word vector list corresponding to the current alarm information, and distributing the current alarm information to an established alarm subset or a newly-built alarm subset based on the time stamp field of the current alarm information and distributing the current alarm information to the newly-built alarm subset.
Specifically, an empty vocabulary may be created for each alert type prior to formally traversing the alert information collection. In the process of traversing each alarm information ordered according to the time stamp in the alarm information set, aiming at the current traversed alarm information, an alarm type field, a time stamp field and an alarm description field of the current alarm information can be acquired from the current alarm information, and meanwhile, a globally unique id is endowed to the current alarm information; then, based on each word of the alarm description field of the current alarm information, updating a word list of the alarm type indicated by the alarm type field of the current alarm information, wherein the word list comprises word vectors of updated or newly built words in the word list (the word vectors can be obtained by using tools such as word2 vec) and occurrence frequencies, if a certain word is in the word list, the occurrence frequency of the certain word is increased by 1, otherwise, an item is newly built in the word list to correspond to the word, and the word vectors of the word are added and the occurrence frequency is set to be 1; and simultaneously, a word vector list corresponding to the current alarm information is also created.
In addition, the alarm information set is traversed while the alarm sub-set is created and updated. For the current alarm information traversed currently, the current alarm information can be distributed to the created alarm subset or the newly-built alarm subset based on the timestamp field of the current alarm information, and the current alarm information can be distributed to the newly-built alarm subset. In some embodiments, if the timestamp field of the current alert information exceeds the maximum time of the current created latest alert sub-set (i.e., the latest created alert sub-set), a new alert sub-set is created, a triplet of alert type fields, ids, and timestamp fields of the current alert information is added to the new alert sub-set, and the maximum time of the new alert sub-set is determined based on the timestamp field of the current alert information and the preset time window length (e.g., the timestamp field of the current alert information and the preset time window length may be added to obtain the maximum time of the new alert sub-set); if the time stamp field of the current alarm information does not exceed the maximum time of the current created latest alarm sub-set, adding a triplet formed by the alarm type field, the id and the time stamp field of the current alarm information into the current created latest alarm sub-set.
After the data preparation is realized by completely traversing the alarm information sequenced by the time stamp in the alarm information set acquired from the alarm information once, the time sequence vector of each alarm type can be determined based on a plurality of alarm subsets. Wherein, the time sequence vector of any alarm type represents the time sequence characteristic of the alarm information belonging to the alarm type. Because there may be some similarity in time sequence between the alarm information of the associated alarm types, extracting the time sequence vector of each alarm type helps to mine out the associated alarm types. In some embodiments, for any alarm type, the number of alarm information belonging to the alarm type included in the alarm subsets may be determined sequentially, and the timing vector of the alarm type is determined based on the number of alarm information belonging to the alarm type included in each alarm subset. For example, assuming that there are m alarm subsets, the timing vector for any alarm type a may be represented as (t 1, t2,..and tm), where ti is the number of alarm information for alarm type a in the ith alarm subset.
In addition, the alarm information of the associated alarm types may have a certain similarity in terms of semantics, so that the semantic vector of each alarm type can be extracted to improve the accuracy of the alarm type association analysis. For any alarm type, the semantic vector of the alarm type can be extracted based on the word list of the alarm type and the word vector list corresponding to the alarm information belonging to the alarm type. The semantic vector of any alarm type contains the common semantic information among the alarm information belonging to the alarm type. In some embodiments, the template text of the alarm type, that is, the part of text shared by the alarm information belonging to the alarm type, may be determined based on the vocabulary of the alarm type and the word vector list corresponding to the alarm information belonging to the alarm type, and then the semantic vector of the alarm type may be extracted from the template text of the alarm type by using a natural language model, such as a Transformer model, based on the template text of the alarm type. The method may determine, based on the frequency of each word segment in the vocabulary of the alarm type, a keyword segment in the vocabulary of the alarm type (e.g., a word segment with a frequency higher than a preset frequency value), and determine, based on a word vector list corresponding to alarm information of the alarm type, an order between the keyword segments in the vocabulary of the alarm type, thereby generating a template text of the alarm type based on the keyword segments in the vocabulary of the alarm type and the order therebetween.
In order to further improve the accuracy of alarm type association analysis, for any two alarm types, an analysis alarm sequence can be obtained from an alarm information set, and the association strength between any two alarm types can be predicted by using a neural network model based on a word vector list corresponding to alarm information in the analysis alarm sequence and word lists of the any two alarm types. Wherein, the association strength between any two alarm types is a numerical value between 0 and 1, and the larger the value is, the stronger the association between any two alarm types is. Here, based on the multiple alert subsets, a preset number of alert subsets may be acquired and combined into a subset to be analyzed; wherein the subset to be analyzed contains triples corresponding to the latest alarm information belonging to one of the two alarm types. Then, after deleting the triples of which the alarm type field in the subset to be analyzed is not any two alarm types, aiming at any triples in the subset to be analyzed, the word vector sequence corresponding to the corresponding alarm information can be spliced to the triples based on the id of the alarm information in the triples, so as to obtain the analysis alarm sequence. The word vector sequence corresponding to the corresponding alarm information is constructed based on the word vector list corresponding to the corresponding alarm information and the word list of the alarm type to which the corresponding alarm information belongs, and word vectors of corresponding word segmentation are contained. Inputting the analysis alarm sequence into a pre-trained association analysis model to obtain association strength between any two alarm types output by the association analysis model. The correlation analysis model may be any neural network model capable of processing time series data, such as a cyclic neural network or a variant thereof, which is not particularly limited in the present invention. The correlation analysis model can be trained based on the sample analysis alarm sequences of two sample alarm types and correlation labels (indicating correlation or non-correlation) of the two sample alarm types, and the sample analysis alarm sequences of the two sample alarm types are similar to the acquisition mode of the analysis alarm sequences of any two alarm types in the above description, and are not repeated here.
For any two alarm types, by integrating the time sequence vector, the semantic vector and the association strength between any two alarm types, whether any two alarm types are associated alarm types can be determined more accurately. The time sequence vector, the semantic vector and the association strength between any two alarm types can be input into the multi-layer perceptron to execute classification tasks, so that whether any two alarm types are associated alarm types or not can be determined.
In summary, the method provided by the embodiment of the invention constructs the word list of each alarm type and the word vector list corresponding to each alarm information by traversing each alarm information in the alarm information set according to the time stamp sequence, divides a plurality of alarm subsets based on the time stamp of each alarm information, then determines the time sequence vector of each alarm type based on the plurality of alarm subsets, extracts the semantic vector of the alarm type based on the word list of the alarm type and the word vector list corresponding to the alarm information belonging to the alarm type for any alarm type, acquires the analysis alarm sequence from the alarm information set for any two alarm types, determines the association strength between any two alarm types based on the word vector list corresponding to the alarm information in the analysis alarm sequence and the word list of any two alarm types, and accordingly determines whether any two alarm types are associated alarm types based on the time sequence vector of any two alarm types, semantic vector and association strength between any two alarm types, and can mine the alarm types by only once in the whole alarm information in the whole mining process, thereby improving the efficiency of the whole mining method, and judging whether the association between the two alarm types are associated alarm types accurately by combining the time sequence information.
The following describes the alarm log analysis device of the crown block system, and the alarm log analysis device of the crown block system and the alarm log analysis method of the crown block system described below can be referred to correspondingly.
Based on any of the above embodiments, fig. 3 is a schematic structural diagram of an alarm log analysis device of an overhead travelling crane system, where, as shown in fig. 3, the device includes:
The data traversing unit 310 is configured to traverse each alarm information ordered according to the time stamp in the alarm information set, construct a word list of each alarm type and a word vector list corresponding to each alarm information, and divide a plurality of alarm subsets based on the time stamp of each alarm information; the alarm subset consists of triplets formed by alarm type fields, id and timestamp fields of each alarm message;
A timing feature extraction unit 320, configured to determine a timing vector of each alarm type based on the plurality of alarm subsets;
The semantic information extracting unit 330 is configured to extract, for any alert type, a semantic vector of the any alert type based on a vocabulary of the any alert type and a list of word vectors corresponding to alert information belonging to the any alert type;
the association strength analysis unit 340 is configured to obtain an analysis alarm sequence for any two alarm types, and determine association strength between the any two alarm types based on a word vector list corresponding to alarm information in the analysis alarm sequence and the word list of the any two alarm types;
A confirmation unit 350, configured to determine, for any two alarm types, whether the any two alarm types are associated alarm types based on the timing vectors, the semantic vectors, and the association strength between the any two alarm types.
According to the device provided by the embodiment of the invention, the word list of each alarm type and the word vector list corresponding to each alarm information are constructed by traversing each alarm information in the alarm information set according to the time stamps, a plurality of alarm subsets are divided based on the time stamps of each alarm information, then the time sequence vector of each alarm type is determined based on the plurality of alarm subsets, the semantic vector of each alarm type is extracted for any alarm type based on the word list of the alarm type and the word vector list corresponding to the alarm information belonging to the alarm type, the analysis alarm sequence is acquired from the alarm information set for any two alarm types, and the association strength between any two alarm types is determined based on the word vector list corresponding to the alarm information in the analysis alarm sequence and the word list of any two alarm types, so that whether any two alarm types are associated alarm types or not can be determined based on the time sequence vectors, the semantic vectors and the association strength between any two alarm types, the whole mining process only needs to traverse the alarm information in the log base once, the mining efficiency is improved, the association accuracy of the two alarm types can be improved, and the association strength between the two alarm types can be judged by combining the time sequence characteristics of the two alarm types.
Based on any of the above embodiments, the traversing each alarm information in the alarm information set according to the time stamp order, constructing a word list of each alarm type and a word vector list corresponding to each alarm information, and dividing a plurality of alarm subsets based on the time stamp of each alarm information, specifically including:
creating an empty vocabulary for each alarm type;
Traversing each alarm information ordered by time stamp in the alarm information set, determining an alarm type field, a time stamp field and an alarm description field of the current alarm information aiming at the current alarm information, updating a word list of an alarm type indicated by the alarm type field of the current alarm information based on each word of the alarm description field of the current alarm information, creating a word vector list corresponding to the current alarm information, and distributing the current alarm information to an established alarm subset or a newly-built alarm subset based on the time stamp field of the current alarm information and distributing the current alarm information to the newly-built alarm subset;
The word list of any alarm type comprises word vectors of word segmentation contained in alarm information belonging to the any alarm type and the occurrence frequency of corresponding word segmentation in the alarm information belonging to the any alarm type; the word vector list corresponding to any alarm information comprises the serial numbers of each word of the alarm description field of the any alarm information in the word list of the alarm type indicated by the alarm type field of the any alarm information.
Based on any of the foregoing embodiments, the assigning the current alert information to the created alert subset or the newly created alert subset based on the timestamp field of the current alert information and assigning the current alert information to the newly created alert subset specifically includes:
If the timestamp field of the current alarm information exceeds the maximum time of the current created latest alarm sub-set, creating a new alarm sub-set, adding a triplet formed by the alarm type field, the id and the timestamp field of the current alarm information into the new alarm sub-set, and determining the maximum time of the new alarm sub-set based on the timestamp field of the current alarm information and a preset time window length;
and if the time stamp field of the current alarm information does not exceed the maximum time of the current created latest alarm sub-set, adding a triplet formed by the alarm type field, the id and the time stamp field of the current alarm information into the current created latest alarm sub-set.
Based on any one of the foregoing embodiments, determining a timing vector for each alarm type based on the plurality of alarm subsets specifically includes:
Sequentially determining the quantity of alarm information belonging to any alarm type contained in the plurality of alarm subsets aiming at any alarm type;
And determining a time sequence vector of any alarm type based on the quantity of alarm information belonging to the alarm type contained in each alarm subset.
Based on any one of the foregoing embodiments, the extracting, for any one of the alert types, the semantic vector of the any one of the alert types based on the vocabulary of the any one of the alert types and the list of word vectors corresponding to the alert information belonging to the any one of the alert types specifically includes:
Determining a template text of any alarm type based on the word list of any alarm type and a word vector list corresponding to alarm information belonging to the any alarm type;
And extracting the semantic vector of any alarm type based on the template text of any alarm type.
Based on any one of the foregoing embodiments, the determining, based on the vocabulary of any one of the alert types and the word vector list corresponding to the alert information belonging to the any one of the alert types, the template text of the any one of the alert types specifically includes:
determining key word segmentation in the word list of any alarm type based on the frequency of each word segmentation in the word list of any alarm type;
Determining the sequence among key segmentation words in the word list of any alarm type based on a word vector list corresponding to the alarm information of the any alarm type;
generating template text of any alarm type based on the sequence among the keyword segmentation words in the word list of the alarm type.
Based on any one of the above embodiments, the obtaining the analysis alert sequence for any two alert types, and determining the association strength between any two alert types based on the word vector list corresponding to the alert information in the analysis alert sequence and the word list of any two alert types specifically includes:
Based on the alarm subsets, acquiring a preset number of alarm subsets to be combined into a subset to be analyzed; wherein the subset to be analyzed contains the latest alarm information belonging to one of the two alarm types;
After deleting the triples of which the alarm type field is not the any two alarm types in the subset to be analyzed, aiming at any triplet in the subset to be analyzed, splicing word vector sequences corresponding to corresponding alarm information into any triplet based on the id of the alarm information in the any triplet to obtain the analysis alarm sequence; the word vector sequence corresponding to the corresponding alarm information is constructed based on the word vector list corresponding to the corresponding alarm information and the word list of the alarm type to which the corresponding alarm information belongs;
Inputting the analysis alarm sequence into a correlation analysis model to obtain the correlation strength between any two alarm types output by the correlation analysis model.
Fig. 4 is a schematic structural diagram of an electronic device according to the present invention, as shown in fig. 4, the electronic device may include: processor 410, memory 420, communication interface (Communications Interface) 430, and communication bus 440, wherein processor 410, memory 420, and communication interface 430 communicate with each other via communication bus 440. The processor 410 may invoke logic instructions in the memory 420 to perform an overhead crane system alarm log analysis method comprising: traversing each alarm information in the alarm information set according to the time stamp sequence, constructing a word list of each alarm type and a word vector list corresponding to each alarm information, and dividing a plurality of alarm subsets based on the time stamp of each alarm information; the alarm subset consists of triplets formed by alarm type fields, id and timestamp fields of each alarm message; determining a timing vector for each alarm type based on the plurality of alarm subsets; extracting semantic vectors of any alarm type based on word lists of the alarm type and word vector lists corresponding to alarm information belonging to the alarm type aiming at the alarm type; aiming at any two alarm types, acquiring an analysis alarm sequence, and determining the association strength between the any two alarm types based on a word vector list corresponding to alarm information in the analysis alarm sequence and the word list of the any two alarm types; for any two alarm types, determining whether the any two alarm types are associated alarm types based on the time sequence vectors, the semantic vectors and the association strength between the any two alarm types.
Further, the logic instructions in the memory 420 described above may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the method of analysis of an alarm log of an overhead travelling crane system provided by the methods described above, the method comprising: traversing each alarm information in the alarm information set according to the time stamp sequence, constructing a word list of each alarm type and a word vector list corresponding to each alarm information, and dividing a plurality of alarm subsets based on the time stamp of each alarm information; the alarm subset consists of triplets formed by alarm type fields, id and timestamp fields of each alarm message; determining a timing vector for each alarm type based on the plurality of alarm subsets; extracting semantic vectors of any alarm type based on word lists of the alarm type and word vector lists corresponding to alarm information belonging to the alarm type aiming at the alarm type; aiming at any two alarm types, acquiring an analysis alarm sequence, and determining the association strength between the any two alarm types based on a word vector list corresponding to alarm information in the analysis alarm sequence and the word list of the any two alarm types; for any two alarm types, determining whether the any two alarm types are associated alarm types based on the time sequence vectors, the semantic vectors and the association strength between the any two alarm types.
In still another aspect, the present invention further provides a non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor is implemented to perform the above-provided method for analyzing an alarm log of an overhead travelling crane system, the method comprising: traversing each alarm information in the alarm information set according to the time stamp sequence, constructing a word list of each alarm type and a word vector list corresponding to each alarm information, and dividing a plurality of alarm subsets based on the time stamp of each alarm information; the alarm subset consists of triplets formed by alarm type fields, id and timestamp fields of each alarm message; determining a timing vector for each alarm type based on the plurality of alarm subsets; extracting semantic vectors of any alarm type based on word lists of the alarm type and word vector lists corresponding to alarm information belonging to the alarm type aiming at the alarm type; aiming at any two alarm types, acquiring an analysis alarm sequence, and determining the association strength between the any two alarm types based on a word vector list corresponding to alarm information in the analysis alarm sequence and the word list of the any two alarm types; for any two alarm types, determining whether the any two alarm types are associated alarm types based on the time sequence vectors, the semantic vectors and the association strength between the any two alarm types.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (7)
1. The method for analyzing the alarm log of the crown block system is characterized by comprising the following steps of:
Traversing each alarm information in the alarm information set according to the time stamp sequence, constructing a word list of each alarm type and a word vector list corresponding to each alarm information, and dividing a plurality of alarm subsets based on the time stamp of each alarm information; the alarm subset consists of triplets formed by alarm type fields, id and timestamp fields of each alarm message;
determining a timing vector for each alarm type based on the plurality of alarm subsets;
Extracting semantic vectors of any alarm type based on word lists of the alarm type and word vector lists corresponding to alarm information belonging to the alarm type aiming at the alarm type;
Aiming at any two alarm types, acquiring an analysis alarm sequence, and determining the association strength between the any two alarm types based on a word vector list corresponding to alarm information in the analysis alarm sequence and the word list of the any two alarm types;
For any two alarm types, determining whether the any two alarm types are associated alarm types based on a time sequence vector, a semantic vector and association strength between the any two alarm types;
The method comprises the steps of obtaining an analysis alarm sequence aiming at any two alarm types, and determining the association strength between any two alarm types based on a word vector list corresponding to alarm information in the analysis alarm sequence and the word list of the any two alarm types, wherein the specific steps include:
Based on the alarm subsets, acquiring a preset number of alarm subsets to be combined into a subset to be analyzed; wherein the subset to be analyzed contains the latest alarm information belonging to one of the two alarm types;
After deleting the triples of which the alarm type field is not the any two alarm types in the subset to be analyzed, aiming at any triplet in the subset to be analyzed, splicing word vector sequences corresponding to corresponding alarm information into any triplet based on the id of the alarm information in the any triplet to obtain the analysis alarm sequence; the word vector sequence corresponding to the corresponding alarm information is constructed based on the word vector list corresponding to the corresponding alarm information and the word list of the alarm type to which the corresponding alarm information belongs;
inputting the analysis alarm sequence into a correlation analysis model to obtain the correlation strength between any two alarm types output by the correlation analysis model;
Traversing each alarm information in the alarm information set according to the time stamp sequence, constructing a word list of each alarm type and a word vector list corresponding to each alarm information, and dividing a plurality of alarm subsets based on the time stamp of each alarm information, wherein the method specifically comprises the following steps:
creating an empty vocabulary for each alarm type;
Traversing each alarm information ordered by time stamp in the alarm information set, determining an alarm type field, a time stamp field and an alarm description field of the current alarm information aiming at the current alarm information, updating a word list of an alarm type indicated by the alarm type field of the current alarm information based on each word of the alarm description field of the current alarm information, creating a word vector list corresponding to the current alarm information, and distributing the current alarm information to an established alarm subset or a newly-built alarm subset based on the time stamp field of the current alarm information and distributing the current alarm information to the newly-built alarm subset;
The word list of any alarm type comprises word vectors of word segmentation contained in alarm information belonging to the any alarm type and the occurrence frequency of corresponding word segmentation in the alarm information belonging to the any alarm type; the word vector list corresponding to any alarm information comprises the serial numbers of each word of the alarm description field of the any alarm information in the word list of the alarm type indicated by the alarm type field of the any alarm information;
The step of distributing the current alarm information to the created alarm subset or the newly-built alarm subset based on the timestamp field of the current alarm information and distributing the current alarm information to the newly-built alarm subset specifically comprises the following steps:
If the timestamp field of the current alarm information exceeds the maximum time of the current created latest alarm sub-set, creating a new alarm sub-set, adding a triplet formed by the alarm type field, the id and the timestamp field of the current alarm information into the new alarm sub-set, and determining the maximum time of the new alarm sub-set based on the timestamp field of the current alarm information and a preset time window length;
and if the time stamp field of the current alarm information does not exceed the maximum time of the current created latest alarm sub-set, adding a triplet formed by the alarm type field, the id and the time stamp field of the current alarm information into the current created latest alarm sub-set.
2. The method for analyzing the alarm log of the crown block system according to claim 1, wherein determining a timing vector for each alarm type based on the plurality of alarm subsets specifically comprises:
Sequentially determining the quantity of alarm information belonging to any alarm type contained in the plurality of alarm subsets aiming at any alarm type;
And determining a time sequence vector of any alarm type based on the quantity of alarm information belonging to the alarm type contained in each alarm subset.
3. The method for analyzing the alarm log of the crown block system according to claim 1, wherein the extracting the semantic vector of any alarm type based on the vocabulary of any alarm type and the word vector list corresponding to the alarm information belonging to the any alarm type specifically comprises:
Determining a template text of any alarm type based on the word list of any alarm type and a word vector list corresponding to alarm information belonging to the any alarm type;
And extracting the semantic vector of any alarm type based on the template text of any alarm type.
4. The method for analyzing the alarm log of the crown block system according to claim 3, wherein the determining the template text of any alarm type based on the word list of any alarm type and the word vector list corresponding to the alarm information belonging to the any alarm type specifically comprises:
determining key word segmentation in the word list of any alarm type based on the frequency of each word segmentation in the word list of any alarm type;
Determining the sequence among key segmentation words in the word list of any alarm type based on a word vector list corresponding to the alarm information of the any alarm type;
generating template text of any alarm type based on the sequence among the keyword segmentation words in the word list of the alarm type.
5. An alarm log analysis device for an overhead travelling crane system, comprising:
the data traversing unit is used for traversing each alarm information in the alarm information set according to the time stamp sequence, constructing a word list of each alarm type and a word vector list corresponding to each alarm information, and dividing a plurality of alarm subsets based on the time stamp of each alarm information; the alarm subset consists of triplets formed by alarm type fields, id and timestamp fields of each alarm message;
a time sequence feature extraction unit, configured to determine a time sequence vector of each alarm type based on the plurality of alarm subsets;
The semantic information extraction unit is used for extracting semantic vectors of any alarm type based on word lists of the any alarm type and word vector lists corresponding to alarm information belonging to the any alarm type aiming at the any alarm type;
the association strength analysis unit is used for acquiring an analysis alarm sequence aiming at any two alarm types, and determining association strength between any two alarm types based on a word vector list corresponding to alarm information in the analysis alarm sequence and word lists of the any two alarm types;
The confirming unit is used for determining whether any two alarm types are associated alarm types or not according to the time sequence vectors, the semantic vectors and the association strength between the any two alarm types for any two alarm types;
The method comprises the steps of obtaining an analysis alarm sequence aiming at any two alarm types, and determining the association strength between any two alarm types based on a word vector list corresponding to alarm information in the analysis alarm sequence and the word list of the any two alarm types, wherein the specific steps include:
Based on the alarm subsets, acquiring a preset number of alarm subsets to be combined into a subset to be analyzed; wherein the subset to be analyzed contains the latest alarm information belonging to one of the two alarm types;
After deleting the triples of which the alarm type field is not the any two alarm types in the subset to be analyzed, aiming at any triplet in the subset to be analyzed, splicing word vector sequences corresponding to corresponding alarm information into any triplet based on the id of the alarm information in the any triplet to obtain the analysis alarm sequence; the word vector sequence corresponding to the corresponding alarm information is constructed based on the word vector list corresponding to the corresponding alarm information and the word list of the alarm type to which the corresponding alarm information belongs;
inputting the analysis alarm sequence into a correlation analysis model to obtain the correlation strength between any two alarm types output by the correlation analysis model;
Traversing each alarm information in the alarm information set according to the time stamp sequence, constructing a word list of each alarm type and a word vector list corresponding to each alarm information, and dividing a plurality of alarm subsets based on the time stamp of each alarm information, wherein the method specifically comprises the following steps:
creating an empty vocabulary for each alarm type;
Traversing each alarm information ordered by time stamp in the alarm information set, determining an alarm type field, a time stamp field and an alarm description field of the current alarm information aiming at the current alarm information, updating a word list of an alarm type indicated by the alarm type field of the current alarm information based on each word of the alarm description field of the current alarm information, creating a word vector list corresponding to the current alarm information, and distributing the current alarm information to an established alarm subset or a newly-built alarm subset based on the time stamp field of the current alarm information and distributing the current alarm information to the newly-built alarm subset;
The word list of any alarm type comprises word vectors of word segmentation contained in alarm information belonging to the any alarm type and the occurrence frequency of corresponding word segmentation in the alarm information belonging to the any alarm type; the word vector list corresponding to any alarm information comprises the serial numbers of each word of the alarm description field of the any alarm information in the word list of the alarm type indicated by the alarm type field of the any alarm information;
The step of distributing the current alarm information to the created alarm subset or the newly-built alarm subset based on the timestamp field of the current alarm information and distributing the current alarm information to the newly-built alarm subset specifically comprises the following steps:
If the timestamp field of the current alarm information exceeds the maximum time of the current created latest alarm sub-set, creating a new alarm sub-set, adding a triplet formed by the alarm type field, the id and the timestamp field of the current alarm information into the new alarm sub-set, and determining the maximum time of the new alarm sub-set based on the timestamp field of the current alarm information and a preset time window length;
and if the time stamp field of the current alarm information does not exceed the maximum time of the current created latest alarm sub-set, adding a triplet formed by the alarm type field, the id and the time stamp field of the current alarm information into the current created latest alarm sub-set.
6. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of analyzing an alarm log of an overhead travelling crane system according to any one of claims 1 to 4 when the program is executed by the processor.
7. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the method of the crown block system alert log analysis of any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410643052.4A CN118227741B (en) | 2024-05-23 | 2024-05-23 | Crown block system alarm log analysis method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410643052.4A CN118227741B (en) | 2024-05-23 | 2024-05-23 | Crown block system alarm log analysis method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118227741A CN118227741A (en) | 2024-06-21 |
| CN118227741B true CN118227741B (en) | 2024-08-27 |
Family
ID=91502894
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410643052.4A Active CN118227741B (en) | 2024-05-23 | 2024-05-23 | Crown block system alarm log analysis method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118227741B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119718682B (en) * | 2025-02-27 | 2025-07-08 | 深圳市倍联德实业有限公司 | AI and edge computing integrated server resource scheduling system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113407507A (en) * | 2018-04-23 | 2021-09-17 | 华为技术有限公司 | Alarm log compression method, device and system and storage medium |
| CN115033688A (en) * | 2022-05-11 | 2022-09-09 | 阿里巴巴（中国）有限公司 | Method, device, equipment and storage medium for identifying alarm event type |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7949529B2 (en) * | 2005-08-29 | 2011-05-24 | Voicebox Technologies, Inc. | Mobile systems and methods of supporting natural language human-machine interactions |
| US10831711B2 (en) * | 2017-09-26 | 2020-11-10 | International Business Machines Corporation | Prioritizing log tags and alerts |
| US11514383B2 (en) * | 2019-09-13 | 2022-11-29 | Schlumberger Technology Corporation | Method and system for integrated well construction |
| US10983771B1 (en) * | 2019-11-21 | 2021-04-20 | Oracle International Corporation | Quality checking inferred types in a set of code |
| CN112003718B (en) * | 2020-09-25 | 2021-07-27 | 南京邮电大学 | A network alarm location method based on deep learning |
-
2024
- 2024-05-23 CN CN202410643052.4A patent/CN118227741B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113407507A (en) * | 2018-04-23 | 2021-09-17 | 华为技术有限公司 | Alarm log compression method, device and system and storage medium |
| CN115033688A (en) * | 2022-05-11 | 2022-09-09 | 阿里巴巴（中国）有限公司 | Method, device, equipment and storage medium for identifying alarm event type |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118227741A (en) | 2024-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10187415B2 (en) | Cognitive information security using a behavioral recognition system | |
| CN106778259B (en) | Abnormal behavior discovery method and system based on big data machine learning | |
| CN111726248A (en) | Alarm root cause positioning method and device | |
| CN112364014B (en) | Data query method, device, server and storage medium | |
| CN114465874B (en) | Fault prediction method, device, electronic equipment and storage medium | |
| CN118227741B (en) | Crown block system alarm log analysis method and device, electronic equipment and storage medium | |
| EP3789882B1 (en) | Automatic configuration of logging infrastructure for software deployments using source code | |
| CN112988509B (en) | Alarm message filtering method and device, electronic equipment and storage medium | |
| CN110147540B (en) | Method and system for generating business security requirement document | |
| CN109858280A (en) | A kind of desensitization method based on machine learning, device and desensitization equipment | |
| CN103577514A (en) | Method and apparatus automated data exploration | |
| CN111339293A (en) | Data processing method and device of alarm event and classification method of alarm event | |
| KR20070011432A (en) | Data Processing in Computerized Systems | |
| CN114666078A (en) | Method and system for detecting SQL injection attack, electronic equipment and storage medium | |
| CN115408236A (en) | Log data auditing system, method, equipment and medium | |
| CN107169092A (en) | Intelligent Recognition and the method and system of sensitive content are handled in interaction | |
| CN117350380A (en) | Abnormal behavior monitoring system and method | |
| CN113407495A (en) | SIMHASH-based file similarity determination method and system | |
| CN108616413B (en) | Information calibration method and device | |
| CN118133207B (en) | Cross-domain log anomaly detection model construction method, device, equipment and medium | |
| CN113535458A (en) | Abnormal false alarm processing method and device, storage medium and terminal | |
| CN119127093A (en) | A method, device, electronic device and storage medium for persistent storage of logs | |
| CN114465875B (en) | Fault processing method and device | |
| CN117932081A (en) | Knowledge graph-based associated retrieval method, device, equipment and medium | |
| CN116136897A (en) | Information processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |