CN117009156A - Probability-based hard disk fault analysis method and device - Google Patents
Probability-based hard disk fault analysis method and device Download PDFInfo
- Publication number
- CN117009156A CN117009156A CN202310671065.8A CN202310671065A CN117009156A CN 117009156 A CN117009156 A CN 117009156A CN 202310671065 A CN202310671065 A CN 202310671065A CN 117009156 A CN117009156 A CN 117009156A
- Authority
- CN
- China
- Prior art keywords
- hard disk
- fault
- trimming
- data
- disk fault
- 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.)
- Pending
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 197
- 238000012549 training Methods 0.000 claims abstract description 119
- 238000002372 labelling Methods 0.000 claims abstract description 31
- 238000000034 method Methods 0.000 claims abstract description 20
- 238000009966 trimming Methods 0.000 claims description 156
- 238000004590 computer program Methods 0.000 claims description 18
- 238000010276 construction Methods 0.000 claims description 16
- 238000003860 storage Methods 0.000 claims description 15
- 238000013507 mapping Methods 0.000 claims description 7
- 238000012423 maintenance Methods 0.000 abstract description 10
- 238000013473 artificial intelligence Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 18
- 238000004364 calculation method Methods 0.000 description 16
- 238000004891 communication Methods 0.000 description 16
- 230000006870 function Effects 0.000 description 14
- 238000005259 measurement Methods 0.000 description 11
- 238000012545 processing Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 5
- 239000000872 buffer Substances 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000012896 Statistical algorithm Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/22—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
- G06F11/2273—Test methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Quality & Reliability (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Debugging And Monitoring (AREA)
Abstract
The application provides a hard disk fault analysis method and device based on probability, which relate to the field of artificial intelligence and can be also used in the financial field, and comprise the following steps: constructing a hard disk fault data graph set according to hard disk operation record data; wherein, each hard disk operation record data corresponds to each point in the hard disk fault data graph set; constructing a hard disk fault analysis training set according to the position information of each point in the hard disk fault data graph set; analyzing the faults of the newly added hard disk by utilizing a pre-trained hard disk fault analysis model to obtain a fault analysis result; the hard disk fault analysis model is obtained by training the hard disk fault analysis training set. The method can construct the hard disk fault analysis model with higher accuracy based on the accuracy of the hard disk fault labeling, thereby improving the automation level of the equipment operation and maintenance.
Description
Technical Field
The application relates to the field of artificial intelligence, can be used in the financial field, in particular to a hard disk fault analysis method and device based on probability.
Background
The hard disk is used as an important medium for data storage, and provides a high-density, high-capacity and high-cost-performance storage space. In the prior art, hard disk failure prediction is a technology for judging whether a disk can fail or not based on a machine learning model, so that the data security of a hard disk storage system can be ensured, and the business influence caused by hard disk failure is avoided.
In the prior art, inaccurate hard disk fault prediction is a common technical problem in a bank data center environment. For example, when a storage type hard disk failure alarm occurs in a bank production environment, in order to ensure production, whether the hard disk is actually failed or not, people often implement replacement of the hard disk as a preferred scheme. In this case, the replaced hard disk may thus be falsely marked as faulty. If the error labeling result is used for constructing a subsequent hard disk fault analysis model, the result of fault prediction analysis according to the model in the future is further influenced, and a new error alarm is generated, so that a vicious circle is formed.
Disclosure of Invention
Aiming at the problems in the prior art, the application provides the hard disk fault analysis method and the device based on probability, which can construct a hard disk fault analysis model with higher accuracy based on the accuracy of hard disk fault labeling, thereby improving the automation level of equipment operation and maintenance.
In order to solve the technical problems, the application provides the following technical scheme:
in a first aspect, the present application provides a probability-based hard disk failure analysis method, including:
Constructing a hard disk fault data graph set according to hard disk operation record data; wherein, each hard disk operation record data corresponds to each point in the hard disk fault data graph set;
constructing a hard disk fault analysis training set according to the position information of each point in the hard disk fault data graph set;
analyzing the faults of the newly added hard disk by utilizing a pre-trained hard disk fault analysis model to obtain a fault analysis result; the hard disk fault analysis model is obtained by training the hard disk fault analysis training set.
Further, the constructing a hard disk failure data graph set according to the hard disk operation record data includes:
retrieving the hard disk operation record data to obtain hard disk operation record data containing fault records as hard disk fault record data; wherein the hard disk fault record data comprises a hard disk fault type;
and mapping the hard disk fault record data to a multidimensional space coordinate system to obtain the hard disk fault data atlas.
Further, the constructing a hard disk failure analysis training set according to the position information of each point in the hard disk failure data graph set includes:
determining the number of the trimming edges and the trimming distance corresponding to each point according to the position information of each point in the hard disk fault data graph set;
Judging whether the hard disk fault record data is correct or not according to the trimming quantity and the trimming distance corresponding to each point;
and constructing the hard disk fault analysis training set according to the determined hard disk fault record data.
Further, the judging whether the hard disk fault record data is correct according to the trimming quantity and the trimming distance corresponding to each point includes:
determining the trimming weight and measured value corresponding to each point according to the trimming quantity and trimming distance corresponding to each point;
determining trimming weights and random values corresponding to each point according to the hard disk fault types and the trimming weights;
and judging whether the hard disk fault record data is correct or not according to the trimming weight and the measured value and the trimming weight and the random value.
Further, the determining the trimming weight and the measured value corresponding to each point according to the trimming number and the trimming distance corresponding to each point includes:
calculating the trimming weight according to the trimming distance;
and calculating the trimming weight and the measured value according to the trimming quantity and the trimming weight.
Further, determining the trimming weight and the random value corresponding to each point according to the hard disk fault type and the trimming weight, including:
Calculating the corresponding fault type duty ratio according to the hard disk fault type;
and calculating the trimming weight and the random value according to the fault type duty ratio.
Further, the determining whether the hard disk fault record data is correct according to the trimming weight and the measured value and the trimming weight and the random value includes:
if the trimming weight and the measured value belong to a first preset area according to the trimming weight and the random value, judging that the hard disk fault record data are correct;
and if the trimming weight and the measured value belong to a second preset area according to the trimming weight and the random value, judging that the hard disk fault record data is wrong.
Further, the constructing the hard disk failure analysis training set according to the determined hard disk failure record data includes:
directly incorporating the hard disk fault record data determined to be correct into the hard disk fault analysis training set;
and after the labeling result of the hard disk fault record data which is judged to be in error is subjected to the negation operation, the hard disk fault record data is included in the hard disk fault analysis training set.
Further, the step of pre-training the hard disk failure analysis model includes:
inputting the hard disk fault analysis training set into an extreme gradient lifting tree for training to obtain a hard disk fault analysis initial model;
And continuously updating the hard disk fault analysis training set, and optimizing parameters of the hard disk fault analysis initial model by using the updated hard disk fault analysis training set to obtain the hard disk fault analysis model.
In a second aspect, the present application provides a probability-based hard disk failure analysis apparatus, including:
the fault atlas construction unit is used for constructing a hard disk fault data atlas according to the hard disk operation record data; wherein, each hard disk operation record data corresponds to each point in the hard disk fault data graph set;
the training set determining unit is used for constructing a hard disk fault analysis training set according to the position information of each point in the hard disk fault data graph set;
the hard disk fault analysis unit is used for analyzing the newly added hard disk faults by utilizing a pre-trained hard disk fault analysis model to obtain a fault analysis result; the hard disk fault analysis model is obtained by training the hard disk fault analysis training set.
Further, the failure atlas construction unit includes:
the record data retrieval module is used for retrieving the hard disk operation record data to obtain hard disk operation record data containing fault records as hard disk fault record data; wherein the hard disk fault record data comprises a hard disk fault type;
And the multidimensional mapping module is used for mapping the hard disk fault record data into a multidimensional space coordinate system to obtain the hard disk fault data atlas.
Further, the training set determining unit includes:
the trimming determining module is used for determining the trimming quantity and the trimming distance corresponding to each point according to the position information of each point in the hard disk fault data graph set;
the fault data judging module is used for judging whether the fault record data of the hard disk is correct or not according to the trimming quantity and the trimming distance corresponding to each point;
the training set construction module is used for constructing the hard disk fault analysis training set according to the judged hard disk fault record data.
Further, the fault data judging module includes:
the measured value calculation sub-module is used for determining the trimming weight and the measured value corresponding to each point according to the trimming quantity and the trimming distance corresponding to each point;
the random value calculation sub-module is used for determining the trimming weight and the random value corresponding to each point according to the hard disk fault type and the trimming weight;
and the fault data judging sub-module is used for judging whether the hard disk fault record data is correct or not according to the trimming weight and the measured value and the trimming weight and the random value.
Further, the measurement value calculation sub-module includes:
the trimming weight determining submodule is used for calculating the trimming weight according to the trimming distance;
and the measured value determining submodule is used for calculating the trimming weight and the measured value according to the trimming quantity and the trimming weight.
Further, the random value calculation submodule includes:
the type duty ratio calculation submodule is used for calculating the corresponding fault type duty ratio according to the fault type of the hard disk;
and the random value determining submodule is used for calculating the trimming weight and the random value according to the fault type duty ratio.
Further, the fault data judging sub-module includes:
the data correct judging sub-module is used for judging that the hard disk fault record data is correct if the trimming weight and the measured value belong to a first preset area according to the trimming weight and the random value;
and the data error judging sub-module is used for judging that the hard disk fault records data errors if the trimming weight and the measured value belong to a second preset area according to the trimming weight and the random value.
Further, the training set construction module includes:
the correct data inclusion module is used for directly incorporating the hard disk fault record data which are judged to be correct into the hard disk fault analysis training set;
And the error data reverse module is used for taking the hard disk fault analysis training set after carrying out reverse operation on the labeling result of the hard disk fault record data which is judged to be in error.
Further, the hard disk failure analysis unit includes:
the initial model building module is used for inputting the hard disk fault analysis training set into an extreme gradient lifting tree for training to obtain a hard disk fault analysis initial model;
and the analysis model construction module is used for continuously updating the hard disk fault analysis training set, optimizing parameters of the hard disk fault analysis initial model by using the updated hard disk fault analysis training set, and obtaining the hard disk fault analysis model.
In a third aspect, the present application provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the probability-based hard disk failure analysis method when the program is executed.
In a fourth aspect, the present application provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the probability-based hard disk failure analysis method.
In a fifth aspect, the present application provides a computer program product comprising computer programs/instructions which, when executed by a processor, implement the steps of the probability-based hard disk failure analysis method.
Aiming at the problems in the prior art, the probability-based hard disk fault analysis method and device provided by the application can label the fault hard disk data set of the bank data center in an optimized mode, and further screen out data with high labeling accuracy based on a probability statistics principle to be used as a training set of a hard disk fault analysis model for model training, so that the problem of low labeling accuracy of the data of the model training set in the prior art is solved, the model training effect is optimized, the accuracy of the hard disk fault analysis is improved, and the automation level of equipment operation and maintenance is improved.
Drawings
In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a probability-based hard disk failure analysis method in an embodiment of the application;
FIG. 2 is a flow chart of constructing a hard disk failure data set in an embodiment of the present application;
FIG. 3 is a flowchart of determining a hard disk failure analysis training set according to an embodiment of the present application;
FIG. 4 is a flowchart of determining hard disk failure record data according to an embodiment of the present application;
FIG. 5 is a flow chart of measurement determination in an embodiment of the application;
FIG. 6 is a flowchart of random value determination in an embodiment of the present application;
FIG. 7 is a second flowchart of determining hard disk failure record data according to an embodiment of the present application;
FIG. 8 is a flowchart of a hard disk failure analysis training set constructed in accordance with an embodiment of the present application;
FIG. 9 is a flowchart of training a hard disk failure analysis model in an embodiment of the present application;
FIG. 10 is a block diagram of a probability-based hard disk failure analysis apparatus in an embodiment of the present application;
FIG. 11 is a block diagram of a failure atlas construction unit in an embodiment of the present application;
FIG. 12 is a block diagram of a training set determination unit in an embodiment of the present application;
FIG. 13 is a block diagram of a failure data determination module according to an embodiment of the present application;
FIG. 14 is a block diagram of a measurement calculation sub-module in an embodiment of the application;
FIG. 15 is a block diagram of a random value calculation sub-module in an embodiment of the application;
FIG. 16 is a block diagram of a failure data determination sub-module in accordance with an embodiment of the present application;
FIG. 17 is a block diagram of a training set construction module in an embodiment of the present application;
FIG. 18 is a block diagram of a hard disk failure analysis unit in an embodiment of the present application;
fig. 19 is a schematic structural diagram of an electronic device in an embodiment of the application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
It should be noted that the probability-based hard disk failure analysis method and device provided by the application can be used in the financial field and any field except the financial field, and the application field of the probability-based hard disk failure analysis method and device provided by the application is not limited.
According to the technical scheme, the acquisition, storage, use, processing and the like of the data meet the relevant regulations of laws and regulations.
In an embodiment, referring to fig. 1, in order to construct a hard disk failure analysis model with higher accuracy based on the accuracy of hard disk failure labeling, so as to improve the automation level of equipment operation and maintenance, the application provides a probability-based hard disk failure analysis method, which includes:
s101: constructing a hard disk fault data graph set according to hard disk operation record data; wherein, each hard disk operation record data corresponds to each point in the hard disk fault data graph set;
s102: constructing a hard disk fault analysis training set according to the position information of each point in the hard disk fault data graph set;
s103: analyzing the faults of the newly added hard disk by utilizing a pre-trained hard disk fault analysis model to obtain a fault analysis result; the hard disk fault analysis model is obtained by training the hard disk fault analysis training set.
It can be understood that the embodiment of the application relates to the field of artificial intelligence, the field of equipment operation and maintenance and the field of application of probability science, optimizes a fault hard disk data set labeling method of a bank data center based on probability statistics, and can solve the problems of poor training effect and low analysis precision of a fault analysis model caused by large fault data labeling error in the prior art. The term "labeling" is mainly used to refer to a process of labeling characteristics such as a hard disk that has failed during operation and a corresponding failure type.
Specifically, first, in order to label the hard disk failure data, hard disk operation record data needs to be acquired first, and then a hard disk failure data graph set is constructed according to the hard disk operation record data. And each hard disk operation record data is each point in the hard disk fault data graph set. It should be noted that, in the embodiment of the present application, instead of the method of recording data in a table format in the prior art, an atlas method is innovatively used to summarize hard disk operation record data. Thus, historically all hard disk drive operational record data can be fully represented in the atlas. Besides, the hard disk operation record data is recorded in an atlas form, and the subsequent fault labeling accuracy judging process through a probability statistical algorithm can be further supported.
Next, after the hard disk failure data graph set is established, a hard disk failure analysis training set may be determined according to the position information of each point in the hard disk failure data graph set. The step is corresponding to the process of realizing the accuracy judgment of the fault marking of the hard disk by using a statistical algorithm. In this step, the positional information of each point may include at least the coordinate position of the point. The distance between the points can be determined by acquiring the position information of the points, so that whether the labeling results corresponding to the points are close or not can be analyzed, and whether the labeling results corresponding to the points are accurate or not can be analyzed. The correct labeling result can be directly incorporated into the training set of the hard disk fault analysis model; the labeling result determined to be wrong is firstly subjected to the operation of 'reverse' (reversing the conclusion about whether the fault exists) and then can be included in the training set of the hard disk fault analysis model, so that the accuracy of the labeling structure is ensured to the greatest extent.
Finally, analyzing the newly added hard disk faults by utilizing a pre-trained hard disk fault analysis model to obtain a fault analysis result; the hard disk failure analysis model is obtained by training a hard disk failure analysis training set, and a specific training method is described in detail below.
From the description, the probability-based hard disk fault analysis method provided by the application can label the fault hard disk data set of the bank data center in an optimized mode, and further screen out data with high labeling accuracy based on a probability statistics principle to be used as a training set of a hard disk fault analysis model for model training, so that the problem of low labeling accuracy of the data of the model training set in the prior art is solved, the model training effect is optimized, the accuracy of the hard disk fault analysis is improved, and the automation level of equipment operation and maintenance is improved.
Steps S101 to S103 are described in detail below.
Step S101: constructing a hard disk fault data graph set according to hard disk operation record data; and each hard disk operation record data corresponds to each point in the hard disk fault data graph set.
Fig. 2 is a schematic diagram of an embodiment of a method for implementing probability-based hard disk failure analysis according to the present application.
In one embodiment, referring to fig. 2, the constructing a hard disk failure data graph set according to hard disk operation record data includes:
s201: retrieving the hard disk operation record data to obtain hard disk operation record data containing fault records as hard disk fault record data; wherein the hard disk fault record data comprises a hard disk fault type;
s202: and mapping the hard disk fault record data to a multidimensional space coordinate system to obtain the hard disk fault data atlas.
It is understood that the hard disk operational status data (also referred to as hard disk operational record data) may be SMART data of the hard disk. Typically, each hard disk records a plurality of data each day to describe the operation state of the hard disk, and each data includes a plurality of current attributes, such as a remapped sector number, a write error rate, a parity error rate, and the like.
The embodiment of the application takes any SMART data of any hard disk as a data setIf any SMART data predicts faults, the hard disk faults, and if all SMART data of the same hard disk are healthy, the hard disk is healthy. Further, the data structure of the hard disk failure data set adopts a graph structure, and therefore, the hard disk failure data set is also referred to as a hard disk failure data graph set. Further, all SMART data of all hard disks in the hard disk fault data graph set are expressed as A in a multi-dimensional space through a form of points in a coordinate system m . Where m represents m points, i.e., m SMART data, A m Representing the mth point.
From the above description, the probability-based hard disk fault analysis method provided by the application can construct a hard disk fault data graph set according to hard disk operation record data.
In one embodiment, referring to fig. 3, the constructing a hard disk failure analysis training set according to the location information of each point in the hard disk failure data graph set includes:
s301: determining the number of the trimming edges and the trimming distance corresponding to each point according to the position information of each point in the hard disk fault data graph set;
s302: judging whether the hard disk fault record data is correct or not according to the trimming quantity and the trimming distance corresponding to each point; specifically, referring to fig. 4, the determining whether the hard disk fault record data is correct according to the trimming number and the trimming distance corresponding to each point includes: determining trimming weights and trimming weight and measured values corresponding to each point according to the trimming quantity and trimming distance corresponding to each point (S401); determining trimming weights and random values corresponding to each point according to the hard disk fault type and the trimming weights (S402); judging whether the hard disk fault record data is correct or not according to the trimming weight and the measured value and the trimming weight and the random value (S403);
It can be understood that the embodiment of the application can show the trimming quantity and the trimming distance of each SMART data of each hard disk by a method of a graph. Suppose there is A in the graph i And A j Two points (i.e. two different SMART data),representation ofA i And A j The distance between the two points (i.e., the difference between the two SMART data). When A is i And A j When the labels of the two SMART data are the same (the label results of the two SMART data are the same), then A is as follows j As A i Will A i And A j Forming an edge therebetween. When A is i And A j When the labels of the edges are different, the edge is called a trimming edge. The embodiment of the application judges whether the marking of the hard disk fault data is accurate or not by judging the trimming quantity and the trimming distance of the SMART data.
Logically, distances between points labeled the same are relatively similar. Therefore, by judging whether one point periphery is marked with the same point count or different point counts, it is also necessary to calculate the distance between them as a basis for the subsequent calculation.
In one embodiment, referring to fig. 5, the determining the trimming weight and the measured value corresponding to each point according to the trimming number and the trimming distance corresponding to each point includes:
s501: calculating the trimming weight according to the trimming distance;
S502: and calculating the trimming weight and the measured value according to the trimming quantity and the trimming weight.
It can be appreciated that the embodiment of the present application sums the trimming weights of SMART data (single) for quantization, and the formula is as follows:
wherein w is ij Weights representing point i (SMART data) to point j trimming, d ij Representing the distance from point i (SMART data) to the edge of point j, S i Represents the sum of all trimming weights (measured values) for point i, n j Representing a total of n j Point, I i (j) Indicating whether the points i to j are trimming or not; if it is a cutThe edge is 1 and if not the trim is 0.
S303: constructing the hard disk fault analysis training set according to the determined hard disk fault record data; specifically, referring to fig. 8, the constructing the hard disk failure analysis training set according to the determined hard disk failure record data includes: directly incorporating the hard disk failure record data determined to be correct into the hard disk failure analysis training set (S801); and after the labeling result of the hard disk fault record data determined to be in error is subjected to the negation operation, the hard disk fault record data is included in the hard disk fault analysis training set (S802).
It will be appreciated that the same class of SMART data is aggregated together in the figure due to data similarity. Therefore, the higher the ratio of different types of examples in the adjacent points of the SMART data, the greater the probability that the annotation of the SMART data is inaccurate, i.e. the probability that the hard disk is annotated with errors. Meanwhile, since the example with accurate annotation should be closer to the example of the same kind and farther from the example of different kind, the smaller the trimming distance of the example is, the greater the probability of inaccurate annotation of SMART data, namely the probability of the hard disk being annotated with errors is.
In one embodiment, referring to fig. 6, determining the trimming weight and the random value corresponding to each point according to the hard disk fault type and the trimming weight includes:
s601: calculating the corresponding fault type duty ratio according to the hard disk fault type;
s602: and calculating the trimming weight and the random value according to the fault type duty ratio.
It will be appreciated that from the analysis of the number of cut edges described above, I i (j) Is subject to p i Is a binomial distribution of (a). Wherein p is i Is the duty cycle of the different classes of example i in the dataset.
Since the number of trimming and the distribution of trimming weights are random, S i Is a random variable (also called random value) which obeys normal distribution, namely
Where μ is the desired, σ 2 Is the variance;
in one embodiment, referring to fig. 7, the determining whether the hard disk fault record data is correct according to the trimming weight and the measured value and the trimming weight and the random value includes:
s701: if the trimming weight and the measured value belong to a first preset area according to the trimming weight and the random value, judging that the hard disk fault record data are correct;
s702: and if the trimming weight and the measured value belong to a second preset area according to the trimming weight and the random value, judging that the hard disk fault record data is wrong.
It will be appreciated that the embodiment of the application can measure S i At random variable S i The positions in (a) are divided into three categories:
class a: measurement S i The number of the trimming edges is small and the trimming distance is short in the sigma left domain, and the trimming edges are regarded as accurate data, namely the marking of the peripheral data of the SMART data is basically the same as the marking of the SMART data, so that the SMART data is regarded as accurate;
class B: measurement S i In the-sigma and sigma domains, as the trimming weight ratio is slightly larger than the ratio of different examples of the example i in the data set, whether the mark is accurate or not cannot be judged, so that the mark of surrounding data of the SMART data is abandoned and is different from the mark of the SMART data to a certain extent, and whether the mark of the SMART data is accurate or not cannot be judged;
class C: measurement S i Since the trimming weight ratio of the example i is far greater than that of the different examples of the example i in the data set, and the periphery of the example i is basically a non-homogeneous point, the labeling of the example i is inaccurate, and the labeling of the example i is corrected. The surrounding data of the SMART data is marked substantially differently from the surrounding data itself, so that the SMART data is marked incorrectly, and thus the SMA is considered to beRT data marking errors need to be reversely modified;
And combining the data accurately marked in the class A with the corrected (inverted) data marked in the class C to form a new data set.
From the above description, the probability-based hard disk fault analysis method provided by the application can construct a hard disk fault analysis training set according to the position information of each point in the hard disk fault data graph set.
In one embodiment, referring to fig. 9, the step of pre-training a hard disk failure analysis model includes:
s901: inputting the hard disk fault analysis training set into an extreme gradient lifting tree for training to obtain a hard disk fault analysis initial model;
s902: and continuously updating the hard disk fault analysis training set, and optimizing parameters of the hard disk fault analysis initial model by using the updated hard disk fault analysis training set to obtain the hard disk fault analysis model.
It can be understood that the constructed hard disk failure analysis training set is put into an extreme gradient lifting tree (also called XGBoost algorithm model) for training, and model parameters are corrected, so that an initial hard disk failure analysis model is obtained, and the model can also be used for failure analysis of new data. Preferably, in order to further improve the accuracy of the analysis of the hard disk failure analysis model, the hard disk failure analysis training set may be continuously updated in an iterative manner, and parameters of the hard disk failure analysis initial model are optimized by using the updated hard disk failure analysis training set, so as to obtain the hard disk failure analysis model. The fault result obtained through the optimized model analysis can further back feed the hard disk fault analysis training set, so that the subsequent model training is facilitated, and a virtuous circle is formed. The training method of the neural network can be seen in the prior art.
From the above description, the probability-based hard disk fault analysis method provided by the application can train the hard disk fault analysis model in advance.
Based on the same inventive concept, the embodiment of the present application also provides a probability-based hard disk failure analysis device, which can be used to implement the method described in the above embodiment, as described in the following embodiment. Because the principle of solving the problem of the hard disk failure analysis device based on the probability is similar to that of the hard disk failure analysis method based on the probability, the implementation of the hard disk failure analysis device based on the probability can be referred to the implementation of the determination method based on the software performance benchmark, and the repetition is omitted. As used below, the term "unit" or "module" may be a combination of software and/or hardware that implements the intended function. While the system described in the following embodiments is preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
In an embodiment, referring to fig. 10, in order to construct a hard disk failure analysis model with higher accuracy based on the accuracy of hard disk failure labeling, so as to improve the automation level of equipment operation and maintenance, the application provides a probability-based hard disk failure analysis device, which comprises: failure atlas construction section 1001, training set determination section 1002, and hard disk failure analysis section 1003.
A failure atlas construction unit 1001, configured to construct a hard disk failure data atlas according to hard disk operation record data; wherein, each hard disk operation record data corresponds to each point in the hard disk fault data graph set;
a training set determining unit 1002, configured to construct a hard disk failure analysis training set according to the position information of each point in the hard disk failure data graph set;
a hard disk failure analysis unit 1003, configured to analyze the newly added hard disk failure by using a hard disk failure analysis model trained in advance, so as to obtain a failure analysis result; the hard disk fault analysis model is obtained by training the hard disk fault analysis training set.
In one embodiment, referring to fig. 11, the fault atlas construction unit 1001 includes: the log data retrieval module 1101 and the multidimensional mapping module 1102.
A record data retrieving module 1101, configured to retrieve the hard disk operation record data, and obtain hard disk operation record data containing a fault record as hard disk fault record data; wherein the hard disk fault record data comprises a hard disk fault type;
and the multidimensional mapping module 1102 is configured to map the hard disk failure record data to a multidimensional space coordinate system to obtain the hard disk failure data atlas.
In one embodiment, referring to fig. 12, the training set determining unit 1002 includes: the training set comprises a trimming determining module 1201, a fault data judging module 1202 and a training set constructing module 1203.
The trimming determining module 1201 is configured to determine the number of trimming and the trimming distance corresponding to each point according to the position information of each point in the hard disk fault data graph set;
the fault data judging module 1202 is configured to judge whether the hard disk fault record data is correct according to the number of trimming edges and the trimming distance corresponding to each point;
the training set construction module 1203 is configured to construct the hard disk failure analysis training set according to the determined hard disk failure record data.
In one embodiment, referring to fig. 13, the fault data determining module 1202 includes: measurement value calculation submodule 1301, random value calculation submodule 1302, and failure data judgment submodule 1303.
The measured value calculation submodule 1301 is configured to determine an edge trimming weight and a measured value corresponding to each point according to the number of edge trimming and the edge trimming distance corresponding to each point;
a random value calculation submodule 1302, configured to determine a trimming weight and a random value corresponding to each point according to the hard disk fault type and the trimming weight;
And the fault data judging submodule 1303 is used for judging whether the hard disk fault record data is correct or not according to the trimming weight and the measured value and the trimming weight and the random value.
In one embodiment, referring to fig. 14, the measurement value calculation sub-module 1301 includes: trimming weight determination submodule 1401 and measurement determination submodule 1402.
A trimming weight determination submodule 1401, configured to calculate the trimming weight according to the trimming distance;
a measurement value determining submodule 1402 is configured to calculate the trimming weight and the measurement value according to the trimming number and the trimming weight.
In one embodiment, referring to fig. 15, the random value calculation submodule 1302 includes: the type duty calculation submodule 1501 and the random value determination submodule 1502.
A type duty ratio calculation submodule 1501, configured to calculate a corresponding fault type duty ratio according to the hard disk fault type;
a random value determination submodule 1502 is configured to calculate the trimming weight and the random value according to the fault type duty cycle.
In one embodiment, referring to fig. 16, the fault data determination submodule 1303 includes: data correct determination submodule 1601 and data error determination submodule 1602.
A data correct judging sub-module 1601, configured to determine that the hard disk fault record data is correct if it is determined that the trimming weight and the measured value belong to a first preset area according to the trimming weight and the random value;
and the data error judging sub-module 1602 is configured to judge that the hard disk fault records data errors if the trimming weight and the measured value belong to a second preset area according to the trimming weight and the random value.
In one embodiment, referring to fig. 17, the training set construction module 1203 includes: the correct data inclusion module 1701 and the error data inversion module 1702.
The correct data inclusion module 1701 is configured to directly include the hard disk fault record data determined to be correct into the hard disk fault analysis training set;
the error data reversing module 1702 is configured to take the hard disk failure analysis training set after reversing the labeling result of the hard disk failure record data determined to be in error.
In one embodiment, referring to fig. 18, the hard disk failure analysis unit 1003 includes: the initial model construction module 1801 and the analytical model construction module 1802.
The initial model building module 1801 is configured to input the hard disk failure analysis training set into an extreme gradient lifting tree for training, so as to obtain a hard disk failure analysis initial model;
The analysis model construction module 1802 is configured to continuously update the hard disk failure analysis training set, and optimize parameters of the hard disk failure analysis initial model by using the updated hard disk failure analysis training set to obtain the hard disk failure analysis model.
In order to construct a hard disk fault analysis model with higher accuracy based on the accuracy of hard disk fault labeling in terms of hardware level, thereby improving the automation level of equipment operation and maintenance, the application provides an embodiment of an electronic device for realizing all or part of contents in the probability-based hard disk fault analysis method, wherein the electronic device specifically comprises the following contents:
a Processor (Processor), a Memory (Memory), a communication interface (Communications Interface), and a bus; the processor, the memory and the communication interface complete communication with each other through the bus; the communication interface is used for realizing information transmission between the probability-based hard disk fault analysis device and related equipment such as a core service system, a user terminal, a related database and the like; the logic controller may be a desktop computer, a tablet computer, a mobile terminal, etc., and the embodiment is not limited thereto. In this embodiment, the logic controller may refer to the embodiment of the probability-based hard disk failure analysis method and the embodiment of the probability-based hard disk failure analysis device in the embodiments, and the contents thereof are incorporated herein, and the repetition is omitted.
It is understood that the user terminal may include a smart phone, a tablet electronic device, a network set top box, a portable computer, a desktop computer, a Personal Digital Assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Wherein, intelligent wearing equipment can include intelligent glasses, intelligent wrist-watch, intelligent bracelet etc..
In practical applications, part of the hard disk fault analysis method based on probability may be executed on the electronic device side as described above, or all operations may be completed in the client device. Specifically, the selection may be made according to the processing capability of the client device, and restrictions of the use scenario of the user. The application is not limited in this regard. If all operations are performed in the client device, the client device may further include a processor.
The client device may have a communication module (i.e. a communication unit) and may be connected to a remote server in a communication manner, so as to implement data transmission with the server. The server may include a server on the side of the task scheduling center, and in other implementations may include a server on an intermediate platform, such as a server on a third party server platform having a communication link with the task scheduling center server. The server may include a single computer device, a server cluster formed by a plurality of servers, or a server structure of a distributed device.
Fig. 19 is a schematic block diagram of a system configuration of an electronic device 9600 according to an embodiment of the present application. As shown in fig. 19, the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. Notably, this fig. 19 is exemplary; other types of structures may also be used in addition to or in place of the structures to implement telecommunications functions or other functions.
In one embodiment, the probability-based hard disk failure analysis method functionality may be integrated into the central processor 9100. The central processor 9100 may be configured to perform the following control:
s101: constructing a hard disk fault data graph set according to hard disk operation record data; wherein, each hard disk operation record data corresponds to each point in the hard disk fault data graph set;
s102: constructing a hard disk fault analysis training set according to the position information of each point in the hard disk fault data graph set;
s103: analyzing the faults of the newly added hard disk by utilizing a pre-trained hard disk fault analysis model to obtain a fault analysis result; the hard disk fault analysis model is obtained by training the hard disk fault analysis training set.
From the description, the probability-based hard disk fault analysis method provided by the application can label the fault hard disk data set of the bank data center in an optimized mode, and further screen out data with high labeling accuracy based on a probability statistics principle to be used as a training set of a hard disk fault analysis model for model training, so that the problem of low labeling accuracy of the data of the model training set in the prior art is solved, the model training effect is optimized, the accuracy of the hard disk fault analysis is improved, and the automation level of equipment operation and maintenance is improved.
In another embodiment, the probability-based hard disk failure analysis apparatus may be configured separately from the central processor 9100, for example, the probability-based hard disk failure analysis apparatus of the data composite transmission apparatus may be configured as a chip connected to the central processor 9100, and the function of the probability-based hard disk failure analysis method is implemented by control of the central processor.
As shown in fig. 19, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is noted that the electronic device 9600 need not include all of the components shown in fig. 19; in addition, the electronic device 9600 may further include components not shown in fig. 19, and reference may be made to the related art.
As shown in fig. 19, the central processor 9100, sometimes also referred to as a controller or operational control, may include a microprocessor or other processor device and/or logic device, which central processor 9100 receives inputs and controls the operation of the various components of the electronic device 9600.
The memory 9140 may be, for example, one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, or other suitable device. The information about failure may be stored, and a program for executing the information may be stored. And the central processor 9100 can execute the program stored in the memory 9140 to realize information storage or processing, and the like.
The input unit 9120 provides input to the central processor 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used for displaying display objects such as images and characters. The display may be, for example, but not limited to, an LCD display.
The memory 9140 may be a solid state memory such as Read Only Memory (ROM), random Access Memory (RAM), SIM card, etc. But also a memory which holds information even when powered down, can be selectively erased and provided with further data, an example of which is sometimes referred to as EPROM or the like. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application/function storage portion 9142, the application/function storage portion 9142 storing application programs and function programs or a flow for executing operations of the electronic device 9600 by the central processor 9100.
The memory 9140 may also include a data store 9143, the data store 9143 for storing data, such as contacts, digital data, pictures, sounds, and/or any other data used by an electronic device. The driver storage portion 9144 of the memory 9140 may include various drivers of the electronic device for communication functions and/or for performing other functions of the electronic device (e.g., messaging applications, address book applications, etc.).
The communication module 9110 is a transmitter/receiver 9110 that transmits and receives signals via an antenna 9111. The communication module (transmitter/receiver) 9110 is coupled to the central processor 9100 to provide an input signal and receive an output signal, which may be the same as in the case of a conventional mobile communication terminal.
Based on different communication technologies, a plurality of communication modules 9110, such as a cellular network module, a bluetooth module, and/or a wireless lan module, may be provided in the same electronic device. The communication module (transmitter/receiver) 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide audio output via the speaker 9131 and to receive audio input from the microphone 9132 to implement usual telecommunications functions. The audio processor 9130 can include any suitable buffers, decoders, amplifiers and so forth. In addition, the audio processor 9130 is also coupled to the central processor 9100 so that sound can be recorded locally through the microphone 9132 and sound stored locally can be played through the speaker 9131.
The embodiment of the present application further provides a computer readable storage medium capable of implementing all steps in the probability-based hard disk failure analysis method in which the execution subject is a server or a client, and the computer readable storage medium stores thereon a computer program that when executed by a processor implements all steps in the probability-based hard disk failure analysis method in which the execution subject is a server or a client, for example, the processor implements the following steps when executing the computer program:
S101: constructing a hard disk fault data graph set according to hard disk operation record data; wherein, each hard disk operation record data corresponds to each point in the hard disk fault data graph set;
s102: constructing a hard disk fault analysis training set according to the position information of each point in the hard disk fault data graph set;
s103: analyzing the faults of the newly added hard disk by utilizing a pre-trained hard disk fault analysis model to obtain a fault analysis result; the hard disk fault analysis model is obtained by training the hard disk fault analysis training set.
From the description, the probability-based hard disk fault analysis method provided by the application can label the fault hard disk data set of the bank data center in an optimized mode, and further screen out data with high labeling accuracy based on a probability statistics principle to be used as a training set of a hard disk fault analysis model for model training, so that the problem of low labeling accuracy of the data of the model training set in the prior art is solved, the model training effect is optimized, the accuracy of the hard disk fault analysis is improved, and the automation level of equipment operation and maintenance is improved.
It will be apparent to those skilled in the art that embodiments of the present application may be provided as a method, apparatus, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The principles and embodiments of the present invention have been described in detail with reference to specific examples, which are provided to facilitate understanding of the method and core ideas of the present invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present invention, the present description should not be construed as limiting the present invention in view of the above.
Claims (13)
1. The hard disk fault analysis method based on probability is characterized by comprising the following steps of:
constructing a hard disk fault data graph set according to hard disk operation record data; wherein, each hard disk operation record data corresponds to each point in the hard disk fault data graph set;
constructing a hard disk fault analysis training set according to the position information of each point in the hard disk fault data graph set;
Analyzing the faults of the newly added hard disk by utilizing a pre-trained hard disk fault analysis model to obtain a fault analysis result; the hard disk fault analysis model is obtained by training the hard disk fault analysis training set.
2. The method for analyzing hard disk failure based on probability of claim 1, wherein the constructing a hard disk failure data graph set from hard disk operation record data comprises:
retrieving the hard disk operation record data to obtain hard disk operation record data containing fault records as hard disk fault record data; wherein the hard disk fault record data comprises a hard disk fault type;
and mapping the hard disk fault record data to a multidimensional space coordinate system to obtain the hard disk fault data atlas.
3. The method for analyzing hard disk failure based on probability of claim 2, wherein constructing a hard disk failure analysis training set based on the position information of each point in the hard disk failure data graph set comprises:
determining the number of the trimming edges and the trimming distance corresponding to each point according to the position information of each point in the hard disk fault data graph set;
judging whether the hard disk fault record data is correct or not according to the trimming quantity and the trimming distance corresponding to each point;
And constructing the hard disk fault analysis training set according to the determined hard disk fault record data.
4. The method for analyzing hard disk failure based on probability according to claim 3, wherein said determining whether the hard disk failure record data is correct according to the number of trimmings and the trimming distance corresponding to each point comprises:
determining the trimming weight and measured value corresponding to each point according to the trimming quantity and trimming distance corresponding to each point;
determining trimming weights and random values corresponding to each point according to the hard disk fault types and the trimming weights;
and judging whether the hard disk fault record data is correct or not according to the trimming weight and the measured value and the trimming weight and the random value.
5. The method for analyzing hard disk failure based on probability according to claim 4, wherein determining the trimming weights and the measured values corresponding to each point according to the trimming number and the trimming distance corresponding to each point comprises:
calculating the trimming weight according to the trimming distance;
and calculating the trimming weight and the measured value according to the trimming quantity and the trimming weight.
6. The probability-based hard disk failure analysis method according to claim 4, wherein determining the trimming weight and the random value corresponding to each point according to the hard disk failure type and the trimming weight comprises:
Calculating the corresponding fault type duty ratio according to the hard disk fault type;
and calculating the trimming weight and the random value according to the fault type duty ratio.
7. The probability-based hard disk failure analysis method according to claim 4, wherein the judging whether the hard disk failure record data is correct based on the trimming weight and the measured value and the trimming weight and the random value, comprises:
if the trimming weight and the measured value belong to a first preset area according to the trimming weight and the random value, judging that the hard disk fault record data are correct;
and if the trimming weight and the measured value belong to a second preset area according to the trimming weight and the random value, judging that the hard disk fault record data is wrong.
8. The probability-based hard disk failure analysis method according to claim 3, wherein constructing the hard disk failure analysis training set from the determined hard disk failure record data comprises:
directly incorporating the hard disk fault record data determined to be correct into the hard disk fault analysis training set;
and after the labeling result of the hard disk fault record data which is judged to be in error is subjected to the negation operation, the hard disk fault record data is included in the hard disk fault analysis training set.
9. The probability-based hard disk failure analysis method according to claim 1, wherein the step of pre-training the hard disk failure analysis model comprises:
inputting the hard disk fault analysis training set into an extreme gradient lifting tree for training to obtain a hard disk fault analysis initial model;
and continuously updating the hard disk fault analysis training set, and optimizing parameters of the hard disk fault analysis initial model by using the updated hard disk fault analysis training set to obtain the hard disk fault analysis model.
10. A probability-based hard disk failure analysis apparatus, comprising:
the fault atlas construction unit is used for constructing a hard disk fault data atlas according to the hard disk operation record data; wherein, each hard disk operation record data corresponds to each point in the hard disk fault data graph set;
the training set determining unit is used for constructing a hard disk fault analysis training set according to the position information of each point in the hard disk fault data graph set;
the hard disk fault analysis unit is used for analyzing the newly added hard disk faults by utilizing a pre-trained hard disk fault analysis model to obtain a fault analysis result; the hard disk fault analysis model is obtained by training the hard disk fault analysis training set.
11. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the probability-based hard disk failure analysis method of any one of claims 1 to 9 when the program is executed by the processor.
12. A computer readable storage medium having stored thereon a computer program, characterized in that the computer program when executed by a processor implements the steps of the probability-based hard disk failure analysis method of any of claims 1 to 9.
13. A computer program product comprising computer programs/instructions which, when executed by a processor, implement the steps of the probability-based hard disk failure analysis method of any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310671065.8A CN117009156A (en) | 2023-06-07 | 2023-06-07 | Probability-based hard disk fault analysis method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310671065.8A CN117009156A (en) | 2023-06-07 | 2023-06-07 | Probability-based hard disk fault analysis method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117009156A true CN117009156A (en) | 2023-11-07 |
Family
ID=88562649
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310671065.8A Pending CN117009156A (en) | 2023-06-07 | 2023-06-07 | Probability-based hard disk fault analysis method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117009156A (en) |
-
2023
- 2023-06-07 CN CN202310671065.8A patent/CN117009156A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108574933B (en) | User track recovery method and device | |
| CN105264555A (en) | Evaluation control | |
| US8856007B1 (en) | Use text to speech techniques to improve understanding when announcing search results | |
| US20200034278A1 (en) | System for refreshing and sanitizing testing data in a low-level environment | |
| US8214373B1 (en) | Systems and methods for assignment of human reviewers using probabilistic prioritization | |
| US10509660B2 (en) | Systems and methods for assessing configuration profiles for a user configurable device | |
| CN110991170A (en) | Chinese disease name intelligent standardization method and system based on electronic medical record information | |
| US11797354B2 (en) | Ensemble machine learning framework for predictive operational load balancing | |
| US9273972B2 (en) | Navigation system with error detection mechanism and method of operation thereof | |
| WO2020170020A1 (en) | Feedback mining with domain-specific modeling | |
| CN112882934B (en) | Test analysis method and system based on defect growth | |
| US20210357806A1 (en) | Machine learning model training method and machine learning model training device | |
| CN110674397A (en) | Method, device, equipment and readable medium for training age point prediction model | |
| CN113033912A (en) | Problem solving person recommendation method and device | |
| US20220276067A1 (en) | Method and apparatus for guiding voice-packet recording function, device and computer storage medium | |
| CN117009156A (en) | Probability-based hard disk fault analysis method and device | |
| US20140195949A1 (en) | Content delivery system with sequence generation mechanism and method of operation thereof | |
| CN110826616A (en) | Information processing method and device, electronic equipment and storage medium | |
| US20230064460A1 (en) | Generating input processing rules engines using probabilistic clustering techniques | |
| CN112905435B (en) | Workload assessment method, device, equipment and storage medium based on big data | |
| CN112163861B (en) | Transaction risk factor feature extraction method and device | |
| US20210365831A1 (en) | Identifying claim complexity by integrating supervised and unsupervised learning | |
| CN113935788A (en) | Model evaluation method, device, equipment and computer readable storage medium | |
| US10579754B1 (en) | Systems and methods for performing a fast simulation | |
| CN112764957A (en) | Application fault delimiting 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 |