CN107871055B - Data analysis method and device - Google Patents
Data analysis method and device Download PDFInfo
- Publication number
- CN107871055B CN107871055B CN201610854748.7A CN201610854748A CN107871055B CN 107871055 B CN107871055 B CN 107871055B CN 201610854748 A CN201610854748 A CN 201610854748A CN 107871055 B CN107871055 B CN 107871055B
- Authority
- CN
- China
- Prior art keywords
- data
- data analysis
- feature
- matrix
- task
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- Stored Programmes (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a data analysis method and a data analysis device, when a data analysis task is received, according to the type of the data analysis task, a characteristic matrix corresponding to the type can be determined from a corresponding matrix library, and data to be analyzed corresponding to the data analysis task can be processed by the characteristic matrix, so that the corresponding characteristic matrix is obtained from the matrix library to be used for processing one data analysis task, the output characteristic of the characteristic matrix is obtained, and the analysis result of the data analysis task is determined according to the output characteristic. And the feature matrix is not required to be specially configured for the data analysis task, so that the time for configuring the feature matrix according to the data to be analyzed and the requirement of the data analysis task originally is saved, and the data analysis efficiency is improved.
Description
Technical Field
The present invention relates to the field of data processing, and in particular, to a data analysis method and apparatus.
Background
With the development of data analysis technology, the importance of data is improved, and the analysis result of data analysis can be often used as an important reference for decision and development of some companies.
The process of data analysis needs to use the feature matrix and the model to process the data to be analyzed, so as to obtain the analysis result. The more accurate the analysis result of the data analysis, the higher the reference value.
At present, in the process of data analysis, when each data analysis task is executed, a feature matrix needs to be configured specially for the data analysis task, a large amount of time is consumed for configuring the feature matrix each time, and the efficiency of data analysis needs to be improved.
Disclosure of Invention
In order to solve the above technical problems, embodiments of the present invention provide a data analysis method and apparatus, which can save time for configuring a feature matrix and improve data analysis efficiency.
In a first aspect, an embodiment of the present invention provides a data analysis method, where the method includes:
receiving a data analysis task; acquiring data to be analyzed corresponding to the data analysis task; determining a characteristic matrix corresponding to the type from a matrix library according to the type of the data analysis task; processing the data according to the processing logic of the feature matrix to obtain the output features of the feature matrix; and determining an analysis result of the data analysis task according to the output characteristic.
Therefore, when a data analysis task is received, according to the type of the data analysis task, a feature matrix corresponding to the type can be determined from a corresponding matrix library, and the data required to be analyzed by the data analysis task can be processed by the feature matrix.
In a first possible implementation manner of the first aspect, the matrix library includes a feature matrix configured in an analysis process of a historical task, the historical task is a completed data analysis task, and a type of the historical task is the same as a type of the data analysis task.
Therefore, the characteristic matrix of the historical task is applied to the analysis of the data analysis task, and the effect of reusing the characteristic matrix is achieved. The time consumed for carrying out data analysis on the data analysis task is reduced, and the efficiency of data analysis is further improved. If the more comprehensive the characteristic matrix stored in the matrix library is, the more types of data analysis tasks corresponding to the characteristic matrix are, the greater the probability that the characteristic matrix stored before being directly reused in the matrix library is processed when a new data analysis task is received.
With reference to the implementation manner of the first aspect, in a second possible implementation manner, the method further includes:
determining a feature project corresponding to the type from a feature project library according to the type, wherein the feature project comprises a data analysis process from the step of obtaining data required to be analyzed to the step of obtaining output features from a feature matrix; the processing logic according to the feature matrix processes the data to obtain the output features of the feature matrix, including: and processing the data according to the processing logic of the feature matrix according to the data analysis process included in the feature engineering to obtain the output features of the feature matrix.
Therefore, by establishing the feature engineering library in advance and storing the feature engineering configured in the analysis process of the historical task and the corresponding relation between the type of the historical task and the configured feature engineering in the feature engineering library, the feature engineering corresponding to the type can be directly matched from the feature engineering library according to the type of the data analysis task when the received data analysis task is performed, the time for configuring the complicated data analysis step is saved, and the data analysis efficiency is improved.
With reference to the second implementation manner of the first aspect, in a third possible implementation manner, the data analysis process further includes a process of preprocessing data to be analyzed, where the preprocessing process includes any one or more of data deduplication, data sampling, and data optimization.
Therefore, by recording the preprocessing process of the data to be analyzed, when the data analysis process is reused for the data analysis task, the complicated preprocessing step of manual configuration is omitted, the time is saved, and the data analysis efficiency is improved.
In a fourth possible implementation manner of the first aspect, the feature matrix included in the matrix library is configured according to features and processing logic stored in a feature library, and the features stored in the feature library are constructed according to historical data.
Therefore, a feature library related to a field can be constructed in advance through data in the field, so that when a corresponding feature matrix needs to be configured for a data analysis task, configuration can be directly performed according to the determined features and processing logic in the feature library, and time consumed for configuring the feature matrix can be saved to a certain extent.
With reference to the fourth implementation manner of the first aspect, in a fifth possible implementation manner, if the historical data belongs to the field of telecommunications, the features stored in the feature library are constructed according to data in the field, and the method includes: the characteristics stored in the characteristic library are obtained by constructing the data in the telecommunication field based on the attributes carried by the data, wherein the attributes comprise any one or combination of a plurality of user attributes, position attributes, service attributes, terminal attributes and network attributes.
Therefore, according to the characteristics of the telecommunication field, a characteristic library about the telecommunication can be constructed according to the historical data of the telecommunication field by combining any one or more items of the user attribute, the position attribute, the service attribute, the terminal attribute and the network attribute, so that the data analysis efficiency of the data analysis task of the telecommunication field is improved.
In a sixth possible implementation manner of the first aspect, the determining an analysis result of the data analysis task according to the output feature includes: searching a model matched with the type of the data analysis task from a model library; and processing the output characteristics according to the searched model to obtain the analysis result.
Therefore, by constructing a model base in advance and storing the model configured in the analysis process of the historical task and the corresponding relation between the type of the historical task and the configured model in the model base, when the received data analysis task is performed, the model corresponding to the type can be directly matched from the model base according to the type of the data analysis task, and the output characteristic of the characteristic matrix can be processed according to the model, so that the time for configuring the model for the data analysis task again is saved, and the data analysis efficiency is improved.
With reference to the sixth implementation manner of the first aspect, in a seventh possible implementation manner, the model library includes a model configured in an analysis process of a historical task, the historical task is a completed data analysis task, and a type of the historical task is the same as a type of the data analysis task.
Therefore, the models configured for the data analysis tasks of the same type are basically similar, the models of the historical tasks are stored in the model base, the models can be reused when the data analysis tasks of the same type are processed, and the time for configuring the models for the data analysis tasks is saved.
In a second aspect, an embodiment of the present invention provides a data analysis apparatus, which includes a receiving unit, an obtaining unit, a determining unit, and a processing unit, and is configured to execute the method described in the first aspect or any implementation manner of the first aspect.
In a third aspect, a data analysis server is provided, which includes a memory for storing a program and a processor for executing the program, wherein when the program is executed, the processor is specifically configured to execute the method described in the first aspect or any implementation manner of the first aspect.
In a fourth aspect, a computer-readable medium is provided, the computer-readable medium being used for storing program code containing instructions for implementing the method described in the first aspect or any implementation manner of the first aspect.
According to the technical scheme, when a data analysis task is received, the feature matrix corresponding to the type can be determined from the corresponding matrix library according to the type of the data analysis task, and the data required to be analyzed by the data analysis task can be processed by the feature matrix.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a flowchart of a method of data analysis according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of data analysis in the field of telecommunications according to an embodiment of the present invention;
FIG. 3 is a flow chart of another method for data analysis according to an embodiment of the present invention;
fig. 4 is a flowchart of a method for analyzing data in the telecommunication field according to an embodiment of the present invention;
FIG. 5 is a flow chart of a method of another data analysis method provided by an embodiment of the present invention;
fig. 6 is a diagram illustrating an apparatus structure of a data analysis apparatus according to an embodiment of the present invention;
fig. 7 is a hardware structure diagram of a data analysis server according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As the importance of data increases, the need for data analysis of data increases. The process of data analysis needs to use the feature matrix and the model to process the data to be analyzed, so as to obtain the analysis result. Although there is analysis software for data analysis at present, such analysis software mainly aims at the general analysis of data, and does not pay attention to the characteristics of data in various fields.
There are some fields in which the types of data are more varied, and in these fields, the data to be analyzed by the same type of data analysis task may not be the same. There are some fields where the kind of data is fixed, such as the telecommunication field, the banking field, etc., and in these fields, the data to be analyzed for the same type of data analysis task may be substantially similar. It can be seen that, in a field where the types of data are fixed, if the feature matrix configured by the data analysis task that has been processed can be saved, and then if a data analysis task of the same type needs to be processed, since the data that needs to be processed by the data analysis task of the same type in this field are substantially similar, the feature matrix saved before may be reused in the data analysis task.
Therefore, when a data analysis task is received, according to the type of the data analysis task, a feature matrix corresponding to the type can be determined from a corresponding matrix library, and data required to be analyzed by the data analysis task can be processed by the feature matrix, so that the corresponding feature matrix is obtained from the matrix library to be used for processing the data analysis task, and a feature matrix is not required to be specially configured for the data analysis task, so that the time for configuring the feature matrix according to the data required to be analyzed and the requirement of the data analysis task originally is saved, and the data analysis efficiency is improved.
The embodiment of the invention is mainly applied to the field with fixed data types and less changes, such as the telecommunication field. When the data analysis tasks in the fields are processed, the feature matrix corresponding to the type can be determined from the matrix library according to the type of the data analysis tasks, the determined feature matrix can be used for processing and analyzing the data required to be analyzed by the data analysis tasks, and the output features required by the data analysis tasks are output from the feature matrix.
In the embodiment of the present invention, the data analysis task may be specific data analysis performed on specific data to fulfill a requirement, and if the data analysis task is an offline data analysis task, data to be analyzed by the data analysis task is offline data, and the performed data analysis is also data analysis for the offline data. The type of a data analysis task may be any one or combination of more of the specific analysis type of data analysis, the data type of the data to be analyzed, the algorithm used in the data analysis, and the type of feature matrix used.
The data analysis task has corresponding data to be analyzed, that is, data to be analyzed by the data analysis task, and in the data to be analyzed corresponding to the data analysis task, one data may be an individual value or record. Taking a data table of a table structure as an example, one data may be one record in the data table. The data may have attributes and the features may be collections of data having the same attributes. Taking a data table of a table structure as an example, one feature may be a general name of a column or a row of data in the table, and the column or the row of data may have the same attribute, for example, 4 columns are included in one data table, the 4 columns are "name", "age", "sex" and "height", respectively, each column may include at least one data having the same attribute, for example, the column of the feature "name" may include data "zhang san", "lie si" and "wang wu", and the column of the feature "age" may include data "20 years", "30 years" and "40 years". The three data of "zhang san", "li si" and "wang wu" all have attributes representing names. The three data of "20 years", "30 years", and "40 years" all have an attribute indicating age. For this data table there may be 4 features, respectively "name", "age", "gender" and "height", each feature comprising data with the same attributes, e.g. feature "name" comprises data "age 20", "age 30" and "age 40", feature "name" comprises data "zhang san", "lie four" and "wang five".
A data analysis task can perform data analysis on data to be analyzed, and the data analysis can include data processing operations of classifying, recombining, converting, operating, analyzing and the like on the data to be analyzed. In the process of data analysis, data to be analyzed needs to be input into the feature matrix, so that output features needed by the data analysis task are output through operation of the feature matrix. In embodiments of the present invention, the input content and the output content of a feature matrix may both be features. What features a feature matrix should input and what features can be output is determined by the processing logic of the feature matrix. For example, in the field of telecommunications, the processing logic of the feature matrix may be an Application Programming Interface (API). When the data to be analyzed by the data analysis task is processed through the feature matrix, the data needs to be firstly recombined and converted into the input features conforming to the feature matrix.
The data analysis method provided by the embodiment of the invention is described in detail below. Fig. 1 is a flowchart of a method of data analysis according to an embodiment of the present invention, where the method includes:
s101: a data analysis task is received.
For example, the data analysis task may be a specific data analysis performed on specific data to fulfill a requirement, and the data analysis task may provide information content such as data to be analyzed, attribution type of the data analysis task to be analyzed, and the like. For example, if data analysis needs to be performed on 8000 service Support systems (BSS) data tables for an off-network project, the data analysis task for the off-network project may include data analysis on 8000 BSS data tables to implement the task of the analysis requirement of the off-network project.
At this time, "8000 BSS data tables" related to the requirement of "performing data analysis on 8000 BSS data tables" may be used as the data to be analyzed by the data analysis task, and further, specific data analysis may be performed on the data included in the "8000 BSS data tables" provided by the data analysis task and the type of the data to which the data belongs, such as BSS. Meanwhile, since the item is executed in an off-network environment, the information content provided by the data analysis task may further include a characteristic condition for analyzing the data to be analyzed, such as an off-line environment, and the data analysis performed on the "8000 BSS data tables" may be data analysis performed on the data in the off-line environment.
The data analysis task received by the embodiment of the invention can belong to a data analysis task in a specific field, and the specific field can be a field with fixed data types and less changes, such as the fields of telecommunication and banking.
S102: and acquiring the data to be analyzed corresponding to the data analysis task.
For example, the data to be analyzed may be carried in the data analysis task, and thus may be obtained together when the data analysis task is obtained; the data to be analyzed can also be stored in the storage position indicated by the data analysis task, and after the data analysis task is acquired, the data required to be analyzed can be extracted from the indicated storage position. The method of acquiring the data to be analyzed is not limited in the present invention, and the data to be analyzed may be extracted through Structured Query Language (SQL).
S103: and determining a characteristic matrix corresponding to the type from a matrix library according to the type of the data analysis task.
For example, the type of data analysis task may be any one or more of a specific analysis type of data analysis, a data type of data required to analyze the data, an algorithm used in the data analysis, a feature matrix type used, and the like. The type of a data analysis task may be determined according to a specific application scenario or need.
Taking a data analysis task of an off-network project with 8000 pieces of BSS table as an example, the type of the data analysis task may be a specific analysis type of data analysis according to different scenarios or requirements: and analyzing the off-grid task, so that a characteristic matrix corresponding to the off-grid task analysis can be determined from the matrix library. Alternatively, the type of data analysis task may be the data type of the data that needs to be analyzed: BSS table, so that the characteristic matrix corresponding to the BSS table can be determined from the matrix library. Alternatively, the type of the data analysis task may be a specific analysis type of data analysis: and an algorithm A used in the off-network task analysis and the data analysis, so that a characteristic matrix corresponding to the off-network task analysis and the algorithm A can be determined from the matrix library.
The matrix library related in the embodiment of the invention can be pre-established and is mainly used for storing the feature matrix, the stored feature matrix can comprise feature matrices corresponding to different types of data analysis tasks, and the feature matrix can be used for processing data required to be analyzed in the corresponding data analysis tasks. The characteristic matrix in the matrix library has an incidence relation with the type of the corresponding data analysis task, so that the characteristic matrix having the incidence relation with the type can be determined from the matrix library through the type of the data analysis task. The feature matrix determined from the matrix library can be used for processing the data analysis task received in step S101, so that a process of specially configuring the feature matrix for the data analysis task is avoided, time required for configuring the feature matrix according to data to be analyzed and the requirements of the data analysis task originally is saved, and data analysis efficiency is improved. In general, the domain to which the data analysis task corresponding to the feature matrix in the matrix library belongs may be the same as the domain to which the data analysis task received in S101 belongs, and both belong to a domain with fixed data types and less changes, such as a telecommunication domain.
Aiming at the fact that the feature matrix corresponding to the type can be determined from the matrix library through the type of the data analysis task in S103, the invention provides several modes for determining the feature matrix from the matrix library according to the type of the data analysis task.
The first determination method:
in the determination method, the feature matrix included in the matrix library may be a feature matrix configured in an analysis process of the historical task. The historical task is a completed data analysis task.
The case where the feature matrix is matched from the matrix library is equivalent to the case where the feature matrix to be configured is determined for the currently received data analysis task (i.e., the data analysis task received at S101) from the feature matrices configured in the analysis process based on the historical tasks.
And determining the type of the historical task corresponding to the characteristic matrix from the matrix base to be the same as the type of the data analysis task received in the step S101. The determined feature matrix is applied to the processing of the data analysis task received in step S101, so that the effect of reusing the feature matrix is achieved. If the more comprehensive the characteristic matrix stored in the matrix library is, the more types of data analysis tasks corresponding to the characteristic matrix are, the greater the probability that the characteristic matrix stored before being directly reused in the matrix library is processed when a new data analysis task is received.
Through the first determination method, the probability of multiplexing the pre-stored feature matrix from the matrix library can be increased, the feature matrix does not need to be reconfigured for the data analysis task received in the step S101, the time consumed for performing data analysis on the data analysis task is reduced, and the efficiency of data analysis is improved.
The second determination method is as follows:
the embodiment of the invention not only can multiplex the characteristic matrix matched from the matrix library, but also can determine the characteristic matrix from the matrix library through a second determination mode. In this case, the feature matrix included in the matrix library may be configured according to the features and the processing logic stored in the feature library, where the features stored in the feature library are constructed according to the historical data. The scenario adopting the second determination method may include a scenario in which the corresponding feature matrix cannot be matched from the matrix library according to the type of the data analysis task, and this scenario may be understood as finding that the matrix library does not have the feature matrix corresponding to the type of the data analysis task when the feature matrix is matched from the matrix library by the first determination method. There are several cases that may mainly result from the feature matrix in the matrix library not having a correspondence to the type of the data analysis task. For example, in one case, the data analysis task of the same type as the data analysis task is not processed before the data analysis task is received in S101, and therefore the feature matrix corresponding to the type of the data analysis task is not stored in the matrix library; for example, in another case, although the data analysis task of the same type as the data analysis task received in S101 has been processed and analyzed previously, the configured feature matrix is not saved in the matrix library.
In a second determination, the feature matrix determined from the matrix library is a feature matrix configured from the features stored in the feature library and the processing logic. The features stored in the feature library may be constructed based on historical data, which is data belonging to a domain, which may be the telecommunications domain. Historical data for a domain may be obtained by pre-collection. The features stored in the feature library may be constructed for the attributes carried by the historical data. For example, when a feature library related to the telecommunication field is constructed, the data collected in the telecommunication field can be clustered according to the attributes carried by the data, so as to construct a feature library belonging to the telecommunication field. The attributes carried by the data in the telecommunications domain may include any one or a combination of user attributes, location attributes, service attributes, terminal attributes and network attributes. For example, data with user attributes are categorized for user attributes, resulting in a data set of the feature "user".
In addition, in the process of constructing the feature library, a knowledge graph technology may be used for carrying out processing operations such as screening and classification on collected data belonging to the telecommunication field, so that the collected data are classified into data sets with different attributes to form a plurality of features, and further, the generation automation of partial features in the feature library is realized.
The processing logic stored in the feature library may be constructed according to the relationship between different features in the feature library, for example, how to obtain the feature c according to the feature a and the feature b may be a processing logic, and it is seen that the feature and the processing logic have a clear association relationship.
For a domain, a feature library related to the domain can be constructed according to historical data of the domain. Then, when a data analysis task requiring data analysis in this field is received, a feature matrix corresponding to the type of the data analysis task can be configured and obtained by using the features and the processing logic included in the feature library.
When the feature matrix is configured by using the pre-constructed feature library, a feature matrix required by the data analysis task type can be generated according to the processing logic and the features stored in the feature library. For example, for a data analysis task in the field of telecommunications, according to the type of the data analysis task, features a, b, and c required to be input by a feature matrix in the data analysis task and features d and e required to be output are determined from a feature library, and how to determine a processing logic x for obtaining the features d and e according to the features a, b, and c is determined from the feature library, so that the feature matrix required by the data analysis task is configured according to the features a, b, c, d, and e and the processing logic x, and can be stored in a matrix library.
In addition, one or more algorithms may be used in configuring the feature matrix from the feature library, and may be part of the processing logic. The specific process comprises the steps of selecting a data set formed by required characteristics such as 'age' column by using the determined processing logic, processing the selected characteristics by using the determined algorithm, automatically generating a characteristic matrix F required to be configured in the data analysis by using the processed characteristics by using the processing logic, and storing the characteristic matrix F in a matrix library. The algorithm that may be used in the data analysis process may be one or a combination of multiple types, such as Machine Learning (ML), Time series analysis (Time series analysis), descriptive study (descriptive study), and the like, and the algorithm used to configure the feature matrix may be determined by the type of the data analysis task at this Time. The data analysis process described in the present invention may include a process from obtaining data to be analyzed to obtaining output characteristics from the characteristic matrix.
How to configure the feature matrix by the feature library is described with reference to fig. 2. In the scenario shown in fig. 2, a scheme of configuring a feature matrix through a feature library is implemented by a module in a data analysis platform. A feature library may be pre-constructed, and a specific construction process may be to construct features with different attributes according to collected data and/or Knowledge Engine (Knowledge Engine) modules belonging to the telecommunication field, obtain telecommunication APIs (telecommunications APIs) modules shown in fig. 2 according to a relationship between the different features, and further form the feature library by the features and the telecommunications APIs modules; the Telecom APIs module can perform information interaction with an Interactive exploration Environment (Interactive exploration Environment) module at any time. Since an algorithm may be used in the process of configuring the feature matrix, an algorithm (Algorithms) module is provided, and the Algorithms module can also perform information interaction with the Interactive application Environment module at any time.
When the Interactive abstraction Environment module receives a data analysis task belonging to the field of telecommunications sent from a telecommunications Feature Derivation (telecommunications Feature Derivation) module, the Interactive abstraction Environment module can firstly search from a matrix library to judge whether a Feature matrix corresponding to the type exists; if the data analysis task does not exist, the Interactive abstraction Environment module can determine the required processing logic from the Telecom APIs module of the feature library according to the type through information interaction with the feature library, and select the required features, so as to generate a feature matrix required to be configured by the data analysis task, and the Interactive abstraction Environment module can continue to perform subsequent data analysis according to the feature matrix, wherein the required processing logic determined from the Telecom APIs module can be any one or a combination of a Home/work Zone identifier (Home Zone), an operating system identifier (OS Identification) and a Content Identification.
In addition, in the process of determining the required processing logic, one or more Algorithms required for this time may be determined from the Algorithms module due to the type of the data analysis task at this time, so that the feature matrix required to be configured for this time of data analysis may be generated by combining the determined Algorithms.
Through the second determination mode, a feature library related to a field can be constructed in advance through data in the field, so that when a corresponding feature matrix needs to be configured for a data analysis task, configuration can be directly performed according to the determined features and processing logic in the feature library, and time consumed for configuring the feature matrix can be saved to a certain extent.
S104: and processing the data according to the processing logic of the feature matrix to obtain the output features of the feature matrix.
For example, after the feature matrix corresponding to the type of the data analysis task is obtained through configuration, the processing logic included in the feature matrix may be used to perform corresponding processing operations on the data to be analyzed by the data analysis task acquired in S102, so as to acquire the output features of the feature matrix.
In the process of performing corresponding processing operations on the data to be analyzed by the data analysis task acquired in step S102 by using the processing logic included in the feature matrix, if the input content to be input to the feature matrix does not meet the input requirement of the processing logic included in the feature matrix, the data to be analyzed by the data analysis task acquired in step S102 needs to be subjected to some data processing operations, such as data recombination, conversion, and the like, to further acquire the input content meeting the input requirement; after the input content such as input characteristics meeting the input requirements is input into the characteristic matrix, the characteristic matrix can operate the input characteristics through the contained processing logic, and finally output content such as output characteristics required by the data analysis task is output.
S105: and determining an analysis result of the data analysis task according to the output characteristic.
For example, after the output characteristics required by the data analysis task are determined, the output characteristics may be calculated through a model, so as to obtain an analysis result of the data analysis task.
According to the embodiment, the data to be analyzed of the data analysis task this time and the type of the data analysis task can be obtained through the received data analysis task, so that the feature matrix corresponding to the type can be determined from the matrix library according to the type of the data analysis task, the data to be analyzed can be processed by the processing logic of the feature matrix, the output feature of the feature matrix is obtained, the corresponding feature matrix is directly obtained from the matrix library to be used for processing one data analysis task, the feature matrix does not need to be specially configured for the data analysis task, the time for configuring the feature matrix according to the data to be analyzed and the requirement of the data analysis task originally is saved, and the data analysis efficiency is improved.
In the process of obtaining the output characteristics from the characteristic matrix according to the required analysis data, the data analysis steps to be implemented are cumbersome, for example, data analysis steps such as converting the data into the input characteristics required by the characteristic matrix, and operating the input characteristics by processing logic in the characteristic matrix. In the embodiment of the invention, the data analysis steps from the acquisition of the data to be analyzed to the acquisition of the output characteristics from the characteristic matrix are collectively referred to as characteristic engineering.
In the fields with fixed data types and less changes, such as the telecommunication field, the characteristic engineering is actually similar in the data analysis tasks of the same type, so in the fields with fixed data types and less changes, the characteristic matrix configured in the data analysis tasks of the same type can be multiplexed, and the characteristic engineering in the data analysis tasks of the same type can be multiplexed.
On the basis of the embodiment corresponding to fig. 1, with respect to step S104, an embodiment of the present invention further provides a specific method for acquiring an output feature of a feature matrix based on a historical task, as shown in fig. 3, the method includes:
s301: and determining a feature project corresponding to the type from a feature project library according to the type of the data analysis task, wherein the feature project comprises a data analysis process from the step of obtaining the data to be analyzed to the step of obtaining the output features from the feature matrix.
For example, the feature engineering library according to the embodiment of the present invention may be pre-constructed, and is used to store a data analysis process from acquiring data to be analyzed of a history task that has been processed to obtaining an output feature from a feature matrix, and a corresponding relationship between a type of the history task and a feature engineering thereof.
The feature engineering may further include a preprocessing process of analyzing data required by the data analysis task, wherein the preprocessing process includes any one or more of data deduplication, data sampling and data optimization. For example, the data preprocessing may include data sampling of the data to be analyzed, extracting a part of the data from the sampled data for data analysis, or data deduplication of the data to be analyzed, removing duplicate data to improve data analysis accuracy, or pre-feature clustering of the data to be analyzed to obtain the input features required by the feature matrix.
Because the feature engineering is similar in the data analysis tasks of the same type, the feature engineering with the corresponding relation with the type can be directly matched from the constructed feature engineering library according to the type of the data analysis tasks, so that the feature engineering can be repeatedly used for carrying out the corresponding data analysis process.
S302: and processing the data according to the processing logic of the feature matrix according to the data analysis process included in the feature engineering to obtain the output features of the feature matrix.
For example, when a data analysis task is received, a corresponding feature engineering may be determined from a feature engineering library according to the type of the data analysis task, and the determined feature engineering may be a feature engineering used in a historical task of the same type. According to the characteristic engineering, the data required to be analyzed by the data analysis task can be preprocessed, although the data content of the data analysis task is different from that of the historical task, the processing steps are basically the same, so that the data content required to be processed by each step can be correspondingly adjusted according to the data required to be analyzed by the data analysis task on the basis of the characteristic engineering, the determined characteristic engineering can be applied to the processing of the data analysis task, and the output characteristic required by the data analysis task can be obtained according to the characteristic matrix.
As further illustrated in fig. 4, after obtaining a data analysis task from a data source module and obtaining data to be analyzed from the data analysis task, if the total amount of the data to be analyzed is too much, a part of the data may be extracted from the data to be analyzed by a data sampling module shown in fig. 4 for analysis; and then the data preprocessing module performs other preprocessing operations on the extracted partial data, such as any one or combination of multiple items of data deduplication, data optimization and the like. Then, the telecom api (telecom api) module processes the data obtained after the data preprocessing module processes the data again to obtain output features, and stores the output features into a Feature Matrix (Feature Matrix) module shown in fig. 4, and simultaneously stores the data analysis process from the data to be analyzed to the Feature Matrix obtained from the Feature Matrix into the Feature engineering module shown in fig. 4 as a Feature engineering of the data analysis task, and then the Feature management module performs screening processing and management on the obtained Feature engineering, so that when the data analysis task of the same type is obtained from the data source module, the Feature engineering stored in the Feature management module can be directly reused to obtain the output features of the Feature Matrix, thereby saving the time consumed by the data analysis process.
According to the embodiment, the characteristic engineering library is established in advance, the characteristic engineering configured in the analysis process of the historical task and the corresponding relation between the type of the historical task and the configured characteristic engineering are stored in the characteristic engineering library, so that the characteristic engineering corresponding to the type can be directly matched from the characteristic engineering library according to the type of the data analysis task when the received data analysis task is performed, the time for configuring complicated data analysis steps is saved, and the data analysis efficiency is improved.
In the process of performing data analysis on the data analysis task, the output characteristics output by the characteristic matrix need to be calculated through a model, so as to obtain an analysis result of the data analysis task. In the embodiment of the present invention, the model may be a rule entity generated jointly according to a predetermined algorithm, algorithm parameters, and training data, and for the model used by the data analysis task, the input content of the model may include the output features of the feature matrix used by the data analysis task, and the output content of the model may be the analysis result of the data analysis task.
In the field with fixed data types and less change, the models configured for the data analysis tasks of the same type are basically similar, so that the time for configuring the models for the data analysis tasks is saved.
In another embodiment, as shown in fig. 5, step S105 may include:
s501: and searching a model matched with the type of the data analysis task from a model library.
For example, the model library may be pre-constructed, and is used to store the models configured in the analysis process of the historical tasks and the corresponding relationships between the types of the historical tasks and the configured models thereof, so that the subsequent data analysis tasks of the same type may directly find the corresponding models from the model library according to the types thereof, thereby realizing the reuse of the models and saving the time for re-configuring the models for the data analysis tasks. For a data analysis task, if a model can be determined from a model library according to the type of the data analysis task, the type of the historical task corresponding to the model may be the same as the type of the data analysis task.
S502: and processing the output characteristics according to the searched model to obtain the analysis result.
For example, as shown in fig. 4, after the feature engineering module in fig. 4 outputs the output features, the model construction module matches a model corresponding to the type from the pre-stored models according to the type of the data analysis task, and performs an operation on the output features by using the model obtained through matching, so as to obtain a corresponding analysis result, thereby avoiding an operation process of reconfiguring the model for the data analysis task; meanwhile, the Model building module can also input the stored Model into the Model (Model) module, and the Model (Model) module performs related training on the input Model by utilizing the Prediction (Prediction) module, so that the accuracy of the Model is improved. In order to increase the reuse rate of the models stored in the constructed model library, one or more combined operations of model training, model evaluation, model updating and the like can be performed on the models stored in the model library. As shown in fig. 2, in order to ensure that when receiving a data analysis task sent by a telecommunications Feature Derivation (telecommunications Feature Derivation) module, a corresponding Model can be matched from a Model Lifecycle Management (Model Lifecycle Management) module according to the type of the data analysis task, any one or more of a Model building (Build) sub-module, a Model verification (valid Model) sub-module, a Model Update (Update Model) sub-module, and a Model monitoring (delay Monitor Model) sub-module included in the Model Lifecycle Management module can be used to manage the Model, so that the Model is more accurate, and the reuse rate of the Model is also improved.
According to the embodiment, the model base can be constructed in advance, the model configured in the analysis process of the historical task and the corresponding relation between the type of the historical task and the configured model are stored in the model base, so that when the received data analysis task is performed, the model corresponding to the type can be directly matched from the model base according to the type of the data analysis task, the output characteristic of the characteristic matrix can be processed according to the model, the time for configuring the model for the data analysis task again is saved, and the data analysis efficiency is improved.
Fig. 6 is a device configuration diagram of a data analysis device according to an embodiment of the present invention, where the data analysis device 600 includes: a receiving unit 601, an obtaining unit 602, a determining unit 603 and a processing unit 604, wherein:
the receiving unit 601 is configured to receive a data analysis task.
An obtaining unit 602, configured to obtain data to be analyzed corresponding to the data analysis task.
The determining unit 603 is configured to determine, according to the type of the data analysis task, a feature matrix corresponding to the type from a matrix library.
A processing unit 604, configured to process the data according to the processing logic of the feature matrix, so as to obtain an output feature of the feature matrix.
The determining unit 603 is further configured to determine an analysis result of the data analysis task according to the output feature.
Optionally, the matrix library includes a feature matrix configured in an analysis process of a historical task, the historical task is a completed data analysis task, and a type of the historical task is the same as a type of the data analysis task.
Optionally, the determining unit 603 is further configured to determine, according to the type, a feature project corresponding to the type from a feature project library, where the feature project includes a data analysis process from obtaining data to be analyzed to obtaining an output feature from a feature matrix;
the processing unit 604 is further configured to process the data according to the processing logic of the feature matrix according to a data analysis process included in the feature engineering, and obtain an output feature of the feature matrix.
Optionally, the data analysis process further includes a process of preprocessing the data to be analyzed, where the preprocessing process includes any one or more of data deduplication, data sampling, and data optimization.
Optionally, the feature matrix included in the matrix library is configured according to features stored in a feature library and processing logic, and the features stored in the feature library are constructed according to historical data.
Optionally, if the historical data belongs to the field of telecommunications, the features stored in the feature library are constructed according to data in the field, and the method includes:
the features stored in the feature library are obtained by constructing data in the telecommunication field based on attributes carried by the data, wherein the attributes comprise any one or combination of a plurality of user attributes, position attributes, service attributes, terminal attributes and network attributes.
Optionally, the determining unit 603 is further configured to search a model matching the type of the data analysis task from a model library; and processing the output characteristics according to the searched model to obtain the analysis result.
Optionally, the model library includes a model configured in an analysis process of a historical task, the historical task is a completed data analysis task, and a type of the historical task is the same as a type of the data analysis task.
In one embodiment, the receiving unit 601 may be a network interface or API for receiving data analysis tasks submitted by an application program; the obtaining unit 602, the determining unit 603, and the processing unit 604 may be implemented by one or more processors, which may specifically be general processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs), or other programmable logic components.
It should be noted that, for details of implementation in the embodiment corresponding to fig. 6, reference may be made to relevant descriptions of the embodiments corresponding to fig. 1, fig. 3, and fig. 5, which are not described herein again.
Fig. 7 is a schematic diagram of a hardware structure of a data analysis server according to an embodiment of the present invention, where the data analysis server 700 includes a memory 701, a receiver 702, and a processor 703 connected to the memory 701 and the receiver 702, where the memory 701 is used to store a set of program instructions, and the processor 703 is used to call the program instructions stored in the memory 701 to perform the following operations:
triggering the receiver 702 to receive a data analysis task;
triggering the receiver 702 to obtain the data to be analyzed corresponding to the data analysis task;
determining a feature matrix corresponding to the type from a matrix library according to the type of the data analysis task;
processing the data according to the processing logic of the feature matrix to obtain the output features of the feature matrix;
and determining an analysis result of the data analysis task according to the output characteristic.
Optionally, the processor 703 is further configured to call the program instructions stored in the memory 701 to perform other steps in the embodiments corresponding to fig. 1, fig. 3, and fig. 5.
In one embodiment, the processor 703 may be a Central Processing Unit (CPU), the Memory 701 may be an internal Memory of a Random Access Memory (RAM) type, and the receiver 702 may include a common physical interface, which may be an Ethernet (Ethernet) interface or an Asynchronous Transfer Mode (ATM) interface. The processor 703, receiver 702, and memory 701 may be integrated into one or more separate circuits or hardware, such as: application Specific Integrated Circuit (ASIC).
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium may be at least one of the following media: various media that can store program codes, such as read-only memory (ROM), RAM, magnetic disk, or optical disk.
It should be noted that, in the present specification, all the embodiments are described in a progressive manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the apparatus and system embodiments, since they are substantially similar to the method embodiments, they are described in a relatively simple manner, and reference may be made to some of the descriptions of the method embodiments for related points. The above-described embodiments of the apparatus and system are merely illustrative, and the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (14)
1. A method of data analysis, the method comprising:
receiving a data analysis task;
acquiring data to be analyzed corresponding to the data analysis task;
determining a feature matrix corresponding to the type from a matrix library according to the type of the data analysis task;
processing the data according to the processing logic of the feature matrix to obtain the output features of the feature matrix;
determining an analysis result of the data analysis task according to the output characteristic;
the feature matrix included in the matrix library is configured according to features stored in a feature library and processing logic, the features stored in the feature library are constructed according to historical data, and the processing logic is used for identifying the association relationship among the features.
2. The method of claim 1, wherein the matrix library comprises a feature matrix configured during an analysis process of a historical task, wherein the historical task is a completed data analysis task, and wherein the type of the historical task is the same as the type of the data analysis task.
3. The method of claim 2, further comprising:
determining a feature project corresponding to the type from a feature project library according to the type, wherein the feature project comprises a data analysis process from the acquisition of data to be analyzed to the acquisition of output features from a feature matrix;
the processing the data according to the processing logic of the feature matrix to obtain the output features of the feature matrix includes:
and processing the data according to the processing logic of the feature matrix according to the data analysis process included in the feature engineering to obtain the output features of the feature matrix.
4. The method of claim 3, wherein the data analysis process further comprises a pre-processing process for the data to be analyzed, the pre-processing process comprising any one or more of data deduplication, data sampling, and data optimization.
5. The method of claim 1, wherein the historical data belongs to a telecommunication domain, and the features stored in the feature library are constructed according to data in the domain, and the method comprises:
the features stored in the feature library are obtained by constructing data in the telecommunication field based on attributes carried by the data, wherein the attributes comprise any one or combination of a plurality of user attributes, position attributes, service attributes, terminal attributes and network attributes.
6. The method of claim 1, wherein determining the analysis results of the data analysis task based on the output features comprises:
searching a model matched with the type of the data analysis task from a model library;
and processing the output characteristics according to the searched model to obtain the analysis result.
7. The method of claim 6, wherein the model library comprises models configured during analysis of historical tasks, the historical tasks being completed data analysis tasks, the historical tasks being of the same type as the data analysis tasks.
8. A data analysis apparatus, characterized in that the apparatus comprises a receiving unit, an obtaining unit, a determining unit, and a processing unit:
the receiving unit is used for receiving a data analysis task;
the acquisition unit is used for acquiring the data to be analyzed corresponding to the data analysis task;
the determining unit is used for determining a feature matrix corresponding to the type from a matrix library according to the type of the data analysis task;
the processing unit is used for processing the data according to the processing logic of the feature matrix so as to obtain the output features of the feature matrix;
the determining unit is further used for determining an analysis result of the data analysis task according to the output characteristic;
the feature matrix included in the matrix library is configured according to features stored in a feature library and processing logic, the features stored in the feature library are constructed according to historical data, and the processing logic is used for identifying the association relationship among the features.
9. The apparatus of claim 8, wherein the matrix library comprises a feature matrix configured during an analysis process of a historical task, wherein the historical task is a completed data analysis task, and wherein the type of the historical task is the same as the type of the data analysis task.
10. The apparatus according to claim 9, wherein the determining unit is further configured to determine, according to the type, a feature engineering corresponding to the type from a feature engineering library, where the feature engineering includes a data analysis process from obtaining data to be analyzed to obtaining output features from a feature matrix;
and the processing unit is also used for processing the data according to the processing logic of the feature matrix according to the data analysis process included in the feature engineering to obtain the output features of the feature matrix.
11. The apparatus of claim 10, wherein the data analysis process further comprises a process of preprocessing the data to be analyzed, the preprocessing process comprising any one or more of data deduplication, data sampling, and data optimization.
12. The apparatus of claim 8, wherein the historical data belongs to a telecommunication domain, and the features stored in the feature library are constructed according to data in the domain, and the method comprises:
the features stored in the feature library are obtained by constructing data in the telecommunication field based on attributes carried by the data, wherein the attributes comprise any one or combination of a plurality of user attributes, position attributes, service attributes, terminal attributes and network attributes.
13. The apparatus of claim 8, wherein the determining unit is further configured to search a model library for a model matching the type of the data analysis task; and processing the output characteristics according to the searched model to obtain the analysis result.
14. The apparatus of claim 13, wherein the model library comprises models configured during an analysis of historical tasks, wherein the historical tasks are completed data analysis tasks, and wherein the historical tasks are of the same type as the data analysis tasks.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210254763.3A CN114791927A (en) | 2016-09-27 | 2016-09-27 | Data analysis method and device |
| CN201610854748.7A CN107871055B (en) | 2016-09-27 | 2016-09-27 | Data analysis method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610854748.7A CN107871055B (en) | 2016-09-27 | 2016-09-27 | Data analysis method and device |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210254763.3A Division CN114791927A (en) | 2016-09-27 | 2016-09-27 | Data analysis method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107871055A CN107871055A (en) | 2018-04-03 |
| CN107871055B true CN107871055B (en) | 2022-03-29 |
Family
ID=61750877
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210254763.3A Pending CN114791927A (en) | 2016-09-27 | 2016-09-27 | Data analysis method and device |
| CN201610854748.7A Active CN107871055B (en) | 2016-09-27 | 2016-09-27 | Data analysis method and device |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210254763.3A Pending CN114791927A (en) | 2016-09-27 | 2016-09-27 | Data analysis method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN114791927A (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110163380B (en) * | 2018-04-28 | 2023-07-07 | 腾讯科技（深圳）有限公司 | Data analysis method, model training method, device, equipment and storage medium |
| CN109299178B (en) * | 2018-09-30 | 2020-01-14 | 北京九章云极科技有限公司 | Model application method and data analysis system |
| CN109740774B (en) * | 2019-02-28 | 2021-07-30 | 中国公路工程咨询集团有限公司 | Correction method of pavement maintenance measure library and electronic equipment |
| CN110275880B (en) * | 2019-05-21 | 2023-05-30 | 创新先进技术有限公司 | Data analysis method, device, server and readable storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101021897A (en) * | 2006-12-27 | 2007-08-22 | 中山大学 | Two-dimensional linear discrimination human face analysis identificating method based on interblock correlation |
| CN101324923A (en) * | 2008-08-05 | 2008-12-17 | 北京中星微电子有限公司 | Method and apparatus for extracting human face recognition characteristic |
| CN101409643A (en) * | 2008-11-24 | 2009-04-15 | 北京中创信测科技股份有限公司 | Method, apparatus and corresponding system for analyzing telecom network modeling |
| CN101685458A (en) * | 2008-09-27 | 2010-03-31 | 华为技术有限公司 | Recommendation method and system based on collaborative filtering |
| CN102184250A (en) * | 2011-05-24 | 2011-09-14 | 东华大学 | Garment fabric sample retrieving method based on colored image matching |
| US20130191186A1 (en) * | 2012-01-24 | 2013-07-25 | International Business Machines Corporation | System, method and computer program for capturing relationships between business outcomes, persons and technical assets |
| CN103268362A (en) * | 2013-06-08 | 2013-08-28 | 国家电网公司 | Auxiliary design method of virtual terminals on basis of general template and key character matching |
| CN104184589A (en) * | 2014-08-26 | 2014-12-03 | 重庆邮电大学 | Identity authentication method, terminal device and system |
-
2016
- 2016-09-27 CN CN202210254763.3A patent/CN114791927A/en active Pending
- 2016-09-27 CN CN201610854748.7A patent/CN107871055B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101021897A (en) * | 2006-12-27 | 2007-08-22 | 中山大学 | Two-dimensional linear discrimination human face analysis identificating method based on interblock correlation |
| CN101324923A (en) * | 2008-08-05 | 2008-12-17 | 北京中星微电子有限公司 | Method and apparatus for extracting human face recognition characteristic |
| CN101685458A (en) * | 2008-09-27 | 2010-03-31 | 华为技术有限公司 | Recommendation method and system based on collaborative filtering |
| CN101409643A (en) * | 2008-11-24 | 2009-04-15 | 北京中创信测科技股份有限公司 | Method, apparatus and corresponding system for analyzing telecom network modeling |
| CN102184250A (en) * | 2011-05-24 | 2011-09-14 | 东华大学 | Garment fabric sample retrieving method based on colored image matching |
| US20130191186A1 (en) * | 2012-01-24 | 2013-07-25 | International Business Machines Corporation | System, method and computer program for capturing relationships between business outcomes, persons and technical assets |
| CN103268362A (en) * | 2013-06-08 | 2013-08-28 | 国家电网公司 | Auxiliary design method of virtual terminals on basis of general template and key character matching |
| CN104184589A (en) * | 2014-08-26 | 2014-12-03 | 重庆邮电大学 | Identity authentication method, terminal device and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107871055A (en) | 2018-04-03 |
| CN114791927A (en) | 2022-07-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yang et al. | A system architecture for manufacturing process analysis based on big data and process mining techniques | |
| CN107871055B (en) | Data analysis method and device | |
| CN111352759B (en) | Alarm root cause judging method and device | |
| CN113392646A (en) | Data relay system, construction method and device | |
| US9706005B2 (en) | Providing automatable units for infrastructure support | |
| CN112199276A (en) | Alteration detection method and device for microservice architecture, server and storage medium | |
| CN113190426B (en) | Stability monitoring method for big data scoring system | |
| CN112182025A (en) | Log analysis method, device, equipment and computer readable storage medium | |
| CN113098888A (en) | Abnormal behavior prediction method, device, equipment and storage medium | |
| US20200082822A1 (en) | System and method for mapping a customer journey to a category | |
| CN113157978B (en) | Data label establishing method and device | |
| CN115204889A (en) | Text processing method and device, computer equipment and storage medium | |
| CN112416800A (en) | Intelligent contract testing method, device, equipment and storage medium | |
| CN112256517A (en) | Log analysis method and device of virtualization platform based on LSTM-DSSM | |
| CN110647537A (en) | Data searching method, device and storage medium | |
| CN116225848A (en) | Log monitoring method, device, equipment and medium | |
| CN115794744A (en) | Log display method, device, equipment and storage medium | |
| CN110895538A (en) | Data retrieval method, device, storage medium and processor | |
| US8090750B2 (en) | Prompting of an end user with commands | |
| CN110727532B (en) | Data restoration method, electronic equipment and storage medium | |
| CN115706695A (en) | Method, device, equipment and storage medium for determining root cause of network fault | |
| CN109785099B (en) | Method and system for automatically processing service data information | |
| CN106469086B (en) | Event processing method and device | |
| CN111311329B (en) | Tag data acquisition method, device, equipment and readable storage medium | |
| CN115052035B (en) | Message pushing method, device and storage medium |
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 |