CN109271374B - Database health degree scoring method and system based on machine learning - Google Patents
Database health degree scoring method and system based on machine learning Download PDFInfo
- Publication number
- CN109271374B CN109271374B CN201811220356.0A CN201811220356A CN109271374B CN 109271374 B CN109271374 B CN 109271374B CN 201811220356 A CN201811220356 A CN 201811220356A CN 109271374 B CN109271374 B CN 109271374B
- Authority
- CN
- China
- Prior art keywords
- data
- model
- training
- database
- scoring
- 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
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/3668—Software testing
- G06F11/3672—Test management
- G06F11/3688—Test management for test execution, e.g. scheduling of test suites
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a database health degree scoring method and a scoring system based on machine learning, wherein the scoring method comprises the following steps: 1. collecting database monitoring indexes, and obtaining health degree scores through an expert model; collecting original data and scoring to serve as a sample set; 2. preprocessing the data in the sample set such as denoising and normalization, and dividing the data into training data, verification data and test data; 3. establishing a regression prediction model by adopting a regression prediction algorithm, training model parameters by using training data, adjusting the model parameters by using verification data, and testing the effect of the model by using test data; 4. and reading the monitoring indexes of the database in a period of time and preprocessing the monitoring indexes to be used as the input of a regression prediction model, wherein the output of the model is the health degree scoring result of the database in the current or future period of time. The method can analyze a large number of database monitoring indexes, and a regression prediction model is established to obtain the database health degree scoring result at the current or future moment.
Description
Technical Field
The invention belongs to the field of operation and maintenance of databases, and particularly relates to a method and a system for scoring and predicting the health degree of a database by adopting an artificial intelligence method.
Background
At present, the operation and maintenance of a large Database system are mainly maintained by a high-end DBA (Database Administrator), and the DBA can score the health degree of the whole operation condition of the Database by checking various indexes of the Database, which is called an expert model. The expert model is that an expert with years of database operation and maintenance experience is relied on, the index which has the greatest influence on the health degree of the database is manually selected, the indexes are scored by adopting a manually set threshold value, and finally the scores are summed up to obtain the final health score. But only depending on the experience of the expert DBA, the method is difficult to deal with various difficulties of database operation and maintenance. The monitoring indexes of the database are huge in quantity, and the cost is too high by manual analysis; the incidence relation among different indexes is complex, and the rules are difficult to find by manual analysis; problems are easy to find depending on manpower, but are difficult to locate; monitoring indexes of different database software are different; the complexity of the system is continuously increased, and the association relationship is more complex. To summarize, the major drawbacks of the expert model are: 1. the indexes are selected by experts according to experience, a large number of indexes are not selected, whether the unselected indexes are important to the health of the database or not can not be provided with answers by the experts, and the unselected indexes have huge analysis workload and cannot be completed by manpower; 2. the expert model cannot analyze the incidence relation among various indexes; the indexes are isolated; 3. the expert model cannot give a health score prediction for a period of time in the future, and can only calculate the current score according to the currently obtained index.
Disclosure of Invention
The purpose of the invention is as follows: aiming at the problems in the prior art, the invention provides a method for scoring and predicting the health degree of a database by using machine learning.
The technical scheme is as follows: the invention provides a database health degree scoring method based on machine learning, which comprises the following steps:
(1) collecting database monitoring indexes, and obtaining health degree scores through an expert model; collecting original data and scoring to serve as a sample set;
(2) preprocessing the data in the sample set such as denoising and normalization, and dividing the data in the sample set into training data, verification data and test data;
(3) establishing a regression prediction model by adopting a regression prediction algorithm, training model parameters by using training data, adjusting the model parameters by using verification data, and testing the effect of the model by using test data;
(4) and (3) reading the monitoring indexes of the database within a period of time, preprocessing the monitoring indexes by adopting the same method as the step (2), taking the preprocessed data as the input of a regression prediction model, and outputting the model, namely the database health degree scoring result at the current or future moment.
Preferably, the step (2) further comprises feature selection after denoising the data in the sample set, and then normalization processing is performed; the feature selection is to calculate the importance of the monitoring indexes and delete the first N monitoring indexes with the minimum importance in the sample set; the method specifically comprises the following steps:
(2.1) establishing a GBM regression model, setting the number m of the lifting trees and the learning rate rho, and setting a loss function as a mean square error;
(2.2) dividing the denoised sample set into a GBM training set and a GBM testing set, and training a GBM regression model by using the GBM training set; in the training process, if the loss function of n rounds of training is not reduced any more, the training is stopped; otherwise, continuing training until the loss function is not reduced any more; obtaining the importance value of each monitoring index in the sample set after the training is stopped;
(2.3) repeating the steps (2.1) and (2.2) W times, obtaining W importance values of each monitoring index, and taking the average value of the importance values as the importance values of the monitoring indexes;
and (2.4) sequencing the importance values of all the monitoring indexes, and deleting the first N monitoring indexes with the minimum importance values.
Constructing an LSTM RNN regression prediction model by adopting an LSTM RNN algorithm in the step (3), wherein the LSTM RNN regression prediction model comprises 1 input layer, 3 LSTM layers, 1 full-connection layer and 1 output layer, and the 3 LSTM layers comprise 32 LSTM units; the full connection layer comprises 32 units and adopts a relu activation function; the output layer comprises 4 units and adopts a sigmoid activation function;
in the step (3), a random forest regression prediction model can be constructed by adopting a random forest algorithm; the random forest regression prediction model comprises p decision trees, and the determining step of the depth q of the decision trees is as follows:
setting an upper limit value Q of the depth of the decision tree, enabling t to carry out Q times of training from 1 to Q, calculating a loss function value of each training, and taking the value of t with the minimum loss function value in Q times of training as the depth Q of the decision tree.
On the other hand, the invention provides a database health degree scoring system based on machine learning, which comprises a database monitoring index acquisition module, an expert model scoring module, a preprocessing module and a regression prediction model; the database monitoring index acquisition module acquires database monitoring indexes and scores the health degree through the expert model scoring module; the preprocessing module carries out preprocessing such as denoising and normalization on a sample set formed by collected database monitoring indexes and health degree scores, and divides data in the sample set into training data, verification data and test data; the regression prediction model trains model parameters by using training data, utilizes the whole model parameters of verification data, and utilizes test data to test the effect of the model.
The system further comprises a characteristic selection module, wherein the characteristic selection module is used for carrying out characteristic selection on the denoised data in the preprocessing module and then carrying out normalization processing.
Further, the regression prediction model is an LSTM RNN regression prediction model, the LSTM RNN regression prediction model includes 1 input layer, 3 LSTM layers, 1 fully-connected layer, and 1 output layer, and each of the 3 LSTM layers includes 32 LSTM units; the full connection layer comprises 32 units and adopts a relu activation function; the output layer comprises 4 units and a sigmoid activation function is adopted.
The regression prediction model can also adopt a random forest regression prediction model; the random forest regression prediction model comprises p decision trees, and the determining step of the depth q of the decision trees is as follows:
setting an upper limit value Q of the depth of the decision tree, enabling t to carry out Q times of training from 1 to Q, calculating a loss function value of each training, and taking the value of t with the minimum loss function value in Q times of training as the depth Q of the decision tree.
Has the advantages that: compared with the prior art, the database health degree scoring method and system based on machine learning, disclosed by the invention, can be used for scoring and predicting the database health degree in the current or future period of time by utilizing a large amount of database monitoring indexes to train and adjusting the regression prediction model.
Drawings
FIG. 1 is a flow chart of a database health scoring method disclosed in the present invention;
FIG. 2 is a block diagram of a database health scoring system according to the present disclosure.
Detailed Description
The invention is further elucidated with reference to the drawings and the detailed description.
Example 1:
a database health degree scoring method based on machine learning, as shown in fig. 1, includes the following steps:
step 1, collecting database monitoring indexes, and obtaining health degree scores through an expert model; collecting original data and scoring to serve as a sample set;
in this embodiment, 250 indexes of database operation are collected as monitoring indexes, including database connection state, CPU utilization, memory utilization, disk read-write, cache size, delay, response time, and the like; scores for health scores from 0 to 100; scoring through an expert model to serve as manually marked sample set data;
step 2, preprocessing such as denoising and normalization is carried out on the data in the sample set, and the data in the sample set is divided into training data, verification data and test data;
and denoising the data in the sample set, including removing abnormal values and missing values and removing indexes with only a single value. In order to remove the indexes, in the embodiment, the denoised data is subjected to feature selection and then normalization processing; the feature selection is to calculate the importance of the monitoring indexes and delete the first N monitoring indexes with the minimum importance in the sample set; the method comprises the following specific steps:
(2.1) establishing a GBM regression model, setting the number m of the lifting trees and the learning rate rho, and setting a loss function as a mean square error;
(2.2) dividing the denoised sample set into a GBM training set and a GBM testing set, and training a GBM regression model by using the GBM training set; in the training process, if the loss function of n rounds of training is not reduced any more, the training is stopped; otherwise, continuing training until the loss function is not reduced any more; obtaining the importance value of each monitoring index in the sample set after the training is stopped;
(2.3) repeating the steps (2.1) and (2.2) W times, obtaining W importance values of each monitoring index, and taking the average value of the importance values as the importance values of the monitoring indexes;
and (2.4) sequencing the importance values of all the monitoring indexes, and deleting the first N monitoring indexes with the minimum importance values.
In this embodiment, 1000 lifting trees are set, the learning rate is 0.05, and the training is stopped when the loss function of 100 rounds of training is not reduced any more in the training process. And (3) training for 10 times, namely repeating the steps (2.1) and (2.2) for 10 times, establishing 10 GBM regression models, obtaining 10 importance values for each monitoring index, and taking the average of 10 times as a final importance value.
For some classification features, one-hot coding is used. Through denoising and feature selection, the 250-dimensional index is reduced to 141-dimensional index. In order to enable the calculation of the model to be more accurate and efficient, all indexes are normalized, namely all index values are scaled between 0-1.
Step 3, establishing a regression prediction model by adopting a regression prediction algorithm, training model parameters by using training data, adjusting the model parameters by using verification data, and testing the effect of the model by using test data;
in the embodiment, an LSTM RNN regression prediction model is constructed by adopting an LSTM RNN algorithm, the LSTM RNN regression prediction model comprises 1 input layer, 3 LSTM layers, 1 full-connection layer and 1 output layer, and the 3 LSTM layers comprise 32 LSTM units; the full connection layer comprises 32 units and adopts a relu activation function; the output layer comprises 4 units and adopts a sigmoid activation function; the LSTM RNN regression prediction model loss function adopts Mean Square Error (MSE), the optimizer adopts rmsprop, and the Batch size is 128. Training was at 50 epochs.
The training data is sorted according to time, and the model can perform scoring and prediction according to the data in a period of time, so the sequence cannot be disturbed. The algorithm of the recurrent neural network is adopted because the recurrent neural network can retain information for a period of time and add the information to the calculation of the final result. The model simultaneously outputs the current score and the score of a future period of time of the input data; both the length of the input time series and the length of the predicted time series can be adjusted as parameters.
Step 4, reading monitoring indexes of the database within a period of time, preprocessing the monitoring indexes by adopting the same method as the step 2, taking the preprocessed data as the input of a regression prediction model, and sequencing the input data according to the time sequence without disturbing; the output of the model is the database health degree scoring result at the current or future moment.
As shown in fig. 2, the block diagram of the database health scoring system disclosed in the present invention includes a database monitoring index collecting module, an expert model scoring module, a preprocessing module, and a regression prediction model; the database monitoring index acquisition module acquires database monitoring indexes and scores the health degree through the expert model scoring module; the preprocessing module carries out preprocessing such as denoising and normalization on a sample set formed by collected database monitoring indexes and health degree scores, and divides data in the sample set into training data, verification data and test data; the regression prediction module trains model parameters using the training data and adjusts the model parameters using the validation data and the test data. The system continuously reads new index data, processes the new data according to a characteristic selection and scaling method during model training, then sends the new data into the model for scoring and forecasting, the model outputs scoring and forecasting results,
the preprocessing module also comprises a characteristic selection module, and the characteristic selection module is used for selecting the characteristics of the denoised data in the preprocessing module and then carrying out normalization processing.
In this embodiment, the regression prediction model is an LSTM RNN regression prediction model, the LSTM RNN regression prediction model includes 1 input layer, 3 LSTM layers, 1 fully-connected layer, and 1 output layer, and each of the 3 LSTM layers includes 32 LSTM units; the full connection layer comprises 32 units and adopts a relu activation function; the output layer comprises 4 units and a sigmoid activation function is adopted.
Example 2:
this example differs from example 1 in that: constructing a random forest regression prediction model by adopting a random forest algorithm in the step (3); the random forest regression prediction model comprises p decision trees, and the determining step of the depth q of the decision trees is as follows:
setting an upper limit value Q of the depth of the decision tree, enabling t to carry out Q times of training from 1 to Q, calculating a loss function value of each training, and taking the value of t with the minimum loss function value in Q times of training as the depth Q of the decision tree.
In this embodiment, the random forest regression prediction model includes 100 decision trees, the upper limit value of the depth of the decision tree is set to 10, and after 10 training tests, the optimal depth of the decision tree is 3.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, 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 application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams 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 foregoing descriptions of specific exemplary embodiments of the present invention have been presented for purposes of illustration and description. It is not intended to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The exemplary embodiments were chosen and described in order to explain certain principles of the invention and its practical application to enable one skilled in the art to make and use various exemplary embodiments of the invention and various alternatives and modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims and their equivalents.
Claims (2)
1. A database health degree scoring method based on machine learning is characterized by comprising the following steps:
(1) collecting database monitoring indexes, and obtaining health degree scores through an expert model; collecting original data and scoring to serve as a sample set;
(2) carrying out denoising and normalization pretreatment on the data in the sample set, and dividing the data in the sample set into training data, verification data and test data;
(3) establishing a regression prediction model by adopting a regression prediction algorithm, training model parameters by using training data, adjusting the model parameters by using verification data, and testing the effect of the model by using test data; the training data are sorted according to time, the model carries out scoring and prediction according to the data in a period of time, and the sequence cannot be disturbed; the algorithm of the recurrent neural network is adopted because the recurrent neural network can keep the information in a period of time and add the information into the calculation of the final result; the model simultaneously outputs the current score and the score of a future period of time of the input data; the length of the input time sequence and the predicted time length are used as parameters for adjustment;
(4) reading monitoring indexes of the database within a period of time, preprocessing the monitoring indexes by adopting the same method as the step (2), taking the preprocessed data as the input of a regression prediction model, and sequencing the input data according to the time sequence without disturbing; the output of the model is the database health degree scoring result at the current or future period of time;
in the step (2), denoising the data in the sample set, and then carrying out normalization processing; the characteristics are selected to calculate the importance of the monitoring index, and the front part with the minimum importance in the sample set is deletedNA monitoring index; the method specifically comprises the following steps:
(2.1) establishing a GBM regression model, and setting the number of the hoisting treesmAnd learning rateρThe loss function is mean square error;
(2.2) dividing the denoised sample set into a GBM training set and a GBM testing set, and training a GBM regression model by using the GBM training set; during the training process ifnIf the loss function of the round training is not reduced, the training is stopped; otherwise, continuing training until the loss function is not reduced any more; obtaining the importance value of each monitoring index in the sample set after the training is stopped;
(2.3) repeating steps (2.1) and (2.2)WThen, each monitoring index is obtainedWTaking the average value of the importance values as the importance value of the monitoring index;
(2.4) sorting the importance value of each monitoring index, and sequencing the front part with the minimum importance valueNDeleting the monitoring indexes;
constructing an LSTM RNN regression prediction model by adopting an LSTM RNN algorithm in the step (3), wherein the LSTM RNN regression prediction model comprises 1 input layer, 3 LSTM layers, 1 full-connection layer and 1 output layer, and the 3 LSTM layers comprise 32 LSTM units; the full connection layer comprises 32 units and adopts a relu activation function; the output layer comprises 4 units and a sigmoid activation function is adopted.
2. The scoring system of the machine learning-based database health scoring method according to claim 1, comprising a database monitoring index collection module, an expert model scoring module, a preprocessing module, a regression prediction model; the database monitoring index acquisition module acquires database monitoring indexes and scores the health degree through the expert model scoring module; the preprocessing module carries out denoising and normalization preprocessing on a sample set formed by collected database monitoring indexes and health degree scores, and divides data in the sample set into training data, verification data and test data; the regression prediction module trains model parameters by using the training data, adjusts the model parameters by using the verification data, and tests the effect of the model by using the test data;
the system also comprises a characteristic selection module, wherein the characteristic selection module is used for carrying out characteristic selection on the denoised data in the preprocessing module and then carrying out normalization processing;
the regression prediction model is an LSTM RNN regression prediction model, the LSTM RNN regression prediction model comprises 1 input layer, 3 LSTM layers, 1 full-connection layer and 1 output layer, and the 3 LSTM layers respectively comprise 32 LSTM units; the full connection layer comprises 32 units and adopts a relu activation function; the output layer comprises 4 units and a sigmoid activation function is adopted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811220356.0A CN109271374B (en) | 2018-10-19 | 2018-10-19 | Database health degree scoring method and system based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811220356.0A CN109271374B (en) | 2018-10-19 | 2018-10-19 | Database health degree scoring method and system based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109271374A CN109271374A (en) | 2019-01-25 |
| CN109271374B true CN109271374B (en) | 2021-01-26 |
Family
ID=65194440
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811220356.0A Active CN109271374B (en) | 2018-10-19 | 2018-10-19 | Database health degree scoring method and system based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109271374B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7326926B2 (en) * | 2019-06-27 | 2023-08-16 | トヨタ自動車株式会社 | LEARNING DEVICE, REHABILITATION SUPPORT SYSTEM, METHOD, PROGRAM, AND LEARNED MODEL |
| CN112303810A (en) * | 2019-08-01 | 2021-02-02 | 山东朗进科技股份有限公司 | Air conditioner health prediction method based on machine learning |
| CN111079941B (en) * | 2019-12-03 | 2024-02-20 | 武汉纺织大学 | Credit information processing method, credit information processing system, terminal and storage medium |
| CN112988529B (en) * | 2019-12-16 | 2024-08-16 | 天翼数字生活科技有限公司 | Method and system for predicting database system performance based on machine learning |
| CN111259953B (en) * | 2020-01-15 | 2023-10-20 | 云南电网有限责任公司电力科学研究院 | Equipment defect time prediction method based on capacitive equipment defect data |
| CN111581045B (en) * | 2020-03-18 | 2024-05-28 | 平安科技（深圳）有限公司 | Database anomaly monitoring method, device, computer device and storage medium |
| CN111625417B (en) * | 2020-05-12 | 2022-11-11 | 国网电力科学研究院有限公司 | Database health monitoring method and device and storage medium |
| CN111930601A (en) * | 2020-05-14 | 2020-11-13 | 国家电网有限公司客户服务中心 | Deep learning-based database state comprehensive scoring method and system |
| CN111897802B (en) * | 2020-08-10 | 2023-08-04 | 中国工商银行股份有限公司 | Database container fault positioning method and system |
| CN111949502A (en) * | 2020-08-14 | 2020-11-17 | 中国工商银行股份有限公司 | Database early warning method and device, computing equipment and medium |
| CN113806453A (en) * | 2021-09-18 | 2021-12-17 | 广东电网有限责任公司 | Automatic confirmation method, device, terminal and medium for scheduling operation ticket comprehensive order |
| KR102703138B1 (en) * | 2021-10-26 | 2024-09-04 | 연세대학교 산학협력단 | Method and apparatus for tuning database based on machine learning |
| CN114490650A (en) * | 2022-01-17 | 2022-05-13 | 成都飞机工业（集团）有限责任公司 | System, device, equipment and storage medium for predicting database space |
| CN114487848B (en) * | 2022-01-17 | 2024-05-03 | 北京和利时系统集成有限公司 | State calculation method and device for storage battery |
| CN115094193A (en) * | 2022-06-27 | 2022-09-23 | 中冶华天南京工程技术有限公司 | Intelligent molten iron pretreatment desulfurization system based on data mining |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106096767A (en) * | 2016-06-07 | 2016-11-09 | 中国科学院自动化研究所 | A kind of link travel time prediction method based on LSTM |
| CN106250306A (en) * | 2016-08-18 | 2016-12-21 | 电子科技大学 | A kind of performance prediction method being applicable to enterprise-level O&M automatization platform |
| CN106897109A (en) * | 2017-02-13 | 2017-06-27 | 云南大学 | Based on the virtual machine performance Forecasting Methodology that random forest is returned |
| CN107291668A (en) * | 2017-07-14 | 2017-10-24 | 中南大学 | A kind of subway based on machine learning passenger flow forecasting in short-term |
| CN108197011A (en) * | 2018-01-29 | 2018-06-22 | 上海洞识信息科技有限公司 | A kind of single index prediction and method for early warning based on artificial intelligence big data platform |
| CN108375808A (en) * | 2018-03-12 | 2018-08-07 | 南京恩瑞特实业有限公司 | Dense fog forecasting procedures of the NRIET based on machine learning |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6991134B2 (en) * | 2015-10-09 | 2022-01-12 | ガーダント ヘルス， インコーポレイテッド | Population-based treatment recommendations using cell-free DNA |
| CN107992454B (en) * | 2017-12-25 | 2020-11-17 | 安徽大学 | Air quality grade prediction method based on online sequential regression |
-
2018
- 2018-10-19 CN CN201811220356.0A patent/CN109271374B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106096767A (en) * | 2016-06-07 | 2016-11-09 | 中国科学院自动化研究所 | A kind of link travel time prediction method based on LSTM |
| CN106250306A (en) * | 2016-08-18 | 2016-12-21 | 电子科技大学 | A kind of performance prediction method being applicable to enterprise-level O&M automatization platform |
| CN106897109A (en) * | 2017-02-13 | 2017-06-27 | 云南大学 | Based on the virtual machine performance Forecasting Methodology that random forest is returned |
| CN107291668A (en) * | 2017-07-14 | 2017-10-24 | 中南大学 | A kind of subway based on machine learning passenger flow forecasting in short-term |
| CN108197011A (en) * | 2018-01-29 | 2018-06-22 | 上海洞识信息科技有限公司 | A kind of single index prediction and method for early warning based on artificial intelligence big data platform |
| CN108375808A (en) * | 2018-03-12 | 2018-08-07 | 南京恩瑞特实业有限公司 | Dense fog forecasting procedures of the NRIET based on machine learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109271374A (en) | 2019-01-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109271374B (en) | Database health degree scoring method and system based on machine learning | |
| CN108595913B (en) | Supervised learning method for identifying mRNA and lncRNA | |
| CN111967971B (en) | Bank customer data processing method and device | |
| CN111429415B (en) | Method for constructing efficient detection model of product surface defects based on network collaborative pruning | |
| CN113901977A (en) | Deep learning-based power consumer electricity stealing identification method and system | |
| CN116153495A (en) | Prognosis survival prediction method for immunotherapy of esophageal cancer patient | |
| CN110110663A (en) | A kind of age recognition methods and system based on face character | |
| CN107368526A (en) | A kind of data processing method and device | |
| CN114220458B (en) | Voice recognition method and device based on array hydrophone | |
| CN110704616B (en) | Equipment alarm work order identification method and device | |
| CN109543693A (en) | Weak labeling data noise reduction method based on regularization label propagation | |
| CN115865483A (en) | Abnormal behavior analysis method and device based on machine learning | |
| CN111930601A (en) | Deep learning-based database state comprehensive scoring method and system | |
| Ridhovan et al. | Disease detection in banana leaf plants using densenet and inception method | |
| CN109242165A (en) | A kind of model training and prediction technique and device based on model training | |
| CN112306816A (en) | Method and system for evaluating entity robot response based on deep learning | |
| CN113824580B (en) | Network index early warning method and system | |
| CN109934352B (en) | Automatic evolution method of intelligent model | |
| CN110852178A (en) | Piano music score difficulty identification method based on decision tree lifting | |
| CN113259158B (en) | Network flow prediction method and equipment, model construction and training method and device | |
| CN113205274B (en) | Quantitative ranking method for construction quality | |
| CN110413774A (en) | A kind of information classification approach based on genetic algorithm | |
| CN115273009A (en) | Road crack detection method and system based on deep learning | |
| CN115511670A (en) | Online education method and online education platform | |
| CN113420733A (en) | Efficient distributed big data acquisition implementation method and system |
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 |