CN107885642A - Business monitoring method and system based on machine learning - Google Patents
Business monitoring method and system based on machine learning Download PDFInfo
- Publication number
- CN107885642A CN107885642A CN201711225245.4A CN201711225245A CN107885642A CN 107885642 A CN107885642 A CN 107885642A CN 201711225245 A CN201711225245 A CN 201711225245A CN 107885642 A CN107885642 A CN 107885642A
- Authority
- CN
- China
- Prior art keywords
- business
- log
- machine learning
- daily record
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3006—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3466—Performance evaluation by tracing or monitoring
- G06F11/3476—Data logging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention provides a kind of business monitoring method and system based on machine learning.Business monitoring method based on machine learning includes：First step：The real-time logs of server are collected, distributed search engine platform is arrived in storage；Second step：Data are read from the log data stream of the magnanimity on distributed search engine platform, carry out log integrity, and the abnormal log of real-time retrieval analysis business；Third step：Using the daily record pre-processed, feature, training data model are extracted；Four steps：The data model trained according to model algorithm training module exports a sample set, then carries out algorithm using sample set and return to judge, identifies that the abnormal stability bandwidth of business monitoring is gone forward side by side industry business alarm.
Description
Technical field
The present invention relates to network communication field, it is more particularly related to a kind of business prison based on machine learning
Control method and system,
Background technology
Generally, business diary can disperse storage on different devices, increase with the magnanimity of business diary, at up to a hundred
On server, if still consulting daily record and statistical service operating index, efficiency with the conventional method for logging in every machine successively
Very low, operation is also troublesome.
In general business monitoring means formulate substantial amounts of threshold values by script or batch installation agent (agent) and detect business
Availability, common are the monitoring methods such as shell, nagios, zabbix, prometheus.But this method can only lead to
The threshold values up and down pre-set is crossed, the property abnormality and the whether wrong false information of business of server is monitored, can not collect in real time
The stability bandwidth of the business diary Data Detection operational indicator of magnanimity.
The content of the invention
The technical problems to be solved by the invention are for drawbacks described above in the prior art be present, there is provided one kind is based on machine
The business monitoring real-time detection method and system of study, wherein providing a sample collecting system, pass through training machine learning algorithm
Model, analyzed using business diary of the distributed search engine log collection instrument to server, and returned according to model in real time
Algorithm forms a set of service exception fluctuation monitoring scheme.
According to the present invention, there is provided a kind of business monitoring method based on machine learning, including：
First step：The real-time logs of server are collected, distributed search engine platform is arrived in storage；
Second step：Data are read from the log data stream of the magnanimity on distributed search engine platform, carry out daily record
Pretreatment, and the abnormal log of real-time retrieval analysis business；
Third step：Using the daily record pre-processed, feature, training data model are extracted；
Four steps：The data model trained according to model algorithm training module exports a sample set, then utilizes
Sample set carries out algorithm and returns to judge, identifies that the abnormal stability bandwidth of business monitoring is gone forward side by side industry business alarm.
Preferably, in first step, monitored in real time by log collection instrument, pull business diary caused by server,
Daily record is transmitted in log collection tool set, daily record is passed into message queue in form of a message, message queue again
Data are written to distributed search engine platform and stored.
Preferably, in second step, daily record dirty data is filtered out on distributed search engine, according to service feature and then
Extract different business diaries to be sorted out, retrieved respectively again using the business diary sorted out, analyze the different of each business
Normal logged result.
Preferably, in third step, the time series data of business diary is extracted, arrange parameter periodic quantity, extracts training
Sample point, using gradient boosting algorithm regression forecasting value, contrasted, taken different according to original abnormal point numerical value and predicted value
The often difference between point data value and predicted value, difference average value is calculated as fluctuation threshold values.
Preferably, in four steps, threshold values is fluctuated in the prediction provided according to model algorithm training module and model algorithm is instructed
Practice the fluctuation range that module calculates, judge current stability bandwidth whether in predetermined fluctuation range.
According to the present invention, a kind of service monitoring system based on machine learning is additionally provided, including：
Log collection module, for collecting the real-time logs of server, distributed search engine platform is arrived in storage, for daily record
Pretreatment module uses；
Log integrity module, for reading data from the log data stream on distributed search engine platform, carry out
Log integrity, and the abnormal log of real-time retrieval analysis business；
Model algorithm training module, for using the daily record pre-processed, extracting feature, training data model；
Business monitoring alarm module, the data model for being trained according to model algorithm training module export a sample
Collection, then carry out algorithm using sample set and return to judge, identify that the abnormal stability bandwidth of business monitoring is gone forward side by side industry business alarm.
Preferably, log collection module is monitored, pulls business diary caused by server in real time by log collection instrument,
Daily record is transmitted in log collection tool set, daily record is passed into message queue in form of a message, message queue again
Data are written to distributed search engine platform and stored.
Preferably, log integrity module filters out daily record dirty data on distributed search engine, according to service feature
And then extract different business diaries and sorted out, retrieved respectively again using the business diary sorted out, analyze each business
Abnormal log result.
Preferably, the time series data of model algorithm training module extraction business diary, arrange parameter periodic quantity, extraction
Training sample point, using gradient boosting algorithm regression forecasting value, contrasted according to original abnormal point numerical value and predicted value,
The difference between abnormal point numerical value and predicted value is taken, calculates difference average value as fluctuation threshold values.
Preferably, threshold values is fluctuated in the prediction that business monitoring alarm module provides according to model algorithm training module and model is calculated
The fluctuation range that method training module calculates, judge current stability bandwidth whether in predetermined fluctuation range.
Brief description of the drawings
With reference to accompanying drawing, and by reference to following detailed description, it will more easily have more complete understanding to the present invention
And be more easily understood its with the advantages of and feature, wherein：
Fig. 1 schematically shows the stream of the business monitoring method according to the preferred embodiment of the invention based on machine learning
Cheng Tu.
Fig. 2 schematically shows the work(of the service monitoring system according to the preferred embodiment of the invention based on machine learning
Can block diagram.
Fig. 3 schematically shows the day of the service monitoring system according to the preferred embodiment of the invention based on machine learning
The concrete operations example of will collection module.
Fig. 4 schematically shows the day of the service monitoring system according to the preferred embodiment of the invention based on machine learning
The concrete operations example of will pretreatment module.
Fig. 5 schematically shows the mould of the service monitoring system according to the preferred embodiment of the invention based on machine learning
The concrete operations example of type Algorithm for Training module.
It should be noted that accompanying drawing is used to illustrate the present invention, it is not intended to limit the present invention.Pay attention to, represent that the accompanying drawing of structure can
It can be not necessarily drawn to scale.Also, in accompanying drawing, same or similar element indicates same or similar label.
Embodiment
In order that present disclosure is more clear and understandable, with reference to specific embodiments and the drawings in the present invention
Appearance is described in detail.
<First embodiment>
Fig. 1 schematically shows the stream of the business monitoring method according to the preferred embodiment of the invention based on machine learning
Cheng Tu.
As shown in figure 1, the business monitoring method according to the preferred embodiment of the invention based on machine learning includes：
First step S1：The real-time logs of server are collected, distributed search engine platform is arrived in storage；
Second step S2：Data are read from the log data stream of the magnanimity on distributed search engine platform, carry out day
Will pre-processes, and the abnormal log of real-time retrieval analysis business；
Third step S3：Using the daily record pre-processed, feature, training data model are extracted；
Four steps S4：The data model trained according to model algorithm training module exports a sample set, Ran Houli
Algorithm is carried out with sample set return to judge, identify that the abnormal stability bandwidth of business monitoring is gone forward side by side industry business alarm.
It can specifically or additionally perform following concrete operations.
Preferably, in first step S1, monitored in real time by log collection instrument, pull business day caused by server
Will, daily record is transmitted in log collection tool set, daily record is passed into message queue in form of a message, message queue is again
Distributed search engine platform is write data into be stored.
Preferably, in second step S2, daily record dirty data is filtered out on distributed search engine, is entered according to service feature
And extract different business diaries and sorted out, retrieved respectively again using the business diary sorted out, analyze each business
Abnormal log result.
Preferably, in third step S3, the time series data of business diary is extracted, arrange parameter periodic quantity, extracts instruction
Practice sample point, using gradient boosting algorithm regression forecasting value, contrasted, taken according to original abnormal point numerical value and predicted value
Difference between abnormal point numerical value and predicted value, difference average value is calculated as fluctuation threshold values.
Preferably, threshold values and model algorithm are fluctuated in four steps S4, the prediction provided according to model algorithm training module
The fluctuation range that training module calculates, judge current stability bandwidth whether in predetermined fluctuation range.
The invention has the advantages that early warning threshold values up and down need not be pre-set, train and extract by model algorithm, export
Business samples of undulations collection, machine learning algorithm can predict current operational indicator fluctuation range in massive logs, and in real time
Statistics current business index fluctuation whether prediction fluctuation threshold values in, calculate stability bandwidth using regression algorithm, reach real-time
Service exception monitoring alarm, be advantageous to more comprehensively, the robustness of more accurate and more efficient monitoring business operation.
<Second embodiment>
Fig. 2 schematically shows the work(of the service monitoring system according to the preferred embodiment of the invention based on machine learning
Can block diagram.
As shown in Fig. 2 the service monitoring system according to the preferred embodiment of the invention based on machine learning includes：
Log collection module 10, it is mainly used in collecting the real-time logs of server, distributed search engine platform is arrived in storage,
Used for log integrity module；
Log integrity module 20, it is mainly used in reading from the log data stream of the magnanimity on distributed search engine platform
Access evidence, carry out log integrity, and the abnormal log of real-time retrieval analysis business；
Model algorithm training module 30, it is mainly used in, using the daily record pre-processed, extracting feature, training data mould
Type；
Business monitoring alarm module 40, it is mainly used in the data model output one trained according to model algorithm training module
Individual sample set, then carry out algorithm and return to judge, identify that the abnormal stability bandwidth of business monitoring is gone forward side by side industry business alarm.
The concrete operations example or additional operations of modules is detailed below.
Log collection module 10, such as accompanying drawing 3, by all business diary real-time collectings caused by server to distributed search
Engine platform, step are as follows：
(1) monitored in real time by instrument, pull business diary caused by server；
(2) daily record is transmitted in log collection tool set, daily record is passed into message queue in form of a message, disappeared
Breath queue writes data into the storage of distributed search engine platform again.
Log integrity module 20, as shown in Figure 4, for carrying out log integrity, including data mistake after log collection
Filter, daily record classification, retrieval analysis etc..Step is as follows：
(1) data filtering, daily record dirty data is filtered out on distributed search engine, avoids influenceing final algorithm output
Sample causes error.
(2) daily record is classified, and according to service feature and then is extracted different business diaries and is sorted out；
(3) retrieval analysis, retrieved respectively again using the business diary sorted out, analyze the abnormal log knot of each business
Fruit.
Model algorithm training module 30, as shown in Figure 5, step is as follows：
(1) time series data of business diary is extracted.
(2) arrange parameter periodic quantity, training sample point is extracted, uses gradient boosting algorithm regression forecasting value.Such as GBDT
Gradient lifts regression algorithm.
(3) threshold values is fluctuated to calculate：Contrasted according to original abnormal point numerical value and predicted value, take abnormal point numerical value
Difference between predicted value, difference average value is calculated as fluctuation threshold values.If for example, difference of actual value and predicted value
It is then abnormity point more than or less than threshold value.
Threshold values and model algorithm are fluctuated in business monitoring alarm module 40, the prediction provided according to model algorithm training module 30
Whether the fluctuation range that training module 30 calculates, machine learning may determine that current stability bandwidth in predetermined normal ripple
In dynamic scope.If result is normal, illustrate that current each operational indicator is normal, if exceeding or falling below fluctuation range, illustrate to work as
Preceding service operation has exception, triggers warning system.
The invention has the advantages that early warning threshold values up and down need not be pre-set, train and extract by model algorithm, export
Business samples of undulations collection, machine learning algorithm can predict current operational indicator fluctuation range in massive logs, and in real time
Statistics current business index fluctuation whether prediction fluctuation threshold values in, calculate stability bandwidth using regression algorithm, reach real-time
Service exception monitoring alarm, be advantageous to more comprehensively, the robustness of more accurate and more efficient monitoring business operation.
Furthermore, it is necessary to explanation, unless otherwise indicated, the otherwise term in specification " first ", " second ", " the 3rd "
Be used only for distinguishing each component in specification, element, step etc. Deng description, without be intended to indicate that each component, element,
Logical relation or ordinal relation between step etc..
It is understood that although the present invention is disclosed as above with preferred embodiment, but above-described embodiment and it is not used to
Limit the present invention.For any those skilled in the art, without departing from the scope of the technical proposal of the invention,
Many possible changes and modifications are all made to technical solution of the present invention using the technology contents of the disclosure above, or are revised as
With the equivalent embodiment of change.Therefore, every content without departing from technical solution of the present invention, the technical spirit pair according to the present invention
Any simple modifications, equivalents, and modifications made for any of the above embodiments, still fall within the scope of technical solution of the present invention protection
It is interior.
Claims (10)
- A kind of 1. business monitoring method based on machine learning, it is characterised in that including：First step：The real-time logs of server are collected, distributed search engine platform is arrived in storage；Second step：Data are read from the log data stream of the magnanimity on distributed search engine platform, daily record is carried out and locates in advance Reason, and the abnormal log of real-time retrieval analysis business；Third step：Using the daily record pre-processed, feature, training data model are extracted；Four steps：The data model trained according to model algorithm training module exports a sample set, then utilizes sample Collection carries out algorithm and returns to judge, identifies that the abnormal stability bandwidth of business monitoring is gone forward side by side industry business alarm.
- 2. the business monitoring method according to claim 1 based on machine learning, it is characterised in that in first step, lead to Cross log collection instrument to monitor in real time, pull business diary caused by server, daily record is passed in log collection tool set It is defeated, daily record is passed into message queue in form of a message, message queue writes data into distributed search engine platform again Stored.
- 3. the business monitoring method according to claim 1 or 2 based on machine learning, it is characterised in that in second step, Daily record dirty data is filtered out on distributed search engine, according to service feature and then different business diaries is extracted and is returned Class, retrieved respectively again using the business diary sorted out, analyze the abnormal log result of each business.
- 4. the business monitoring method according to claim 1 or 2 based on machine learning, it is characterised in that in third step, The time series data of business diary is extracted, arrange parameter periodic quantity, training sample point is extracted, is returned using gradient boosting algorithm Predicted value, contrasted according to original abnormal point numerical value and predicted value, take the difference between abnormal point numerical value and predicted value Value, difference average value is calculated as fluctuation threshold values.
- 5. the business monitoring method according to claim 1 or 2 based on machine learning, it is characterised in that in four steps, The fluctuation range that the prediction fluctuation threshold values and model algorithm training module provided according to model algorithm training module calculates, judges Whether current stability bandwidth is in predetermined fluctuation range.
- A kind of 6. service monitoring system based on machine learning, it is characterised in that including：Log collection module, for collecting the real-time logs of server, storage is arrived distributed search engine platform, located in advance for daily record Module is managed to use；Log integrity module, for reading data from the log data stream on distributed search engine platform, carry out daily record Pretreatment, and the abnormal log of real-time retrieval analysis business；Model algorithm training module, for using the daily record pre-processed, extracting feature, training data model；Business monitoring alarm module, the data model for being trained according to model algorithm training module export a sample set, Then carry out algorithm using sample set and return to judge, identify that the abnormal stability bandwidth of business monitoring is gone forward side by side industry business alarm.
- 7. the service monitoring system according to claim 6 based on machine learning, it is characterised in that log collection module is led to Cross log collection instrument to monitor in real time, pull business diary caused by server, daily record is passed in log collection tool set It is defeated, daily record is passed into message queue in form of a message, message queue writes data into distributed search engine platform again Stored.
- 8. the service monitoring system based on machine learning according to claim 6 or 7, it is characterised in that log integrity Module filters out daily record dirty data on distributed search engine, according to service feature and then extracts different business diaries and enters Row is sorted out, and is retrieved respectively again using the business diary sorted out, analyzes the abnormal log result of each business.
- 9. the service monitoring system based on machine learning according to claim 6 or 7, it is characterised in that model algorithm is instructed Practice the time series data of module extraction business diary, arrange parameter periodic quantity, extract training sample point, lifted and calculated using gradient Method regression forecasting value, contrasted according to original abnormal point numerical value and predicted value, take abnormal point numerical value and predicted value it Between difference, calculate difference average value as fluctuation threshold values.
- 10. the service monitoring system based on machine learning according to claim 6 or 7, it is characterised in that business monitoring report The fluctuation model that the prediction fluctuation threshold values and model algorithm training module that alert module provides according to model algorithm training module calculate Enclose, judge current stability bandwidth whether in predetermined fluctuation range.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711225245.4A CN107885642A (en) | 2017-11-29 | 2017-11-29 | Business monitoring method and system based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711225245.4A CN107885642A (en) | 2017-11-29 | 2017-11-29 | Business monitoring method and system based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107885642A true CN107885642A (en) | 2018-04-06 |
Family
ID=61775936
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711225245.4A Pending CN107885642A (en) | 2017-11-29 | 2017-11-29 | Business monitoring method and system based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107885642A (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108897674A (en) * | 2018-07-12 | 2018-11-27 | 郑州云海信息技术有限公司 | A kind of log analysis method and device |
| CN109039727A (en) * | 2018-07-24 | 2018-12-18 | 中国银行股份有限公司 | Message queue monitoring method and device based on deep learning |
| CN109086186A (en) * | 2018-07-24 | 2018-12-25 | 中国联合网络通信集团有限公司 | log detection method and device |
| CN109582529A (en) * | 2018-09-29 | 2019-04-05 | 阿里巴巴集团控股有限公司 | A kind of setting method and device of alarm threshold value |
| CN109656922A (en) * | 2018-12-19 | 2019-04-19 | 国网北京市电力公司 | Data processing method and device |
| CN109753408A (en) * | 2018-12-11 | 2019-05-14 | 江阴逐日信息科技有限公司 | A kind of process predicting abnormality method based on machine learning |
| CN109756395A (en) * | 2018-12-28 | 2019-05-14 | 易票联支付有限公司 | A kind of business datum monitoring method and system |
| CN110188015A (en) * | 2019-04-04 | 2019-08-30 | 北京升鑫网络科技有限公司 | A kind of host access relation abnormal behaviour self-adapting detecting device and its monitoring method |
| CN110460591A (en) * | 2019-07-26 | 2019-11-15 | 南京理工大学 | Based on the CDN Traffic anomaly detection device and method for improving separation time memory network |
| CN110619406A (en) * | 2018-06-19 | 2019-12-27 | 中移信息技术有限公司 | Method and device for determining business abnormity |
| CN110928718A (en) * | 2019-11-18 | 2020-03-27 | 上海维谛信息科技有限公司 | Exception handling method, system, terminal and medium based on correlation analysis |
| CN110958136A (en) * | 2019-11-11 | 2020-04-03 | 国网山东省电力公司信息通信公司 | Deep learning-based log analysis early warning method |
| CN111177095A (en) * | 2019-12-10 | 2020-05-19 | 中移（杭州）信息技术有限公司 | Log analysis method and device, computer equipment and storage medium |
| CN111726248A (en) * | 2020-05-29 | 2020-09-29 | 北京宝兰德软件股份有限公司 | Alarm root cause positioning method and device |
| CN111897788A (en) * | 2020-07-14 | 2020-11-06 | 中电福富信息科技有限公司 | Log retrieval analysis and visual mining method based on algorithm selection |
| CN112506750A (en) * | 2020-09-28 | 2021-03-16 | 国网甘肃省电力公司信息通信公司 | Big data processing system for mass log analysis and early warning |
| CN112579728A (en) * | 2020-12-18 | 2021-03-30 | 成都民航西南凯亚有限责任公司 | Behavior abnormity identification method and device based on mass data full-text retrieval |
| CN114185848A (en) * | 2020-09-15 | 2022-03-15 | 中国移动通信集团山东有限公司 | Interface state generation method and device, computer equipment and storage medium |
| CN115292150A (en) * | 2022-10-09 | 2022-11-04 | 帕科视讯科技(杭州)股份有限公司 | Method for monitoring health state of IPTV EPG service based on AI algorithm |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105046327A (en) * | 2015-06-03 | 2015-11-11 | 王宝会 | Intelligent electrical network information system and method based on machine learning technology |
| CN105721194A (en) * | 2016-01-13 | 2016-06-29 | 广州衡昊数据科技有限公司 | Intelligent positioning system of faults and hidden dangers of mobile network |
| CN106844138A (en) * | 2016-12-14 | 2017-06-13 | 北京奇艺世纪科技有限公司 | O＆M warning system and method |
| CN107229976A (en) * | 2017-06-08 | 2017-10-03 | 郑州云海信息技术有限公司 | A kind of distributed machines learning system based on spark |
| US20170300473A1 (en) * | 2016-04-18 | 2017-10-19 | Microsoft Technology Licensing, Llc | Correlating distinct events using linguistic analysis |
-
2017
- 2017-11-29 CN CN201711225245.4A patent/CN107885642A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105046327A (en) * | 2015-06-03 | 2015-11-11 | 王宝会 | Intelligent electrical network information system and method based on machine learning technology |
| CN105721194A (en) * | 2016-01-13 | 2016-06-29 | 广州衡昊数据科技有限公司 | Intelligent positioning system of faults and hidden dangers of mobile network |
| US20170300473A1 (en) * | 2016-04-18 | 2017-10-19 | Microsoft Technology Licensing, Llc | Correlating distinct events using linguistic analysis |
| CN106844138A (en) * | 2016-12-14 | 2017-06-13 | 北京奇艺世纪科技有限公司 | O＆M warning system and method |
| CN107229976A (en) * | 2017-06-08 | 2017-10-03 | 郑州云海信息技术有限公司 | A kind of distributed machines learning system based on spark |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110619406A (en) * | 2018-06-19 | 2019-12-27 | 中移信息技术有限公司 | Method and device for determining business abnormity |
| CN108897674A (en) * | 2018-07-12 | 2018-11-27 | 郑州云海信息技术有限公司 | A kind of log analysis method and device |
| CN109039727B (en) * | 2018-07-24 | 2021-08-06 | 中国银行股份有限公司 | Deep learning-based message queue monitoring method and device |
| CN109039727A (en) * | 2018-07-24 | 2018-12-18 | 中国银行股份有限公司 | Message queue monitoring method and device based on deep learning |
| CN109086186A (en) * | 2018-07-24 | 2018-12-25 | 中国联合网络通信集团有限公司 | log detection method and device |
| CN109086186B (en) * | 2018-07-24 | 2022-02-15 | 中国联合网络通信集团有限公司 | Log detection method and device |
| CN109582529A (en) * | 2018-09-29 | 2019-04-05 | 阿里巴巴集团控股有限公司 | A kind of setting method and device of alarm threshold value |
| CN109753408A (en) * | 2018-12-11 | 2019-05-14 | 江阴逐日信息科技有限公司 | A kind of process predicting abnormality method based on machine learning |
| CN109753408B (en) * | 2018-12-11 | 2022-03-25 | 江阴逐日信息科技有限公司 | Flow abnormity prediction method based on machine learning |
| CN109656922A (en) * | 2018-12-19 | 2019-04-19 | 国网北京市电力公司 | Data processing method and device |
| CN109756395A (en) * | 2018-12-28 | 2019-05-14 | 易票联支付有限公司 | A kind of business datum monitoring method and system |
| CN110188015A (en) * | 2019-04-04 | 2019-08-30 | 北京升鑫网络科技有限公司 | A kind of host access relation abnormal behaviour self-adapting detecting device and its monitoring method |
| CN110460591B (en) * | 2019-07-26 | 2021-11-09 | 南京理工大学 | CDN flow abnormity detection device and method based on improved hierarchical time memory network |
| CN110460591A (en) * | 2019-07-26 | 2019-11-15 | 南京理工大学 | Based on the CDN Traffic anomaly detection device and method for improving separation time memory network |
| CN110958136A (en) * | 2019-11-11 | 2020-04-03 | 国网山东省电力公司信息通信公司 | Deep learning-based log analysis early warning method |
| CN110928718A (en) * | 2019-11-18 | 2020-03-27 | 上海维谛信息科技有限公司 | Exception handling method, system, terminal and medium based on correlation analysis |
| CN110928718B (en) * | 2019-11-18 | 2024-01-30 | 上海维谛信息科技有限公司 | Abnormality processing method, system, terminal and medium based on association analysis |
| CN111177095B (en) * | 2019-12-10 | 2023-10-27 | 中移（杭州）信息技术有限公司 | Log analysis method, device, computer equipment and storage medium |
| CN111177095A (en) * | 2019-12-10 | 2020-05-19 | 中移（杭州）信息技术有限公司 | Log analysis method and device, computer equipment and storage medium |
| CN111726248A (en) * | 2020-05-29 | 2020-09-29 | 北京宝兰德软件股份有限公司 | Alarm root cause positioning method and device |
| CN111897788A (en) * | 2020-07-14 | 2020-11-06 | 中电福富信息科技有限公司 | Log retrieval analysis and visual mining method based on algorithm selection |
| CN114185848A (en) * | 2020-09-15 | 2022-03-15 | 中国移动通信集团山东有限公司 | Interface state generation method and device, computer equipment and storage medium |
| CN112506750A (en) * | 2020-09-28 | 2021-03-16 | 国网甘肃省电力公司信息通信公司 | Big data processing system for mass log analysis and early warning |
| CN112579728A (en) * | 2020-12-18 | 2021-03-30 | 成都民航西南凯亚有限责任公司 | Behavior abnormity identification method and device based on mass data full-text retrieval |
| CN115292150B (en) * | 2022-10-09 | 2023-04-07 | 帕科视讯科技(杭州)股份有限公司 | Method for monitoring health state of IPTV EPG service based on AI algorithm |
| CN115292150A (en) * | 2022-10-09 | 2022-11-04 | 帕科视讯科技(杭州)股份有限公司 | Method for monitoring health state of IPTV EPG service based on AI algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107885642A (en) | Business monitoring method and system based on machine learning | |
| CN110366031B (en) | Vision-based abnormal state monitoring and fault diagnosis method for MES (manufacturing execution system) of digital workshop | |
| CN111507376B (en) | Single-index anomaly detection method based on fusion of multiple non-supervision methods | |
| CN108537544A (en) | A kind of transaction system method for real-time monitoring and its monitoring system | |
| CN100412993C (en) | System for intelligent maintaince of muclear power paltn based on state monitoring | |
| CN112925279A (en) | Fault comprehensive analysis system based on MES system | |
| CN112787860B (en) | Root alarm analysis and identification method and device | |
| AU2019275633B2 (en) | System and method of automated fault correction in a network environment | |
| CN108398934B (en) | equipment fault monitoring system for rail transit | |
| CN110738255A (en) | device state monitoring method based on clustering algorithm | |
| CN106454331A (en) | A video signal quality detection system and method | |
| CN108965340A (en) | A kind of industrial control system intrusion detection method and system | |
| AU2013273841A1 (en) | A risk assessment method and system for the security of an industrial installation | |
| CN110580492A (en) | Track circuit fault precursor discovery method based on small fluctuation detection | |
| CN116382217A (en) | Intelligent operation and maintenance monitoring system for production line | |
| CN111666978B (en) | Intelligent fault early warning system for IT system operation and maintenance big data | |
| CN107844067A (en) | A kind of gate of hydropower station on-line condition monitoring control method and monitoring system | |
| CN116292241A (en) | Fault diagnosis early warning method and system for oil delivery pump unit | |
| CN113962308A (en) | Aviation equipment fault prediction method | |
| CN110728381A (en) | Intelligent power plant inspection method and system based on RFID and data processing | |
| CN117150418B (en) | Transformer operation detection period formulation method and system based on state characteristic fault tree | |
| CN115883163A (en) | Network safety alarm monitoring method | |
| CN117170303B (en) | PLC fault intelligent diagnosis maintenance system based on multivariate time sequence prediction | |
| CN109115271B (en) | Digit control machine tool remote monitoring system | |
| CN106910334B (en) | Method and device for predicting road section conditions based on big data |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180406 |
|
| RJ01 | Rejection of invention patent application after publication |