CN109685101A - A kind of adaptive acquisition method of multidimensional data and system - Google Patents
A kind of adaptive acquisition method of multidimensional data and system Download PDFInfo
- Publication number
- CN109685101A CN109685101A CN201811345413.8A CN201811345413A CN109685101A CN 109685101 A CN109685101 A CN 109685101A CN 201811345413 A CN201811345413 A CN 201811345413A CN 109685101 A CN109685101 A CN 109685101A
- Authority
- CN
- China
- Prior art keywords
- data
- acquisition
- multidimensional
- dimensional
- adaptive
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
Abstract
The invention belongs to big data acquisition technique field, a kind of adaptive acquisition method of multidimensional data and system are disclosed.The present invention carries out dimensionality reduction to multidimensional data using dimensionality reduction technology, by multidimensional data dimensionality reduction to one-dimensional, obtains the one-dimensional principal component of multidimensional data；It is input in one-dimensional self-adapting data gathering algorithm using the one-dimensional principal component of original multi-dimensional data as the reference data for judging data variation；Utilize the collection process of one-dimensional self-adapting data gathering algorithm adjustment multidimensional big data.Because the PCA in dimensionality reduction technology carries out dimensionality reduction using the covariance of multidimensional data, and the frequency acquisition for adjusting data in one-dimensional data acquisition is also based on the variation size adjustment of data, so the present invention is feasible, and tests and show feasibility and validity of the invention.Application of the invention is extremely wide, including all business scenarios acquired using multidimensional big data, the performance of data acquisition can be improved on the basis of guaranteeing accuracy of data acquisition, and then improve the efficiency of applied business.
Description
Technical field
The invention belongs to big data acquisition technique field more particularly to a kind of adaptive acquisition method of multidimensional data and it is
System.
Background technique
Currently, the prior art commonly used in the trade is such that under current internet application scenarios, data become increasingly to weigh
It wants.Data are to support very multiple services optimized integration, and data acquisition is the performance bottle of most of operation systems associated with the data
Neck.Such as in network safety filed by acquisition communication data, and then the feature of data is analyzed, to detect attack and invasion, from
And it realizes network system and protects.But big data era, data have 5V characteristic, traditional based on the statistical sampling methods (period
Sampling, Poisson sampling and random sampling) data acquisition be no longer satisfied current demand.It further sees, with artificial intelligence
The development of energy, the every aspect that the intelligentized penetration of service to people is lived, therefore the target of current data acquisition is usually
The data of multidimensional rather than one-dimensional data.In conclusion adaptive multidimensional big data acquisition method is current big data era
The problem of institute's urgent need to resolve.
In addition to traditional collecting method based on statistical sampling.It has been proposed being directed in existing work one-dimensional
The adaptive acquisition method of data, such as the prediction algorithm based on regression analysis and time series analysis.It may be implemented adaptive
The frequency for the adjustment data acquisition answered improves data and adopts to reduce data collection capacity on the basis of guaranteeing accuracy of data acquisition
Collect performance.But these methods cannot be directed to multidimensional data, can not solve the adaptive acquisition problems of multidimensional data.One-dimensional
It is the variation based on data itself during the adaptive acquisition adjustment of data: when data in self-adapting data gathering algorithm
Data volume is big, then improves data acquiring frequency, acquires more data, to guarantee accuracy of data acquisition；And work as the data of data
It measures small, then reduces data acquiring frequency, reduce data acquisition to the burden of application system.But for multidimensional data, data acquisition
It is adaptive adjustment during, it should be a unsolved problem using which kind of data in multidimensional as base reference data.
It is not provided in current work in solution, that is, current research work not about the adaptive of multidimensional data acquisition
Answer acquisition scheme.
In conclusion problem of the existing technology is: most in the current enterprise or traditional statistical sampling side of use
Method is acquired, such as the period, at random, layering and Poisson sampling etc..It can directly acquire multidimensional data but can not achieve certainly
Adapt to acquisition.And the data acquisition for being directed to data analysis at present is all to acquire entirely.But in current big data era, data
It measures increasing, needs to use adaptive sampling to solve to reduce data collection capacity.So we have proposed be directed to multidimensional number
According to adaptive sampling method.Do not propose the adaptive acquisition method for being directed to multidimensional big data.But multidimensional big data
Adaptive acquisition method is very high in the value of the following big data era, can to avoid data acquisition performance bottleneck problem, from
And the realization of preferably supporting business.
It solves the difficulty and meaning of above-mentioned technical problem: not being directed to multidimensional big data acquisition problems in current work
It is proposed associated solutions.It is current usually or using traditional sampling algorithm (period, random and Poisson sampling) to realize data
Acquisition.But the problem is that adaptive acquisition adjustment can not be realized based on context, to acquire data volume reducing
While also reduce data acquisition precision.And it currently proposes adaptive based on regression forecasting or time series analysis
It answers acquisition scheme both in one-dimensional data, and may not apply to multidimensional data.Because it is not solved in multidimensional data
The problem of finding reference data, i.e. none reference data are used to adjust the process of data acquisition, also just cannot achieve multidimensional
The adaptive acquisition of big data.
Summary of the invention
In view of the problems of the existing technology, the present invention provides a kind of adaptive acquisition method of multidimensional data and systems.
The invention is realized in this way a kind of adaptive acquisition method of multidimensional data, is adaptively acquired using one-dimensional data
Algorithm realizes the adaptive adjustment of multidimensional data acquisition, realizes the acquisition of multidimensional data in conjunction with dimensionality reduction technology.The multidimensional number
Include: according to adaptive acquisition method
Step 1 utilizes dimensionality reduction technology: dimensionality reduction is carried out to the multidimensional data of acquisition target, by multidimensional data dimensionality reduction to one
Dimension；
Step 2, the one-dimensional principal component obtained using original multi-dimensional Data Dimensionality Reduction, as the base value for judging data variation
According to adjusting the frequency acquisition of multidimensional data, realize the adaptive acquisition of multidimensional data.
Further, realize that the multidimensional data under current big data scene adaptively acquires.
The adaptive acquisition method of multidimensional data specifically includes:
(1) target for assuming data acquisition is Yi=(y1, y2, y3..., yn), it is multidimensional data, yj(j=1,2,
3 ..., be n) target data per one-dimensional.Wherein i is t+1, t+2, t+3 ..., t+Nr；Wherein t is some acquisition time
Point.Defining data predicted value isYiIt is actual value.NrIt is actual acquisition number, NpIt is Prediction and Acquisition number, and Np=Nr；
(2) by multidimensional data dimensionality reduction to one-dimensional principal component, as reference data, PCA algorithm is mainly utilized herein:
yi=PCA (Yi)
(3) mean value based on one-dimensional principal component calculating target data actual value and predicted value is as follows:
(4) the average ratio R of predicted value and actual valueMIt indicates: working as RMIt is 1, indicates that data do not change substantially；And work as RM
Significantly greater than 1 and the variation that represents target data values less than 1 it is big.
Theoretically, data variation ratio can also be calculated with variance, use RDIt indicates.Calculating process is as follows:
(5) the variation ratio based on data, the specific method of adjustment of data acquisition are as follows:
Wherein TiRepresent current sampling interval, Ti-1Represent the previous sampling interval.TincRepresent the value added in sampling interval；
TdecRepresent the reduced value in sampling interval.ThruAnd ThrlIt is the threshold value for judging data variation.Work as RMGreater than ThrlAnd it is less than
Thru, represent that data variation is small, and acquisition interval should increase Tinc；Work as RMGreater than ThruOr it is less than Thrl, it is big to represent data variation,
Acquisition interval should reduce Tdec。TmmaxAnd TminRepresent the maximum value and minimum value of data acquisition intervals, that is, data acquisition
The adjustment maximum at interval is no more than Tmmax, minimum cannot be below Tmin, the constraint condition of adjustment is acquired as data.
Another object of the present invention is to provide a kind of social networks using the adaptive acquisition method of the multidimensional data
Recommend control system.
Another object of the present invention is to provide a kind of intrusion detections using the adaptive acquisition method of the multidimensional data
System.
Another object of the present invention is to provide a kind of assets portraits using the adaptive acquisition method of the multidimensional data
Acquisition system.
Another object of the present invention is to provide a kind of any business using the adaptive acquisition method of the multidimensional data
Application system.
The present invention provides the method for optimizing data acquisition under big data scene.Big data era, data have number
Greatly according to amount, the features such as flow velocity is fast.Therefore more optimal method is needed to improve the performance of data acquisition, preferably to support industry
The realization of business.
In conclusion advantages of the present invention and good effect are as follows: by dimensionality reduction technology by original multi-dimensional Data Dimensionality Reduction to one
Principal component is tieed up, in conjunction with the adaptive gathering algorithm of one-dimensional data, realizes the adaptive acquisition method of multidimensional big data.Of the invention answers
With extremely wide, including all scenes acquired using multidimensional big data.Such as network security, recommender system, social networks etc..This hair
It is bright by by original multi-dimensional Data Dimensionality Reduction, using the collection process of one-dimensional principal component adjustment data, to realize adaptive multidimensional
Collecting method.Advantages of the present invention and good effect are as follows: realize Automatic-searching base value using dimensionality reduction technology (such as PCA)
According to, and the adaptive acquisition method of one-dimensional data is combined, realize the adaptive acquisition method of multidimensional data.Most important advantage is
Solve the acquisition problems of multidimensional data under big data scene.The present invention realizes that simply applicable scene is very more, any to relate to
And the field of multidimensional big data acquisition, and the computation complexity of the dimensionality reduction technologies such as PCA is low.With the development of big data era,
Present invention can apply to any required scenes for realizing multidimensional data acquisition, on the basis of guaranteeing accuracy of data acquisition, greatly
It is big to reduce data collection capacity, to improve the performance of big data acquisition, and data acquisition operations itself are reduced to application system
Burden.Data required for adaptively acquiring under the acquisition demand scene of multidimensional big data may be implemented.Data acquisition
Application field is very wide, including recommender system, social networks, intrusion detection etc..Traditional data acquisition is generally based on statistics
The methods of sampling.But in current big data era, data have 5V characteristic (Volume, Variety, Velocity, Value,
Veracity).Simultaneously as the development of the technologies such as artificial intelligence, while the demand of required data is acquired in multiple dimensions
It greatly increases；It needs to find the adaptive method of one kind on the basis of not influencing multidimensional data accuracy, substantially reduces data
Collection capacity improve acquisition performance to reduce data acquisition to the burden of application system.The present invention is mainly in combination with machine learning
In regression analysis and dimensionality reduction technology, devise a kind of general adaptive acquisition method of multidimensional data, can satisfy big number
According to the acquisition demand in epoch, the adaptive acquisition of multidimensional data is realized under specific business scenario.
Detailed description of the invention
Fig. 1 is the adaptive acquisition method flow chart of multidimensional data provided in an embodiment of the present invention.
Fig. 2 is the adaptive acquisition system structural schematic diagram of multidimensional data provided in an embodiment of the present invention；
Fig. 3 is the principle comparison diagram result schematic diagram of one-dimensional data acquisition and multidimensional data acquisition provided by the invention.
Fig. 4 is the step result schematic diagram of multidimensional data acquisition provided by the invention.
Fig. 5 is the self-adapting data collection result schematic diagram of one-dimensional principal component provided in an embodiment of the present invention.
Fig. 6 is the self-adapting data collection result schematic diagram of internal storage data provided in an embodiment of the present invention.
Fig. 7 is the self-adapting data collection result schematic diagram that CPU provided in an embodiment of the present invention occupies data.
Fig. 8 is the self-adapting data collection result schematic diagram of amount of batteries data provided in an embodiment of the present invention.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to embodiments, to the present invention
It is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to
Limit the present invention.
The present invention greatly reduces the collection capacity of data on the basis of guaranteeing multidimensional data acquisition precision, to prevent from counting
The normal operation of application system is influenced according to acquisition, improves the performance of multidimensional data acquisition.
Application principle of the invention is explained in detail with reference to the accompanying drawing.
As shown in Figure 1, the adaptive acquisition method of multidimensional data provided in an embodiment of the present invention the following steps are included:
S101: it utilizes dimensionality reduction technology: dimensionality reduction being carried out to multiple data i.e. multidimensional data, by multidimensional data dimensionality reduction to one
Dimension, that is, obtain the one-dimensional principal component of multidimensional data；
S102: using the one-dimensional principal component of this original multi-dimensional data, as the reference data for judging data variation, adjustment
The frequency acquisition of multidimensional data.
As shown in Fig. 2, the adaptive acquisition method of multidimensional data main four provided in an embodiment of the present invention extend it is one-dimensional from
Data acquisition is adapted to, allows to be applied under multidimensional data acquisition scene.Also or say it is to combine dimensionality reduction technology that multidimensional is big
Data collection problems are reduced to one-dimensional data acquisition problems.
Present invention mainly solves such problems: under big data scene, solve since the collection capacity of big data is big,
The problem of burden is caused to application system；Base is determined when the adaptive acquisition adjustment of one-dimensional data gathering algorithm acquisition multidimensional data
The problem of quasi- data.Its main target is to be directed to multidimensional data, on the basis of guaranteeing certain acquisition precision, is substantially reduced
The data volume of acquisition, to improve the performance of data acquisition, the operation for preventing data from acquiring influences the normal operation of application system.
Be directed to the adaptive acquisition problems of one-dimensional data, some research work have been proposed at present, be mostly using regression analysis and
The prediction of time series analysis (such as arma modeling (Auto-Regressive Moving Average Model)) technological means
The variation of data, according to the process of the variation adjustment data acquisition of one-dimensional data itself.At the time of data volume is big, data are improved
The frequency of acquisition, to acquire more data；At the time of data volume is small, the frequency of data acquisition is reduced, reduction needs to acquire
Data.Utilize the present invention, it is ensured that the precision of data acquisition, and the collection capacity of data is reduced, it improves data and adopts
Collect performance, can be adapted under big data scene.But it under currently acquisition environment, is especially dug for machine learning and data
Pick technology when handling data, needs to acquire multiple data under same time dimension, that is, should be in a time
It puts while acquiring multiple data, that is, multiple feature, that is, multidimensional datas.But if it is multiple data are acquired simultaneously, acquire system
System should be using which data as reference data, that is, the acquisition of multidimensional data should be adjusted according to the variation of which data
Frequency is a problem.
Using dimensionality reduction technology (such as PCA (Principal Component Analysis, principal component analysis)) to multiple
Data i.e. multidimensional data carry out dimensionality reduction, by multidimensional data dimensionality reduction to one-dimensional, that is, obtain multidimensional data it is one-dimensional it is main at
Point.Then the one-dimensional principal component for utilizing this original multi-dimensional data is further adjusted as the reference data for judging data variation
The frequency acquisition of multidimensional data；PCA using be that the covariance of multidimensional data carries out dimensionality reduction, and adjust the frequency acquisition of data
It is to be adjusted based on the variation size of data.As shown in figure 3, adaptive acquisition problems of the present invention by comparison one-dimensional data,
In conjunction with dimensionality reduction technology, by multidimensional data dimensionality reduction to one-dimensional principal component, using one-dimensional principal component as reference data, with it is one-dimensional it is main at
That divides changes to adjust the collection process of multidimensional data.
When acquiring one-dimensional data, self-adapting data gathering algorithm is to be gone adjustment should according to the variation of this one-dimensional data itself
The frequency acquisition of data.But the acquisition for multidimensional data, acquisition system should be adjusted based on any data
It is a problem.It is directed to the acquisition problems of multidimensional data, it is believed that there are two types of solutions: 1, in the multi-dimensional data of acquisition
In, a main data is found as adjustment acquisition frequency in the reference data adaptively adjusted, that is, one-dimensional adaptive acquisition
When rate, the frequency of multidimensional data acquisition is adjusted according to the variation of main data.It is done so that may be inaccuracy,
It is how a main data to be determined in multiple data first, because needing the main of selection under different business scenarios
Data is different, needs to carry out a large amount of statistics to each data in multidimensional data under each business scenario to calculate and could obtain
To main data, so the selection method of a general main data can not be designed.Moreover this main
The different variation tendency for surely reacting other data of data, the accuracy that will lead to the data of acquisition in this case is low, no
It is able to satisfy the demand of multidimensional big data acquisition.Thus second workaround and core of the invention are devised.2, drop is utilized
Dimension technology (such as PCA (Principal Component Analysis, principal component analysis)) is to multiple data i.e. multidimensional
Data carry out dimensionality reduction, by multidimensional data dimensionality reduction to one-dimensional principal component that is one-dimensional, that is, obtaining multidimensional data.Then this is utilized
The one-dimensional principal component of original multi-dimensional data further adjusts the acquisition of multidimensional data as the reference data for judging data variation
Frequency.Doing so is that comparison is reasonable because PCA using be that the covariance of multidimensional data carries out dimensionality reduction, and adjusts data
Frequency acquisition is also based on the variation size of data to adjust.Specific operating procedure is as shown in Figure 4: (1) by original multi-dimensional number
According to utilization PCA dimensionality reduction to one-dimensional principal component；(2) using one-dimensional principal component as reference data, original multi-dimensional data are as original number
According to being input in one-dimensional self-adapting data gathering algorithm；(3) multidimensional data is acquired using one-dimensional self-adapting data gathering algorithm.
Traditional methods of sampling can reduce the collection capacity of data, but can not achieve the adjustment of adaptive data acquisition, so
One-dimensional data gathering algorithm used in multidimensional data acquisition scheme can be using any one-dimensional self-adapting data gathering algorithm such as
Regression analysis or time series analysis.
Application effect of the invention is explained in detail below with reference to emulation.
This experiment is emulated in mobile terminal, the case where when mainly considering while acquiring four kinds of data, including in system
Occupancy is deposited, system CPU occupies, total amount of batteries, cpu temperature.First with PCA by 4 D data dimensionality reduction to one-dimensional principal component；Benefit
Multidimensional data frequency acquisition is adjusted with the adaptive gathering algorithm of one-dimensional data (regression analysis), Simulation results is obtained, goes forward side by side
Row analysis.
As shown in Fig. 4-Fig. 8, experimental result includes the adaptive collection result and multidimensional of extracted one-dimensional principal component
Data include EMS memory occupation, and CPU is occupied, and four kinds of data such as amount of batteries and cpu temperature are adjusted adaptive based on the variation of principal component
Answer collection result.From experimental result as can be seen that the variation tendency of each data can be become with the variation of one-dimensional principal component
Gesture reflects.The individual collection result of four kinds of data is more satisfactory, and the collection process of these four data is based on master
The variation tendency adjustment of ingredient, it is feasible that the adaptive acquisition of multidimensional data/variable/attribute/feature is made of PCA.
The present invention is discussed in detail and introduces one-dimensional self-adapting data gathering algorithm ACFAS_PAR (Adaptive first before
CollectionFrequencyAdjustmentStrategyBasedon Predicted Accuracy Ratio).This is it
Preceding work.Basic principle is that the frequency of data acquisition is adjusted based on data variation, to realize adaptive sampling.Data variation
Size can be indicated by calculating the difference between the actual value of data and the predicted value of data.When the actual value of data and pre-
Between measured value very close to, illustrate that current data varies less, frequency acquisition needs to reduce, reduce data acquisition content；Work as number
According to actual value and predicted value between difference it is very big, illustrate current data variation very greatly, frequency acquisition needs to increase, to increase
The content of data acquisition, improves accuracy of data acquisition.
Multidimensional data adaptively acquires that detailed process is as follows:
(1) target of data acquisition is Yi=(y1, y2, y3, y4), it is 4 D data.It is assumed that y1It is EMS memory occupation, y2It is
CPU is occupied, y3It is amount of batteries, y4It is cpu temperature.Wherein i is t+1, t+2, t+3 ..., t+Nr；Wherein t is some acquisition
Time point.Defining data predicted value isYiIt is actual value.NrIt is actual acquisition number, NpIt is Prediction and Acquisition number, and Np
=Nr；
(2) by multidimensional data dimensionality reduction to one-dimensional principal component, as reference data, PCA algorithm is mainly utilized herein, it will be four-dimensional
Data Dimensionality Reduction is to one-dimensional principal component:
yi=PCA (Yi)
(3) mean value based on one-dimensional principal component calculating target data actual value and predicted value is as follows:
(4) the average ratio R of predicted value and actual valueMIt indicates: working as RMIt is 1, indicates that data do not change substantially；And work as RM
Significantly greater than 1 and the variation that represents target data values less than 1 it is big.
Theoretically, data variation ratio can also be calculated with variance, use RDIt indicates.Calculating process is as follows:
(5) the variation ratio based on data, the specific method of adjustment of data acquisition are as follows:
Wherein TiRepresent current sampling interval, Ti-1Represent the previous sampling interval.TincRepresent the value added in sampling interval；
TdecRepresent the reduced value in sampling interval.ThruAnd ThrlIt is the threshold value for judging data variation.Work as RMGreater than ThrlAnd it is less than
Thru, represent that data variation is small, and acquisition interval should increase Tinc；Work as RMGreater than ThruOr it is less than Thrl, it is big to represent data variation,
Acquisition interval should reduce Tdec。TmmaxAnd TminRepresent the maximum value and minimum value of data acquisition intervals, that is, data acquisition
The adjustment maximum at interval is no more than Tmax, minimum cannot be below Tmin, the constraint condition of adjustment is acquired as data.
The realization pseudocode of the adaptive acquisition method of the multidimensional data mentioned in the present invention is as follows:
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (7)
1. a kind of adaptive acquisition method of multidimensional data, which is characterized in that the adaptive gathering algorithm of one-dimensional data is utilized, in conjunction with drop
Dimension technology realizes the adaptive adjustment of multidimensional data acquisition, realizes the acquisition of multidimensional data；The multidimensional data adaptively acquires
Method main process includes:
Step 1 utilizes dimensionality reduction technology: dimensionality reduction is carried out to the multidimensional data of acquisition target, by multidimensional data dimensionality reduction to one-dimensional；
Step 2, the one-dimensional principal component obtained using original multi-dimensional Data Dimensionality Reduction are adjusted as the reference data for judging data variation
The frequency acquisition of whole multidimensional data realizes the adaptive acquisition of multidimensional data.
2. the adaptive acquisition method of multidimensional data as described in claim 1, which is characterized in that realize in current big data scene
Under multidimensional data adaptively acquire.
3. the adaptive acquisition method of multidimensional data as described in claim 1, which is characterized in that the multidimensional data is adaptively adopted
Set method specifically includes:
(1) target for assuming data acquisition is Yi=(y1, y2, y3..., yn), it is multidimensional data, yj(j=1,2,3 ...,
It n) is target data per one-dimensional.Wherein i is t+1, t+2, t+3 ..., t+Nr；Wherein t is some acquisition time.Definition
Data predicted value isYiIt is actual value.NrIt is actual acquisition number, NpIt is Prediction and Acquisition number, and Np=Nr；
(2) by multidimensional data dimensionality reduction to one-dimensional principal component, as reference data, PCA algorithm is mainly utilized herein:
yi=PCA (Yi)
(3) mean value based on one-dimensional principal component calculating target data actual value and predicted value is as follows:
(4) the average ratio R of predicted value and actual valueMIt indicates: working as RMIt is 1, indicates that data do not change substantially；And work as RMObviously
The variation for representing target data values greater than 1 and less than 1 is big.
Theoretically, data variation ratio can also be calculated with variance, use RDIt indicates.Calculating process is as follows:
(5) the variation ratio based on data, the specific method of adjustment of data acquisition are as follows:
Wherein TiRepresent current sampling interval, Ti-1Represent the previous sampling interval.TincRepresent the value added in sampling interval；TdecGeneration
The reduced value in table sampling interval.ThruAnd ThrlIt is the threshold value for judging data variation.Work as RMGreater than ThrlAnd it is less than Thru, represent
Data variation is small, and acquisition interval should increase Tinc；Work as RMGreater than ThruOr it is less than Thrl, it is big to represent data variation, acquisition interval
T should be reduceddec。
Each of the specific calculating process of above-mentioned formula parameter should be based on the statistical distribution of target data acquired in business
Characteristic determines.
4. a kind of social networks using the adaptive acquisition method of multidimensional data described in claims 1 to 3 any one recommends control
System processed.
5. a kind of intruding detection system using the adaptive acquisition method of multidimensional data described in claims 1 to 3 any one.
6. a kind of assets portrait acquisition system using the adaptive acquisition method of multidimensional data described in claims 1 to 3 any one
System.
7. a kind of any service application system using the adaptive acquisition method of multidimensional data described in claims 1 to 3 any one
System etc..
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811345413.8A CN109685101B (en) | 2018-11-13 | 2018-11-13 | Multi-dimensional data self-adaptive acquisition method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811345413.8A CN109685101B (en) | 2018-11-13 | 2018-11-13 | Multi-dimensional data self-adaptive acquisition method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109685101A true CN109685101A (en) | 2019-04-26 |
| CN109685101B CN109685101B (en) | 2021-09-28 |
Family
ID=66185393
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811345413.8A Active CN109685101B (en) | 2018-11-13 | 2018-11-13 | Multi-dimensional data self-adaptive acquisition method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109685101B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112507208A (en) * | 2020-11-02 | 2021-03-16 | 北京迅达云成科技有限公司 | Network data acquisition system based on big data |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010057119A2 (en) * | 2008-11-14 | 2010-05-20 | Neurovigil, Inc. | Methods of identifying sleep and waking patterns and uses |
| US8385662B1 (en) * | 2009-04-30 | 2013-02-26 | Google Inc. | Principal component analysis based seed generation for clustering analysis |
| CN103970092A (en) * | 2014-04-13 | 2014-08-06 | 北京工业大学 | Multi-stage fermentation process fault monitoring method based on self-adaption FCM algorithm |
| CN104809476A (en) * | 2015-05-12 | 2015-07-29 | 西安电子科技大学 | Multi-target evolutionary fuzzy rule classification method based on decomposition |
| CN105678270A (en) * | 2016-01-12 | 2016-06-15 | 哈尔滨工程大学 | Screening method for stress wave signal characteristics of one-dimensional member |
| CN106250819A (en) * | 2016-07-20 | 2016-12-21 | 上海交通大学 | Based on face's real-time monitor and detection facial symmetry and abnormal method |
-
2018
- 2018-11-13 CN CN201811345413.8A patent/CN109685101B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010057119A2 (en) * | 2008-11-14 | 2010-05-20 | Neurovigil, Inc. | Methods of identifying sleep and waking patterns and uses |
| US8385662B1 (en) * | 2009-04-30 | 2013-02-26 | Google Inc. | Principal component analysis based seed generation for clustering analysis |
| CN103970092A (en) * | 2014-04-13 | 2014-08-06 | 北京工业大学 | Multi-stage fermentation process fault monitoring method based on self-adaption FCM algorithm |
| CN104809476A (en) * | 2015-05-12 | 2015-07-29 | 西安电子科技大学 | Multi-target evolutionary fuzzy rule classification method based on decomposition |
| CN105678270A (en) * | 2016-01-12 | 2016-06-15 | 哈尔滨工程大学 | Screening method for stress wave signal characteristics of one-dimensional member |
| CN106250819A (en) * | 2016-07-20 | 2016-12-21 | 上海交通大学 | Based on face's real-time monitor and detection facial symmetry and abnormal method |
Non-Patent Citations (1)
| Title |
|---|
| ZHIHUI JI 等: "A Novel Two-Dimension Adaptive Data Collection Method for Network Management", 《2009 WRI INTERNATIONAL CONFERENCE ON COMMUNICATIONS AND MOBILE COMPUTING》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112507208A (en) * | 2020-11-02 | 2021-03-16 | 北京迅达云成科技有限公司 | Network data acquisition system based on big data |
| CN112507208B (en) * | 2020-11-02 | 2021-07-20 | 北京迅达云成科技有限公司 | Network data acquisition system based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109685101B (en) | 2021-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105204971B (en) | A kind of dynamic supervision interval method of adjustment based on Naive Bayes Classification technology | |
| CN110968426B (en) | Edge cloud collaborative k-means clustering model optimization method based on online learning | |
| CN112114986B (en) | Data anomaly identification method, device, server and storage medium | |
| CN105022761A (en) | Group search method and apparatus | |
| US20190132747A1 (en) | Generation of access point configuration change based on a generated coverage monitor | |
| EP4167149A1 (en) | Method and apparatus for building predictive model, computing device, and storage medium | |
| EP3771152B1 (en) | Network analysis program, network analysis device, and network analysis method | |
| CN108989880A (en) | A kind of code rate adaptive method for switching and system | |
| CN112398700B (en) | Service degradation method and device, storage medium and computer equipment | |
| CN109104731A (en) | Construction method, device and the computer equipment of cell scenario category classification model | |
| CN113379176A (en) | Telecommunication network abnormal data detection method, device, equipment and readable storage medium | |
| CN108961316A (en) | Image processing method, device and server | |
| CN108762936A (en) | Distributed computing system based on artificial intelligence image recognition and method | |
| KR20070008210A (en) | Method and apparatus for blocking the objectionable multimedia information | |
| CN108471357A (en) | A kind of terminal access scheduling method and device based on narrowband Internet of Things | |
| CN109685101A (en) | A kind of adaptive acquisition method of multidimensional data and system | |
| CN100493001C (en) | Automatic clustering method for multi-particle size network under G bit flow rate | |
| JP2023525112A (en) | Method and device for detecting user data of user equipment UE and storage medium | |
| CN113055333B (en) | Network flow clustering method and device capable of adaptively and dynamically adjusting density grid | |
| CN111278039B (en) | User perception suppression identification method, device, equipment and medium | |
| CN116385857A (en) | Calculation power distribution method based on AI intelligent scheduling | |
| CN103455525B (en) | The method and apparatus of popularization account number state is determined based on the search popularization behavior of user | |
| US11736562B1 (en) | Method and system for achieving high availability of service under high-load scene in distributed system | |
| CN110868304B (en) | PCC strategy issuing method and system | |
| CN112597323A (en) | Remote sensing image storage and migration method and 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 |