CN104125584A - Service index realization prediction method aiming at network service and apparatus thereof - Google Patents
Service index realization prediction method aiming at network service and apparatus thereof Download PDFInfo
- Publication number
- CN104125584A CN104125584A CN201310155387.3A CN201310155387A CN104125584A CN 104125584 A CN104125584 A CN 104125584A CN 201310155387 A CN201310155387 A CN 201310155387A CN 104125584 A CN104125584 A CN 104125584A
- Authority
- CN
- China
- Prior art keywords
- prediction
- operational indicator
- index
- data
- algorithm
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a service index realization prediction method aiming at a network service and an apparatus thereof and relates to the data service field. In the prior art, a prediction technology to a service index needs to be supported by a lot of data samples and a self-adaption performance according to a characteristic of the service index can not be achieved. By using the method and the apparatus of the invention, the above technical problems are solved. The method comprises the following steps: analyzing prediction factors and index dimensions which are related to the service index according to an index standard of a prediction service index; selecting a main association factor from the prediction factors and extracting a data sample from network management data according to the main association factor and the index dimensions; carrying out prediction on the data sample according to a prediction algorithm. The method and the apparatus are mainly used for carrying out prediction on the service index.
Description
Technical field
The present invention relates to data service technical field, particularly a kind of operational indicator for Network realizes method and the device of prediction.
Background technology
Along with communications industry scale is constantly expanded, diversified, the complicated and magnanimity of network management data trend, its management and Optimization Work are more difficult, and the requirement of the forecast analysis to operational indicator is also more and more higher.Forecast model based on mobile network manager at present, most of Predicting Technique adopting only limits to matching, the normalizing prediction of simple function, this is a kind of forecast model for data trend, can only open certain positive role, but the network having changed for order Structure and Scale be not suitable for.
The Predicting Technique of existing simple matching, normalizing prediction at least has following technical problem:
1, can only start at present the effect of trend analysis for the operational indicator prediction of a certain Network, need a large amount of data samples; And exist the excessive problem of workload of safeguarding mass data sample.
2, Forecasting Methodology is too simple, poor for some more complicated type of service prediction accuracy, bad adaptability, foresight a little less than; And for some specific operational indicators, as the operational indicator of festivals or holidays cannot, in conjunction with the particularity of operational indicator, cause predicting the outcome unstable.
Summary of the invention
The Predicting Technique of operational indicator is needed to a large amount of data sample support, Forecasting Methodology are simple, predictability is poor in order to solve in prior art, cannot be according to the adaptive technical problem of the characteristic of operational indicator, the present invention proposes a kind of operational indicator for Network and realizes method and the device predicted.
A method that realizes prediction for the operational indicator of Network, comprising:
Go out the predictive factor relevant to described operational indicator and index dimension according to the index standard analysis of prediction operational indicator;
From described predictive factor, select main association factor, and according to described main association factor and described index dimension extracted data sample from network management data;
According to prediction algorithm, described data sample is predicted.
A device of realizing prediction for the operational indicator of Network, comprising:
Analytic unit, for going out the predictive factor relevant to described operational indicator and index dimension according to the index standard analysis of prediction operational indicator;
Extracting unit, for selecting main association factor from described predictive factor, and according to described main association factor and described index dimension extracted data sample from network management data;
Predicting unit, for predicting described data sample according to prediction algorithm.
The scheme that the present embodiment provides adopts according to the index spectroscopic analysis of prediction operational indicator and goes out the predictive factor relevant to the characteristic of this operational indicator and index dimension, and the means of predicting again according to the corresponding data sample of this characteristic extracting, and then solve while prediction in prior art and to have needed a large amount of data samples, poor for some more complicated type of service prediction accuracy, the technical problem of bad adaptability, and then obtain and can predict for the characteristic of operational indicator, reduce data maintenance amount, prediction scheme strong adaptability, can ensure the technique effect of prediction accuracy simultaneously.
Brief description of the drawings
Accompanying drawing is used to provide a further understanding of the present invention, and forms a part for specification, for explaining the present invention, is not construed as limiting the invention together with embodiments of the present invention.In the accompanying drawings:
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.
A kind of operational indicator for Network that Fig. 1 provides for the embodiment of the present invention 1 realizes the method flow diagram of prediction;
Fig. 2 for the embodiment of the present invention 1 provide for determine the structure chart of the index dimension relevant to described operational indicator according to the type of service of operational indicator and estimation range;
The method flow diagram of described data sample being predicted according to prediction algorithm that Fig. 3 provides for the embodiment of the present invention 1;
A kind of operational indicator for Network that Fig. 4 provides for the embodiment of the present invention 3 realizes the structure drawing of device of prediction.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiment.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.And following embodiment is possibility of the present invention, embodiment puts in order and the numbering of embodiment and the sequence independence that it is preferably carried out.
Embodiment 1
The present embodiment provides a kind of operational indicator for Network to realize the method for prediction, and the method can be deployed on server, also can be deployed on a user terminal.As shown in Figure 1, the method comprises:
Step 101, analyzes the predictive factor relevant to operational indicator and index dimension according to the index standard (being the index algorithm of operational indicator) of prediction operational indicator;
Particularly, this step can realize in the following way: using the index parameter relating in the index standard of prediction operational indicator as the predictive factor relevant to operational indicator, and determine the index dimension relevant to operational indicator according to the type of service of operational indicator with estimation range, wherein, definite index dimension at least comprises time dimension.
Or, this step can also be by business personnel for possible multiple prediction operational indicator, respectively according to the index standard of prediction operational indicator, people analyzes the predictive factor relevant to operational indicator and index dimension, being organized into list presets, in the time need to knowing predictive factor and index dimension for a certain prediction operational indicator, table look-up; Or or can have business personnel directly to analyze, artificial for the analysis of a certain prediction operational indicator and input corresponding predictive factor and index dimension etc.
Data prediction belongs to the category of data mining.But with respect to other data mining applications, the network management data of mobile network, statistics derives from the bottom of equipment and measures, and has fixing index standard (being index algorithm), and human factor is less.Therefore there is more intense regularity, change comparatively steadily and upper regular strong feature of time cycle.Can (predict a certain operational indicator of a certain Network forecasting object in conjunction with these features, also be prediction operational indicator) analyze, combing goes out predictive factor and the index dimension of this forecasting object, for data preparation is done in main association factor and the data sample extraction of next step.Wherein, thus due to mobile network's operational indicator will taking the time as main dimension under normal circumstances, index dimension includes time dimension.
In the present embodiment, in the early stage of a certain operational indicator prediction to a certain business, need to analyze combing for index dimension and the predictive factor of forecasting object.(predictive factor refers to the object relevant with forecasting object, more specifically, is the index parameter relating in the index algorithm/formula of forecasting object, with reference to 1 in following embodiment 2).One of following table is according to the combing of analysis for many years of mobile network manager data, and the data characteristic of mobile network manager has following index dimension.
Table one
The index dimension of describing in conjunction with above-mentioned table one, in conjunction with every business datum of mobile network, can draw as shown in Figure 2 for determine the structure chart of the index dimension relevant to operational indicator according to the type of service of operational indicator and estimation range.In the structure of Fig. 1, time dimension is the main dimension of network management data (data that in the present embodiment, network management data refers in network management system to be taken care of), it is the principal element that decision predicts the outcome, with cycle time, the data correlation degree of dimension is the highest, so need the data sample in the cycle of introducing in the data sample of prediction use, therefore need time index dimension, for example predict non-festivals or holidays of index when GPRS index busy 18 every day, can utilize non-every day festivals or holidays historical data when busy 18 as data sample.Analyze data analysis combing by above index dimension, can effectively divide combing and go out rational data sample, so can effectively improve preparatory, the authenticity of prediction.
Step 102 is selected main association factor from predictive factor, and according to main association factor and index dimension extracted data sample from network management data;
Same forecasting object has the predictive factor of multiple associateds, if do not add distinguish these factors are all introduced in prediction algorithm, certainly will increase the complexity of algorithm and the difficulty of analytical calculation meaninglessly, therefore in this programme, with PCA, the predictive factor of forecasting object be carried out to Analysis and Screening.
PCA is a kind of statistical method of dimensionality reduction, it is by means of an orthogonal transform, the former random vector that its component is relevant changes into the incoherent new random vector of its component, this shows as the covariance matrix of former random vector is transformed into diagonal form battle array on algebraically, geometrically showing as the orthogonal coordinate system of former coordinate system transformation Cheng Xin, make it to point to sample point and scatter p the orthogonal direction of opening most, then multidimensional variable system is carried out to dimension-reduction treatment, make it to convert low-dimensional variable system to a higher precision, again by the suitable cost function of structure, further low-dimensional system is changed into unidimensional system.
By PCA, can from the multiple predictive factor associated with operational indicator, analyze and the maximally related main association factor of this operational indicator, so the predictive factor of eliminate redundancy effectively reduces the complexity of calculating.Finally extract corresponding sample data by corresponding index dimension and main association factor.Due to the principle ripe application at present of PCA, therefore those skilled in the art can run away with main association factor in conjunction with the principle of PCA above-mentioned predictive factor that this enforcement is provided after input variable, so detailed process is not repeated herein.
Step 103, predicts data sample according to prediction algorithm.
Utilize above-mentioned steps 101-102 to complete after the preparation of data sample, can predict for this operational indicator according to data sample and prediction algorithm.The present embodiment can at least be suitable for three kinds of algorithms and realize data prediction, and prediction algorithm comprises: SVM(SVMs) algorithm, Grey Prediction Algorithm or BP(backpropagation) neural network algorithm.
Particularly, as shown in Figure 3, according to prediction algorithm, data sample is predicted and can be realized in the following way:
301, data sample is normalized
Forecast sample data are normalized and are incorporated in the equation (i.e. prediction data model) of prediction algorithm lift scheme operation efficiency.
After preliminary treatment, the prediction of data sample is mainly divided into two stages: training stage and forecast period.This example will utilize above-mentioned three kinds of prediction algorithms, and the data after normalized are carried out to model training and prediction.
302 training stages: train according to the data category of operational indicator, for example, when the telephone traffic of Fuzhou area prediction, telephone traffic sample is divided into idle telephone traffic festivals or holidays, festivals or holidays telephone traffic when busy, non-festivals or holidays busy telephone traffic, non-festivals or holidays idle telephone traffic four class data samples, generally speaking the data sample of dividing is thinner, and the data model precision of training is higher.When the precision of model reaches default precision (be error coefficient lower than default error) between True Data and prediction output, can think model convergence or reach optimum, and record this training precision;
303, utilize the model after training to predict normalized data sample.
Forecast period: import the model parameter training in prediction algorithm, predict and predicted the outcome in conjunction with data sample.
In preferred version, the method also can comprise the steps 104:
Step 104, carries out precision checking, the validity of evaluation prediction result to the result of prediction.
After prediction algorithm has been trained, (can think that the prediction algorithm after training is forecast model here), employing accuracy test mode be carried out to model checking and predict the outcome the most accurately with this acquisition that circulates.The mode of precision checking has: residual error checking, degree of association checking, the poor checking of posteriority.
Wherein, residual error checking: the error between comparison prediction value and actual value is one describing mode intuitively.
Wherein, degree of association inspection: be mainly the fitting degree of response prediction data to historical data, by inquiry experience table, we can estimate whether forecast model foundation is reasonable.
Wherein, the poor verification of posteriority: the probability distribution situation of main manifestations error, further prove the reasonability of model.
More preferably in scheme, also comprise: after completing steps 104, can quote predicting the outcome of output checks with True Data verification, therefrom embody accuracy, the fluctuation etc. of data, thereby embody the effect of data mining, in existing network, also may occur that measurement data arrives the situation that postpones loss, can utilize predicts the outcome does the effect of data filling, embodies the flatness of data.
The method that the present embodiment provides adopts according to the index spectroscopic analysis of prediction operational indicator and goes out the predictive factor relevant to the characteristic of this operational indicator and index dimension, and the means of predicting again according to the corresponding data sample of this characteristic extracting, and then solve while prediction in prior art and to have needed a large amount of data samples, poor for some more complicated type of service prediction accuracy, the technical problem of bad adaptability, and then obtain and can predict for the characteristic of operational indicator, reduce data maintenance amount, prediction scheme strong adaptability, can ensure the technique effect of prediction accuracy simultaneously.
Embodiment 2
The example that provides a kind of operational indicator for Network to realize the method for prediction in conjunction with mobile service application scenarios is provided the present embodiment, to further the method for step 101-104 in embodiment 1 is explained.
Case using the IMS network completion rate (the whole province) in webmaster as prediction operational indicator, utilizes this operational indicator to predict practice to this forecast model, and index standard/index algorithm is as follows:
Number of times+called connection number of times/caller number of call attempts+called number of call attempts is connected in the IMS network completion rate=caller
1, for forecasting object: the These parameters standard analysis of the IMS of the whole province network completion rate index goes out relevant predictive factor and index dimension is as follows:
The index relating to, with reference to as predictive factor, comprising: number of times, called connection number of times, caller number of call attempts, called number of call attempts are connected in caller.
Be IMS network according to the type of operational indicator, estimation range is that the whole province and time dimension determine that the index dimension obtaining is following table two:
Table two
2, four predictive factor that go out for analysis combing are carried out a principal component analysis, and the program of developing by the main program analytical method of above-mentioned introduction is calculated, and result is as follows, number of times is connected in caller, called connection number of times, caller call attempt, the contribution rate of four predictive factor of called call attempt is respectively
| Index name | Number of times is connected in caller | Called connection number of times | Caller number of call attempts | Called number of call attempts |
| Contribution rate (%) | 95.9315 | 2.9513 | 0.6821 | 0.4351 |
Known by above data analysis, it is main association factor that number of times is connected in caller.
Connect number of times according to main association factor for bishop, index dimension is that time dimension, business dimension and region dimension extract corresponding data as data sample from network management data.After this step completes, predict that so the preparatory condition at initial stage just completes, following step only need to be carried out data modeling prediction according to prediction algorithm.
3-4, utilize three kinds of prediction algorithms to carry out data prediction, the predicting the outcome respectively as shown in following table three-table five of corresponding three kinds of algorithms:
Table three
Degree of association verification:
The degree of association of three period prediction data and True Data is for being respectively: 0.820,0.801,0,794, and be 50 because we get maximum poor percentage,
Be ρ=0.5, all effective according to above three degrees of association of professional's verification.
The poor inspection of posteriority:
Three periods are predicted the outcome, carry out the poor inspection of posteriority, the following professional of being verifies the table of comparisons of the precision of prediction grade drawing.
C is that the variance ratio P of prediction data is little probability of error ratio
The computing formula of C is:
The computing formula of P is:
Order
S
0＝0.6745S
1
:
The checking data of above three periods is:
Variance ratio is 0.431, and the little probability of error is 0.81,
Variance ratio is 0.423, and the little probability of error is 0.84,
Variance ratio is 0.476, and the little probability of error is 0.78,
All in 5%, all effective according to professional's checking data with upside deviation verification.
Table four
Degree of association verification:
The degree of association of three period prediction data and True Data is for being respectively: 0.816,0.7122,0,824, and be 50 because we get maximum poor percentage,
Be ρ=0.5, all effective according to above three degrees of association of professional's verification.
The poor inspection of posteriority:
Variance ratio is 0.492, and the probability of error is 0.77,
Variance ratio is 0.423, and the little probability of error is 0.74,
Variance ratio is 0.398, and the little probability of error is 0.81,
With upside deviation verification, all in 5%, checking data is all effective.
Table five
Degree of association verification:
The degree of association of three period prediction data and True Data is for being respectively: 0.784,0.813,0,845, and be 50 because we get maximum poor percentage,
Be ρ=0.5, all effective according to above three degrees of association of professional's verification.
The poor inspection of posteriority:
Variance ratio is 0.424, and the little probability of error is 0.84,
Variance ratio is 0.437, and the little probability of error is 0.77,
Variance ratio is 0.345, and the little probability of error is 0.83,
With upside deviation verification, all in 5%, checking data is all effective.
Obtain each prediction data by above prediction algorithm and compare, within prediction data and True Data error are all less than the scope that professional formulates error 5%, verification has confirmed the validity of data.
This invention of method that embodiment 1 and 2 provides is in conjunction with the index standard of actual mobile network manager business, operational indicator is predicted, and without whole network management datas, analyze according to the feature of forecasting object but adopt, the technological means of extracted data sample from network management data, so in the time reducing data reserve capacity, ensure to predict the outcome accuracy, also reduced the amount of calculation of budget cycle.
Embodiment 3
Realize for the ease of the method in embodiment 1 or 2, the present embodiment provides a kind of operational indicator for Network to realize the device of prediction, and this device can be arranged on a user terminal, also can be installed on one or more server.As shown in Figure 4, comprising: analytic unit 41, extracting unit 42, predicting unit 43.
Analytic unit 41, for going out the predictive factor relevant to operational indicator and index dimension according to the index standard analysis of prediction operational indicator; Extracting unit 42, selects main association factor for the predictive factor analyzing from analytic unit, and according to main association factor and index dimension extracted data sample from network management data; Predicting unit 43, predicts for data sample extracting unit 42 being extracted according to prediction algorithm.
In preferred version, this device also comprises:
Verification unit 44, for carrying out precision checking, the validity of evaluation prediction result to predicting the outcome of predicting unit 43.
Particularly, wherein, analytic unit 41, specifically for using the index parameter that relates in the index standard of prediction operational indicator as the predictive factor relevant to operational indicator, and determine the index dimension relevant to operational indicator according to the type of service of operational indicator with estimation range, wherein, definite index dimension at least comprises time dimension.
Extracting unit 42, specifically for obtaining main association factor by PCA to predictive factor analysis.
Predicting unit 43, specifically for being normalized data sample; Arrive optimum or convergence according to the data category training prediction algorithm of prediction operational indicator; Utilize the prediction algorithm after training to predict normalized data sample; Wherein, prediction algorithm comprises: support vector machines algorithm, Grey Prediction Algorithm or BP neural network algorithm.
The device that the present embodiment provides has according to the index spectroscopic analysis of prediction operational indicator and goes out relevant operational indicator characteristic the function of predicting again according to the corresponding data sample of this characteristic extracting predicted such as corresponding predictive factor and index dimension, and then solve while prediction in prior art and to have needed a large amount of data samples, poor for some more complicated type of service prediction accuracy, the technical problem of bad adaptability, and then obtain minimizing data maintenance amount, characteristic for operational indicator is predicted, prediction scheme strong adaptability, the technique effect that accuracy degree is high.
The product such as the said equipment or device that the embodiment of the present invention provides is to belong to flow and method taking computer program as foundation, and according to each step of method flow in embodiment of the method (at least one in embodiment 1-2) and/or accompanying drawing corresponding consistent mode completely, the functional module providing.And because this functional module is the software service of realizing by the mode of computer program, so functional module of specifically not mentioning for device embodiment 3, owing to considering that the content of recording according to said method embodiment has enough made those skilled in the art determine directly, expectedly and realize the functional module that described step institute must foundation from each process step of method record, so be not repeated herein.
The part that technical scheme of the present invention contributes to prior art in essence is in other words the function embodying with the form of software product, that is to say: even if the function body of its performed method of each equipment of device of the present invention, equipment or composition system or realization is hardware, the part that still in fact realizes above-mentioned functions of the present invention is but module or the unit of computer software product.And this computer software product can be stored in the storage medium can read, as the floppy disk of computer, hard disk or CD etc., comprise that some instructions are in order to make a method described in each embodiment of equipment execution the present invention.
The above, it is only the specific embodiment of the present invention, but the present invention can have multiple multi-form embodiment, above by reference to the accompanying drawings the present invention is illustrated, this does not also mean that the applied embodiment of the present invention can only be confined in these specific embodiments, those skilled in the art should understand, the embodiment that above provided is some examples in multiple preferred implementation, and the embodiment of any embodiment the claims in the present invention all should be within the claims in the present invention scope required for protection; Those skilled in the art can modify to the technical scheme of recording in each embodiment above, or part technical characterictic is wherein equal to replacement.Within the spirit and principles in the present invention all, any amendment of doing, be equal to and replace or improvement etc., within all should being included in the protection range of the claims in the present invention.
Claims (10)
1. a method that realizes prediction for the operational indicator of Network, is characterized in that, comprising:
Go out the predictive factor relevant to described operational indicator and index dimension according to the index standard analysis of prediction operational indicator;
From described predictive factor, select main association factor, and according to described main association factor and described index dimension extracted data sample from network management data;
According to prediction algorithm, described data sample is predicted.
2. method according to claim 1, is characterized in that, the method also comprises:
Result to prediction is carried out precision checking, the validity of evaluation prediction result.
3. method according to claim 1 and 2, is characterized in that, the described index standard analysis according to prediction operational indicator goes out the predictive factor relevant to described operational indicator and index dimension, specifically comprises:
Using the index parameter relating in the index standard of prediction operational indicator as the predictive factor relevant to described operational indicator, and determine the index dimension relevant to described operational indicator according to the type of service of described operational indicator with estimation range, wherein, described definite index dimension at least comprises time dimension.
4. method according to claim 1 and 2, is characterized in that, describedly from described predictive factor, selects main association factor, specifically comprises:
By PCA, described predictive factor analysis is obtained to main association factor.
5. method according to claim 1 and 2, is characterized in that, described prediction algorithm comprises: support vector machines algorithm, Grey Prediction Algorithm or BP neural network algorithm;
Describedly according to prediction algorithm, described data sample is predicted, is specifically comprised:
Described data sample is normalized;
Arrive optimum or convergence according to the data category training prediction algorithm of prediction operational indicator;
Utilize the prediction algorithm after training to predict normalized data sample.
6. a device of realizing prediction for the operational indicator of Network, is characterized in that, comprising:
Analytic unit, for going out the predictive factor relevant to described operational indicator and index dimension according to the index standard analysis of prediction operational indicator;
Extracting unit, for selecting main association factor from described predictive factor, and according to described main association factor and described index dimension extracted data sample from network management data;
Predicting unit, for predicting described data sample according to prediction algorithm.
7. device according to claim 6, is characterized in that, this device also comprises:
Verification unit, for carrying out precision checking, the validity of evaluation prediction result to the result of prediction.
8. according to the device described in claim 6 or 7, it is characterized in that,
Described analytic unit, specifically for using the index parameter that relates in the index standard of prediction operational indicator as the predictive factor relevant to described operational indicator, and determine the index dimension relevant to described operational indicator according to the type of service of described operational indicator with estimation range, wherein, described definite index dimension at least comprises time dimension.
9. according to the device described in claim 6 or 7, it is characterized in that,
Described extracting unit, specifically for obtaining main association factor by PCA to described predictive factor analysis.
10. according to the device described in claim 6 or 7, it is characterized in that,
Described predicting unit, specifically for being normalized described data sample; Arrive optimum or convergence according to the data category training prediction algorithm of prediction operational indicator;
Utilize the prediction algorithm after training to predict normalized data sample; Wherein, described prediction algorithm comprises: support vector machines algorithm, Grey Prediction Algorithm or BP neural network algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310155387.3A CN104125584A (en) | 2013-04-27 | 2013-04-27 | Service index realization prediction method aiming at network service and apparatus thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310155387.3A CN104125584A (en) | 2013-04-27 | 2013-04-27 | Service index realization prediction method aiming at network service and apparatus thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104125584A true CN104125584A (en) | 2014-10-29 |
Family
ID=51770817
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310155387.3A Pending CN104125584A (en) | 2013-04-27 | 2013-04-27 | Service index realization prediction method aiming at network service and apparatus thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104125584A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105554782A (en) * | 2015-12-09 | 2016-05-04 | 中国联合网络通信集团有限公司 | Prediction method and device for user perception index |
| CN106326461A (en) * | 2016-08-30 | 2017-01-11 | 杭州东方通信软件技术有限公司 | Real time processing guarantee method and system based on network signaling record |
| WO2017045472A1 (en) * | 2015-09-16 | 2017-03-23 | 华为技术有限公司 | Resource prediction method and system, and capacity management apparatus |
| CN107346464A (en) * | 2016-05-06 | 2017-11-14 | 腾讯科技（深圳）有限公司 | Operational indicator Forecasting Methodology and device |
| CN107527070A (en) * | 2017-08-25 | 2017-12-29 | 江苏赛睿信息科技股份有限公司 | Recognition methods, storage medium and the server of dimension data and achievement data |
| CN108829638A (en) * | 2018-06-01 | 2018-11-16 | 阿里巴巴集团控股有限公司 | A kind of business datum fluctuation processing method and processing device |
| CN109711606A (en) * | 2018-12-13 | 2019-05-03 | 平安医疗健康管理股份有限公司 | A kind of data predication method and device based on model |
| CN111985651A (en) * | 2019-05-22 | 2020-11-24 | 中国移动通信集团福建有限公司 | Operation and maintenance method and device for business system |
| CN112330003A (en) * | 2020-10-27 | 2021-02-05 | 电子科技大学 | Periodic capacity data prediction method, system and storage medium based on bidirectional cyclic neural network |
| CN112488871A (en) * | 2020-10-23 | 2021-03-12 | 广西电网有限责任公司电力科学研究院 | Method and system for eliminating redundant data of original input features of power grid |
| CN112734195A (en) * | 2020-12-31 | 2021-04-30 | 平安科技（深圳）有限公司 | Data processing method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080201281A1 (en) * | 2004-10-29 | 2008-08-21 | Nec Laboratories America, Inc. | Parallel support vector method and apparatus |
| CN101541030A (en) * | 2009-05-06 | 2009-09-23 | 华为技术有限公司 | Method for predicting data based on support vector machine and equipment thereof |
| CN102111284A (en) * | 2009-12-28 | 2011-06-29 | 北京亿阳信通软件研究院有限公司 | Method and device for predicting telecom traffic |
| CN102889598A (en) * | 2012-10-24 | 2013-01-23 | 华南理工大学 | Control method for assisting stable combustion of garbage by predicting garbage calorific value |
| CN103024762A (en) * | 2012-12-26 | 2013-04-03 | 北京邮电大学 | Service feature based communication service forecasting method |
-
2013
- 2013-04-27 CN CN201310155387.3A patent/CN104125584A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080201281A1 (en) * | 2004-10-29 | 2008-08-21 | Nec Laboratories America, Inc. | Parallel support vector method and apparatus |
| CN101541030A (en) * | 2009-05-06 | 2009-09-23 | 华为技术有限公司 | Method for predicting data based on support vector machine and equipment thereof |
| CN102111284A (en) * | 2009-12-28 | 2011-06-29 | 北京亿阳信通软件研究院有限公司 | Method and device for predicting telecom traffic |
| CN102889598A (en) * | 2012-10-24 | 2013-01-23 | 华南理工大学 | Control method for assisting stable combustion of garbage by predicting garbage calorific value |
| CN103024762A (en) * | 2012-12-26 | 2013-04-03 | 北京邮电大学 | Service feature based communication service forecasting method |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017045472A1 (en) * | 2015-09-16 | 2017-03-23 | 华为技术有限公司 | Resource prediction method and system, and capacity management apparatus |
| CN106549772A (en) * | 2015-09-16 | 2017-03-29 | 华为技术有限公司 | Resource prediction method, system and capacity management device |
| CN106549772B (en) * | 2015-09-16 | 2019-11-19 | 华为技术有限公司 | Resource prediction method, system and capacity management device |
| CN105554782B (en) * | 2015-12-09 | 2019-01-01 | 中国联合网络通信集团有限公司 | The prediction technique and device of user's perception index |
| CN105554782A (en) * | 2015-12-09 | 2016-05-04 | 中国联合网络通信集团有限公司 | Prediction method and device for user perception index |
| CN107346464B (en) * | 2016-05-06 | 2021-04-16 | 腾讯科技（深圳）有限公司 | Service index prediction method and device |
| CN107346464A (en) * | 2016-05-06 | 2017-11-14 | 腾讯科技（深圳）有限公司 | Operational indicator Forecasting Methodology and device |
| CN106326461A (en) * | 2016-08-30 | 2017-01-11 | 杭州东方通信软件技术有限公司 | Real time processing guarantee method and system based on network signaling record |
| CN106326461B (en) * | 2016-08-30 | 2019-07-30 | 杭州东方通信软件技术有限公司 | A kind of real-time processing support method and system based on network signal record |
| CN107527070A (en) * | 2017-08-25 | 2017-12-29 | 江苏赛睿信息科技股份有限公司 | Recognition methods, storage medium and the server of dimension data and achievement data |
| CN107527070B (en) * | 2017-08-25 | 2020-03-24 | 南京小睿软件有限公司 | Identification method of dimension data and index data, storage medium and server |
| CN108829638A (en) * | 2018-06-01 | 2018-11-16 | 阿里巴巴集团控股有限公司 | A kind of business datum fluctuation processing method and processing device |
| CN109711606A (en) * | 2018-12-13 | 2019-05-03 | 平安医疗健康管理股份有限公司 | A kind of data predication method and device based on model |
| CN111985651A (en) * | 2019-05-22 | 2020-11-24 | 中国移动通信集团福建有限公司 | Operation and maintenance method and device for business system |
| CN112488871A (en) * | 2020-10-23 | 2021-03-12 | 广西电网有限责任公司电力科学研究院 | Method and system for eliminating redundant data of original input features of power grid |
| CN112330003A (en) * | 2020-10-27 | 2021-02-05 | 电子科技大学 | Periodic capacity data prediction method, system and storage medium based on bidirectional cyclic neural network |
| CN112330003B (en) * | 2020-10-27 | 2022-11-08 | 电子科技大学 | Periodic capacity data prediction method, system and storage medium based on bidirectional cyclic neural network |
| CN112734195A (en) * | 2020-12-31 | 2021-04-30 | 平安科技（深圳）有限公司 | Data processing method and device, electronic equipment and storage medium |
| CN112734195B (en) * | 2020-12-31 | 2023-07-07 | 平安科技（深圳）有限公司 | Data processing method, device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104125584A (en) | Service index realization prediction method aiming at network service and apparatus thereof | |
| CN108564286B (en) | Artificial intelligent financial wind-control credit assessment method and system based on big data credit investigation | |
| US20180081629A1 (en) | Method and apparatus for providing ordered sets of arbitrary percentile estimates for varying timespans | |
| CN110363387A (en) | Portrait analysis method, device, computer equipment and storage medium based on big data | |
| WO2019165673A1 (en) | Reimbursement form risk prediction method, apparatus, terminal device, and storage medium | |
| CN103024762A (en) | Service feature based communication service forecasting method | |
| CN111695830A (en) | Power resource allocation method, system and equipment | |
| CN114048436A (en) | Construction method and construction device for forecasting enterprise financial data model | |
| CN113869801A (en) | Maturity state evaluation method and device for enterprise digital middleboxes | |
| US20140289007A1 (en) | Scenario based customer lifetime value determination | |
| CN116523262A (en) | Intelligent planning method, system and medium for production plan based on big data | |
| CN115422788A (en) | Power distribution network line loss analysis management method, device, storage medium and system | |
| CN112765463B (en) | Data management method for big data and user requirements and cloud computing server | |
| Deng et al. | A novel method for elimination of inconsistencies in ordinal classification with monotonicity constraints | |
| US20230376800A1 (en) | Predicting runtime variation in big data analytics | |
| CN115619430A (en) | User value evaluation method and device | |
| CN111209105A (en) | Capacity expansion processing method, capacity expansion processing device, capacity expansion processing equipment and readable storage medium | |
| CN113743994A (en) | Provider's season-busy prediction method, system, equipment and storage medium | |
| CN114372835A (en) | Comprehensive energy service potential customer identification method, system and computer equipment | |
| CN113935407A (en) | Abnormal behavior recognition model determining method and device | |
| CN112308294A (en) | Default probability prediction method and device | |
| US20230394069A1 (en) | Method and apparatus for measuring material risk in a data set | |
| Addabbo et al. | Children capabilities and family characteristics in Italy | |
| CN116484230B (en) | Method for identifying abnormal business data and training method of AI digital person | |
| CN117807411B (en) | Server performance index prediction method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20141029 |