CN107623924A - It is a kind of to verify the method and apparatus for influenceing the related Key Performance Indicator KPI of Key Quality Indicator KQI - Google Patents
It is a kind of to verify the method and apparatus for influenceing the related Key Performance Indicator KPI of Key Quality Indicator KQI Download PDFInfo
- Publication number
- CN107623924A CN107623924A CN201610565431.1A CN201610565431A CN107623924A CN 107623924 A CN107623924 A CN 107623924A CN 201610565431 A CN201610565431 A CN 201610565431A CN 107623924 A CN107623924 A CN 107623924A
- Authority
- CN
- China
- Prior art keywords
- kpi
- key
- key performance
- kqi
- strong correlation
- 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.)
- Withdrawn
Links
Landscapes
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a kind of method and apparatus for the Key Performance Indicator KPI for verifying and influenceing Key Quality Indicator KQI correlations, by determining the KQI of business and the KPI of corresponding at least one strong correlation, the KPI of all strong correlations one-dimensional threshold value and multidimensional threshold value can be got according to the KPI of each strong correlation and KQI degree of association and the KPI of each strong correlation weighted value, so as to obtain the complete degree of the KPI of at least one strong correlation corresponding to the KQI of the business, make it possible to all-sidedly and accurately verify the Key Performance Indicator of Key Quality Indicator, improve the efficiency of the network optimization of a certain business Key Quality Indicator of service network in communication system.
Description
Technical field
The present invention relates to the network communications technology, espespecially a kind of verify influences the related key performances of Key Quality Indicator KQI
Index KPI method and apparatus.
Background technology
At present, in a wireless communication system, Key Quality Indicator (Key Quality Indicators, abbreviation:KQI) it is
The QoS parameter of the impression of being close to the users proposed for different business, is customer-centric, embodies the pass of service layer
Key index.Wherein, when a certain quality of service index KQI of the service network of communication system deteriorates, it would be desirable to which finding out causes the KQI
Chief performance index (Key Performance Indication, the abbreviation of deterioration:KPI), with carry out parameter adjustment or
The network optimization.
In the prior art, generally use is curve-fitting method and correlation coefficient process, it is assumed that needs the KQI that is calculated to be
Http response time delay, the KPI for influenceing the KQI is negative 110dBm coverage rates, and Fig. 1 is influences of the KPI to KQI in the prior art
Graph of a relation, as shown in figure 1, being drawn a conclusion using curve-fitting method:Coverage rate is poorer, and http response time delay is lower, and uses phase
Y-factor method Y is closed to draw a conclusion:Coverage rate and http response time delay degree of correlation very little.
But be not consistent with engineering experience and the field network optimization fact using prior art, above-mentioned two conclusion, this
Sample judges whether KQI transfinites merely by single KPI quality, can not solve field network optimization it is true in " KPI is
Good KQI is poor, or the good KPI of KQI poor " situation so as to a certain quality of service index of the service network of communication system
Network optimization efficiency is low.
The content of the invention
In order to solve the above-mentioned technical problem, the invention provides a kind of verify to influence the related passes of Key Quality Indicator KQI
Key performance indications KPI method and apparatus, the Key Performance Indicator of Key Quality Indicator can be all-sidedly and accurately calculated, so as to carry
The efficiency of the network optimization of a certain business Key Quality Indicator of service network in high communication system.
In order to reach the object of the invention, in a first aspect, being verified the invention provides one kind influences Key Quality Indicator KQI phases
The Key Performance Indicator KPI of pass method, this method include:
Determine the Key Quality Indicator KQI of business and at least one and business Key Quality Indicator KQI strong correlations
Key Performance Indicator KPI;
It is according to the Key Performance Indicator KPI of the strong correlation and Key Quality Indicator KQI degree of association and each strong
Related Key Performance Indicator KPI weighted value obtains the Key Performance Indicator KPI of each strong correlation one-dimensional thresholding and multiple
The multidimensional thresholding of the Key Performance Indicator KPI combinations of strong correlation;
According to the key of the Key Performance Indicator KPI of acquired each strong correlation one-dimensional thresholding and multiple strong correlations
Can the multidimensional thresholding of index KPI combination obtain the Key Performance Indicator KPI of the strong correlation and sentence to rate;
If it is determined that described sentence when meeting rate predetermined threshold value, the Key Performance Indicator KPI of the strong correlation is obtained in institute
State misdetection rate when Key Quality Indicator KQI exceedes predetermined deterioration thresholding.
The embodiments of the invention provide a kind of verify to influence the related Key Performance Indicator KPI's of Key Quality Indicator KQI
Method, by determining the KQI of business and the KPI of corresponding at least one strong correlation, according to the KPI's of each strong correlation and KQI
The KPI of the degree of association and each strong correlation weighted value can get the KPI of all strong correlations one-dimensional threshold value and multidimensional door
Limit value, so as to obtain the complete degree of the KPI of at least one strong correlation corresponding to the KQI of the business, enabling complete
The Key Performance Indicator of Key Quality Indicator is verified in face exactly, and a certain business for improving service network in communication system is crucial
The efficiency of the network optimization of quality index.
Further, in one embodiment, one of the Key Performance Indicator KPI according to acquired each strong correlation
The multidimensional thresholding of the Key Performance Indicator KPI of dimension thresholding and multiple strong correlations combinations obtains the Key Performance Indicator of the strong correlation
KPI's sentences to after rate, in addition to:
When the rising scale of False Rate is prescribed a time limit more than offset gate, redefine corresponding to the Key Quality Indicator KQI of business
The Key Performance Indicator KPI of strong correlation, wherein, the False Rate is that 1- is sentenced to rate.
By carrying out bias correction to acquired False Rate, so as to can further more accurately to find out influence
The KQI of business KPI
Further, in one embodiment, one of the Key Performance Indicator KPI according to acquired each strong correlation
The multidimensional thresholding of the Key Performance Indicator KPI of dimension thresholding and multiple strong correlations combinations obtains the Key Performance Indicator of the strong correlation
KPI's sentences to after rate, in addition to:
The predetermined threshold to rate is sentenced described in adjustment, however, it is determined that sentence described in adjustment and 0 is approached to the misdetection rate obtained after rate, then really
The Key Performance Indicator KPI of the fixing strong correlation for ringing the Key Quality Indicator KQI is not omitted.
Predetermined threshold to rate is sentenced by adjustment, can effectively verify the strong correlation that the identified KQI on business influences
KPI be comprehensive and accurate.
Further, in one embodiment, after sentencing described in adjustment to the predetermined threshold of rate, in addition to:
If it is determined that 0 can not be approached to the misdetection rate obtained after rate by sentencing described in adjustment, it is determined that influenceed the Key Quality and referred to
The Key Performance Indicator KPI for marking KQI strong correlation has omission.
Predetermined threshold to rate is sentenced by adjustment, can effectively verify the strong correlation that the identified KQI on business influences
KPI be comprehensive and accurate.
Further, in one embodiment, it is determined that influenceing the key of the strong correlation of the Key Quality Indicator KQI
After energy index KPI is not omitted, in addition to:
According to the key of the Key Performance Indicator KPI of acquired each strong correlation one-dimensional thresholding and multiple strong correlations
The multidimensional thresholding of energy index KPI combinations carries out data training to the Key Performance Indicator KPI of the strong correlation, estimates without described
Misdetection rate during Key Quality Indicator KQI data.
, can be right in the case where not knowing KQI data by carrying out data training to the KQI-KPI models of foundation
Network parameter carries out pre-optimized, so as to prevent failure, improves efficiency.
Second aspect, the invention provides a kind of verify to influence the related Key Performance Indicator KPI of Key Quality Indicator KQI
Device, the device includes:Determining unit, threshold cell, judging unit and authentication unit;
The determining unit is arranged to the Key Quality Indicator KQI of determination business and at least one and business key
The Key Performance Indicator KPI of quality index KQI strong correlations;
The threshold cell is arranged to Key Performance Indicator KPI and the Key Quality Indicator according to the strong correlation
The Key Performance Indicator KPI of the KQI degree of association and each strong correlation weighted value obtains the Key Performance Indicator of each strong correlation
The multidimensional thresholding of the Key Performance Indicator KPI combinations of KPI one-dimensional thresholding and multiple strong correlations;
The judging unit be arranged to according to the Key Performance Indicator KPI of acquired each strong correlation one-dimensional thresholding and
The multidimensional thresholding of the Key Performance Indicator KPI combinations of multiple strong correlations obtains sentencing for the Key Performance Indicator KPI of the strong correlation
To rate;
The authentication unit is arranged to if it is determined that described sentence when meeting rate predetermined threshold value, obtains the pass of the strong correlation
Key performance indications KPI exceedes predetermined misdetection rate when deteriorating thresholding in the Key Quality Indicator KQI.
The embodiments of the invention provide a kind of verify to influence the related Key Performance Indicator KPI's of Key Quality Indicator KQI
Device, by determining the KQI of business and the KPI of corresponding at least one strong correlation, according to the KPI's of each strong correlation and KQI
The KPI of the degree of association and each strong correlation weighted value can get the KPI of all strong correlations one-dimensional threshold value and multidimensional door
Limit value, so as to obtain the complete degree of the KPI of at least one strong correlation corresponding to the KQI of the business, enabling complete
The Key Performance Indicator of Key Quality Indicator is verified in face exactly, and a certain business for improving service network in communication system is crucial
The efficiency of the network optimization of quality index.
Further, in one embodiment, the determining unit is also configured to according to acquired each strong correlation
Key Performance Indicator KPI one-dimensional thresholding and multiple strong correlations Key Performance Indicator KPI combination multidimensional thresholding obtain institute
Sentencing to after rate for the Key Performance Indicator KPI of strong correlation is stated, when the rising scale of False Rate is prescribed a time limit more than offset gate, again really
Determine the Key Performance Indicator KPI of strong correlation corresponding to the Key Quality Indicator KQI of business, wherein, the False Rate is that 1- sentences pair
Rate.
By carrying out bias correction to acquired False Rate, so as to can further more accurately to find out influence
The KQI of business KPI.
Further, in one embodiment, described device also includes adjustment unit, and the adjustment unit is arranged in root
According to the Key Performance Indicator KPI of acquired each strong correlation one-dimensional thresholding and the Key Performance Indicator KPI of multiple strong correlations
The multidimensional thresholding of combination obtains the sentencing to after rate, sentencing described in adjustment to the pre- of rate of the Key Performance Indicator KPI of the strong correlation
Determine threshold value, however, it is determined that sentence described in adjustment and 0 is approached to the misdetection rate obtained after rate, it is determined that influence the Key Quality Indicator KQI
The Key Performance Indicator KPI of strong correlation do not omit.
Predetermined threshold to rate is sentenced by adjustment, can effectively verify the strong correlation that the identified KQI on business influences
KPI be comprehensive and accurate.
Further, in one embodiment, the adjustment unit is also configured to sentencing the predetermined threshold to rate described in adjustment
After value, however, it is determined that 0 can not be approached to the misdetection rate obtained after rate by sentencing described in adjustment, it is determined that influence the Key Quality Indicator
The Key Performance Indicator KPI of KQI strong correlation has omission.
Predetermined threshold to rate is sentenced by adjustment, can effectively verify the strong correlation that the identified KQI on business influences
KPI be comprehensive and accurate.
Further, in one embodiment, the device also includes:Estimate unit;
The unit of estimating is arranged to it is determined that influenceing the Key Performance Indicator of the strong correlation of the Key Quality Indicator KQI
After KPI is not omitted, according to the Key Performance Indicator KPI of acquired each strong correlation one-dimensional thresholding and multiple strong correlations
The multidimensional thresholdings of Key Performance Indicator KPI combinations data training is carried out to the Key Performance Indicator KPI of the strong correlation, estimate
There is no the misdetection rate during Key Quality Indicator KQI data.
, can be right in the case where not knowing KQI data by carrying out data training to the KQI-KPI models of foundation
Network parameter carries out pre-optimized, so as to prevent failure, improves efficiency.
Other features and advantages of the present invention will be illustrated in the following description, also, partly becomes from specification
Obtain it is clear that or being understood by implementing the present invention.The purpose of the present invention and other advantages can be by specification, rights
Specifically noted structure is realized and obtained in claim and accompanying drawing.
Brief description of the drawings
Accompanying drawing is used for providing further understanding technical solution of the present invention, and a part for constitution instruction, with this
The embodiment of application is used to explain technical scheme together, does not form the limitation to technical solution of the present invention.
Fig. 1 is influence graphs of a relation of the KPI to KQI in the prior art;
Fig. 2 is a kind of side for verifying the Key Performance Indicator KPI for influenceing Key Quality Indicator KQI correlations provided by the invention
The schematic flow sheet of method embodiment one;
Fig. 3 is a kind of side for verifying the Key Performance Indicator KPI for influenceing Key Quality Indicator KQI correlations provided by the invention
Method parameter input and output schematic diagram;
Fig. 4 is a kind of side for verifying the Key Performance Indicator KPI for influenceing Key Quality Indicator KQI correlations provided by the invention
The KPI model schematics for the KQI- strong correlations that method Application Example one is built;
Fig. 5 is that a kind of verify provided in an embodiment of the present invention influences the related Key Performance Indicators of Key Quality Indicator KQI
One every data output schematic diagram of KPI methods Application Example.
Fig. 6 is that a kind of verify provided in an embodiment of the present invention influences the related Key Performance Indicators of Key Quality Indicator KQI
The result of implementation schematic diagram of KPI methods Application Example one;
Fig. 7 is that a kind of verify provided in an embodiment of the present invention influences the related Key Performance Indicators of Key Quality Indicator KQI
The structural representation of KPI device embodiments one;
Fig. 8 is a kind of Key Quality Indicator KQI provided in an embodiment of the present invention and related Key Performance Indicator KPI analyses
System schematic.
Embodiment
For the object, technical solutions and advantages of the present invention are more clearly understood, below in conjunction with accompanying drawing to the present invention
Embodiment be described in detail.It should be noted that in the case where not conflicting, in the embodiment and embodiment in the application
Feature can mutually be combined.
Can be in the computer system of such as one group computer executable instructions the flow of accompanying drawing illustrates the step of
Perform.Also, although logical order is shown in flow charts, in some cases, can be with suitable different from herein
Sequence performs shown or described step.
The present embodiments relate to method can apply to communication system, internet, ecommerce, E-Government, public affairs
Department's operation flow etc. needs to carry out the scene that data analysis and system value-added service consider, and be not limited to computer industry or
The communications industry, it is embodied and not particularly restricted.
The present embodiments relate to method, it is intended to solution sentence in the prior art merely by single KPI quality
Whether disconnected KQI transfinites, and can not solve the feelings of " the good KQI of KPI are poor, or the good KPI of KQI poor " in the field network optimization fact
Condition so that the technical problem low to the network optimization efficiency of a certain quality of service index of the service network of communication system.
Technical scheme is described in detail with specifically embodiment below.These specific implementations below
Example can be combined with each other, and may be repeated no more for same or analogous concept or process in some embodiments.
Fig. 2 is a kind of side for verifying the Key Performance Indicator KPI for influenceing Key Quality Indicator KQI correlations provided by the invention
The schematic flow sheet of method embodiment one.The present embodiment, which refers to checking, influences the related key performances of Key Quality Indicator KQI
Index KPI detailed process.As shown in figure 1, this method includes:
S101, the Key Quality Indicator KQI for determining business and at least one strong with the Key Quality Indicator KQI of the business
Related Key Performance Indicator KPI.
Specifically, according to the service conditions of specific service network, the Key Quality Indicator KQI and extremely of the business is determined
The Key Performance Indicator KPI of the Key Quality Indicator KQI strong correlations of a few business, for example, in a communications system, it is mobile
Common carrier finds the strong correlation KPI of certain provincial capital's main city zone online low rate, wherein it is possible to select " during http response
Prolong " KQI is used as, wherein the KPI related to KQI can have one or more, can according to each related KPI weighted value
With determine with the KPI of " http response time delay " strong correlation, can be " up initial HARQ retransmit ratio ", " up to be put down per PRB
Equal handling capacity ", " negative 110dBm coverage rates ", " user plane averagely activates UE numbers " etc. are as the strong correlation of " http response time delay "
KPI, but do not limited with this.
S102, according to the Key Performance Indicator KPI and the Key Quality Indicator KQI of the strong correlation degree of association and every
The Key Performance Indicator KPI of individual strong correlation weighted value obtain the Key Performance Indicator KPI of each strong correlation one-dimensional thresholding and
The multidimensional thresholding of the Key Performance Indicator KPI combinations of multiple strong correlations.
Specifically, all KPI corresponding to KQI and the KQI that business needs can be obtained each by normalizing quantization operations
The KPI and KQI degree of association, general KQI-KPI degrees of association scope are [0,1], set unrelated thresholding and dependent threshold, you can to sentence
Continuous item KPI, reminder item KPI nothing to do with items KPI are decided., can basis to reduce the overlapping degree between continuous item KPI as far as possible
Cluster table, rules out the KPI with KQI strong correlations from same cluster correlation item KPI, and is obtained by normalizing quantization operations each strong
Related weighted value, then calculate the KPI's of strong correlation successively according to the KPI of the obtained strong correlation degree of association and weighted value
One-dimensional thresholding and multidimensional thresholding, wherein, when the KPI of strong correlation only has one, then it need to only calculate the one of the KPI of the strong correlation
Thresholding is tieed up, when the KPI of strong correlation is more than or equal to two, then the KPI of the strong correlation one-dimensional thresholding is first calculated, then,
It is combined to calculate the KPI of these strong correlations multidimensional door to the KPI of multiple strong correlations according to the KPI of strong correlation weighted value
Limit, continue the example above, the KPI of above-mentioned 4 strong correlations can carry out two-dimensional combination, be arranged according to the KPI of 4 strong correlations weight
Sequence, 4 strong correlation item KPI are divided to for two groups, every group of 2 KPI, wherein can will " up initial HARQ retransmits ratio " with " on
Row is per PRB average throughputs " it is used as the 1st group;" negative 110dBm coverage rates " is used as the 2nd group with " user plane averagely activates UE numbers "
Etc., specific calculate can select according to specific actual conditions, but be not limited to this.
S103, according to the Key Performance Indicator KPI of acquired each strong correlation one-dimensional thresholding and multiple strong correlations
The Key Performance Indicator KPI that the multidimensional thresholding of Key Performance Indicator KPI combinations obtains the strong correlation is sentenced to rate.
Rate is referred to according to the KPI of each strong correlation default to be sentenced to accurate judgement KQI under rate meeting specifically, this is sentenced
The probability for deteriorating thresholding is crossed, is combined according to the KPI of the KPI of all strong correlations one-dimensional thresholding and plurality of strong correlation more
Dimension thresholding come be calculated all strong correlation KPI it is default sentence to accurate judgement KQI under rate cross deteriorate thresholding probability, together
When corresponding False Rate be that 1- is sentenced to rate.Continue the example above, ratio can be retransmitted according to up initial HARQ ", it is " up every
PRB average throughputs ", " negative 110dBm coverage rates ", " user plane averagely activates UE numbers " this 4 strong correlations KPI one-dimensional door
Limit, and the two-dimentional thresholding of combination of two, sentence to accurate judgement KQI under rate to calculate the KPI of this 4 strong correlations default
" http response time delay " crosses the probability for deteriorating thresholding, and specific calculate can select according to actual conditions, but not with this
It is limited.
S104, if it is determined that described sentence when meeting rate predetermined threshold value, obtain the Key Performance Indicator KPI of the strong correlation
Exceed predetermined misdetection rate when deteriorating thresholding in the Key Quality Indicator KQI.
As long as specifically, according to it is obtained above sentence to be more than or equal to rate default sentence to rate, it is possible to be further continued for examining
KQI is crossed in the KPI data sample (be designated as data for K) for the strong correlation for deteriorating thresholding, how many data be crossed it is above-mentioned
The KPI of all strong correlations one-dimensional thresholding or multidimensional thresholding group (be designated as data for J), then can obtain misdetection rate=
1-J/K, according to misdetection rate, it is possible to which checking determines the KPI of strong correlation completeness according to KQI, and misdetection rate meets default
Threshold value, it may be determined that the related Key Performance Indicator KPI of Key Quality Indicator KQI completeness height is influenceed, so as to comprehensive
The Key Performance Indicator KPI for the strong correlation for influenceing Key Quality Indicator KQI is obtained exactly, is taken so as to improve in communication system
The efficiency of the network optimization of a certain business Key Quality Indicator of business network.
The embodiment of the invention discloses a kind of verify to influence the related Key Performance Indicator KPI's of Key Quality Indicator KQI
Method, by determining the KQI of business and the KPI of corresponding at least one strong correlation, according to the KPI's of each strong correlation and KQI
The KPI of the degree of association and each strong correlation weighted value can get the KPI of all strong correlations one-dimensional threshold value and multidimensional door
Limit value, so as to obtain the complete degree of the KPI of at least one strong correlation corresponding to the KQI of the business, enabling complete
The Key Performance Indicator of Key Quality Indicator is verified in face exactly, and a certain business for improving service network in communication system is crucial
The efficiency of the network optimization of quality index.
Further, in one embodiment, one of the Key Performance Indicator KPI according to acquired each strong correlation
The multidimensional thresholding of the Key Performance Indicator KPI of dimension thresholding and multiple strong correlations combinations obtains the Key Performance Indicator of the strong correlation
KPI's sentences to after rate, in addition to:
According to it is described sentence rate is obtained corresponding to False Rate, when the rising scale of the False Rate is prescribed a time limit more than offset gate,
Redefine the Key Performance Indicator KPI of strong correlation corresponding to the Key Quality Indicator KQI of business.
Specifically, False Rate with sentence to rate be it is corresponding existing for, refer to according to the KPI of each strong correlation meet it is default
Sentence the probability crossed to false judgment KQI under rate and deteriorate thresholding, i.e.,:False Rate=1- is sentenced to rate, according to acquired each
The multidimensional thresholding of the Key Performance Indicator KPI combinations of the Key Performance Indicator KPI of strong correlation one-dimensional thresholding and multiple strong correlations
Obtain when sentencing to rate of the Key Performance Indicator KPI of the strong correlation, False Rate can be got, if by data sample come
Calculate, the False Rate is constantly in propradation, it is necessary to look at whether the ratio of the rising exceedes deviation threshold, if
The rising scale deviation of the False Rate is too big, then the KPI of explanation influence KQI strong correlation is looked for problematic, it is necessary to redefine
The KPI of strong correlation corresponding to the KQI of the business, that is, rebuild new model and sample data, to find out influence exactly
The KPI of the KQI strong correlations of business.
By carrying out bias correction to acquired False Rate, so as to can further more accurately to find out influence
The KQI of business KPI.
Further, in one embodiment, according to the one-dimensional of the Key Performance Indicator KPI of acquired each strong correlation
The multidimensional thresholding of the Key Performance Indicator KPI of thresholding and multiple strong correlations combinations obtains the Key Performance Indicator of the strong correlation
KPI's sentences to after rate, in addition to:
The predetermined threshold to rate is sentenced described in adjustment, however, it is determined that sentence described in adjustment and 0 is approached to the misdetection rate obtained after rate, then really
The Key Performance Indicator KPI of the fixing strong correlation for ringing the Key Quality Indicator KQI is not omitted.
If, can be by sentencing the predetermined threshold to rate specifically, when can not accurately being provided to predetermined threshold value
Be adjusted, usually from 0 to 100% between be adjusted, come obtain after the adjustment sentence under the predetermined threshold to rate gained
To misdetection rate, if the KPI that the misdetection rate can approach these strong correlations for for 0, proving to determine before is not omitted.
Predetermined threshold to rate is sentenced by adjustment, can effectively verify the strong correlation that the identified KQI on business influences
KPI be comprehensive and accurate.
Further, in one embodiment, after sentencing described in adjustment to the predetermined threshold of rate, in addition to:
If it is determined that 0 can not be approached to the misdetection rate obtained after rate by sentencing described in adjustment, it is determined that influenceed the Key Quality and referred to
The Key Performance Indicator KPI for marking KQI strong correlation has omission.
Specifically, if 0, before proving can not be approached by sentencing the misdetection rate resulting afterwards of the threshold value to rate by adjustment
The KPI of these strong correlations determined has omission, cannot characterize the reason for KQI deteriorates comprehensively, i.e. the KQI determined before
Do not have in the KPI set of corresponding all strong correlations comprising the KQI whole factors deteriorated, it is necessary to go to determine that the KQI is corresponding again
KPI in look for other and have the KPI of strong correlation, it is necessary to verified again.
Predetermined threshold to rate is sentenced by adjustment, can effectively verify the strong correlation that the identified KQI on business influences
KPI be comprehensive and accurate.
Further, in one embodiment, it is determined that influenceing the key of the strong correlation of the Key Quality Indicator KQI
After energy index KPI is not omitted, in addition to:
According to the key of the Key Performance Indicator KPI of acquired each strong correlation one-dimensional thresholding and multiple strong correlations
The multidimensional thresholding of energy index KPI combinations carries out data training to the Key Performance Indicator KPI of the strong correlation, estimates without described
Misdetection rate during Key Quality Indicator KQI data.
Specifically, it is determined that the Key Performance Indicator KPI for influenceing the strong correlation of the Key Quality Indicator KQI is not omitted
Afterwards, can be in the case of no acquisition KQI data, only after the especially KQI-KPI undergoes enough practical data training
By the data for collecting these KPI, it is possible to estimate whether KQI deteriorates, so as to carry out pre-optimized to network parameter.
, can be right in the case where not knowing KQI data by carrying out data training to the KQI-KPI models of foundation
Network parameter carries out pre-optimized, so as to prevent failure, improves efficiency.
It should be noted that Fig. 3, which is a kind of verifies provided by the invention, influences the key of Key Quality Indicator KQI correlations
Energy index KPI method parameter input and output schematic diagram, as shown in figure 3, a kind of checking provided in an embodiment of the present invention influences to close
Key Performance Indicator KPI related key quality index KQI method, some belonged in whole KQI-KPI association analysis, its
In, it is the incidence relation for first finding out KQI-KPI first, specific implementation steps are such as when verifying KQI-KPI incidence relation
Under:
1) some KQI data and corresponding N number of KPI data (such as N=30) are inputted;
2) data cleansing, pretreatment, cluster;
3) the quantum chemical method each KPI and KQI degree of association;
4) according to M " the strong correlation item KPI " (such as M=4), and calculate M for quantifying the degree of association, cluster table and ruling out KQI
Individual " strong correlation item KPI " weight;
5) according to above-mentioned each KPI and the KQI degree of association, M " strong correlation item KPI " weight, KQI and { strong correlations
KPI } data record etc. calculated quantify association, so as to obtain following input results:
A) one-dimensional thresholding;
B) multidimensional thresholding;
C) False Rate whether thresholding group judgement KQI transfinites;
D) misdetection rate whether thresholding group judgement KQI transfinites;
E) the complete degree estimation and confidence level that KPI set descriptions KQI transfinites.
Illustrate below and the detailed process of the present invention in actual applications is described in detail:
The practical application is to find certain provincial capital main city zone online low rate this business in mobile communication carrier
Main related KPI, multidimensional threshold matrix corresponding to output simultaneously assess continuous item KPI complete degree.Primary Construction model, use
Selected date, 24 hours granularity cell DBMSs of 4000 cells, investigates KQI:Http response time delay is wireless with 30
Side KPI correlation degree.
Step 1, table 1 influence the key of Key Quality Indicator KQI correlations for a kind of verify provided in an embodiment of the present invention
KQI-KPI mapping tables in energy index KPI method Application Example one;According to abnormal traffic distribute leaflets, KQI to be assessed is read
With corresponding 30 wireless sides KPI item lists, if it is " http response time delay " that business KQI titles, which are assumed, its corresponding 30 nothing
Line side KPI titles, it is as shown in table 1 below:
Table 1 is KQI-KPI mapping tables
Step 2, Fig. 4 are that a kind of verify provided by the invention influences the related Key Performance Indicators of Key Quality Indicator KQI
The KPI model schematics for the KQI- strong correlations that KPI method Application Example one is built, as shown in figure 4, according to above-mentioned KQI-
KPI association components carry out normalizing quantization operations, obtain the degrees of association of each KPI with corresponding KQI, general KQI-KPI degree of association models
Enclose for [0,1], unrelated thresholding and dependent threshold are set, you can rule out continuous item KPI, reminder item KPI nothing to do with items KPI.For
The overlapping degree between continuous item KPI is reduced as far as possible, can call cluster table, strong phase is ruled out from same cluster correlation item KPI
Item KPI is closed, other are then referred to as related similar terms KPI with the continuous item KPI in cluster, then calculate returning for strong correlation item KPI
After one changes weight, each KQI related KPI matrixes are exported, include weight.
Step 3, aforementioned four strong correlation item KPI individual event door can be calculated according to Bayesian formula with it is expected False Rate
Limit.
Step 4, according to weight sequencing, 4 strong correlation item KPI are divided to for two groups, every group of 2 KPI are one group, are calculated above-mentioned
Four strong correlation item KPI two dimension joint False Rate desired distribution, such as:
" up initial HARQ retransmits ratio " is used as the 1st group with " up per PRB average throughputs ";
" negative 110dBm coverage rates " is used as the 2nd group with " user plane averagely activates UE numbers ".
Step 5, the two dimension exported according to strong correlation item KPI weight proportions and above-mentioned steps 4 it is expected False Rate distribution, example
Such as, it is quantified as 500*500 matrix.
Step 6, the numerical index numbering for determining according to KPI weight size adaptations strong correlation item KPI in every group, can be false
If using the bigger strong correlation item KPI of weight as main KPI in every group.For example, offset is set as 100, and in the 2nd group, main KPI:
The index number that user plane averagely activates UE numbers is 500/2+100*0.244/ (0.244+0.226)=302, then default to meet
False Rate<24.71%, two dimension can be inquired about and it is expected that False Rate distribution matrix obtains, secondary KPI:The index of negative 100dBm coverage rates
Numbering is 387.According to the index number of each KPI in group, the KPI value inquired is joint threshold value.It can thus obtain
To main KPI and time KPI two dimension joint thresholding.
Step 7, Fig. 5 are that a kind of verify provided in an embodiment of the present invention influences the key of Key Quality Indicator KQI correlations
One every data output schematic diagram of energy index KPI methods Application Example, as shown in figure 5, the individual event door according to strong correlation item KPI
Limit combines thresholding with two dimension, with reference to obtained KQI-KPI matched datas space, calculates KQI:Http response time delay is final to be sentenced pair
Rate and misdetection rate.
Step 8, Fig. 6 are that a kind of verify provided in an embodiment of the present invention influences the key of Key Quality Indicator KQI correlations
The result of implementation schematic diagram of energy index KPI methods Application Example one, with KPI:Exemplified by negative 100dBm coverage rates, it can be estimated that multidimensional
The gain of thresholding vs individual event thresholdings.KPI according to Fig. 5:The individual event thresholding of negative 100dBm coverage rates is with combining KPI:" user
Averagely activate UE numbers in face " two-dimentional thresholding, calculating the KQI that two-dimentional thresholding additionally rules out transfinites number of cells, as a result such as Fig. 5 institutes
Show.After two-dimentional thresholding, transfinite KQI of a large amount of coverage rates between 88.97% and 96.02% additionally have found.
Step 9, it is expected False Rate as variable, under given KQI warning thresholdings, carry out above-mentioned steps 3 to step 5
Calculate, obtain KQI:The misdetection rate curve of http response time delay.Whether can be reached according to misdetection rate or approach 0 judgement strong correlation
Whether item KPI is complete.
In this example, table 2 influences the related key performances of Key Quality Indicator KQI for a kind of verify provided by the invention
Proof list in index KPI method Application Example one, when http response time delay takes 90ms, 95ms, 100ms and 105ms,
Zero is can reach to fail to judge, it is as shown in table 2 below.Thus, it is possible to association analysis scheme and system that the present invention describes are verified, from 30
Complete KQI continuous items are have found in KPI items.
Table 2 influences the related Key Performance Indicator KPI of Key Quality Indicator KQI table for checking
Http response time delay | 90ms | 95ms | 100ms | 105ms |
Multidimensional thresholding False Rate | 15.9% | 23% | 31.9% | 41.5% |
Corresponding multidimensional thresholding misdetection rate | 0 | 0 | 0 | 0 |
By above-described embodiment, identify whether KQI exceedes by using the KPI of strong correlation one-dimensional thresholding and multidimensional thresholding
The deterioration thresholding alerted is needed, so that False Rate is reduced to the present embodiment from minimum 50% or so of previous methods
Less than 20%;Simultaneously on the premise of so low False Rate is ensured, misdetection rate is reduced to from minimum 45% or so of previous methods
Less than 20%.That is, the strong correlation KPI that judgement KQI deteriorates is accurate and missing rate is low.
A kind of verify provided in an embodiment of the present invention influences the related Key Performance Indicator KPI's of Key Quality Indicator KQI
Method, it can be determined that whether KPI set to be assessed can completely explain that KQI transfinites, and assesses KPI and gather incomplete probability,
For use in the search and discovery of hiding associations is instructed, i.e.,:Whether there is unknown strong correlation KPI not to be included into be assessed
Set, and the quantization threshold of the final continuous item KPI by providing KQI, can be relatively accurately for judging whether KQI transfinites
The network optimization is instructed to be adjusted with systematic parameter, so as to preferably solve the feelings of " the good KQI of KPI are poor, or the good KPI of KQI poor "
Condition.
Fig. 7 is that a kind of verify provided in an embodiment of the present invention influences the related Key Performance Indicators of Key Quality Indicator KQI
The structural representation of KPI device embodiments one, as shown in fig. 7, the device includes determining unit 10, threshold cell 20, judges list
Member 30 and authentication unit 40;
The determining unit 10 is arranged to the Key Quality Indicator KQI of determination business and at least one and business pass
The Key Performance Indicator KPI of key quality index KQI strong correlations;
The threshold cell 20 is arranged to Key Performance Indicator KPI and the Key Quality Indicator according to the strong correlation
The Key Performance Indicator KPI of the KQI degree of association and each strong correlation weighted value obtains the Key Performance Indicator of each strong correlation
The multidimensional thresholding of the Key Performance Indicator KPI combinations of KPI one-dimensional thresholding and multiple strong correlations;
The judging unit 30 is arranged to the one-dimensional thresholding of the Key Performance Indicator KPI according to acquired each strong correlation
Obtain the Key Performance Indicator KPI's of the strong correlation with the Key Performance Indicator KPI of the multiple strong correlations multidimensional thresholdings combined
Sentence to rate;
The authentication unit 40 is arranged to if it is determined that described sentence when meeting rate predetermined threshold value, obtains the strong correlation
Key Performance Indicator KPI exceedes predetermined misdetection rate when deteriorating thresholding in the Key Quality Indicator KQI.
A kind of verify provided in an embodiment of the present invention influences the related Key Performance Indicator KPI's of Key Quality Indicator KQI
Device, including:Determining unit, threshold cell, judging unit and authentication unit, by by determining the KQI of business and corresponding
The KPI of at least one strong correlation, according to the KPI of each strong correlation and KQI degree of association and the KPI of each strong correlation weighted value
The KPI of all strong correlations one-dimensional threshold value and multidimensional threshold value can be got, so as to obtain KQI pairs of the business
The KPI at least one strong correlation answered complete degree, enabling all-sidedly and accurately verify the key of Key Quality Indicator
Energy index, improve the efficiency of the network optimization of a certain business Key Quality Indicator of service network in communication system.
Further, on the basis of above-described embodiment, the determining unit 10 is also configured to according to acquired every
The multidimensional door of the Key Performance Indicator KPI combinations of the Key Performance Indicator KPI of individual strong correlation one-dimensional thresholding and multiple strong correlations
Limit obtains the sentencing to after rate of the Key Performance Indicator KPI of the strong correlation, when the rising scale of False Rate exceedes skew thresholding
When, the Key Performance Indicator KPI of strong correlation corresponding to the Key Quality Indicator KQI of business is redefined, wherein, the False Rate
Sentence for 1- to rate.
Device provided in an embodiment of the present invention, above method embodiment can be performed, its implementing principle and technical effect class
Seemingly, will not be repeated here.
Further, on the basis of above-described embodiment, described device also includes adjustment unit 50, the adjustment unit 50
Be arranged to the Key Performance Indicator KPI according to acquired each strong correlation one-dimensional thresholding and multiple strong correlations it is key
Can the multidimensional thresholding of index KPI combination obtain the Key Performance Indicator KPI of the strong correlation and sentence to after rate, sentencing described in adjustment
To the predetermined threshold of rate, however, it is determined that sentence described in adjustment and approach 0 to the misdetection rate obtained after rate, it is determined that influence the Key Quality
The Key Performance Indicator KPI of index KQI strong correlation is not omitted.
Device provided in an embodiment of the present invention, above method embodiment can be performed, its implementing principle and technical effect class
Seemingly, will not be repeated here.
Further, on the basis of above-described embodiment, the adjustment unit 50 is also configured to sentencing to rate described in adjustment
Predetermined threshold after, however, it is determined that 0 can not be approached to the misdetection rate obtained after rate by sentencing described in adjustment, it is determined that influence the key
The Key Performance Indicator KPI of quality index KQI strong correlation has omission.
Device provided in an embodiment of the present invention, above method embodiment can be performed, its implementing principle and technical effect class
Seemingly, will not be repeated here.
Further, on the basis of above-described embodiment, the device also includes:Estimate unit 60;
The unit 60 of estimating is arranged to it is determined that the key performance for influenceing the strong correlation of the Key Quality Indicator KQI refers to
Marking KPI does not have after omitting, according to the Key Performance Indicator KPI of acquired each strong correlation one-dimensional thresholding and multiple strong phases
The multidimensional thresholding of the Key Performance Indicator KPI combinations of pass carries out data training to the Key Performance Indicator KPI of the strong correlation, in advance
Estimate misdetection rate during no Key Quality Indicator KQI data.
Device provided in an embodiment of the present invention, above method embodiment can be performed, its implementing principle and technical effect class
Seemingly, will not be repeated here.
It should be noted that Fig. 8 is a kind of Key Quality Indicator KQI provided in an embodiment of the present invention and related key
Energy index KPI analysis system schematic diagrames, as shown in figure 8, the system includes:Basic database 100, expert's interface 200, KQI/
KPI memory cell 300, data pre-processing unit 400, intelligent clustering unit 500, vector space resolving cell 600, quantization association
Computing unit 700, multidimensional threshold computation unit 800, mining analysis algorithm pond 900 and data craft wisdom storehouse 1000, wherein, it is right
In it is provided in an embodiment of the present invention it is a kind of verify influence the related Key Performance Indicator KPI devices of Key Quality Indicator KQI can be with
It is arranged in the multidimensional threshold computation unit 800 of the system.Below by the fortune to multidimensional threshold computation unit 800 within the system
Capable process is described in detail as follows:
S801, quantify association computing unit 700 input KQI items and corresponding strong correlation item KPI and its quantify the degree of association with
Normalized weight.Set the default of KQI to sentence quasi- rate, calculate each strong correlation item KPI one-dimensional thresholding.That is, when certain strong correlation item
When KPI crosses its one-dimensional thresholding, KQI is transfinited with the default probability for sentencing quasi- rate.
S802, intelligent clustering unit 500 call mining analysis algorithm pond 900, with data pre-processing unit 400 in S801
Obtained part index number characteristic value, the initial hierarchical cluster of categorical measure is not limited, as shown in figure 8, then intelligent clustering
Unit 500 calls data craft wisdom storehouse 1000 to carry out final cluster judgement, and exports cluster table.Optionally, data prediction
Unit 400 calls expert's interface 200 to carry out live expert and aids in cluster judgement.
S803, vector space resolving cell 600 call the KQI and KPI data that data pre-processing unit 400 exports, and form
KQI-KPI data vectors space, mining analysis algorithm pond 900 carry out vector space decomposition operation, isolate quantifiable KQI-
KPI associates component.
S804, quantify association computing unit 700 to KQI-KPI association component progress normalizing quantization operations, obtain each KPI
With the corresponding KQI degree of association.KQI-KPI degrees of association scope is [0,1], sets unrelated thresholding and dependent threshold, you can rule out
Continuous item KPI, reminder item KPI nothing to do with items KPI.To reduce the overlapping degree between continuous item KPI as far as possible, quantify association and calculate
Unit 700 calls the cluster table of the output of intelligent clustering unit 500, and strong correlation item KPI is ruled out from same cluster correlation item KPI,
Other are then referred to as related similar terms KPI with the continuous item KPI in cluster.Quantify association computing unit 700 and calculate strong correlation item
After KPI normalized weight, each KQI related KPI matrixes are exported, include weight.
S805, multidimensional threshold computation unit 800 are according to the strong correlation item for quantifying the KQI that association computing unit 700 rules out
KPI and weight, strong correlation item KPI individual event thresholding and multidimensional thresholding are calculated successively.According to business and system accuracy requirement, multidimensional
The dimension of thresholding is from two dimension, and the quantity no more than strong correlation item KPI.
Although disclosed herein embodiment as above, described content be only readily appreciate the present invention and use
Embodiment, it is not limited to the present invention.Technical staff in any art of the present invention, taken off not departing from the present invention
On the premise of the spirit and scope of dew, any modification and change, but the present invention can be carried out in the form and details of implementation
Scope of patent protection, still should be subject to the scope of the claims as defined in the appended claims.
Claims (10)
1. a kind of verify the method for influenceing the related Key Performance Indicator KPI of Key Quality Indicator KQI, it is characterised in that described
Method includes:
Determine the Key Quality Indicator KQI of business and at least one and Key Quality Indicator KQI strong correlations of business pass
Key performance indications KPI;
According to the Key Performance Indicator KPI of the strong correlation and Key Quality Indicator KQI degree of association and each strong correlation
Key Performance Indicator KPI weighted value obtain each strong correlation Key Performance Indicator KPI one-dimensional thresholding and multiple strong phases
The multidimensional thresholding of the Key Performance Indicator KPI combinations of pass;
Referred to according to the key performance of the Key Performance Indicator KPI of acquired each strong correlation one-dimensional thresholding and multiple strong correlations
The Key Performance Indicator KPI that the multidimensional thresholding of mark KPI combinations obtains the strong correlation is sentenced to rate;
If it is determined that described sentence when meeting rate predetermined threshold value, the Key Performance Indicator KPI of the strong correlation is obtained in the pass
Key quality index KQI exceedes misdetection rate during predetermined deterioration thresholding.
2. according to the method for claim 1, it is characterised in that refer to according to the key performance of acquired each strong correlation
The multidimensional thresholding for marking the Key Performance Indicator KPI combinations of KPI one-dimensional thresholding and multiple strong correlations obtains the pass of the strong correlation
Key performance indications KPI's sentences to after rate, in addition to:
When the rising scale of False Rate is prescribed a time limit more than offset gate, strong phase corresponding to the Key Quality Indicator KQI of business is redefined
The Key Performance Indicator KPI of pass, wherein, the False Rate is that 1- is sentenced to rate.
3. according to the method for claim 1, it is characterised in that refer to according to the key performance of acquired each strong correlation
The multidimensional thresholding for marking the Key Performance Indicator KPI combinations of KPI one-dimensional thresholding and multiple strong correlations obtains the pass of the strong correlation
Key performance indications KPI's sentences to after rate, in addition to:
The predetermined threshold to rate is sentenced described in adjustment, however, it is determined that sentence described in adjustment and 0 is approached to the misdetection rate obtained after rate, it is determined that shadow
The Key Performance Indicator KPI for ringing the strong correlation of the Key Quality Indicator KQI is not omitted.
4. according to the method for claim 3, it is characterised in that after sentencing described in adjustment to the predetermined threshold of rate, also wrap
Include:
If it is determined that 0 can not be approached to the misdetection rate obtained after rate by sentencing described in adjustment, it is determined that influence the Key Quality Indicator KQI
The Key Performance Indicator KPI of strong correlation have omission.
5. according to the method for claim 3, it is characterised in that it is determined that influenceing the strong phase of the Key Quality Indicator KQI
After the Key Performance Indicator KPI of pass is not omitted, in addition to:
Referred to according to the key performance of the Key Performance Indicator KPI of acquired each strong correlation one-dimensional thresholding and multiple strong correlations
The multidimensional thresholding for marking KPI combinations carries out data training to the Key Performance Indicator KPI of the strong correlation, estimates no key
Misdetection rate during quality index KQI data.
6. a kind of verify the device for influenceing the related Key Performance Indicator KPI of Key Quality Indicator KQI, it is characterised in that the dress
Put including:Determining unit, threshold cell, judging unit and authentication unit;
The determining unit is arranged to the Key Quality Indicator KQI of determination business and at least one and business Key Quality
The Key Performance Indicator KPI of index KQI strong correlations;
The threshold cell is arranged to according to the Key Performance Indicator KPI and the Key Quality Indicator KQI of the strong correlation
The Key Performance Indicator KPI of the degree of association and each strong correlation weighted value obtains the Key Performance Indicator KPI's of each strong correlation
The multidimensional thresholding of the Key Performance Indicator KPI of one-dimensional thresholding and multiple strong correlations combinations;
The judging unit is arranged to according to the Key Performance Indicator KPI of acquired each strong correlation one-dimensional thresholding and multiple
The Key Performance Indicator KPI that the multidimensional thresholding of the Key Performance Indicator KPI combinations of strong correlation obtains the strong correlation is sentenced to rate;
The authentication unit is arranged to if it is determined that described sentence when meeting rate predetermined threshold value, obtains the key of the strong correlation
Can misdetection rates of the index KPI when the Key Quality Indicator KQI exceedes predetermined deterioration thresholding.
7. device according to claim 6, it is characterised in that the determining unit is also configured to:
In the one-dimensional thresholding of the Key Performance Indicator KPI according to acquired each strong correlation and the key performance of multiple strong correlations
The multidimensional thresholding of index KPI combinations obtains the sentencing to after rate of the Key Performance Indicator KPI of the strong correlation, upper when False Rate
The ratio of liter is prescribed a time limit more than offset gate, redefines the Key Performance Indicator of strong correlation corresponding to the Key Quality Indicator KQI of business
KPI, wherein, the False Rate is that 1- is sentenced to rate.
8. device according to claim 6, it is characterised in that described device also includes adjustment unit, the adjustment unit
Be arranged to the Key Performance Indicator KPI according to acquired each strong correlation one-dimensional thresholding and multiple strong correlations it is key
Can the multidimensional thresholding of index KPI combination obtain the Key Performance Indicator KPI of the strong correlation and sentence to after rate, sentencing described in adjustment
To the predetermined threshold of rate, however, it is determined that sentence described in adjustment and approach 0 to the misdetection rate obtained after rate, it is determined that influence the Key Quality
The Key Performance Indicator KPI of index KQI strong correlation is not omitted.
9. device according to claim 6, it is characterised in that the adjustment unit is also configured to sentencing to rate described in adjustment
Predetermined threshold after, however, it is determined that 0 can not be approached to the misdetection rate obtained after rate by sentencing described in adjustment, it is determined that influence the key
The Key Performance Indicator KPI of quality index KQI strong correlation has omission.
10. device according to claim 9, it is characterised in that the device also includes:Estimate unit;
The unit of estimating is arranged to it is determined that influenceing the Key Performance Indicator KPI of the strong correlation of the Key Quality Indicator KQI
After not omitting, according to the Key Performance Indicator KPI of acquired each strong correlation one-dimensional thresholding and multiple strong correlations
The multidimensional thresholding of Key Performance Indicator KPI combinations carries out data training to the Key Performance Indicator KPI of the strong correlation, estimates not
There is the misdetection rate during Key Quality Indicator KQI data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610565431.1A CN107623924A (en) | 2016-07-15 | 2016-07-15 | It is a kind of to verify the method and apparatus for influenceing the related Key Performance Indicator KPI of Key Quality Indicator KQI |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610565431.1A CN107623924A (en) | 2016-07-15 | 2016-07-15 | It is a kind of to verify the method and apparatus for influenceing the related Key Performance Indicator KPI of Key Quality Indicator KQI |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107623924A true CN107623924A (en) | 2018-01-23 |
Family
ID=61087950
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610565431.1A Withdrawn CN107623924A (en) | 2016-07-15 | 2016-07-15 | It is a kind of to verify the method and apparatus for influenceing the related Key Performance Indicator KPI of Key Quality Indicator KQI |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107623924A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110690982A (en) * | 2018-07-05 | 2020-01-14 | 北京亿阳信通科技有限公司 | Telecommunication network management performance data correlation analysis method and system |
WO2020244650A1 (en) * | 2019-06-06 | 2020-12-10 | 华为技术有限公司 | Method, device and system for acquiring performance intention indicator |
CN112258027A (en) * | 2020-10-21 | 2021-01-22 | 平安科技(深圳)有限公司 | KPI (Key performance indicator) optimization method, device, equipment and medium |
CN112996015A (en) * | 2019-12-18 | 2021-06-18 | 中国移动通信集团河南有限公司 | Index association relationship construction method and device |
WO2021147370A1 (en) * | 2020-01-24 | 2021-07-29 | 华为技术有限公司 | Method, apparatus and system for training fault detection model |
CN114095387A (en) * | 2020-07-29 | 2022-02-25 | 中国移动通信集团北京有限公司 | Information determination method, device, equipment and medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102143507A (en) * | 2010-11-23 | 2011-08-03 | 北京中创信测科技股份有限公司 | Method and system for monitoring service quality, and analytical method and system therefor |
CN102625344A (en) * | 2012-03-13 | 2012-08-01 | 重庆信科设计有限公司 | Model and method for evaluating user experience quality of mobile terminal |
US20130316701A1 (en) * | 2011-02-24 | 2013-11-28 | Datang Mobile Communications Equipment Co., Ltd | Data processing method and device for essential factor lost score |
CN104683998A (en) * | 2013-12-02 | 2015-06-03 | 中兴通讯股份有限公司 | Method and device for processing business quality |
JP2016051985A (en) * | 2014-08-29 | 2016-04-11 | 日本電信電話株式会社 | Quality management system, method, and program |
-
2016
- 2016-07-15 CN CN201610565431.1A patent/CN107623924A/en not_active Withdrawn
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102143507A (en) * | 2010-11-23 | 2011-08-03 | 北京中创信测科技股份有限公司 | Method and system for monitoring service quality, and analytical method and system therefor |
US20130316701A1 (en) * | 2011-02-24 | 2013-11-28 | Datang Mobile Communications Equipment Co., Ltd | Data processing method and device for essential factor lost score |
CN102625344A (en) * | 2012-03-13 | 2012-08-01 | 重庆信科设计有限公司 | Model and method for evaluating user experience quality of mobile terminal |
CN104683998A (en) * | 2013-12-02 | 2015-06-03 | 中兴通讯股份有限公司 | Method and device for processing business quality |
JP2016051985A (en) * | 2014-08-29 | 2016-04-11 | 日本電信電話株式会社 | Quality management system, method, and program |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110690982A (en) * | 2018-07-05 | 2020-01-14 | 北京亿阳信通科技有限公司 | Telecommunication network management performance data correlation analysis method and system |
CN110690982B (en) * | 2018-07-05 | 2023-07-14 | 北京亿阳信通科技有限公司 | Method and system for correlation analysis of management performance data of telecommunication network |
WO2020244650A1 (en) * | 2019-06-06 | 2020-12-10 | 华为技术有限公司 | Method, device and system for acquiring performance intention indicator |
CN112996015A (en) * | 2019-12-18 | 2021-06-18 | 中国移动通信集团河南有限公司 | Index association relationship construction method and device |
CN112996015B (en) * | 2019-12-18 | 2023-11-03 | 中国移动通信集团河南有限公司 | Index association relation construction method and device |
WO2021147370A1 (en) * | 2020-01-24 | 2021-07-29 | 华为技术有限公司 | Method, apparatus and system for training fault detection model |
CN114095387A (en) * | 2020-07-29 | 2022-02-25 | 中国移动通信集团北京有限公司 | Information determination method, device, equipment and medium |
CN114095387B (en) * | 2020-07-29 | 2023-09-05 | 中国移动通信集团北京有限公司 | Information determination method, device, equipment and medium |
CN112258027A (en) * | 2020-10-21 | 2021-01-22 | 平安科技(深圳)有限公司 | KPI (Key performance indicator) optimization method, device, equipment and medium |
CN112258027B (en) * | 2020-10-21 | 2021-05-18 | 平安科技(深圳)有限公司 | KPI (Key performance indicator) optimization method, device, equipment and medium |
WO2021164361A1 (en) * | 2020-10-21 | 2021-08-26 | 平安科技(深圳)有限公司 | Kpi optimization method, apparatus, device, and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107623924A (en) | It is a kind of to verify the method and apparatus for influenceing the related Key Performance Indicator KPI of Key Quality Indicator KQI | |
EP3327582A1 (en) | Method and apparatus for completing a knowledge graph | |
Md Saad et al. | Hamming distance method with subjective and objective weights for personnel selection | |
Xiao | An intelligent complex event processing with numbers under fuzzy environment | |
WO2021051917A1 (en) | Artificial intelligence (ai) model evaluation method and system, and device | |
WO2023071121A1 (en) | Multi-model fusion-based object detection method and apparatus, device and medium | |
Zuo et al. | A large group decision-making method and its application to the evaluation of property perceived service quality | |
CN108989092A (en) | A kind of wireless network predicting method, electronic equipment and storage medium | |
CN107995682A (en) | Wireless access independent positioning method, device, electronic equipment and storage medium | |
CN115378988A (en) | Data access abnormity detection and control method and device based on knowledge graph | |
Ye et al. | Global optimization method using adaptive and parallel ensemble of surrogates for engineering design optimization | |
US20150169794A1 (en) | Updating location relevant user behavior statistics from classification errors | |
WO2022242181A1 (en) | Method and apparatus for evaluating health degree indexes of layers of smart substation | |
Zhang et al. | Federated Multi-Task Learning with Non-Stationary and Heterogeneous Data in Wireless Networks | |
CN116468967B (en) | Sample image screening method and device, electronic equipment and storage medium | |
CN109711555B (en) | Method and system for predicting single-round iteration time of deep learning model | |
CN109728958A (en) | A kind of network node trusts prediction technique, device, equipment and medium | |
CN111159768A (en) | Evaluation method for link privacy protection effect of social network | |
CN108133234B (en) | Sparse subset selection algorithm-based community detection method, device and equipment | |
CN116226336A (en) | Mobile base station complaint root cause analysis method based on Bayesian network | |
CN108268982B (en) | Large-scale active power distribution network decomposition strategy evaluation method and device | |
Yu et al. | Analysis of distributed database access path prediction based on recurrent neural network in internet of things | |
WO2022217712A1 (en) | Data mining method and apparatus, and computer device and storage medium | |
Wang et al. | Research on comprehensive performance evaluation of communication network based on the fuzzy number intuitionistic fuzzy information | |
CN114385403A (en) | Distributed cooperative fault diagnosis method based on double-layer knowledge graph framework |
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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180123 |
|
WW01 | Invention patent application withdrawn after publication |