CN104113869B - A kind of potential report user's Forecasting Methodology and system based on signaling data - Google Patents
A kind of potential report user's Forecasting Methodology and system based on signaling data Download PDFInfo
- Publication number
- CN104113869B CN104113869B CN201410279999.8A CN201410279999A CN104113869B CN 104113869 B CN104113869 B CN 104113869B CN 201410279999 A CN201410279999 A CN 201410279999A CN 104113869 B CN104113869 B CN 104113869B
- Authority
- CN
- China
- Prior art keywords
- user
- report user
- report
- potential
- characteristic vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a kind of potential report user's Forecasting Methodology and system based on signaling data, the business support technical field belonged in moving communicating field, being established first based on A interface signaling data includes the whole network user characteristics vector of report user's characteristic vector and non-report user's characteristic vector, the business similarity of non-report user and report user are calculated further according to report user's characteristic vector and non-report user's characteristic vector, the potential report user in non-report user is finally determined according to business similarity, the higher user of business similarity is bigger for the possibility of potential report user.This method and system, by carrying out characteristic vector modeling to mass users data, potential report user is excavated according to report user, before user really complains or turns net, give warning in advance, it will complain or turn net situation and eliminate in bud, user perceives and reduction client of operator maintenance cost provides the foundation to improve.
Description
Technical field
The present invention relates to the business support technology in moving communicating field, and in particular to a kind of based on the potential of signaling data
Report user's Forecasting Methodology and system.
Background technology
With the continuous development of the communications industry, the arrival in 4G epoch, various new communication networking problems continue to bring out, user
Complaint amount is also growing day by day.Meanwhile contemporary client has no longer been to make a phone call, send short messages that for the service of the offer of operator
The simple mode of sample, more emphasis on personalized, good client perception, user satisfaction is only improved, reduces customer complaint, and handle
The complaint advanced processing that may occur, just can guarantee that the continuous development of Operator Specific Service.Operator possesses the user of magnanimity at present
Data, either magnanimity signaling data, or a large amount of customer profile datas, all with great value.It is but most of
Operator only goes to consider, does not make full use of sea for the processing and prevention of customer complaint from the angle of network index
The advantage of data resource is measured, by the method for data mining, personalized complain formulated for user solves and complained prediction side
Case.
In moving communicating field, prior art, handled primarily directed to the complaint occurred, based on set thing
Real remedies, and the user that may be complained can not be predicted in advance, prevented trouble before it happens.Complaint handling flow is general at present
For：Service of the client for operator, which produces, to be discontented with, and is then dialed service calls or is complained by the network platform, customer service
After center receives customer complaint, distribute leaflets support center to new business, and support center is analyzed according to complaint particular content, be fixed
Position, solve the problems, such as the complaint of user, then reply work order.Therefore, such complaint handling experienced " customer service -- monitoring -- customer service "
Flow, treatment effeciency is low, and the wasting of resources is big, and the process limited is difficult to ensure that, CSAT is poor.It is existing to complain prediction
Method is also only to calculate some network element indexs, such as weak covering, anomalous event, establishes warning system, judges that user is led to this
Quality is talked about, so as to predict the user that may be complained.
On the whole, existing complaint Forecasting Methodology mainly has following several：
One kind is Qualitative Forecast Methods, is that the history and present situation of things development are made explanations, analyzed and judged, so as to comprehensive
The one or more possibilities for the future trend for pointing out things development closed, such as market survey method, Delphi method.
Another kind of is quantitative forecast method, mainly calculating network index, the poor user's inventory of output network index, as
Complain prediction inventory.
Problems be present in existing complaint prediction scheme：
1. the method for market survey can only be directed to limited sample and customer group, the user of substantial amounts of entirety can not be grasped
Complain feature.
2. it is existing to calculate the complaint Forecasting Methodology of network element index, using some rigid indexs of network as judgement bar
Part, it is difficult to reflect mobile subscriber's subjective perception situation, and personality analysis is carried out according to user itself differentiation feature.
3. existing complaint handling is all after complaint has occurred and that, the mode remedied and explained afterwards, customer service
Personnel are only confined in complaining the pacifying of client, handle above, or the plain analysis of causes is done to complaint content, have no profound level
Customer complaint analysis.
4. existing method is based on the work order for having occurred and that complaint, and many users produce severe perception for service,
They can't go to complain, but directly turn net, then existing complaint handling system can not just carry to this kind of user
It is preceding to show loving care for and keep.According to marketing principle, the cost developed needed for a new user is maintain old user 6 times, thus is carried
Before predict may complaint user it is particularly important to the income and profit of operator.
The content of the invention
For the needs of defect present in prior art and practical application, it is an object of the invention to provide one kind to be based on
Potential the report user's Forecasting Methodology and system of signaling data, by this method and system can look-ahead go out potential complain and use
Family, give warning in advance, improve the perception of user, reduce maintenance cost.
To achieve the above object, the technical solution adopted by the present invention is as follows：
A kind of potential report user's Forecasting Methodology based on signaling data, comprises the following steps：
(1) the whole network user characteristics vector is established based on A interface signaling data；The whole network user characteristics vector includes
Report user's characteristic vector and non-report user's characteristic vector；Parameter in user characteristics vector is the business ginseng of corresponding business
Number；
(2) non-report user and report user are calculated according to report user's characteristic vector and non-report user's characteristic vector
Business similarity；
(3) the potential report user in non-report user is determined according to business similarity, the higher user of business similarity is
The possibility of potential report user is bigger.
Further, a kind of potential report user's Forecasting Methodology based on signaling data as described above, in step (1) and step
Suddenly between (2), in addition to：
The whole network user characteristics vector is standardized by (1-2).
Further, a kind of potential report user's Forecasting Methodology based on signaling data as described above, in step (1), with
Single business is that granularity establishes the whole network user characteristics vector；The business includes call, short message, no-response paging, location updating
And switching on and shutting down.
Further, a kind of potential report user's Forecasting Methodology based on signaling data as described above, chosen distance are complained
The customer complaint time is less than setting duration, corresponding with complaint business business and establishes the whole network user characteristics vector.
Further, a kind of potential report user's Forecasting Methodology based on signaling data as described above, in step (1-2),
Described be standardized the whole network user characteristics vector refers to the parameter values of characteristic vector being normalized into [- 1,1].
Further, a kind of potential report user's Forecasting Methodology based on signaling data as described above, by characteristic vector
The formula that parameter values are standardized is：
Wherein, t represents the numerical value of some parameter of some user in the whole network user characteristics vector, and min represents the whole network user
Described in minimum value in some parameter values, max represents the maximum in some parameter values described in the whole network user；
As max=min, the parameter values that parameter is corresponded in the whole network characteristic vector are unified into assignment.
Further, a kind of potential report user's Forecasting Methodology based on signaling data as described above, its step (2)
In, the Euclidean distance that parameter is corresponded to by calculating report user's characteristic vector with non-report user's characteristic vector is calculated and not complained
User and the business similarity of report user, the smaller similarity of Euclidean distance are higher；The calculation formula of the Euclidean distance is：
Wherein, d (x, y) represents the business similarity of non-report user and report user, x represent the feature of report user to
Amount, y represent the characteristic vector of non-report user, and n represents the number of service parameter in the whole network user characteristics vector, xkRepresent to complain
The numerical value of k-th of service parameter, y in user characteristics vectorkRepresent the number of k-th of service parameter in non-report user's characteristic vector
Value.
Further, a kind of potential report user's Forecasting Methodology based on signaling data as described above, in step (3),
When the similarity of non-report user and report user meet the decision threshold of setting, the user is defined as potential report user.
A kind of potential report user's forecasting system based on signaling data, including：
Characteristic vector establishes module：For establishing the whole network user characteristics vector based on A interface signaling data；It is described complete
Network users characteristic vector includes report user's characteristic vector and non-report user's characteristic vector；Parameter in user characteristics vector is
The service parameter of corresponding business；
Similarity calculation module：Do not complained for being calculated according to report user's characteristic vector and non-report user's characteristic vector
User and the business similarity of report user；
Potential report user's determining module：Potential report user in non-report user, industry are determined according to business similarity
The higher user of similarity of being engaged in is bigger for the possibility of potential report user.
Further, a kind of potential report user's forecasting system based on signaling data as described above, the system also include：
Vectorial standardized module：For the whole network user characteristics vector to be standardized.
The beneficial effects of the present invention are：Method and system of the present invention, according to the data that report user has occurred
Feature, the potential report user that may be complained is found, improve the satisfaction of the whole network entirety user, use can be effectively reduced
Family turns net rate.Complaint modeling of this method based on report user, so as to which Auto-matching goes out in network to be likely present same problem
User, give warning in advance, will complain or turn net situation and eliminate in bud, drive the whole network to use according to report user so as to realize
Family, the focusing orientation problem of network problem is realized, can effectively reduce the network operation cost of operator, and this method and tradition
Forecasting Methodology compare, the accuracy of prediction can be effectively improved.
Brief description of the drawings
Fig. 1 is a kind of structure of potential report user's forecasting system based on signaling data in the specific embodiment of the invention
Block diagram；
Fig. 2 is a kind of flow of potential report user's Forecasting Methodology based on signaling data in the specific embodiment of the invention
Figure.
Embodiment
With reference to Figure of description, the present invention is described in further detail with embodiment.
Fig. 1 shows a kind of structural frames of potential report user's forecasting system based on signaling data in embodiment
Figure, the system mainly establish module 11, vectorial standardized module 12, similarity calculation module 13 and potential throwing including characteristic vector
User's determining module 14 is told, wherein：
Characteristic vector is established module 11 and is used to be established the whole network user characteristics vector based on A interface signaling data；It is described
The whole network user characteristics vector includes report user's characteristic vector and non-report user's characteristic vector；Parameter in user characteristics vector
For the service parameter of corresponding business；
Vectorial standardized module 12 is used to the whole network user characteristics vector being standardized；
Similarity calculation module 13, which is used to be calculated according to report user's characteristic vector and non-report user's characteristic vector, does not throw
Tell the business similarity of user and report user；
Potential report user's determining module 14 determines the potential report user in non-report user, industry according to business similarity
The higher user of similarity of being engaged in is bigger for the possibility of potential report user.
Fig. 2 shows a kind of potential throwing based on signaling data based on system shown in Fig. 1 in present embodiment
The flow chart of user in predicting method is told, this method comprises the following steps：
Step S21：The whole network user characteristics vector is established based on A interface signaling data；
The Forecasting Methodology of the present invention is based on the modeling and analysis that A interface signalings data are carried out with characteristic vector, according to
The data characteristics of report user occurs, finds the potential report user that may be complained.That is this programme is to complain in work order
The data for having occurred and that report user are Sample Establishing model, and it is modeling object to select the signaling data related to complaint.
First, the whole network user characteristics vector, the whole network user characteristics vector bag are established based on A interface signaling data
Include report user's characteristic vector and non-report user's characteristic vector, the parameter in user characteristics vector is the business ginseng of corresponding business
Number, that is to say, that the user characteristics vector in present embodiment is business (communication service) characteristic vector, report user's feature to
Amount is that the related service parameter of report user is modeled, and non-report user's vector is pair also without pair of the user complained
Relevant parameter is answered to be modeled.
Communication interface of the A interfaces between network subsystem (NSS) and base station sub-system (BSS), passes through A interface signaling data
The service parameter data for obtaining different communication business are any technique commonly known.
When establishing the whole network user characteristics vector, user characteristics vector is established by granularity of single business, wherein, the industry
Business includes call, short message, no-response paging, location updating, switching on and shutting down etc..Specifically select any time business establish user characteristics to
During amount, in present embodiment preferably with chosen distance report user complain the time be less than setting duration, with complaint business phase
Corresponding business establishes the whole network user characteristics vector.For example, a report user was thrown talk business in 10 points of certain day
Tell, then the session parameter data of 10 points or so of the whole network user can be selected to establish characteristic vector.
Different business, the parameter of its characteristic vector is different, and user can be according to the industry for being actually needed selected needs
Business parameter.For example, for this business of conversing, the detailed record sheet of business diagnosis that can utilize system is that BDR tables establish call
Service feature is vectorial, and talk business characteristic vector field is as shown in table 1 below in present embodiment.
The row of field one are the parameter of talk business characteristic vector in table 1.
Table 1
Step S22：The whole network user characteristics vector is standardized；
In present embodiment, the whole network user characteristics vector aims of standardization are the whole network users that will be established in step S21
The parameter values of characteristic vector are normalized between -1 to 1, naturally it is also possible to other standards are selected, because some parameters
Index value is very big but importance is not maximum, by standardization, makes all index importances identical, is characteristic vector
Parameter there is identical value scope.The formula of present embodiment Plays is as follows：
Wherein, t represents the numerical value of some parameter of some user in the whole network user characteristics vector, and min represents the whole network user
Described in (all users in the whole network user characteristics vector, including report user and non-report user) in some parameter values
Minimum value, max represent the maximum in some parameter values described in the whole network user；
As max=min, i.e., when the maximum of the numerical value of a certain parameter is identical with minimum value in the whole network user, then can not
It is standardized using above-mentioned formula, at this moment the parameter values that parameter is corresponded in characteristic vector can be unified assignment, such as
It is entered as average value or is entered as 0.
For example, there is 5 users in the whole network user, the 1st user is the user for actually occurring complaint, and other 4 users are
Non- report user in existing network, and potential report user may be included in this 4 users.Talk business for this 5 users,
From five parameter indexs in the talk business characteristic vector shown in table 1 in present embodiment：Rise exhale the time, rise exhale longitude,
Rise and exhale latitude, on-hook longitude, on-hook latitude, the vector of the whole network user characteristics comprising this five parameters of foundation is as shown in table 2.
The characteristic vector of a user is represented in table 2 per a line, wherein representated by report user's a line for report user
Characteristic vector.
| StartTime | Rise and exhale longitude | Rise and exhale latitude | On-hook longitude | On-hook latitude | |
| Report user | 16 | 111.65865 | 40.81055 | 111.65865 | 40.81055 |
| User 1 | 20 | 111.412865 | 40.028756 | 111.412865 | 40.028756 |
| User 2 | 8 | 111.52961 | 40.704372 | 111.52961 | 40.704372 |
| User 3 | 11 | 111.67084 | 40.82067 | 111.67084 | 40.82067 |
| User 4 | 10 | 111.644173 | 40.82008 | 111.64417 | 40.82008 |
Table 2
The maximum of each row and minimum value are maximum and the minimum value such as institute of table 3 of each service parameter in above-mentioned table 2
Show：
| StartTime | Rise and exhale longitude | Rise and exhale latitude | On-hook longitude | On-hook latitude | |
| max | 20 | 111.67084 | 40.82067 | 111.67084 | 40.82067 |
| min | 8 | 111.412865 | 40.028756 | 111.412865 | 40.028756 |
| max-min | 12 | 0.257975 | 0.791914 | 0.257975 | 0.791914 |
Table 3
The talk business characteristic vector in table 2 is standardized using standardization formula, with report user's
Exemplified by StartTime fields, the field value is 16, then its standardized calculation formula is as follows：
Data in table 2 after the characteristic vector standard parameter of all users are as shown in table 4：
| StartTime | Rise and exhale longitude | Rise and exhale latitude | On-hook longitude | On-hook latitude | |
| Report user | 0.33 | 0.91 | 0.97 | 0.91 | 0.97 |
| User 1 | 1 | -1 | -1 | -1 | -1 |
| User 2 | -1 | -0.09 | 0.71 | -0.09 | 0.71 |
| User 3 | -0.5 | 1 | 1 | 1 | 1 |
| User 4 | -0.67 | 0.79 | 1.00 | 0.79 | 1.00 |
Table 4
Step S23：Non- report user is calculated with complaining according to report user's characteristic vector and non-report user's characteristic vector
The business similarity of user；
Do not complained after the whole network user characteristics vector is established, it is necessary to be calculated by the method for comparing in the whole network user
User and the business similarity of report user, used so as to be found in non-report user with immediate potential complain of report user
Family.Specific lookup method is in present embodiment, by calculating report user's characteristic vector and non-report user's characteristic vector
The Euclidean distance of middle corresponding parameter calculates the similarity of the whole network user and report user, and Euclidean distance is smaller, and similarity is higher, is
The possibility of potential report user is bigger.The calculation formula of Euclidean distance is as follows in present embodiment：
Wherein, d (x, y) represents the business similarity of non-report user and report user, x represent the feature of report user to
Amount, y represent the characteristic vector of non-report user, and n represents the number of service parameter in the whole network user characteristics vector, xkRepresent to complain
The numerical value of k-th of service parameter, y in user characteristics vectorkRepresent the number of k-th of service parameter in non-report user's characteristic vector
Value.
In actual applications, if the report user in the whole network user characteristics vector has and multiple do not thrown, it is necessary to calculate respectively
Tell the business similarity between user and each report user.
For the whole network user characteristics vector in above-mentioned table 2 in present embodiment, the number n of service parameter is 5, user
1 is calculated as follows with the Euclidean distance of report user：
Euclidean distance in table 2 between 4 non-report users and report user is as shown in table 5 below：
| Customs Assigned Number | Euclidean distance |
| User 1 | 3.95 |
| User 2 | 1.98 |
| User 3 | 0.86 |
| User 4 | 1.02 |
Table 5
Step S24：Potential report user in non-report user, the higher use of business similarity are determined according to business similarity
Family is bigger for the possibility of potential report user.
Because the Euclidean distance of characteristic vector just represents the similarity between the whole network user and report user, apart from smaller,
Illustrate that both speech quality, client perception situation are closer.It can be seen that, immediate with report user is to use in table 5
Family 3, although illustrating that user 3 does not complain, it is bad to be particularly likely that user perceives, speech quality or network quality compared with
The user of difference, if handled not in time, user may be caused not complain but directly turn net.
In actual applications, K and the immediate potential report user of report user can be selected, as prediction inventory.K's
Selection should be arranged in suitable scope by experience, if K values are too small, be only able to find seldom potential report user, scope
It is too small, if K values are too big, it will much perceive good user and also divide potential report user into, cause erroneous judgement to increase.Therefore,
By setting rational judgment threshold, when the user in non-report user and the similarity of report user meet the decision threshold of setting
During value, the user is defined as potential report user.
In addition, the seeking scope of potential report user, also must be not necessarily the whole network user, because the increase meaning of data volume
Taste amount of calculation and searches the increase of time, the Hot Spot that Railway Project takes place frequently can also be selected to search according to actual conditions
Potential report user, the hit rate of prediction can also be improved.
In order to verify prediction effect of this programme for potential report user, work order is complained by obtaining certain districts and cities, is being thrown
Tell that the serious area of the network problem that takes place frequently is tested, potential report user's inventory is outputed by this programme, passes through phone
The mode of return visit is accurate to determine to predict whether.The scheme for testing traditional calculations network index simultaneously complains the accurate of prediction
Property.
One week report user 100 of work order is complained, by analysis of history report user's data, is analyzed and exported using this programme
Potential 1000 users of report user's inventory, and call-on back by phone is carried out to this 1000 users, as a result show wherein 675 users
Think that speech quality has problem, when call drop, call be present the problems such as voiceless sound, short message sending failure, illustrate that this programme is accurate
Rate is about 67.5%.Next all complaint work orders are investigated simultaneously, totally 110 complaints, wherein 58 people are present in this programme prediction
In potential report user's inventory, the hit rate for illustrating this programme is 52.7%.
And by calculating the network indexes such as weak covering, anomalous event, 1000 potential report users are exported, call-on back by phone
As a result, speech quality problem is reflected with the presence of 376 users, it is about 37.6% to illustrate traditional scheme predictablity rate.And
In actual 110 report users of two weeks, only 8 people are appeared in prediction inventory, illustrate that traditional scheme complains prediction effect
Far below this programme.
The data paid a return visit by existing network test and client show that potential report user's forecasting accuracy of this programme is far above
Traditional scheme.
Obviously, those skilled in the art can carry out the essence of various changes and modification without departing from the present invention to the present invention
God and scope.So, if these modifications and variations of the present invention belong to the scope of the claims in the present invention and its equivalent technology
Within, then the present invention is also intended to comprising including these changes and modification.
Claims (9)
1. a kind of potential report user's Forecasting Methodology based on signaling data, comprises the following steps：
(1) the whole network user characteristics vector is established based on A interface signaling data；The whole network user characteristics vector includes complaining
The non-report user's characteristic vector of user characteristics vector sum；Parameter in user characteristics vector is the service parameter of corresponding business；
(2) business of non-report user and report user are calculated according to report user's characteristic vector and non-report user's characteristic vector
Similarity；
(3) the potential report user in non-report user is determined according to business similarity, the higher user of business similarity is potential
The possibility of report user is bigger, when the similarity of non-report user and report user meet the decision threshold of setting, the use
Family is defined as potential report user.
A kind of 2. potential report user's Forecasting Methodology based on signaling data as claimed in claim 1, it is characterised in that：In step
Suddenly between (1) and step (2), in addition to：
The whole network user characteristics vector is standardized by (1-2).
A kind of 3. potential report user's Forecasting Methodology based on signaling data as claimed in claim 1 or 2, it is characterised in that：
In step (1), the whole network user characteristics vector is established by granularity of single business；The business includes call, short message, paging without sound
Should, location updating and switching on and shutting down.
A kind of 4. potential report user's Forecasting Methodology based on signaling data as claimed in claim 3, it is characterised in that：Selection
Apart from report user complain the time be less than setting duration, corresponding with complaint business business establish the whole network user characteristics to
Amount.
A kind of 5. potential report user's Forecasting Methodology based on signaling data as claimed in claim 2, it is characterised in that：Step
In (1-2), it is described the whole network user characteristics vector is standardized refer to the parameter values of characteristic vector being normalized into [- 1,
1]。
A kind of 6. potential report user's Forecasting Methodology based on signaling data as claimed in claim 5, it is characterised in that：By spy
Levying the formula that vectorial parameter values are standardized is：
Wherein, t represents the numerical value of some parameter of some user in the whole network user characteristics vector, and min represents institute in the whole network user
The minimum value in some parameter values is stated, max represents the maximum in some parameter values described in the whole network user；
As max=min, the parameter values that parameter is corresponded in the whole network characteristic vector are unified into assignment.
A kind of 7. potential report user's Forecasting Methodology based on signaling data as claimed in claim 1, it is characterised in that：Step
(2) in, the Euclidean distance calculating of parameter is corresponded to not with non-report user's characteristic vector by calculating report user's characteristic vector
Report user and the business similarity of report user, the smaller similarity of Euclidean distance are higher；The calculation formula of the Euclidean distance
For：
Wherein, d (x, y) represents the business similarity of non-report user and report user, and x represents the characteristic vector of report user, y
The characteristic vector of non-report user is represented, n represents the number of service parameter in the whole network user characteristics vector, xkRepresent report user
The numerical value of k-th of service parameter, y in characteristic vectorkRepresent the numerical value of k-th of service parameter in non-report user's characteristic vector.
8. a kind of potential report user's forecasting system based on signaling data, including：
Characteristic vector establishes module：For establishing the whole network user characteristics vector based on A interface signaling data；Described the whole network is used
Family characteristic vector includes report user's characteristic vector and non-report user's characteristic vector；Parameter in user characteristics vector is corresponding
The service parameter of business；
Similarity calculation module：For calculating non-report user according to report user's characteristic vector and non-report user's characteristic vector
With the business similarity of report user；
Potential report user's determining module：Potential report user in non-report user, business phase are determined according to business similarity
It is bigger for the possibility of potential report user like higher user is spent, when the similarity of non-report user and report user meet setting
Decision threshold when, the user is defined as potential report user.
9. a kind of potential report user's forecasting system based on signaling data as claimed in claim 8, it is characterised in that this is
System also includes：
Vectorial standardized module：For the whole network user characteristics vector to be standardized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410279999.8A CN104113869B (en) | 2014-06-20 | 2014-06-20 | A kind of potential report user's Forecasting Methodology and system based on signaling data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410279999.8A CN104113869B (en) | 2014-06-20 | 2014-06-20 | A kind of potential report user's Forecasting Methodology and system based on signaling data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104113869A CN104113869A (en) | 2014-10-22 |
| CN104113869B true CN104113869B (en) | 2017-12-22 |
Family
ID=51710469
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410279999.8A Active CN104113869B (en) | 2014-06-20 | 2014-06-20 | A kind of potential report user's Forecasting Methodology and system based on signaling data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104113869B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106685674B (en) | 2015-11-05 | 2020-01-10 | 华为技术有限公司 | Method and device for predicting network event and establishing network event prediction model |
| CN106910078A (en) * | 2015-12-22 | 2017-06-30 | 阿里巴巴集团控股有限公司 | Risk identification method and device |
| CN107147521B (en) * | 2017-05-10 | 2020-02-14 | 浪潮天元通信信息系统有限公司 | Early warning and monitoring method for complaint service |
| CN112101692B (en) * | 2019-06-18 | 2023-11-24 | 中国移动通信集团浙江有限公司 | Identification method and device for mobile internet bad quality users |
| CN111353792A (en) * | 2020-05-25 | 2020-06-30 | 广东电网有限责任公司惠州供电局 | Client portrait system with visual display and data analysis functions |
| CN112584407B (en) * | 2020-12-04 | 2022-07-22 | 重庆玖舆博泓科技有限公司 | LTE user complaint qualitative method and device based on space-time combination |
| CN112699099B (en) * | 2020-12-30 | 2024-06-04 | 中电科普天科技股份有限公司 | User complaint database expansion method, device and storage medium |
| CN115049092A (en) * | 2021-03-08 | 2022-09-13 | 中国移动通信集团江苏有限公司 | Broadband potential complaint user prediction method and system |
| CN115103073B (en) * | 2022-08-25 | 2022-12-13 | 深圳市星网信通科技有限公司 | Revisit task processing method and device and computer readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101414351A (en) * | 2008-11-03 | 2009-04-22 | 章毅 | Fingerprint recognition system and control method |
| CN101998472A (en) * | 2009-08-27 | 2011-03-30 | 中国移动通信集团甘肃有限公司 | Complaint pre-treatment method as well as complaint treatment method, device and system |
| CN102045357A (en) * | 2010-12-29 | 2011-05-04 | 深圳市永达电子股份有限公司 | Affine cluster analysis-based intrusion detection method |
| CN102170446A (en) * | 2011-04-29 | 2011-08-31 | 南京邮电大学 | Fishing webpage detection method based on spatial layout and visual features |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8042031B2 (en) * | 2006-12-21 | 2011-10-18 | Industrial Technology Research Institute | Maximum likelihood detection method and system |
-
2014
- 2014-06-20 CN CN201410279999.8A patent/CN104113869B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101414351A (en) * | 2008-11-03 | 2009-04-22 | 章毅 | Fingerprint recognition system and control method |
| CN101998472A (en) * | 2009-08-27 | 2011-03-30 | 中国移动通信集团甘肃有限公司 | Complaint pre-treatment method as well as complaint treatment method, device and system |
| CN102045357A (en) * | 2010-12-29 | 2011-05-04 | 深圳市永达电子股份有限公司 | Affine cluster analysis-based intrusion detection method |
| CN102170446A (en) * | 2011-04-29 | 2011-08-31 | 南京邮电大学 | Fishing webpage detection method based on spatial layout and visual features |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104113869A (en) | 2014-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104113869B (en) | A kind of potential report user's Forecasting Methodology and system based on signaling data | |
| KR101663278B1 (en) | Data processing method and device for essential factor lost score | |
| WO2020125716A1 (en) | Method for realizing network optimization and related device | |
| US8488772B2 (en) | Grouping of contact center agents | |
| Jin et al. | Nevermind, the problem is already fixed: proactively detecting and troubleshooting customer dsl problems | |
| CN105357691B (en) | LTE wireless network user perceives monitoring method and system | |
| AU2009311534B2 (en) | Two step routing procedure in a call center | |
| US10637990B1 (en) | Call center load balancing and routing management | |
| CN103796183B (en) | A method and device for identifying spam text messages | |
| AU2015213817A1 (en) | Systems, apparatuses and methods for communication flow modification | |
| WO2020134523A1 (en) | User identification method and device | |
| CN101562830A (en) | Customer perception evaluation method and system | |
| CN109548036B (en) | A mobile network potential complaint user prediction method and device | |
| CN113947260A (en) | User satisfaction prediction method and device and electronic equipment | |
| US11716422B1 (en) | Call center load balancing and routing management | |
| CN104244314B (en) | A kind of potential group customer recognition methods based on Mc interface signaling | |
| CN111917574B (en) | Social network topology model and construction method, user confidence and affinity calculation method and telecom fraud intelligent interception system | |
| CN106998256A (en) | A kind of communication failure localization method and server | |
| CN107046480A (en) | A kind of user's perception evaluating method and device | |
| CN115334560B (en) | Base station abnormality monitoring method, device, equipment and computer readable storage medium | |
| CN102149113B (en) | Mobile user perception quantification method | |
| CN108243046B (en) | A service quality assessment method and device based on data auditing | |
| CN109788501A (en) | 2G+4G mobile network's voice quality joint assessment method and device | |
| CN109963292B (en) | Complaint prediction method, complaint prediction device, electronic apparatus, and storage medium | |
| CN110070392A (en) | Customer churn method for early warning and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |