CN104168535A - 3G client development model method and 3G client development model - Google Patents
3G client development model method and 3G client development model Download PDFInfo
- Publication number
- CN104168535A CN104168535A CN201410418345.9A CN201410418345A CN104168535A CN 104168535 A CN104168535 A CN 104168535A CN 201410418345 A CN201410418345 A CN 201410418345A CN 104168535 A CN104168535 A CN 104168535A
- Authority
- CN
- China
- Prior art keywords
- rete mirabile
- user
- network users
- call
- roaming
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Mobile Radio Communication Systems (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention discloses a 3G client development model method. The method includes the ARPU value step of calculating ARPU value information of a user of a different network and then conducting correction and segmentation, the different network positioning step of obtaining address information of the user of the different network through judgment and screening of variable coefficients, the roaming probability step of collecting voice information and short message information of a network-access user of the different network, integrating data so as to conduct standard deviation processing and calculating the possibility of user roaming, the close contact person step of obtaining a different-network close contact person of a user of the home network, and the student step of predicting a student group. The invention further discloses a 3G client development model which comprises a long-roaming model module, an ARPU value model module, an address positioning model module, a roaming probability model module, a close contact person model module and a student model module. The 3G client development model method and the 3G client development model can better and more accurately serve users so that the different-network user can obtain more preferential and superior services after having access to the home network.
Description
Technical field
The present invention relates to the communications field, relate in particular to a kind of 3G visitor's open model method and 3G visitor open model.
Background technology
Along with the fast development of 3G, customer demand presents obvious new feature, communication service also manifests the trend making new advances, customer quantity explosive growth and client are to service response speed, the requirement of accuracy improves constantly, traditional business hall, customer service hot line channel can't bear the heavy load, and each operator is in the customer retaining of trying every possible means, for the amount of extending one's service, allow more users enjoy the good service of oneself, start target to concentrate one's gaze on rival's especially high-end user of user, 3G visitor opens exactly in order to comply with this needs rete mirabile growing up that arises at the historic moment and instigates rebellion within enemy camp means.
It is mutual according between different operators user that existing 3G visitor opens, and extracts rete mirabile user's Back ground Information, makes taking rete mirabile client view as basic service framework, utilizes the service product of telemarketing means and customer communication popularization oneself.A kind of improved procedure is to utilize data processing technique, with computer collection, record data, again from a large amount of, may be that rambling, elusive data, extracting and derive for some specific people is valuable, significant data, and transmission after data analysis and processing is paid to the data of selecting.By pattern analysis technology, the method can gather client's effective information, and the user who provides production capacity value serves.For the performing step of the method, data handling procedure is comparatively single, and customer basis information could not effectively combine reduction client real demand, to judge that user is worth accounting larger for artificial subjective factor in addition, in the time serving popularization, can cause the reduction of data utilance, increase operation costs.
Existing 3G visitor on-mode can not meet growing customer demand and the service becoming more meticulous, need do more detailed classification to user's consumption feature and user's type, researchs and develops new 3G visitor open model imperative.
Summary of the invention
The object of the present invention is to provide a kind of 3G visitor's open model method and 3G visitor open model, can be better, more precision serve user, make rete mirabile user can obtain how preferential and better service adding after Home Network.
The technical scheme that realizes above-mentioned purpose is:
A kind of 3G visitor open model method, comprises the following steps:
ARPU value step, crosses the information of network users by gathering rete mirabile, calculate rete mirabile user's ARPU value information, then revise according to known this network users ARPU value by logistic regression algorithm, finally ARPU value is carried out to segment processing;
Rete mirabile positioning step, crosses the call detailed list of network users and Home Network user preset time period according to rete mirabile, match the several base station informations stable with Home Network user's communication rule, obtains rete mirabile user's address information by the judgement screening of the coefficient of variation;
Roam probability step, cross voice and the short message of network users by gathering the rete mirabile of described Preset Time section, integrate the data of this time period and carry out standard deviation processing, then obtain the possibility size of user's roaming by probability calculation formula;
The people's step that is closely connected, by rete mirabile is crossed network users the moon call rule and the ticket of week call rule gather, calculate the corresponding coefficient of variation, the rete mirabile that draws accordingly this network users people that is closely connected;
Student's step, by gathering, this network users and rete mirabile are crossed the call of network users and note is single in detail, gather user's usage behavior, in conjunction with the people's information that is closely connected, dope student group.
In above-mentioned 3G visitor open model method, also comprise long strolling suddenly, this length is strolled suddenly and is comprised:
Step S61, extracting we is called rete mirabile call-information, ownership place, ticket spot and the corresponding province of regular rete mirabile number;
Step S62, processes rear segmentation to the duration of call, frequency;
Step S63, the section of checking the number is processed, excellent number of mark;
Step S64, gathers targeted customer's information.
In above-mentioned 3G visitor open model method, described ARPU value step comprises:
Step S11, extracts rete mirabile and crosses the voice of network users and the characteristic information of conversing;
Step S12, the note that extraction rete mirabile is crossed network users receives and transmission information;
Step S13, gathers rete mirabile and crosses the voice of network users, call, note reception and note transmission information, calculates ARPU value by logistic regression algorithm;
Step S14, ARPU value step S13 being calculated according to known this network users ARPU value is revised;
Step S15, is on average divided into 10 sections by revised ARPU value, and result gathers.
In above-mentioned 3G visitor open model method, described rete mirabile positioning step comprises:
Step S21, extracts Preset Time section rete mirabile call detailed list;
Step S22, locates the first district, relocates the first small towns and second small towns in this district;
Step S23, locates the second district, relocates the first small towns and second small towns in this district;
Step S24, calculates two the first small towns and two the second small towns call average and standard deviation separately, tries to achieve the coefficient of variation;
Step S25, according to the coefficient of variation, screening obtains rete mirabile user's address information.
In above-mentioned 3G visitor open model method, described roaming probability step comprises:
Step S31, gathers rete mirabile and crosses voice and the short message of network users;
Step S32, calculates mean value and the standard deviation of the long-distance roaming duration of call in described Preset Time section;
Step S33, calculates roaming probability by probability calculation formula;
Step S34, is on average divided into 10 sections by the roaming probability calculating, and end product gathers.
In above-mentioned 3G visitor open model method, described in the people's step that is closely connected comprise:
Step S41, extraction rete mirabile is crossed the phone number communication ticket of network users;
Step S42, extracts a month rule contact user, mark week situation;
Step S43, extracts all rule contact user;
Step S44, asks all rule users' the coefficient of variation;
Step S45, determines the rete mirabile of the Home Network people that is closely connected according to the coefficient of variation of trying to achieve.
In above-mentioned 3G visitor open model method, described student's step comprises:
Step S51, gathers the voice messaging of this network users;
Step S52, calculates the duration of call and number of times, and averages and standard deviation;
Step S53, gathers the short message of this network users;
Step S54, calculates note transmission times, averages;
Step S55, gathers the analysis result of voice and note, asks for probable value, and in conjunction with being closely connected, people predicts student group.
In above-mentioned 3G visitor open model method, described Preset Time section is two months.
A kind of 3G visitor open model, comprising:
Long unrestrained model module, gathers local rete mirabile and crosses the call scenarios of network users under long-distance and roaming state;
ARPU value model module, prediction rete mirabile is crossed the ARPU value of network users, by with the mapping of Home Network bill, restore user's true consumption value;
Address location model module, simulation rete mirabile is crossed ownership district and the small towns of network users;
Roaming probabilistic model module, infers the possibility size that will occur roaming without the user of roaming in future;
The human model module that is closely connected, crosses the call rule of network users and this network users according to rete mirabile, draw two class users' contact tightness degree; And
Student model module, according to receiving transmission situation feature in specific time period voice call and note, deduction rete mirabile is crossed the probability that network users is student.
The invention has the beneficial effects as follows: the present invention is in order effectively to solve the problem of rete mirabile user service, the rete mirabile user profile that has call and note to record with this network users is carried out regular, unknown user profile is reduced by related algorithm.Simultaneously, algorithm of the present invention reduces and provides user's information, from service aspect, create objective condition to the service of becoming more meticulous, under considerable resources supplIes and manpower, can make resource maximize the use, reduce human cost, and high-quality user is effectively put together and served, advantageously expand Home Network customer group.
Brief description of the drawings
Fig. 1 is the structural representation of 3G visitor open model of the present invention;
Fig. 2 is the flow chart of 3G visitor open model method of the present invention;
Fig. 3 is the flow chart of the ARPU value step of 3G visitor open model method of the present invention;
Fig. 4 is the flow chart of the rete mirabile positioning step of 3G visitor open model method of the present invention;
Fig. 5 is the flow chart of the roaming probability step of 3G visitor open model method of the present invention;
Fig. 6 is the flow chart of the people's step that is closely connected of 3G visitor open model method of the present invention;
Fig. 7 is the flow chart of student's step of 3G visitor open model method of the present invention;
Fig. 8 is that the length of 3G visitor open model method of the present invention is strolled rapid flow chart.
Embodiment
Below in conjunction with accompanying drawing, the invention will be further described.
Refer to Fig. 1,3G visitor open model of the present invention, comprise: long unrestrained model module 1 and objective open model module 2, visitor's open model module 2 comprises ARPU value model module 21, address location model module 22, roaming probabilistic model module 23, the human model module 24 that is closely connected and student model module 25, wherein:
Long unrestrained model module 1 is crossed the call scenarios of network users under long-distance and roaming state for gathering local rete mirabile, user within the scope of certain call relatively beats roaming demand, therefore the advantage of this model is that can be good at having reflected long-distance roaming demand obtains user, concerning rete mirabile 2G user, rate are unified in our long-distance roaming of 3G single deck tape-recorder very large attraction.
ARPU value model module 21 is crossed the ARPU value of network users for predicting rete mirabile, by with the mapping of Home Network bill, restore user's true consumption value, embody user's value.Under certain level of consumption, can carry out targetedly the marketing of 3G single deck tape-recorder different product, this model plays effect indispensable and that replace in whole model system.
Address location model module 22 is crossed ownership district and the small towns of network users for simulating rete mirabile, can draw a circle to approve out the user group within the scope of certain dispensing on the one hand, can market concentrated area on the other hand, can save human cost, acts on huge.
Roaming probabilistic model module 23 is for inferring the possibility size that will occur roaming without the user of roaming in future, complement each other with ARPU model, be combined with and can obtain the impact of prediction roaming on customer consumption value, can restore the applicable Home Network 3G single deck tape-recorder product of rete mirabile user;
The human model module 24 that is closely connected is crossed the call rule of network users and this network users according to rete mirabile, draw two class users' contact tightness degree, considers the impact of population effect, predicts the rete mirabile colony that this class and this network users be closely connected and is easier to instigate rebellion within enemy camp.The in the situation that of resource abundance, this crowd of user plays good regulating action.
Student model module 25 foundations are in the specific time period, voice call and note receive transmission situation feature, infer the probability that rete mirabile user is student, stay close in addition close contact person and combine, under the prerequisite of known Home Network User, can orient more accurately rete mirabile User.On the one hand, because User consumption is on the low side, loyalty is higher, and User information faces is wider on the other hand, causes this class user to instigate rebellion within enemy camp difficulty larger, therefore needs on stream to shield, and avoids the waste of human resources.
Refer to Fig. 2,3G visitor open model method of the present invention, comprises the following steps:
Step S1, ARPU value step, crosses the information of network users by gathering rete mirabile, calculate rete mirabile user's ARPU value information by logistic regression algorithm, then revise according to known this network users ARPU value, finally ARPU value is carried out to segment processing, refer to Fig. 3, specifically comprise:
Step S11, extracts rete mirabile and crosses the voice of network users and the characteristic information of conversing;
Step S12, the note that extraction rete mirabile is crossed network users receives and transmission information;
Step S13, gathers rete mirabile and crosses the voice of network users, call, note reception and note transmission information, calculates ARPU value by logistic regression algorithm; Wherein, logistic regression algorithm is as follows:
Can inquire Home Network customer consumption data, extract and wherein have mutual this month voice SMS data with rete mirabile user, it is gathered according to rete mirabile Subscriber Number, when combined data, classified information under voice SMS is gathered respectively, using each user and comprise every summary information (x) as a sample, obtain a sample set A.Can set up multiple linear regression model:
ARPU＝b0+b1*x1+b2*x2+…+bn*xn
Obtain the actual ARPU value of rete mirabile user by survey, in substitution model, obtain b1, b2 ..., bn}, will obtain estimating the multiple linear regression model of ARPU value in parameter substitution model.
Step S14, ARPU value step S13 being calculated according to known this network users ARPU value is revised, that is: use by the actual ARPU value of this network users and compare with prediction ARPU value, obtain standard error estimate, according to error amount correction ARPU value estimation result.
Step S15, is on average divided into 10 sections by revised ARPU value, and result gathers.
Step S2, rete mirabile positioning step, crosses the call detailed list of network users and Home Network user preset time period according to rete mirabile, match the several base station informations stable with Home Network user's communication rule, the address information that obtains rete mirabile user by the judgement screening of the coefficient of variation, refers to Fig. 4, specifically comprises:
Step S21, the rete mirabile call detailed list of extraction Preset Time section (being two months in the present embodiment), whether this rete mirabile call detailed list comprises opposite end number, one's own side's ticket spot, calling and called, is long-distance roaming call.
Step S22, locates the first district, relocates the first small towns and second small towns in this district;
Step S23, locates the second district, relocates the first small towns and second small towns in this district;
Step S24, calculate two the first small towns and two the second small towns call average and standard deviation separately, that is: by dialing call detailed list, obtain the base station information of each call, count two base stations that occurrence number is maximum, calculate average and the standard deviation of voice duration in same base station, thereby try to achieve the coefficient of variation, that is: obtain the coefficient of variation with token sound is poor divided by speech mean.
Step S25, locates last small towns information according to the coefficient of variation, i.e. screening obtains rete mirabile user's address information, using the base station of coefficient of variation minimum as the base station of orienting, mates the address information that obtains user with base station information that is:.
Step S3, roaming probability step, cross voice and the short message of network users by gathering the rete mirabile of Preset Time section (in the present embodiment two months), integrate the data of this time period and carry out standard deviation processing, obtain again the possibility size of user's roaming by probability calculation formula, refer to Fig. 5, specifically comprise:
Step S31, gathers rete mirabile and crosses voice and the short message of network users;
Step S32, calculates long-distance roaming voice duration mean value and standard deviation in Preset Time section (in the present embodiment two months);
Step S33, calculates roaming probability by probability calculation formula, as follows:
Extract user's the various information such as voice and note, obtain a z value:
Z=w0+w1*x1+w2*x2+...+wm*xm (wherein x1, x2 ..., xm is user's various information, dimension is m);
Obtain roaming probability: p=1/ (1+exp (z)) according to the form of sigmoid function afterwards.
Step S34, is on average divided into 10 sections by the roaming probability calculating, and end product gathers.
Step S4, the people's step that is closely connected, by rete mirabile being crossed to the moon call rule of network users and the ticket of week call rule gathers, calculates the corresponding coefficient of variation, and the rete mirabile that draws accordingly this network users people that is closely connected, refers to Fig. 6, specifically comprises:
Step S41, extraction rete mirabile is crossed the phone number communication ticket of network users;
Step S42 extracts a month rule contact user from phone number communication ticket, mark week situation so that find same contact person's call rule;
Step S43 extracts all rule contact user from phone number communication ticket;
Step S44, asks all rule users' the coefficient of variation, that is: statistics have call weekly all several be rule day, try to achieve in a few days standard deviation and the mean value of talk times of rule, the coefficient of variation is that standard deviation is divided by mean value;
Step S45, determines the rete mirabile of the Home Network people that is closely connected according to the coefficient of variation of trying to achieve, and obtains All Contacts's the coefficient of variation that is:, and the coefficient of variation is less, contacts tightr.
Step S5, student's step, by gathering, this network users and rete mirabile are crossed the call of network users and note is single in detail, gather user's usage behavior, in conjunction with the people's information that is closely connected, dope student group, refer to Fig. 7, specifically comprise:
Step S51, gathers the voice messaging of this network users;
Step S52, calculates the duration of call and number of times, and averages and standard deviation;
Step S53, gathers the short message of this network users;
Step S54, calculates note transmission times, averages;
Step S55, gathers the analysis result of voice and note, asks for probable value by known student's message registration, analyzes the rule of student's call, comprises air time point and the duration of call, and user's message registration is compared with it, obtains probable value, and concrete grammar is:
Extract user's the various information such as voice and note, obtain a z value:
Z=w0+w1*x1+w2*x2+...+wm*xm (wherein x1, x2 ..., xm is user's various information, dimension is m);
Obtain the probability into student: p=1/ (1+exp (z)) according to the form of sigmoid function afterwards, in conjunction with being closely connected, people predicts student group, that is: proposition is defined as student's user, these users' the people that is closely connected is also defined as to student, predicts thus student group.
Above five steps neutralizes on the one hand can carry out reasonably screening recommendation on the basis of the customer consumption reducing and roaming possibility, can, according to existing adjusting condition, filter out high-quality resource on the other hand, saves existing cost.
Step S6, refers to Fig. 8, and long strolling suddenly comprises:
Step S61, extracting we is called rete mirabile call-information, ownership place, ticket spot and the corresponding province of regular rete mirabile number;
Step S62, processes rear segmentation to the duration of call, frequency;
Step S63, the section of checking the number is processed, excellent number of mark;
Step S64, gathers targeted customer's information.
Length is strolled the data that suddenly draw and can be used separately.
Above embodiment is used for illustrative purposes only, but not limitation of the present invention, person skilled in the relevant technique, without departing from the spirit and scope of the present invention, can also make various conversion or modification, therefore all technical schemes that are equal to also should belong to category of the present invention, should be limited by each claim.
Claims (9)
1. a 3G visitor open model method, is characterized in that, comprises the following steps:
ARPU value step, crosses the information of network users by gathering rete mirabile, calculate rete mirabile user's ARPU value information, then revise according to known this network users ARPU value by logistic regression algorithm, finally ARPU value is carried out to segment processing;
Rete mirabile positioning step, crosses the call detailed list of network users and Home Network user preset time period according to rete mirabile, match the several base station informations stable with Home Network user's communication rule, obtains rete mirabile user's address information by the judgement screening of the coefficient of variation;
Roam probability step, cross voice and the short message of network users by gathering the rete mirabile of described Preset Time section, integrate the data of this time period and carry out standard deviation processing, then obtain the possibility size of user's roaming by probability calculation formula;
The people's step that is closely connected, by rete mirabile is crossed network users the moon call rule and the ticket of week call rule gather, calculate the corresponding coefficient of variation, the rete mirabile that draws accordingly this network users people that is closely connected;
Student's step, by gathering, this network users and rete mirabile are crossed the call of network users and note is single in detail, gather user's usage behavior, in conjunction with the people's information that is closely connected, dope student group.
2. 3G visitor open model method according to claim 1, is characterized in that, also comprise long strolling suddenly, this length is strolled suddenly and comprised:
Step S61, extracting we is called rete mirabile call-information, ownership place, ticket spot and the corresponding province of regular rete mirabile number;
Step S62, processes rear segmentation to the duration of call, frequency;
Step S63, the section of checking the number is processed, excellent number of mark;
Step S64, gathers targeted customer's information.
3. 3G visitor open model method according to claim 1, is characterized in that, described ARPU value step comprises:
Step S11, extracts rete mirabile and crosses the voice of network users and the characteristic information of conversing;
Step S12, the note that extraction rete mirabile is crossed network users receives and transmission information;
Step S13, gathers rete mirabile and crosses the voice of network users, call, note reception and note transmission information, calculates ARPU value by logistic regression algorithm;
Step S14, ARPU value step S13 being calculated according to known this network users ARPU value is revised;
Step S15, is on average divided into 10 sections by revised ARPU value, and result gathers.
4. 3G visitor open model method according to claim 1, is characterized in that, described rete mirabile positioning step comprises:
Step S21, extracts Preset Time section rete mirabile call detailed list;
Step S22, locates the first district, relocates the first small towns and second small towns in this district;
Step S23, locates the second district, relocates the first small towns and second small towns in this district;
Step S24, calculates two the first small towns and two the second small towns call average and standard deviation separately, tries to achieve the coefficient of variation;
Step S25, according to the coefficient of variation, screening obtains rete mirabile user's address information.
5. 3G visitor open model method according to claim 1, is characterized in that, described roaming probability step comprises:
Step S31, gathers rete mirabile and crosses voice and the short message of network users;
Step S32, calculates mean value and the standard deviation of the long-distance roaming duration of call in described Preset Time section;
Step S33, calculates roaming probability by probability calculation formula;
Step S34, is on average divided into 10 sections by the roaming probability calculating, and end product gathers.
6. 3G according to claim 1 visitor open model method, is characterized in that, described in the people's step that is closely connected comprise:
Step S41, extraction rete mirabile is crossed the phone number communication ticket of network users;
Step S42, extracts a month rule contact user, mark week situation;
Step S43, extracts all rule contact user;
Step S44, asks all rule users' the coefficient of variation;
Step S45, determines the rete mirabile of the Home Network people that is closely connected according to the coefficient of variation of trying to achieve.
7. 3G visitor open model method according to claim 1, is characterized in that, described student's step comprises:
Step S51, gathers the voice messaging of this network users;
Step S52, calculates the duration of call and number of times, and averages and standard deviation;
Step S53, gathers the short message of this network users;
Step S54, calculates note transmission times, averages;
Step S55, gathers the analysis result of voice and note, asks for probable value, and in conjunction with being closely connected, people predicts student group.
8. 3G visitor open model method according to claim 1, is characterized in that, described Preset Time section is two months.
9. a 3G visitor open model, is characterized in that, comprising:
Long unrestrained model module, gathers local rete mirabile and crosses the call scenarios of network users under long-distance and roaming state;
ARPU value model module, prediction rete mirabile is crossed the ARPU value of network users, by with the mapping of Home Network bill, restore user's true consumption value;
Address location model module, simulation rete mirabile is crossed ownership district and the small towns of network users;
Roaming probabilistic model module, infers the possibility size that will occur roaming without the user of roaming in future;
The human model module that is closely connected, crosses the call rule of network users and this network users according to rete mirabile, draw two class users' contact tightness degree; And
Student model module, according to receiving transmission situation feature in specific time period voice call and note, deduction rete mirabile is crossed the probability that network users is student.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410418345.9A CN104168535B (en) | 2014-08-22 | 2014-08-22 | The method for establishing 3G visitor's open model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410418345.9A CN104168535B (en) | 2014-08-22 | 2014-08-22 | The method for establishing 3G visitor's open model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104168535A true CN104168535A (en) | 2014-11-26 |
| CN104168535B CN104168535B (en) | 2018-01-19 |
Family
ID=51912131
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410418345.9A Active CN104168535B (en) | 2014-08-22 | 2014-08-22 | The method for establishing 3G visitor's open model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104168535B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105160547A (en) * | 2015-08-07 | 2015-12-16 | 北京思特奇信息技术股份有限公司 | Cross-network identification method and system based on stable contact index |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070112614A1 (en) * | 2005-11-11 | 2007-05-17 | Matteo Maga | Identifying target customers for campaigns to increase average revenue per user |
| CN102215504A (en) * | 2010-04-08 | 2011-10-12 | 中国移动通信集团甘肃有限公司 | Method and system for identifying class of newly network-accessed user |
| CN103607691A (en) * | 2013-11-26 | 2014-02-26 | 中国联合网络通信集团有限公司 | Flow package recommendation method and device |
-
2014
- 2014-08-22 CN CN201410418345.9A patent/CN104168535B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070112614A1 (en) * | 2005-11-11 | 2007-05-17 | Matteo Maga | Identifying target customers for campaigns to increase average revenue per user |
| CN102215504A (en) * | 2010-04-08 | 2011-10-12 | 中国移动通信集团甘肃有限公司 | Method and system for identifying class of newly network-accessed user |
| CN103607691A (en) * | 2013-11-26 | 2014-02-26 | 中国联合网络通信集团有限公司 | Flow package recommendation method and device |
Non-Patent Citations (2)
| Title |
|---|
| 何帆: "移动通信企业精细化营销客户细分研究", 《万方在线数据库》 * |
| 张俐: "浅谈ARPU值的细化与分析", 《财经界》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105160547A (en) * | 2015-08-07 | 2015-12-16 | 北京思特奇信息技术股份有限公司 | Cross-network identification method and system based on stable contact index |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104168535B (en) | 2018-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109978608A (en) | The marketing label analysis extracting method and system of target user's portrait | |
| Karacuka et al. | Competition in Turkish mobile telecommunications markets: Price elasticities and network substitution | |
| US8346208B2 (en) | Method and system for identifying mobile user cluster by call detail record | |
| US8537991B2 (en) | Toll-free reverse least cost routing price management system and method | |
| CN101021927A (en) | Unified marketing supporting system based on analysis of user behaviour and habit and method thereof | |
| CN102541886A (en) | System and method for identifying relationship among user group and users | |
| US20120185540A1 (en) | Method and Arrangement for Supporting Analysis of Social Networks in a Communication Network | |
| US8732295B2 (en) | Method and apparatus for providing real friends count | |
| CN105869035A (en) | Mobile user credit evaluation method and apparatus | |
| CN102215504A (en) | Method and system for identifying class of newly network-accessed user | |
| Grzybowski et al. | Estimating demand for fixed-mobile bundles and switching costs between tariffs | |
| CN101883133A (en) | Accurate influence marketing system based on signalling analysis and method thereof | |
| CN101321070B (en) | Monitoring system and method for suspicious user | |
| TW201233099A (en) | Method and system for carrying out predictive analysis relating to nodes of a communication network | |
| CN105824818A (en) | Informationized management method, platform and system | |
| Rohman et al. | An assessment of mobile broadband access in Indonesia: a demand or supply problem? | |
| CN104168535A (en) | 3G client development model method and 3G client development model | |
| Kwak et al. | Estimating demand curve in the Korean VoIP telecommunications market | |
| Pattanavichai et al. | A pricing model and sensitivity analysis for MNO's investment decision making in 3G UMTS networks | |
| CN107154875B (en) | Method for node sensitivity sequencing in telephone communication network | |
| CN105160547A (en) | Cross-network identification method and system based on stable contact index | |
| CN112307075B (en) | User relationship identification method and device | |
| Inoue et al. | Mobile-carrier choice behavior analysis between three major mobile-carriers and mobile virtual network operators | |
| CN107819924A (en) | A kind of recognition methods of spam phone number, device and equipment | |
| Stork | Are mobile phones replacing the use of public phones in Africa? |
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 |