WO2018203843A2 - A system for making subscriber classification - Google Patents
A system for making subscriber classification Download PDFInfo
- Publication number
- WO2018203843A2 WO2018203843A2 PCT/TR2017/000138 TR2017000138W WO2018203843A2 WO 2018203843 A2 WO2018203843 A2 WO 2018203843A2 TR 2017000138 W TR2017000138 W TR 2017000138W WO 2018203843 A2 WO2018203843 A2 WO 2018203843A2
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- unit
- estimation
- subscribers
- subscriber
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3438—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment monitoring of user actions
Definitions
- the present invention relates to a system for determining which subscriber class the said subscribers belong to by using the data about mobile network, fixed network and value-added services usages of mobile network operator subscribers.
- preventing realization of subscriber classification in a correct way is another case that is common among mobile network operator subscribers.
- This case is such that a subscriber does not always use a line that is registered to his/her name. Some subscribers are only the user of the lines being used at that moment and real line holders are completely different persons (for example, another family member). Because subscribers using the lines that are registered to another person's name cause acquisition of particularly subscriber information in a wrong way, it is not always possible to include these subscribers into a correct subscriber class.
- Including mobile network operator subscribers into a correct subscriber class is a determination of vital importance in terms of mobile network operators in order to know the subscriber better, adjust services to be provided and recommended to the subscriber accordingly.
- the Chinese patent document no. CN103428370 discloses a method for determining whether a mobile phone has multiple users or not. In the method, it is enabled to detect whether the user is line holder or not by comparing the user information wherein the mobile device is registered with the SIM card information being used in the mobile device.
- US2010293165 discloses a subscriber identification system. In the said system, it is enabled to detect and identify a certain subscriber within a group of subscribers by examining subscriber usage data.
- An objective of the present invention is to realize a system for determining which subscriber class the said subscribers belong to by using the data about mobile network, fixed network and value-added services usages of mobile network operator subscribers.
- Figure 1 is a schematic view of the inventive system.
- the inventive system for making subscriber classification (1) comprises:
- At least one data collection unit (2) which collects data about subscribers' usages of mobile network, fixed network and value-added services from different data sources (V);
- At least one data interpretation unit (3) whereto the data collected by the data collection unit (2) are transferred and which enriches the data by carrying out summarization and interpretation transactions on these data and generates meaningful variables for each subscriber; at least one data learning unit (4) which carries out learning transaction by means of machine learning methods on the data of the type determined by the data interpretation unit (3) and generates an estimation method; at least one estimation unit (5) which realizes an estimation about in which subscriber class each subscriber should take part by running the model generated by the data learning unit (4) for the subscribers and can transfer the estimation results to various terminal systems (U); - at least one database (51) which keeps the data about the estimations managed by the estimation unit (5) and realized by the estimation unit (5). ( Figure 1)
- the data collection unit (2) is a unit which collects data about subscribers' usages of mobile network, fixed network and value-added services from different data sources (V) and carries out summarization and interpretation transactions on these data.
- the data collected by the data collection unit (2) from data sources (V) are history of network performance indicators; location history of subscribers that emerge with their signalling and call detail information occurring by their mobile network usages; data about their mobile data usages; data of TV and entertainment usage and purchase that is received from fixed and mobile network system; history of music usage; data about data usages performed over fixed networks; information of the mobile device that is used by the subscriber; data about club, line ownership or usership, membership information of the subscribers on the mobile network operator.
- the data interpretation unit (3) is a unit whereto the data collected by the data collection unit (2) are transferred and which enriches the data by carrying out summarization and interpretation transactions on these data and generates meaningful variables for each subscriber.
- the data interpretation unit (3) determines POI (Point-of-Interest) information (such as university, ' hospital, plaza, AVM, concert, stadium) over the location history collected by the data collection unit (2) and enriches these location information. Similarly, the data interpretation unit (3) analyses the data about the mobile data usage collected by the data collection unit (2) and the data about the data usages realized over the fixed networks and determines over which applications or sites these data usages are realized by exhibiting fields of these applications and sites (such as game, social media, health, music). Thus, the said data usage data are enriched.
- POI Point-of-Interest
- the data learning unit (4) is a unit which carries out learning transaction by means of machine learning methods on the data whereon the data interpretation unit (3) carries out summarization and interpretation transactions and the variables generated by it and creates an estimation model.
- the data learning unit (4) carries out learning transaction by means of regression method which is one of machine learning methods.
- the data learning unit (4) is a unit which periodically realizes learning and model development applications in preferred embodiment of the invention. Thus, it is ensured that the estimation model is updated together with new data occurred in time and new variables generated.
- the estimation unit (5) is a unit which realizes an estimation about in which subscriber class each subscriber should take part by running the model generated by the data learning unit (4) for the subscribers and can transfer the estimation results to various terminal systems (U) such as marketing, planning systems.
- the estimation unit (5) can runs the said model for all subscribers in one embodiment of the invention whereas it can run the model for a certain group of subscribers as well.
- the said subscriber groups in different embodiments of the invention, can be subscriber groups such as subscribers changing tariff/bill, new subscribers, subscribers realizing new service purchase, subscribers experiencing/realizing change of classification/club/membership, subscribers changing device, subscribers whose data on the HLR (Home Location Register) changes.
- the estimation unit (5) is also a unit which has the quality to realize the estimation transaction again for all subscribers or one or several subscriber group in the event that the estimation model changes/is updated as a result of the transactions realized by the data learning unit (4).
- the estimation unit (5) is a unit which has the quality to test the model created by the data learning unit (4) by using a control data and to determine whether the model has exceeded the correct estimation threshold value or not.
- classes that can be determined for subscribers as a result of estimation realized by the estimation unit (5) can consists of various subscriber classes such as "High Income Group Subscribers”, “White Collar Employee”, “University Student”, “The Teen Under the Age of 26" in different embodiments of the invention.
- the database (51) is a unit which keeps the data about the estimations managed by the estimation unit (5) and realized by the estimation unit (5).
- the data collection unit (2) collects data about subscribers' usages of mobile network, fixed network and value-added services from various data sources (V).
- the data interpretation unit (3) enriches suitable data by carrying out summarization, interpretation transactions on these data received by the data collection unit (2).
- the data learning unit (4) carries out learning transaction by means of machine learning methods and it creates an estimation model for determining into which subscriber class a subscriber should be included, on the data wherein the data interpretation unit (3) carries out summarization and interpretation transaction and the variables generated by it, based on the information about which subscriber class does the subscriber belong to for the cases in these said data.
- this model is a model wherein environmental factors are also taken into account together with information of calling/being called and information of social network analysis.
- This model is tested by the estimation unit (5) at first and if it is detected to have exceeded a certain estimation threshold value, it is started to be used for all subscribers or a selected subscriber group by the estimation unit (5) for the purpose of determining into which subscriber class subscribers should be included according to the usage data and thus it is ensured that subscribers are placed into a subscriber class suitable for their usage data.
- the estimation unit (5) is a unit which has the quality to transfer estimation results to different terminal systems (U) and also to write them to a database (51).
Abstract
The present invention relates to a system (1) for determining which subscriber class the said subscribers belong to by using the data about mobile network, fixed network and value-added services usages of mobile network operator subscribers. The data collection unit (2), the data interpretation unit (3), the learning unit (4), the estimation unit (5) and the database (51) are included within the inventive system (1).
Description
A SYSTEM FOR MAKING SUBSCRIBER CLASSIFICATION Technical Field
The present invention relates to a system for determining which subscriber class the said subscribers belong to by using the data about mobile network, fixed network and value-added services usages of mobile network operator subscribers.
Background of the Invention
Today, mobile network operators can evaluate their subscribers under various classifications in order to provide customized services to the said subscribers and make offers suited for subscriber characteristics. Information such as age, income group, profession, place of residence, gender should be known so that a subscriber can be included into a correct subscriber class. Although mobile network operators can sometimes access subscribers in order to obtain these information, it is not always possible to obtain these information completely and correctly.
In addition, preventing realization of subscriber classification in a correct way is another case that is common among mobile network operator subscribers. This case is such that a subscriber does not always use a line that is registered to his/her name. Some subscribers are only the user of the lines being used at that moment and real line holders are completely different persons (for example, another family member). Because subscribers using the lines that are registered to another person's name cause acquisition of particularly subscriber information in a wrong way, it is not always possible to include these subscribers into a correct subscriber class.
Including mobile network operator subscribers into a correct subscriber class is a determination of vital importance in terms of mobile network operators in order to know the subscriber better, adjust services to be provided and recommended to the subscriber accordingly. When laborious and inadequate methods for determining the information to make this classification is taken into account in the state of the art, together with importance and criticality of this determination, it is understood that a more effective solution wherein classification of a mobile subscriber is made by taking usage data into consideration is necessary in the state of the art.
The Chinese patent document no. CN103428370, an application in the state of the art, discloses a method for determining whether a mobile phone has multiple users or not. In the method, it is enabled to detect whether the user is line holder or not by comparing the user information wherein the mobile device is registered with the SIM card information being used in the mobile device.
The United States patent document no. US2010293165, another application in the state of the art, discloses a subscriber identification system. In the said system, it is enabled to detect and identify a certain subscriber within a group of subscribers by examining subscriber usage data.
Summary of the Invention
An objective of the present invention is to realize a system for determining which subscriber class the said subscribers belong to by using the data about mobile network, fixed network and value-added services usages of mobile network operator subscribers.
Detailed Description of the Invention
The "System for Making Subscriber Classification" realized to fulfil the objective of the present invention is shown in the figure attached, in which:
Figure 1 is a schematic view of the inventive system.
The components illustrated in the figure are individually numbered, where the numbers refer to the following:
System
Data collection unit
Data interpretation unit
Learning unit
Estimation unit
51. Database
V. Data source
U. Terminal systems
The inventive system for making subscriber classification (1) comprises:
at least one data collection unit (2) which collects data about subscribers' usages of mobile network, fixed network and value-added services from different data sources (V);
at least one data interpretation unit (3) whereto the data collected by the data collection unit (2) are transferred and which enriches the data by carrying out summarization and interpretation transactions on these data and generates meaningful variables for each subscriber; at least one data learning unit (4) which carries out learning transaction by means of machine learning methods on the data of the type determined by the data interpretation unit (3) and generates an estimation method;
at least one estimation unit (5) which realizes an estimation about in which subscriber class each subscriber should take part by running the model generated by the data learning unit (4) for the subscribers and can transfer the estimation results to various terminal systems (U); - at least one database (51) which keeps the data about the estimations managed by the estimation unit (5) and realized by the estimation unit (5). (Figure 1)
The data collection unit (2) is a unit which collects data about subscribers' usages of mobile network, fixed network and value-added services from different data sources (V) and carries out summarization and interpretation transactions on these data.
In preferred embodiment of the invention, the data collected by the data collection unit (2) from data sources (V) are history of network performance indicators; location history of subscribers that emerge with their signalling and call detail information occurring by their mobile network usages; data about their mobile data usages; data of TV and entertainment usage and purchase that is received from fixed and mobile network system; history of music usage; data about data usages performed over fixed networks; information of the mobile device that is used by the subscriber; data about club, line ownership or usership, membership information of the subscribers on the mobile network operator.
The data interpretation unit (3) is a unit whereto the data collected by the data collection unit (2) are transferred and which enriches the data by carrying out summarization and interpretation transactions on these data and generates meaningful variables for each subscriber.
In one embodiment of the invention, the data interpretation unit (3) determines POI (Point-of-Interest) information (such as university, ' hospital, plaza, AVM, concert, stadium) over the location history collected by the data collection unit (2)
and enriches these location information. Similarly, the data interpretation unit (3) analyses the data about the mobile data usage collected by the data collection unit (2) and the data about the data usages realized over the fixed networks and determines over which applications or sites these data usages are realized by exhibiting fields of these applications and sites (such as game, social media, health, music). Thus, the said data usage data are enriched.
The data learning unit (4) is a unit which carries out learning transaction by means of machine learning methods on the data whereon the data interpretation unit (3) carries out summarization and interpretation transactions and the variables generated by it and creates an estimation model.
In preferred embodiment of the invention, the data learning unit (4) carries out learning transaction by means of regression method which is one of machine learning methods.
The data learning unit (4) is a unit which periodically realizes learning and model development applications in preferred embodiment of the invention. Thus, it is ensured that the estimation model is updated together with new data occurred in time and new variables generated.
The estimation unit (5) is a unit which realizes an estimation about in which subscriber class each subscriber should take part by running the model generated by the data learning unit (4) for the subscribers and can transfer the estimation results to various terminal systems (U) such as marketing, planning systems.
The estimation unit (5) can runs the said model for all subscribers in one embodiment of the invention whereas it can run the model for a certain group of subscribers as well. The said subscriber groups, in different embodiments of the invention, can be subscriber groups such as subscribers changing tariff/bill, new subscribers, subscribers realizing new service purchase, subscribers
experiencing/realizing change of classification/club/membership, subscribers changing device, subscribers whose data on the HLR (Home Location Register) changes. The estimation unit (5) is also a unit which has the quality to realize the estimation transaction again for all subscribers or one or several subscriber group in the event that the estimation model changes/is updated as a result of the transactions realized by the data learning unit (4). In one embodiment of the invention, the estimation unit (5) is a unit which has the quality to test the model created by the data learning unit (4) by using a control data and to determine whether the model has exceeded the correct estimation threshold value or not. As can be understood by a skilled person in the art, classes that can be determined for subscribers as a result of estimation realized by the estimation unit (5) can consists of various subscriber classes such as "High Income Group Subscribers", "White Collar Employee", "University Student", "The Youth Under the Age of 26" in different embodiments of the invention.
The database (51) is a unit which keeps the data about the estimations managed by the estimation unit (5) and realized by the estimation unit (5).
With the inventive system (1), transaction of determining which subscriber class the said subscribers belong to by using the data about mobile network, fixed network and value-added services usages of mobile network operator subscribers is carried out. When the said transaction is carried out, firstly the data collection unit (2) collects data about subscribers' usages of mobile network, fixed network and value-added services from various data sources (V). Then, the data interpretation unit (3) enriches suitable data by carrying out summarization, interpretation transactions on these data received by the data collection unit (2).
Thus, a large number of variables whereon analysis can be made are created specific to each subscriber. The data learning unit (4) carries out learning transaction by means of machine learning methods and it creates an estimation model for determining into which subscriber class a subscriber should be included, on the data wherein the data interpretation unit (3) carries out summarization and interpretation transaction and the variables generated by it, based on the information about which subscriber class does the subscriber belong to for the cases in these said data. In one embodiment of the invention, this model is a model wherein environmental factors are also taken into account together with information of calling/being called and information of social network analysis. This model is tested by the estimation unit (5) at first and if it is detected to have exceeded a certain estimation threshold value, it is started to be used for all subscribers or a selected subscriber group by the estimation unit (5) for the purpose of determining into which subscriber class subscribers should be included according to the usage data and thus it is ensured that subscribers are placed into a subscriber class suitable for their usage data. The estimation unit (5) is a unit which has the quality to transfer estimation results to different terminal systems (U) and also to write them to a database (51). Within these basic concepts; it is possible to develop various embodiments of the inventive system (1), the invention cannot be limited to examples disclosed herein and it is essentially according to claims.
Claims
A system (1) for making subscriber classification; comprising:
at least one data collection unit (2) which collects data about subscribers' usages of mobile network, fixed network and value-added services from different data sources (V);
at least one data interpretation unit (3);
at least one data learning unit (4);
at least one estimation unit (5) which realizes an estimation about in which subscriber class each subscriber should take part by running the model generated by the data learning unit (4) for the subscribers and can transfer the estimation results to various terminal systems (U); at least one database (51) which keeps the data about the estimations managed by the estimation unit (5) and realized by the estimation unit (5);
and characterized by
at least one data interpretation unit (3) whereto the data collected by the data collection unit (2) are transferred and which enriches the data by carrying out summarization and interpretation transactions on these data and generates meaningful variables for each subscriber;
at least one data learning unit (4) which carries out learning transaction by means of machine learning methods on the data of the type determined by the data interpretation unit (3) and generates an estimation method.
A system (1) according to Claim 1 ; characterized by the data collection unit
(2) which collects history of network performance indicators; location history of subscribers that emerge with their signalling and call detail information occurring by their mobile network usages; data about their mobile data usages; data of TV and entertainment usage and purchase that is received from fixed and mobile network system; history of music usage; data about data usages
performed over fixed networks; information of the mobile device that is used by the subscriber; data about club, line ownership or usership, membership information of the subscribers on the mobile network operator from data sources (V).
3. A system (1) according to Claim 1 ; characterized by the data interpretation unit (3) which determines POI (Point-of-Interest) information (such as university, hospital, plaza, AVM, concert, stadium) over the location history collected by the data collection unit (2) and enriches these location information.
4. A system (1) according to Claim 1 ; characterized by the data interpretation unit (3) which analyses the data about the mobile data usage collected by the data collection unit (2) and the data about the data usages realized over the fixed networks and determines over which applications or sites these data usages are realized by exhibiting fields of these applications and sites (such as game, social media, health, music).
5. A system (1) according to Claim 1; characterized by the data learning unit (4) which carries out learning transaction by means of regression method -that is one of machine learning methods.
6. A system (1) according to Claim 1; characterized by the data learning unit (4) which periodically realizes learning and model development applications.
7. A system (1) according to Claim 1 ; characterized by the estimation unit (5) which has the quality to run the estimation model for all subscribers or a a certain group of subscribers.
8. A system (1) according to Claim 1 ; characterized by the estimation unit (5) which has the quality to realize the estimation transaction again for all
subscribers or one or several subscriber group in the event that the estimation model changes/is updated as a result of the transactions realized by the data learning unit (4).
9. A system (1) according to Claim 1 ; characterized by the estimation unit (5) which has the quality to test the model created by the data learning unit (4) by using a control data and to determine whether the model has exceeded the correct estimation threshold value or not.
10. A system (1) according to Claim 1; characterized by the data learning unit (4) which creates an estimation model wherein environmental factors are also taken into account together with information of calling/being called and information of social network analysis.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TR2016/18562A TR201618562A3 (en) | 2016-12-14 | 2016-12-14 | A SYSTEM THAT ENABLES SUBSCRIBER CLASSIFICATION |
| TR2016/18562 | 2016-12-14 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| WO2018203843A2 true WO2018203843A2 (en) | 2018-11-08 |
| WO2018203843A3 WO2018203843A3 (en) | 2019-01-24 |
Family
ID=64016977
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/TR2017/000138 WO2018203843A2 (en) | 2016-12-14 | 2017-12-13 | A system for making subscriber classification |
Country Status (2)
| Country | Link |
|---|---|
| TR (1) | TR201618562A3 (en) |
| WO (1) | WO2018203843A2 (en) |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120226554A1 (en) * | 2011-03-01 | 2012-09-06 | Jeffrey C Schmidt | System and method for providing data to a portable communications device based on real-time subscriber behavior |
| US8583471B1 (en) * | 2011-06-13 | 2013-11-12 | Facebook, Inc. | Inferring household income for users of a social networking system |
| IL232316A (en) * | 2014-04-28 | 2017-04-30 | Verint Systems Ltd | System and method for demographic profiling of mobile terminal users based on network-centric estimation of installed mobile applications and their usage patterns |
-
2016
- 2016-12-14 TR TR2016/18562A patent/TR201618562A3/en unknown
-
2017
- 2017-12-13 WO PCT/TR2017/000138 patent/WO2018203843A2/en active Application Filing
Also Published As
| Publication number | Publication date |
|---|---|
| TR201618562A3 (en) | 2018-10-22 |
| TR201618562A2 (en) | 2018-07-23 |
| WO2018203843A3 (en) | 2019-01-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210319375A1 (en) | Churn prediction in a broadband network | |
| Argon et al. | Priority assignment under imperfect information on customer type identities | |
| US20150371163A1 (en) | Churn prediction in a broadband network | |
| CN107885742B (en) | Service recommendation method and device | |
| Khan et al. | One label, one billion faces: Usage and consistency of racial categories in computer vision | |
| CN106354797B (en) | Data recommendation method and device | |
| US20190246968A1 (en) | System and a method for enabling responsive cognitive behavioral therapy | |
| CN107230154A (en) | The recognition methods of life insurance Claims Resolution case with clique's risk of fraud and device | |
| CN105408894B (en) | A kind of user identity classification determines method and device | |
| CN110309060A (en) | Detection method, device, storage medium and the computer equipment that recognizer updates | |
| CN109302301A (en) | A kind of appraisal procedure and device of business game | |
| CN110442791A (en) | Data push method and system | |
| CN110197402B (en) | User label analysis method, device, equipment and storage medium based on user group | |
| KR20130065849A (en) | System for recommend the customized application, method thereof and recordable medium storing the method | |
| Exler et al. | Smartphone-based estimation of a user being in company or alone based on place, time, and activity | |
| CN113411453A (en) | Intelligent management method and device of outbound object, medium and computing equipment | |
| WO2018203843A2 (en) | A system for making subscriber classification | |
| WO2018226173A2 (en) | A system for demographic analysis of mobile operator subscribers visiting the areas wherein indoor base station is established | |
| KR20150096573A (en) | Apparatus and method for displaying coupon in connection with coupon usage information | |
| US20190320072A1 (en) | A system for determining whether mobile subscribers are line holders or not | |
| CN106656943B (en) | A kind of matching process and device of network user's attribute | |
| CN109830290A (en) | Method for managing medical information, device, server and readable storage medium storing program for executing | |
| CN104933058A (en) | Network access activity monitoring method and system | |
| CN104994431B (en) | A kind for the treatment of method and apparatus of web video attention rate | |
| CN111382343B (en) | Label system generation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17908528 Country of ref document: EP Kind code of ref document: A2 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 17908528 Country of ref document: EP Kind code of ref document: A2 |