Applications of cluster analysis for cellular operators
Subscriber travel behavior is defined using cellular call location data. The defined travel behavior is used to segment the customer population. In a further aspect of the disclosed principles, a method to consolidate numerous of price plans is disclosed wherein price plans are grouped using cluster analysis. In the context of this disclosure, the term “cluster analysis” encompasses a number of different algorithms and methods for grouping objects of similar kind into respective categories to thus organize observed data into meaningful structures. In this context, cluster analysis is a data analysis process for sorting different objects into groups in a way that the degree of association between two objects is maximal if they belong to the same group and minimal otherwise.
1. A method for creating a customized marketing plan for a subscriber in a cellular network having a current price plan, the method implemented by a computing device executing computer readable instructions causing the computing device to perform a plurality of steps comprising:
defining, for a subscriber, a call pattern as a cluster profile, the call pattern comprising a plurality of call origin locations of the subscriber;
creating a plurality of travel pattern archetypes and associating a price plan archetype, respectively, with each of the travel pattern archetypes;
associating, based on the cluster profile, the subscriber with a travel pattern archetype from the plurality of travel plan archetypes, the travel pattern archetype having an associated price plan archetype;
comparing, for the subscriber, the current price plan with the associated price plan archetype; and
constructing a micro-marketing strategy for the subscriber based on the comparing.
2. The method according to claim 1 , wherein comparing the current price plan with the associated price plan archetype comprises executing a comparison based on profitability and net subscriber cost.
3. The method according to claim 1 , wherein the comparing the current price plan with the associated price plan archetype and constructing the micro-marketing strategy for the subscriber are executed in parallel.
4. The method according to claim 1 , wherein the defining a call pattern as a cluster profile comprises:
selecting a sample of subscribers from a total population;
for each subscriber, from the sample of subscribers, plotting a location of a call origin for each subscriber call;
for each subscriber, from the sample of subscribers, employing cluster analysis to identify a number of clusters for the subscriber, wherein clusters are denoted by respective centroids and radii;
using statistical grouping, on the number of clusters for each subscriber, to identify one or more call pattern archetypes; and
establishing the cluster profile by associating the call pattern, for the subscriber, with one of the call pattern archetypes.
5. The method according to claim 4 , wherein plotting a location comprises plotting a latitude and longitude of an originating tower.
6. The method according to claim 4 , wherein the plotting a location of a call origin for each call comprises plotting a location of a call origin for each call in a specified time interval.
7. The method according to claim 4 , wherein the plotting a location of a call origin for each call comprises determining whether the call origin is within a coverage space of a predetermined operator.
8. The method according to claim 4 , wherein the employing cluster analysis to identify a number of clusters and respective centroids and radii is executed via a blind random sample of subscribers.
9. The method according to claim 4 , wherein the employing cluster analysis to identify a number of clusters and respective centroids and radii is executed via a structured sample.
10. The method according to claim 4 , further comprising:
defining a specific price plan for each travel pattern archetype;
associating a stereotypical role with each call pattern archetype; and
associating each subscriber with a price plan by calculating the call pattern based on a call history of the subscriber.
11. A computer-readable medium having thereon computer-executable instructions for creating a customized marketing plan for a subscriber in a cellular network having a current price plan, the computer-executable instructions comprising instructions for:
defining, for a subscriber, a call pattern as a cluster profile, the call pattern comprising a plurality of call origin locations of the subscriber;
creating a plurality of travel pattern archetypes and associating a price plan archetype, respectively, with each of the travel pattern archetypes;
associating, based on the cluster profile, the subscriber with a travel pattern archetype from the plurality of travel plan archetypes, the travel pattern archetype having an associated price plan archetype;
comparing, for the subscriber, the current price plan with the associated price plan archetype; and
constructing a micro-marketing strategy for the subscriber based on the comparing.
12. The computer-readable medium according to claim 11 , wherein the instructions for comparing the current price plan with the associated price plan archetype comprises instructions for executing a comparison based on profitability and net subscriber cost.
13. The computer-readable medium according to claim 11 , wherein the instructions for comparing the current price plan with the associated price plan archetype and constructing the micro-marketing strategy for the subscriber are executed substantially in parallel.
14. The computer-readable medium according to claim 11 , wherein the instructions for the defining a call pattern as a cluster profile comprises instructions for:
selecting a sample of subscribers from a total population;
plotting, for each subscriber from the sample of subscribers, a location of a call origin for each subscriber call;
employing, for each subscriber from the sample of subscribers, cluster analysis to identify a number of clusters for the subscriber, wherein clusters are denoted by respective centroids and radii for each subscriber;
using statistical grouping, on the number of clusters for each subscriber, to identify one or more call pattern archetypes; and
establishing the cluster profile by associating the call pattern, for the subscriber, with one of the call pattern archetypes.
15. The computer-readable medium according to claim 14 , wherein the instructions for plotting a location comprise instructions for plotting a latitude and longitude of an originating tower.
16. The computer-readable medium according to claim 14 , wherein the instructions for the plotting a location of a call origin for each call comprise instructions for plotting a location of a call origin for each call in a specified time interval.
17. The computer-readable medium according to claim 14 , wherein the instructions for the plotting a location of a call origin for each call comprise instructions for determining whether the call origin is within a coverage space of a predetermined operator.
18. The computer-readable medium according to claim 14 , wherein the instructions for the employing cluster analysis to identify a number of clusters and respective centroids and radii are executed via a blind random sample of subscribers.
19. The computer-readable medium according to claim 14 , further comprising instructions for:
defining a specific price plan for each travel pattern archetype;
associating a stereotypical role with each call pattern archetype; and
associating each subscriber with a price plan by calculating the call pattern based on a call history of the subscriber.